Journals of Gerontology: bioLogical SCIENCES Cite journal as: J Gerontol A Biol Sci Med Sci doi:10.1093/gerona/glu019
© The Author 2014. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: [email protected].
Targeting of DNA Damage Signaling Pathway Induced Senescence and Reduced Migration of Cancer cells Ran Gao,1,2 Rumani Singh,1,2 Zeenia Kaul,1,3 Sunil C. Kaul,1,2 and Renu Wadhwa1,2 1 Cell Proliferation Research Group and DBT-AIST International Laboratory for Advanced Biomedicine, National Institute of Advanced Industrial Science & Technology (AIST), Tsukuba, Ibaraki, Japan. 3 Department of Molecular Virology, Immunology and Medical Genetics, Wexner Cancer Center, College of Medicine, The Ohio State University, Columbus. 2
The heat shock 70 family protein, mortalin, has pancytoplasmic distribution pattern in normal and perinuclear in cancer human cells. Cancer cells when induced to senesce by either chemicals or stress showed shift in mortalin staining pattern from perinuclear to pancytoplasmic type. Using such shift in mortalin staining as a reporter, we screened human shRNA library and identified nine senescence-inducing siRNA candidates. An independent Comparative Genomic Hybridization analysis of 35 breast cancer cell lines revealed that five (NBS1, BRCA1, TIN2, MRE11A, and KPNA2) of the nine genes located on chromosome regions identified as the gain of locus in more than 80% cell lines. By gene-specific PCR, these five genes were found to be frequently amplified in cancer cell lines. Bioinformatics revealed that the identified targets were connected to MRN (MRE11-RAD50-NBS1) complex, the DNA damage–sensing complex. We demonstrate that the identified shRNAs triggered DNA damage response and induced the expression of tumor suppressor protein p16INK4A causing growth arrest of cancer cells. Furthermore, cells showed decreased migration, mediated by decrease in matrix metalloproteases. Taken together, we demonstrate that the MRN complex is a potential target of cancer cell proliferation and migration, and staining pattern of mortalin could serve as an assay to identify senescence-inducing/anticancer reagents. Key Words: Cancer—Induced senescence—Mortalin staining pattern—shRNA screening—Bioinformatics—DNA damage pathway—p16INK4A—Growth arrest. Received August 20, 2013; Accepted January 22, 2014 Decision Editor: Rafael de Cabo, PhD
N
ormal human cells possess limited proliferative capacity and after a certain number of divisions in vitro enter a stage of permanent growth arrest (1). During this stage (called replicative senescence), cells remain viable and metabolically active; however, they are refractory to proliferation stimuli providing a barrier to immortalization, a crucial step in human carcinogenesis (2–4). Senescent cells are characterized by altered cell morphology such as flat, irregular, and giant size; increased cytoplasmic granularity and large nucleus; G1 arrest; expression of β-galactosidase; increased expression of tumor suppressor proteins including p53, p21WAF1, p16INK4A, pRB, and plasminogen activator protein (5–7). Genetic alterations that override senescence have been shown to be tightly associated with the development of tumors providing evidence that senescence is an innate antitumor mechanism in normal cells and offers new perspectives for cancer therapy (7,8). In order to form tumors, incipient cancer cells must evade senescence and attain additional characteristics such as growth factor–independent proliferation, anchorage independence growth, resistant to contact inhibition, invasion and migration by enhanced degradation
of matrix components, deregulation of apoptosis, and angiogenesis. Although these features may serve as good targets for anticancer drug, induction of senescence/permanent growth arrest has been predicted to yield beneficial outcomes (8). On the other hand, senescence cells show resistance to chemotherapy, and hence, induction of senescence remains to be validated as an effective and superior approach. Nevertheless, induction of cellular senescence has been demonstrated to be a good model to identify components and understand the functioning of tumor suppression mechanisms (7) that may lead to development of new anticancer drugs (8). Mortalin, a member of heat shock 70 (hsp70) family proteins, exists in multiple subcellular sites including mitochondria, endoplasmic reticulum, plasma membrane, cytosol, and nucleus. In addition to its essential role in mitochondrial import, energy generation, intracellular trafficking, and chaperoning (9), it promotes tumorigenesis by inactivation of tumor suppressor protein p53, deactivation of apoptosis, and stimulation of angiogenesis (9–12). It is upregulated in tumors and tumor-derived cell lines (13–16). In normal cells, mortalin is widely distributed in Page 1 of 13
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Address correspondence to Renu Wadhwa, MPhil, PhD, DBT-AIST International Laboratory for Advanced Biomedicine, National Institute of Advanced Industrial Science & Technology (AIST), Central 4, 1-1-1 Higashi, Tsukuba, Ibaraki 305–8562, Japan. Email: [email protected]
Page 2 of 13
Gao et al.
duplicate, covering the entire human genome with 1 Mbp resolution) provided by Macrogen (MAC Array KARYO 4,000) using Hybri-station (Genomic Solutions, Ann Arbor, MI) (24). The array slides were scanned at 532 and 635 nm using GenePix4000A (Axon Instruments, Union, CA) and were analyzed by Mac Viewer software (Macrogen) as described (25,26). Genomic PCR Genomic DNA of 22 cancer cell lines as well as human normal genomic DNA was obtained from ATCC. PCR conditions for linear range were established using normal genomic DNA and internal gene (GAPDH) amplification. A total of 100 ng of DNA was amplified in a reaction volume of 20 μL consisting of 0.4 μM each primer, 200 μM each dNTP, 10× PCR buffer, and 0.2 U of Taq DNA Polymerase (Takara, Japan), using the following PCR protocol: 3 min initial denaturation at 94°C, followed by 30 cycles of 30 s at 94°C denaturation, 45 s at separate annealing temperature (KPNA2 at 50°C, TIN2 at 50°C, MRE11A at 46°C, NBS1 at 48°C, BRCA1 at 52°C, and GAPDH at 54°C), and 2 min at 72°C elongation, and then a final 7 min at 72°C for extension. PCR products were resolved by 1% agarose gel electrophoresis and stained with Red gel (BIOTIUM, CA). Gene-specific primer sequences were as follows: Gene
Materials and Methods Cell Culture and Transfections Human biliary tract carcinoma (SKCHA-1), gall bladder carcinoma (TGBC-2), breast carcinomas (SKBR3, T47D, MDA-MB231, MDA-MB157, and MCF7), and osteogenic sarcoma (U2OS) were cultured in Dulbecco’s modified Eagle’s minimal essential medium (Gibco) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin in a humidified incubator (37°C and 5% CO2). Screening of shRNA library was done as described earlier (23,24). For biochemical and visual analyses, selected shRNA expression plasmids (2 μg) were transfected into cells plated in 6-well plate at 70% confluency. The cells were processed for various assays as described subsequently. BAC Array and CGH Analysis CGH analysis, as described earlier (25), was performed on 24 breast, 7 gall bladder, and 4 bile duct cancer cell lines. Genomic DNA (500 ng) from the test (cancer cells) and the reference (human genomic DNA, Promega) were labeled by random priming reaction using cyanine 5-dCTP and cyanine 3-dCTP (Perkin Elmer; Fremont, CA), Bioprime DNA Labeling System (Invitrogen, Carlsbad, CA), and Array Kit (Macrogen, Korea). Labeled test and reference DNAs were hybridized to the CGH array (4030 BAC clone DNAs in
Sense primer
Antisense primer
KPNA2
CACCTTCATGAAGTCATTGTCTT
TCTAGAATGAAAGGCTGGGG
TIN2
CTTCATTCCTACTAAACTACTTG
ACTGTAGAGACAGTTCTAGACCT
MRE11A
TGTCTCAATTTGTTTGAATATCCTTT
CCAAGGGAATATGGTAGATAGTGC
NBS1
CACTCCGTTTACAATTTAATAGC
CACAAAATCCCAAAATGAAATACG
BRCA1
GCTAAAAATACACGGATGG
TGCAAGACTGCGTCTC
GAPDH
TTGCCATCAATGACCCCTTCA
CGCCCCACTTGATTTTGGA
Cell Viability Assays For cell viability assays, AlamarBlue was used. Cells were treated at the density of 4,500 cells/well on 96-well plates and incubated at 37°C and 5% CO2 overnight. Next day, 1/10th volume of AlamarBlue (Invitrogen) was directly added to the culture medium and incubated at 37°C for 1–4 hours. Conversion of the blue nonfluorescent active ingredient (resazurin) to its reduced form resorufin that produced bright red fluorescence was monitored at 570 nm. For colony forming assay, 500 cells were plated in 10-cm dishes and incubated at 37°C and 5% CO2 for 1 week. Cells were fixed in acetone and ethanol (1:1) followed by staining with 0.1% crystal violet in distilled water for 30 min. The number of colonies was counted. Cell Migration Assay Cells were grown in a monolayer, and a wound was made by completely scratching the cells in a line with a 200 μL pipette tip. Cells were washed a few times with PBS to remove cell debris and incubated with fresh medium. The time of scratching wound was designated as time 0. Cells
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
the cytoplasm showing pancytoplasmic staining pattern. In tumor-derived cells, it shows perinuclear distribution (17). Interestingly, induction of senescence in cancer cells by drugs (MKT-077, BrdU, Withaferin A, H2O2, 5Aza-dC, and Epoximicin), by peptides, or by single chromosomes/chromosome fragments caused reallocation of pancytoplamic mortalin staining (18–22), suggesting that it could serve as an useful reporter to screen cancer targets. Furthermore, the shift in mortalin staining pattern was associated with abrogation of mortalin-p53 interaction, release of p53 from mortalin-p53 complex, and translocation to nucleus resulting in an activation of its transcriptional activation function (18,20,21). By using mortalin staining as an internal reporter to examine the induction of senescence in cancer cells, we screened human shRNA library and identified nine anticancer candidates (23). In the present study, we investigated these nine genes in the Comparative Genomic Hybridization (CGH) outcomes of 24 breast cancer cell lines and found that 5/9 gene targets (NBS1, BRCA1, TIN2, MRE11A, and KPNA2) were located on chromosomal regions that were frequently amplified in cancer cell lines. We demonstrate that these five genes are commonly amplified in cancer cells. Knockdown of their expression by shRNAs triggered DNA damage response and activation of tumor suppressor, p16INK4A, causing growth arrest and loss of malignant properties suggesting their potential as cancer targets.
Senescence and Reduced Migration of Cancer cells
were allowed to proliferate and migrate into the wound during the next 24 hours. Migration of cells into the wound was recorded under a phase contrast microscope with a 10× phase objective. Migration capacity was quantitated by measuring the percent of open area in 6–10 randomly captured images.
