Represses Human Telomerase Reverse Transcriptase (Htert)

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

Biochimica et Biophysica Acta 1843 (2014) 565–579

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

journal homepage: www.elsevier.com/locate/bbamcr

The 58-kDa microspherule protein (MSP58) represses human telomerase reverse transcriptase (hTERT) gene expression and cell proliferation by interacting with telomerase transcriptional element-interacting factor (TEIF)

Che-Chia Hsu a, Chang-Han Chen h, Tsung-I Hsu a,c,Jan-JongHunga,b,c,d, Jiunn-Liang Ko i,BoZhangj, Yi-Chao Lee f,g, Han-Ku Chen k, Wen-Chang Chang b,c,e,f,⁎,Ding-YenLina,b,c,d,⁎⁎

a Institute of Bioinformatics and Biosignal Transduction, College of Bioscience and Biotechnology, National Cheng Kung University, Tainan 70101, Taiwan, ROC b Department of Pharmacology, College of Medicine, National Cheng Kung University, Tainan 70101, Taiwan, ROC c Infectious Diseases and Signaling Research Center, National Cheng Kung University, Tainan 70101, Taiwan, ROC d Institute for Cancer Biology and Drug Discovery, College of Medical Science and Technology, Taipei Medical University, Taipei 11031, Taiwan, ROC e Graduate Institute of Medical Sciences, College of Medicine, Taipei Medical University, Taipei 11031, Taiwan, ROC f Center for Neurotrauma and Neuroregeneration, Taipei Medical University, Taipei 11031, Taiwan, ROC g Program for Neural Regenerative Medicine, College of Medical Science and Technology, Taipei Medical University, Taipei 11031, Taiwan, ROC h Center for Translational Research in Biomedical Sciences, Kaohsiung Chang Gung Memorial Hospital, Kaohsiung 83301, Taiwan, ROC i Institute of Medicine, Chung Shan Medical University, Taichung 40201, Taiwan, ROC j Department of Pathology, Health Science Center of Peking University, Beijing 10091, China k Department of Pathology, Yuan's General Hospital, Kaohsiung 80249, Taiwan, ROC

article info abstract

Article history: 58-kDa microspherule protein (MSP58) plays an important role in a variety of cellular processes including tran- Received 17 July 2013 scriptional regulation, cell proliferation and oncogenic transformation. Currently, the mechanisms underlying the Received in revised form 13 November 2013 oncogenic effect of MSP58 are not fully understood. The human telomerase reverse transcriptase (hTERT)gene, Accepted 4 December 2013 which encodes an essential component for telomerase activity that is involved in cellular immortalization and Available online 18 December 2013 transformation, is strictly regulated at the gene transcription level. Our previous study revealed a novel function Keywords: of MSP58 in cellular senescence. Here we identify telomerase transcriptional element-interacting factor (TEIF) as 58-kDa microspherule protein a novel MSP58-interacting protein and determine the effect of MSP58 on hTERT transcription. This study thus Telomerase provides evidence showing MSP58 to be a negative regulator of hTERT expression and telomerase activity. Human telomerase reverse transcriptase Luciferase reporter assays indicated that MSP58 could suppress the transcription of hTERT promoter. Additionally, Telomerase transcriptional element-interacting stable overexpression of MSP58 protein in HT1080 and 293T cells decreased both endogenous hTERT expression factor and telomerase activity. Conversely, their upregulation was induced by MSP58 silencing. Chromatin immunoprecipitation assays showed that MSP58 binds to the hTERT proximal promoter. Furthermore, overexpression of MSP58 inhibited TEIF-mediated hTERT transactivation, telomerase activation, and cell proliferation promotion. The inhibitory effect of MSP58 occurred through inhibition of TEIF binding to DNA. Ultimately, the HT1080- implanted xenograft mouse model conﬁrmed these cellular effects. Together, our ﬁndings provide new insights into both the biological function of MSP58 and the regulation of telomerase/hTERT expression. © 2013 Elsevier B.V. All rights reserved.

Abbreviations: MSP58, 58-kDa microspherule protein; hTERT, human telomerase reverse transcriptase; TEIF, Telomerase transcriptional element-interacting factor; USF1, upstream stimulatory factor 1; TCF-4, T cell transcription factor 4; PI3K, phosphoinositide 3-kinase; NFAT, nuclear factor of activated T cell; HMGA2, high mobility group A2; Bmi-1, B-lymphoma Moloney murine leukemia virus insertion region-1; WT1, Wilms tumor 1; Mad1, max dimer protein 1; MZF-2, myeloid zinc-ﬁnger protein 2; NFX1, nuclear factor binds to the X1 box; Smad3, SMAD family member 3; menin (MEN1), multiple endocrine neoplasia type 1; IRF-4, interferon regulatory factor 4; IRF8, interferon regulatory factor 8; UBF, upstream binding factor; qPCR, quantitative PCR; MOF, males absent on the ﬁrst; NuRD, nucleosome remodeling and deacetylase; Rb, retinoblastoma protein; Nde1, nudE nuclear distribution gene E homolog 1; STRA13, stimulated by retinoic acid 13; TOJ3, target of Jun 3 ⁎ Correspondence to: W.-C. Chang, Graduate Institute of Medical Sciences, College of Medicine, and Center for Neurotrauma and Neuroregeneration, Taipei Medical University, Taipei 11031, Taiwan, ROC. ⁎⁎ Correspondence to: D.-Y. Lin, Institute of Bioinformatics and Biosignal Transduction, College of Bioscience and Biotechnology; National Cheng Kung University, Tainan 70101, Taiwan, ROC. Tel.: +886 6 2757575x31207; fax: +886 6 2083663. E-mail addresses: [email protected] (W.-C. Chang), [email protected] (D.-Y. Lin).

1. Introduction as a centrosomal protein that is required for maintaining centrosome homeostasis [49,50]. Importantly, a number of studies have indicated Telomerase is a ribonucleoprotein complex with reverse-transcriptase that MSP58 plays a key role in oncogenesis. For instance, MSP58 and enzymatic activity which is responsible for adding hexameric repeats TOJ3 (a quail homologue of MSP58) behave as oncogenes in fibroblast (TTAGGG) to telomeric ends of linear chromosomes [1]. The human transformation assays, whereas the tumor suppressor, phosphatase core telomerase enzyme consists of a catalytic subunit (human telome- and tensin homologue (PTEN), suppresses the transforming activity of rase reverse transcriptase; hTERT) which bears the enzymatic activity of MSP58 [51,52]. telomerase and an RNA moiety (human telomerase RNA; hTR) that is In a previous study, we demonstrated that MSP58 induces cellular used as a template for telomeric repeat synthesis [2]. Maintenance of senescence through the p53/p21 pathway [48]. Identifying potential telomere lengths by telomerase is required for cells to be able to prolif- novel binding partner(s) of MSP58 may help elucidate the role it plays erate indefinitely and to escape from replicative senescence. In adult in controlling cell proliferation; to do this, we performed yeast two- humans, telomerase is present in human stem cells and germ cells but hybrid screening using MSP58 as bait. We identified a novel MSP58- is barely detectable in the vast majority of differentiated somatic cells interacting protein, telomerase transcriptional element-interacting [3]. However, telomerase can be reactivated in most cancers and not factor (TEIF), that was found to bind to the hTERT promoter and acti- only stabilize their telomere size but also directly regulate multiple vates its transcription and telomerase activities [53]. Notably, TEIF over- cancer-promoting pathways [4]. Activation of telomerase is a critical expression has been detected in tumor tissues and cell lines, parallel to step during cellular immortalization and malignant transformation [5]. hTERT [53,54]. In this study, we examined the potential contribution of Notably, telomerase or hTERT reactivation is detected in up to 90% MSP58 to the transcriptional regulation of hTERT. Our data revealed of human cancers [6]. Moreover, it has been shown that ectopic that overexpression of MSP58 repressed hTERT transcription and expression of hTERT can restore telomerase activity in many telomerase- telomerase activity, in part, through interaction with TEIF. Our findings negative cell types, thereby facilitating immortalization [7,8]. Further- reveal MSP58 to be a negative regulator of hTERT expression and telo- more, introduction of a dominant negative form of hTERT or an anti- merase activity. sense vector targeted against hTERT inhibits telomerase activity in human cancer cells and eventually induces apoptosis or differentiation 2. Materials and methods [9]. Therefore, hTERT expression is defined as the rate-limiting factor in regulating telomerase activity and is a hallmark of cancer. 2.1. Plasmids and antibodies Numerous studies suggest that telomerase enzymatic activity and expressions of telomerase components are regulated at multiple levels, Yeast constructs expressing LexA-MSP58 and pGal-AD-TEIF, a mam- including transcription and post-transcription, and by chaperone- malian vector expressing Flag-MSP58 and EGFP-MSP58, a pQCXIP-GFP- mediated folding and proper localization; however, the major control MSP58 retroviral vector, and a bacterial vector expressing GST-MSP58 mechanism of telomerase regulation seems to be at the level of hTERT were as described previously [40,48,53]. MSP58 deletion mutants for transcription [10]. Deletion analyses in reporter assays showed that expressing LexA fusion in yeast and Flag-tagged fusion in mammalian the approximately 200-bp proximal region of the hTERT promoter is im- cells were generated by inserting polymerase chain reaction (PCR) portant for maintaining basal transcriptional activity [11,12].ThehTERT fragments respectively coding MSP58 amino acids 1–300 and 300– promoter contains several putative transcription factor-binding sites, 462 into the pBTM116 and pCMV-Tag2 (Stratagene, La Jolla, CA, USA) including c-Myc [13–15],Sp1[11,12],E2F[16],USF[17], hypoxia- vectors. The full-length TEIF (amino acids 1–786) was cloned into the inducible factor (HIF)-1 [18], and TCF4 [19]. Additional studies have pcDNA3.1-HA expression vector (Invitrogen) to generate HA-TEIF. The demonstrated that PI3K [20],NFAT[21],HMGA2[22],theBmi-1onco- TEIF and its deletion mutants for expressing Gal4 activation domain fu- gene [23], and the papillomavirus E6 protein [24] can also stimulate sion in yeast and HA-tagged fusion in mammalian cells were generated hTERT promoter expression. In contrast, WT1 [25], Mad1 [26], MZF-2 by inserting PCR fragments respectively coding the full-length TEIF, and [27], NFX1 [28], Smad3 [29],menin[30], and p53 [31,32] can negatively amino acids 1–395 and 396–786 into the pACT2 (BD Biosciences regulate hTERT promoter expression. Distinct repression and activation Clontech) and pcDNA3.1-HA vectors. Human Daxx complementary (c) mechanisms likely operate in different cell contexts. For example, estro- DNA was purchased from GeneCopoeia (Germantown, MD, USA). Full- gen upregulates telomerase in both mammary and ovarian epithelial length Daxx was cloned into the pCDNA3.1-HA-expressing vector cells [33,34]; IRF-4 and IRF-8, two lymphoid cell-specific transcription (Invitrogen) to generate HA-Daxx. hTERT retrovirus vector pLPC- factors, activate hTERT transcription in immune cells [35]; and E2F-1 hTERT (Clontech laboratories, CA, USA) was obtained from Dr. Yu-Ten has opposing actions toward hTERT gene expression in human tumors Ju (National Taiwan University, Taipei, Taiwan). Full-length hTERT and normal somatic cells [16,36]. Epigenetic regulatory mechanisms, was cloned into the pCMV-Tag2 vector (Stratagene) to generate Flag- such as promoter DNA methylation [37] and histone acetylation [38], hTERT. The hTERT promoter deletion mutants p548 (−548 to +50), also modulate hTERT expression. p212 (−212 to +50), p196 (−196 to +50), p177 (−177 to +50), The 58-kDa microspherule protein (MSP58, also known as MCRS1) and p95 (−95 to +50) were cloned upstream of the firefly luciferase was originally identified as the interaction partner of the p120 nucleolar reporter in the pGL3-Basic vector (Promega, Madison, WI, USA) follow- protein [39]. Further studies showed that in nuclei and nucleoli, MSP58 ing a previously described protocol [55]. The luciferase reporter plas- functions to regulate transcription through its interactions with the mid, TERTLuc800, as previously described [15],wasagiftfromDr. transcription factors Daxx, STRA13, and UBF [40–42]. Interestingly, Kou-Juey Wu (National Yang-Ming University, Taipei, Taiwan). Short- microspherule protein 2 (MCRS2) was identified as an interacting hairpin (sh)RNA targeting MSP58 was expressed from the pLKO.1 hair- partner of a potent inhibitor of telomerase, LPTS/PinX1, and hTERT pin vector, which harbors an expression cassette for a puromycin resis- [43]. MCRS2 inhibits telomerase activity in vitro and long-term overex- tance gene. shMSP58 sequences used were shMSP58 1: 5′-CCTGCGAT pression of MCRS2 in cancer cell lines leads to telomere shortening. TCGTCTTCCTTAT-3′ and shMSP58 2: 5′-CAACAACTCTGTGGTGGAGAT- Drosophila MCRS2 is co-purified with RNA polymerase II complexes 3′. pLKO.1-shLuc (TRCN0000072243; shLuc) was used as a control. and is required for normal levels of cyclin gene expression [44]. Notably, These vectors were obtained from The RNAi Consortium (TRC) library MSP58 and MCRS2 have been shown to be associated with several (National RNAi Core Facility, Academia Sinica, Taipei, Taiwan). The chromatin-remodeling complexes including Mi-2β (a component of pSUPER vector expressing small-interfering (si)RNA of human MSP58 the NuRD complex), INO80, Brg1 (a subunit of the SWI/SNF complex), was constructed according to instructions from OligoEngine (Seattle, and histone acetyltransferase males absent on the first (MOF) WA, USA). The 19-nucleotide target sequences for siMSP58 were si-3 [42,45–48]. Furthermore, p78, an isoform of MSP58, has been identified (5′-GAAGAAGAAGGTATCCAAA-3′) and si-5 (5′-CAAGGTGTCATCAAGC C.-C. Hsu et al. / Biochimica et Biophysica Acta 1843 (2014) 565–579 567

