Pituitary tumor transforming gene 1 regulates Aurora kinase A activity

Article metrics

Abstract

Pituitary tumor transforming gene 1 (PTTG1), a transforming gene highly expressed in several cancers, is a mammalian securin protein regulating both G1/S and G2/M phases. Using protein array screening, we showed PTTG1 interacting with Aurora kinase A (Aurora-A), and confirmed the interaction using co-immunoprecipitation, His-tagged pull-down assays and intracellular immunofluorescence colocalization. PTTG1 transfection into HCT116 cells prevented Aurora-A T288 autophosphorylation, inhibited phosphorylation of the histone H3 Aurora-A substrate and resulted in abnormally condensed chromatin. PTTG1-null cell proliferation was more sensitive to Aurora-A knock down and to Aurora kinase Inhibitor III treatment. The results indicate that PTTG1 and Aurora-A interact to regulate cellular responses to anti-neoplastic drugs. PTTG1 knockdown is therefore a potential approach to improve the efficacy of tumor Aurora kinase inhibitors.

Introduction

Pituitary tumor transforming gene 1 (PTTG1), isolated from rat pituitary tumor cells (Pei and Melmed, 1997), was subsequently identified as a securin protein (Zou et al., 1999), which binds separase, inhibits cohesin cleavage and facilitates sister chromatid separation. PTTG1 is required for several cellular processes (Vlotides et al., 2007), including human fetal brain development (Boelaert et al., 2003), telencephalic neurogenesis (Tarabykin et al., 2000) and rat liver regeneration (Akino et al., 2005). PTTG1-null mice exhibit testicular, pituitary, and pancreatic beta cell hypoplasia (Wang et al., 2001b, 2003) and male-selective insulinopenic diabetes (Wang et al., 2003). PTTG1 is abundantly expressed in most cancers and PTTG1 levels correlate with tumor development and size (Saez et al., 1999; McCabe et al., 2003). PTTG1 is induced early in the pathogenesis of estrogen-induced rat prolactinomas (Heaney et al., 1999), and has been suggested as a prognostic marker for thyroid, breast (Solbach et al., 2004) and colon cancer invasiveness (Heaney et al., 2000). PTTG and HTLV-1 Tax exhibit a cooperative transforming activity (Sheleg et al., 2007), whereas small interfering RNA (siRNA) directed against PTTG suppressed lung cancer growth in nude mice (Kakar and Malik, 2006) and was also proposed as a subcellular therapy for ovarian cancer (El-Naggar et al., 2007).

Overexpressed PTTG1 results in chromosome instability and aneuploidy, which has been suggested as a mechanism underlying PTTG1 transforming activity (Wang and Melmed, 2000; Yu et al., 2003). PTTG1 inhibits p53 transcriptional activity (Bernal et al., 2002), and p53 stabilization is uncoupled by the loss of PTTG1 (Bernal and Hernandez, 2007). PTTG interacts with Ku, the regulatory subunit of DNA-dependent protein kinase (Chien and Pei, 2000; Pei, 2000, 2001; Wang and Melmed, 2000), further indicating a role for the protein in DNA damage repair. PTTG1 induces genetic instability in colorectal cancer cells by inhibiting double-stranded DNA repair activity (Kim et al., 2007), activates c-Myc (Pei, 2001) and basic fibroblast growth factor (bFGF) (Chien and Pei, 2000) and promotes tumor angiogenesis (Zou et al., 1999; Kim et al., 2006). Dysregulated PTTG1 likely prevents mitotic exit as doubly mutant Cdc20 and PTTG1 embryos were unable to maintain metaphase arrest (Li et al., 2007). Although non-homologous end joining is intact in securin-deficient cells, the process occurs through aberrant end processing (Bernal et al., 2008). Furthermore, securin-deficient cells exhibit enhanced sensitivity to DNA damage-inducing agents (Bernal et al., 2008). Using a ChIP (chromatin immunoprecipitation)-on-Chip assay, PTTG1 was shown to bind multiple gene promoters (Tong et al., 2007), and PTTG1 interacts with Sp1 to promote the G1/S transition (Tong et al., 2007). This action may also contribute to PTTG1-induced cell transformation. To further investigate PTTG1 functions, we screened an array of 5000 proteins for PTTG1 interactions and showed that PTTG1 interacts with Aurora kinase A (Aurora-A), and also suppresses Aurora-A activity. Aurora-A is activated by TPX-2, Ajuba and Bora (Eyers et al., 2003b; Hirota et al., 2003; Hutterer et al., 2006), suppressed by its N-terminal domain (Zhang et al., 2007) and its oncogenic activity suppressed by p53 (Chen et al., 2002). Our findings thus add knowledge to mechanisms for Aurora-A regulation.

Results

PTTG1 interacts with Aurora-A in vitro and in vivo

PTTG1 is shown to interact with more than 700 gene promoters (Tong et al., 2007). Here we show PTTG1 specifically interacts with several proteins (Supplementary data 1). Thus, we built a PTTG1 interaction network by combining the results. With this assay, we confirmed the reported interaction of PTTG1 and ribosomal protein S10 (Pei, 1999), with a Z-score of 3.34 (Figure 1a). PTTG1 was shown to bind specifically to Aurora-A with a Z-score of 4.90, indicative of a high probability of true interaction (Figure 1a). Using co-immunoprecipitation and His-tag pull-down assay to confirm this interaction (Figure 1b), and using Aurora-A antibody, we readily co-immunoprecipitated PTTG1 protein from whole cell extracts. Aurora-A protein was also detected in the immunoprecipitated complex using PTTG1 antibody, indicating an in vivo interaction. In His-Tag pull-down assays, wherein expressed Aurora-A, PTTG1 and their respective fragments served as bait, their respective counterparts were captured. As shown in Figure 1c, His-tagged Aurora-A and fragment 1 co-precipitated with PTTG1, whereas His-tagged PTTG1 and fragment 1 also pulled down Aurora-A. Thus, PTTG1 and Aurora-A appear to interact through their respective N termini.