Immunostaining For immunostaining, cells were plated on glass coverslips placed in a 12-well plate (104 cells/coverslip) and processed as described (23). Antibodies used were monoclonal anti-mortalin (23), CARF (27), NBS1 (Y112), BRCA1 (MS13), KPNA2 (ab99288), TIN2 (ab64386; Abcam), pRB (Ser795), γH2AX (Ser139), MRE11 (Cell Signaling), MMP-2 (H-76), MMP-3 (AA07), p16 (F-12), p21 (C-19), and p53 (DO-1; Santa Cruz Biotech, CA) for 1 hour at room temperature, washed thrice with 0.2% Triton X-100 in PBS, and then incubated with Alexa-594-conjugated and Alexa488-conjugated (Molecular Probes, Eugene, OR) secondary antibodies. The stained cells were washed twice in PBS and then observed immediately on a Carl Zeiss microscope (Axiovert 200 M). Western Blotting Protein extraction was carried out using standard methods, and protein concentrations were determined by BCA assay (Pierce, Thermo Scientific). Lysates containing 20 μg protein were boiled in sodium dodecyl sulfate sample buffer for 5 min, subjected to 10% sodium dodecyl sulfate–polyacrylamide
gel electrophoresis, and then transferred onto a polyvinylidene difluoride membrane. Membranes were blocked in TBS-T containing 5% skim milk. After blocking, membranes were incubated with antibodies: BRCA1 (MS13), TIN2 (ab64386), NBS1 (Y112; ab32074; Abcam), MMP-9 (C20), MMP-3 (AA07), MMP-2 (H-76), p16 (C-20), (Santa Cruz Biotech, CA), Rad 50 (Cell Signaling), and CARF. Actin was used as an internal loading control (Chemicon International, CA). After probing with the primary antibody as indicated, the blots were incubated with horse radish peroxidase– conjugated secondary antibody (sc-2004 or sc-2005; Santa Cruz Biotech). Finally, the blots were developed using enhanced chemiluminescence (RPN2132; GE Healthcare, Buckinghamshire, UK). Senescence Assay SA-β-galactosidase activity was evaluated using the “Senescence-β-Gal Staining Kit” (Cell Signaling Technology) following the manufacturer’s instructions. Results and Discussion Identification of BRCA1, TIN2, NBS1, KPNA2, and MRE11A as Candidate Senescence-Inducing shRNAs We transfected human breast (MCF7) and bone (U2OS) cancer cells with expression vectors encoding 768 human shRNAs. Transfected cells were selected in the culture medium supplemented with puromycin (0.5 µg/mL) and examined for mortalin staining pattern under the automated scanning system attached to Axiovert 200M microscope (Zeiss). As shown in Figure 1, the shRNAs that caused shift in mortalin staining from perinuclear to pancytoplasmic in cancer cells were selected for the second round of screening. The process was repeated 4 times by transfecting the selected shRNAs from each round, and finally 22 shRNAs were identified (23) that resulted in induction of senescence. The phenomenon was previously demonstrated when senescence was induced in cancer cells with anticancer drugs, phytochemicals, peptides, and chromosome fragments (18–22,28). Interestingly, all the 22 identified gene targets have been predicted to be involved in human carcinogenesis (23). We previously investigated the effect of the 22 shRNAs on six other cancer cell lines (four breast carcinomas—SKBR3, T47D, MDA-MB231, MDA-MB157; gall bladder carcinoma—TGBC-2; biliary tract carcinoma— SKCHA-1) and found that 9 of the 22 shRNAs triggered the shift in mortalin staining pattern in all the six cell lines tested, suggesting that the gene targets of these 9 shRNAs may play an important or essential role during carcinogenesis (23). Furthermore, shift in mortalin staining pattern was observed in cell lines with various genetic background including cells with wild-type p53 and lack of p16INK4A function (U2OS), compromised p53 and lack of p16INK4A function (MCF7), mutant p53 and wild-type pRB (SKBR3,
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Cell Invasion Assay Invasion chamber (QCM ECMatrix Cell Invasion Assay, 24-well [8 μm], colorimetric) was brought to room temperature in a tissue culture hood. Serum-free medium (SFM, 300 μL, warmed at 37°C) was added to inserts and incubated for 2 hours at room temperature for rehydration of ECM layer. After rehydration, media were carefully removed from the inserts without disturbing the membrane. Cell suspension containing 0.5–1.0 × 106 cells/mL in serum-free media was prepared and 300 μL was added to each insert. Media (500 μL) containing 10% FBS was added to the lower chamber. The invasion chambers were incubated for 24–72 hours in CO2 incubator. Using cottontipped swabs, the ECMatrix gel and noninvading cells were gently removed from the interior of the inserts. In order to stain the invasive cells on lower surface of the membrane, the inserts were dipped in 500 μL of staining solution that was aliquoted in the unused well of the plate for 20 min. The stained inserts were rinsed in water and air-dried. Cells were quantitated by dissolving stained cells in 10% acetic acid (100 μL/well) and transferred to 96-well plate for colorimetric reading of optical density at 560 nm.
Page 3 of 13
Page 4 of 13
Gao et al.
T47D, MDA-MB 231), compromised p53 and wild-type pRB function (MCF7), and lack of both p53 and pRB (Saos-2, MDA-MB157, and TGBC-2), suggesting that the phenomenon is independent to the status of p53 and pRB, the two main tumor suppressor genes. Furthermore, these cell lines possessed different tumor-forming and malignant characteristics. Although some of these including U2OS, Saos-2, and MCF7 were nontumorigenic and nonmetastatic, the others including SKBR3, T47D, MDA-MB 231, and MDA-MB 157 were tumorigenic and metastatic, suggesting that the shift in mortalin staining pattern is independent of the malignant properties of the tumor cells. In order to further reveal the significance of the identified gene targets in cancer, we screened the outcomes of an independent CGH array analysis in 35 cancer cell lines (25,26) and found that the loci of five (BRCA1, TIN2, NBS1, KPNA2, and MRE11A) of the nine gene targets were located on the chromosomal regions that were identified as
gain of locus in cancer cells (Figure 1). We then performed genomic PCR with specific primers for the five selected genes on 22 cell lines and found their amplification in majority of them. As shown in Figure 1C, 19/22 cell lines showed amplification of BRCA1, 14/22 showed amplification of TIN2, 20/22 showed amplification of NBS1, 20/22 showed amplification of KPNA2, and 16/22 showed amplification of MRE11A gene loci. Furthermore, these proteins have been reported as upregulated in a variety of cancers and contribute to carcinogenesis in several studies (23). BRCA1, TIN2, NBS1, KPNA2, or MRE11A Silencing Affected Cancer Cell Migration We first investigated the induction of senescence by selected shRNAs in U2OS and MCF7 cells. The cells were transfected with specific shRNAs. Transfection efficacies were examined by cotransfection of a reporter gene (GFP in a
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 1. Identification of candidate anticancer shRNAs, based on the shift in mortalin staining pattern and CGH analysis in cancer cell lines. (A) Schematic presentation of shRNA library screening using mortalin staining as a reporter. Human breast carcinoma (MCF7) and osteosarcoma (U2OS) were used. In four rounds of screening, 22 shRNA were identified that were seen to shift the mortalin staining from perinuclear to pancytoplasmic type. Subsequent assays in six cancer cell lines (4 breast and 2 gall bladder) showed the shift in staining pattern in 9 out of 22. (B) CGH screening of 35 breast cancer cell lines revealed that five (BRCA1, TIN2, NBS1, KPNA2, and MRE11A) of nine genes were located on the chromosomal regions that are frequently amplified in cancer cells. (C) Expression analysis of five selected shRNAs target genes in human cancer cell lines. Gene-specific PCR was carried on 22 cancer cell lines and human normal genomic DNA. Quantitative analysis of gene-specific PCR showed their amplification in majority of the cell lines.
Senescence and Reduced Migration of Cancer cells
Page 5 of 13
ratio of 50:1/shRNA:GFP plasmid; Figure 2A). Knockdown effect of shRNAs was examined by immunostaining each target gene using specific antibodies (Figure 2B). We found that the transfected cells (both MCF7 and U2OS) showed decreased (about 70%) expression of the targeted genes in more than 60%–70% of cells. At 48–72 hours posttransfections, no immediate cell death was observed in response to knockdown of either of the five (BRCA1, TIN2, NBS1, KPNA2, and MRE11A) genes. The cells were examined for mortalin and senescence-associated β-gal staining. We found that the shift in mortalin staining pattern was associated with increase in β-gal staining in cells compromised of either of these five genes (Figure 2C, D, and E). Cell viability (determined by Vi-CELL Series Cell Viability Analyzers and Alamar Blue assay) and invasion (migration through Matrigel and wound scratch assay) assays revealed decrease in cell proliferation (20%–50%) and invasion through Matrigel (15%–35%) and migration to the wound (~2.5- to 3-fold), respectively (Figure 2F, G, and H). Based on these data, we subjected the transfected cells to quantitative in vitro real-time invasion assay using RTCA DP instrument (Roche Diagnostics GmbH, Germany). The data also
revealed that the knockdown of the selected genes caused decrease in their migration capacity (data not shown). Pathway Analysis of Gene Targets To get further insights into the functional mechanism of shRNA-induced senescence in cancer cells, we performed bioinformatics analysis on the selected genes (BRCA1, TIN2, NBS1, KPNA2, and MRE11A) by MetaCore Pathway and Data Mining (http://www.genego.com/genego_lp.php). The analysis revealed that besides their role in diverse pathways including apoptosis, telomere maintenance, immune response, and protein stability, the five genes were most closely connected to the DNA damage response signaling (Figure 3A). The latter spans a complex network of pathways, ranging from sensing of the DNA lesion, localization of the DNA repair proteins to the lesion, and DNA repair and growth arrest by network of phosphorylating key players as described subsequently. BRCA1, the breast- and ovarian cancer–specific tumor suppressor, is a multifunctional nuclear protein involved in the regulation of DNA repair, cell cycle, transcription, chromatin remodeling, and genome integrity. Inherited
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 1. Continued.
Page 6 of 13
Gao et al.
mutations or loss of BRCA1 confers a high risk for breast cancer development, whereas its altered expression occurs frequently in sporadic forms of cancer. It has been shown that BRCA1 is an autoregulated transcription factor that selectively titrates its levels to maintain genome integrity in response to genotoxic insults (29). MRE11A (Meiotic recombination 11) is an essential DNA double-strand break (DSB) repair protein. It has exonuclease and endonuclease activities and functions in homologous recombination and telomere length maintenance in cancer cells (30). NBS1 is a protein involved in Nijmegen breakage syndrome (NBS, a chromosomal instability syndrome) associated with cancer predisposition, radiosensitivity, microcephaly, and growth retardation. It is a component of the MRN (MRE11-RAD50-NBS1) complex, a central player associated with DSB repair (31). MRN is one of the first factors to be localized in the DNA lesion (hence also called lesion sensor), where it plays a structural role of tethering together the broken chromosomes and initiate a cascade of DNA repair signaling events (32) controlled by the ATM protein kinase, a major regulator of the DSB. NBS1 is phosphorylated by ATM in response to radiation,
gets transported to the nucleus by KPNA2, and controls the nuclear localization of the MRN complexes (33). It functions as a tumor suppressor in preserving genome integrity in the nucleus and acts as an oncogenic protein in the cytoplasm by PI3-kinase/AKT-activation pathway (33). MRN functions to target ATM and Tip60 acetyltransferase to DSBs (34). Overexpression of NBS1 has been shown to induce epithelial–mesenchymal transition, contribute to transformation, and serve as a marker of aggressive head and neck cancer (35). Depletion of NBS1 resulted in an inhibition of ALT-, but not telomerase-mediated, telomere maintenance (36). TRF1 interacting nuclear protein 2 (TIN2), essential for embryonic development and cell viability, is a negative regulator of telomere length elongation and is enriched in hepatocarcinoma (37). Karyopherinalpha2 (KPNA2) was found upregulated in a variety of cancers. It mediates nuclear localization of NBS1 complex in response to DNA damage (38–40). In light of the above known functions of these proteins, we examined whether shRNA-mediated knockdown of these genes evoked DNA damage response in cancer cells. shRNA-transfected cells were examined for DNA
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 2. Cell viability and migration analysis compromised for the selected gene targets. (A) Transfection efficacy of the shRNA plasmids as revealed by cotransfection of GFP reporter. (B) Knockdown efficacy of the shRNAs as revealed by immunostaining of cells with specific antibodies. (C) Shift in mortalin staining pattern from perinuclear (vector) to pancytoplasmic (shRNA-transfected cells) type. (D and E) Senescence-associated β-gal (SA-β-gal) staining in control (vector) and shRNA-transfected cells and their quantitation from three independent experiments. (F) Viability of MCF7 cells transfected with selected shRNAs was examined at approximately 48–72 hours posttransfection. Invasion index as examined by Matrigel invasion assays (G) and invasive capacity of cells in wound scratch assays (H). *P < .05; **P < .01; ***P < .001.
