PITX1 protein interacts with ZCCHC10 to regulate hTERT mRNA transcription

Telomerase is a ribonucleoprotein ribonucleic enzyme that is essential for cellular immortalization via elongation of telomere repeat sequences at the end of chromosomes. Human telomerase reverse transcriptase (hTERT), the catalytic subunit of telomerase holoenzyme, is a key regulator of telomerase activity. Telomerase activity, which has been detected in the majority of cancer cells, is accompanied by hTERT expression, suggesting that this enzyme activity contributes to an unlimited replication potential of cancer cells via regulation of telomere length. Thus, hTERT is an attractive target for cancer-specific treatments. We previously reported that pared-like homeodomain 1 (PITX1) is a negative regulator of hTERT through direct binding to the hTERT promoter. However, the mechanism by which the function of PITX1 contributes to transcriptional silencing of the hTERT gene remains to be clarified. Here, we show that PITX1 and zinc finger CCHC-type containing 10 (ZCCHC10) proteins cooperate to facilitate the transcriptional regulation of the hTERT gene by functional studies via FLAG pull-down assay. Co-expression of PITX1 and ZCCHC10 resulted in inhibition of hTERT transcription, in melanoma cell lines, whereas mutate-deletion of homeodomain in PITX1 that interact with ZCCHC10 did not induce similar phenotypes. In addition, ZCCHC10 expression levels showed marked decrease in the majority of melanoma cell lines and tissues. Taken together, these results suggest that ZCCHC10-PITX1 complex is the functional unit that suppresses hTERT transcription, and may play a crucial role as a novel tumor suppressor complex.


Introduction
Telomerase ribonucleic enzyme is associated with extend cell life span by elongation of telomere repeat sequences on the end of chromosomes, and sustain cell proliferation in cancer PLOS ONE | https://doi.org/10.1371/journal.pone.0217605 August 12, 2019 1 / 21 a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 cells [1,2]. Human telomerase consists of essential enzyme subunits; the protein catalytic subunit human telomerase reverse transcriptase (hTERT) and the RNA subunit, human telomerase RNA component (hTERC) and the accessory proteins dyskerin, NOP10, NHP2 and GAR1 [3,4]. It is important to note that hTERT molecules are very low rather than other telomerase components in cancer cell lines. This is suggested that hTERT transcription is tightly regulated more than other telomerase components [5]. The expression of hTERT is critical for telomerase enzyme activity. Indeed, ectopic hTERT expression in telomerase-negative normal cells can extend lifespan and establish immortalized cell lines via elongation of telomeres [6,7]. Expression of hTERT is down-regulated in most human adult somatic cells, except in germ cells and some stem cells. On the other hand, its expression was detected in the majority of human cancer cells (around 85-90%) [8,9]. This is consistent with telomerase conferring a strong selective advantage for continued growth of malignant cells, suggesting that telomerase activity is essential for most cancer cell immortalization and it may be possible to inhibit of cancer development by the control of hTERT expression. Furthermore, hTERT has noncanonical functions in addition to that of maintaining telomere length. It was reported that hTERT acts as a transcriptional modulator of Wnt/beta-catenin and NF-kappa B signaling pathways, resulting in the enhanced expression of Wnt and NF-kappa B target genes that facilitate cancer promoting functions such as proliferation and resistance to apoptosis [10,11]. Additionally, hTERT protein directly associates with the RNA polymerase III subunit RPC32, which restore tRNA levels and promote cell metabolism and proliferation in cancer cells [12]. Although it is known that expression of hTERT is regulated by various activating and repressing transcription factors and epigenetic modification [13,14], the underlying molecular mechanisms that are involved in regulation of hTERT transcription during cellular differentiation and cancer development remains unclear. We previously confirmed that human chromosomes 3, 5, and 10 carry hTERT regulatory genes using microcell mediated chromosome transfer (MMCT) [15]. In particular, we identified paired-like homeodomain 1 (PITX1) as a novel hTERT suppressor gene, located on human chromosome 5 by a combination of MMCT and gene expression profile analysis. PITX1 regulates hTERT transcription through binding to its promoter [16,17]. PITX1 was originally identified as a transcription factor gene that is able to activate pituitary transcription of a pro-opiomelanocortin gene. PITX1, which belongs to bicoid-related homeobox genes, plays a role in the development of the Rathke pouch and adult pituitary [18], and consists of an N-terminal homeodomain and a C-terminal Otp-Aristaless-Rax (OAR) domain. The homeodomain has DNA binding function and also acts as a protein-protein interaction for target molecule [19,20]. On the other hand, the OAR domain may modulate transactivation capacities or be involved in protein-protein interactions [21,22]. Pitx1 knockout mice developed fetuses with abnormal hindlimbs, thus suggesting that it regulates the developmental limb [23]. In addition, PITX1 is known as a tumor suppressor gene that inhibits the RAS pathway through Ras protein activator-like 1 (RASAL1), which is a member of the Ras GTPase-activating protein family [24] and induces activation of p53 transcription [25]. Furthermore, we provided important evidence that PITX1 directly binds to specific PITX1 response element sites in the hTERT promoter region, resulting in telomerase inhibition [17]. Downregulation of PITX1 is observed in various cancers including malignant melanoma, oral, gastric, colon, lung, and bladder cancers [24,[26][27][28][29][30]. Collectively, this evidence suggests that PITX1 plays a crucial role in cancer development, though telomerase-dependent pathways. Interestingly, the introduction of an intact human chromosome 5 into melanoma A2058 cells more strongly suppressed hTERT transcription when compared with PITX1 cDNA-overexpressing clones [16,31]. Therefore, human chromosome 5 carries one or more genes that are involved in the suppression of hTERT transcription, in addition to the PITX1 gene.
The zinc fingers Lys-Cys-His-Cys-type 10 (ZCCHC10) gene belongs to the CCHC-type zinc finger nucleic acid binding protein family. DNA methylation level of CpG site in ZCCHC10 gene using offspring cord blood DNA was showed potentially ZCCHC10 related to apoptosis, tumorigenesis and inflammation pathways [32]. In addition, ZCCHC10 protein level is downregulated in atopic dermatitis patients-derived serum [33]. However, the functional role of ZCCHC10 gene in tumorigenesis is poorly understood. ZCCHC10 contains a single CCHC domain, which is known that mediate protein-protein interactions. For example, zinc finger protein FOG family member 1 (FOG1) contains five CCHC domain and each CCHC domain bind to GATA bunding protein 1 (GATA1) [34]. FOG1-GATA1 complexes function as activators for several genes, which required for normal erythroid differentiation and megakaryocyte maturation [35]. Taken together, these previously studies suggest that the CCHC domain in ZCCHC10 have the potential to interact with other biomolecules containing transcription factors and the complex may have a function of regulation for tumorigenesis related genes.
In this study, we identified ZCCHC10 as a protein that interacted with PITX1, and interaction sites were mapped to the CCHC domain of ZCCHC10 and the homeodomain of PITX1. Moreover, expression levels of ZCCHC10 in human melanoma cells were significantly lower when compared with normal melanocyte cells. In addition, overexpression of both the PITX1 and ZCCHC10 genes showed significant suppression of hTERT transcription when compared to that of each gene in melanoma cell lines.