Figure 1
figure1

Pituitary tumor transforming gene 1 (PTTG1) interacts with Aurora kinase A (Aurora-A). (a) PTTG1 was biotin-labeled and incubated with the protein microarray. Attached biotinylated PTTG1 was probed with Streptavidin-Alexa Fluor 647 conjugate. The results show that PTTG1 binds Aurora-A and ribosomal protein S10 (RPS10) but not bovine serum albumin (BSA). (b) Aurora-A antibody was used to co-immunoprecipitate PTTG1 and PTTG1 antibody used to co-immunoprecipitate Aurora-A, respectively. Western blots show that PTTG1 co-immunoprecipitated with Aurora kinase A. (c) His-Tag Aurora-A, PTTG1 and their fragments were expressed using the TNT Quick Coupled Transcription/Translation system. Aurora-A and its fragments were used to pull down PTTG1, or PTTG1 and its fragments were used to pull down Aurora-A. The results show that PTTG1 protein binds Aurora-A and its fragment 1 and Aurora-A binds to PTTG1 and its fragment 1, suggesting that the two proteins interact through their respective N termini. (d) Aurora-A and PTTG1 interaction domains.

PTTG1 colocalizes with centrosomic and spindle Aurora-A

Intracellular Aurora-A localization is important for mitotic control and spindle formation, and intracellular Aurora-A locations undergo striking changes during the cell cycle. In late G1/early S phase, Aurora-A is found in pericentriolar centrosome material, and during prophase. The centrosomic association continues at the mitotic poles and, subsequently at metaphase, resides in adjacent spindle microtubules (Li and Li, 2006). Owing to lack of a sufficiently robust PTTG1 antibody for intracellular immunofluorescent location, we monitored intracellular PTTG1 location by using green fluorescent protein (GFP)-tagged PTTG1. As shown in Figure 2a, PTTG1-GFP colocalized with Aurora-A in the spindle and centrosome during metaphase. PTTG1 inhibits separase proteolytic activity (Hornig et al., 2002; Waizenegger et al., 2002), and when degraded, releases separase at the metaphase to anaphase transition. As separase cleaves cohesin and promotes sister chromatid separation (Hornig et al., 2002), we tested separase and Aurora-A localization in HCT116 cells. As shown in Figure 2a, separase and Aurora-A colocalized mainly to the centrosome, clearly distinct from PTTG1 and Aurora-A colocalization, suggesting that PTTG1 and Aurora-A interact independently of separase. Furthermore, separase and Aurora-A centrosome colocalization was not evident in cells devoid of PTTG1 (Figure 2b), further indicating the PTTG1 requirement for centrosomic separase location. When separase levels were knocked down by specific RNA interference, PTTG1 and Aurora-A colocalization persisted (data not shown). Thus, in vivo PTTG1 binding to Aurora kinase A appears not to involve separase, a known protein partner for PTTG1. Furthermore, PTTG1 also appears to direct the centrosomic location of separase during mitosis, besides inhibiting separase activity. Centrosomic separase accumulation may be important for separase activation as the duration of accumulation approximates the time of separase activation.

Figure 2
figure2

Cellular location of Pituitary tumor transforming gene 1 (PTTG1). PTTG1-GFP was used to identify intracellular PTTG1 location. Aurora kinase A (Aurora-A) and separase localization was detected by antibodies as indicated in Materials and methods. Immunofluorescence shows (a) in HCT116 cells PTTG1 colocalizes with Aurora-A on the spindle and centrosome in HCT116 cells as indicated by the white arrow. (b) Separase is not detected in the centrosome (white arrow) in PTTG1−/− HCT116 cells. (c) Separase is localized on the centrosome area (white arrow) in HCT116 cells. (c) PTTG1 overexpression caused abnormal chromatin structure (white arrow). Scale bar=5 μm.

PTTG1 inhibits Aurora-A phosphorylation of histone H3

To assess whether PTTG1 serves as a substrate competitor for Aurora-A activity, we tested whether PTTG1 inhibits phosphorylation of the histone H3 substrate. Aurora-A is an arginine-directed kinase (Ohashi et al., 2006) modulating spindle function and chromatin condensation, and which phosphorylates histone H3 (Crosio et al., 2002), crucial for the onset of mitosis in the early G2 phase (Pascreau et al., 2003). Non-phosphorylated histone H3 participates in chromatin compaction, and during mitosis, Ser10 phosphorylation weakens histone tail–DNA interactions and favors DNA–polyamine binding, thus enabling the formation of highly compacted mitotic chromosomes (Prigent and Dimitrov, 2003). Similar to the histone H3 Ser10 motif (Crosio et al., 2002), the PTTG1 protein sequence contains a putative Aurora-A consensus motif (RXS/T) (Cheeseman et al., 2002) in the basic N-terminal domain (Figure 1d). We assayed cell-free kinase activity and as shown in Figure 3e, when Aurora-A was omitted from the reaction, endogenous phosphorylated histone H3 was low, whereas the presence of Aurora-A enabled histone H3 phosphorylation. PTTG1 appeared to directly inhibit Aurora-A activity, as evidenced by dose-dependent inhibition of Aurora-A histone H3 phosphorylation.

Figure 3
figure3

Pituitary tumor transforming gene 1 (PTTG1) attenuates Aurora kinase A (Aurora-A) function. (a) Overexpressed PTTG1 caused abnormal spindle formation as evidenced by unevenly distributed PTTG1-GFP and Aurora-A (white arrows). (b) Western blot showing that overexpressed PTTG1 suppressed Aurora-A T288 autophosphorylation and histone H3 Ser10 phosphorylation in HCT116 cells compared with controls. (c) Overexpressed PTTG1 repressed histone H3 Ser10 phosphorylation as shown by immunofluorescence (white arrows). Scale bar=5 μm. (d) Degradation of mitotic proteins. HCT116 cells were synchronized with 400 ng/ml nocodazole for 12 h and mitotic cells collected by shake off, washed and cultured in fresh medium at 37 °C to allow progression of G2/M, and cells harvested at the indicated time points. Western blot showed that PTTG1 degraded 4.5 h and Aurora-A 5.5 h after cells were released from G2/M block. (e) In vitro kinase assay. Aurora-A, histone H3 and increasing PTTG1 or GST-His levels (0.5, 1, 2 μg) were brought to a 30 μl volume reaction system as described in Materials and methods. After incubations at 37 °C for 20 min, reactions were halted and analysed by western blotting using histone H3 ser10 antibody. PTTG1 dose-dependently inhibits Aurora-A phosphorylation of histone H3 substrate, whereas GST-His has no effect at the same concentration. C1 is a control without Aurora-A; C2 is a positive control (without Aurora-A inhibitor).