TGA-3′). All plasmids were veriﬁed by restriction enzyme digestion and 2.6. Yeast two-hybrid screen and β-galactosidase assay DNA sequencing analyses. The rabbit MSP58 antibody was described previously [48]. The following commercial antibodies were used in The LexA-MSP58 full-length bait construct was used to screen this study: Flag (M2; Sigma, St. Louis, MO, USA), HA (HA.11; Babco/ against a human testis cDNA library (Clontech). Yeast two-hybrid Covance, Berkley, CA, USA), TEIF (653A; Bethyl Laboratories, Montgom- screening and analysis were performed as described previously [40]. ery, TX, USA). hTERT (Y182; GeneTex, San Antonio, TX, USA), Daxx (D7810; Sigma), and actin (clone AC-74; Sigma). 2.7. Cell-free transcription and translation (TNT) and GST pull-down assays

The TEIF, Daxx, hTERT, MSP58 and its deletion mutants were 2.2. Cell culture, transfection, and viral production and infection expressed in vitro using plasmids containing a T7 or SP6 promoter in a rabbit reticulocyte lysate system (Promega). The expression and puri- 293T cells were cultured in Dulbecco's modiﬁed Eagle's medium ﬁcation of GST and GST-MSP58 fusion proteins were performed as (DMEM; Invitrogen) supplemented with 10% fetal bovine serum described previously [40].About40μg of TEIF proteins from a cell- (FBS). HT1080 cells were maintained in minimum essential medium free transcription and translation system were incubated with 2 μgof (Invitrogen) with 10% FBS. All cells were maintained at 37 °C under a GST or GST-MSP58 fusion proteins in 0.2 ml of binding buffer (10 mM

5% CO2 atmosphere. Plasmids were transfected using PolyJet™ reagent Hepes (pH 7.5), 50 mM NaCl, 0.1% Nonidet P-40, 0.5 mM dithiothreitol, (SignaGen Laboratories, Gaithersburg, MD, USA). For stable transfection, and 0.5 mM EDTA) for 1–2 h, washed four times, and analyzed by sodi- cells were selected with either 400 μg/ml neomycin (G418) or 2 μg/ml um dodecylsulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) puromycin. To create retroviral stocks, 293T cells were cotransfected and by Western blotting using an anti-HA antibody. A fraction of the with the pQCXIP-GFP control or pQCXIP-GFP-MSP58 expression vector reaction mixture was analyzed by Coomassie blue staining to visualize and pCL-ampho packaging plasmid. To create lentiviral stocks, 293T GST fusion proteins. cells were co-transfected with the appropriate lentiviral expression vector, pCMVΔR8.74 packaging vector, and pMD2 VSVG envelope 2.8. Assay for telomerase activity vector. Culture supernatants were harvested at 48 h post-transfection and filtered through a 0.45-μm filter, split into aliquots, and stored For the telomerase activity analysis, protein extracts were prepared at −80 °C. Infections were carried out overnight in the presence of using a CHAPS lysis buffer (10 mM Tris–HCl at pH 7.5, 1 mM MgCl2, 8 μg/ml polybrene (Sigma). Cells were subsequently selected for an- 1mMEGTA,0.1mMPMSF,5mMβ-mercaptoethanol, 0.5% CHAPS, tibiotic resistance (2 μg/ml puromycin and/or 400 μg/ml neomycin). and 10% glycerol) and then extracted for 30 min on ice. Extracts were centrifuged at 13,000 ×g for 20 min, and the supernatants were collected. Supernatant extracts were quantified for protein with a BCA Protein 2.3. Western blot, coimmunoprecipitation, immunofluorescence, and Assay Kit (Pierce Biotechnology, Rockford, IL, USA). The telomeric repeat luciferase analyses amplification protocol (TRAP) assay was carried out as originally described [4] with modifications [57,58].Briefly, each reaction was The complete protocols for the Western blot, coimmunopre- carried out using 2 μl of cell lysate, 1 μl of each primer, 0.5 μl of Taq- cipitation, immunofluorescence, and luciferase reporter analyses polymerase (New England BioLabs, Ipswich, MA, USA), 5 μl of a 10× were described previously [48]. For the reporter assays, cells were TRAP reaction buffer (200 mM Tris–HCl at pH 8.3, 15 mM MgCl2, cotransfected with TERTLuc800, hTERT-p548, hTERT-p212, hTERT- 630 mM KCl, 0.5% Tween 20, and 10 mM EGTA), and 2 μl of dNTPs in p196, hTERT-p177, hTERT-p95, or Flag-MSP58 and the internal control DEPC-treated water in a final volume of 50 μl. Primers used for the reporter pRL-TK. Cells were harvested at 48 h post-transfection, and TRAP-assay PCR were the TS oligonucleotide (5′-AATCCGTCGAGCAG firefly luciferase activity was measured with the dual-luciferase AGTT-3′), which served as both the telomerase template and forward reporter assay system (Promega) and normalized against Renilla lucif- primer for the PCR; the CXa oligonucleotide (5′-GTGTAACCCTAACCCT erase activity. All experiments were performed at least three times AACCC-3′), which served as the reverse primer and the substrate for with each plasmid, and the mean relative luciferase activities were the 36-bp internal standard control (ITAS; 5′-AATCCGTCGAGCAGAGTT determined. AAAAGGCCGAGAAGCGAT-3′); and the reverse primer (NT; 5′-ATCGCT TCTCGGCCTTTT-3′) for amplification of the internal standard control (IC). The PCR program consisted of incubation at 30 °C for 30 min and 2.4. Isolation of RNA and real-time quantitative PCR (qPCR) then 30 rounds of 94 °C for 2 min, 94 °C for 30 s, 55 °C for 30 s, and 72 °C for 1 min, with a final extension at 72 °C for 2 min. TRAP products The detailed steps of RNA isolation and the reverse-transcription were resolved using electrophoresis in 12.5% non-denaturing polyacryl- (RT)-PCR were described previously [48,56]. The real-time qPCR was amide gels (acrylamide/N,N′-methylenebisacrylamide, 19: 1, w/w) in performed using the SYBR Green Advantage qPCR Premix (Clontech) 0.5× TBE (Tris-borate EDTA). Gels were stained with 5 μlofSYBR and C1000™ Thermal Cycler (Bio-Rad Laboratories). PCRs were then Green (Lonza, Rockland, ME, USA) for 30 min in 50 ml of TBE 0.5× performed using the following conditions for 30 cycles: 95 °C for 30 s, buffer, then exposed to ultraviolet (UV) light, and visualized using a 60 °C for 30 s, and 72 °C for 30 s. Values for hTERT were normalized Kodak image acquisition station. to the glyceraldehyde-3-phosphate dehydrogenase (GAPDH) housekeeping control. The following primer pairs were used: hTERT, forward 2.9. Chromatin immunoprecipitation (ChIP) analysis (5′-TGTTTCTGGATTTGCAGGTG-3′) and reverse (5′-GTTCTTGGCTTTCA GGATGG-3′); and GAPDH, forward (5′-CCCACTCCTCCACCTTTGAC-3′) ChIP assays were performed as described previously [56].Briefly, and reverse (5′-TCTCTCTTCCTCTTGTGCTCTTG-3′). sheared chromatin fragments were immunoprecipitated with antibodies specific to Flag, TEIF, or control immunoglobulin G (IgG) at 4 °C overnight. After dissociating the DNAs from the immunoprecipitated 2.5. Focus-formation and soft-agar assays chromatin, the DNAs were PCR-amplified. The association of TEIF and Flag-MSP58 with the hTERT promoter was measured with a nested Focus-formation and soft-agar assays of HT1080 or 293T cells over- PCR using the following primers: outer primers, 5′-AGTGGATTCGCG expressing HA-TEIF and GFP-MSP58 or MSP58 shRNAs were performed GGCACAGA-3′ (sense) and 5′-TTCCCACGTGCGCAGCAGGA-3′ (anti- by a method we described previously [48]. sense); and inner primers, 5′-TCCTGCCCCTTCACCTTCCAG-3′ (sense) 568 C.-C. Hsu et al. / Biochimica et Biophysica Acta 1843 (2014) 565–579

293T HT1080 A HT1080 stable cell lines

shLuc. shMSP58shMSP58 1 shLuc. 2 shMSP58shMSP58 1 2 par. vec. #11 #14

α MSP58 α Flag

α hTERT α hTERT

α actin α actin

B 293T HT1080 293T HT1080 Real-time qPCR Real-time qPCR Real-time qPCR Real-time qPCR

3.5 3 1.2 1.2

3 2.5 1 1 2.5 2 0.8 0.8 2 1.5 0.6 0.6 1.5 1 0.4 0.4 1 expression (Fold) expression expression (Fold) expression expression (Fold) expression 0.5 (Fold) expression 0.5 0.2 0.2 Relative hTERT mRNA hTERT Relative Relative hTERT mRNA hTERT Relative Relative hTERT mRNA hTERT Relative Relative hTERT mRNA hTERT Relative 0 0 0 0 r 8 r to 5 to 58 c P c P shLuc. shLuc. e S e S v M v M shMSP58 1 shMSP58 2 shMSP58 1 shMSP58 2 C Myc/Max Sp1 Sp1 Sp1 Sp1 Sp1 Myc/Max 7 Blank E2F ATG MSP58 hTERT 6 - 200 - 150 - 100 -50 +50 - 800 5 Luc p800 ) - 548 5 4 Luc p548 0

- 212 p212 3 Luc LA (X1 R - 196 Luc p196 2 - 177 Luc p177 1

-95 Luc p95 0 p800 p548 p212 p196 p177 p95 293T; p800 D 293T; p196 293T; p800

2.5

3 6 2 E

2.5 5 ) 3 1.5 293T HT1080 ) ) 10 3 5

2 4 X r ( er e 1 ff ff

A u u b r Flag-MSP58 b r Flag-MSP58 1.5 3 s o s o i t si t RL s c c y e ly e 0.5 l v v 1 2 RLA (X10 RLA (X10