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 2. Continued.
Page 7 of 13 Senescence and Reduced Migration of Cancer cells
Page 8 of 13
Gao et al.
damage response proteins, γH2AX, CARF, and phosphoCHK2 (pCHK2) (41,42), and their downstream regulators. As shown in Figure 3B, immunostaining assay in MCF7 cells transfected with shRNA for BRCA1, TIN2, NBS1, KPNA2, or MRE11A showed increase in γH2AX, CARF, and pCHK2 that signified DNA damage response. We examined the status of p53, p21, and p16INK4A in control and transfected cells and found that the shRNAs did not cause nuclear translocation of p53. The expression of p21 also remained unchanged in three of the four cell lines. On the other hand, the level of p16INK4A was found to be consistently enhanced when BRCA1, NBS1, or MRE11A were knockdown in all the four cell lines in several independent experiments (Figure 3C). Induction of p16INK4A level was also confirmed by Western blotting of control and shRNA-transfected MCF7 and U2OS cells (Figure 3D–F). p16INK4A is known to be silenced by promoter methylation in these cells. The present data strongly suggested that the
knockdown of target genes may cause demethylation of the p16INK4A promoter. Based on these results, it was suggestive that besides their role in DNA damage and repair, the target genes might be associated with histone methylation. Although conventionally, histone demethylases are believed to regulate DNA methylation and gene transcription, several recent studies have shown a cross talk between the DNA damage/growth arrest and DNA demethylation. These include role of histone demethylase LSD1 in DNA damage response (43) and role of growth arrest and DNA damage–induced protein, GADD45 alpha, in DNA demethylation (44,45). PPARγ was also shown to promote DNA demethylation (46). In view of these recent reports and our data, it is suggestive that the knockdown of DNA damage response proteins (NBS1, BRCA1, TIN2, MRE11A, and KPNA2) may cause DNA demethylation as supported by expression of p16INK4A. The mechanism of such induction of DNA demethylation warrants further studies.
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 3. Selected shRNAs evoke DNA damage response in cancer cells. (A) Bioinformatics analysis on the selected genes (NBS1, BRCA1, TIN2, MRE11A, and KPNA2) by MetaCore Pathway and Data Mining showed that they are closely connected to the DNA damage response pathway. (B) MCF7 cell transfected with BRCA1, TIN2, NBS1, KPNA2, or MRE11A showed increase in γH2AX, CARF (Collaborator of ARF), and pCHK2. (C) Selected candidate shRNAs induced senescence in cancer cells through p16INK4A pathway. Selected candidate shRNAs were transfected into four cell lines (SKBR3, MCF7, U2OS, and TGBC-2) and examined for p53, p21WAF1, and p16INK4A by immunostaining. Representative staining for mortalin, p53, p21WAF1, and p16INK4A is shown in control and transfected cells. Results in four cancer cell lines are summarized showing consistent induction of p16INK4A. (D) Induction of p16INK4A in shRNA-transfected cells was confirmed by Western blotting. Quantitation of induction of p16INK4A expression from four independent experiments (E). (F) Immunostaining of p16INK4A in control and shRNA-transfected MCF7 and U2OS cells. *P < .05.
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 3. Continued.
Page 9 of 13 Senescence and Reduced Migration of Cancer cells
Page 10 of 13
Gao et al.
increased β-gal staining (Figure 4A). Cell migration analyses by wound scratch assay revealed that the p16INK4A overexpression also caused decrease in migration compared to the control cells (Figure 4C and D). Although the role of p16INK4A in growth arrest has been well documented, its contribution to cell migration remains unclear. Therefore, we examined the expression of cell migration markers in p16INK4A overexpressing cells. As shown in Figure 4E, we found that these cells have decrease in phosphorylation of pRB and increase in CARF (27) signifying the growth arrest. At the same time regulators of cell migration such as MMP-2 and -3 were downregulated (Figure 4E and F). Interestingly, four (MRE11A, NBS1, BRCA1, and TIN2) target genes showed decrease in p16INK4A overexpressing cells indicating a feedback regulation between these proteins and reflecting molecular functional economy in which the cells under growth arrest downregulate the DNA damage signaling proteins. In other words, p16INK4A acted as an effector protein of DNA damage response initiated by MRN complex proteins and triggered growth arrest
Figure 4. Overexpression of p16INK4A caused growth arrest and loss of migration in cancer cells. (A) p16INK4A overexpressing cells showed growth arrest in colony forming assay. Both colony number and size were reduced, and cells showed SA-β-gal staining (B). These cells exhibited loss of cell migration as shown by wound scratch (C) and Matrigel assays (D). (E) The p16INK4A overexpressing cells showed cell cycle arrest as marked by decrease in phospho-RB. Loss of migration was marked by decrease in MMP-2 and MMP-3 expression. There was a decrease in NBS1, BRCA1, TIN2, MRE11, and RAD50. (F) Immunostaining of control and p16INK4A overexpressing cells for the selected gene targets, matrix metalloproteases, and phospho-RB is shown. *P < .05; **P < .01.
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Loss of Migration in shRNA-Transfected Cancer Cells was Mediated by Induction of p16INK4A As shown earlier, the knockdown of BRCA1, KPNA2, MRE11A, NBS1, or TIN2 triggered growth arrest and loss of migration in cancer cells by DNA damage response mediated by induction of p16INK4A in four cancer cell lines. The molecular mechanism of induction of p16INK4A by knockdown of five shRNAs in different cell lines remains to be resolved in an independent study. It is suggestive that the compromised expression of DNA damage and repair signaling proteins served as a DNA damage signal, triggered the formation of DNA damage foci (increase in γH2AX), and caused growth arrest (increase in CARF, p21WAF1, and p16INK4A). We verified the effect of p16INK4A on cell growth and migration by an alternative approach. Cells were stably transfected with p16INK4A expression vector. Cell proliferation assays including colony forming efficacy, colony size, and viability revealed significant growth arrest in p16INK4A overexpressing cells (Figure 4A and B). Furthermore, p16INK4A overexpressing cells showed
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 4. Continued.
Page 11 of 13 Senescence and Reduced Migration of Cancer cells
Page 12 of 13
Gao et al.
in cells. Growth arrested p16INK4A overexpressing cells, on the other hand, downregulated the DNA damage response genes. The DNA DSBs initiate DNA damage response that in turn activates relocalization of several proteins like MRE11A, NBS1, and BRCA1 to nuclear foci that colocalize with γH2AX and result in foci formation (47). It has been reported that γH2AX interacts directly with BRCA1 and NBS1, required for formation of MRN complex at DNA damage sites (30,32,47–49). Furthermore, KPNA2 was shown to be essential for nuclear localization of NBS1 (50). Identification of multiple components of this nuclear focus formation complex in our screening revealed that the DNA damage signaling is a strong candidate target for cancer therapy. Disruption of any element of this complex triggered senescence response in cancer cells as also marked by shift in staining pattern of mortalin, increase in p16INK4A and CARF expression, and decrease in phosphorylation of pRB and matrix metalloproteases (MMP-2 and -3). Changes in mortalin, CARF, p16Ink4A, and pRB signified growth arrest of cancer cells, whereas change in matrix metalloproteases (established regulators of tumor environment) revealed decrease in their migration and invasion capacity as has been reported in several other studies (51). Knockdown of either of the five targets appeared to cause induction of p16INK4A that inhibited cyclin/CDK complexes, resulting in inhibition of pRB phosphorylation and cell cycle. Increase in CARF was previously shown to activate p53-p21WAF1 tumor suppression pathway causing growth arrest of cancer cells (42). Taken together, knockdown of the identified gene targets caused activation of two major tumor suppressor pathways, p16INK4A-pRB and p53-p21WAF1 (Figure 5). Downregulation of p16INK4A occurs frequently with human cell transformation (52). It has also been correlated with the development of invasive characteristics of cancer cells (53). Introduction of p16INK4A was shown to result in decreased invasion and angiogenesis of human glioma (54,55). Consistent with these reports, we found that the overexpression of p16INK4A caused decrease in migration of both U2OS and MCF7 cells. Furthermore, p16INK4A overexpressing cells exhibited decreased expression of MMP-2 and MMP-3 (Figures 4 and 5) that may account for decreased invasion capacity of cells. Taken together, we report an outcome of a
novel endogenous reporter (mortalin staining) based screening for cancer gene targets that revealed DNA damage signaling pathway as a strong candidate target for induction of senescence and compromise the migration capacity of cancer cells. Funding This study was partly supported by grants from the New Energy and Industrial Technology Development Organization (NEDO) of Japan. Acknowledgment Authors thank Drs. Saito and T. Hirano for CGH analysis and T. Yaguchi for assistance. R.S. was a receipient of the Ministry of Education, Culture, Sports, Science and Technology (MEXT) scholarship. References 1. Hayflick L, Moorhead PS. The serial cultivation of human diploid cell strains. Exp Cell Res. 1961;25:585–621. 2. Campisi J. Aging and cancer: the double-edged sword of replicative senescence. J Am Geriatr Soc. 1997;45:482–488. 3. Reddel RR. The role of senescence and immortalization in carcinogenesis. Carcinogenesis. 2000;21:477–484. 4. Prieur A, Peeper DS. Cellular senescence in vivo: a barrier to tumorigenesis. Curr Opin Cell Biol. 2008;20:150–155. doi:10.1016/j. ceb.2008.01.007 5. Duncan EL, Wadhwa R, Kaul SC. Senescence and immortalization of human cells. Biogerontology. 2000;1:103–121. 6. Campisi J. Between Scylla and Charybdis: p53 links tumor suppression and aging. Mech Ageing Dev. 2002;123:567–573. 7. Campisi J. Senescent cells, tumor suppression, and organismal aging: good citizens, bad neighbors. Cell. 2005;120:513–522. 8. Lleonart ME, Artero-Castro A, Kondoh H. Senescence induction; a possible cancer therapy. Mol Cancer. 2009;8:3. doi:10.1186/ 1476-4598-8-3 9. Kaul SC, Deocaris CC, Wadhwa R. Three faces of mortalin: a housekeeper, guardian and killer. Exp Gerontol. 2007;42:263–274. 10. Wadhwa R, Takano S, Robert M, et al. Inactivation of tumor suppressor p53 by mot-2, a hsp70 family member. J Biol Chem. 1998;273:29586–29591. 11. Kaul SC, Aida S, Yaguchi T, Kaur K, Wadhwa R. Activation of wild type p53 function by its mortalin-binding, cytoplasmically localizing carboxyl terminus peptides. J Biol Chem. 2005;280:39373–39379. 12. Ma Z, Izumi H, Kanai M, Kabuyama Y, Ahn NG, Fukasawa K. Mortalin controls centrosome duplication via modulating centrosomal localization of p53. Oncogene. 2006;25:5377–5390. 13. Wadhwa R, Takano S, Kaur K, et al. Upregulation of mortalin/ mthsp70/Grp75 contributes to human carcinogenesis. Int J Cancer. 2006;118:2973–2980.