We then tested whether PTTG1 inhibits Aurora-A activity in vivo. The phosphorylation state of histone H3 determines chromatin condensation and de-condensation (Crosio et al., 2002), and Aurora-A action depends on T288 autophosphorylation in the activation loop (Walter et al., 2000). As shown in Figure 3b, PTTG1-GFP overexpression in HCT116 cells resulted in decreased phosphorylation of both Aurora-A and histone H3 as demonstrated by western blot. Reduced histone H3 Ser10 phosphorylation levels also correlated with chromatin abnormalities, as shown in Figure 3c. Next, we tested whether separase inhibition induces similar effects as those observed with overexpressed PTTG1. Although separase was reported to impact Aurora kinase B (Pereira and Schiebel, 2003), reduction of separase levels caused aneuploidy but had little effect on the spindle, Aurora-A distribution or chromatin condensation (data not shown), conforming with previous reports (Waizenegger et al., 2002). These observations also indicate that PTTG1 action on Aurora-A and chromatin occurs independently of effects on separase.

The small-molecule Aurora-A inhibitor, ZM 447439, retards the progression of chromosome condensation, leading to a disorganized spindle (Gadea and Ruderman, 2005). As shown in Figure 3a, PTTG1 overexpression caused unequal PTTG1 and Aurora-A distribution, suggestive of spindle disorganization. Also, transfectants overexpressing PTTG1-GFP contain abnormal chromatin, likely reflecting the interference with chromatin condensation and de-condensation (Figure 3c). Taken together, these results confirm that PTTG1 inhibits Aurora-A activity in vivo. As both PTTG1 and Aurora-A are substrates of the adenomatous polyposis coli (APC) degradation pathway, we investigated the time course of PTTG1 and Aurora-A degradation during mitosis (Figure 3d), and PTTG1 degraded about 60 min before Aurora-A. Thus, PTTG1 appears to inhibit Aurora-A activity, and PTTG1 degradation may release Aurora-A for cell cycle activity. Unlike results shown in HeLa cells (Lindon and Pines, 2004), PTTG1 degradation starts at 4.5 h and Aurora-A at 5.5 h, suggesting that HCT116 cell mitosis might be delayed upon synchronization treatments.

PTTG1–Aurora-A regulates Aurora kinase inhibitor III and doxorubicin response

Aurora-A regulates mitotic entry at the G2 checkpoint (Marumoto et al., 2002), and abrogating DNA damage-induced G2 checkpoints may induce topoisomerase-based drug sensitivity (Vogel et al., 2005, 2007). As PTTG1 modulates Aurora-A activity, we tested whether PTTG1 modulates responses to antineoplastic drugs whose efficacy relies on a functional G2 checkpoint.

Aurora-A is overexpressed in a variety of tumor cell lines and acts to transform fibroblasts and format multipolar mitotic spindles associated with genomic instability (Zhou et al., 1998). As several Aurora-A inhibitors exhibit antineoplastic effects (Mountzios et al., 2008), we tested the effects of Aurora-A siRNA knock down in HCT116 wild-type (WT) and PTTG1−/− cells. As shown by western blot (Figure 4a), RNA interference efficiently knocked down Aurora kinase A expression in both WT and PTTG1−/− cells, but PTTG1−/− cells were more sensitive to Aurora-A knock down, with 62% remaining viable compared with 88% of WT cells (P<0.05) (Figure 4a).

Figure 4
figure4

Pituitary tumor transforming gene 1 (PTTG1) regulates cell response to Aurora kinase inhibition. (a) PTTG1−/− cell proliferation was attenuated in response to Aurora kinase A (Aurora-A) siRNA (mean±s.d., n=3. *P<0.05 vs WT (wild type)). Cells were seeded into 96-well plates, cultured overnight and then transfected with siRNAs (100 nM). After 72 h, premixed WST-1 cell proliferation reagent (10 μl) was added to each well. Cells were incubated for a further 4 h and shaken thoroughly for 1 min before measurement. (b) PTTG1−/− cell growth is more sensitive to Aurora kinase inhibitor III (mean±s.d., n=3. *P<0.05 vs WT). Cells were seeded into 96-well plates, cultured overnight and then treated with drug as indicated for 48 h. Viable cells were measured using premixed WST-1 cell proliferation reagent. (c) Anchorage-dependent colony formation assay was performed as described in Materials and methods. PTTG1−/− cells were more sensitive to Aurora-A inhibitor III than WT cells (mean±s.d., n=3. *P<0.05 vs WT). (d) Overexpressed PTTG1 rescued PTTG1−/− sensitivity to Aurora kinase inhibitor III (12.5 μM) (mean±s.d., n=3. P>0.05 vs WT); PTTG1 knockdown enhanced WT cell sensitivity to Aurora kinase inhibitor III (12.5 μM) (mean±s.d., n=3. P>0.05 vs PTTG1−/−). PTTG1−/−, PTTG1-null or knockout (KO) cells.

Replication of PTTG1−/− cells treated with 12.5 μM Aurora kinase inhibitor III Cyclopropanecarboxylic acid-(3-(4-(3-trifluoromethyl-phenylamino)-pyrimidine-2-ylamino)-phenyl)-amide (Figure 4b) was more sensitive than WT responses, with 56% remaining viable as compared to 73% of WT cells treated for 48 h (P<0.05). Archorage-dependent colony formation by PTTG1−/− cells was abrogated as compared with WT colonies in the presence of Aurora kinase inhibitor III (Figure 4c). We then either re-introduced PTTG1 into PTTG1−/− HCT116 cells or transfected PTTG1 siRNA into WT cells to investigate the PTTG1 role in the drug response. As shown in Figures 4d and e, knock down of PTTG1 increased HCT116 cell sensitivity to Aurora kinase inhibitor III and overexpressed PTTG1 rescued the sensitivity of PTTG1−/− cells. Other less-specific Aurora kinase inhibitors, including Aurora kinase Inhibitor II (4-(4′-benzamidoanilino)-6,7-dimethoxyquinazoline), Aurora kinase/CDK inhibitor 4-(5-amino-1-(2,6-difluorobenzoyl)-1H-[1,2,4]triazol-3-yl-amino)-benzenesulfonamide, were more effective in suppressing HCT116 PTTG1−/− cells (Supplementary data 2).