Identification of novel proteins that formed complex with PITX1 in hTERT-negative cells
Cellular immortalization of all human tumors requires maintenance of telomere length, which needs reactivation of telomerase enzyme (telomerase dependent pathways) or activation of telomerase independent Alternative Lengthening of Telomere (ALT) pathways. We propose that functional PITX1 complexes that regulate hTERT transcription are present in telomerase-negative tumor cells in which hTERT transcription is not detectable, but not in telomerase-positive tumor cells. To determine protein interactions with PITX1, we performed pull-down assay using hTERT-positive HeLa229, 293T cells [36,37] and hTERT-negative ALT U2OS [38] cells with FLAG-tagged PITX1. We used control cells that expressed only FLAG-tag protein ( Fig 1A). First, FLAG-tagged PITX1 was transfected into HeLa229, 293T and U2OS cells, respectively. Enrichment of FLAG-tagged PITX1 protein in cell lysates obtained by immunoprecipitation with anti-FLAG antibody was confirmed using western blotting analysis ( Fig  1B). A band at 49 kDa corresponding to FLAG-tagged PITX1 was detected in these cancer cells, but not in control cells, on Coomassie brilliant blue (CBB) and silver stained gels ( Fig  1C). nanoLC/MS/MS-based proteomic analysis of FLAG immunoprecipitation extracts obtained from these cell lines yielded 27 (HeLa229), 34 (293T), and 101 (U2OS) proteins, respectively ( Fig 1D and S1 Table). Ultimately, 57 proteins were identified only in hTERT-negative U2OS cells (Fig 1D). Therefore, it is likely that PITX1-interacting proteins function as suppressors of hTERT transcription.