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 5. A model showing shift in mortalin staining pattern (gray) from the perinuclear to the pancytoplasmic type in cancer cells compromised for BRCA1, KPNA2, NBS1, MRE11, and TIN2 leading to the upregulation of genes involved in growth arrest and downregulation of genes involved in cell migration.
Senescence and Reduced Migration of Cancer cells
36. Zhong ZH, Jiang WQ, Cesare AJ, Neumann AA, Wadhwa R, Reddel RR. Disruption of telomere maintenance by depletion of the MRE11/ RAD50/NBS1 complex in cells that use alternative lengthening of telomeres. J Biol Chem. 2007;282:29314–29322. 37. Oh BK, Kim YJ, Park C, Park YN. Up-regulation of telomere-binding proteins, TRF1, TRF2, and TIN2 is related to telomere shortening during human multistep hepatocarcinogenesis. Am J Pathol. 2005;166:73–80. 38. Jensen JB, Munksgaard PP, Sørensen CM, et al. High expression of karyopherin-α2 defines poor prognosis in non-muscle-invasive bladder cancer and in patients with invasive bladder cancer undergoing radical cystectomy. Eur Urol. 2011;59:841–848. doi:10.1016/j. eururo.2011.01.048 39. Zheng M, Tang L, Huang L, et al. Overexpression of karyo pherin-2 in epithelial ovarian cancer and correlation with poor prognosis. Obstet Gynecol. 2010;116:884–891. doi:10.1097/ AOG.0b013e3181f104ce 40. Wang CI, Wang CL, Wang CW, et al. Importin subunit alpha-2 is identified as a potential biomarker for non-small cell lung cancer by integration of the cancer cell secretome and tissue transcriptome. Int J Cancer. 2011;128:2364–2372. doi:10.1002/ijc.25568 41. Dickey JS, Redon CE, Nakamura AJ, Baird BJ, Sedelnikova OA, Bonner WM. H2AX: functional roles and potential applications. Chromosoma. 2009;118:683–692. doi:10.1007/s00412-009-0234-4 42. Cheung CT, Singh R, Yoon AR, et al. Molecular characterization of apoptosis induced by CARF silencing in human cancer cells. Cell Death Differ. 2011;18:589–601. doi:10.1038/cdd.2010.129 43. Mosammaparast N, Kim H, Laurent B, et al. The histone demethylase LSD1/KDM1A promotes the DNA damage response. J Cell Biol. 2013;203:457–470. doi:10.1083/jcb.201302092 44. Schäfer A. Gadd45 proteins: key players of repair-mediated DNA demethylation. Adv Exp Med Biol. 2013;793:35–50. doi:10.1007/978-1-4614-8289-5_3 45. Li Y, Zhao M, Yin H, et al. Overexpression of the growth arrest and DNA damage-induced 45alpha gene contributes to autoimmunity by promoting DNA demethylation in lupus T cells. Arthritis Rheum. 2010;62:1438–1447. doi:10.1002/art.27363 46. Fujiki K, Shinoda A, Kano F, Sato R, Shirahige K, Murata M. PPARγinduced PARylation promotes local DNA demethylation by production of 5-hydroxymethylcytosine. Nat Commun. 2013;4:2262–2265. doi:10.1038/ncomms3262 47. Paull TT, Rogakou EP, Yamazaki V, Kirchgessner CU, Gellert M, Bonner WM. A critical role for histone H2AX in recruitment of repair factors to nuclear foci after DNA damage. Curr Biol. 2000;10:886–895. 48. Huen MS, Sy SM, Chen J. BRCA1 and its toolbox for the maintenance of genome integrity. Nat Rev Mol Cell Biol. 2010;11:138–148. doi:10.1038/nrm2831 49. Rodier F, Coppé JP, Patil CK, et al. Persistent DNA damage signalling triggers senescence-associated inflammatory cytokine secretion. Nat Cell Biol. 2009;11:973–979. doi:10.1038/ncb1909 50. Tseng SF, Chang CY, Wu KJ, Teng SC. Importin KPNA2 is required for proper nuclear localization and multiple functions of NBS1. J Biol Chem. 2005;280:39594–39600. 51. Kessenbrock K, Plaks V, Werb Z. Matrix metalloproteinases: regulators of the tumor microenvironment. Cell. 2010;141:52–67. doi:10.1016/j.cell.2010.03.015 52. Serrano M, Lee H, Chin L, Cordon-Cardo C, Beach D, DePinho RA. Role of the INK4a locus in tumor suppression and cell mortality. Cell. 1996;85:27–37. 53. Grover R, Chana JS, Wilson GD, Richman PI, Sanders R. An analysis of p16 protein expression in sporadic malignant melanoma. Melanoma Res. 1998;8:267–272. 54. Chintala SK, Fueyo J, Gomez-Manzano C, et al. Adenovirus-mediated p16/CDKN2 gene transfer suppresses glioma invasion in vitro. Oncogene. 1997;15:2049–2057. 55. Harada H, Nakagawa K, Iwata S, et al. Restoration of wild-type p16 down-regulates vascular endothelial growth factor expression and inhibits angiogenesis in human gliomas. Cancer Res. 1999;59:3783–3789.
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
14. Dundas SR, Lawrie LC, Rooney PH, Murray GI. Mortalin is overexpressed by colorectal adenocarcinomas and correlates with poor survival. J Pathol. 2005;205:74–81. 15. Yi X, Luk JM, Lee NP, et al. Association of mortalin (HSPA9) with liver cancer metastasis and prediction for early tumor recurrence. Mol Cell Proteomics. 2008;7:315–325. 16. Lu WJ, Lee NP, Kaul SC, et al. Mortalin-p53 interaction in cancer cells is stress dependent and constitutes a selective target for cancer therapy. Cell Death Differ. 2011;18:1046–1056. doi:10.1038/cdd.2010.177 17. Wadhwa R, Kaul SC, Mitsui Y, Sugimoto Y. Differential subcellular distribution of mortalin in mortal and immortal mouse and human fibroblasts. Exp Cell Res. 1993;207:442–448. 18. Wadhwa R, Sugihara T, Yoshida A, et al. Selective toxicity of MKT-077 to cancer cells is mediated by its binding to the hsp70 family protein mot-2 and reactivation of p53 function. Cancer Res. 2000;60:6818–6821. 19. Michishita E, Nakabayashi K, Suzuki T, et al. 5-Bromodeoxyuridine induces senescence-like phenomena in mammalian cells regardless of cell type or species. J Biochem. 1999;126:1052–1059. 20. Widodo N, Deocaris CC, Kaur K, et al. Stress chaperones, mortalin, and pex19p mediate 5-aza-2’ deoxycytidine-induced senescence of cancer cells by DNA methylation-independent pathway. J Gerontol A Biol Sci Med Sci. 2007;62:246–255. 21. Widodo N, Takagi Y, Shrestha BG, Ishii T, Kaul SC, Wadhwa R. Selective killing of cancer cells by leaf extract of Ashwagandha: components, activity and pathway analyses. Cancer Lett. 2008;262:37–47. doi:10.1016/j.canlet.2007.11.037 22. Kaul Z, Yaguchi T, Chiura HX, Kaul SC, Wadhwa R. Quantum dotbased mortalin staining as a visual assay for detection of induced senescence in cancer cells. Ann N Y Acad Sci. 2007;1100:368–372. 23. Gao R, Yaguchi T, Horimoto K, et al. Identification of anti-cancer shRNAs based on the staining pattern of mortalin. Tiss Cult Res Commun. 2010;29:147–153. 24. Wadhwa R, Kaul SC, Ikawa Y, Sugimoto Y. Identification of a novel member of mouse hsp70 family. Its association with cellular mortal phenotype. J Biol Chem. 1993;268:6615–6621. 25. Saito S, Morita K, Hirano T. High frequency of common DNA copy number abnormalities detected by bacterial artificial chromosome array comparative genomic hybridization in 24 breast cancer cell lines. Hum Cell. 2009;22:1–10. doi:10.1111/j.1749-0774.2008.00061.x 26. Saito S, Ghosh M, Morita K, Hirano T, Miwa M, Todoroki T. The genetic differences between gallbladder and bile duct cancer cell lines. Oncol Rep. 2006;16:949–956. 27. Hasan K, Cheung C, Kaul Z, et al. CARF Is a vital dual regulator of cellular senescence and apoptosis. J Biol Chem. 2009;284:1664–1672. doi:10.1074/jbc.M805778200 28. Bertram MJ, Bérubé NG, Hang-Swanson X, et al. Identification of a gene that reverses the immortal phenotype of a subset of cells and is a member of a novel family of transcription factor-like genes. Mol Cell Biol. 1999;19:1479–1485. 29. De Siervi A, De Luca P, Byun JS, et al. Transcriptional autoregulation by BRCA1. Cancer Res. 2010;70:532–542. doi:10.1158/0008-5472. CAN-09-1477 30. Deng R, Tang J, Ma JG, et al. PKB/Akt promotes DSB repair in cancer cells through upregulating Mre11 expression following ionizing radiation. Oncogene. 2011;30:944–955. doi:10.1038/onc.2010.467 31. Kobayashi J, Antoccia A, Tauchi H, Matsuura S, Komatsu K. NBS1 and its functional role in the DNA damage response. DNA Repair (Amst). 2004;3:855–861. 32. van den Bosch M, Bree RT, Lowndes NF. The MRN complex: coordinating and mediating the response to broken chromosomes. EMBO Rep. 2003;4:844–849. 33. Teng SC, Wu KJ, Tseng SF, Wong CW, Kao L. Importin KPNA2, NBS1, DNA repair and tumorigenesis. J Mol Histol. 2006;37:293–299. 34. Sun Y, Jiang X, Price BD. Tip60: connecting chromatin to DNA damage signaling. Cell Cycle. 2010;9:930–936. 35. Yang MH, Chang SY, Chiou SH, et al. Overexpression of NBS1 induces epithelial-mesenchymal transition and co-expression of NBS1 and Snail predicts metastasis of head and neck cancer. Oncogene. 2007;26:1459–1467.