PTTG1−/− cells are sensitive to stress and enter the G2/M phase prematurely (Bernal et al., 2008). Moreover, G2 DNA damage checkpoint abrogation renders cells more sensitive to DNA damage-inducing drugs (Vogel et al., 2005, 2007). Our results suggest that in the absence of PTTG1, Aurora-A activity is enhanced, likely disrupting the G2 checkpoint, with subsequently enhanced sensitivity to DNA damage-inducing drugs. To test whether PTTG1−/− cell sensitivity to DNA damage is due to attenuated G2 checkpoint control related to a PTTG1–Aurora-A interaction, we used the DNA damage agent doxorubicin to treat WT and PTTG1−/− HCT116 cells. As shown in Figure 5a, we confirmed that PTTG1−/− cell proliferation was more sensitive to 1.25 μM doxorubicin (Bernal et al., 2008); 27% of PTTG1−/− cells remained viable as compared with 55% of treated WT HCT116 cells. This result was confirmed by colony formation assays, showing that after treatment with 0.16 μM doxorubicin, 36% of PTTG1−/− colonies survived as compared with 74% of WT colonies (Figure 5b). As shown in Figure 5c, overexpressing PTTG1 in PTTG1−/− cells rescued tumor cell susceptibility to doxorubicin, whereas knock down of PTTG1 in HCT116 WT cells enhanced cell sensitivity to doxorubicin treatment. We then tested the requirement for Aurora-A in response to DNA-damaging agents by using Aurora-A inhibitor or siRNA. Low-dose (3 μM) Aurora kinase inhibitor III exhibited similar proliferation inhibitory effects on both WT and PTTG1−/− cell proliferations (Figure 4b). After preincubation with 3 μM Aurora kinase inhibitor III, WT and PTTG1−/− cells exhibited a similar response to doxorubicin (Figure 5d), suggesting that abrogated G2 PTTG1–Aurora-A may underlie PTTG1−/− cell sensitivity to doxorubicin. These results were further confirmed by knocking down Aurora-A expression levels and showing similar responses to doxorubicin upon Aurora-A siRNA transfection as those observed in PTTG1−/− cells (Figure 5e).

Figure 5
figure5

Pituitary tumor transforming gene 1 (PTTG1) regulates cell response to doxorubicin. (a) PTTG1−/−cell growth was more sensitive to doxorubicin (mean±s.d., n=3; *P<0.05 vs WT (wild type)). Cells were seeded into 96-well plates, cultured overnight and treated with test drug for 48 h. Premixed WST-1 cell proliferation reagent (10 μl) was used to measure viable cells. (b) Results of colony formation assay after Aurora kinase inhibitor III treatment (mean±s.d., n=3. *P<0.05 vs WT). (c) Overexpressed PTTG1 rescued PTTG1−/− doxorubicin sensitivity (mean±s.d., n=3. P>0.05 vs WT); PTTG1 knockdown enhanced WT cell sensitivity to doxorubicin (mean±s.d., n=3; P>0.05 vs PTTG1−/−). (d) Low-dose (3 μM) Aurora kinase inhibitor III treatment abolished the different response of WT and PTTG1−/− HCT116 cells to 0.63 μM doxorubicin (mean±s.d., n=3). (e) Aurora kinase A siRNA abolished the differences in WT and PTTG1−/− cell responses to 0.63 μM doxorubicin (mean±s.d., n=3).

PTTG1–Aurora-A interaction is not related to impaired phosphorylation

As Aurora-A activity is inhibited by DNA damage (Krystyniak et al., 2006), we tested Aurora-A activity in HCT116 WT and PTTG1−/− cells treated with doxorubicin. Immunoprecipitation and the measurement of kinase activity showed that after doxorubicin treatment, Aurora-A activity was higher in PTTG1−/− than in WT cells (Figure 6a). We then tested PTTG1 responses to DNA damage (Figure 6b) and showed enhanced PTTG1 phosphorylation induced by DNA-damaging drug treatment. Although phosphorylated PTTG1 may act with Aurora-A at the DNA damage-induced G2 check point, several phosphorylation-deficient PTTG1 mutants tested did not impair the PTTG1–Aurora-A interaction (data not shown).

Figure 6
figure6

Pituitary tumor transforming gene 1 (PTTG1) and DNA damage pathway. (a) Aurora kinase A (Aurora-A) activity was less inhibited after DNA damage treatment in PTTG1−/− cells. Lane C1, positive control with Aurora-A; lane C2, negative control without Aurora-A; lane WT, Aurora-A co-immunoprecipitated from HCT116 WT cells after doxorubicin treatment; Lane−/−, Aurora-A immunoprecipitated from HCT116 PTTG1−/− cells after doxorubicin treatment. Aurora-A activity is higher in PTTG1−/− than in WT HCT116 cells. (b) Increased phosphorylated PTTG1 (arrowhead, shifted band) after 24 h of 5 μM doxorubicin treatment. (c) Proposed signaling pathway for PTTG1 action at the G2 DNA damage checkpoint.

Discussion

Our results indicate that PTTG1 directly inhibits Aurora-A activity both in vitro and in vivo. However, PTTG1 may also act in concert with other proteins to inhibit Aurora-A activity, as PP2A binds to PTTG1 throughout the cell cycle, stabilizes PTTG1 by dephosphorylation (Gil-Bernabe et al., 2006) and also dephosphorylates and inactivates Aurora-A (Eyers et al., 2003a; Horn et al., 2007). Whether PTTG1 facilitates Aurora-A dephosphorylation by targeting PP2A to Aurora-A requires further investigation.

PTTG1 overexpression inhibited histone H3 phosphorylation and chromatin condensation. Abnormal chromatin structure was also observed in a PTTG1 transgenic mouse model (Donangelo et al., 2006) where heterochromatin was less abundant in PTTG1-GFP overexpressing cells, suggesting the inhibition of chromatin condensation. However, whether these in vivo observations are related to Aurora-A is as yet unclear. Using algorithm to analyse the PTTG1 sequence further identifies several putative Aurora-A phosphorylation sites (Xue et al., 2008). Whether Aurora-A phosphorylates the putative sites needs further investigation.

