Upregulation of survivin during immortalization of ... - Nature

Oncogene (2009) 28, 2678–2689

& 2009 Macmillan Publishers Limited All rights reserved 0950-9232/09 $32.00 www.nature.com/onc

ORIGINAL ARTICLE

Upregulation of survivin during immortalization of nontransformed human fibroblasts transduced with telomerase reverse transcriptase J Yuan1,2, BM-P Yang1,2, Z-H Zhong3,4, I Shats5, M Milyavsky5, V Rotter5, RB Lock1,2, RR Reddel3,4 and KL MacKenzie1,2 1 Children’s Cancer Institute Australia for Medical Research, Randwick, New South Wales, Australia; 2University of New South Wales, Faculty of Medicine, Sydney, New South Wales, Australia; 3Children’s Medical Research Institute, Westmead, New South Wales, Australia; 4University of Sydney, Faculty of Medicine, Sydney, New South Wales, Australia and 5Weizmann Institute of Science, Department of Molecular Cell Biology, Rehovot, Israel

These investigations demonstrate that expression of the inhibitor of apoptosis family member, survivin, is dramatically increased during immortalization of nontransformed human fibroblasts that were transduced with telomerase reverse transcriptase (hTERT). Expression of survivin in immortalized fibroblasts peaked during G2/M phase of the cell cycle. However, the upregulation of survivin was dissociated from the rate of proliferation and proportion of G2/M cells. Depletion of survivin from immortal fibroblasts increased sensitivity to stress-induced apoptosis and resulted in an accumulation of cells with 4N DNA content. Conversely, overexpression of survivin in mortal fibroblasts conferred resistance to apoptosis. In contrast, very low levels of survivin in proliferating parental fibroblasts had no bearing on sensitivity to apoptosis. The upregulation of survivin did not appear to be a direct consequence of hTERT transduction. However, repression of hTERT resulted in the rapid downregulation of survivin in telomerase-immortalized fibroblasts and tumor cell lines, but not in cells immortalized via an Alternative Lengthening of Telomeres mechanism. These results have important therapeutic implications, as telomerase and survivin are both broadly expressed in human cancers. Selection during the immortalization process for cells expressing high levels of survivin may account for the abundance of survivin in diverse tumor types. Oncogene (2009) 28, 2678–2689; doi:10.1038/onc.2009.136; published online 1 June 2009 Keywords: survivin; telomerase; apoptosis; immortalization; fibroblasts Introduction The activation of a telomere maintenance mechanism during carcinogenesis prevents telomere shortening and Correspondence: Dr KL MacKenzie, Cancer Cell Development, Children’s Cancer Institute Australia for Medical Research, Sydney Children’s Hospital, Level 1, High St, PO Box 81, Randwick, New South Wales 2031, Australia. E-mail: [email protected] Received 28 October 2008; revised 17 April 2009; accepted 26 April 2009; published online 1 June 2009

enables proliferation beyond senescence (reviewed in Stewart and Weinberg, 2006). In 80–90% of cancers, telomeres are maintained through the activity of the enzyme telomerase, whereas approximately 10% of cancers maintain telomere length via a recombinationbased mechanism known as Alternative Lengthening of Telomeres (ALT). Telomerase is also expressed in germ cells and various stem and progenitor cells, but is downregulated during differentiation (Hiyama et al., 1995; Wright et al., 1996). The human telomerase holoenzyme is a ribonucleoprotein complex that includes a reverse transcriptase (hTERT) as a catalytic domain, an RNA component (hTR) that functions as a template for reverse transcription, and the RNA binding and modifying protein, dyskerin (Cohen et al., 2007). Ectopic expression of hTERT is sufficient to reconstitute telomerase enzyme activity, elongate telomeres and extend the lifespan of BJ foreskin fibroblasts and retinal epithelial cells (Bodnar et al., 1998). However, telomere length independent mechanisms that involve the p16INK4a/RB and/or p53 tumor suppressor pathways and/or other uncharacterized molecular events restrict the immortalization of many cell types, including human lung fibroblasts, bone marrow endothelial cells, keratinocytes and mammary epithelial cells (Kiyono et al., 1998; MacKenzie et al., 2000, 2002; Milyavsky et al., 2003; Taylor et al., 2004; Wen et al., 2006). Another common molecular feature of diverse tumor types is the abundant expression of survivin. Similar to telomerase, survivin is generally not detectable within the normal tissue surrounding tumors, although it has been demonstrated in normal human fetal tissue, testes and progenitor cells (Adida et al., 1998; Fukuda and Pelus, 2001). Survivin was assigned to the inhibitor of apoptosis protein (IAP) family because it has a N-terminal baculoviral IAP repeat (Ambrosini et al., 1997). However, it lacks a caspase recruitment domain and the carboxyterminal RING finger domain found in some other mammalian IAPs (Verhagen et al., 2001). An evolutionarily conserved nuclear export signal located between the baculoviral IAP repeat domain and carboxy terminus appears to be a key determinant of the dual functions of survivin in cell division and apoptosis (Li et al., 1999; Colnaghi et al., 2006; Knauer et al., 2006).

Upregulation of survivin during immortalization J Yuan et al

2679

Within the nucleus, survivin functions as an essential component of the chromosomal passenger complex (Uren et al., 2000). Hence, depletion of survivin in tumor cells was shown to delay mitosis and result in defective cell division and polyploidy (Li et al., 1999; Carvalho et al., 2003; Lens et al., 2003). Expression of survivin in tumor cells has also been shown to protect against apoptotic stimuli (Ambrosini et al., 1997; Wang et al., 2005; Knauer et al., 2007). The precise mechanisms underlying the cytoprotective functions of survivin remain to be fully elucidated, however, its cytoplasmic localization provides an opportunity for direct or indirect effects on components of the apoptotic machinery (Dohi et al., 2004; Arora et al., 2007). The cytoprotective function of survivin may account for the inverse correlation between the abundance of survivin in tumor cells and patient response to chemotherapy and radiotherapy (reviewed in Pennati et al., 2007). Elevated levels of survivin in tumor cells positively correlate with various indicators of aggressive disease and shorter disease-free or overall survival. These observations suggest that survivin is upregulated during carcinogenesis and contributes to disease progression. However, as expression of survivin is cell cycle and developmentally regulated, an alternative explanation is that high levels of survivin detected in neoplastic versus normal cells simply reflects the high rate of proliferation of undifferentiated neoplastic cells. Aside from correlations, the functional significance of survivin in premalignant cells of human origin has not yet been unequivocally demonstrated. In this study, it is shown that survivin is upregulated during immortalization of normal diploid fibroblasts. Elevated expression of survivin in immortalized fibroblasts protected against stress-induced apoptosis and showed no simple correlation with rate of proliferation. These results support the notion that the upregulation of survivin in cancer cells is reflective of an oncogenic event.