Validation of PITX1-interacting proteins
We previously reported that microcell hybrid clones obtained by the introduction of human chromosome 5 completely suppressed TERT transcription [16]. On the other hand, the suppression effects of hTERT transcription in transfectant cells that were overexpressed by introduction of PITX1 cDNA were weak when compared to those in microcell hybrid cells with human chromosome 5 in A2058 human melanoma cells [16,31]. This suggests that PITX1 interacts with other proteins on human chromosome 5 to form a functional complex essential for suppression of hTERT transcription. Therefore, we focused on genes located on chromosome 5 as novel PITX1 interacting proteins. As a result, among 57 candidate genes, six genes localized on human chromosome 5 were identified as PITX1-interacting proteins in hTERTnegative U2OS cells (Table 1 and S1 Fig). Interestingly, three of these genes, namely, heterogeneous nuclear ribonucleoprotein A0 (hnRNP A0), ZCCHC10 and 75 kDa glucose-regulated protein (HSPA9), were localized in the peripheral PITX1 region (5q31.1). We first aimed to identity PITX1-interacting proteins using the Human Protein Atlas (HPA) database, which includes information on gene expression in normal and cancer cells and tissues (http://www. proteinatlas.org/). Among these, ZCCHC10 showed lower expression in melanoma tissue   samples when compared with normal melanocyte cells. To further validate the authenticity of the PITX1-ZCCHC10 complex indicated by nanoLC/MS/MS analysis, we investigated whether ZCCHC10 is present in the PITX1 immunocomplex in FLAG-PITX1-expressing U2OS cells by western blotting analysis using anti-ZCCHC10 antibody. Western blotting detected bands suggesting that ZCCHC10 protein forms a complex with FLAG-PITX1 protein (Fig 2). Thus, ZCCHC10 directly or indirectly interacts with PITX1.

Expression profiles of ZCCHC10 expression level in melanoma cell lines and tissues
We next performed western blotting analysis to investigate the expression profiles of ZCCHC10 in human melanoma cell lines. The expression levels of ZCCHC10 proteins were significantly reduced in all six melanoma cell lines (A2058, SKMEKL28, G361, HMV1, CRL1597 and GAK) when compared to normal human epidermal melanocyte cells (Fig 3A  and 3B). To further explore the relationship between ZCCHC10 and hTERT gene, we performed qRT-PCR analysis to determine hTERT mRNA expression level in melanoma cell lines (S2 Fig). Interestingly, ZCCHC10 protein levels negatively correlated with hTERT mRNA levels in melanoma cell lines ( Fig 3C). Furthermore, qRT-PCR analysis indicated that ZCCHC10 mRNA expression was markedly decreased in 11 out of 12 human clinical melanoma  specimens when compared with normal melanocyte cells ( Fig 3D). Moreover, our previous data demonstrated that PITX1 expression is also down-regulated in melanomas [26]. These data suggest that both PITX1 and ZCCHC10 play an important role as tumor suppressor genes and the expression profiles of those genes are involved in melanoma development through dysregulation of hTERT transcription.

ZCCHC10-PITX1 complex suppress hTERT expression level
In order to investigate the suppression effects of hTERT transcription by the PITX1-ZCCHC10 complex, we transiently co-transfected A2058 cells with expression vectors containing the PITX1 cDNA and ZCCHC10 cDNA sequences. Transient overexpression of PITX1 or ZCCHC10 in A2058 cells reduced hTERT transcription to 84% or 82% of control cells, respectively. On the other hand, transient co-expression of PITX1 and ZCCHC10 vector induced a significant decrease in transcription to 59% of that in control cells ( Fig 4A). To further examine whether PITX1 and ZCCHC10 form protein complexes, we carried out immunoprecipitation of cell lysates from FLAG-PITX1 or ZCCHC10 overexpressing A2058 cells using anti-PITX1 and ZCCHC10 antibodies. Bands corresponding to PITX1 and ZCCHC10 were detected by western blotting in both pull-downs ( Fig 4B and 4C). These results suggest that PITX1 is able to form a complex with ZCCHC10, and is functionally involved in the suppression of hTERT transcription in A2058 cells.