PTTG1−/− HCT116 cells exhibited enhanced sensitivity to Aurora-A siRNA, several Aurora kinase inhibitors and to doxorubicin. Aurora-A overactivity in PTTG1−/− HCT116 cells might be the reason for the cell more sensitive to Aurora kinase inhibitor III according to similar reports (Ikezoe et al., 2007). Aurora-A overexpression could cause drug resistance under specific conditions (Yang et al., 2006; Sun et al., 2007). In our experiments, Aurora-A overactivity in the absence of PTTG1 more likely promotes cell death similar to reported (Zhang et al., 2004, 2008). The notice of cell death and related mechanisms remain to be elucidated.

Pretreatment with low-dose Aurora kinase inhibitor III or siRNA also abolished differences between WT and PTTG1−/− cell responses to doxorubicin, suggesting alternative mechanisms for cell survival after different treatments. PTTG1 nuclear–cytoplasmic translocation upon DNA damage (Kim et al., 2007) may underlie PTTG1 suppression of Aurora-A. Unlike results obtained after ultraviolet and etoposide treatment (Kim et al., 2007; Lai et al., 2007), HCT116 cells treated with 5 μM doxorubicin for 24 h showed increased PTTG1 levels and phosphorylation, which might be due to different treatments or doses employed.

Although PTTG1 phosphorylation is enhanced after DNA damage, impaired phosphorylation caused by the S181A mutation did not alter the interaction. After DNA damage, extranuclear PTTG1 translocation likely enables DNA damage repair (Kim et al., 2007), and also prevents cytoplasmic Aurora-A from exerting a G2 check point action, allowing time for DNA repair. However, the kinase responsible for PTTG1 phosphorylation upon DNA damage is not known yet. Pds1 (a PTTG1 analog) is phosphorylated by Chk1 on nine sites upon DNA damage in yeast (Wang et al., 2001a), suggesting that Chk1 might be one of the candidates.

PTTG1 is a negative regulator of p53 transcriptional activity (Bernal et al., 2002). p53 stabilization in the absence of PTTG1 may also increase cell susceptibility to genotoxic stress (Bernal and Hernandez, 2007). PTTG1 regulates Ku heterodimer binding to DNA at double-strand break sites, and intracellular PTTG1 suppresses DNA damage repair by binding to Ku70 (Kim et al., 2007). Although non-homologous end joining is quantitatively normal in securin-deficient cells, this occurs through aberrant end processing (Bernal et al., 2008), which may also account for cell susceptibility to genotoxic stress. Thus, PTTG1 signaling after DNA damage could release Ku, translocating from the nucleus to the cytoplasm, allowing Ku binding to DNA, and enabling damage repair. The accumulation of cytoplasmic PTTG1 in turn inhibits Aurora-A from exerting G2 checkpoint control (Figure 6c).

PTTG1 has been proposed as a prognostic marker for differentiated thyroid cancer, lymph node invasion and breast cancer recurrence (Solbach et al., 2004), and colon cancer invasiveness (Heaney et al., 2000). Our results show that PTTG1 plays a role at the G2 checkpoint by interacting with Aurora kinase A and regulating responses to Aurora kinase inhibitors and DNA damage agents, suggesting that PTTG1 may act as a marker for predicting chemotherapy responses. The results also suggest that PTTG1 downregulation might enhance Aurora kinase inhibitor efficacy.

Materials and methods

Protein microarray screening

The human Protoarray nc3.0 study was performed as described by the manufacturer (Invitrogen, Carlsbad, CA, USA). PTTG1 protein was biotinylated using biotin-XX sulfosuccinimidyl ester, purified by gel filtration, assessed by western blot and appropriate protein aliquots used to probe a protein microarray (Protoarray nc-v3, Invitrogen) and attached biotinylate PTTG1 probed with Streptavidin-Alexa Fluor 647 conjugate. Arrays were scanned by GenePix 4000B (Molecular Devices Corporation, Sunnyvale, CA, USA), and images analysed using ProtoArray Prospector (Invitrogen).

Poly-His pull-down assays, in vitro transcription and co-immunoprecipitation

TNT Quick Coupled Transcription/Translation system (Promega, Madison, WI, USA) was used to generate His–PTTG1 and His–Aurora-A fusion proteins. The pull-down assay was performed with ProFound Pull-Down PolyHis Protein:Protein Interaction Kit (Pierce, IL, USA). Briefly, fusion proteins acting as bait were adsorbed to immobilized cobalt chelate gels. Prey proteins were biotin-lysine-labeled during synthesis, and incubated with bait proteins for 1 h at 4 °C. Bound proteins were eluted and analysed by SDS–PAGE (polyacrylamide gel electrophoresis) and detected by Transcend Non-Radioactive Translation Detection Systems (Promega, Madison, MI, USA). Co-immunoprecipitation was carried out with a ProFound Mammalian Co-Immunoprecipitation Kit. Control IgG and PTTG1 polyclonal antibody were conjugated to Antibody Coupling Gel and incubated overnight with HCT116 cell lysates in a rotating platform. Immunoprecipitates were eluted and analysed by western blotting.

Immunofluorescence microscopy

Cells grown on coverslips were fixed using 4% paraformaldehyde in phosphate buffer pH 7.4 for 20 min, permeabilized with 0.2% Triton X-100 in phosphate-buffered saline (PBS) for 10 min, and incubated with 1% bovine serum albumin (BSA) for 1 h. The following primary antibodies were included: polyclonal Aurora-A (Santa Cruz Technology, Santa Cruz, CA, USA), Phospho-histone (Ser10) (Cell Signaling, Boston, MA, USA) and monoclonal separase (Novus, Littleton, CO, USA) antibodies. Slides were incubated overnight at 4 °C with primary antibodies suspended in BSA 0.1% containing PBS. Coverslips were incubated with the relevant Alexa Fluor 488 or Alexa Fluor 568 secondary antibodies for 2 h and with TO-PRO-3 (Invitrogen) for DNA detection for 20 min, then washed with PBS-BSA 0.1%, dried at room temperature protected from light and mounted using Prolong Gold (Invitrogen).

Images were obtained using a Leica TCS SP confocal microscope. Ar laser 488 and ArKr laser 568 were used for the detection of Alexa fluorophores 488 and 568, respectively. The exclusion of autofluorescence was achieved by imaging with a narrow spectral detection window with the pinhole set to 1.0 Airy unit (AU) for optimal resolution. Depicted images were derived from maximum intensity projections of confocal stacks.