Results MRC5 cells acquire resistance to apoptosis during hTERT-mediated immortalization We previously demonstrated that hTERT-transduced MRC5 (MRC5hTERT) fibroblasts underwent a twofold increase in replicative lifespan before entering a telomere length-independent growth crisis, characterized by a reduced rate of expansion and an increased rate of cell death (MacKenzie et al., 2000; Taylor et al., 2004). A subset of cells eventually escaped crisis and became immortalized. The immortalized cells retained functional properties of normal cells; they retained mitogenic requirements, were anchorage dependent, arrested in response to DNA damage and antimitotic agents, and were nontumorigenic. The rates of proliferation (0.40±0.16 population doublings (PD)/day) and cell death (4.5±0.5% cell death by propidium iodide (PI) staining) in cultures of immortal cells were unaltered compared to early passage (25–55 PD) parental MRC5

cells (0.44±0.13 PD/day and 4.4±1.3% cell death). To determine whether the cells that escaped crisis and were immortalized had acquired resistance to cell death, the apoptotic potential of pre- and postcrisis MRC5hTERT cells was compared with parental MRC5 cells. In preliminary investigations, we determined that 100 nM okadaic acid (OA) and 1 mg/ml actinomycin D (ActD) very effectively induced apoptosis in MRC5 cells, whereas other agents, such as tumor necrosis factor-a, H2O2, Paclitaxel and Vincristine induced growth arrest and/or nonapoptotic cell death (data not shown). Apoptotic cell death was measured in MRC5 and MRC5hTERT cells treated with OA or ActD for 24 h by staining with Annexin V and PI. The results showed that postcrisis MRC5hTERT cells were significantly more resistant to apoptosis than parental MRC5 cells or precrisis MRC5hTERT cells (Po0.001; Figure 1a). In addition, precrisis MRC5hTERT cells were slightly more resistant to ActD than parental MRC5 cells (Po0.05). To confirm that apoptotic signaling was impaired in the immortal MRC5hTERT cells, caspase activation was investigated. Immunoblotting for the effector caspase, caspase-7, showed that the level of the proenzyme (35 kDa) in parental MRC5 and precrisis MRC5hTERT cells was reduced by treatment with OA or ActD (Figure 1b). The reduction in pro-caspase-7 in these cells was accompanied by the appearance of a band of the expected size for active caspase-7 (20 kDa). No active caspase-7 was detected in drug-treated postcrisis MRC5hTERT cells (320 PD). ELISA assay confirmed that the induction of caspase-3/7 enzyme activity was blunted in postcrisis MRC5hTERT cells (Student’s t-test, Po0.01; Figure 1c). The enzymatic activity of caspase-9, which acts in the intrinsic apoptotic pathway upstream of effector caspase-3 and caspase-7, was also significantly reduced in OA- and ActD-treated MRC5hTERT cells compared with treated MRC5 cells (Figure 1d). The reduction in caspase-9 activity was more dramatic in postcrisis than precrisis MRC5hTERT cells. Together, these results demonstrate that hTERT-immortalized MRC5 cells were defective in their ability to activate caspases involved in the intrinsic apoptotic pathway and consequently were resistant to cell death. The results also demonstrate a more moderate apoptotic defect in precrisis MRC5hTERT cells. Survivin is upregulated during immortalization of human fibroblasts The expression of survivin was assayed at various time points during immortalization. Quantitative real-time RT–PCR (qRT–PCR) and immunoblot analyses demonstrated very low levels of survivin mRNA and protein in proliferating MRC5 cells at 45 PD (Figure 2a). Survivin protein and mRNA expression were both moderately upregulated in the precrisis MRC5hTERT mass culture and were expressed at very high levels in postcrisis MRC5hTERT cells. Expression of survivin was also upregulated in seven out of nine Oncogene


2680

immortalized clones that were derived by limiting dilution of the MRC5hTERT mass culture at early passage (Taylor et al., 2004; Figure 2a, bottom panel). The level of expression of survivin in MRC5hTERT 40

Untreated

% Apoptotic cells

OA 30

ActD *

20

*

** *

** * * **

** *

10 0 PD

45

57

100

208

Pre-crisis

MRC5

316

Post-crisis

MRC5-hTERT

Jurkat H2O2

320 MRC5hTERT

57 MRC5hTERT

OA

45 MRC5

320 MRC5hTERT

57 MRC5hTERT

AcD

45 MRC5

320 MRC5hTERT

57 MRC5hTERT

PD

45 MRC5

Untreated

Pro-Casp 7 (35kDa) Active (20kDa) β-Actin (42kDa)

Caspase 3/7 activity

20 Untreated OA ActD

15 10

** * *

5 0 PD

45

57

MRC5

Pre-crisis

320 Post-crisis

MRC5-hTERT

Caspase 9 activity

6

**

4

0 PD

Untreated OA ActD

** *

*

2

** *

45

57

320

MRC5

Pre-crisis

Post-crisis

MRC5-hTERT Oncogene

cells bore no relationship to the proliferative rate, as the MRC5hTERT mass culture at 217 PD proliferated at the same rate as parental MRC5 cells at 45 PD (0.4 PD/ day; Figure 2a, bottom panel). Similarly, the rate of proliferation of MRC5hTERT clones (0.3–0.6 PD/day) was comparable to parental MRC5 cells at 45 PD, and did not correlate with the level of survivin expressed. Notwithstanding the lack of correlation between the proliferation and expression of survivin in early passage MRC5 and immortal MRC5hTERT cells, it was noted that survivin was not expressed in senescent MRC5 cells. To determine whether expression of survivin was cell cycle dependent, immortalized MRC5hTERT cells (316 PD) and parental MRC5 cells (45 PD) were arrested in G1, S and G2/M phases by treatment with mimosine, thymidine and nocodazole, respectively (Figure 2b). In immortal MRC5hTERT cells, survivin mRNA and protein were found to be highest in cells arrested in G2/M and lowest in G1-arrested cells. In contrast, expression of survivin was low in parental MRC5 cells at all phases of the cell cycle. Thus, the contrasting levels of survivin in unsynchronized MRC5 versus MRC5hTERT cells was not a reflection of the percentage of cells in a particular phase of the cell cycle, nor a consequence of increased proliferation. Immunofluorescence staining demonstrated that survivin was expressed at a very high level within both the cytoplasm and nucleus of immortal MRC5hTERT cells (316 PD; Figure 2c). There was a generally lower level of staining in both the cytoplasm and nucleus of precrisis MRC5hTERT cells (67 PD) and infrequent staining of parental MRC5 cells (45 PD). Nuclear and cytoplasmic expression of survivin was confirmed by immunoblotting proteins fractionated from the nucleus and cytosol (data not shown). There was no significant difference in the frequency of survivin-positive cells in cultures of immortal cells compared with precrisis MRC5hTERT cultures. The proportion of cells expressing nuclear versus cytoplasmic survivin was also very similar in precrisis and immortal MRC5hTERT cultures. These results indicate that the progressive upregulation of survivin in MRC5hTERT cells was principally due to an increased level of expression, rather than a substantially changed proportion of cells expressing survivin. Expression of survivin was also assessed in hTERTimmortalized WI-38 lung fibroblasts and JFCF6 jejunal fibroblasts, as well as in JFCF6 cells that were immortalized by transfection with SV40 early region

Figure 1 Immortal MRC5hTERT cells are resistant to apoptosis. MRC5 and MRC5hTERT cells were treated with 100 nM okadaic acid (OA) or 1 mg/ml actinomycin D (ActD) for 24 h. (a) Quantification of apoptotic cells by Annexin V-APC/PI staining. (b) Immunoblot detection of pro-caspase-7, and caspase-7 cleavage products. As a positive control for detection of active caspase-7, Jurkat cells were incubated with 100 mM H2O2 for 4 h. (c) Caspase-3/7 and (d) caspase-9 enzyme activity measured using caspase ELISA assay plate. Values were normalized to untreated cells and are presented as mean±s.d. calculated from three independent experiments. ***Po0.001, **Po0.01 and *Po0.05, Student’s t-test comparison with MRC5 cells.