Homeodomain of PITX1 interacts with CCHC domain of ZCCHC10
To determine the protein-protein interaction sites between ZCCHC10 and a PITX1, we performed mutation analysis. PITX1 is a transcription factor that contains a homeodomain and an OAR, which are the predicted protein-protein interaction domain [18][19][20]. First, we prepared a FLAG-tagged PITX1 deleted homeodomain 1 (half of the N-terminal side of the homeodomain deleted: FLAG-PITX1 ΔHD1), FLAG-PITX1 homeodomain 2 (half of the C-terminal side of the homeodomain deleted: FLAG-PITX1 ΔHD2) and FLAG-PITX1 deleted OAR (OAR domain deleted: FLAG-PITX1 ΔOAR) protein expression plasmid vectors ( Fig  5A). FLAG-tagged wild-type PITX1 (FLAG-PITX1 wt) or each FLAG-tagged mutation-type PITX1 vector was co-expressed with an expression vector containing the wild-type ZCCHC10 expression vector in A2058 cells. Immunoprecipitation analysis was performed using anti- (B) Total protein extracts from A2058 cells, transfected with FLAG-ZCCHC10 and PITX1 expression vectors, were subjected to immunoprecipitation with anti-FLAG M2 affinity gel followed by western blotting with anti-PITX1 and anti-ZCCHC10 antibodies. As a control, empty FLAG vector was transfected into A2058 cells. (C) Total protein extracts from A2058 cells, transfected with FLAG-PITX1 and ZCCHC10 expression vectors, were subjected to immunoprecipitation with anti-FLAG M2 affinity gel, followed by western blotting with anti-ZCCHC10 and anti-PITX1 antibodies. As a negative control, empty FLAG vector was transfected into A2058 cells. FLAG antibody. Western blotting using anti-ZCCHC10 antibody showed co-precipitated ZCCHC10 with PITX1 wt and ΔOAR, but not with ΔHD1 and ΔHD2 (Fig 5B). This indicates that the homeodomain in PITX1 is essential for forming a protein complex with ZCCHC10. On the other hand, the ZCCHC10 protein possesses the CCHC domain, which is known as a protein interaction domain. To determine the region of ZCCHC10 that interacted with PITX1, we performed a deletion analysis of the ZCCHC10 protein. We generated a FLAGtagged ZCCHC10 plasmid vector, in which the CCHC domain was deleted (FLAG-ZCCHC10 ΔCCHC) (Fig 5C). FLAG-tagged wild-type ZCCHC10 (FLAG-ZCCHC10 wt) or FLAG-ZCCHC10 ΔCCHC was co-expressed with the expression vector coding the wild-type of PITX1 expression vector in A2058 cells. Immunoprecipitation analysis was performed using anti-FLAG antibody. Western blotting using anti-PITX1 antibody showed co-precipitation of PITX1 with wild-type ZCCHC10 protein, but not with ZCCHC10 ΔCCHC. This indicates that PITX1 forms a protein complex through the CCHC domain in ZCCHC10 (Fig 5D). Furthermore, we performed qRT-PCR analysis to investigate the effects of PITX1 and ZCCHC10 protein interaction regions on hTERT transcription. Expression levels of hTERT mRNA in A2058 cells co-transfected with ZCCHC10 wt and PITX1 wt expressing vectors were markedly decreased when compared to empty vector-transfected control cells (reduced hTERT transcription to 71% of control cells with empty vector). On the other hand, co-transfection to ZCCHC ΔCCHC with PITX1 wt vector in A2058 cells had no effect on hTERT transcription when compared to that in control cells (Fig 5E). These results suggest that interaction between PITX1 and ZCCHC10 proteins is mediated by CCHC on ZCCHC10 and that the homeodomain on PITX1 is essential for formation of a functional complex that regulates hTERT transcription.

Overexpression of both PITX1 and ZCCHC10 lead to inhibition of cell proliferation
To further explore the effects of ZCCHC10-PITX1 complex on hTERT transcription, we generated two A2058 cell lines that stably co-overexpressed FLAG-tagged ZCCHC10 and PITX1 (cl.1 and cl.2) or control clones that transfected empty vector (control cl.1 and cl.2). Significantly increased protein expression level of ZCCHC10 and PITX1 transfected cells compared to control transfected cells was confirmed using western blotting analysis (Fig 6A). The expression of hTERT mRNA transcription significantly decreased in ZCCHC10/PITX1 clones compared that in the vector control clones (each clone reduced hTERT transcription to 27-28% of control cl.1) (Fig 6B). To further examine the down-regulation effects of hTERT transcription on cell kinetics, we investigated expression profiles in senescence and apoptosis-related genes. The protein expression level of p53 was not significantly difference between co-overexpressed ZCCHC10/PITX1 clones and control clones (Fig 6A and 6C). On the other hand, the transcription level of p21, known a senescence marker gene, was partially increased in ZCCHC10/ PITX1 clones (cl.2) when compared to those in control clones (increased 4.0-fold compared to control cl.1) (Fig 6D). Additionally, co-overexpression of ZCCHC10/PITX1 significantly induced upregulation of BAX (BCL-2-associated X) mRNA transcription (increased 1.4-fold and 2.1-fold compared to control cl.1) and down-regulation of BCL-2 (B-cell leukemia/lymphoma 2) mRNA transcription were observed in ZCCHC10/PITX1 clones (reduced BCL-2 transcription to 50% and 61% of control cl.1) (Fig 6E and 6F). The BAX/BCL-2 ratio which known that associated with an increased apoptotic response was significantly increased in co-overexpressed ZCCHC10/PITX1 clones (Fig 6G). To confirm whether apoptosis cells increased by ZCCHC10/PITX1, we performed flow cytometric analysis using Annexin V stanning of co-expressed ZCCHC10/PITX1 clones and those of control clones. The number of Annexin V positive cells were increased in ZCCHC10/PITX1 clones when compared to those in control clones (increased 4.0-fold and 1.9-fold compared to control cl.1) (Fig 6H). Furthermore, those clones inhibited cell growth compared with control clones (Fig 6I). These data demonstrate that co-expressed ZCCHC10/PITX1 down-regulates hTERT expression, accompanied with significantly inhibited cell proliferation and may lead to p53-independent cellular senescence and apoptosis.