Protein degradation during mitosis

Mitotic cells for degradation sequence assessment were prepared by a modified synchronization regime. Cells were released from aphidicolin (10 ng/μl) (Sigma Aldrich, St Louis, MO, USA) into medium containing 400 ng/μl of nocodazole (Sigma Aldrich) and incubated for 12 h before harvesting by shake off. Mitotic cells were washed three times in ice-cold PBS and replaced in 37 °C prewarmed medium. Mitotic cells were collected for analysis.

In vitro kinase activity assay

Assays were performed using a modified Aurora-A assay kit from Cell Signaling. Histone H3 was purchased from Promega and control His-GST protein from Upstate (Billerica, MA, USA). His-PTTG1 was expressed in insect cells and purified by the Caltech Protein Expression Center. Aurora-A and substrate histone H3 were brought to a 30 μl volume containing 25 mM Tris-HCl (pH 7.5), 10 mM MgCl2, 5 mM, β-glycerophosphate, 0.1 mM Na3VO4, 20–200 μM ATP, 1 μg histone 3, 10 U GST-Aurora-A and increasing PTTG1 levels or control proteins (0.5, 1 and 2 μg). Reaction mixtures were incubated at 37 °C for 30 min, reactions stopped by the addition of 10 μl 4 × NuPAGE LDS sample buffer and reaction products denatured for 10 min at 95 °C. Proteins were separated by standard SDS–PAGE and analysed by western blotting.

Western blot

Protein extracts were resolved by NuPAGE 4–12% Bis-Tris Gel (Invitrogen). Samples were electroblotted onto polyvinylidene fluoride (PVDF) membrane (Invitrogen), and membranes blocked and incubated with primary antibody. Donkey anti-rabbit or anti-mouse (GE Healthcare, Piscataway, NJ, USA) antibodies were conjugated to horseradish peroxide to reveal immunocomplexes by enhanced chemiluminescence (Pierce).

Cell proliferation assay

Cell proliferation was evaluated using WST-1 cell proliferation reagent (Clontech, Mountain View, CA, USA). Briefly, 90 μl cells were seeded into 96-well plates and cultured overnight, treated with drugs for 48 h and 10 μl of premixed WST-1 cell proliferation reagent was added. Cells were incubated for a further 4 h, shaken and absorbance measured at 450 nm using a Victor 3 multiwell plate reader (Perkin Elmer, Waltham, MA, USA). Cell viability rate was calculated for each well as A450-treated cells/A450 control cells × 100% (A450: OD value at 450 nm).

Anchorage-dependent colony formation assay

Briefly, 20 000 cells were plated in a six-well culture dish, allowed to attach overnight and treated with depicted drug concentrations. Cells were cultured in standard culture medium containing indicated treatments for 12 days and medium re-freshened every 3 days. Colonies were stained with 0.5% crystal violet in methanol/acetic acid (3:1) and those composed of >50 cells were counted.

siRNA and transfection

Aurora kinase A siRNA (AM51331, sense: IndexTermGGCAACCAGUGUACCUCAUtt; anti-sense: 5′-IndexTermAUGAGGUACACUGGUUGCCtg-3′) and PTTG1 siRNA (16706, sense: 5′-IndexTermGUCUGUAAAGACCAAGGGAtt-3′; anti-sense: 5′-IndexTermUCCCUUGGUCUUUACAGACtt-3′) were purchased from Ambion (Austin, TX, USA). Transfections were carried out with lipofectamine 2000 according to protocol.

Statistics

Analysis was performed by a standard two-tailed Student's t-test and results plotted as mean±s.d.

Abbreviations

Aurora-A:

Aurora kinase A

PTTG1:

pituitary tumor transforming gene 1

References

  1. Akino K, Akita S, Mizuguchi T, Takumi I, Yu R, Wang XY et al. (2005). A novel molecular marker of pituitary tumor transforming gene involves in a rat liver regeneration. J Surg Res 129: 142–146.

  2. Bernal JA, Hernandez A . (2007). p53 stabilization can be uncoupled from its role in transcriptional activation by loss of PTTG1/securin. J Biochem (Tokyo) 141: 737–745.

  3. Bernal JA, Luna R, Espina A, Lazaro I, Ramos-Morales F, Romero F et al. (2002). Human securin interacts with p53 and modulates p53-mediated transcriptional activity and apoptosis. Nat Genet 32: 306–311.

  4. Bernal JA, Roche M, Mendez-Vidal C, Espina A, Tortolero M, Pintor-Toro JA . (2008). Proliferative potential after DNA damage and non-homologous end joining are affected by loss of securin. Cell Death Differ 15: 202–212.

  5. Boelaert K, Tannahill LA, Bulmer JN, Kachilele S, Chan SY, Kim D et al. (2003). A potential role for PTTG/securin in the developing human fetal brain. FASEB J 17: 1631–1639.

  6. Cheeseman IM, Anderson S, Jwa M, Green EM, Kang J, Yates III JR et al. (2002). Phospho-regulation of kinetochore-microtubule attachments by the Aurora kinase Ipl1p. Cell 111: 163–172.

  7. Chen SS, Chang PC, Cheng YW, Tang FM, Lin YS . (2002). Suppression of the STK15 oncogenic activity requires a transactivation-independent p53 function. EMBO J 21: 4491–4499.

  8. Chien W, Pei L . (2000). A novel binding factor facilitates nuclear translocation and transcriptional activation function of the pituitary tumor-transforming gene product. J Biol Chem 275: 19422–19427.

  9. Crosio C, Fimia GM, Loury R, Kimura M, Okano Y, Zhou H et al. (2002). Mitotic phosphorylation of histone H3: spatio-temporal regulation by mammalian Aurora kinases. Mol Cell Biol 22: 874–885.

  10. Donangelo I, Gutman S, Horvath E, Kovacs K, Wawrowsky K, Mount M et al. (2006). Pituitary tumor transforming gene overexpression facilitates pituitary tumor development. Endocrinology 147: 4781–4791.

  11. El-Naggar SM, Malik MT, Kakar SS . (2007). Small interfering RNA against PTTG: a novel therapy for ovarian cancer. Int J Oncol 31: 137–143.

