Aurora-A Kinase Regulates Telomerase Activity through c-Myc in Human Ovarian and Breast Epithelial Cells

Introduction

Aurora-A (also named BTAK, STK15, aurora-2, ARKI, and AIKI) is a serine/threonine protein kinase that belongs to the Drosophila aurora and Saccharomyces cerevisiae Ipl1 (Aurora/Ipl1p) kinase family. Three mammalian members have been identified from this family, including Aurora-A, -B, and -C. A conserved kinase catalytic domain is positioned toward the carboxyl termini of the proteins encoded by mammalian members of the Aurora family, whereas a long NH₂-terminal region exhibits much divergence (1) . Aurora-A has attracted intense interest after the discovery that the chromosomal region (20q13.2) in which it is located commonly undergoes amplification in epithelial cancers (2 , 3) . It has been shown that 20q13.2 amplifications involving the Aurora-A gene occur in as many as 12–50% of ovarian, breast, colorectal, and gastric cancers. Moreover, up to 57 and 62% of ovarian and breast cancers, respectively, show overexpression and/or activation of Aurora-A, even where gene amplification is not detected (3 , 4) . In addition, ectopic expression of Aurora-A in murine fibroblasts as well as mammary epithelia induces centrosome amplification, aneuploidy, and oncogenic phenotype (5 , 6) . However, the underlying molecular mechanism is not well understood.

Telomere length and telomerase activity have been implicated in the control of the proliferative capacity of normal and malignant cells (7) . In most human somatic cells, except for regenerating tissues and activated lymphocytes, telomerase activity is undetectable. However, most human cancer cells exhibit stabilized telomere lengths and are positive for telomerase activity. Introduction of telomerase into primary human cells stabilizes telomeres, prevents both senescence and crisis, and endows cells with unlimited proliferative potential (8) . The ability of telomerase to rescue cells from the adverse consequences of telomere dysfunction is likely critical for its role in facilitating malignant transformation of primary human cells and in maintaining the viability and proliferation of established cancer cells.

Human telomerase complex is composed of human telomerase RNA, telomerase-associated protein 1 (TEP1), and human telomerase reverse transcriptase (hTERT). Human telomerase RNA functions as a template for telomere elongation by telomerase (9) . The function of telomerase-associated protein 1 is thought to be associated with RNA and protein binding. hTERT contains reverse transcriptase motifs and functions as the catalytic subunit of telomerase (10 , 11) . Ectopic expression of hTERT in normal human cells restores telomerase activity and extends the replicative life span of the cells (12 , 13) . Moreover, expression of dominant-negative hTERT or addition of small molecule telomerase inhibitors in cancer cells inhibits telomerase activity, leading to drastic telomere shortening, senescence, and apoptosis (14 , 15) . These studies indicate that maintenance of telomeric length by hTERT is required for cells to escape from replicative senescence and to acquire the ability to proliferate indefinitely. Therefore, expression of hTERT/activation of telomerase plays an important role in cellular immortality and malignant transformation.

In the present study, we investigated the regulation of telomerase by Aurora-A in human breast and ovarian epithelial cells. We observed that telomerase activity is significantly induced by Aurora-A through the regulation of hTERT transcription in a c-Myc-dependent manner.

Materials and Methods

Reagents and Plasmids.

Anti-Aurora-A was generated by immunization of rabbit with glutathione S-transferase-aurora/box-2 fusion protein that has been described previously (3) . Anti-HA and anti-FLAG antibodies were from Roche. DMEM and fetal bovine serum were purchased from Invitrogen Co. Hemagglutinin (HA)-tagged Aurora-A was created by PCR using pcDNA3-Aurora-A as template, which was kindly provided by Dr. Subrata Sen (The University of Texas M. D. Anderson Cancer Center, Houston, TX). The PCR products were subcloned to the NotI sites of the mammalian expression vector pHM6 (Roche). Full-length (hTERT/−1375) and core (hTERT/−180) hTERT promoter-Luc plasmids were gifts from Dr. Saturo Kyo (Kanazawa University, Ishikawa-ken, Japan). Myc/MT1 and Myc/MT1–2 mutant core promoter constructs were created using a QuikChange multiple site-directed mutagenesis kit (Stratagene).

Cell Lines, Cell Culture, and Transfection.