2681

(ER) and spontaneous activation of either telomerase or ALT (Figure 3). Consistent with the trend observed in MRC5 cells, expression of survivin mRNA and protein was dramatically increased in immortalized WI38hTERT and JFCF6 cell lines, irrespective of the mode of immortalization of the JFCF6 cells. Survivin was also found to be upregulated in hTERT-immortalized BJ foreskin fibroblasts (data not shown). Elevated expression of survivin confers resistance to apoptosis in immortalized fibroblasts To investigate whether resistance to apoptosis was a direct consequence of the upregulation of survivin in MRC5hTERT cells, siRNA-mediated knockdown of survivin was performed. Postcrisis MRC5hTERT cells were transfected with four previously described siRNAs (S1–S4) or a scrambled oligonucleotide (Sc) as a control (Figure 4a). The most effective siRNA (S1), which consistently resulted in >95% mRNA knockdown, was used to test the antiapoptotic effect of survivin. Consistent with results in Figure 1, immortal MRC5hTERT cells (320 PD) transfected with Sc were resistant to treatment with OA or ActD. However, transfection of the immortal cells with S1 resulted in a 3.5-fold increase in spontaneous apoptosis (untreated cells; Po0.05, Student’s t-test) and significantly increased their sensitivity to OA- and ActD-induced apoptosis (Po0.05 and Po0.01, respectively, Student’s t-test; Figure 4a, right graph). In contrast to the immortal cells, transfection of precrisis MRC5hTERT and parental MRC5 cells with S1 had no significant effect on the level of spontaneous or drug-induced apoptosis. Similar results were obtained when survivin was functionally ablated in MRC5 and MRC5hTERT cells by transduction with a replication-defective adenoviral vector encoding a dominant negative mutant survivin (Adv-T34A; Supplementary Figure 1; Mesri et al., 2001). Consistent with the siRNA knockdown experiment, functional inactivation of survivin by Adv-T34A transduction restored the apoptotic response of postcrisis MRC5hTERT cells, but did not alter the level of apoptosis in OA- and ActD-treated precrisis MRC5hTERT and MRC5 cultures. Together, these results demonstrate that the high level of survivin expressed in immortal MRC5hTERT cells conferred resistance to stress-induced apoptotic cell death, whereas the low level of survivin expressed in normal MRC5 cells and precrisis MRC5hTERT cells was not protective. Additional knockdown experiments confirmed that survivin also confers resistance to apoptosis in other immortalized fibroblast cell lines, including hTERT-immortalized WI38 and ALT-immortalized MRC5 cells (Supplementary Figure 2). The effect of survivin upregulation was also investigated by transducing precrisis MRC5hTERT cells that had bypassed senescence (82 PD) and parental MRC5 cells (45 PD) with a retroviral vector encoding survivin (MIGR1-Survivin) or a control vector encoding green fluorescent protein (GFP) alone (MIGR1; Gurbuxani et al., 2005). Overexpression of survivin was verified by

immunoblot analysis and was shown to confer both MRC5 and MRC5hTERT cells with resistance to OAand ActD-induced apoptosis (Figure 4b). There was no difference in the rate of proliferation of MRC5 and MRC5hTERT cells transduced with survivin compared with mock or GFP-transduced cells over 30 days posttransduction (data not shown). These results confirm that an abundance of survivin confers resistance to apoptosis in nontransformed human fibroblasts. Repression of hTERT results in the downregulation of survivin in telomerase-immortalized fibroblasts and tumor cells To determine whether the upregulation of survivin in immortalized MRC5hTERT cells was linked to hTERT expression and/or telomerase enzyme activity, expression of hTERT was suppressed with siRNA. After screening eight different hTERT siRNAs for their effect on hTERT mRNA expression and telomerase enzyme activity, expression of survivin mRNA and protein were assessed in five transfectants that expressed substantially less hTERT mRNA (Figures 5a and b). A dramatic reduction in the amount of survivin protein and mRNA was detected in the MRC5hTERT cells that were transfected with the siRNAs that most effectively suppressed hTERT and telomerase activity (T7 and T8). To confirm the specificity of the siRNA, six additional telomerase-positive cell lines and two telomerase-negative (ALT) cell lines were transfected with T8 siRNA. The cell line panel included MRC5 cells immortalized by infection with SV40 ER and spontaneous activation of either telomerase (MRC5V1) or ALT (MRC5V2) (Huschtscha and Holliday, 1983), as well as four telomerase-positive tumor cell lines. qRT– PCR and Q-TRAP analyses confirmed that there was a dramatic downregulation of hTERT and telomerase enzyme activity in the telomerase-positive cells transfected with T8 (Figure 5c). Survivin mRNA and protein expression were also suppressed in telomerase-immortalized fibroblasts and tumor cells transfected with T8 (Figure 5d). In contrast, survivin was not downregulated in the T8-transfected ALT cell lines. This result demonstrates that the downregulation of survivin in telomerase-positive cells was a specific effect of suppression of hTERT and/or telomerase activity and not an off-target effect of the T8 siRNA. To discern between the possibilities that survivin was downregulated in T8-transfected cells as a consequence of repression of hTERT or inhibition of telomerase, immortalized MRC5hTERT cells and the A549 lung tumor cell line were transfected with siRNAs that targeted the RNA component of telomerase, hTR (TR2 and TR3). Transfection with TR2 suppressed expression of hTR and telomerase enzyme activity, but did not significantly alter survivin expression (Figures 6a–c). Thus, maintenance of high levels of survivin in telomerase-immortalized fibroblasts and tumor cell lines was dependent on continued expression of hTERT, but appeared to be independent of hTR and telomerase enzyme activity. Oncogene


2682 Survivin mRNA (% of HL60)

75

MRC5

50 PD

45

MRC5hTERT 58

100 210

315 HL60 Survivin (16.5 kDa)

25

β-actin (42 kDa) 0 PDs 45

60

57

MRC5

100

208

Pre-crisis

316

Post-crisis

MRC5hTERT MRC5 E

S

MRC5hTERT M

2

9

11 12 13 20 35 36 42 HL60 Survivin β-actin

0.4

PD/Day

MRC5-hTERT 316PD

G0/G1: 59% S: 21% G2/M: 20%

G0/G1: 55% S: 22% G2/M: 23% Control

G0/G1: 83%

G0/G1: 83%

6 5 Survivin mRNA

MRC5 45PD

- 0.4, 0.5, 0.5, 0.3, 0.4, 0.5, 0.6, 0.5, 0.5, 0.6

4 3 2 1

Cell count

Mimosine

0 C

S: 43%

G1

S G2/M C

MRC5 (45PD)