ZCCHC10 and PITX1 protein directly bind to a specific region in the hTERT promoter
Previously studies from our laboratory showed PITX1 directly bound to three PITX1 transcription factor regulatory element (RPE) regions in the hTERT promoter (Fig 7A) [17]. To confirm whether PITX1 and ZCCHC10 could bind to these PRE regions of the hTERT promoter, we carried out Chromatin Immunoprecipitation (ChIP) assays with antibodies recognizing PITX1 and ZCCHC10 using A2058 cell lines that stably overexpression of FLAGtagged ZCCHC10 and PITX1 or the control that empty vector transfected. The immunoprecipitated chromatins were amplified by qPCR using hTERT promoter-specific primers ( Fig  7A). The result shows both PITX1 and ZCCHC10 directly bind to the RPE region in hTERT promoter (Fig 7B).
To further examine the suppressive effects of hTERT within endogenous PITX1 and ZCCHC10, we performed ChIP assay using telomerase positive HeLa229 cell line and telomerase negative U2OS cell line. Both PITX1 and ZCCHC10 did not bind to the hTERT promoter in HeLa cells (Fig 7C). In contrast, we found that endogenous PITX1 and ZCCHC10 also bind to the RPE of hTERT promoter in U2OS (Fig 7D). These results suggest that PITX1-ZCCHC10 complex plays a significant role in the negative regulation of hTERT promoter.