Page 13 of 13
© The Author 2014. Published by Oxford University Press on behalf of The Gerontological Society of America. All rights reserved. For permissions, please e-mail: [email protected].
Targeting of DNA Damage Signaling Pathway Induced Senescence and Reduced Migration of Cancer cells Ran Gao,1,2 Rumani Singh,1,2 Zeenia Kaul,1,3 Sunil C. Kaul,1,2 and Renu Wadhwa1,2 1 Cell Proliferation Research Group and DBT-AIST International Laboratory for Advanced Biomedicine, National Institute of Advanced Industrial Science & Technology (AIST), Tsukuba, Ibaraki, Japan. 3 Department of Molecular Virology, Immunology and Medical Genetics, Wexner Cancer Center, College of Medicine, The Ohio State University, Columbus. 2
The heat shock 70 family protein, mortalin, has pancytoplasmic distribution pattern in normal and perinuclear in cancer human cells. Cancer cells when induced to senesce by either chemicals or stress showed shift in mortalin staining pattern from perinuclear to pancytoplasmic type. Using such shift in mortalin staining as a reporter, we screened human shRNA library and identified nine senescence-inducing siRNA candidates. An independent Comparative Genomic Hybridization analysis of 35 breast cancer cell lines revealed that five (NBS1, BRCA1, TIN2, MRE11A, and KPNA2) of the nine genes located on chromosome regions identified as the gain of locus in more than 80% cell lines. By gene-specific PCR, these five genes were found to be frequently amplified in cancer cell lines. Bioinformatics revealed that the identified targets were connected to MRN (MRE11-RAD50-NBS1) complex, the DNA damage–sensing complex. We demonstrate that the identified shRNAs triggered DNA damage response and induced the expression of tumor suppressor protein p16INK4A causing growth arrest of cancer cells. Furthermore, cells showed decreased migration, mediated by decrease in matrix metalloproteases. Taken together, we demonstrate that the MRN complex is a potential target of cancer cell proliferation and migration, and staining pattern of mortalin could serve as an assay to identify senescence-inducing/anticancer reagents. Key Words: Cancer—Induced senescence—Mortalin staining pattern—shRNA screening—Bioinformatics—DNA damage pathway—p16INK4A—Growth arrest. Received August 20, 2013; Accepted January 22, 2014 Decision Editor: Rafael de Cabo, PhD
N
ormal human cells possess limited proliferative capacity and after a certain number of divisions in vitro enter a stage of permanent growth arrest (1). During this stage (called replicative senescence), cells remain viable and metabolically active; however, they are refractory to proliferation stimuli providing a barrier to immortalization, a crucial step in human carcinogenesis (2–4). Senescent cells are characterized by altered cell morphology such as flat, irregular, and giant size; increased cytoplasmic granularity and large nucleus; G1 arrest; expression of β-galactosidase; increased expression of tumor suppressor proteins including p53, p21WAF1, p16INK4A, pRB, and plasminogen activator protein (5–7). Genetic alterations that override senescence have been shown to be tightly associated with the development of tumors providing evidence that senescence is an innate antitumor mechanism in normal cells and offers new perspectives for cancer therapy (7,8). In order to form tumors, incipient cancer cells must evade senescence and attain additional characteristics such as growth factor–independent proliferation, anchorage independence growth, resistant to contact inhibition, invasion and migration by enhanced degradation
of matrix components, deregulation of apoptosis, and angiogenesis. Although these features may serve as good targets for anticancer drug, induction of senescence/permanent growth arrest has been predicted to yield beneficial outcomes (8). On the other hand, senescence cells show resistance to chemotherapy, and hence, induction of senescence remains to be validated as an effective and superior approach. Nevertheless, induction of cellular senescence has been demonstrated to be a good model to identify components and understand the functioning of tumor suppression mechanisms (7) that may lead to development of new anticancer drugs (8). Mortalin, a member of heat shock 70 (hsp70) family proteins, exists in multiple subcellular sites including mitochondria, endoplasmic reticulum, plasma membrane, cytosol, and nucleus. In addition to its essential role in mitochondrial import, energy generation, intracellular trafficking, and chaperoning (9), it promotes tumorigenesis by inactivation of tumor suppressor protein p53, deactivation of apoptosis, and stimulation of angiogenesis (9–12). It is upregulated in tumors and tumor-derived cell lines (13–16). In normal cells, mortalin is widely distributed in Page 1 of 13
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Address correspondence to Renu Wadhwa, MPhil, PhD, DBT-AIST International Laboratory for Advanced Biomedicine, National Institute of Advanced Industrial Science & Technology (AIST), Central 4, 1-1-1 Higashi, Tsukuba, Ibaraki 305–8562, Japan. Email: [email protected]
Page 2 of 13
Gao et al.
duplicate, covering the entire human genome with 1 Mbp resolution) provided by Macrogen (MAC Array KARYO 4,000) using Hybri-station (Genomic Solutions, Ann Arbor, MI) (24). The array slides were scanned at 532 and 635 nm using GenePix4000A (Axon Instruments, Union, CA) and were analyzed by Mac Viewer software (Macrogen) as described (25,26). Genomic PCR Genomic DNA of 22 cancer cell lines as well as human normal genomic DNA was obtained from ATCC. PCR conditions for linear range were established using normal genomic DNA and internal gene (GAPDH) amplification. A total of 100 ng of DNA was amplified in a reaction volume of 20 μL consisting of 0.4 μM each primer, 200 μM each dNTP, 10× PCR buffer, and 0.2 U of Taq DNA Polymerase (Takara, Japan), using the following PCR protocol: 3 min initial denaturation at 94°C, followed by 30 cycles of 30 s at 94°C denaturation, 45 s at separate annealing temperature (KPNA2 at 50°C, TIN2 at 50°C, MRE11A at 46°C, NBS1 at 48°C, BRCA1 at 52°C, and GAPDH at 54°C), and 2 min at 72°C elongation, and then a final 7 min at 72°C for extension. PCR products were resolved by 1% agarose gel electrophoresis and stained with Red gel (BIOTIUM, CA). Gene-specific primer sequences were as follows: Gene
Materials and Methods Cell Culture and Transfections Human biliary tract carcinoma (SKCHA-1), gall bladder carcinoma (TGBC-2), breast carcinomas (SKBR3, T47D, MDA-MB231, MDA-MB157, and MCF7), and osteogenic sarcoma (U2OS) were cultured in Dulbecco’s modified Eagle’s minimal essential medium (Gibco) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin–streptomycin in a humidified incubator (37°C and 5% CO2). Screening of shRNA library was done as described earlier (23,24). For biochemical and visual analyses, selected shRNA expression plasmids (2 μg) were transfected into cells plated in 6-well plate at 70% confluency. The cells were processed for various assays as described subsequently. BAC Array and CGH Analysis CGH analysis, as described earlier (25), was performed on 24 breast, 7 gall bladder, and 4 bile duct cancer cell lines. Genomic DNA (500 ng) from the test (cancer cells) and the reference (human genomic DNA, Promega) were labeled by random priming reaction using cyanine 5-dCTP and cyanine 3-dCTP (Perkin Elmer; Fremont, CA), Bioprime DNA Labeling System (Invitrogen, Carlsbad, CA), and Array Kit (Macrogen, Korea). Labeled test and reference DNAs were hybridized to the CGH array (4030 BAC clone DNAs in
Sense primer
Antisense primer
KPNA2
CACCTTCATGAAGTCATTGTCTT
TCTAGAATGAAAGGCTGGGG
TIN2
CTTCATTCCTACTAAACTACTTG
ACTGTAGAGACAGTTCTAGACCT
MRE11A
TGTCTCAATTTGTTTGAATATCCTTT
CCAAGGGAATATGGTAGATAGTGC
NBS1
CACTCCGTTTACAATTTAATAGC
CACAAAATCCCAAAATGAAATACG
BRCA1
GCTAAAAATACACGGATGG
TGCAAGACTGCGTCTC
GAPDH
TTGCCATCAATGACCCCTTCA
CGCCCCACTTGATTTTGGA
Cell Viability Assays For cell viability assays, AlamarBlue was used. Cells were treated at the density of 4,500 cells/well on 96-well plates and incubated at 37°C and 5% CO2 overnight. Next day, 1/10th volume of AlamarBlue (Invitrogen) was directly added to the culture medium and incubated at 37°C for 1–4 hours. Conversion of the blue nonfluorescent active ingredient (resazurin) to its reduced form resorufin that produced bright red fluorescence was monitored at 570 nm. For colony forming assay, 500 cells were plated in 10-cm dishes and incubated at 37°C and 5% CO2 for 1 week. Cells were fixed in acetone and ethanol (1:1) followed by staining with 0.1% crystal violet in distilled water for 30 min. The number of colonies was counted. Cell Migration Assay Cells were grown in a monolayer, and a wound was made by completely scratching the cells in a line with a 200 μL pipette tip. Cells were washed a few times with PBS to remove cell debris and incubated with fresh medium. The time of scratching wound was designated as time 0. Cells
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
the cytoplasm showing pancytoplasmic staining pattern. In tumor-derived cells, it shows perinuclear distribution (17). Interestingly, induction of senescence in cancer cells by drugs (MKT-077, BrdU, Withaferin A, H2O2, 5Aza-dC, and Epoximicin), by peptides, or by single chromosomes/chromosome fragments caused reallocation of pancytoplamic mortalin staining (18–22), suggesting that it could serve as an useful reporter to screen cancer targets. Furthermore, the shift in mortalin staining pattern was associated with abrogation of mortalin-p53 interaction, release of p53 from mortalin-p53 complex, and translocation to nucleus resulting in an activation of its transcriptional activation function (18,20,21). By using mortalin staining as an internal reporter to examine the induction of senescence in cancer cells, we screened human shRNA library and identified nine anticancer candidates (23). In the present study, we investigated these nine genes in the Comparative Genomic Hybridization (CGH) outcomes of 24 breast cancer cell lines and found that 5/9 gene targets (NBS1, BRCA1, TIN2, MRE11A, and KPNA2) were located on chromosomal regions that were frequently amplified in cancer cell lines. We demonstrate that these five genes are commonly amplified in cancer cells. Knockdown of their expression by shRNAs triggered DNA damage response and activation of tumor suppressor, p16INK4A, causing growth arrest and loss of malignant properties suggesting their potential as cancer targets.
Senescence and Reduced Migration of Cancer cells
were allowed to proliferate and migrate into the wound during the next 24 hours. Migration of cells into the wound was recorded under a phase contrast microscope with a 10× phase objective. Migration capacity was quantitated by measuring the percent of open area in 6–10 randomly captured images.