S: 48%

MRC5 (45PD)

Thymidine

C

G1

G1

S G2/M

MRC5hTERT(316PD)

MRC5-hTERT (316PD)

S G2/M C

G1

S

G2/M Survivin

G2/M: 59%

G2/M: 54%

β-actin

Nocodazole

DNA content

MRC5 45PD Nuclear

Cytoplasmic

100

MRC5hTERT 67PD

% positive

80 60 40 20

MRC5hTERT 316PD

Oncogene

0 PD

45 MRC5

67

316

MRC5hTERT


2683

Cell cycle analyses were performed to determine whether the downregulation of survivin in T8-transfected cells was due to depletion of cells in G2/M. In four independent experiments there was a 10–14% reduction in the percentage of S phase cells in T8 transfectants (Figure 6d; Supplementary Figure 3). This was concomitant to a small increase in the proportion of cells in G1 (9–11%) and G2/M (2–5%). There was no substantial increase in sensescence-associated b-galactosidase activity, which remained below approximately 2%

following repression of hTERT (data not shown). Thus, the lower level of survivin expressed in T8-transfected cells was not explained by a reduced proportion of cells in G2/M or by senescence. However, the downregulation of survivin may have been an indirect consequence of apparently defective S phase progression associated with repression of hTERT. There was no apparent effect of hTR knockdown on cell cycle profile. Cells depleted of survivin accumulated with 4N DNA content and developed polyploidy, as previously described

WI-38hTERT

100 Survivin mRNA

WI-38 80

25

clone #1 56

clone #2 59

102

102

HL60

PD

60 Survivin 40 β-actin

20 0 PD

25

40

55

93

53

clone # 1

WI38

92

HL60

clone #2

WI38hTERT SV40ER (Telomerase)

SV40ER (ALT)

T.1M

T.1L

HL60

T.1J/1-4D

T.1H

80, 113, 160, 150, 120, 140 116, 145

T.1F

9,

T.1J/6B

100

hTERT1

150

hTERT3

200 JFCF6

Survivin mRNA

250

PD

50

SV40ER (Telomerase)

HL60

T.1M

T.1L

T.1J/1-4D

T.1H

T.1F

T.1J/6B

80 113 160 150 120 140 116 145 hTERT3

9

hTERT1

0 PD

JFCF6

Survivin

β-actin

SV40ER (ALT)

Figure 3 Survivin is upregulation during immortalization of WI-38 and JFCF6 fibroblasts. Survivin mRNA (left panels) and protein (right panels) detected by quantitative real-time RT–PCR (qRT–PCR) and immunoblot, respectively. qRT–PCR values were normalized to HL60 and are mean±s.d. calculated from three independent experiments. (a) WI-38 parental and hTERT-transduced clones. (b) JFCF6 parental cells and clones immortalized following transfection with hTERT (hTERT1 and hTERT3), or by transfection with SV40 ER and spontaneous activation of telomerase or Alternative Lengthening of Telomeres (ALT).

Figure 2 Upregulation of survivin during hTERT-driven immortalization of MRC5 cells. (a) Survivin mRNA (left panel) and protein (top right and bottom panels) were detected by quantitative real-time RT–PCR (qRT–PCR) and immunoblot, respectively. Top immunoblot shows parental MRC5 and MRC5hTERT mass culture at the indicated population doublings (PD). Bottom immunoblot shows MRC5 parental cells at 45 PD (E), senescent MRC5 at 63 PD (S), MRC5hTERT mass culture (M) at 217 PD and MRC5hTERT clones at 194–221 PD. The clone numbers are indicated above, and rate of proliferation below the blot. (b) Left: Cell cycle analysis by propidium iodide (PI) staining MRC5 and immortal MRC5hTERT cells arrested in G0/G1, S and G2/M phases by 24 h treatment with 400 mM mimosine, 2 mM thymidine or 0.1 mg/ml nocodazole, respectively. Right: Expression of survivin mRNA and protein in asynchronous and arrested cells assessed by qRT–PCR (top right) and immunoblotting (bottom right), respectively. C: control (no treatment). (c) Immunofluorescence detection of survivin (red) with 40 ,6-diamidino-2-phenylindole (DAPI; blue) stained nuclei. qRT–PCR values were normalized to HL60 and are mean±s.d. calculated from three independent experiments. Oncogene


2684 30

120

ActD

OA

Untreated % apoptotic cells

Survivin mRNA

100 48 h 72 h

80 60 40 20

** 20

* *

10

0 M

Sc

S1

S2

S3

48h

0

S4

S1

Sc

72h

Sc

MRC5 45PD

HL60 M Sc S1 S2 S3 S4 M Sc S1 S2 S3 S4

S1

Sc

MRC5hTERT 57PD

S1

MRC5hTERT 320PD

Survivin β-actin

Untreated 30

Survivin β-actin

% apoptotic cells

HL60

*

M

S

Surv

GFP

** **

MRC5hTERT (82PD) Mock

Surv

GFP

Mock

MRC5 (52PD)

ActD

OA

* 20

10

0 GFP

Survivin

MRC5

GFP

Survivin MRC5hTERT

Figure 4 Upregulation of survivin confers resistance to stress-induced apoptosis. (a) Immortal MRC5hTERT cells (320 population doublings (PD)) were transfected with siRNAs targeting survivin (S1–S4) or a scrambled oligonucleotide (Sc), or were mocktransfected (M). Survivin mRNA and protein expression were quantified in transfected cells by quantitative real-time RT–PCR (qRT– PCR; top left) and immunoblot (bottom left). Twenty-four hours after transfection with S1 or Sc, MRC5 and MRC5hTERT cells were treated with okadaic acid (OA) or actinomycin D (ActD) for another 24 h. Apoptosis was quantified by Annexin V-APC/PI staining (right graph). Results are mean±s.d. of three independent experiments. (b) MRC5 and precrisis MRC5hTERT cells were transduced with MIGR1-survivin (Surv) or control MIGR1 (green fluorescent protein; GFP) retroviral vectors. GFP-positive cells were FACS sorted. Survivin protein was detected by immunoblotting (left). S: senescent MRC5, M: immortalized MRC5hTERT cells. Transduced cells were treated with OA or ActD for 24 h and apoptosis was determined by Annexin V-APC/PI staining (right). Results are mean±s.d. of three independent assays. **Po0.01 and *Po0.05, Student’s t-test.

(Figure 6d; Supplementary Figure 3) (Carvalho et al., 2003; Lens et al., 2003).