Discussion
TERT, which plays a crucial role in the regulation of telomerase activity, contributing to stem cell self-renewal and immortalization of malignant cells, is regulated by many different genes in response to a wide variety of oncogenic and suppressive signaling pathways. We previously identified PITX1 as one of the hTERT suppressor, which directly binds to its promoter region [17]. In this study, we found that ZCCHC10 is one of the components in PITX1 complexes that significantly regulate hTERT transcription. ZCCHC10 is also located in the same human chromosome region 5q31.1 as the PITX1 gene. Recently, another group reported that a long noncoding RNA esophagus epithelial intergenic specific transcript (Epist) can down-regulate hTERT mRNA transcription [39]. Interestingly, the Epist RNA transcript is located 4.5-kbp downstream of the PITX1 coding region. Furthermore, it has been reported that chromosome rearrangements that contain interstitial deletions and chromosome breakpoints were present on the long arm of chromosome 5 in acute myeloid leukemia (AML) and myelodysplasia (MDS) [40]. In particular, the most commonly deleted segment of 5q in MDS is within the 5q31 region [41]. This circumstantial evidence suggests that the human chromosome 5q31 region, which contains a negative regulator of hTERT, plays a significant role as the hTERT regulatory center. hTERT expression is apparent in stem cells of various compartments, while down-regulation of the transcription is seen immediately after differentiation of stem cells [42]. Therefore, future studies that include expression profile analysis of PITX1 and ZCCHC10 in human embryonic stem (hES) cells or induced pluripotent stem (iPS) cells (hTERT positive) and its differential cells (hTERT negative) should facilitate our understanding of the physiological mechanism of hTERT regulation that involved in cancer development.
In this study, three genes (ZCCHC10, hnRNPA0, and HSPA9), which were localized in the human chromosome 5q31.1-q31.2 region near PITX1, were identified as candidate proteins that interacted with PITX1 using pull-down assay. hnRNP A0 belongs to the A/B subfamily of ubiquitously expressed heterogeneous nuclear ribonucleoproteins (hnRNPs). The hnRNAPs are complexes of RNA and protein that localize in the cell nucleus during gene transcription and subsequent post-transcriptional modification of newly synthesized RNA (pre-mRNA). The hnRNAPs have important roles in multiple aspects of nucleic acid metabolism and in the regulation of different cellular processes, and some of them are reportedly associated with cancer development [43]. As little is known about the role of hnRNP A0 in the maintenance of telomerase activity, including the regulation of hTERT transcription, functional analysis of this gene may provide future insights into the regulation telomerase. Other groups have previously reported that HSPA9 directly interacts with p53 protein and inhibits nuclear translocation and activation of p53 mRNA transcription. In addition, targeting HSPA9-p53 complex with shRNA may induce p53-mediated apoptosis in hepatocellular carcinoma [44]. Therefore, HSPA9 was identified as an oncogene through the physical inhibition of p53. On the other hand, we have provided evidence to suggest that HSPA9 protein interacts with PITX1 protein in hTERT-negative U2OS cells, which are known to express wild-type p53 in response to radiation [45]. Thus, these results might suggest that the mechanism by which PITX1 suppresses HSPA9 oncogenic function involves the inhibition of telomerase activity. Interestingly, PITX1 was identified as an activator of p53 mRNA transcription via direct binding to the promoter [25]. It is possible that PITX1 not only plays a role in regulating transcription via directly binding to the target gene promoter region, but may also play a role as a p53 transcription factor and an inhibitor of p53 negative regulator, ultimately controlling telomerase activity.
Homeodomain is one of the most important eukaryotic DNA-binding motifs and is highly conserved in sequence, structure, and mechanism of DNA binding. The homeodomain proteins, which include Nanog homeodomain protein (NANOG) and POU class 5 homeobox 1 (POU5F1 also known as OCT4), play an essential role in the development of pluripotent cells in the embryo, and in the self-renewal of embryonic stem (ES) cells and induced pluripotent stem (iPS) cells. These proteins are the transcription factors that regulate target genes, including epigenetic regulators directly binding to these promoter regions using homeodomain DNA-binding motifs [46]. However, little is known regarding the mechanisms by which these domains act in protein-protein interactions for target molecules. It has been reported that the interferon α (INF-α) promoter is regulated by PITX1 in a manner similar to that of the TERT promoter. Interestingly, the homeodomains in PITX1 protein interacts with interferon regulatory transcription factor 3 (IRF3) and IRF7 proteins that control INF-α gene expression [19]. In addition, another group has shown that PITX families containing PITX1, PITX2 and PITX3 proteins interact via homeodomains with the DNA-binding domain of glial cell missing a (GCMa) proteins, which are known be transcription factors. The PITX-GCMa protein complex binds to the target promoter and influences GCMa-dependent promoter activity in a cellspecific manner [20]. Here, we found that interaction in ZCCHC10 protein through homeodomain in PITX1 controls hTERT mRNA transcription, suggesting that the homeodomain also plays a crucial role in regulating the target molecule. These data suggest that the homeodomain of PITX gene plays a distinctive role in transcriptional regulation of target genes when compared with other homeobox genes.
Mutations of the hTERT promoter is known to occur in over 70% of human melanomas. These mutations occur in the core promoter region (-146C>T and -124C>T), and are known to generate a new motif for transcription factors, which have been linked to reactivation of the epigenetically silenced hTERT transcription. Firstly, the E-twenty-six (ETS) family was reported binding to hTERT promoter mutation site [47]. GA binding protein transcription factor subunit alpha (GABPA) was also reported specially recruitment to mutant hTERT promoter and its mediated long-range chromatin interaction and enrichment of active histone marks which drives hTERT transcription [48]. Additionally, aberrant activation of RAS signaling during BRAF V600E mutation maintains an active chromatin state at the mutate hTERT promoters of human melanoma, which facilitates the recruit of RNA polymerase II thereby leading to transcriptional activation of hTERT [49]. In this study, we showed ectopic expression of PITX1 and ZCCHC10 suppressed hTERT mRNA transcription using A2058 cells that had hTERT promoter mutation (-124C>T). We demonstrated PITX1-ZCCHC10 protein complex directly bound hTERT promoter region (-1325). Therefore, PITX1-ZCCHC10 complex may regulate chromatin modification active to repress even under the hTERT promoter active mutation state.
We reported here that PITX1 interacted with ZCCHC10, resulting in significant suppression of hTERT transcription when compared to overexpression of PITX1 or ZCCHC10 alone. PITX1 has been shown to activate the p53 tumor suppressor gene, which regulates cell cycle progression, DNA integrity and cell survival, by directly binding to the p53 promoter [25]. In addition, we previously showed that PITX1 binds to the hTERT promoter to suppress hTERT transcription [17]. Interestingly, p53 is predicted to interact with ZCCHC10 based on highthroughput Protein-Protein Interaction Network data (http://genomenetwork.nig.ac.jp/). Furthermore, p53 can also suppress hTERT transcription in various tumors, such as lung, prostate, and breast cancer cells [50]. These findings provide evidence that ZCCHC10 plays an important role in the regulation of hTERT through PITX1-and p53-dependent pathways.
Introduction of ZCCHC10 and PITX1 into A2058 cells did not affect expression level of p53 proteins (Fig 6C). Additionally, p53 in A2058 cells has dysfunctional mutation [51]. However, BAX gene known one of p53 downstream effector was increased in ZCCHC10 and PITX1 co-expressed clones (Fig 6E). Therefore, our finding may suggest there are p53 independent pathway upregulate of BAX via suppression of hTERT transcription by introduction of PITX1 and ZCCHCC10. Indeed, PITX1 is associated with p53 independent induced apoptosis in osteogenic sarcoma cell lines [52]. On the other hand, p21 was increased in ZCCHC10/ PITX1 cl.2 but not in ZCCHC10/PTX1 cl.1 (Fig 6D). Associate with this p21 transcription level, cell growth rate was lower in ZCCHC10/PITX1 cl.2 than in ZCCHC10/PITX1 cl.1 ( Fig  6I). These results suggest that the suppression effect of hTERT transcription by introduction of ZCCHC10 and PITX1 involved in senescence and apoptosis, however these introduction of tumor suppression effects may be influenced balances of hTERT canonical and non-canonical function that telomere shortening, restoration of tRNA level, beta-catenin and NF-kappa B signaling pathways [10][11][12]. Further studies aimed at identification of influence of hTERT canonical and non-canonical function in our study, will be required in order to clarify the PIITX1-ZCCHC10 role of tumor suppressing pathways in cancer development.
We previously showed that introduction of an intact human chromosome 5 markedly suppresses hTERT transcription in melanoma A2058 cells [16]. On the other hand, in this study, transient co-expression of PITX1 and ZCCHC10 induced a partial decrease in hTERT transcription ( Fig 4A). It is possible that there are other negative regulatory factors and component proteins in the PITX1-ZCCHC10 complex on human chromosome 5 for functional hTERT control, which would ultimately lead to a large discrepancy in suppression of hTERT transcription in cancer cells. In fact, although PITX1 is localized in the 5q31 region, we have previously found that a putative telomerase repressor gene was mapped to the 5p11 to 5p13 region by combination of functional analysis using chromosome engineering technology [53]. In this study, we found that at least six distinct PITX1-interacting proteins are present on human chromosome 5. Therefore, further studies that include identification of novel hTERT suppressor genes and functional analysis of protein or genes, should facilitate our understanding of the molecular mechanisms involved in the hTERT regulatory network via PITX1.

