Nuclear import of hTERT requires a bipartite nuclearlocalization signal and Akt-mediated phosphorylation

Jeeyun Chung1,*,`, Prabhat Khadka1,2,* and In Kwon Chung1,2,§

1Department of Systems Biology, Yonsei University, Seoul, 120-749, Korea2Department of Integrated Omics for Biomedical Science, WCU Program of Graduate School, Yonsei University, Seoul, 120-749, Korea

*These authors contributed equally to this work`Present address: Program in the Biological and Biomedical Sciences, Yale University, New Haven, CT 06520, USA§Author for correspondence (topoviro@yonsei.ac.kr)

Accepted 24 January 2012Journal of Cell Science 125, 2684–2697� 2012. Published by The Company of Biologists Ltddoi: 10.1242/jcs.099267

SummarySustained cell proliferation requires telomerase to maintain functional telomeres that are essential for chromosome integrity andprotection. Although nuclear import of telomerase transcriptase (hTERT) is required for telomerase activity to elongate telomeres in

vivo, the molecular mechanism regulating nuclear localization of hTERT is unclear. We have identified a bipartite nuclear localizationsignal (NLS; amino acid residues 222–240) that is responsible for nuclear import of hTERT. Immunofluorescence imaging of hTERTrevealed that mutations in any of the bipartite NLS sequences result in decreased nuclear fluorescence intensity compared with wild-typehTERT. We also show that Akt-mediated phosphorylation at serine 227 is necessary for directing nuclear translocation of hTERT.

Interestingly, serine 227 is located between two clusters of basic amino acids in the bipartite NLS. Inactivation of Akt activity by adominant-negative mutant or wortmannin treatment attenuated nuclear localization of hTERT. We further show that both bipartite NLSand serine 227 in hTERT are required for cell immortalization of normal human foreskin fibroblast cells. Taken together, our findings

reveal a previously unknown regulatory mechanism for nuclear import of hTERT through a bipartite NLS mediated by Aktphosphorylation, which represents an alternative pathway for modulating telomerase activity in cancer.

Key words: Telomerase, hTERT, Nuclear localization signal, Akt, Phosphorylation

IntroductionTelomeres, the specialized nucleoprotein complexes at the ends

of linear eukaryotic chromosomes, are essential for chromosome

integrity and protection (Blackburn, 2001). In most organisms,

telomere DNA consists of long tracts of duplex telomere repeats

(TTAGGG in vertebrates) with 39 single-stranded G overhangs

and is tightly associated with the six-protein complex, shelterin,

which protects chromosome termini from being recognized as

sites for DNA damage (Palm and de Lange, 2008; Liu et al.,

2004). In the absence of a telomere maintenance pathway, most

eukaryotic cells progressively lose telomeric DNA with each cell

division because of the end replication problem (Lingner et al.,

1995). If telomere shortening is not balanced by elongation, it

leads to cell death, cell senescence or abnormal cell proliferation.

Therefore, eukaryotic cells with extended replicative lifespan

should have a mechanism to counteract the loss of telomeric

DNA (Stewart and Weinberg, 2006). The maintenance of

telomeric repeats in most eukaryotic organisms requires

telomerase, which consists of a telomerase reverse transcriptase

(TERT) and a telomerase RNA component (TERC) (Autexier

and Lue, 2006). Telomerase activity is expressed in a majority of

human tumors but repressed in most normal somatic cells,

suggesting that the activation of telomerase is a crucial step in

human oncogenesis (Kim et al., 1994).