Discussion In these investigations, isogenic cells were employed to demonstrate that expression of survivin increases dramatically during hTERT-driven immortalization of nontransformed fibroblasts. Increased expression of survivin was demonstrated in three hTERT-immortalized fibroblast strains (MRC5, WI-38 and JFCF6), and was also found to be upregulated in hTERT-immortaOncogene

lized BJ foreskin fibroblasts (additional data not shown). The expression of survivin in hTERT-immortalized MRC5 fibroblasts was cell cycle dependent, peaking in G2/M phase, as previously shown in tumor cells (Li et al., 1998). However, there was no relationship between the level of survivin and the doubling rate of early passage mortal MRC5 and immortal MRC5hTERT cells. We have previously shown by bromodeoxyuridine and PI staining that hTERT-immortalized MRC5 and early passage MRC5 cells have a very similar cell cycle profile (MacKenzie et al., 2000). These observations, together with data in this report showing that survivin was expressed at a very low level in G2/M arrested parental MRC5 cells, indicate that the


2685

high level of survivin expressed in immortal MRC5hTERT cells was not simply a reflection of the proportion of mitotic cells.

a

120

hTERT mRNA Telomerase activity

100 % Sc

80 60 40 20 0 Sc T1 T2 T3 T4 T5 T6 T7 T8 siRNAhTERT

b Survivin mRNA

100 80 60 40 20 0 Sc

T1

T4

T5

T7

T8

Survivin β-actin hTERT mRNA

c

Telomerase activity

140 120 % Sc

100 80 60 40 20 0

d

T1 08 0

H

M C F7

49

Telomerase +

140

Sc

120 Survivin mRNA

A5

eL a H

F6 C JF

W I3

8h

hT

TE

RT 1

ER T1

Sc T8 Sc T8 Sc T8 Sc T8 Sc T8 Sc T8

Immunoblotting and qRT–PCR analyses demonstrated that survivin was moderately upregulated in precrisis MRC5hTERT cells and was expressed at a very high level in immortal MRC5hTERT cells. Immunofluorescence staining revealed that survivin was upregulated in most of the cells (approximately 80%) in both the pre- and postcrisis cultures, although the intensity of the staining was higher in the postcrisis cells. These results indicate that the progressive upregulation of survivin detected by immunoblotting reflected an increased level of survivin expressed at the cellular level, rather than an increased proportion of survivin-positive cells in the culture. However, the upregulation of survivin appears to be an indirect effect of hTERT expression, as there was no substantial induction of survivin immediately upon transduction of MRC5 cells with hTERT (data not shown). The progressive upregulation of survivin observed in MRC5hTERT cultures most likely reflects selection during immortalization for cells that express very high levels of survivin. A subpopulation of cells expressing higher levels of survivin may have arisen as a consequence of spontaneous molecular alterations occurring during propagation of the hTERT-transduced cells. Higher levels of survivin may have provided a selective advantage by protecting cells against stresses encountered during immortalization. This possibility is consistent with our survivin knockdown experiments that showed the propensity of immortal MRC5hTERT cells to resist stress-induced apoptosis was dependent on high levels of survivin, whereas there was no apparent survival advantage conferred by low to moderate levels of survivin expressed in mortal MRC5 and precrisis MRC5hTERT cells. Selection for cells expressing high levels of survivin during immortalization may account for the very broad and abundant expression of survivin among diverse cancer types, as most cancers arise from an immortalized cell (Reddel, 2000).

ALT + T8

100 80 60 40 20 0 WI38 JFCF6 HeLa hTERT hTERT Sc T8

Sc T8

Sc T8

A549

MCF7 HT1080

MRC5 MRC5 SaOS-2 V1 V2

Sc T8

Sc T8

Sc T8

Sc T8 Sc T8 Sc T8

Survivin

β-actin

Figure 5 siRNA-mediated suppression of hTERT downregulates telomerase enzyme activity and survivin in telomerasepositive immortal fibroblasts and tumor cell lines. (a) Immortal MRC5hTERT cells were transfected with siRNAs targeting hTERT (T1–T8) or a scrambled siRNA control (Sc). hTERT mRNA and telomerase enzyme activity were measured by quantitative real-time RT–PCR (qRT–PCR) and Q-TRAP, respectively, 48 h after transfection. (b) Expression of survivin in five hTERT siRNA transfectants that had low hTERT expression as assessed by qRT–PCR (top) and immunoblot analysis (bottom) 48 h after transfection. (c) Quantitation of hTERT mRNA by qRT–PCR and telomerase enzyme activity by Q-TRAP in immortal fibroblasts and tumor cell lines transfected with T8 siRNA or Sc. (d) Expression of survivin mRNA determined by qRT–PCR and survivin protein detected by immunoblot analysis of immortal fibroblasts and tumor cell lines transfected with T8 siRNA or Sc. The black vertical lines within the immunoblot image separates different gels and/or indicate where lanes were rearranged for cohesion in this composite figure. For all mRNA and Q-TRAP quantitations, values were normalized to cells transfected with Sc and are presented as mean±s.d. of three independent experiments. Oncogene


2686

a

A549

MRC5hTERT

b Telomerase activity

hTR mRNA

A549

MRC5hTERT 100

120 100 80 60 40 20

80 60 40 20

0 Sc

TR2

0

TR3

Sc A549

MRC5hTERT

d

TR3 G2/M

S

G1 100 75 50 25

2) -h

si

R

TR

T

(T

R

(T

(S n

-s

ur

vi

vi

ra si

Survivin

Sc

Sc TR2 TR3 T8 Sc TR2 TR3 T8 HL60

8)

1)

0

A549

TE

T8

-h

TR3

m

MRC5hTERT

TR2

si

Sc

e

140 120 100 80 60 40 20 0

bl

Survivin mRNA

c

TR2

β-actin

Figure 6 Downregulation of survivin following suppression of hTERT is not due to suppression of telomerase enzyme activity or depletion of G2/M cells. (a–c) Postcrisis MRC5hTERT and A549 cells were transfected with siRNA targeting hTR (TR2 and TR3), or hTERT (T8). (a) Quantitation of hTR mRNA expression by quantitative real-time RT–PCR (qRT–PCR) 48 h after siRNA transfection. (b) Quantitation of telomerase activity by Q-TRAP 48 h after transfection. (c) Quantitation of survivin mRNA expression by qRT–PCR (top panel) and survivin protein detection by immunoblot (bottom panel) 48 h after transfection. Values were normalized to cells transfected with Sc. Results are mean±s.d. of three independent experiments. (d) Cell cycle analysis was performed on immortalized MRC5hTERT cells transfected with Sc, S1, T8 or TR2 by propidium iodide (PI) staining 48 h after transfection. The data are average values from four independent experiments. An example of FACS results showing the cell cycle profiles of siRNAtransfected cells is provided in Supplementary Figure 3.