  12. Eyers PA, Erikson E, Chen LG, Maller JL . (2003a). A novel mechanism for activation of the protein kinase Aurora A. Curr Biol 13: 691–697.

  13. Eyers PA, Erikson E, Chen LG, Maller JL . (2003b). A novel mechanism for activation of the protein kinase Aurora A. Curr Biol 13: 691–697.

  14. Gadea BB, Ruderman JV . (2005). Aurora kinase inhibitor ZM447439 blocks chromosome-induced spindle assembly, the completion of chromosome condensation, and the establishment of the spindle integrity checkpoint in Xenopus egg extracts. Mol Biol Cell 16: 1305–1318.

  15. Gil-Bernabe AM, Romero F, Limon-Mortes MC, Tortolero M . (2006). Protein phosphatase 2A stabilizes human securin, whose phosphorylated forms are degraded via the SCF ubiquitin ligase. Mol Cell Biol 26: 4017–4027.

  16. Heaney AP, Horwitz GA, Wang Z, Singson R, Melmed S . (1999). Early involvement of estrogen-induced pituitary tumor transforming gene and fibroblast growth factor expression in prolactinoma pathogenesis. Nat Med 5: 1317–1321.

  17. Heaney AP, Singson R, McCabe CJ, Nelson V, Nakashima M, Melmed S . (2000). Expression of pituitary-tumour transforming gene in colorectal tumours. Lancet 355: 716–719.

  18. Hirota T, Kunitoku N, Sasayama T, Marumoto T, Zhang D, Nitta M et al. (2003). Aurora-A and an interacting activator, the LIM protein Ajuba, are required for mitotic commitment in human cells. Cell 114: 585–598.

  19. Horn V, Thelu J, Garcia A, biges-Rizo C, Block MR, Viallet J . (2007). Functional interaction of Aurora-A and PP2A during mitosis. Mol Biol Cell 18: 1233–1241.

  20. Hornig NC, Knowles PP, McDonald NQ, Uhlmann F . (2002). The dual mechanism of separase regulation by securin. Curr Biol 12: 973–982.

  21. Hutterer A, Berdnik D, Wirtz-Peitz F, Zigman M, Schleiffer A, Knoblich JA . (2006). Mitotic activation of the kinase Aurora-A requires its binding partner Bora. Dev Cell 11: 147–157.

  22. Ikezoe T, Yang J, Nishioka C, Tasaka T, Taniguchi A, Kuwayama Y et al. (2007). A novel treatment strategy targeting Aurora kinases in acute myelogenous leukemia. Mol Cancer Ther 6: 1851–1857.

  23. Kakar SS, Malik MT . (2006). Suppression of lung cancer with siRNA targeting PTTG. Int J Oncol 29: 387–395.

  24. Kim DS, Buchanan MA, Stratford AL, Watkinson JC, Eggo MC, Franklyn JA et al. (2006). PTTG promotes a novel VEGF–KDR–ID3 autocrine mitogenic pathway in thyroid cancer. Clin Otolaryngol 31: 246.

  25. Kim DS, Franklyn JA, Smith VE, Stratford AL, Pemberton HN, Warfield A et al. (2007). Securin induces genetic instability in colorectal cancer by inhibiting double-stranded DNA repair activity. Carcinogenesis 28: 749–759.

  26. Krystyniak A, Garcia-Echeverria C, Prigent C, Ferrari S . (2006). Inhibition of Aurora A in response to DNA damage. Oncogene 25: 338–348.

  27. Lai Y, Xin D, Bai J, Mao Z, Na Y . (2007). The important anti-apoptotic role and its regulation mechanism of PTTG1 in UV-induced apoptosis. J Biochem Mol Biol 40: 966–972.

  28. Li JJ, Li SA . (2006). Mitotic kinases: the key to duplication, segregation, and cytokinesis errors, chromosomal instability, and oncogenesis. Pharmacol Ther 111: 974–984.

  29. Li M, York JP, Zhang P . (2007). Loss of Cdc20 causes a securin-dependent metaphase arrest in two-cell mouse embryos. Mol Cell Biol 27: 3481–3488.

  30. Lindon C, Pines J . (2004). Ordered proteolysis in anaphase inactivates Plk1 to contribute to proper mitotic exit in human cells. J Cell Biol 164: 233–241.

  31. Marumoto T, Hirota T, Morisaki T, Kunitoku N, Zhang D, Ichikawa Y et al. (2002). Roles of aurora-A kinase in mitotic entry and G2 checkpoint in mammalian cells. Genes Cells 7: 1173–1182.

  32. McCabe CJ, Khaira JS, Boelaert K, Heaney AP, Tannahill LA, Hussain S et al. (2003). Expression of pituitary tumour transforming gene (PTTG) and fibroblast growth factor-2 (FGF-2) in human pituitary adenomas: relationships to clinical tumour behaviour. Clin Endocrinol (Oxf) 58: 141–150.

  33. Mountzios G, Terpos E, Dimopoulos MA . (2008). Aurora kinases as targets for cancer therapy. Cancer Treat Rev 34: 175–182.

  34. Ohashi S, Sakashita G, Ban R, Nagasawa M, Matsuzaki H, Murata Y et al. (2006). Phospho-regulation of human protein kinase Aurora-A: analysis using anti-phospho-Thr288 monoclonal antibodies. Oncogene 25: 7691–7702.

  35. Pascreau G, Arlot-Bonnemains Y, Prigent C . (2003). Phosphorylation of histone and histone-like proteins by aurora kinases during mitosis. Prog Cell Cycle Res 5: 369–374.

  36. Pei L . (1999). Pituitary tumor-transforming gene protein associates with ribosomal protein S10 and a novel human homologue of DnaJ in testicular cells. J Biol Chem 274: 3151–3158.

  37. Pei L . (2000). Activation of mitogen-activated protein kinase cascade regulates pituitary tumor-transforming gene transactivation function. J Biol Chem 275: 31191–31198.

  38. Pei L . (2001). Identification of c-myc as a down-stream target for pituitary tumor-transforming gene. J Biol Chem 276: 8484–8491.

  39. Pei L, Melmed S . (1997). Isolation and characterization of a pituitary tumor-transforming gene (PTTG). Mol Endocrinol 11: 433–441.