The MCF-10A cell line was obtained from the American Type Culture Collection, and ovarian surface epithelium cell line HIOSE118 was described previously (16) . MCF-10A was maintained in DMEM (DMEM/F12) supplemented with 5% fetal bovine serum. HIOSE118 cells was grown in 199/MDCB 105 (1:1) medium (Sigma) supplemented with 5% fetal bovine serum. Transfection was performed using LipofectAmine Plus (Invitrogen).

RNA Interference (RNAi).

The RNAi duplexes with a 2-deoxthymidine overhang were synthesized by Dharmacon Research Inc. The cDNA-targeted region and the sequence of the RNAi duplexes are: c-Myc RNAi (nucleotides 568–588) AACGTTAGCTTCACCAACAGG. The RNAi duplexes were reconstituted to 20 μm in sterile RNase-free water. Transfection of RNAi for targeting endogenous genes was performed using oligofectamine (Invitrogen) as described previously (17) .

Immunoblotting Analysis, Telomerase Assay, and Reverse Transcription (RT)-PCR.

Western blot was performed as described previously (3) . Telomerase activity was measured with a telomerase ELISA kit (Roche) according to the manufacturer’s instructions. In addition, a telomeric repeat amplification protocol (TRAP) assay was also performed using a TRAPeze Telomerase Detection Kit (Intergen) according to the manufacturer’s instructions. Autoradiography was quantified with a PhosphorImager (Molecular Dynamics). RT-PCR was carried out as described previously (16) , using primers derived from hTERT 5′-TCACCTCGAGGGTGAAGGCACTGTT-3′ and 5′-ATGCTGGCGATGACCTCCGTGA-3′.

Luciferase Reporter Assay and in Vivo [³²P]Pi Labeling.

MCF-10A cells were transiently transfected with hTERT-Luc (0.5 μg), Aurora-A (0∼2.0 μg), and β-galactosidase (0.5 μg). The amount of DNA in each transfection was kept constant by the addition of empty pcDNA3 vector. After 36 h of transfection, luciferase activity was measured using a luciferase assay reagent (Promega). Transfection efficiency was normalized by cotransfection with β-galactosidase-expressing vector. The β-galactosidase activity was measured by using Galato-Light (Tropix). Luciferase activity was expressed as relative luciferase activity. [³²P]P_i in vivo labeling was performed as described previously (18) .

Results

Aurora-A Induces Telomerase Activity in HIOSE118 and MCF-10A Cells.

Previous studies have demonstrated that telomerase activity is regulated by a number of oncogenic and tumor suppressor proteins (16 , 19, 20, 21, 22) . To examine the role of Aurora-A in the regulation of telomerase, “normal” ovarian and breast epithelial cells HIOSE118 and MCF-10A cells were transiently transfected with HA-Aurora-A. After 48 h of transfection, telomerase activity was examined by TRAP-ELISA assay. As shown in Fig. 1, A and B ⇓ , low levels of telomerase and endogenous Aurora-A protein were detected in both cell lines. Ectopic expression of Aurora-A significantly induced telomerase activity in a dose-dependent manner. Moreover, Aurora-A-induced telomerase activity in MCF-10A cells was confirmed by a traditional TRAP assay, which was repeated three times with TRAPeze Telomerase Detection kit (Fig. 1, C and D) ⇓ . These results suggest that Aurora-A is a positive regulator of telomerase in human ovarian and breast epithelial cells.

Induction of Telomerase Activity by Aurora-A through Activation of hTERT Transcription but not Phosphorylation.

Previous studies have shown that hTERT, a catalytic subunit of telomerase, is phosphorylated and activated by a number of protein kinases (16 , 22 , 23) . Because Aurora-A is a serine/threonine protein kinase, we next examined whether Aurora-A regulates telomerase through phosphorylation of hTERT. HIOSE118 cells were transfected with FLAG-tagged hTERT expression plasmid together with Aurora-A or pcDNA3 and labeled with [³²P]P_i. After 3 h of the labeling, immunoprecipitations were performed with anti-FLAG antibody and analyzed by Western blot. As shown in Fig. 2A ⇓ , FLAG-hTERT was expressed as an expected size in the cells transfected with hTERT/pcDNA3 and hTERT/Aurora-A; however, phosphorylated hTERT was not induced by ectopic expression of Aurora-A. As a positive control, HIOSE118 cells were cotransfected with constitutively active Akt (myr-Akt) and hTERT. The in vivo labeling showed that hTERT is phosphorylated by myr-Akt (Fig. 2A) ⇓ . Thus, we conclude that Aurora-A does not phosphorylate hTERT. We next examined whether Aurora-A regulates hTERT at transcription levels. RT-PCR analyses revealed that mRNA levels of hTERT were undetectable in MCF-10A and HIOSE118 cells and that ectopic expression of Aurora-A induced hTERT mRNA in both cell lines in a dose-dependent manner (Fig. 2B) ⇓ . Therefore, these data indicate that Aurora-A regulates hTERT at transcriptional but not a posttranslation modification level.