0.5 1 0 α MSP58 0 0 --2050 100 20 100 MSP58 (ng) α actin

α Flag α Flag (MSP58) (MSP58) Control si αactin α actin MSP58 si-3

HT1080; p800 HT1080; p196 HT1080; p800

IC 5 IC 5 3

2.5 4 ) ) ) 4 3 5 2 2 α α Flag 3 Flag 3 1.5 α actin α actin 2

2 RLA (X10 RLA (X10 RLA (X10 1 1 1 0.5

0 0 0 --2050 100 20 100 α MSP58 (ng) MSP58

α Flag α Flag αactin (MSP58) (MSP58)

α actin α actin ControlMSP58 si si-3 C.-C. Hsu et al. / Biochimica et Biophysica Acta 1843 (2014) 565–579 569 and 5′-TTCCCACGTGCGCAGCAGGA-3′ (antisense). The amplified DNA National Cheng Kung University. The mice were divided randomly into was separated on 2% agarose gels and visualized by ethidium bromide four groups. Before implanting on Day 0, HT1080 GFP HA empty, staining. HT1080 GFP-MSP58 HA, HT1080 HA-TEIF GFP, and HT1080 GFP-MSP58 HA-TEIF cells were harvested and counted. The cells (2.5 × 106 cells) 2.10. Electrophoretic mobility shift assay (EMSA) were suspended in 0.1 ml of PBS, and then inoculated subcutaneously into the back of mice (five mice for each group), which led to palpable TEIF and MSP58 proteins used in the EMSA were in vitro-translated nodules on Day 8. The growth of tumors was measured with calipers by using a TNT quick coupled transcription–translation system (Promega) every two days from Day 8 to 23. The width (a) and length (b) of the according to the manufacturer's procedures. An EMSA with biotin-labeled tumors were measured by calipers to evaluate tumor volume according probes was performed using the LightShift chemiluminescent EMSA kit to the formula V = ba2/2. Mice were sacrificed on Day 24 and tumors (Pierce Biotechnology). The 234-bp double-stranded probe of the were collected to measure their weight. For Western blot and telomerase hTERT promoter was amplified by a PCR with a 5′-biotin-labeled or activity assays, tumor tissues were dissected after 24 days, quick-frozen 5′-unlabeled primer and purified using the PCR-M Clean Up System in dry ice, and stored at −80 °C. (Viogene, Taipei, Taiwan). The following specific primer pairs were used: forward 5′-TTCACCTTCCAGCTCCGCCTCCTCCGCGCG-3′ (−144 to 2.13. Statistical analysis −127) and reverse 5′-GAGCGCACGGCTCGGCAGCGGGGA-3′ (+90 to +74). Biotin-labeled DNA fragments were incubated for 30 min on Statistical comparisons were performed using a two-tailed Student's ice with the in vitro-translated protein (2 μl) in EMSA binding buffer t-test. p values were calculated using the GraphPad Prism software vers. (10mMTrisatpH7.5,50mMKCl,1mMDTT,5%glycerol,0.05% 3.03 package (GraphPad Software, San Diego, CA, USA). NP-40, and 0.1 μg sheared salmon sperm DNA). For the probe competition experiments, the TNT-TEIF protein was preincubated with an unla- 3. Results beled probe for 10 min at 4 °C before a 20-min incubation with the biotin-labeled probe. For competition assays, in vitro-translated protein 3.1. Effects of MSP58 on TERT transcription and telomerase activity in of MSP58FL,MSP581–300, or MSP58300–462 (increasing volumes 2 to 4 μl) HT1080 and 293T cells were preincubated with the TEIF for 10 min followed by the addition of the biotin-labeled probe. For the supershift assay, the TNT-TEIF recom- Our previous data showed that stable MSP58 overexpression in- binant protein was preincubated with an anti-HA antibody for 10 min duced cellular senescence in HT1080 fibrosarcoma cells [48]. Interest- before the addition of the biotin-labeled probe. The reaction mixture ingly, those cells displayed a marked reduction in telomerase activity. was loaded onto a 4% non-denaturing polyacrylamide gel, which was We further demonstrated that MSP58 overexpression can also induce run in 0.5× TBE buffer at 200 V for 2 h. The DNA–protein complex senescence-associated growth arrest in 293T cells (Supplemental was transferred to a nylon membrane and UV-crosslinked. The biotin Fig. 1A, B). In the previous report [43], it has been demonstrated that end-labeled DNA probe was detected using a streptavidin-horseradish MCRS2, a splice isoform of human MSP58, can associate with the peroxidase conjugate and a chemiluminescent substrate. catalytic subunit hTERT of telomerase. Forced expression of MCRS2 inhibited telomerase activity in vitro. These findings suggest that 2.11. Cell cycle analysis MSP58 inhibits telomerase activity might also through physical interaction with hTERT protein. To examine this possibility, we investigated Cells transfected and infected with the indicated plasmids were the interaction between MSP58 and hTERT in vivo. HT1080 MSP58 trypsinized and washed with ice-cold PBS. After fixation with 4 °C 75% stable clone (no.14) cell lysates were immunoprecipitated with anti- ethanol for 24 h, the cells were treated with 0.2 mg/ml RNase A Flag antibody, and Western blot analysis was performed with anti- (Invitrogen™, Grand Island, NY) and stained with 20 μg/ml PI (Sigma- Daxx or hTERT antibody. Using a rabbit monoclonal antibody Y182, Aldrich, St Louis, MO, USA) for 1 h at room temperature. To analyze we could detect the basal levels of hTERT protein in HT1080 cells and cell distribution in the cell cycle, PI-stained cells were analyzed by a other cancer cell lines (data not shown). Endogenous Daxx protein flow cytometer (Beckman coulter Cell Lab Quanta™ SC flow cytometry). could be coimmunoprecipitated by anti-Flag antibody but not by non- Data were analyzed with WinMDI 2.8 software. specific mouse IgG (Supplemental Fig. 2A), consistent with our previous report [40]. In contrast, no appreciable interaction was detected 2.12. Xenograft analysis between Flag-MSP58 and hTERT (Supplemental Fig. 2A). Next, we examined the interaction between MSP58 and hTERT in vitro by GST Twenty female athymic SCID mice (6–8 weeks of age) were pur- pull-down assay. In vitro transcriptional and translational lysates con- chased from the National Laboratory Animal Center of National Cheng taining full-length Daxx or hTERT protein were incubated with the Kung University in Taiwan. The mice were housed in individually venti- immobilized GST or GST-MSP58 fusion protein, analyzed by SDS- lated sterile cages. All cage accessories were sterilized and autoclaved PAGE, and detected by immunoblotting. A strong retention of Daxx pro- before they were placed in the cage. The animal experiment was tein, but not hTERT protein was obtained in the sample containing approved by the Institutional Animal Care and Use Committee at immobilized GST-MSP58 (Supplemental Fig. 2B). These results are

Fig. 1. Effect of the MSP58 on hTERT expression and telomerase activity in 293T and HT1080 cells. A, left; 293T and HT1080 cells were infected with two independent lentiviruses encoding MSP58 short-hairpin (sh)RNAs (shMSP58 1 and 2) or a luciferase control sequence (shLuc) and then cultured at 37 °C for 3 days. Cell extracts were prepared and analyzed by immunoblotting assays using antibodies against MSP58, hTERT, or actin. Right, MSP58 and hTERT expressions were also analyzed using immunoblotting in HT1080 parental, vector, and stably MSP58-overexpressing cells (clones 11 and 14). B, quantitative RT-PCR analysis of hTERT mRNA in 293T and HT1080 cells 3 days after lentiviral shRNA transduction or transient transfection with a control or Flag-MSP58 vector. Values for hTERT were normalized to the GAPDH housekeeping control. The expression of hTERT in shLuc or Flag vector control cells was deﬁned as 1.0, and other values were accordingly normalized. Columns, mean of three independent experiments; bars, standard deviation. *p b 0.05; **p b 0.01. C, schematic representation of various hTERT promoter constructs used in this assay. Luc, luciferase. Numbers refer to the number of bases upstream (−) or downstream (+) of the ATG initiation codon of the hTERT gene. Relative luciferase activities (RLA) of extracts from HT1080 cells cotransfected with the luciferase reporter plasmid containing hTERT promoter deletion mutants TERTLuc800 (−800 to +1), p548 (−548 to +50), p212 (−212 to +50), p196 (−196 to +50), p177 (−177 to +50), p95 (−95 to +50), and pRL-TK (internal control vector) for 24 h were subjected to a luciferase assay. Results represent the mean ± standard deviation from three independent experiments. *p b 0.05. D, 293T and HT1080 cells were cotransfected with the TERTLuc800 or p-196 reporter construct, pRL-TK, and the indicated amounts of MSP58-overexpressing (Flag-MSP58) or -knockdown (pSuper-MSP58 si-3) plasmids, followed by a luciferase assay at 48 and 72 h post-transfection, respectively. MSP58 protein levels were assessed by Western blotting. Column and error bars indicate the mean ± standard deviation obtained from three independent experiments. *p b 0.05; **p b 0.01; ***p b 0.001. E, telomerase activity assayed by a telomeric repeat ampliﬁcation protocol as described under “Materials and methods” in cell lysates from the same experiments as in panel D. CHAPS lysis buffer for a negative control was included. The arrow points to the 36-bp internal control (IC). 570 C.-C. Hsu et al. / Biochimica et Biophysica Acta 1843 (2014) 565–579

A B WCL IP: Flag WCL IP: HA - T t S 58 u T G P p S S In G M HA-TEIF HA-TEIF α HA HA-TEIF Flag-MSP58 Flag-MSP58

kDa kDa 170 kDa 170 130 130 130 90 GST-MSP58 90 72 90 αHA αFlag 55 72 72

43 55 55

130 34 α Flag α HA 55 90 26 GST HT1080 MSP58 #11

IF g t G E t G la u Ig T u Ig F p : : p : : In IP IP In IP IP α TEIF α Flag α Flag α TEIF

CERelativeβ -gal activity

Coiled-coil 0 2500 5000 7500 10000 NoLS domain 1 462 aa Gal4 AD FL MSP58 FL LexA- Gal4 AD-TEIF NLS FHA domain lamin Gal4 AD-TEIF 1-395 Gal4 AD-TEIF 396-786 1 300 MSP58 1-300 Gal4 AD LexA- Gal4 AD-TEIF FL 300 462 MSP58 300-462 MSP58 Gal4 AD-TEIF 1-395 FL Gal4 AD-TEIF 396-786 putative STYKc domain Gal4 AD 786 aa FL 1 LexA- Gal4 AD-TEIF FL TEIF MSP58 Gal4 AD-TEIF 1-395 1-300 Gal4 AD-TEIF 396-786 1 395 TEIF 1-395 Gal4 AD 786 Gal4 AD-TEIF FL 396 LexA- TEIF 396-786 MSP58 Gal4 AD-TEIF 1-395 300-462 Gal4 AD-TEIF 396-786

WCL IP: Flag WCL IP: HA WCL IP: Flag WCL IP: HA

HA-TEIF Flag-MSP58 FL Flag-MSP581-300 HA-TEIF 1-395 Flag-MSP58300-462 HA-TEIF 396-786 170 72 95 95 kDa 130 55 72 43 72 90 55 34 α HA α Flag 43 αHA 55 α Flag 72 26 34 43 55 26 17 34