Telomerase activity correlates with telomerase transcriptase

(hTERT) expression, implicating the catalytic subunit as the rate-

limiting component of the telomerase holoenzyme. Whereas the

enzymatic activity of telomerase is controlled by the

transcriptional regulation of hTERT, substantial evidence

suggests that posttranslational modifications of hTERT play a

role in modulating telomerase activity (Autexier and Lue, 2006;

Cong et al., 2002). Several kinases have been reported to act on

hTERT. Upregulation of telomerase activity is associated with

phosphorylation of hTERT by protein kinase C (PKC) (Li et al.,

1998). Akt kinase is able to phosphorylate hTERT and activate

telomerase activity (Kang et al., 1999). Two putative Akt

phosphorylation sites within hTERT (serine residues at 227 and

824) have been identified. As Akt is a key effector of the

phosphatidylinositide 3-kinase (PI3K) signaling pathway (Vivanco

and Sawyers, 2002), these findings might provide new insights into

the molecular mechanisms by which the PI3K pathway promotes

cell proliferation and survival. By contrast, telomerase activity is

inhibited by c-Abl-kinase-mediated phosphorylation of hTERT

(Kharbanda et al., 2000). In the case of ubiquitylation, MKRN1 is

the first factor identified as an E3 ubiquitin ligase for hTERT in

mammalian cells (Kim et al., 2005). Overexpression of MKRN1

promotes ubiquitylation of hTERT and subsequently causes a

decrease in telomerase activity as well as in telomere length.

Recently, two other factors, CHIP and Hdm2 have been shown to

function as ubiquitin ligases for hTERT and negatively regulate

telomerase activity (Lee et al., 2010; Oh et al., 2010).

As the nuclear localization of hTERT is required for

telomerase activity to elongate telomeres in vivo, newly

synthesized hTERT should be maintained in a conformation

2684 Research Article

Journ

alof

Cell

Scie

nce

that is competent for nuclear import. Because the size of hTERT

precludes a passive nuclear transport mechanism, it could

undergo active nucleocytoplasmic shuttling to set cellular

telomerase activity. Several proteins that regulate the nuclear

localization of hTERT have been identified. Although the

molecular chaperone complex Hsp90–p23 is required for

proper folding of nascent hTERT as well as efficient

telomerase assembly (Holt et al., 1999; Forsythe et al., 2001),

Akt-mediated phosphorylation is essential for nuclear

translocation of hTERT (Kimura et al., 2004; Kawagoe et al.,

2003). Both Hsp90 and Akt physically interact with hTERT, and

their associations are necessary for nuclear translocation of

hTERT and telomerase activity (Kawauchi et al., 2005).

Subcellular localization of hTERT is also regulated through its

interaction with NF-kB (Akiyama et al., 2003). Tumor necrosis

factor a (TNFa) modulates telomerase activity by inducing the

nuclear translocation of hTERT bound to NF-kB p65 (Akiyama

et al., 2004). The 14-3-3 signaling protein interacts with hTERT

and enhances nuclear localization of hTERT by inhibiting CRM

binding to the nuclear export signal (NES)-like motif (Seimiya

et al., 2000). Although the nuclear localization of hTERT is an

important step in regulating telomerase activity, the molecular

mechanism regulating it remains unclear.

In this study, we have identified a bipartite nuclear localization

signal (NLS) that is responsible for nuclear import of hTERT. We

also describe that Akt-mediated phosphorylation at serine 227 is

required to direct nuclear translocation of hTERT. Furthermore,

we show that mutations in the bipartite NLS region interfere with

the cellular immortalization of normal human foreskin fibroblast

(HFF) cells. Overall, our data demonstrate that both bipartite

NLS and Akt-mediated phosphorylation at serine 227 are

previously unknown pathways for ensuring proper cellular

telomerase activity by facilitating nuclear import of hTERT.

ResultsIdentification of the functional domain essential for nuclearlocalization of hTERT

To identify the functional domain responsible for nuclear

localization of hTERT, we generated a series of deletion

mutants of hTERT (Fig. 1A) and determined their subcellular

localization. H1299 cells were transfected with FLAG-tagged

hTERT or various deletion mutants and subjected to

immunofluorescence staining with anti-FLAG antibody.