Our results showing that telomerase activity was abundantly expressed in SV40 ER-transformed fibroblasts that were immortalized via ALT indicate that telomerase activity was not necessary for the upregulation of survivin. The upregulation of survivin in SV40 ER-transformed cells was not unexpected, because SV40 large T antigen inactivates p53 and pRB, both of which negatively regulate the expression of survivin (Hoffman et al., 2002; Mirza et al., 2002; Jiang et al., 2004). In concordance with these observations, a recent study demonstrated that survivin was upregulated in human melanocytes transfected with SV40 ER (Raj et al., 2008). However, the upregulation of survivin in MRC5hTERT cells cannot be attributed to inactivity of p53 and pRB, which were shown to function in response to DNA damage in immortal MRC5hTERT cells (Taylor et al., 2004). Immunoblot analysis of the panel of MRC5hTERT clones also showed no correlation between the abundance of survivin and inactivation p16INK4a, which is one of the upstream regulators of pRB (data not shown). A number of other signaling pathways that are implicated in the upregulation of survivin in Oncogene

cancer cells warrant investigation for potential involvement in the upregulation of survivin during immortalization of hTERT-transduced fibroblasts (Zhang et al., 2006). In particular, it will be important to determine whether a defect in transforming growth factor-b (TGFb) signaling, which was identified in immortal MRC5hTERT and WI38hTERT cells, contributes to the high level of survivin expressed in these cells (Milyavsky et al., 2007). This possibility is supported by two recent studies that have shown TGFb suppresses expression of survivin (Wang et al., 2008; Yang et al., 2008). The abundant expression of survivin in the immortal cells was dependent on continued expression of hTERT in all telomerase-positive cell lines tested, including cell lines with defective p53. Notably, survivin was downregulated within 48 h of hTERT-siRNA transfection, which precludes the possible involvement of critical telomere shortening, particularly since immortal MRC5hTERT cells have telomeres that are greater than 20 kbp in length (MacKenzie et al., 2000; Taylor et al., 2004). Downregulation of survivin following


2687

ablation of hTERT was also not due to a reduction in the proportion of cells in G2/M phase. However, hTERT knockdown did result in an apparent cell cycle defect, evidenced by a reduced proportion of cells in S phase, which may have contributed to the downregulation of survivin. Downregulation of telomerase enzyme activity due to siRNA-mediated repression of hTR had no effect on cell cycle progression or the expression of survivin. Thus, it appears that it was the downregulation of hTERT, rather than telomerase enzyme activity, that caused the cell cycle defect and suppression of survivin in T8-transfected cells. It is also possible that the downregulation of survivin was a direct or indirect effect of chromatin reorganization, as hTERT repression in human fibroblasts was previously shown to alter chromatin architecture (Masutomi et al., 2005). However, the repression of survivin in hTERTsiRNA-transfected cells was not reflective of a generalized effect on gene expression, as qRT–PCR analyses showed that the expression of two other IAPs (cIAP-1 and cIAP-2) were unaltered in hTERT siRNA-transfected immortal MRC5hTERT cells (data not shown). In contrast to telomerase-positive cells, downregulation of hTERT had no effect on the expression of survivin in ALT-immortalized cells lines, including MRC5 cells immortalized by the combined action of SV40 ER and ALT. These results confirm the specificity the hTERT-siRNA and underscore the crucial role of hTERT in maintaining high levels of survivin in telomerase-positive cells. It was previously shown that hTERT-immortalized fibroblasts are resistant to stress-induced death, although the mechanism underlying this effect was not elucidated (Gorbunova et al., 2002). The present investigation provides compelling evidence that resistance to cell death in telomerase-immortalized cells is a consequence of the upregulation of survivin. siRNAmediated knockdown of survivin in immortalized MRC5hTERT cells was shown to significantly increase the rate of spontaneous apoptosis and sensitize cells to drug-induced apoptosis. Conversely, overexpression of survivin in parental MRC5 and precrisis MRC5hTERT cells conferred resistance to drug-induced cell death. In contrast to immortal cells, spontaneous and druginduced apoptosis was not significantly altered by the downregulation of survivin in parental MRC5 or precrisis MRC5hTERT cells. Thus it appears that the abundant expression of endogenous survivin in immortal cells promoted their survival, whereas the low level of survivin expressed in precrisis MRC5hTERT and normal parental MRC5 cells was not protective. These results are consistent with studies that have shown survivin was dispensable for survival of normal human cells, as well as numerous reports that have demonstrated cytoprotection associated with high levels of survivin in tumor cells (Li et al., 1998; Mesri et al., 2001; Yang et al., 2004; Rodel et al., 2005). The sensitivity of the immortal, but not mortal cells, to the ablation of survivin is also consistent with the oncogene addiction hypothesis posed to explain the dependence of cancer cells on continued expression of specific oncogenes (Weinstein and Joe, 2008).

In summary, our investigations show that the expression of survivin is upregulated during hTERT-mediated immortalization of nontransformed human cells, and renders telomerase-immortalized cells resistant to apoptosis. These results support the notion that the upregulation of survivin in cancer cells is reflective of an oncogenic event and not simply a reflection of a higher mitotic index. Furthermore, we have shown that inhibition of hTERT very effectively suppresses both telomerase activity and expression of survivin. These results have broad therapeutic implications, as the activities of telomerase and survivin together underpin cellular survival, immortality and drug resistance; three properties that commonly manifest in diverse tumor types.

Materials and methods Cell culture Primary and immortal fetal lung fibroblasts, MRC5 and WI-38 were grown in a-Minimal Essential Medium (Invitrogen, Carlsbad, CA, USA) plus 10% fetal bovine serum (FBS; Thermo Scientific, Waltham, MA, USA) as previously described (Huschtscha and Holliday, 1983; MacKenzie et al., 2000; Milyavsky et al., 2003). All other cell lines were maintained in Dulbecco’s Modified Eagle’s Medium (Invitrogen) plus 10% FBS. PD were calculated as PD ¼ logN/log2, where N is expansion. JFCF6 jejunal fibroblasts were immortalized following transfection with pCI-hTERT (Colgin et al., 2000) or SV40 ER. SV40 ER-transfected JFCF6 clones that spontaneously escaped crisis were isolated and determined to have either activated telomerase (JFCF6 T.1J/6B, T.1F, and T.1H) or maintained telomeres via ALT (JFCF-6T.1J/1-4D, T.1L, and T.1M; P Bonnefin and R Reddel, unpublished). Apoptosis assays The proportion of apoptotic cells was assessed by costaining with allophycocyanin (APC)-labeled Annexin V (BD Biosciences, San Jose, CA, USA) and PI (BD Biosciences) according to the manufacturer’s instructions. Annexin V-positive and PI-negative cells were quantified using a FACS Calibur flow cytometer (BD Biosciences). Following 24-h treatment with 100 nM OA (Sigma, St Louis, MO, USA) or 1 mg/ml Act D (Sigma), the enzymatic activities of caspase-3/7 and caspase-9 were measured using a BD Apolert Caspase Assay Plate and incorporating inhibitors of caspase-3/7 and caspase-9 (DEVD-fmk and LEHD-fmk, respectively) as controls (BD Biosciences). Immunoblot and immunofluorescence detection of survivin Immunoblotting was performed as previously described (Taylor et al., 2004) using a rabbit anti-survivin antibody (NB500-201), purchased from Novus Biologicals (Littleton, CO, USA; Carvalho et al., 2003), and an anti-caspase-7 rabbit polyclonal antibody from BD Biosciences. The membranes were stripped with Restore Western Blot Stripping Buffer (Pierce, Rockford, IL, USA) and reprobed using a rabbit antihuman actin antibody (Sigma). For immunofluorescence staining, cells were grown on a chamber slide (BD Biosciences), then fixed with freshly prepared 4% cold paraformaldehyde (ICN, Bryan, OH, USA) for 20 min, washed with phosphate-buffered saline (PBS) and permeabilized with cold 0.02% Triton X-100 (Sigma) for 30 s. The cells were then Oncogene


2688 incubated with anti-survivin antibody in 0.5% bovine serum albumin/PBS and incubated with Alexa Fluor 594 goat antirabbit immunoglobulin-G (Invitrogen) at room temperature for 1 h. After washing with PBS, the cells were incubated in 0.3 mM 40 ,6-diamidino-2-phenylindole (DAPI; Invitrogen) for nuclear staining. The stained cells were viewed under a Zeiss Axioplan2 fluorescence microscope using an EpiPlan NeoFluor HD DIC 20 Objective (Carl Zeiss, Germany). Images were captured with a SensiCam CCD camera (Pco Computer Optics, Germany) using Image-ProPlus v4.1 software (Media Cybernetics Inc. Bethesda, MD, USA). Two investigators, blinded to the treatment, quantified survivin expression within the cytoplasm and nucleus. Five to six fields of view, each including 17–54 cells, were counted by each investigator.