72 130 72 kDa 72 55 55 90 43 55 α αFlag αHA αFlag 43 HA 34 72 43 34 26 55 34 26 26 43 C.-C. Hsu et al. / Biochimica et Biophysica Acta 1843 (2014) 565–579 571 consistent with those of the coimmunoprecipitation experiments, TEIF could form a complex in cells. Furthermore, we mapped the in- suggesting that MSP58 does not associate with hTERT. hTERT gene teraction domain(s) using deletion derivatives of MSP58 and TEIF in expression is highly regulated and generally correlates with telomerase yeast two-hybrid assays and coimmunoprecipitation assays. As activity. We further examined whether MSP58 can regulate hTERT shown in Fig. 2D and E, the results implied that the C-terminal region expression. To assess this, we silenced gene expression using lentivirus of MSP58 (amino acids 300–462) was sufficienttointeractwithTEIF, particle carrying shRNA against MSP58 in the HT1080 and 293T cell and the C-terminal region of TEIF (amino acids 396–786) mediates lines and then immunoblotted extracts for expressions of MSP58 and interaction with MSP58. These results suggest that MSP58 may mod- hTERT. As shown in Fig. 1A (left panel), two different shRNAs against ulate hTERT transcription by directly interacting with TEIF. MSP58 (1 and 2) induced increased the steady-state levels of hTERT protein in both the HT1080 and 293T cell lines, compared to the controls 3.3. Overexpression of MSP58 suppresses TEIF-mediated transactivation of which expressed luciferase shRNA. In contrast, hTERT protein levels the TERT promoter and telomerase activity decreased in both MSP58-overexpressing HT1080 stable cell lines (clones 11 and 14) compared to control cells (Fig. 1A, right panel). Con- Previous studies have demonstrated that TEIF transactivates hTERT sistently, MSP58 knockdowns, as shown by RT-qPCR analysis, upregu- promoter activity [53]. We performed luciferase reporter assays to lated hTERT mRNA expression in both the HT1080 and 293T cell lines elucidate whether MSP58 affects TEIF-mediated transcriptional activa- (Fig. 1B). A similar decrease in hTERT mRNA expression was obtained tion of the hTERT promoter. The MSP58 expression construct was by transiently transfecting HT1080 and 293T cells with the MSP58 ex- cotransfected into 293T or HT1080 cells along with the TEIF expression pression construct (Fig. 1B). The finding that MSP58 downregulates construct and TERTLuc800 reporter. We demonstrated that TEIF overex- hTERT mRNA suggests that hTERT expression is controlled by MSP58 at pression induced hTERT promoter activity that was two to three times the transcriptional level. To examine whether MSP58-mediated down- higher than the vector control (Fig. 3A). Overexpression of MSP58 itself regulation of hTERT mRNA is due to promoter repression, we performed significantly suppressed promoter activity. Also, overexpression of aseriesofhTERT promoter-luciferase gene reporter assays. As demon- MSP58 resulted in inhibition of luciferase activity activated by TEIF in strated in Fig. 2C, MSP58 repressed transcriptional activities of the HT1080 and 293T cells in a dose-dependent fashion (Fig. 3A). Converse- TERTLuc800 (p800) [15], hTERT-p548, hTERT-p212, hTERT-p196, ly, compared to the control vector, both MSP58 shRNAs (si-3 and si-5) hTERT-p177, and hTERT-p95 [55] constructs when cotransfected into significantly elevated TEIF-mediated activation of the hTERT promoter HT1080 cells. Furthermore, results showed that MSP58 reduced (Fig. 3A). Similar results were observed using the hTERT-p196 construct TERTLuc800 and hTERT-p196 luciferase activities in a dose-dependent (Fig. 3B). To establish a correlation between the MSP58/TEIF interac- manner in both HT1080 and 293T cells, while knockdown of MSP58 in- tion and the TEIF transcriptional repression by MSP58, the ability of creased the promoter activity (Fig. 1D). To examine the effects of MSP58 specific MSP58 deletion mutants to repress TEIF-mediated trans- overexpression on telomerase activity, we performed a TRAP assay. activation was analyzed. As illustrated in Fig. 3C, overexpression of Consistently, cells expressing MSP58 showed low levels of telomerase MSP58-(300–462), like full-length MSP58, was capable of suppressing activity in a dose-dependent manner (Fig. 1E). These data suggest TEIF-mediated hTERT transcriptional activation, whereas MSP58- that MSP58 downregulates telomerase activity by inhibiting hTERT (1–300) failed to do so, indicating that the C-terminal region of MSP58 expression. was sufficient for MSP58-mediated TEIF trans-repression. These data provide direct evidence that the interaction between MSP58 and TEIF 3.2. Identification of TEIF as a novel MSP58-interacting protein was clearly correlated with MSP58's function as a TEIF trans-repressor. Furthermore, we tested whether MSP58 could repress TEIF-mediated To further identify MSP58-binding proteins that may be involved in activation of telomerase activity. As shown in Fig. 3D, telomerase activity telomerase regulation and/or cellular senescence, we performed yeast was activated by about 1.8 fold in 293T cells 48 h after TEIF transfection. two-hybrid screening using the full-length of MSP58 as bait. One We also observed that knockdown of TEIF gene expression by TEIF siRNA positive clone encoding telomerase transcriptional element-interacting inhibits telomerase activity in HT1080 cells (data not shown). Notably, factor (TEIF) cDNA was isolated from a Matchmaker human testis MSP58 suppressed TEIF-mediated enhancement of endogenous telome- cDNA library (data not shown). TEIF was first identified as a protein rase activity of 293T cells in a dose-dependent manner (Fig. 3D, left that binds to the hTERT promoter and acts as a positive modulator of panel). Consistently, telomerase activity was significantly inhibited in the hTERT gene [53]. To further verify the protein interaction between the TEIF-transfected HT1080 MSP58 clone 11 and 14 cells (Fig. 3D, MSP58 and TEIF in vitro, GST pull-down experiments were carried right panel) compared to the parental and vector control cells. Together, out using the GST-MSP58 fusion protein and a battery of in vitro- these results suggest that MSP58 serves as a negative regulator of TEIF in translated TEIF. As shown in Fig. 2A, the full-length TEIF was specifically controlling the hTERT promoter and telomerase activity. pulled down by GST-MSP58, but not by the GST protein. The interaction between MSP58 and TEIF was further validated by co-immuno 3.4. MSP58 does not alter nuclear localization of the TEIF or hTERT precipitation assays in mammalian cells. Overexpressed MSP58 and TEIF proteins were reciprocally coimmunoprecipitated in 293T cells To elucidate the molecular mechanism underlying MSP58-mediated (Fig. 2B, top panel). Using another approach, endogenous TEIF pro- modulation of the TEIF transactivation potential and telomerase teins were coimmunoprecipitated in the stable HT1080 MSP58- activation, immunofluorescence microscopic analyses were performed. overexpressing cell line (no. 11) by an anti-Flag antibody (Fig. 2B, Because we previously reported that MSP58 can relieve the transcrip- bottom panel). These results provided evidence that MSP58 and tional repressor activity of Daxx through a nucleolar sequestration

Fig. 2. MSP58 interacts with the TEIF in vitro and in vivo. A, GST or GST-MSP58 fusion proteins were expressed in Escherichia coli BL21 cells and immobilized on glutathione-Sepharose. The TEIF was obtained with a cell-free transcription and translation (TNT) system in vitro and incubated with GST or GST-MSP58 proteins. Bound proteins were detected by immunoblotting with an anti-HA antibody. B, coimmunoprecipitation assays. 293T cells were transfected with the indicated expression vectors. Whole-cell lysates (WCL) from 293T and HT1080 MSP58 clone 11 cells were subjected to immunoprecipitation (IP) experiments followed by a Western blot analysis with the indicated antibodies. Arrows indicate the positions of HA-TEIF and Flag-MSP58. Stars indicate the heavy chain of IgG. C, schematic presentation of wild-type MSP58 and TEIF, and different deletion mutants tested in mammalian transfection and yeast two- hybrid assays. Numbers indicate the amino acid position. NLS, nuclear-localization signal; NoLS, nucleolar-localization signal; FHA, forkhead-associated; STYKc, serine/threonine/tyrosine kinase. D, mapping of the MSP58 and TEIF interaction domains. 293T cells co-transfected with HA-TEIF and Flag-MSP58 full-length or deletion mutants of expression constructs were subjected to IP experiments followed by a Western blot analysis with the indicated antibodies. Arrows indicate the positions of HA-TEIF and Flag-MSP58. Stars indicate the heavy chain or light chain of IgG. E, Yeast cotransformed with bait and prey as indicated were analyzed by quantitative β-Gal assays. Data represent the mean ± the standard deviation of three separate experiments. LexA-lamin served as a negative control. 572 C.-C. Hsu et al. / Biochimica et Biophysica Acta 1843 (2014) 565–579

A 293T; p800 293T; p800HT1080; p800 HT1080; p800

3 7 3 2 2.5 6 2.5 ) ) ) ) 1.5 3 2 3 3 2 5 2 4 1.5 1.5 1 3 RLA (X10 RLA (X10 RLA (X10 RLA (X10 1 1 2 0.5 0.5 1 0.5 0 0 0 0 HA-TEIF HA-TEIF HA-TEIF HA-TEIF Control si Flag-MSP58 20 50 100 Flag-MSP58 20 50 100 Control si (ng) MSP58 si-3 (ng) MSP58 si-3

α HA MSP58 si-5 MSP58 si-5 α HA

α Flag α HA α HA α Flag α actin α MSP58 α MSP58 α actin

α actin α actin

BCHT1080; p196 HT1080; p196 293T; p800

2.5 7 2.5 6 ) ) ) 4

2 3 5 2 5 1.5 4 1.5 3 RLA (X10 RLA (X10 1 RLA (X10 1 2 0.5 0.5 1 0 0 0 α HA-TEIF HA-TEIF HA-TEIF HA Control si Flag-MSP58 FL Flag-MSP58 20 50 100 (ng) MSP58 si-3 Flag-MSP58 1-300 αFlag MSP58 si-5 Flag-MSP58 300-462 α HA α HA α Flag α MSP58 α actin α actin