Representative fluorescence images of the hTERT derivatives

are shown in Fig. 1B. The deletion mutants lacking the N-

terminal 100 or 200 amino acids (101–1132 or 201–1132)

showed a pronounced nuclear localization of hTERT similar to

the finding with wild-type hTERT, whereas deletions of the

N-terminal 300 amino acids (301–1132, 401–1132, 543–1132)

resulted in a cytoplasmic distribution. Furthermore, the C-

terminal deletion mutants containing residues 201–300 (1–542,

1–400 and 1–300) were exclusively localized to the nucleus. The

Fig. 1. Subcellular localization of hTERT mutants. (A) Schematic

representation of hTERT constructs used in these experiments. RT, reverse

transcriptase motif; NES, nuclear export signal. (B) The wild-type and

various mutant constructs of hTERT were expressed as FLAG-tagged

proteins in H1299 cells. Cells were subjected to immunofluorescence

staining with anti-FLAG antibody (red), followed by fluorescence

microscopy. The nuclei were stained with DAPI (blue). (C) Lysates

obtained from H1299 cells transfected with the various mutant constructs

were analyzed by immunoblotting with anti-FLAG antibody.

Nuclear translocation of hTERT 2685

Journ

alof

Cell

Scie

nce

C-terminal deletion mutant lacking this region (1–200) was

evenly distributed between nucleus and cytoplasm possibly due

to a passive diffusion. These results demonstrate that the amino

acid residues from 201 to 300 are essential for nuclear

localization of hTERT and could contain a NLS. Western blot

analysis revealed that the hTERT mutant proteins were correctly

expressed (Fig. 1C).

We next examined whether the requirement of the amino acid

residues 201–300 for nuclear localization of hTERT could be

confirmed to be quantitatively significant. For quantification

experiments, more than 100 cells per transfection were scored

according to whether hTERT was higher in the nucleus (N),

evenly distributed between nucleus and cytoplasm (N+C) or

higher in the cytoplasm (C). Whereas wild-type hTERT and the

201–1132 mutant showed predominant localization in the nucleus

(74.1% and 77.2% nuclear, respectively), the majority of the

301–1132 mutant signal was detected in the cytoplasm (66.7%

cytoplasmic, 20.4% pan-cellular; Fig. 2A,B). However, the 1–

300 mutant exclusively localized to the nucleus (100% nuclear).

A similar behavior of hTERT was observed in HT1080 and

MCF7 cells (supplementary material Fig. S1).

Nuclear localization of hTERT has been shown to be regulated

by an NES, which is recognized by a specific export receptor.

The 14-3-3 signaling protein was reported to enhance nuclear

localization of hTERT by inhibiting the CRM1 (also known as

exportin-1) binding to the hTERT NES motif (Seimiya et al.,

2000). To investigate the effect of the NES on nuclear

localization of hTERT, H1299 cells transfected with various

deletion mutants were treated with leptomycin B (LMB), an

inhibitor of the nuclear export by CRM1. When the subcellular

localization was scored quantitatively, the nuclear fluorescence

signal of wild-type hTERT and the 201–1132 mutant was slightly

increased by the presence of LMB (Fig. 2B). However, most of

the 301–1132 mutant signal was found in the cytoplasm even in

the presence of LMB. Because the 301–1132 mutant contains a

NES at the C-terminus, these results suggest that the observed

cytoplasmic signals could be due to failure of nuclear import but

not due to enhancement of nuclear export. The 1–300 mutant,

Fig. 2. Identification of the functional

domain essential for nuclear

localization of hTERT. (A) H1299

cells transfected with various mutant

constructs of hTERT were treated with

20 nM leptomycin B (LMB), or left

untreated, for 6 hours and subjected to

immunofluorescence staining with anti-

FLAG antibody (red). The nuclei were

stained with DAPI (blue). (B) After

transfection, H1299 cells were treated

with 20 nM LMB, or left untreated, for

6 hours and then analyzed quantitatively

for the cellular location of the hTERT

constructs. In each experiment, more

than 100 hTERT-expressing cells were

evaluated quantitatively for mainly

nuclear (N, blue bars), nuclear and

cytoplasmic (N+C, red bars) and mainly

cytoplasmic (C, green bars)