Aurora-A Activates hTERT Promoter Activity.

Because hTERT mRNA was induced by Aurora-A, we next examined whether its promoter is regulated by Aurora-A. HIOSE118 and MCF-10A cells were cotransfected with hTERT promoter-Luc [−1375/+78/hTERT-Luc (long promoter) or −181/+78/hTERT-Luc (core promoter)] and Aurora-A or pcDAN3 vector. As illustrated in Fig. 3A ⇓ , both long and core hTERT promoters were induced by expression of Aurora-A. However, Aurora-A-induced hTERT core promoter activity is about two times higher than that of the long promoter, suggesting that Aurora-A targets to a transcription factor(s) that regulates the core promoter. We further observed that Aurora-A induces hTERT core promoter activity to a similar level as that of ectopic expression of c-Myc (Fig. 3B) ⇓ . Compared with the cells transfected with Aurora-A alone, coexpression of c-Myc and Aurora-A did not exhibit synergistic effect on the promoter activity (Fig. 3B) ⇓ , implying that c-Myc could mediate Aurora-A action in hTERT.

c-Myc Binding Sites Are Required for Induction of hTERT Promoter Activity by Aurora-A.

The hTERT core promoter contains two c-Myc binding sites (E-box), and c-Myc is known to be a key activator of hTERT transcription (20) . Because expression of c-Myc could not further enhance Aurora-A-induced hTERT promoter activity, we reasoned that c-Myc binding site(s) of the core promoter might be required for Aurora-A-induced hTERT promoter activity. To test this hypothesis, hTERT luciferase reporters with mutation of E-box 1 (hTERT/-181E1/m) and E-box 2 (hTERT/-181E1, 2/m) as well as the Sp1 site of the core promoter were created by mutagenesis assay as described in “Materials and Methods.” Luciferase assay revealed that Aurora-A-induced hTERT promoter activity was completely abrogated by mutation of either E-box 1 and/or 2 but not Sp1 (Fig. 3C) ⇓ . Moreover, ectopic expression of c-Myc alone or coexpression of c-Myc and Aurora-A failed to stimulate hTERT/−181E1/m-Luc and hTERT/−181E1, 2/m-Luc activity (Fig. 3B) ⇓ . These data strongly suggest that Aurora-A activates hTERT by targeting the c-Myc transcription factor.

Aurora-A Up-Regulates c-Myc; Knockdown of c-Myc Attenuates Aurora-A-induced hTERT and Telomerase Activity.

Because coexpression of c-Myc and Aurora-A did not show a synergistic effect on hTERT promoter activity, it is unlikely that Aurora-A regulates c-Myc through posttranslational modification. In fact, our in vivo [³²P]P_i cell labeling and coimmunoprecipitation experiments revealed that c-Myc was neither phosphorylated by nor interacted with Aurora-A (data not shown). We next examined whether Aurora-A regulates c-Myc expression. Parental and Aurora-A stably transfected MCF-10A and HIOSE118 cells were analyzed for expression of c-Myc. As illustrated in Fig. 4A ⇓ , protein levels of c-Myc were significantly elevated in Aurora-A stably transfected MCF-10A and HIOSE118 cells as compared with pcDNA3-transfected cells.

Because c-Myc is a major regulator of hTERT transcription and is up-regulated by Aurora-A, we reasoned that Aurora-A-induced telomerase activation is mediated by its action on c-Myc. To test this hypothesis, Aurora-A stably transfected MCF-10A cells were treated with RNAi of c-Myc. Immunoblotting analysis showed that Aurora-A-up-regulated c-Myc is abrogated by the RNAi treatment (Fig. 4B) ⇓ . Accordingly, mRNA levels of hTERT that were induced by Aurora-A were also attenuated (Fig. 4B) ⇓ . Furthermore, Aurora-A-stimulated hTERT promoter and telomerase activities were inhibited by RNAi of c-Myc (Fig. 4, C and D) ⇓ . Taken collectively, these data indicate that Aurora-A-induced telomerase activity and hTERT transcription result from its induction of c-Myc expression.