α actin

D 293T HT1080 control HA-TEIF

HA-TEIF Flag-MSP58

bp 2.5 bp 100 2 90 100 80 1.5 90

(fold) 80 70 1 70 60 0.5

Telomerase activity 60 50 0 50 HA-TEIF

40 Flag-MSP58 50 20 100 (ng) 40 IC IC

α HA α HA α Flag α actin α actin C.-C. Hsu et al. / Biochimica et Biophysica Acta 1843 (2014) 565–579 573 mechanism [40], we then examined whether overexpression of MSP58 MSP58 clones 11 and 14 was also observed in a ChIP assay using an anti- could alter subcellular localization of TEIF or hTERT. As illustrated in Flag antibody (Fig. 5C, lanes 4 and 5). Notably, the amount of TEIF bound Fig. 4, 293T cells were transiently transfected with individual construct to the hTERT promoter significantly decreased in HT1080 MSP58 clones or co-transfected with combinations of the EGFP-MSP58 and HA-tagged 11 and 14 cells compared to parental and vector control cells (Fig. 5C, TEIF or Flag-tagged hTERT constructs and subsequently stained with an lanes 4 and 5). The repression by MSP58 was not associated with a anti-HA or Flag antibody followed by an immunofluorescence analysis. change in the expression level of TEIF protein (data not shown). Togeth- Consistent with a previous report [53,59], TEIF and hTERT were pre- er, these results suggest that MSP58 inhibits TEIF activity through a phys- dominantly localized in nuclei (Fig. 4E–H), while MSP58 was mostly ical interaction, interfering with the ability of TEIF to interact with DNA. localized in nuclei and concentrated in nucleoli (Fig. 4A, B). In the con- These findings point to a novel mechanism whereby MSP58 represses trol, the co-transfection of MSP58 drastically recruited Daxx protein TEIF-mediated hTERT transactivation. from the nucleoplasm to nucleoli in 293T cells (Fig. 4C, D, I–L). However, overexpression of MSP58 did not alter nuclear staining of TEIF or hTERT 3.6. MSP58 suppresses TEIF-mediated promotion of cell proliferation (Fig. 4M–T), suggesting that MSP58's repressive effect did not result in vitro and in vivo from alterations of TEIF or hTERT nuclear localization. Telomerase or hTERT inhibition compromises sustained prolifera- 3.5. MSP58 prevents TEIF from binding to hTERT promoter DNA tion and tumorigenic ability of cancer cells [60,61]. To further evaluate the effect of the MSP58–TEIF complex on hTERT gene regulation and Given MSP58's ability to bind to the C-terminal hTERT promoter cellular proliferation, we first established human HT1080 and 293T DNA-binding domain of TEIF, one possibility is that MSP58 represses cell lines that exhibited stable expression of ectopic HA-tagged TEIF. TEIF activity through direct interference with TEIF's DNA-binding capac- HT1080 and 293T TEIF cells exhibited remarkable colonies in mono- ity. To test this hypothesis, we performed EMSAs. The in vitro-translated layers, and marked colony formation in soft agar compared to vector full-length TEIF was incubated with a biotin-labeled hTERT promoter control cells (Fig. 6A and Supplemental Fig. 3). These results indicate DNA probe (Fig. 5A). The protein–DNA complexes were resolved by na- that TEIF has pro-proliferative properties. To establish the functional tive PAGE. Clearly, the TEIF protein and hTERT probe formed a protein– relationship between the MSP58/TEIF interaction and cell-growth con- DNA complex, whereas the control vector did not form a visible com- trol, HT1080 and 293T cells with stable expression of TEIF were infected plex with the probe (Fig. 5B, left, lane 4 vs. lane 1). This protein–DNA with a retrovirus carrying a GFP-MSP58 or control vector. Protein ex- complex was specific, since formation of the complex could be pression was determined by Western blotting with anti-HA and anti- abolished by a 100 fold excess of the unlabeled hTERT oligonucleotide GFP antibodies (Fig. 6A and Supplemental Fig. 3). MSP58 overexpres- (Fig. 5B, left, lane 5 vs. lane 4). Moreover, this complex could be further sion alone in HT1080 cells significantly reduced the colony size in soft supershifted when an anti-HA antibody was added (Fig. 5B, right, lane 8 agar and colony numbers in monolayers (Fig. 6A, middle and right). Re- vs. lane 9), indicating specific complex formation between the HA-TEIF markably, colony numbers of HT1080 and 293T cells co-expressing TEIF and hTERT. We next tested the effect of MSP58 on this complex forma- and MSP58 were lower than those in TEIF-expressing cells (Fig. 6Aand tion. Remarkably, preincubation of the TEIF reaction mixture with Supplemental Fig. 3). In cell cycle analysis, TEIF-overexpressing cells recombined wild-type MSP58 protein prevented formation of the com- showed a decrease in G1 and an increase in S populations relative to plex of HA-TEIF with the hTERT probe in a dose-dependent fashion control cells. Consistent with our previous study, a significant decrease (Fig. 5B, right, lanes 9–11). Of note, we found that MSP58 did not directly in the number of G1 cells and a concomitant increase in S and G2/M bind to the hTERT promoter in vitro (Fig. 5B, left, lane 2), indicating that cells were seen in the MSP58-overexpressing cells (Fig. 6B). Lack of its prevention of TEIF–hTERT probe complex formation was not due to sub-G1 suggested that MSP58 upregulation does not induce apoptosis competition for hTERT promoter binding. Additionally, the C-terminal of HT1080 cells. Combining MSP58 and TEIF expression mainly region of MSP58 (300–462) more significantly abrogated complex for- sustained an increase in S population. Further, we used an in vivo subcu- mation between the TEIF and hTERT probe than did full-length MSP58 taneous tumor formation assay to examine the proliferative ability of (Fig. 5B, right, lanes 14 and 15), whereas the N-terminal region of MSP58- and TEIF-overexpressing cells in the mice. Fig. 6C (left) shows MSP58 (1–300) was unable to affect TEIF binding to the hTERT probe the time course of the growth of tumors initiated by different HT1080 (Fig. 5B, right, lanes 12 and 13), consistent with results of the luciferase transfectants. As compared with control cells carrying empty vectors, reporter assays (Fig. 3C). These findings suggest that the interaction be- injecting mice with HT1080-MSP58 cells resulted in a significantly tween MSP58 and the TEIF results in the DNA-binding domain of TEIF decreased in growth rate of tumors, whereas mice injected with are masked and further repressed TEIF-mediated activation of hTERT HT1080-TEIF cells showed significantly increased tumor growth. The transcription. Next, we examined whether the expression of MSP58 growth kinetic of tumors in the MSP58/TEIF coexpression group was prevented hTERT-binding activity of the endogenous TEIF in stable similar to that of the tumors in the HT1080-MSP58 group. The tumors MSP58-overexpressing HT1080 cells using a ChIP analysis. After formal- were removed from each mouse on Day 24 and weighed. Consistently, dehyde cross-linking and precipitation of chromatin with an anti-TEIF or overexpression of TEIF significantly increased solid tumor mass, where- Flag antibody, the precipitated DNA was PCR-amplified with a set of spe- as overexpression of MSP58 decreased solid tumor mass compared with cific primers flanking the hTERT proximal core promoter region (Fig. 5A). the control group (Fig. 6C, middle). The tumor weight in the HT1080- As a control, TEIF clearly associated with the endogenous hTERT promoter MSP58/TEIF xenografts was significantly lower than that of the (Fig. 5C, lanes 2 and 3). MSP58 binding to the hTERT promoter in HT1080 HT1080-TEIF group. Notably, the hTERT protein level and telomerase

Fig. 3. MSP58 represses TEIF-mediated activation of the hTERT promoter and telomerase activity. A and B, 293T and HT1080 cells were transiently transfected in 6-well plates with 500 ng of TERTLuc800 or p-196 reporter construct, 50 ng of pRL-TK, 200 ng of a TEIF expression vector (HA-TEIF), and the indicated amounts of MSP58-overexpressing (Flag-MSP58) or -knockdown (500 ng of pSuper-MSP58 si-3 or si-5) plasmids, followed by a luciferase assay 48 and 72 h post-transfection, respectively. The Western blot shows expression of Flag-MSP58, HA-TEIF, endogenous MSP58 and actin. Columns, mean of three independent experiments; bars, standard deviation. *p b 0.05; **p b 0.01. C, 293T cells were cotransfected with 500 ng of TERTLuc800, 50 ng of pRL-TK, 200 ng of HA-TEIF, together with 100 ng of expression plasmids for either of the different MSP58 mutants (Flag-MSP58FL, Flag-MSP581–300,orFag-MSP58300–462), followed by luciferase reporter assays. Western blot shows expression of HA-TEIF and Flag-MSP58FL or mutants. Error bars denote one standard deviation of triplicate measurements. *p b 0.05; **p b 0.01. NS, non-signiﬁcant. D, the TEIF and indicated amounts of MSP58 or control expression vectors were transfected into 293T cells. Western blot analyses of the TEIF, MSP58, and actin proteins, and TRAP assays were performed 48 h after transfection (left). Positive 293T-cell extract control, and CHAPS lysis buffer for negative controls were included. IC, internal control. The ImageJ (NIH) program was used to measure the relative amounts of bands in the TRAP assays after normalization to a 36-bp band (middle). Error bars denote one standard deviation of quadruplicate measurements. Right, HT1080 parental, control vector, and MSP58 stable clone 11 and 14 cells were transfected with the TEIF or control vectors, and TRAP assays were performed 48 h after transfection. Western blot analyses of TEIF and actin proteins. 574 C.-C. Hsu et al. / Biochimica et Biophysica Acta 1843 (2014) 565–579

A B C D

EGFP- MSP58 DAPI HA-Daxx DAPI

E F G H

Flag- HA-TEIF DAPI hTERT DAPI I J K L

EGFP-MSP58 HA-Daxx Merge DAPI M N O P

EGFP-MSP58 HA-TEIF Merge DAPI

Q R S T

EGFP- Flag- MSP58 hTERT Merge DAPI

Fig. 4. MSP58 does not alter TEIF or hTERT nuclear localization. Green fluorescence images of 293T cells transfected with a vector expressing EGFP-MSP58 (A, B). Additionally, immunofluorescence images of 293T cells expressing HA-tagged Daxx, TEIF, or Flag-tagged hTERT alone (C–H), or with EGFP-MSP58 (I–T) as indicated, immunostained with a mouse anti-HA or Flag antibody, followed by the secondary antibody of Texas red-conjugated anti-mouse IgG. DAPI staining revealed the position of the nucleus. Superimposition of green and red images (Merge) resulted in a yellow signal. activity were reduced in cells and tumor tissues co-expressing TEIF and mediated by MSP58 protein which interferes with the DNA-binding MSP58, compared to those expressing TEIF alone (Fig. 6A, left and C, activity of TEIF through interaction with the TEIF DNA-binding domain. right). These results further substantiate the notion that MSP58 func- These findings provide insight into the mechanisms modulating hTERT tions as a negative TEIF coregulator in downregulating TEIF-mediated transcription and cell proliferation at the molecular level that involve hTERT expression and promoted cell proliferation. MSP58 and its interplay with TEIF. MSP58 interacts with multiple proteins to exert distinct cellular 4. Discussion functions. In particular, MCRS2, an isoform of human MSP58, interacts with a potent telomerase inhibitor, LPTS/PinX1 and hTERT [43].MCRS2 Regulation of hTERT gene expression by a variety of factors is crucial inhibited telomerase activity in vitro, and long-term overexpression of for the progression and maintenance of human tumors [10].Inthis MCRS2 resulted in progressive telomere shortening. Here, we demon- study we showed that MSP58 is an important regulator of hTERT expres- strated that MSP58 repressed telomerase activity by inhibiting TEIF- sion. In summary, our results indicated that (i) ectopic expression of mediated transactivation of the hTERT promoter (Fig. 3). Although a MSP58 in both 293T and HT1080 cells resulted in decreased hTERT direct interaction between MSP58 and hTERT was not detected (Supple- transcription and telomerase activity, whereas knockdown of MSP58 mental Fig. 2), identification of MSP58 binding to the hTERT promoter exerted the opposite effect; (ii) MSP58 specifically suppressed TEIF- and modulation of hTERT expression by MSP58 indicate major control mediated hTERT transactivation, telomerase activation, and cell growth- points at the transcriptional level (Fig. 1). Our data indicate that pro- promoting activity in vitro and in vivo; and (iii) this regulation was tein–protein interactions between MSP58 and TEIF are significant for C.-C. Hsu et al. / Biochimica et Biophysica Acta 1843 (2014) 565–579 575

A C HT1080 ATG Transcription Star t markerparentalvector MSP58 MSP58#11 #14 12345 Sp binding sites 250 bp IP Flag 184 bp 184 bp ChIP-PCR -153 – (+31) 250 bp IP TEIF EMSA probe -144 – (+90) 184 bp

250 bp IP IgG 184 bp

250 bp Input 184 bp B Control vector

TEIF

MSP58 FL

Control vector MSP58 1-300

TEIF MSP58 300-462

MSP58 competitor

competitor HA antibody 12345 6 7 8 9 10 11 12 13 14 15

TEIF supershift TEIF/probe complex TEIF/probe complex

Free probe Free probe

kDa α HA

90 26 * 55 * * * α Flag 43 * *

Fig. 5. MSP58 inhibits TEIF DNA-binding activity. A, schematic presentation of the hTERT promoter and PCR primer locations (relative to ATG) for the EMSA and ChIP assays. B, EMSA analysis of DNA binding activity of in vitro-translated TEIF, MSP58 and deletion mutants using the biotin-labeled double-stranded oligonucleotide corresponding to the hTERT promoter sequence −144 to +90 as a probe. In lanes 1 and 6, negative controls with control vector in vitro-translated lysate (2 μl), and in vitro-translated MSP58 (2 μl) failed to bind to the hTERT promoter (lane 2). In lanes 4 and 9, in vitro-translated TEIF (2 μl) binds efﬁciently to the hTERT promoter. In lanes 5 and 7, TEIF (2 μl) binding could be competed away with an unlabeled hTERT probe (competitor, 100-fold molar excess). In lane 3, a negative control with control vector in vitro-translated lysate (2 μl) and competitor. In lane 8, supershifting of the DNA-bound TEIF band with anti-HA antibody. The arrows indicate the position of shifted and supershifted bands. Lanes 10, 11, 14 and 15 show addition of in vitro-translated MSP58FL or MSP58300–462 proteins (2 and 4 μl as indicated) to the TEIF-binding reaction mixture interferes with TEIF binding to the probe in a dose-responsive manner (compare with lane 9), suggesting a physical interaction between the two proteins. Lanes 12 and 13 show that addition of increasing amounts of in vitro-translated MSP581–300 protein (2 and 4 μlas indicated) to the TEIF-binding reaction mixture does not interfere with TEIF binding to the probe (compare with lane 9). The bottom panel shows the recombinant TEIF or MSP58 protein levels used in the above EMSA by immunoblotting. Stars indicate the positions of MSP58. Data are representative of three experiments with similar results. C, The ChIP assay was performed as described under “Materials and methods”. HT1080 parental, control vector, and MSP58 stable clone 11 and 14 cell lysates were incubated with different antibodies as indicated, the precipitated DNA was analyzed by a PCR using primers corresponding to the hTERT promoter sequence, and the resultant DNA fragments of 184 bp for hTERT were separated on 2% agarose gels. Normal IgG was used as the negative control. “Input” bands were obtained from DNA puriﬁed from chromatin not yet immunoprecipitated. Experiments were repeated three times.