fluorescence. (C) Schematic

representation of internal deletion

constructs of hTERT used in these

experiments. (D) Lysates obtained from

H1299 cells transfected with deletion

constructs were analyzed by

immunoblotting with anti-FLAG

antibody. (E) H1299 cells transfected

with various deletion constructs were

treated with 20 nM LMB, or left

untreated, for 6 hours and subjected to

immunofluorescence staining with anti-

FLAG antibody (red). (F) After

transfection, H1299 cells were treated

with 20 nM LMB, or left untreated, for

6 hours and then analyzed quantitatively

for the cellular location of the

hTERT constructs.

Journal of Cell Science 125 (11)2686

Journ

alof

Cell

Scie

nce

which does not contain a NES, was exclusively detected in the

nucleus irrespective of LMB treatment (Fig. 2A,B).

To further confirm the requirement of the residues 201–300, we

generated mutant constructs lacking the internal amino acid

residues 101–200 and 201–300 (Fig. 2C). We also generated a

mutant construct deleted in the basic amino acid cluster (amino

acids 193–199), which is predicted as a putative NLS by the

Nuclear Protein Database (http://npd.hgu.mrc.ac.uk). The proper

expression of each deletion mutant was examined by immunoblot

analysis using anti-FLAG antibody (Fig. 2D). Nuclear localization

of hTERT was not prevented by deletion of residues 101–200

(Fig. 2E,F). However, deletion of residues 201–300 substantially

impaired nuclear localization of hTERT even in the presence of

LMB. A mutant lacking the basic amino acid cluster (Del193–199)

was predominantly localized in the nucleus. Therefore, amino acid

residues 193–199 are not required for nuclear import of hTERT.

Taken together, these data further support that residues 201–300

are essential for nuclear import of hTERT.

Identification of a bipartite nuclear localization signal

in hTERT

A motif scan of the amino acid sequence revealed that hTERT

contains a consensus bipartite NLS where two clusters of basic

amino acids, residues 222–224 (RRR) and residues 236–240

(KRPRR) are aligned in tandem (Fig. 3A). Interestingly, this

putative NLS appears to be conserved among different species

including human, mouse, rat and Xenopus. In hTERT, the amino

acid spacer between the two putative NLS motifs is composed of

11 residues, which is consistent with a typical bipartite NLS

spacer (Lange et al., 2007; Suntharalingam and Wente, 2003). To

demonstrate that this region functions as a bipartite NLS, we

generated several mutant forms of hTERT, in which the basic

amino acids in either the first cluster, second cluster or both

clusters were mutated to alanines (Fig. 3B). Whereas wild-type

hTERT predominantly localized to the nucleus (72.1% nuclear),

the mutations at the first cluster (3A) or second cluster (4A)

resulted in reduced nuclear localization of hTERT (11.0% and

3.7% nuclear, respectively; Fig. 3C,D). When both clusters were

mutated (7A), the nuclear signals were further decreased (2.5%

nuclear). These results indicate that both of the basic amino acid

clusters (between residues 222 and 240) contribute to nuclear

import of hTERT and function as a bipartite NLS.

It is well documented that the nucleotide-bound state of the

small GTPase Ran plays an important role in nuclear

translocation through nuclear pore complexes (NPCs) (Kalab

et al., 2002). A dominant-negative mutant of Ran (RanQ69L)

Fig. 3. Identification of a bipartite nuclear localization

signal in hTERT. (A) Sequence alignments showing

conservation of the bipartite NLS sequences in the different

species. Numbering of the amino acids is based on the

sequence of human TERT. (B) Several mutant forms of

hTERT in which the basic amino acids in the first cluster

(3A), second cluster (4A) or both clusters (7A) are mutated

to alanines, respectively. (C) H1299 cells were transfected

with various mutant constructs and subjected to

immunofluorescence staining with anti-FLAG antibody

(red). The nuclei were stained with DAPI (blue). (D) After

transfection, H1299 cells were analyzed quantitatively for

the cellular location of the hTERT constructs. (E) H1299 and

HT1080 cells were transfected with Myc-tagged RanQ69L

and analyzed by immunofluorescence staining with anti-

hTERT (green) or anti-Myc (red) antibodies. (F) Lysates

obtained from H1299, HT1080, U2OS and Saos-2 cells were

analyzed by immunoblotting with anti-hTERT antibody.