Cell culture
293T, A2058 and GAK cells were obtained from Japanese Collection of Research Bioresources Cell Bank (Osaka, Japan). SKMEL28, G361, HMV1 and CRL1579 cells were obtained from the Cell Resource Center for Biomedical Research, Institute of Development, Aging and Cancer, Tohoku University, Japan. HeLa229 and U2OS cells were obtained from American Type Culture Collection. Cells were cultured in Dulbecco's Eagle's medium (DMEM; Sigma, St. Louis, MO, USA) supplemented with 10% fetal bovine serum (HyClone, Logan, UT, USA). Human Epidermal Melanocyte cells (Invitrogen) were cultured in Medium 254 (Invirogen, Portland, OR, USA) supplemented with Human Melanocyte Growth supplement (Invitrogen). All cells were cultured at 37˚C in a humidified incubator under 5% CO 2 . FLAG-PITX1 stable expression clones were maintained in DMEM supplemented with 10% FBS and G418 (HeLa229 and U2OS 500 μg/ml, 293T 600 μg/ml; Calbiochem, La Jolla, CA, USA). FLAG-ZCCHC10 and PITX1 stable co-expression A2058 clones were maintained in DMEM supplemented with 10% FBS and G418 (300 μg/ml, Calbiochem) and blasticidine S (3 μg/ml, sigma). All Cell lines detected as Mycoplasma-free by MycoAlert mycoplasma detection kit (Lonza, Walkersville, MD, USA) and no more than 20 passage from the validated stocks.

Cell lysis and immunoprecipitation
Cells were harvested and resuspended in 5 ml of PBS at a concentration of 2×10 7 cells/ml. HeLa229, 293T and U2OS cells stably expressing FLAG-tagged proteins were lysed in 5 ml of ice-cold lysis buffer [100 mM Tris HCl (pH 7.5), with 300 mM NaCl, 2 mM EDTA, 2% Triton X-100, and 400 μl of 25×EDTA-free protease inhibitors (Sigma) per 5 ml]. Soluble fractions from cell lysates were isolated by centrifugation at 15,000 × g for 30 minutes in a microfuge. For immunoprecipitation, 50 μl of anti-FLAG affinity gel (Sigma) was added to each lysate, followed by incubation with rotation for 24 hours at 4˚C. Immunoprecipitates were washed three times with lysis buffer, and 50 μl of 150 μg/ml FLAG peptide was used to elute FLAG tagged proteins. Elution was performed at 4˚C and lasted for 30 minutes for each elution. Gels were centrifuged for 1 min at 5,000 × g, and supernatants were transferred to fresh test tubes. A similar protocol was employed when preparing samples for mass spectrometry. Immunoprecipitated proteins were denatured by addition of sample buffer and boiling for 5 minutes, resolved by 8%-16% SDS-PAGE, and analyzed by immunoblotting, CBB staining and silver staining. CBB staining was performed using a Quick-CBB kit (Wako, Osaka, Japan) according to the manufacturer's instructions. Silver staining was performed using a Sil-Best Stain One kit (Nacalai Tesque, Kyoto, Japan) according to the manufacturer's instructions.