Discussion

Aurora-A locates to the spindle pole during mitosis and is regulated in a cell cycle-dependent manner; its protein is low in G₁-S, up-regulated during G₂-M, and reduced rapidly after mitosis (24) . Therefore, almost all of the Aurora-A functional studies published thus far have focused on its role of mitosis regulation (1) . It has been shown that Aurora-A phosphorylates several proteins that are important for mitosis, including: histone H3 (25 , 26) , a key molecule in conversion of the relaxed interphase chromatin to mitotic condensed chromosomes; CPEB (cytoplasmic polyadenylation element-binding protein), best known for its role in promoting polyadenylation of cyclin B mRNA (27) ; TACC3, a protein required for stabilization and organization of microtubules (28) ; Eg5, a kinesin-like protein involved in both centrosome separation and spindle assembly and stability (29) ; and TPX2, which is required to generate a stable bipolar spindle (30) . However, we demonstrate in this report that Aurora-A induces telomerase activity by up-regulation of hTERT at a transcriptional level in human ovarian and breast epithelial cells. Moreover, we define the underlined mechanism by identification of Aurora-A stimulation of c-Myc expression. Thus, we provide evidence of Aurora-A function beyond its cell cycle control.

Activation of telomerase in a majority of human cancers but not in most normal somatic cells and inhibition of telomerase activity in cancer cells, abolishing their telomere maintenance and immortal growth, establish an important role for telomerase-mediated telomere maintenance in human cell immortalization and carcinogenesis. The expression level of the hTERT gene represents a major determinant of telomerase activity in human cells (11 , 12) . Thus, investigation of transcriptional regulation of the hTERT gene should be essential for elucidating molecular mechanisms of telomerase regulation, immortalization, and carcinogenesis in humans. We have shown previously frequent overexpression and/or activation of Aurora-A in human ovarian carcinoma (3) . In this report, we demonstrated that ectopic expression of Aurora-A induces telomerase activity and hTERT expression in human ovarian and breast epithelial cells, which could be one of the major mechanisms for Aurora-A contribution to ovarian and breast oncogenesis.

Regulation of telomerase activity has been widely studied, and mechanisms for posttranslational regulation, i.e., phosphorylation and methylation, of hTERT have been suggested. However, accumulated evidence has indicated a major control point at the level of transcription. Several transcription factors have thus far been identified to regulate hTERT transcription: the E-box-binding oncoprotein c-Myc (20) and a ubiquitous transcription factor Sp1 (31) are activators; the E-box-binding factor Mad1 (32) , the tumor suppressor proteins p53 and WT1 (21) , and the zinc-finger factor MZF2 (33) are possible repressors. Among the consensus binding sequences of these regulators, two canonical E-box (CACGTG) elements located upstream and downstream of the transcription initiation site (187 to 182 and +22 to +27, respectively), which are the binding sites for c-Myc and Mad1, have been most analyzed extensively. The data presented in this study showed mutation of the E-box upstream the transcription initiation site is sufficient to abrogate Aurora-A-stimulated hTERT promoter activity. Ectopic expression of Aurora-A up-regulates c-Myc protein (Fig. 4) ⇓ but does not phosphorylate hTERT (Fig. 2A) ⇓ . Furthermore, RNAi of c-Myc significantly inhibited Aurora-A-induced telomerase activation and hTERT mRNA expression. These results indicate that Aurora-A regulates telomerase through c-Myc rather than direct binding to the promoter or phosphorylation of hTERT.

In summary, we demonstrate in this study that Aurora-A induces telomerase activity in human ovarian and breast epithelial cells by up-regulation of hTERT, which results from Aurora-A-induced c-Myc expression. Because Aurora-A is frequently altered in human ovarian and breast carcinomas, these findings provide an additional mechanism for Aurora-A in ovarian and breast carcinogenesis in addition to its role in G₂-M cell cycle control. Further investigations are required to characterize the mechanism by which Aurora-A up-regulates c-Myc.