Nuclear translocation of hTERT 2687

Journ

alof

Cell

Scie

nce

deficient in GTPase activity has been shown to inhibit NLS-mediated nuclear protein import (Dickmanns et al., 1996;

Palacios et al., 1996) but promote nuclear export of substratescarrying an NES (Richards et al., 1997; Kehlenbach et al., 1998).Thus, we investigated the effect of RanQ69L on nuclearlocalization of hTERT in H1299 and HT1080 cells. In the

absence of RanQ69L expression, endogenous hTERTaccumulated in the nucleus (Fig. 3E). By contrast, thefluorescence signals of both hTERT and RanQ69L were

detected around the perimeter of the nucleus in cells expressingRanQ69L (Fig. 3E), suggesting the involvement of Ran in thenuclear import of hTERT. An antibody used in these experiments

specifically detected endogenous hTERT in H1299 and HT1080cells but not in telomerase-negative U2OS and Saos-2 cells(Fig. 3F). In parallel, as a control we verified that the Ran–GTP-dependent nuclear import of hnRNP A1 was inhibited by

RanQ69L expression (Siomi et al., 1997) (supplementarymaterial Fig. S2A). Because Ran–GTP hydrolysis does notappear to be necessary for nuclear protein export (Kehlenbach

et al., 1999), the pre-existing nuclear hTERT would be exportedto the cytoplasm by RanQ69L expression. We also observed thatectopically expressed hTERT was excluded from the nucleus

when RanQ69L was co-expressed (supplementary material Fig.S2B). Taken together, these results suggest that the identifiedbipartite NLS sequence functions as a typical NLS that is

dependent on the Ran–importin system.

Phosphorylation of serine 227 is required for efficientnuclear translocation of hTERT

Akt kinase has been reported to enhance telomerase activitythrough phosphorylation and nuclear translocation of hTERT(Kimura et al., 2004; Kawagoe et al., 2003). hTERT contains two

putative Akt phosphorylation sites, serine residues 227 and 824(Kang et al., 1999). However, it is not clear whichphosphorylation site is involved in nuclear localization of

hTERT. Because serine 227 is located between two clusters ofbasic amino acids in the bipartite NLS sequence, we predictedthat phosphorylation at serine 227 might be involved in nuclearlocalization of hTERT. To test this possibility, we performed

site-directed mutagenesis of hTERT, mutating the serine residuesat 227 and 824 to alanine (S227A, S824A, S227,824A) (Fig. 4A).We also generated a phosphorylation-mimicking mutant in which

the serine residue at 227 was replaced by glutamate (S227E).H1299 cells were transfected with FLAG–hTERT or variousmutant constructs and subjected to immunofluorescence staining.

Representative fluorescence images of hTERT are shown inFig. 4B. Whereas wild-type hTERT and the S824A mutantpredominantly localized to the nucleus (74.1% and 76.4%nuclear, respectively), the nuclear fluorescence signals of

S227A and S227,824A mutants were considerably decreased(31.2% and 32.5% nuclear, respectively; Fig. 4C). Interestingly,phosphorylation-mimicking mutant S227E showed an increased

ability to translocate into the nucleus (83% nuclear). Theseresults suggest that phosphorylation of serine 227, but not serine824, is important for nuclear translocation of hTERT. The

identical results were observed in telomerase-negative Saos-2cells (supplementary material Fig. S3), demonstrating that theeffects of point mutations were not related to the cellular

background in which the mutant constructs were expressed. Wealso note that the fluorescence signals of wild-type and mutanthTERT were slightly increased in the nucleus when the

subcellular localization was scored quantitatively in the

presence of LMB (Fig. 4C).