Mass spectrometry
Immunoprecipitates from HeLa229, 293T and U2OS cells stably expressing FLAG-PITX1 were prepared as described above. Proteins with the FLAG peptide were eluted from anti-FLAG affinity beads and air dried for nanoLC-MS/MS analysis. Eluted proteins were also detected by CBB or silver staining. Protein pellets were reconstituted in 13 μL of 8 M urea followed by addition of 2 μL of 40 mg/ml DTT. Tubes were gently vortexed and incubated at 37˚C for 90 min, and 5 μL of 40 mg/ml IAA was added after reducing, followed by incubation for another 30 min at 37˚C in the dark. Then, 60 μl of water was added to tubes. Dilute proteins were digested overnight with 5 μl of 25 ng/μl trypsin gold (Promega, Madison, WI, USA) and desalted with a ZipTipC18 pipette tip (Millipore, Eschborn, Germany), according to the manufacturer's instructions. Peptides were diluted with water and air dried to remove acetonitrile. Formic acid was added to the peptide solution to a final concentration of 3.3%. The resulting digests were analyzed by DiNa nano liquid chromatography system (KYA Technologies) coupled with a QSTAR Elite hybrid mass spectrometer (AB Sciex, Ontario, Canada). MS/ MS spectra were searched against the NCBInr and SwissProt databases using ProteinPilot software 2.0 (AB Sciex).

Transfection and co-immunoprecipitation assay
Cells were transfected with plasmid vectors using Lipofectamine 2000 (Invitrogen). For stable overexpression of FLAG tagged PITX1, 1×10 6 HeLa229, 293T or U2OS cells were seeded in each well of 6-well plates, and were transfected 24 h after seeding with 0.5 μg of plasmid. Stable cell lines were generated using G418.
Transient transfection was performed with Lipofectamine 2000 according to the manufacturer's protocol (Invitrogen). Transfected cells were harvested after 24 h in the following icecold lysis buffer; 100 mM Tris HCl (pH 7.5), with 300 mM NaCl, 2mM EDTA, 2% Triton X-100, and 400 μl of 25× EDTA-free protease inhibitors (Sigma) per 5 ml. Immunoprecipitation was performed according to the manufacturer's instructions for the FLAG Immunoprecipitation Kit (Sigma).

Tissue samples
Biopsy samples of human melanoma were obtained from the Tottori University Hospital. All materials were obtained with written informed consent, and procedures were approved by the Institutional Review Board of Tottori University (Permission No.1558). All experiments were performed in accordance with the guidelines of the Ethics Committee of Tottori University. Total RNA was extracted from tissue samples using the RNeasy plus kit (Qiagen, Valencia, CA, USA), according to the supplier's instructions.

ChIP assay
The chromatin immunoprecipitation (ChIP) assay was performed with SimpleChIP Plus Sonication Chromatin IP kit (Cell Signaling Technology, Danvers, MA, USA) according to the manufacturer's protocol. Briefly, to cross-link the DNA in chromatin to histones, the cells were incubated in 1% formaldehyde for 10 min at 37˚C. After being washed with cold phosphate-buffered saline (PBS) containing protease inhibitors (CST), the cells were resuspended in Cell lysis buffer (CST) containing protease inhibitors. DNA then was broken into 100-to 300-bp fragments by Covaris S220 sonicator (Covaris, Woburn, MA, USA). The remainder of the sample was immunoprecipitated using the anti-PITX1 antibody (ab70273, Abcam), anti-ZCCHC10 antibody (HPA038944, Sigma), or anti-rabbit IgG antibody (#2729, CST) for 16 hr at 4˚C. Protein G magnetic Beads (CST) then was used to collect the immunoprecipitated complexes, which were eluted using elution buffer (CST) after extensive washing. The crosslink then was reversed by the addition of 5 M NaCl, which was followed by protease K treatment for 12hr at 65˚C. DNA was recovered using DNA Purification Columns (CST) and was used as a template for PCR to amplify the region of the PITX1 binding sites in the hTERT promoter. The forward and reverse PCR primers used were forward; 5'-GAGCGACCCGTAAT CCTAAGT -3' and reverse; 5'-GGAGCTAGCATTTGAACAGGC-3'. DNA was amplified using an Applied Biosystems StepOne thermal cycler system and a SYBR green PCR kit (Applied Biosystems).

Analysis of apoptosis
The apoptosis cells were measured with the Annexin V staining by APC Annexin V apoptosis detection kit with 7-AAD according to the manufacturer's (BioLegend, San Diego, CA). Annexin V positive but 7-AAD negative (early apoptotic cells) and Annexin V positive and 7-AAD positive (late stage apoptosis) was determined by Gallios flow cytometer (Beckman Coulter, Brea, CA, USA).

Statistics
Data from more than three separate experiments are presented as means ±S.D. Significance was established at P-values less than 0.05 using an unpaired two-tailed Student's t-test. Analysis of relationship of genes expression level was calculated by Pearson's correlation coefficient (r).