  40. Pereira G, Schiebel E . (2003). Separase regulates INCENP-Aurora B anaphase spindle function through Cdc14. Science 302: 2120–2124.

  41. Prigent C, Dimitrov S . (2003). Phosphorylation of serine 10 in histone H3, what for? J Cell Sci 116: 3677–3685.

  42. Saez C, Japon MA, Ramos-Morales F, Romero F, Segura DI, Tortolero M et al. (1999). hpttg is over-expressed in pituitary adenomas and other primary epithelial neoplasias. Oncogene 18: 5473–5476.

  43. Sheleg SV, Peloponese JM, Chi YH, Li Y, Eckhaus M, Jeang KT . (2007). Evidence for co-operative transforming activity of human pituitary tumor transforming gene (PTTG) and HTLV-1 Tax. J Virol 81: 7894–7901.

  44. Solbach C, Roller M, Fellbaum C, Nicoletti M, Kaufmann M . (2004). PTTG mRNA expression in primary breast cancer: a prognostic marker for lymph node invasion and tumor recurrence. Breast 13: 80–81.

  45. Sun C, Chan F, Briassouli P, Linardopoulos S . (2007). Aurora kinase inhibition downregulates NF-kappaB and sensitises tumour cells to chemotherapeutic agents. Biochem Biophys Res Commun 352: 220–225.

  46. Tarabykin V, Britanova O, Fradkov A, Voss A, Katz LS, Lukyanov S et al. (2000). Expression of PTTG and prc1 genes during telencephalic neurogenesis. Mech Dev 92: 301–304.

  47. Tong Y, Tan Y, Zhou C, Melmed S . (2007). Pituitary tumor transforming gene interacts with Sp1 to modulate G1/S cell phase transition. Oncogene 26: 5596–5605.

  48. Vlotides G, Eigler T, Melmed S . (2007). Pituitary tumor-transforming gene: physiology and implications for tumorigenesis. Endocr Rev 28: 165–186.

  49. Vogel C, Hager C, Bastians H . (2007). Mechanisms of mitotic cell death induced by chemotherapy-mediated G2 checkpoint abrogation. Cancer Res 67: 339–345.

  50. Vogel C, Kienitz A, Muller R, Bastians H . (2005). The mitotic spindle checkpoint is a critical determinant for topoisomerase-based chemotherapy. J Biol Chem 280: 4025–4028.

  51. Waizenegger I, Gimenez-Abian JF, Wernic D, Peters JM . (2002). Regulation of human separase by securin binding and autocleavage. Curr Biol 12: 1368–1378.

  52. Walter AO, Seghezzi W, Korver W, Sheung J, Lees E . (2000). The mitotic serine/threonine kinase Aurora2/AIK is regulated by phosphorylation and degradation. Oncogene 19: 4906–4916.

  53. Wang H, Liu D, Wang Y, Qin J, Elledge SJ . (2001a). Pds1 phosphorylation in response to DNA damage is essential for its DNA damage checkpoint function. Genes Dev 15: 1361–1372.

  54. Wang Z, Melmed S . (2000). Pituitary tumor transforming gene (PTTG) transforming and transactivation activity. J Biol Chem 275: 7459–7461.

  55. Wang Z, Moro E, Kovacs K, Yu R, Melmed S . (2003). Pituitary tumor transforming gene-null male mice exhibit impaired pancreatic beta cell proliferation and diabetes. Proc Natl Acad Sci USA 100: 3428–3432.

  56. Wang Z, Yu R, Melmed S . (2001b). Mice lacking pituitary tumor transforming gene show testicular and splenic hypoplasia, thymic hyperplasia, thrombocytopenia, aberrant cell cycle progression, and premature centromere division. Mol Endocrinol 15: 1870–1879.

  57. Xue Y, Ren J, Gao X, Jin C, Wen L, Yao X . (2008). GPS 2.0: prediction of kinase-specific phosphorylation sites in hierarchy. Mol Cell Proteomics; e-pub ahead of print.

  58. Yang H, He L, Kruk P, Nicosia SV, Cheng JQ . (2006). Aurora-A induces cell survival and chemoresistance by activation of Akt through a p53-dependent manner in ovarian cancer cells. Int J Cancer 119: 2304–2312.

  59. Yu R, Lu W, Chen J, McCabe CJ, Melmed S . (2003). Overexpressed pituitary tumor-transforming gene causes aneuploidy in live human cells. Endocrinology 144: 4991–4998.

  60. Zhang D, Hirota T, Marumoto T, Shimizu M, Kunitoku N, Sasayama T et al. (2004). Cre-loxP-controlled periodic Aurora-A overexpression induces mitotic abnormalities and hyperplasia in mammary glands of mouse models. Oncogene 23: 8720–8730.

  61. Zhang D, Shimizu T, Araki N, Hirota T, Yoshie M, Ogawa K et al. (2008). Aurora A overexpression induces cellular senescence in mammary gland hyperplastic tumors developed in p53-deficient mice. Oncogene; e-pub ahead of print.

  62. Zhang Y, Ni J, Huang Q, Ren W, Yu L, Zhao S . (2007). Identification of the auto-inhibitory domains of Aurora-A kinase. Biochem Biophys Res Commun 357: 347–352.

  63. Zhou H, Kuang J, Zhong L, Kuo WL, Gray JW, Sahin A et al. (1998). Tumour amplified kinase STK15/BTAK induces centrosome amplification, aneuploidy and transformation. Nat Genet 20: 189–193.

  64. Zou H, McGarry TJ, Bernal T, Kirschner MW . (1999). Identification of a vertebrate sister-chromatid separation inhibitor involved in transformation and tumorigenesis. Science 285: 418–422.

Download references

Acknowledgements

PTTG1 WT and KO HCT116 cells were kindly provided by Dr Bert Vogelstein, Johns Hopkins University. PTTG1-GFP and PTTG1-PCDNA3.1 plasmids were kindly provided by Dr Run Yu. Supported by NIH Grant CA 75979 (SM), T32 DK007770, and The Doris Factor Molecular Endocrinology Laboratory.

Author information

Correspondence to S Melmed.

Additional information

Disclosure

The authors state no conflict of interest.

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)

Supplementary information

Supplementary data 1 (PDF 51 kb)

Supplementary data 2 (PDF 31 kb)

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • PTTG1
  • Aurora kinase A
  • drug response

Further reading