Cell cycle analysis Cells were fixed in 70% ethanol, washed with PBS and stained with 100 mg/ml PI in a solution containing 50 mg/ml DNasefree RNase (Roche, Indianapolis, IN, USA). The percentage of cells in G1, S and G2/M phases were determined by FACS analysis using Cell Quest (BD Biosciences) and Modfit software (Verity Software House, Topsham, ME, USA).

Retroviral transductions Transductions with retroviral MIGR1-survivin or MIGR1 were performed using Phoenix A packaging cells as previously described (Taylor et al., 2004; Gurbuxani et al., 2005). Transduced cells were FACS-sorted based on GFP expression using a FACS Vantage cell sorter (BD Biosciences). siRNA transfections Oligonucleotide siRNAs with previously published sequence were chemically synthesized (Sigma) with a two-nucleotide overhang at the 30 end (Supplementary Table 1). Five additional hTERT siRNAs were designed using the BLOCK-iT RNAi designer program (Invitrogen) and the BIOPREDsi algorithm (Novartis Institutes for Biomedical Research, Basel, Switzerland) provided online by the Friedrich Miescher Institute for Biomedical Research, Basel. A scrambled siRNA (Sc) was used as a control. Cells were transfected with 100 nM siRNAs using Lipofectamine RNAiMAX reagent (Invitrogen) according to the manufacturer’s recommended protocol.

Conflict of interest The authors declare no conflict of interest.

qRT–PCR and Q-TRAP RNA was isolated using Trizol reagent and reverse transcribed using Superscript III reverse transcriptase and random primers (Invitrogen). Quantitative PCR reactions were performed on an ABI PRISM 7500 real-time PCR machine using ABI universal SYBR Green Master Mix (Applied Biosystems, Foster City, CA, USA). The sequence of primers for survivin were forward 50 -tctgcttcaaggagctggaagg-30 , reverse 50 -ttgttcttggctctttctctgtcc30 ; primers for hTERT were forward 50 -tgacacctcacctcacccac-30 , reverse 50 -cactgtcttccgcaagttcac-30 ; primers for hTR were forward 50 -cgctgtttttctcgctgactt-30 , reverse 50 -tgctctagaatgaacggtggaa-30 ; primers for glyceraldehyde 3-phosphate dehydrogenase (GAPDH) were forward 50 -acccactcctccacctttga-30 , reverse 50 -ctgttgctgtagccaaattcgt-30 . All samples were amplified in duplicate and results were normalized to expression of GAPDH. Telomerase enzyme activity was quantified using the real-time PCR based Quantitative Telomerase Detection kit (Allied Biotech, Ijamsville, MD, USA) with previously described modifications (Schuller et al., 2007).

Acknowledgements We thank Dr John D Crispino (The Ben May Institute for Cancer Research, IL) for providing us with retroviral MIGRSurvivin, Dr Dario C Altieri (University of Massachusetts Medical School, MA, USA) for providing the Adv-T34A adenoviral vector, Dr Gary Nolan (Stanford University, CA, USA) for Phoenix A packaging cells and Dr Lily Huschtscha, Jane Noble and Dr Paul Bonnefin (Children’s Medical Research Institute, New South Wales Australia) for providing immortalized cells. We also thank Laura Veas for proofreading this article. This work was supported by funds from the National Health and Medical Research Council (ID no. 510378 and 568704; KLM), New South Wales Cancer Council (KLM) and Cancer Institute New South Wales (KLM). Children’s Cancer Institute Australia is affiliated with the University of New South Wales and Sydney Children’s Hospital.

References Adida C, Crotty PL, McGrath J, Berrebi D, Diebold J, Altieri DC. (1998). Developmentally regulated expression of the novel cancer anti-apoptosis gene survivin in human and mouse differentiation. Am J Pathol 152: 43–49. Ambrosini G, Adida C, Altieri DC. (1997). A novel anti-apoptosis gene, survivin, expressed in cancer and lymphoma. Nat Med 3: 917–921. Arora V, Cheung HH, Plenchette S, Micali OC, Liston P, Korneluk RG. (2007). Degradation of survivin by the X-linked inhibitor of apoptosis (XIAP)-XAF1 complex. J Biol Chem 282: 26202– 26209. Bodnar AG, Ouellette M, Frolkis M, Holt SE, Chiu CP, Morin GB et al. (1998). Extension of life-span by introduction of telomerase into normal human cells. Science 279: 349–352. Carvalho A, Carmena M, Sambade C, Earnshaw WC, Wheatley SP. (2003). Survivin is required for stable checkpoint activation in taxoltreated HeLa cells. J Cell Sci 116: 2987–2998. Oncogene

Cohen SB, Graham ME, Lovrecz GO, Bache N, Robinson PJ, Reddel RR. (2007). Protein composition of catalytically active human telomerase from immortal cells. Science 315: 1850–1853. Colgin LM, Wilkinson C, Englezou A, Kilian A, Robinson MO, Reddel RR. (2000). The hTERTalpha splice variant is a dominant negative inhibitor of telomerase activity. Neoplasia 2: 426–432. Colnaghi R, Connell CM, Barrett RM, Wheatley SP. (2006). Separating the anti-apoptotic and mitotic roles of survivin. J Biol Chem 281: 33450–33456. Dohi T, Okada K, Xia F, Wilford CE, Samuel T, Welsh K et al. (2004). An IAP-IAP complex inhibits apoptosis. J Biol Chem 279: 34087–34090. Fukuda S, Pelus LM. (2001). Regulation of the inhibitor-of-apoptosis family member survivin in normal cord blood and bone marrow CD34(+) cells by hematopoietic growth factors: implication of survivin expression in normal hematopoiesis. Blood 98: 2091–2100.