To examine whether Akt kinase directly phosphorylateshTERT, H1299 cells transfected with FLAG–hTERT or variousmutants were subjected to immunoprecipitation with anti-FLAG

antibody, followed by immunoblotting with anti-phosphorylated-Akt substrate antibody. As shown in Fig. 4D, S227A and S824Aproteins were phosphorylated by Akt kinase as efficiently as wild-

type hTERT. Although the hTERT proteins were similarlyexpressed, no phosphorylation was observed in the S227,824Amutant. These results are consistent with the previous report that

both serine residues are phosphorylated by Akt kinase (Kang et al.,1999). To further confirm the phosphorylation of hTERT by Aktkinase, we used wortmannin to inhibit Akt kinase activity. It hasbeen well documented that wortmannin inactivates Akt activity

through inhibition of PI3K (Vivanco and Sawyers, 2002). H1299cells were treated with wortmannin and subjected toimmunoprecipitation. The amounts of phosphorylated hTERT

proteins were substantially reduced by the treatment withwortmannin (Fig. 4D).

Because phosphorylation of serine 227, but not serine 824, isimportant for nuclear translocation of hTERT, we wished to

determine the effect of point mutations of hTERT on telomeraseactivity. In these experiments, we chose telomerase-negativeSaos-2 cells to exclude the effect of endogenous telomerase

activity. Saos-2 cells were transfected with FLAG–hTERT orvarious mutants, and telomerase activity was measured by thetelomeric repeat amplification protocol (TRAP) assay. We

observed no difference in the signal intensity between wild-type and point mutant hTERT (Fig. 4E). Because the mutantproteins were almost equally expressed in Saos-2 cells (Fig. 4F),

these results indicate that alanine substitution for serine residues227 and 824 has no effect on telomerase activity, at least asdetermined by the conventional TRAP assay.

Inhibition of Akt activity attenuates nuclear translocationof endogenous hTERT

To further verify the requirement of Akt kinase activity, wedetermined the effect of dominant-negative Akt (DN-Akt) on

subcellular localization of hTERT in H1299 and HT1080 cells.Endogenous hTERT was localized mostly to the nucleus in bothcell lines, but nuclear accumulation of endogenous hTERT was

considerably reduced by transfection of DN-Akt (Fig. 5A,B).When H1299 cells were co-transfected with FLAG–hTERT andDN-Akt, nuclear localization of ectopically expressed hTERT was

also inhibited (supplementary material Fig. S4A,B). However, thephosphorylation-mimicking mutant S227E was mostly localized tothe nucleus even when DN-Akt was overexpressed. We note thatthe intensity of hTERT fluorescence seems to be reduced in DN-

Akt-transfected cells. These findings are consistent with theprevious reports that cytoplasmic hTERT is readily degradedthrough ubiquitylation by E3 ligases, such as MKRN1 and

CHIP, which reside in the cytoplasm (Kim et al., 2005; Leeet al., 2010). We next examined whether wortmannin attenuatesnuclear localization of hTERT. Wortmannin treatment caused

endogenous hTERT to accumulate in the cytoplasm (Fig. 5C,D).We also found that ectopically expressed hTERT was notefficiently translocated into the nucleus in the presence of

wortmannin (supplementary material Fig. S4C,D). These resultsindicate that nuclear localization of endogenous and ectopicallyexpressed hTERT requires Akt activity.

Journal of Cell Science 125 (11)2688

Journ

alof

Cell

Scie

nce

Both the bipartite NLS motifs and Akt-mediated

phosphorylation are necessary for efficient nuclear

localization of hTERT

We next examined to what extent the two factors, bipartite NLS

and Akt-mediated phosphorylation, contribute to nuclear

translocation of hTERT. To address this issue, we constructed

three mutants, each including S227A, in which the first cluster

(3A), second cluster (4A) or both clusters (7A) of basic amino

acids were mutated to alanines (Fig. 6A). The S227A mutant had

a pronounced cytoplasmic localization (49% cytoplasmic, 15.6%

pan-cellular). Mutations of the basic amino acid clusters (S227A/

3A, S227A/4A and S227A/7A) almost completely abolished

nuclear localization of hTERT (85.8%, 82.7% and 85.6%

cytoplasmic, respectively; Fig. 6B,C). When the basic amino

acids around S227E were substituted with alanines (Fig. 6D),

these hTERT mutants (S227E/3A, S227E/4A and S227E/7A)