Immunostaining For immunostaining, cells were plated on glass coverslips placed in a 12-well plate (104 cells/coverslip) and processed as described (23). Antibodies used were monoclonal anti-mortalin (23), CARF (27), NBS1 (Y112), BRCA1 (MS13), KPNA2 (ab99288), TIN2 (ab64386; Abcam), pRB (Ser795), γH2AX (Ser139), MRE11 (Cell Signaling), MMP-2 (H-76), MMP-3 (AA07), p16 (F-12), p21 (C-19), and p53 (DO-1; Santa Cruz Biotech, CA) for 1 hour at room temperature, washed thrice with 0.2% Triton X-100 in PBS, and then incubated with Alexa-594-conjugated and Alexa488-conjugated (Molecular Probes, Eugene, OR) secondary antibodies. The stained cells were washed twice in PBS and then observed immediately on a Carl Zeiss microscope (Axiovert 200 M). Western Blotting Protein extraction was carried out using standard methods, and protein concentrations were determined by BCA assay (Pierce, Thermo Scientific). Lysates containing 20 μg protein were boiled in sodium dodecyl sulfate sample buffer for 5 min, subjected to 10% sodium dodecyl sulfate–polyacrylamide
gel electrophoresis, and then transferred onto a polyvinylidene difluoride membrane. Membranes were blocked in TBS-T containing 5% skim milk. After blocking, membranes were incubated with antibodies: BRCA1 (MS13), TIN2 (ab64386), NBS1 (Y112; ab32074; Abcam), MMP-9 (C20), MMP-3 (AA07), MMP-2 (H-76), p16 (C-20), (Santa Cruz Biotech, CA), Rad 50 (Cell Signaling), and CARF. Actin was used as an internal loading control (Chemicon International, CA). After probing with the primary antibody as indicated, the blots were incubated with horse radish peroxidase– conjugated secondary antibody (sc-2004 or sc-2005; Santa Cruz Biotech). Finally, the blots were developed using enhanced chemiluminescence (RPN2132; GE Healthcare, Buckinghamshire, UK). Senescence Assay SA-β-galactosidase activity was evaluated using the “Senescence-β-Gal Staining Kit” (Cell Signaling Technology) following the manufacturer’s instructions. Results and Discussion Identification of BRCA1, TIN2, NBS1, KPNA2, and MRE11A as Candidate Senescence-Inducing shRNAs We transfected human breast (MCF7) and bone (U2OS) cancer cells with expression vectors encoding 768 human shRNAs. Transfected cells were selected in the culture medium supplemented with puromycin (0.5 µg/mL) and examined for mortalin staining pattern under the automated scanning system attached to Axiovert 200M microscope (Zeiss). As shown in Figure 1, the shRNAs that caused shift in mortalin staining from perinuclear to pancytoplasmic in cancer cells were selected for the second round of screening. The process was repeated 4 times by transfecting the selected shRNAs from each round, and finally 22 shRNAs were identified (23) that resulted in induction of senescence. The phenomenon was previously demonstrated when senescence was induced in cancer cells with anticancer drugs, phytochemicals, peptides, and chromosome fragments (18–22,28). Interestingly, all the 22 identified gene targets have been predicted to be involved in human carcinogenesis (23). We previously investigated the effect of the 22 shRNAs on six other cancer cell lines (four breast carcinomas—SKBR3, T47D, MDA-MB231, MDA-MB157; gall bladder carcinoma—TGBC-2; biliary tract carcinoma— SKCHA-1) and found that 9 of the 22 shRNAs triggered the shift in mortalin staining pattern in all the six cell lines tested, suggesting that the gene targets of these 9 shRNAs may play an important or essential role during carcinogenesis (23). Furthermore, shift in mortalin staining pattern was observed in cell lines with various genetic background including cells with wild-type p53 and lack of p16INK4A function (U2OS), compromised p53 and lack of p16INK4A function (MCF7), mutant p53 and wild-type pRB (SKBR3,
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Cell Invasion Assay Invasion chamber (QCM ECMatrix Cell Invasion Assay, 24-well [8 μm], colorimetric) was brought to room temperature in a tissue culture hood. Serum-free medium (SFM, 300 μL, warmed at 37°C) was added to inserts and incubated for 2 hours at room temperature for rehydration of ECM layer. After rehydration, media were carefully removed from the inserts without disturbing the membrane. Cell suspension containing 0.5–1.0 × 106 cells/mL in serum-free media was prepared and 300 μL was added to each insert. Media (500 μL) containing 10% FBS was added to the lower chamber. The invasion chambers were incubated for 24–72 hours in CO2 incubator. Using cottontipped swabs, the ECMatrix gel and noninvading cells were gently removed from the interior of the inserts. In order to stain the invasive cells on lower surface of the membrane, the inserts were dipped in 500 μL of staining solution that was aliquoted in the unused well of the plate for 20 min. The stained inserts were rinsed in water and air-dried. Cells were quantitated by dissolving stained cells in 10% acetic acid (100 μL/well) and transferred to 96-well plate for colorimetric reading of optical density at 560 nm.
Page 3 of 13
Page 4 of 13
Gao et al.
T47D, MDA-MB 231), compromised p53 and wild-type pRB function (MCF7), and lack of both p53 and pRB (Saos-2, MDA-MB157, and TGBC-2), suggesting that the phenomenon is independent to the status of p53 and pRB, the two main tumor suppressor genes. Furthermore, these cell lines possessed different tumor-forming and malignant characteristics. Although some of these including U2OS, Saos-2, and MCF7 were nontumorigenic and nonmetastatic, the others including SKBR3, T47D, MDA-MB 231, and MDA-MB 157 were tumorigenic and metastatic, suggesting that the shift in mortalin staining pattern is independent of the malignant properties of the tumor cells. In order to further reveal the significance of the identified gene targets in cancer, we screened the outcomes of an independent CGH array analysis in 35 cancer cell lines (25,26) and found that the loci of five (BRCA1, TIN2, NBS1, KPNA2, and MRE11A) of the nine gene targets were located on the chromosomal regions that were identified as
gain of locus in cancer cells (Figure 1). We then performed genomic PCR with specific primers for the five selected genes on 22 cell lines and found their amplification in majority of them. As shown in Figure 1C, 19/22 cell lines showed amplification of BRCA1, 14/22 showed amplification of TIN2, 20/22 showed amplification of NBS1, 20/22 showed amplification of KPNA2, and 16/22 showed amplification of MRE11A gene loci. Furthermore, these proteins have been reported as upregulated in a variety of cancers and contribute to carcinogenesis in several studies (23). BRCA1, TIN2, NBS1, KPNA2, or MRE11A Silencing Affected Cancer Cell Migration We first investigated the induction of senescence by selected shRNAs in U2OS and MCF7 cells. The cells were transfected with specific shRNAs. Transfection efficacies were examined by cotransfection of a reporter gene (GFP in a
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 1. Identification of candidate anticancer shRNAs, based on the shift in mortalin staining pattern and CGH analysis in cancer cell lines. (A) Schematic presentation of shRNA library screening using mortalin staining as a reporter. Human breast carcinoma (MCF7) and osteosarcoma (U2OS) were used. In four rounds of screening, 22 shRNA were identified that were seen to shift the mortalin staining from perinuclear to pancytoplasmic type. Subsequent assays in six cancer cell lines (4 breast and 2 gall bladder) showed the shift in staining pattern in 9 out of 22. (B) CGH screening of 35 breast cancer cell lines revealed that five (BRCA1, TIN2, NBS1, KPNA2, and MRE11A) of nine genes were located on the chromosomal regions that are frequently amplified in cancer cells. (C) Expression analysis of five selected shRNAs target genes in human cancer cell lines. Gene-specific PCR was carried on 22 cancer cell lines and human normal genomic DNA. Quantitative analysis of gene-specific PCR showed their amplification in majority of the cell lines.
Senescence and Reduced Migration of Cancer cells
Page 5 of 13
ratio of 50:1/shRNA:GFP plasmid; Figure 2A). Knockdown effect of shRNAs was examined by immunostaining each target gene using specific antibodies (Figure 2B). We found that the transfected cells (both MCF7 and U2OS) showed decreased (about 70%) expression of the targeted genes in more than 60%–70% of cells. At 48–72 hours posttransfections, no immediate cell death was observed in response to knockdown of either of the five (BRCA1, TIN2, NBS1, KPNA2, and MRE11A) genes. The cells were examined for mortalin and senescence-associated β-gal staining. We found that the shift in mortalin staining pattern was associated with increase in β-gal staining in cells compromised of either of these five genes (Figure 2C, D, and E). Cell viability (determined by Vi-CELL Series Cell Viability Analyzers and Alamar Blue assay) and invasion (migration through Matrigel and wound scratch assay) assays revealed decrease in cell proliferation (20%–50%) and invasion through Matrigel (15%–35%) and migration to the wound (~2.5- to 3-fold), respectively (Figure 2F, G, and H). Based on these data, we subjected the transfected cells to quantitative in vitro real-time invasion assay using RTCA DP instrument (Roche Diagnostics GmbH, Germany). The data also
revealed that the knockdown of the selected genes caused decrease in their migration capacity (data not shown). Pathway Analysis of Gene Targets To get further insights into the functional mechanism of shRNA-induced senescence in cancer cells, we performed bioinformatics analysis on the selected genes (BRCA1, TIN2, NBS1, KPNA2, and MRE11A) by MetaCore Pathway and Data Mining (http://www.genego.com/genego_lp.php). The analysis revealed that besides their role in diverse pathways including apoptosis, telomere maintenance, immune response, and protein stability, the five genes were most closely connected to the DNA damage response signaling (Figure 3A). The latter spans a complex network of pathways, ranging from sensing of the DNA lesion, localization of the DNA repair proteins to the lesion, and DNA repair and growth arrest by network of phosphorylating key players as described subsequently. BRCA1, the breast- and ovarian cancer–specific tumor suppressor, is a multifunctional nuclear protein involved in the regulation of DNA repair, cell cycle, transcription, chromatin remodeling, and genome integrity. Inherited
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 1. Continued.
Page 6 of 13
Gao et al.
mutations or loss of BRCA1 confers a high risk for breast cancer development, whereas its altered expression occurs frequently in sporadic forms of cancer. It has been shown that BRCA1 is an autoregulated transcription factor that selectively titrates its levels to maintain genome integrity in response to genotoxic insults (29). MRE11A (Meiotic recombination 11) is an essential DNA double-strand break (DSB) repair protein. It has exonuclease and endonuclease activities and functions in homologous recombination and telomere length maintenance in cancer cells (30). NBS1 is a protein involved in Nijmegen breakage syndrome (NBS, a chromosomal instability syndrome) associated with cancer predisposition, radiosensitivity, microcephaly, and growth retardation. It is a component of the MRN (MRE11-RAD50-NBS1) complex, a central player associated with DSB repair (31). MRN is one of the first factors to be localized in the DNA lesion (hence also called lesion sensor), where it plays a structural role of tethering together the broken chromosomes and initiate a cascade of DNA repair signaling events (32) controlled by the ATM protein kinase, a major regulator of the DSB. NBS1 is phosphorylated by ATM in response to radiation,
gets transported to the nucleus by KPNA2, and controls the nuclear localization of the MRN complexes (33). It functions as a tumor suppressor in preserving genome integrity in the nucleus and acts as an oncogenic protein in the cytoplasm by PI3-kinase/AKT-activation pathway (33). MRN functions to target ATM and Tip60 acetyltransferase to DSBs (34). Overexpression of NBS1 has been shown to induce epithelial–mesenchymal transition, contribute to transformation, and serve as a marker of aggressive head and neck cancer (35). Depletion of NBS1 resulted in an inhibition of ALT-, but not telomerase-mediated, telomere maintenance (36). TRF1 interacting nuclear protein 2 (TIN2), essential for embryonic development and cell viability, is a negative regulator of telomere length elongation and is enriched in hepatocarcinoma (37). Karyopherinalpha2 (KPNA2) was found upregulated in a variety of cancers. It mediates nuclear localization of NBS1 complex in response to DNA damage (38–40). In light of the above known functions of these proteins, we examined whether shRNA-mediated knockdown of these genes evoked DNA damage response in cancer cells. shRNA-transfected cells were examined for DNA
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 2. Cell viability and migration analysis compromised for the selected gene targets. (A) Transfection efficacy of the shRNA plasmids as revealed by cotransfection of GFP reporter. (B) Knockdown efficacy of the shRNAs as revealed by immunostaining of cells with specific antibodies. (C) Shift in mortalin staining pattern from perinuclear (vector) to pancytoplasmic (shRNA-transfected cells) type. (D and E) Senescence-associated β-gal (SA-β-gal) staining in control (vector) and shRNA-transfected cells and their quantitation from three independent experiments. (F) Viability of MCF7 cells transfected with selected shRNAs was examined at approximately 48–72 hours posttransfection. Invasion index as examined by Matrigel invasion assays (G) and invasive capacity of cells in wound scratch assays (H). *P < .05; **P < .01; ***P < .001.