MSP58-mediated downregulation of the hTERT promoter. Notably, DNA-binding domains and must therefore be recruited to DNA through several studies have shown that MSP58 binds to individual transcription interactions with transcriptional regulators, most likely through an factors, modulating their effects on target genes [40–42,44]. Our ChIP interaction with TEIF. Accordingly, we propose that MSP58 participates assay indicated that MSP58 associates with the hTERT promoter in the negative effect of TEIF on hTERT. (Fig. 5C). However, using an EMSA, we did not observe binding of Another aspect that should be explored is the interaction be- MSP58 to the hTERT promoter (Fig. 5B). MSP58 family proteins lack tween MSP58 and p53 or BRG1, a subunit of the SWI/SNF chromatin- 576 C.-C. Hsu et al. / Biochimica et Biophysica Acta 1843 (2014) 565–579

HT1080 stable cell lines Focus A assay HA empty HA-TEIF HA empty GFP GFP 300 HA-TEIF 250 GFP

ffoci 200 GFP-MSP58 o

er 150 αHA 100 130 kDa Numb 95 50 72 0 55 α GFP 43 HA emptyHA-TEIF GFP GFP HA-TEIF HA empty 34 HA-TEIFHA GFP-MSP58 empty GFP-MSP58 GFP-MSP58 GFP-MSP58 α hTERT

α actin lysis Soft agar buffer HA empty HA-TEIF assay GFP GFP 160

ies 140

lon 120 o TRAP 100

rofc 80

assay e

b 60 40 Num 20 0 P P 8 8 F F P5 5 G G S SP y F -M -M pt EI P P m T F F e A- G G HA H IF ty HA-TEIF HA empty TE p A- em IC H A GFP-MSP58 GFP-MSP58 H

B HA empty HA-TEIF G0/G1 S G2/M GFP GFP 100% G0/G1 : 73.27 % G0/G1 : 64.81 %

S : 6.72 % S:13.98% 80% G2/M : 20.01 % G2/M : 21.2 %

60%

DNA content 40%

HA empty HA-TEIF

Cell cycle population 20% GFP-MSP58 GFP-MSP58

G0/G1 : 50.85 % G0/G1 : 58.48 % 0% S:20.42% S:18.4% G2/M : 28.73 % G2/M : 23.12 % cell number cell number

DNA content

C HA empty , GFP Cell type of injection HA empty , GFP-MSP58 5 HA-TEIF , GFP α 1000 hTERT HA-TEIF , GFP-MSP58 4 900 α actin

) 800

3 3 lysis 700 buffer 2 600

500 Tumor weight (g) 1

400 0 8 8 300 FP 5 FP 5 TRAP G SP G SP Tumor volume (mm y -M F -M pt P EI P assay 200 m F T F e G A- G HA ty H IF p TE 100 em - HA HA 0 0 8 12 14 16 19 21 23

Days after cells injection IC C.-C. Hsu et al. / Biochimica et Biophysica Acta 1843 (2014) 565–579 577 remodeling complex, since they are also regulators of hTERT transcrip- for the interaction with this protein. Obviously, we could not exclude tion. p53 has been reported to repress hTERT expression and downreg- the possibility that MSP58 operates on the hTERT gene by partnering ulate telomerase enzymatic activity [31,32]. A mutational analysis of the with other protein(s) in a different cell context; thus, its effect on hTERT core promoter suggested a mechanism of hTERT repression in hTERT modulation is either positive or negative. In addition, by gene which p53 binds to Sp1 and renders it inaccessible to hTERT promoter expression array analysis, we identified several novel MSP58 target activation. In an additional study, Shats et al. showed that p53 induces genes known to regulate hTERT transcription, including E2F-1 and p21 to repress hTERT transcription via assembly of repressive E2F- E2F-5 (unpublished data). Interestingly, the effects of some target pocket protein complexes on an atypical E2F-binding site in the proxi- genes (such as E2F-1) are known to act as the positive or negative reg- mal core promoter [62]. Previous studies indicated that BRG1 functions ulators of hTERT transcription in different cellular systems, further as a co-repressor of E2F-targeted gene transcription by associating with supporting the view of cell type specificity regarding MSP58 in hTERT Rb [63,64]. Interestingly, it was proposed that BRG1 functions as a transcriptional regulation. Elucidation of the molecular basis of such transactivator of the hTERT gene in human tumor cell lines and pos- cell type-specific effects will require further investigation. Furthermore, itively regulates exon-inclusion of hTERT transcripts, and therefore, because up-regulation of telomerase activity is a common step in cell it contributes to maintaining active telomerase [65]. We previously immortalization, all cell lines amenable to the assays utilized here reported that the MSP58-BRG1-p53 trimeric complex is associated have significant detectable telomerase activity. We previously reported with and is transactivated by the p21 promoter, and additionally, that MSP58 expression was upregulated during replicative senescence MSP58-induced senescence is associated with activation of the p53/ in normal human fibroblasts [48]. It would be of interest to further p21 signaling pathway [48]. Deletion analyses in reporter assays im- examine the MSP58-mediated hTERT transcriptional modulation in nor- plied that the proximal region of the hTERT promoter was involved in mal somatic cells. MSP58-mediated repression (Fig. 1C). This region contains consensus Our work demonstrated that alternations (increases or decreases) in binding sites for Sp1 and E2F transcription factors [11,12,16].Itispossi- the expression of MSP58 protein modulate hTERT expression. Notably, ble that upregulation of p53/p21 by MSP58 works on the promoter of our previous observations that long-term MSP58 gene silencing in hTERT and suppresses it. MSP58 can thus play both a direct and indirect HeLa and HT1080 cells using siRNA induces apoptosis [48] are consistent role in the suppression of hTERT transcription. MSP58 might also act as a with recent reports [49,66].BecausehTERT expression is also necessary transcription modulator of E2F target genes such as hTERT by associat- for the survival of some cancer cells, it is likely that MSP58 siRNA- ing with BRG1. The combination of these upregulation and protein– induced apoptosis is at least in part due to dysregulation of hTERT tran- proteinassociationscausedbybindingofMSP58tothehTERT promoter scription. However, we cannot assume that all the cell growth effects through p53, Brg1, or TEIF may decide the transcriptional state of seen upon MSP58 depletion are due only to the dysregulation in hTERT the hTERT gene. Furthermore, previous studies implicated histone levels, since MSP58 previously has been shown to regulate cell cycle deacetylation in the transcriptional repression of the hTERT gene in nor- progression [44,67]. The observed effects of MSP58 siRNAs on tumor mal cells [38]. Of note, MSP58 can form a complex with Mi-2β, a compo- growth inhibition and apoptosis induction may be partially interpreted nent of the nucleosome-remodeling and deacetylase complex [42]. by the fact that MSP58 is also a general transcription modulator that MCRS2 was purified as part of a complex containing the histone acetyl- controls multiple biologically important gene activities, and that knock- transferase males absent on the first (MOF) in both humans and flies down of MSP58 expression by siRNA may therefore block other cell [45]. Additionally, the MSP58 protein is an integral component of proliferation-associated genes under the influence of MSP58, in addi- both the non-specific lethal (NSL) complex and the INO80 chromatin- tion to activation of the hTERT promoter. For example, MSP58 or remodeling complex [46,47]. Therefore, it is likely that MSP58-mediated MCRS2 has been implicated in regulating the expression of several repression of the hTERT promoter also occurs due to epigenetic histone cyclins [44,48,67]. In addition, a recent study showed that MSP58 is re- modifications and chromatin remodeling. Our studies have provided quired for spindle assembly and regulates the stability of kinetochore clues for elucidating the mechanisms and functional significance of spindles [68]. Our results add complexity to our knowledge of the regu- MSP58-mediated telomerase/hTERT regulation. lation of growth-related gene expression by MSP58. It should be noted that the negative regulation of hTERT transcription The mechanism underlying MSP58-mediated oncogenesis is incom- by MSP58 was observed in some but not all cell types. Two different pletely understood and its target genes essential for transformation shRNAs against MSP58 consistently induced increased levels of both await further characterization. Identifying the hTERT as a target of endogenous hTERT mRNA and protein in 293T, HT1080, and HCT116 MSP58 contributes to our understanding of how this factor modulates cell lines compared to control Luc-shRNA cells (Fig. 1A and data not cell proliferation and cancer development. In this study, we report shown). In contrast, MSP58 shRNAs suppressed levels of both hTERT that MSP58 represses telomerase activity through the inhibition of mRNA and protein in HeLa cell lines (data not shown). Thus, a proper TEIF-mediated transactivation of the hTERT promoter via direct TEIF amount of MSP58 is important for regulating hTERT expression in binding. TEIF, which was initially identified as a transactivator of the cells. The reason for this cell-type specificity is not yet known. The fact hTERT gene, can stimulate telomerase activity [53]. TEIF can also upreg- that we did not detect an obvious difference in the expression of TEIF ulate the expression of DNA polymerase-β [69]. A recent study demon- in all the examined cells (data not shown) suggests that MSP58- strated that the TEIF protein is localized on the centrosome and is mediated suppression of the hTERT promoter might not be necessary implicated in linking centrosome amplification of DNA damage and