2689 Gorbunova V, Seluanov A, Pereira-Smith OM. (2002). Expression of human telomerase (hTERT) does not prevent stress-induced senescence in normal human fibroblasts but protects the cells from stress-induced apoptosis and necrosis. J Biol Chem 277: 38540–38549. Gurbuxani S, Xu Y, Keerthivasan G, Wickrema A, Crispino JD. (2005). Differential requirements for survivin in hematopoietic cell development. Proc Natl Acad Sci USA 102: 11480–11485. Hiyama K, Hirai Y, Kyoizumi S, Akiyama M, Hiyama E, Piatyszek MA et al. (1995). Activation of telomerase in human lymphocytes and hematopoietic progenitor cells. J Immunol 155: 3711–3715. Hoffman WH, Biade S, Zilfou JT, Chen J, Murphy M. (2002). Transcriptional repression of the anti-apoptotic survivin gene by wild type p53. J Biol Chem 277: 3247–3257. Huschtscha LI, Holliday R. (1983). Limited and unlimited growth of SV40-transformed cells from human diploid MRC-5 fibroblasts. J Cell Sci 63: 77–99. Jiang Y, Saavedra HI, Holloway MP, Leone G, Altura RA. (2004). Aberrant regulation of survivin by the RB/E2F family of proteins. J Biol Chem 279: 40511–40520. Kiyono T, Foster SA, Koop JI, McDougall JK, Galloway DA, Klingelhutz AJ. (1998). Both Rb/p16INK4a inactivation and telomerase activity are required to immortalize human epithelial cells. Nature 396: 84–88. Knauer SK, Bier C, Habtemichael N, Stauber RH. (2006). The Survivin-Crm1 interaction is essential for chromosomal passenger complex localization and function. EMBO Rep 7: 1259–1265. Knauer SK, Kramer OH, Knosel T, Engels K, Rodel F, Kovacs AF et al. (2007). Nuclear export is essential for the tumor-promoting activity of survivin. FASEB J 21: 207–216. Lens SM, Wolthuis RM, Klompmaker R, Kauw J, Agami R, Brummelkamp T et al. (2003). Survivin is required for a sustained spindle checkpoint arrest in response to lack of tension. EMBO J 22: 2934–2947. Li F, Ackermann EJ, Bennett CF, Rothermel AL, Plescia J, Tognin S et al. (1999). Pleiotropic cell-division defects and apoptosis induced by interference with survivin function. Nat Cell Biol 1: 461–466. Li F, Ambrosini G, Chu EY, Plescia J, Tognin S, Marchisio PC et al. (1998). Control of apoptosis and mitotic spindle checkpoint by survivin. Nature 396: 580–584. MacKenzie KL, Franco S, May C, Sadelain M, Moore MA. (2000). Mass cultured human fibroblasts overexpressing hTERT encounter a growth crisis following an extended period of proliferation. Exp Cell Res 259: 336–350. MacKenzie KL, Franco S, Naiyer AJ, May C, Sadelain M, Rafii S et al. (2002). Multiple stages of malignant transformation of human endothelial cells modelled by co-expression of telomerase reverse transcriptase, SV40 T antigen and oncogenic N-ras. Oncogene 21: 4200–4211. Masutomi K, Possemato R, Wong JM, Currier JL, Tothova Z, Manola JB et al. (2005). The telomerase reverse transcriptase regulates chromatin state and DNA damage responses. Proc Natl Acad Sci USA 102: 8222–8227. Mesri M, Wall NR, Li J, Kim RW, Altieri DC. (2001). Cancer gene therapy using a survivin mutant adenovirus. J Clin Invest 108: 981–990. Milyavsky M, Shats I, Cholostoy A, Brosh R, Buganim Y, Weisz L et al. (2007). Inactivation of myocardin and p16 during malignant transformation contributes to a differentiation defect. Cancer Cell 11: 133–146.

Milyavsky M, Shats I, Erez N, Tang X, Senderovich S, Meerson A et al. (2003). Prolonged culture of telomerase-immortalized human fibroblasts leads to a premalignant phenotype. Cancer Res 63: 7147–7157. Mirza A, McGuirk M, Hockenberry TN, Wu Q, Ashar H, Black S et al. (2002). Human survivin is negatively regulated by wild-type p53 and participates in p53-dependent apoptotic pathway. Oncogene 21: 2613–2622. Pennati M, Folini M, Zaffaroni N. (2007). Targeting survivin in cancer therapy: fulfilled promises and open questions. Carcinogenesis 28: 1133–1139. Raj D, Liu T, Samadashwily G, Li F, Grossman D. (2008). Survivin repression by p53, Rb and E2F2 in normal human melanocytes. Carcinogenesis 29: 194–201. Reddel RR. (2000). The role of senescence and immortalization in carcinogenesis. Carcinogenesis 21: 477–484. Rodel F, Hoffmann J, Distel L, Herrmann M, Noisternig T, Papadopoulos T et al. (2005). Survivin as a radioresistance factor, and prognostic and therapeutic target for radiotherapy in rectal cancer. Cancer Res 65: 4881–4887. Schuller CE, Jankowski K, Mackenzie KL. (2007). Telomere length of cord blood-derived CD34(+) progenitors predicts erythroid proliferative potential. Leukemia 21: 983–991. Stewart SA, Weinberg RA. (2006). Telomeres: cancer to human aging. Annu Rev Cell Dev Biol 22: 531–557. Taylor LM, James A, Schuller CE, Brce J, Lock RB, Mackenzie KL. (2004). Inactivation of p16INK4a, with retention of pRB and p53/ p21cip1 function, in human MRC5 fibroblasts that overcome a telomere-independent crisis during immortalization. J Biol Chem 279: 43634–43645. Uren AG, Wong L, Pakusch M, Fowler KJ, Burrows FJ, Vaux DL et al. (2000). Survivin and the inner centromere protein INCENP show similar cell-cycle localization and gene knockout phenotype. Curr Biol 10: 1319–1328. Verhagen AM, Coulson EJ, Vaux DL. (2001). Inhibitor of apoptosis proteins and their relatives: IAPs and other BIRPs. Genome Biol 2: REVIEWS3009. Wang J, Yang L, Yang J, Kuropatwinski K, Wang W, Liu XQ et al. (2008). Transforming growth factor beta induces apoptosis through repressing the phosphoinositide 3-kinase/AKT/survivin pathway in colon cancer cells. Cancer Res 68: 3152–3160. Wang Z, Sampath J, Fukuda S, Pelus LM. (2005). Disruption of the inhibitor of apoptosis protein survivin sensitizes Bcr-abl-positive cells to STI571-induced apoptosis. Cancer Res 65: 8224–8232. Weinstein IB, Joe A. (2008). Oncogene addiction. Cancer Res 68: 3077–3080; discussion 3080. Wen VW, Wu K, Baksh S, Hinshelwood RA, Lock RB, Clark SJ et al. (2006). Telomere-driven karyotypic complexity concurs with p16INK4a inactivation in TP53-competent immortal endothelial cells. Cancer Res 66: 10691–10700. Wright WE, Piatyszek MA, Rainey WE, Byrd W, Shay JW. (1996). Telomerase activity in human germline and embryonic tissues and cells. Dev Genet 18: 173–179. Yang D, Welm A, Bishop JM. (2004). Cell division and cell survival in the absence of survivin. Proc Natl Acad Sci USA 101: 15100–15105. Yang J, Song K, Krebs TL, Jackson MW, Danielpour D. (2008). Rb/ E2F4 and Smad2/3 link survivin to TGF-beta-induced apoptosis and tumor progression. Oncogene 27: 5326–5338. Zhang M, Yang J, Li F. (2006). Transcriptional and post-transcriptional controls of survivin in cancer cells: novel approaches for cancer treatment. J Exp Clin Cancer Res 25: 391–402.

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

Oncogene