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 2. Continued.
Page 7 of 13 Senescence and Reduced Migration of Cancer cells
Page 8 of 13
Gao et al.
damage response proteins, γH2AX, CARF, and phosphoCHK2 (pCHK2) (41,42), and their downstream regulators. As shown in Figure 3B, immunostaining assay in MCF7 cells transfected with shRNA for BRCA1, TIN2, NBS1, KPNA2, or MRE11A showed increase in γH2AX, CARF, and pCHK2 that signified DNA damage response. We examined the status of p53, p21, and p16INK4A in control and transfected cells and found that the shRNAs did not cause nuclear translocation of p53. The expression of p21 also remained unchanged in three of the four cell lines. On the other hand, the level of p16INK4A was found to be consistently enhanced when BRCA1, NBS1, or MRE11A were knockdown in all the four cell lines in several independent experiments (Figure 3C). Induction of p16INK4A level was also confirmed by Western blotting of control and shRNA-transfected MCF7 and U2OS cells (Figure 3D–F). p16INK4A is known to be silenced by promoter methylation in these cells. The present data strongly suggested that the
knockdown of target genes may cause demethylation of the p16INK4A promoter. Based on these results, it was suggestive that besides their role in DNA damage and repair, the target genes might be associated with histone methylation. Although conventionally, histone demethylases are believed to regulate DNA methylation and gene transcription, several recent studies have shown a cross talk between the DNA damage/growth arrest and DNA demethylation. These include role of histone demethylase LSD1 in DNA damage response (43) and role of growth arrest and DNA damage–induced protein, GADD45 alpha, in DNA demethylation (44,45). PPARγ was also shown to promote DNA demethylation (46). In view of these recent reports and our data, it is suggestive that the knockdown of DNA damage response proteins (NBS1, BRCA1, TIN2, MRE11A, and KPNA2) may cause DNA demethylation as supported by expression of p16INK4A. The mechanism of such induction of DNA demethylation warrants further studies.
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 3. Selected shRNAs evoke DNA damage response in cancer cells. (A) Bioinformatics analysis on the selected genes (NBS1, BRCA1, TIN2, MRE11A, and KPNA2) by MetaCore Pathway and Data Mining showed that they are closely connected to the DNA damage response pathway. (B) MCF7 cell transfected with BRCA1, TIN2, NBS1, KPNA2, or MRE11A showed increase in γH2AX, CARF (Collaborator of ARF), and pCHK2. (C) Selected candidate shRNAs induced senescence in cancer cells through p16INK4A pathway. Selected candidate shRNAs were transfected into four cell lines (SKBR3, MCF7, U2OS, and TGBC-2) and examined for p53, p21WAF1, and p16INK4A by immunostaining. Representative staining for mortalin, p53, p21WAF1, and p16INK4A is shown in control and transfected cells. Results in four cancer cell lines are summarized showing consistent induction of p16INK4A. (D) Induction of p16INK4A in shRNA-transfected cells was confirmed by Western blotting. Quantitation of induction of p16INK4A expression from four independent experiments (E). (F) Immunostaining of p16INK4A in control and shRNA-transfected MCF7 and U2OS cells. *P < .05.
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 3. Continued.
Page 9 of 13 Senescence and Reduced Migration of Cancer cells
Page 10 of 13
Gao et al.
increased β-gal staining (Figure 4A). Cell migration analyses by wound scratch assay revealed that the p16INK4A overexpression also caused decrease in migration compared to the control cells (Figure 4C and D). Although the role of p16INK4A in growth arrest has been well documented, its contribution to cell migration remains unclear. Therefore, we examined the expression of cell migration markers in p16INK4A overexpressing cells. As shown in Figure 4E, we found that these cells have decrease in phosphorylation of pRB and increase in CARF (27) signifying the growth arrest. At the same time regulators of cell migration such as MMP-2 and -3 were downregulated (Figure 4E and F). Interestingly, four (MRE11A, NBS1, BRCA1, and TIN2) target genes showed decrease in p16INK4A overexpressing cells indicating a feedback regulation between these proteins and reflecting molecular functional economy in which the cells under growth arrest downregulate the DNA damage signaling proteins. In other words, p16INK4A acted as an effector protein of DNA damage response initiated by MRN complex proteins and triggered growth arrest
Figure 4. Overexpression of p16INK4A caused growth arrest and loss of migration in cancer cells. (A) p16INK4A overexpressing cells showed growth arrest in colony forming assay. Both colony number and size were reduced, and cells showed SA-β-gal staining (B). These cells exhibited loss of cell migration as shown by wound scratch (C) and Matrigel assays (D). (E) The p16INK4A overexpressing cells showed cell cycle arrest as marked by decrease in phospho-RB. Loss of migration was marked by decrease in MMP-2 and MMP-3 expression. There was a decrease in NBS1, BRCA1, TIN2, MRE11, and RAD50. (F) Immunostaining of control and p16INK4A overexpressing cells for the selected gene targets, matrix metalloproteases, and phospho-RB is shown. *P < .05; **P < .01.
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Loss of Migration in shRNA-Transfected Cancer Cells was Mediated by Induction of p16INK4A As shown earlier, the knockdown of BRCA1, KPNA2, MRE11A, NBS1, or TIN2 triggered growth arrest and loss of migration in cancer cells by DNA damage response mediated by induction of p16INK4A in four cancer cell lines. The molecular mechanism of induction of p16INK4A by knockdown of five shRNAs in different cell lines remains to be resolved in an independent study. It is suggestive that the compromised expression of DNA damage and repair signaling proteins served as a DNA damage signal, triggered the formation of DNA damage foci (increase in γH2AX), and caused growth arrest (increase in CARF, p21WAF1, and p16INK4A). We verified the effect of p16INK4A on cell growth and migration by an alternative approach. Cells were stably transfected with p16INK4A expression vector. Cell proliferation assays including colony forming efficacy, colony size, and viability revealed significant growth arrest in p16INK4A overexpressing cells (Figure 4A and B). Furthermore, p16INK4A overexpressing cells showed
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 4. Continued.
Page 11 of 13 Senescence and Reduced Migration of Cancer cells
Page 12 of 13
Gao et al.
in cells. Growth arrested p16INK4A overexpressing cells, on the other hand, downregulated the DNA damage response genes. The DNA DSBs initiate DNA damage response that in turn activates relocalization of several proteins like MRE11A, NBS1, and BRCA1 to nuclear foci that colocalize with γH2AX and result in foci formation (47). It has been reported that γH2AX interacts directly with BRCA1 and NBS1, required for formation of MRN complex at DNA damage sites (30,32,47–49). Furthermore, KPNA2 was shown to be essential for nuclear localization of NBS1 (50). Identification of multiple components of this nuclear focus formation complex in our screening revealed that the DNA damage signaling is a strong candidate target for cancer therapy. Disruption of any element of this complex triggered senescence response in cancer cells as also marked by shift in staining pattern of mortalin, increase in p16INK4A and CARF expression, and decrease in phosphorylation of pRB and matrix metalloproteases (MMP-2 and -3). Changes in mortalin, CARF, p16Ink4A, and pRB signified growth arrest of cancer cells, whereas change in matrix metalloproteases (established regulators of tumor environment) revealed decrease in their migration and invasion capacity as has been reported in several other studies (51). Knockdown of either of the five targets appeared to cause induction of p16INK4A that inhibited cyclin/CDK complexes, resulting in inhibition of pRB phosphorylation and cell cycle. Increase in CARF was previously shown to activate p53-p21WAF1 tumor suppression pathway causing growth arrest of cancer cells (42). Taken together, knockdown of the identified gene targets caused activation of two major tumor suppressor pathways, p16INK4A-pRB and p53-p21WAF1 (Figure 5). Downregulation of p16INK4A occurs frequently with human cell transformation (52). It has also been correlated with the development of invasive characteristics of cancer cells (53). Introduction of p16INK4A was shown to result in decreased invasion and angiogenesis of human glioma (54,55). Consistent with these reports, we found that the overexpression of p16INK4A caused decrease in migration of both U2OS and MCF7 cells. Furthermore, p16INK4A overexpressing cells exhibited decreased expression of MMP-2 and MMP-3 (Figures 4 and 5) that may account for decreased invasion capacity of cells. Taken together, we report an outcome of a
novel endogenous reporter (mortalin staining) based screening for cancer gene targets that revealed DNA damage signaling pathway as a strong candidate target for induction of senescence and compromise the migration capacity of cancer cells. Funding This study was partly supported by grants from the New Energy and Industrial Technology Development Organization (NEDO) of Japan. Acknowledgment Authors thank Drs. Saito and T. Hirano for CGH analysis and T. Yaguchi for assistance. R.S. was a receipient of the Ministry of Education, Culture, Sports, Science and Technology (MEXT) scholarship. References 1. Hayflick L, Moorhead PS. The serial cultivation of human diploid cell strains. Exp Cell Res. 1961;25:585–621. 2. Campisi J. Aging and cancer: the double-edged sword of replicative senescence. J Am Geriatr Soc. 1997;45:482–488. 3. Reddel RR. The role of senescence and immortalization in carcinogenesis. Carcinogenesis. 2000;21:477–484. 4. Prieur A, Peeper DS. Cellular senescence in vivo: a barrier to tumorigenesis. Curr Opin Cell Biol. 2008;20:150–155. doi:10.1016/j. ceb.2008.01.007 5. Duncan EL, Wadhwa R, Kaul SC. Senescence and immortalization of human cells. Biogerontology. 2000;1:103–121. 6. Campisi J. Between Scylla and Charybdis: p53 links tumor suppression and aging. Mech Ageing Dev. 2002;123:567–573. 7. Campisi J. Senescent cells, tumor suppression, and organismal aging: good citizens, bad neighbors. Cell. 2005;120:513–522. 8. Lleonart ME, Artero-Castro A, Kondoh H. Senescence induction; a possible cancer therapy. Mol Cancer. 2009;8:3. doi:10.1186/ 1476-4598-8-3 9. Kaul SC, Deocaris CC, Wadhwa R. Three faces of mortalin: a housekeeper, guardian and killer. Exp Gerontol. 2007;42:263–274. 10. Wadhwa R, Takano S, Robert M, et al. Inactivation of tumor suppressor p53 by mot-2, a hsp70 family member. J Biol Chem. 1998;273:29586–29591. 11. Kaul SC, Aida S, Yaguchi T, Kaur K, Wadhwa R. Activation of wild type p53 function by its mortalin-binding, cytoplasmically localizing carboxyl terminus peptides. J Biol Chem. 2005;280:39373–39379. 12. Ma Z, Izumi H, Kanai M, Kabuyama Y, Ahn NG, Fukasawa K. Mortalin controls centrosome duplication via modulating centrosomal localization of p53. Oncogene. 2006;25:5377–5390. 13. Wadhwa R, Takano S, Kaur K, et al. Upregulation of mortalin/ mthsp70/Grp75 contributes to human carcinogenesis. Int J Cancer. 2006;118:2973–2980.