Fig. 6. MSP58 suppresses TEIF-mediated promotion of cell proliferation in vitro and in vivo. A, HT1080 cells stably expressing the TEIF or a control empty vector were infected with ret- roviruses containing an EGFP or EGFP-MSP58 plasmid for 72 h and then split for the following assays: immunoblotting for the TEIF, MSP58, and endogenous hTERT expression (left), TRAP assay, focus-formation assays, and soft-agar anchorage-independent growth (middle). For the focus-formation assay, cells were plated at a low density on dishes. After 2 weeks, colonies were detected by staining cells with 2% methylene blue. For the soft-agar assay, cells were plated in complete growth medium containing 0.3% agarose and overlaid with 0.6% agarose. Representative plates are shown. Data represent the average number of foci or colonies ± standard deviation from three independent experiments performed in triplicate. **p b 0.01. B, MSP58 and TEIF regulate cell cycle progression. The HT1080-HA empty/GFP, HT1080-HA-TEIF/GFP, HT1080-HA empty/GFP-MSP58, and HT1080-HA-TEIF/GFP-MSP58 cells were harvested and stained with propidium iodide, followed by flow cytometry. Each graph depicts fluorescent intensity versus cell count for about 20,000 total events. The percentages of cells in G1, S, and G2/M phases are listed within each graph. Bar plots summarize the percentages of cells in the G1, S, and G2/M phases. C, the HT1080-HA empty/GFP, HT1080-HA-TEIF/GFP, HT1080-HA empty/GFP-MSP58, and HT1080-HA-TEIF/GFP-MSP58 cells were injected subcutaneously into the back of SCID mice to initiate tumor formation (5 mice per group). Tumor growth curves were determined by measuring the tumor volume every two days from Day 8 after injection to Day 23 (left). Tumors were excised and photographed 24 days after injection. Tumor weight of mice 24 days after injection (middle). Data are mean ± SD of 40 samples from two experiments. *p b 0.05; **p b 0.01. The established HT1080 xenograft tumors were collected and processed for Western blot and telomerase activity (right). The TEIF, MSP58 and hTERT levels were determined by Western blot and equal loading was controlled by a corresponding actin blot. The telomerase activity was measured for 125 ng protein extract per lane using the TRAP assay. CHAPS lysis buffer alone served as a negative control. IC, internal control. 578 C.-C. Hsu et al. / Biochimica et Biophysica Acta 1843 (2014) 565–579 telomere dysfunction in cancer development [70]. Interestingly, a previ- Appendix A. Supplementary data ous study reported that p78, an isoform of MSP58, interacts and colocal- izes with the Nde1, Su48, and δ-interacting protein A centrosomal Supplementary data to this article can be found online at http://dx. proteins, and may have a role in the maintenance of centrosome doi.org/10.1016/j.bbamcr.2013.12.004. homeostasis [49,50]. Similarly, we also observed that experimentally induced MSP58 depletion, like TEIF depletion [70], induced centrosome amplification and aneuploidy [48]. The concurrence of centrosome References abnormalities and telomere dysfunction plays an important role in generating chromosomal instability that is characteristically found in [1] C.W. Greider, E.H. Blackburn, Identification of a specific telomere terminal transfer- – neoplasia [71], but little is known about the mechanism linking these ase activity in Tetrahymena extracts, Cell 43 (1985) 405 413. [2] K. Collins, J.R. Mitchell, Telomerase in the human organism, Oncogene 21 (2002) two distinct cellular events. It is quite possible that the ways MSP58 564–579. exerts an inhibitory effect on cell growth are complex, through inter- [3] V. Urquidi, D. Tarin, S. Goodison, Role of telomerase in cell senescence and oncogen- – acting with TEIF, may be prompted by a series of interchanges includ- esis, Annu. Rev. Med. 51 (2000) 65 79. [4] N.W. Kim, M.A. Piatyszek, K.R. Prowse, C.B. Harley, M.D. West, P.L. Ho, G.M. Coviello, ing telomerase inhibition, telomere dysfunction, and centrosome W.E. Wright, S.L. Weinrich, J.W. Shay, Specific association of human telomerase abnormalities. activity with immortal cells and cancer, Science 266 (1994) 2011–2015. We previously proposed that MSP58 has both oncogenic and tumor- [5] C.M. Counter, A.A. Avilion, C.E. LeFeuvre, N.G. Stewart, C.W. Greider, C.B. Harley, S. Bacchetti, Telomere shortening associated with chromosome instability is arrested suppressive properties [48]. Our previous study showed that MSP58- in immortal cells which express telomerase activity, EMBO J. 11 (1992) 1921–1929. driven oncogene induced senescence through the DNA damage re- [6] J.W. Shay, Y. Zou, E. Hiyama, W.E. Wright, Telomerase and cancer, Hum. Mol. Genet. sponse pathway [48]. Combined with our current observations, this 10 (2001) 677–685. [7] J. Nakayama, H. Tahara, E. Tahara, M. Saito, K. Ito, H. Nakamura, T. Nakanishi, T. Ide, F. raises the possibility that by inhibiting hTERT,MSP58favorsanun- Ishikawa, Telomerase activation by hTRT in human normal fibroblasts and hepato- stable genomic state at early stages of transformation, which proba- cellular carcinomas, Nat. Genet. 18 (1998) 65–68. bly drives a mutagenic effect in favor of the development of a fully [8] C.M. Counter, W.C. Hahn, W. Wei, S.D. Caddle, R.L. Beijersbergen, P.M. Lansdorp, J.M. Sedivy, R.A. Weinberg, Dissociation among in vitro telomerase activity, telomere transformed phenotype. Alternatively, MSP58-triggered inhibition maintenance, and cellular immortalization, Proc. Natl. Acad. Sci. U. S. A. 95 (1998) of telomerase/hTERT may reflect potential anti-tumor activity of 14723–14728. MSP58. Overexpression of MSP58 increases S/G2M phases and [9] S. Kondo, Y. Tanaka, Y. Kondo, M. Hitomi, G.H. Barnett, Y. Ishizaka, J. Liu, T. Haqqi, A. decreases tumorigenicity of HT1080 cells. MSP58 might have other, Nishiyama, B. Villeponteau, J.K. Cowell, B.P. Barna, Antisense telomerase treatment: induction of two distinct pathways, apoptosis and differentiation, FASEB J. 12 as yet unidentified functions that contribute to its tumor suppressing (1998) 801–811. effect. Various levels of TEIF expression have been reported in many [10] S. Kyo, M. Takakura, T. Fujiwara, M. Inoue, Understanding and exploiting hTERT pro- tumor types [72,73]. MSP58 may act, in part, by regulating the onco- moter regulation for diagnosis and treatment of human cancers, Cancer Sci. 99 (2008) 1528–1538. genic potential of TEIF through hTERT transcriptional inhibition. Our [11] M. Takakura, S. Kyo, T. Kanaya, H. Hirano, J. Takeda, M. Yutsudo, M. Inoue, Cloning of findings provide insight into one newly identified mechanism of human telomerase catalytic subunit (hTERT) gene promoter and identification of telomerase/hTERT modulation by pointing to MSP58 as a potential proximal core promoter sequences essential for transcriptional activation in immor- talized and cancer cells, Cancer Res. 59 (1999) 551–557. repressor, thereby facilitating our understanding of the role of [12] I. Horikawa, P.L. Cable, C. Afshari, J.C. Barrett, Cloning and characterization of the MSP58 in tumorigenesis. promoter region of human telomerase reverse transcriptase gene, Cancer Res. 59 (1999) 826–830. [13] R.A. Greenberg, R.C. O'Hagan, H. Deng, Q. Xiao, S.R. Hann, R.R. Adams, S. Lichtsteiner, L. Chin, G.B. Morin, R.A. DePinho, Telomerase reverse transcriptase gene is a direct 5. Conclusion target of c-Myc but is not functionally equivalent in cellular transformation, Oncogene 18 (1999) 1219–1226. [14] J. Wang, L.Y. Xie, S. Allan, D. Beach, G.J. Hannon, Myc activates telomerase, Genes We show in this report that MSP58 protein can function as a nega- Dev. 12 (1998) 1769–1774. tive regulator of hTERT transcription, and that interaction between [15] K.J. Wu, C. Grandori, M. Amacker, N. Simon-Vermot, A. Polack, J. Lingner, R. MSP58 and TEIF may be involved in this regulation. MSP58 inhibits Dalla-Favera, Direct activation of TERT transcription by c-MYC, Nat. Genet. 21 – TEIF-mediated hTERT transactivation, telomerase activation, and cell (1999) 220 224. [16] D.L. Crowe, D.C. Nguyen, K.J. Tsang, S. Kyo, E2F-1 represses transcription of the growth-promoting activity in vitro and in vivo. Identification of the human telomerase reverse transcriptase gene, Nucleic Acids Res. 29 (2001) function of MSP58 in telomerase/hTERT regulation provides new 2789–2794. insights for understanding the role of MSP58 in carcinogenesis. [17] B.S. Goueli, R. Janknecht, Regulation of telomerase reverse transcriptase gene activity by upstream stimulatory factor, Oncogene 22 (2003) 8042–8047. [18] H. Nishi, T. Nakada, S. Kyo, M. Inoue, J.W. Shay, K. Isaka, Hypoxia-inducible factor 1 mediates upregulation of telomerase (hTERT), Mol. Cell. Biol. 24 (2004) 6076–6083. Acknowledgements [19] Y. Zhang, L. Toh, P. Lau, X. Wang, Human telomerase reverse transcriptase (hTERT) is a novel target of the Wnt/beta-catenin pathway in human cancer, J. Biol. Chem. 287 (2012) 32494–32511. We thank Dr. Kou-Juey Wu for the luciferase reporter plasmid [20] M. Bellon, C. Nicot, Central role of PI3K in transcriptional activation of hTERT in TERTLuc800 and Dr. Yu-Ten Ju for the hTERT retrovirus vector pLPC- HTLV-I-infected cells, Blood 112 (2008) 2946–2955. [21] A. Chebel, J.P. Rouault, I. Urbanowicz, L. Baseggio, W.W. Chien, G. Salles, M. Ffrench, hTERT. We also thank Drs. Ming-Derg Lai, Chi-Wu Chiang, Ju-Ming Transcriptional activation of hTERT, the human telomerase reverse transcriptase, by Wang, Tsunglin Liu and Joseph T. Tseng for helpful discussion, Chia- nuclear factor of activated T cells, J. Biol. Chem. 284 (2009) 35725–35734. Yang Hung, Shin-jin Lin and Yi-Chang Wang for assistance during ani- [22] A.Y. Li, H.H. Lin, C.Y. Kuo, H.M. Shih, C.C. Wang, Y. Yen, D.K. Ann, High-mobility group A2 protein modulates hTERT transcription to promote tumorigenesis, Mol. mal experiments. The shRNA lentiviral vectors were obtained from the Cell. Biol. 31 (2011) 2605–2617. National RNAi Core Facility (The Institute of Molecular Biology/Genomic [23] G.P. Dimri, J.L. Martinez, J.J. Jacobs, P. Keblusek, K. Itahana, M. Van Lohuizen, J. Research Center, Academia Sinica, Taiwan). We also thank the Center Campisi, D.E. Wazer, V. Band, The Bmi-1 oncogene induces telomerase activity and – for Translational Research in Biomedical Sciences, Kaohsiung Chang immortalizes human mammary epithelial cells, Cancer Res. 62 (2002) 4736 4745. [24] A.J. Klingelhutz, S.A. Foster, J.K. McDougall, Telomerase activation by the E6 gene Gung Memorial Hospital, for providing the instruments used for this product of human papillomavirus type 16, Nature 380 (1996) 79–82. study (CLRPG871342-3). This work was supported by grants from the [25] S. Oh, Y. Song, J. Yim, T.K. Kim, The Wilms' tumor 1 tumor suppressor gene represses National Science Council of Taiwan NSC98-2320-B-006-035-MY3 (to transcription of the human telomerase reverse transcriptase gene, J. Biol. Chem. 274 (1999) 37473–37478. D.-Y. L.), NSC101-2320-B-038-022 (to W.-C. C.), NSC 100-2320-B- [26] C. Gunes, S. Lichtsteiner, A.P. Vasserot, C. Englert, Expression of the hTERT gene is 182A-001 (NMRPG8A0011) (to C.-H. C.), National Cheng Kung Univer- regulated at the level of transcriptional initiation and repressed by Mad1, Cancer sity landmark grant (C015) (to D.-Y. L.), and Kaohsiung Chang Gung Res. 60 (2000) 2116–2121. [27] K. Fujimoto, S. Kyo, M. Takakura, T. Kanaya, Y. Kitagawa, H. Itoh, M. Takahashi, M. Memorial Hospital CMRPG8A0391, CMRPG8A0392, CMRPG890921, Inoue, Identification and characterization of negative regulatory elements of the CMRPG8B1251 (to C.-H. C.). human telomerase catalytic subunit (hTERT) gene promoter: possible role of C.-C. Hsu et al. / Biochimica et Biophysica Acta 1843 (2014) 565–579 579