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
Figure 5. A model showing shift in mortalin staining pattern (gray) from the perinuclear to the pancytoplasmic type in cancer cells compromised for BRCA1, KPNA2, NBS1, MRE11, and TIN2 leading to the upregulation of genes involved in growth arrest and downregulation of genes involved in cell migration.
Senescence and Reduced Migration of Cancer cells
36. Zhong ZH, Jiang WQ, Cesare AJ, Neumann AA, Wadhwa R, Reddel RR. Disruption of telomere maintenance by depletion of the MRE11/ RAD50/NBS1 complex in cells that use alternative lengthening of telomeres. J Biol Chem. 2007;282:29314–29322. 37. Oh BK, Kim YJ, Park C, Park YN. Up-regulation of telomere-binding proteins, TRF1, TRF2, and TIN2 is related to telomere shortening during human multistep hepatocarcinogenesis. Am J Pathol. 2005;166:73–80. 38. Jensen JB, Munksgaard PP, Sørensen CM, et al. High expression of karyopherin-α2 defines poor prognosis in non-muscle-invasive bladder cancer and in patients with invasive bladder cancer undergoing radical cystectomy. Eur Urol. 2011;59:841–848. doi:10.1016/j. eururo.2011.01.048 39. Zheng M, Tang L, Huang L, et al. Overexpression of karyo pherin-2 in epithelial ovarian cancer and correlation with poor prognosis. Obstet Gynecol. 2010;116:884–891. doi:10.1097/ AOG.0b013e3181f104ce 40. Wang CI, Wang CL, Wang CW, et al. Importin subunit alpha-2 is identified as a potential biomarker for non-small cell lung cancer by integration of the cancer cell secretome and tissue transcriptome. Int J Cancer. 2011;128:2364–2372. doi:10.1002/ijc.25568 41. Dickey JS, Redon CE, Nakamura AJ, Baird BJ, Sedelnikova OA, Bonner WM. H2AX: functional roles and potential applications. Chromosoma. 2009;118:683–692. doi:10.1007/s00412-009-0234-4 42. Cheung CT, Singh R, Yoon AR, et al. Molecular characterization of apoptosis induced by CARF silencing in human cancer cells. Cell Death Differ. 2011;18:589–601. doi:10.1038/cdd.2010.129 43. Mosammaparast N, Kim H, Laurent B, et al. The histone demethylase LSD1/KDM1A promotes the DNA damage response. J Cell Biol. 2013;203:457–470. doi:10.1083/jcb.201302092 44. Schäfer A. Gadd45 proteins: key players of repair-mediated DNA demethylation. Adv Exp Med Biol. 2013;793:35–50. doi:10.1007/978-1-4614-8289-5_3 45. Li Y, Zhao M, Yin H, et al. Overexpression of the growth arrest and DNA damage-induced 45alpha gene contributes to autoimmunity by promoting DNA demethylation in lupus T cells. Arthritis Rheum. 2010;62:1438–1447. doi:10.1002/art.27363 46. Fujiki K, Shinoda A, Kano F, Sato R, Shirahige K, Murata M. PPARγinduced PARylation promotes local DNA demethylation by production of 5-hydroxymethylcytosine. Nat Commun. 2013;4:2262–2265. doi:10.1038/ncomms3262 47. Paull TT, Rogakou EP, Yamazaki V, Kirchgessner CU, Gellert M, Bonner WM. A critical role for histone H2AX in recruitment of repair factors to nuclear foci after DNA damage. Curr Biol. 2000;10:886–895. 48. Huen MS, Sy SM, Chen J. BRCA1 and its toolbox for the maintenance of genome integrity. Nat Rev Mol Cell Biol. 2010;11:138–148. doi:10.1038/nrm2831 49. Rodier F, Coppé JP, Patil CK, et al. Persistent DNA damage signalling triggers senescence-associated inflammatory cytokine secretion. Nat Cell Biol. 2009;11:973–979. doi:10.1038/ncb1909 50. Tseng SF, Chang CY, Wu KJ, Teng SC. Importin KPNA2 is required for proper nuclear localization and multiple functions of NBS1. J Biol Chem. 2005;280:39594–39600. 51. Kessenbrock K, Plaks V, Werb Z. Matrix metalloproteinases: regulators of the tumor microenvironment. Cell. 2010;141:52–67. doi:10.1016/j.cell.2010.03.015 52. Serrano M, Lee H, Chin L, Cordon-Cardo C, Beach D, DePinho RA. Role of the INK4a locus in tumor suppression and cell mortality. Cell. 1996;85:27–37. 53. Grover R, Chana JS, Wilson GD, Richman PI, Sanders R. An analysis of p16 protein expression in sporadic malignant melanoma. Melanoma Res. 1998;8:267–272. 54. Chintala SK, Fueyo J, Gomez-Manzano C, et al. Adenovirus-mediated p16/CDKN2 gene transfer suppresses glioma invasion in vitro. Oncogene. 1997;15:2049–2057. 55. Harada H, Nakagawa K, Iwata S, et al. Restoration of wild-type p16 down-regulates vascular endothelial growth factor expression and inhibits angiogenesis in human gliomas. Cancer Res. 1999;59:3783–3789.
Downloaded from http://biomedgerontology.oxfordjournals.org/ at University of Sydney on September 1, 2014
14. Dundas SR, Lawrie LC, Rooney PH, Murray GI. Mortalin is overexpressed by colorectal adenocarcinomas and correlates with poor survival. J Pathol. 2005;205:74–81. 15. Yi X, Luk JM, Lee NP, et al. Association of mortalin (HSPA9) with liver cancer metastasis and prediction for early tumor recurrence. Mol Cell Proteomics. 2008;7:315–325. 16. Lu WJ, Lee NP, Kaul SC, et al. Mortalin-p53 interaction in cancer cells is stress dependent and constitutes a selective target for cancer therapy. Cell Death Differ. 2011;18:1046–1056. doi:10.1038/cdd.2010.177 17. Wadhwa R, Kaul SC, Mitsui Y, Sugimoto Y. Differential subcellular distribution of mortalin in mortal and immortal mouse and human fibroblasts. Exp Cell Res. 1993;207:442–448. 18. Wadhwa R, Sugihara T, Yoshida A, et al. Selective toxicity of MKT-077 to cancer cells is mediated by its binding to the hsp70 family protein mot-2 and reactivation of p53 function. Cancer Res. 2000;60:6818–6821. 19. Michishita E, Nakabayashi K, Suzuki T, et al. 5-Bromodeoxyuridine induces senescence-like phenomena in mammalian cells regardless of cell type or species. J Biochem. 1999;126:1052–1059. 20. Widodo N, Deocaris CC, Kaur K, et al. Stress chaperones, mortalin, and pex19p mediate 5-aza-2’ deoxycytidine-induced senescence of cancer cells by DNA methylation-independent pathway. J Gerontol A Biol Sci Med Sci. 2007;62:246–255. 21. Widodo N, Takagi Y, Shrestha BG, Ishii T, Kaul SC, Wadhwa R. Selective killing of cancer cells by leaf extract of Ashwagandha: components, activity and pathway analyses. Cancer Lett. 2008;262:37–47. doi:10.1016/j.canlet.2007.11.037 22. Kaul Z, Yaguchi T, Chiura HX, Kaul SC, Wadhwa R. Quantum dotbased mortalin staining as a visual assay for detection of induced senescence in cancer cells. Ann N Y Acad Sci. 2007;1100:368–372. 23. Gao R, Yaguchi T, Horimoto K, et al. Identification of anti-cancer shRNAs based on the staining pattern of mortalin. Tiss Cult Res Commun. 2010;29:147–153. 24. Wadhwa R, Kaul SC, Ikawa Y, Sugimoto Y. Identification of a novel member of mouse hsp70 family. Its association with cellular mortal phenotype. J Biol Chem. 1993;268:6615–6621. 25. Saito S, Morita K, Hirano T. High frequency of common DNA copy number abnormalities detected by bacterial artificial chromosome array comparative genomic hybridization in 24 breast cancer cell lines. Hum Cell. 2009;22:1–10. doi:10.1111/j.1749-0774.2008.00061.x 26. Saito S, Ghosh M, Morita K, Hirano T, Miwa M, Todoroki T. The genetic differences between gallbladder and bile duct cancer cell lines. Oncol Rep. 2006;16:949–956. 27. Hasan K, Cheung C, Kaul Z, et al. CARF Is a vital dual regulator of cellular senescence and apoptosis. J Biol Chem. 2009;284:1664–1672. doi:10.1074/jbc.M805778200 28. Bertram MJ, Bérubé NG, Hang-Swanson X, et al. Identification of a gene that reverses the immortal phenotype of a subset of cells and is a member of a novel family of transcription factor-like genes. Mol Cell Biol. 1999;19:1479–1485. 29. De Siervi A, De Luca P, Byun JS, et al. Transcriptional autoregulation by BRCA1. Cancer Res. 2010;70:532–542. doi:10.1158/0008-5472. CAN-09-1477 30. Deng R, Tang J, Ma JG, et al. PKB/Akt promotes DSB repair in cancer cells through upregulating Mre11 expression following ionizing radiation. Oncogene. 2011;30:944–955. doi:10.1038/onc.2010.467 31. Kobayashi J, Antoccia A, Tauchi H, Matsuura S, Komatsu K. NBS1 and its functional role in the DNA damage response. DNA Repair (Amst). 2004;3:855–861. 32. van den Bosch M, Bree RT, Lowndes NF. The MRN complex: coordinating and mediating the response to broken chromosomes. EMBO Rep. 2003;4:844–849. 33. Teng SC, Wu KJ, Tseng SF, Wong CW, Kao L. Importin KPNA2, NBS1, DNA repair and tumorigenesis. J Mol Histol. 2006;37:293–299. 34. Sun Y, Jiang X, Price BD. Tip60: connecting chromatin to DNA damage signaling. Cell Cycle. 2010;9:930–936. 35. Yang MH, Chang SY, Chiou SH, et al. Overexpression of NBS1 induces epithelial-mesenchymal transition and co-expression of NBS1 and Snail predicts metastasis of head and neck cancer. Oncogene. 2007;26:1459–1467.
Page 13 of 13