MZF-2 in transcriptional repression of hTERT, Nucleic Acids Res. 28 (2000) [50] X. Du, Q. Wang, Y. Hirohashi, M.I. Greene, DIPA, which can localize to the 2557–2562. centrosome, associates with p78/MCRS1/MSP58 and acts as a repressor of gene [28] M. Xu, W. Luo, D.J. Elzi, C. Grandori, D.A. Galloway, NFX1 interacts with transcription, Exp. Mol. Pathol. 81 (2006) 184–190. mSin3A/histone deacetylase to repress hTERT transcription in keratinocytes, Mol. [51] A.G. Bader, M.L. Schneider, K. Bister, M. Hartl, TOJ3, a target of the v-Jun transcription Cell. Biol. 28 (2008) 4819–4828. factor, encodes a protein with transforming activity related to human microspherule [29] H. Li, D. Xu, J. Li, M.C. Berndt, J.P. Liu, Transforming growth factor beta suppresses protein 1 (MCRS1), Oncogene 20 (2001) 7524–7535. human telomerase reverse transcriptase (hTERT) by Smad3 interactions with [52] K. Okumura, M. Zhao, R.A. Depinho, F.B. Furnari, W.K. Cavenee, Cellular transforma- c-Myc and the hTERT gene, J. Biol. Chem. 281 (2006) 25588–25600. tion by the MSP58 oncogene is inhibited by its physical interaction with the PTEN [30] M. Hashimoto, S. Kyo, X. Hua, H. Tahara, M. Nakajima, M. Takakura, J. Sakaguchi, Y. tumor suppressor, Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 2703–2706. Maida, M. Nakamura, T. Ikoma, Y. Mizumoto, M. Inoue, Role of menin in the regulation [53] Z. Tang, Y. Zhao, F. Mei, S. Yang, X. Li, J. Lv, L. Hou, B. Zhang, Molecular cloning and oftelomeraseactivityinnormalandcancercells,Int.J.Oncol.33(2008)333–340. characterization of a human gene involved in transcriptional regulation of hTERT, [31] D. Xu, Q. Wang, A. Gruber, M. Bjorkholm, Z. Chen, A. Zaid, G. Selivanova, C. Peterson, Biochem. Biophys. Res. Commun. 324 (2004) 1324–1332. K.G. Wiman, P. Pisa, Downregulation of telomerase reverse transcriptase mRNA [54] F. Mei, B. Zhang, Z.W. Tang, L. Hou, Expression of a telomerase-associated gene in expression by wild type p53 in human tumor cells, Oncogene 19 (2000) 5123–5133. normal, atrophic, and tumorous testes, Chin. Med. Sci. J. 20 (2005) 217–220. [32] T. Kanaya, S. Kyo, K. Hamada, M. Takakura, Y. Kitagawa, H. Harada, M. Inoue, Adeno- [55] C.H. Liao, Y.M. Hsiao, C.P. Hsu, M.Y. Lin, J.C. Wang, Y.L. Huang, J.L. Ko, Transcription- viral expression of p53 represses telomerase activity through down-regulation of ally mediated inhibition of telomerase of fungal immunomodulatory protein from human telomerase reverse transcriptase transcription, Clin. Cancer. Res. 6 (2000) Ganoderma tsugae in A549 human lung adenocarcinoma cell line, Mol. Carcinog. 1239–1247. 45 (2006) 220–229. [33] S. Bayne, J.P. Liu, Hormones and growth factors regulate telomerase activity in [56] D.Y. Lin, H.I. Fang, A.H. Ma, Y.S. Huang, Y.S. Pu, G. Jenster, H.J. Kung, H.M. Shih, ageing and cancer, Mol. Cell. Endocrinol. 240 (2005) 11–22. Negative modulation of androgen receptor transcriptional activity by Daxx, Mol. [34] S. Misiti, S. Nanni, G. Fontemaggi, Y.S. Cong, J. Wen, H.W. Hirte, G. Piaggio, A. Sacchi, Cell. Biol. 24 (2004) 10529–10541. A. Pontecorvi, S. Bacchetti, A. Farsetti, Induction of hTERT expression and telomerase [57] W.E. Wright, J.W. Shay, M.A. Piatyszek, Modifications of a telomeric repeat amplifi- activity by estrogens in human ovary epithelium cells, Mol. Cell. Biol. 20 (2000) cation protocol (TRAP) result in increased reliability, linearity and sensitivity, 3764–3771. Nucleic Acids Res. 23 (1995) 3794–3795. [35] R. Hrdlickova, J. Nehyba, H.R. Bose Jr., Regulation of telomerase activity by interferon [58] M.L. Falchetti, A. Levi, P. Molinari, R. Verna, E. D'Ambrosio, Increased sensitivity and regulatory factors 4 and 8 in immune cells, Mol. Cell. Biol. 29 (2009) 929–941. reproducibility of TRAP assay by avoiding direct primers interaction, Nucleic Acids [36] J. Won, J. Yim, T.K. Kim, Opposing regulatory roles of E2F in human telomerase Res. 26 (1998) 862–863. reverse transcriptase (hTERT) gene expression in human tumor and normal somatic [59] J.M. Wong, L. Kusdra, K. Collins, Subnuclear shuttling of human telomerase induced cells, FASEB J. 16 (2002) 1943–1945. by transformation and DNA damage, Nat. Cell Biol. 4 (2002) 731–736. [37] I. Guilleret, P. Yan, F. Grange, R. Braunschweig, F.T. Bosman, J. Benhattar, Hyperme- [60] X. Zhang, V. Mar, W. Zhou, L. Harrington, M.O. Robinson, Telomere shortening and thylation of the human telomerase catalytic subunit (hTERT) gene correlates with apoptosis in telomerase-inhibited human tumor cells, Genes Dev. 13 (1999) telomerase activity, Int. J. Cancer 101 (2002) 335–341. 2388–2399. [38] Y.S. Cong, S. Bacchetti, Histone deacetylation is involved in the transcriptional [61] W.C. Hahn, S.A. Stewart, M.W. Brooks, S.G. York, E. Eaton, A. Kurachi, R.L. repression of hTERT in normal human cells, J. Biol. Chem. 275 (2000) 35665–35668. Beijersbergen, J.H. Knoll, M. Meyerson, R.A. Weinberg, Inhibition of telomerase [39] Y. Ren, R.K. Busch, L. Perlaky, H. Busch, The 58-kDa microspherule protein (MSP58), limits the growth of human cancer cells, Nat. Med. 5 (1999) 1164–1170. a nucleolar protein, interacts with nucleolar protein p120, Eur. J. Biochem. 253 [62] I. Shats, M. Milyavsky, X. Tang, P. Stambolsky, N. Erez, R. Brosh, I. Kogan, I. Braunstein, (1998) 734–742. M. Tzukerman, D. Ginsberg, V. Rotter, p53-Dependent down-regulation of telome- [40] D.Y. Lin, H.M. Shih, Essential role of the 58-kDa microspherule protein in the rase is mediated by p21waf1, J. Biol. Chem. 279 (2004) 50976–50985. modulation of Daxx-dependent transcriptional repression as revealed by nucleolar [63] J.L. Dunaief, B.E. Strober, S. Guha, P.A. Khavari, K. Alin, J. Luban, M. Begemann, G.R. sequestration, J. Biol. Chem. 277 (2002) 25446–25456. Crabtree, S.P. Goff, The retinoblastoma protein and BRG1 form a complex and [41] A.V. Ivanova, S.V. Ivanov, M.L. Lerman, Association, mutual stabilization, and cooperate to induce cell cycle arrest, Cell 79 (1994) 119–130. transcriptional activity of the STRA13 and MSP58 proteins, Cell. Mol. Life Sci. 62 [64] H.S. Zhang, M. Gavin, A. Dahiya, A.A. Postigo, D. Ma, R.X. Luo, J.W. Harbour, D.C. (2005) 471–484. Dean, Exit from G1 and S phase of the cell cycle is regulated by repressor complexes [42] K. Shimono, Y. Shimono, K. Shimokata, N. Ishiguro, M. Takahashi, Microspherule containing HDAC-Rb-hSWI/SNF and Rb-hSWI/SNF, Cell 101 (2000) 79–89. protein 1, Mi-2beta, and RET finger protein associate in the nucleolus and [65] T. Ito, H. Watanabe, N. Yamamichi, S. Kondo, T. Tando, T. Haraguchi, T. Mizutani, K. up-regulate ribosomal gene transcription, J. Biol. Chem. 280 (2005) 39436–39447. Sakurai, S. Fujita, T. Izumi, T. Isobe, H. Iba, Brm transactivates the telomerase reverse [43] H. Song, Y. Li, G. Chen, Z. Xing, J. Zhao, K.K. Yokoyama, T. Li, M. Zhao, Human MCRS2, transcriptase (TERT) gene and modulates the splicing patterns of its transcripts in a cell-cycle-dependent protein, associates with LPTS/PinX1 and reduces the concert with p54(nrb), Biochem. J. 411 (2008) 201–209. telomere length, Biochem. Biophys. Res. Commun. 316 (2004) 1116–1123. [66] Y. Liang, M. Liu, P. Wang, X. Ding, Y. Cao, Analysis of 20 genes at chromosome band [44] D.S. Andersen, S.J. Raja, J. Colombani, R.L. Shaw, P.F. Langton, A. Akhtar, N. Tapon, 12q13: RACGAP1 and MCRS1 overexpression in nonsmall-cell lung cancer, Genes Drosophila MCRS2 associates with RNA polymerase II complexes to regulate tran- Chromosomes Cancer 52 (2013) 305–315. scription, Mol. Cell. Biol. 30 (2010) 4744–4755. [67] M. Benavides, L.F. Chow-Tsang, J. Zhang, H. Zhong, The novel interaction between [45] S.J. Raja, I. Charapitsa, T. Conrad, J.M. Vaquerizas, P. Gebhardt, H. Holz, J. Kadlec, S. microspherule protein Msp58 and ubiquitin E3 ligase EDD regulates cell cycle pro- Fraterman, N.M. Luscombe, A. Akhtar, The nonspecific lethal complex is a transcrip- gression, Biochim. Biophys. Acta 1833 (2013) 21–32. tional regulator in Drosophila,Mol.Cell38(2010)827–841. [68] S. Meunier, I. Vernos, K-fibre minus ends are stabilized by a RanGTP-dependent [46] L. Chen, Y. Cai, J. Jin, L. Florens, S.K. Swanson, M.P. Washburn, J.W. Conaway, R.C. mechanism essential for functional spindle assembly, Nat. Cell Biol. 13 (2011) Conaway, Subunit organization of the human INO80 chromatin remodeling com- 1406–1414. plex: an evolutionarily conserved core complex catalyzes ATP-dependent nucleo- [69] Y. Zhao, J. Zheng, Y. Ling, L. Hou, B. Zhang, Transcriptional upregulation of some remodeling, J. Biol. Chem. 286 (2011) 11283–11289. DNA polymerase beta by TEIF, Biochem. Biophys. Res. Commun. 333 (2005) [47] Y. Cai, J. Jin, S.K. Swanson, M.D. Cole, S.H. Choi, L. Florens, M.P. Washburn, J.W. 908–916. Conaway, R.C. Conaway, Subunit composition and substrate specificity of a [70] Y. Gong, Y. Sun, M.A. McNutt, Q. Sun, L. Hou, H. Liu, Q. Shen, Y. Ling, Y. Chi, B. Zhang, MOF-containing histone acetyltransferase distinct from the male-specific lethal Localization of TEIF in the centrosome and its functional association with centro- (MSL) complex, J. Biol. Chem. 285 (2010) 4268–4272. some amplification in DNA damage, telomere dysfunction and human cancers, [48] C.C. Hsu, Y.C. Lee, S.H. Yeh, C.H. Chen, C.C. Wu, T.Y. Wang, Y.N. Chen, L.Y. Hung, Y.W. Oncogene 28 (2009) 1549–1560. Liu, H.K. Chen, Y.T. Hsiao, W.S. Wang, J.H. Tsou, Y.H. Tsou, M.H. Wu, W.C. Chang, D.Y. [71] D. Gisselsson, Chromosome instability in cancer: how, when, and why? Adv. Cancer Lin, 58-kDa microspherule protein (MSP58) is novel Brahma-related gene 1 Res. 87 (2003) 1–29. (BRG1)-associated protein that modulates p53/p21 senescence pathway, J. Biol. [72] Y.L. Gong, T. Li, H. Guo, Y. Sun, Y.K. Chi, Y. Ling, Q. Shen, H.J. Liu, L. Hou, B. Zhang, Chem. 287 (2012) 22533–22548. Expression of TEIF protein in soft tissue tumors and its significance, Zhonghua [49] Y. Hirohashi, Q. Wang, Q. Liu, X. Du, H. Zhang, N. Sato, M.I. Greene, p78/MCRS1 forms Bing Li Xue Za Zhi 35 (2006) 651–655. a complex with centrosomal protein Nde1 and is essential for cell viability, [73] Y. Gao, B. Zhang, Expression of TEIF protein in colorectal tumors and its correlation Oncogene 25 (2006) 4937–4946. with centrosome abnormality, Ai Zheng 28 (2009) 1277–1282.