were predominantly localized to the cytoplasm, whereas a

majority of the S227E form was found in the nucleus

(Fig. 6E,F). These results further confirm that both the two

intact NLS motifs and phosphorylation of serine 227 are

necessary for efficient nuclear import of hTERT.

hTERT has been shown to interact with Akt and Hsp90, and

their associations are required for nuclear translocation of hTERT

and telomerase activation (Kawauchi et al., 2005). Thus, we

determined whether mutations in S227A/7A affect hTERT

association with these proteins. As shown in Fig. 6G, S227A

and S227A/7A interact with Akt and Hsp90 as efficiently as the

wild-type hTERT, indicating that the inability of the mutant

proteins to translocate into the nucleus is not due to impaired

interactions with Akt and Hsp90.

Fig. 4. Phosphorylation of serine

227 is required for nuclear

localization of hTERT. (A) The

serine residues at 227 and 824 were

replaced with alanines, singly and

together. To generate a

phosphorylation-mimicking mutant,

serine 227 was replaced with

glutamate. (B) H1299 cells

transfected with various point mutant

constructs were treated with 20 nM

LMB, or left untreated, for 6 hours

and subjected to immunofluorescence

staining with anti-FLAG antibody

(red). The nuclei were stained with

DAPI (blue). (C) After transfection,

H1299 cells were treated with 20 nM

LMB, or left untreated, for 6 hours

and then analyzed quantitatively for

the cellular location of the hTERT

constructs. (D) H1299 cells

transfected with various point mutant

constructs were treated with 1 mM

wortmannin, or left untreated, for

1 hour and subjected to

immunoprecipitation with anti-FLAG

antibody, followed by

immunoblotting with anti-

phosphorylated-Akt substrate

antibody. (E) Lysates obtained from

Saos-2 cells transfected with various

point mutant constructs were

analyzed for telomerase activity by

the TRAP assay. ITAS, internal

telomerase assay standard.

(F) Lysates obtained from Saos-2

cells transfected with various point

mutant constructs were analyzed by

immunoblotting with anti-FLAG-

antibody.

Nuclear translocation of hTERT 2689

Journ

alof

Cell

Scie

nce

Mutations of the bipartite NLS and serine 227 interfere withthe ability of hTERT to immortalize human foreskinfibroblast cells

Most normal human somatic cells have undetectable levels of

telomerase activity. Exogenous expression of hTERT cDNA innormal human fibroblasts was reported to permit senescencebypass, enabling cells to obtain an extended lifespan (Bodnaret al., 1998). Because both the bipartite NLS and Akt

phosphorylation of serine 227 are required for nucleartranslocation of hTERT, we investigated the ability of S227A/7A to immortalize a primary HFF cell strain. HFF cells were

infected at ,40 population doublings (PDs) with the lentivirusparticles expressing the empty vector, hTERT or S227A/7A, andstable cell lines were established as mass cultures from separate

transductions to monitor PD levels at regular intervals. Whereasparental and empty-vector-expressing cells stopped dividing ,50days post-transduction, hTERT-expressing cells continued todivide throughout the duration of the experiments (Fig. 7A). The

wild-type and mutant-hTERT-expressing cells divided at thesame rate for ,70 days, and then the growth rate of mutant cellsstarted to slow down compared with the wild-type cells. Similar

results were obtained in independent infections using each typeof lentivirus (supplementary material Fig. S5A). The hTERTproteins were almost equally expressed in HFF cell lines

throughout the duration of the experiments (Fig. 7B). Inaddition, no difference in stability was observed betweenwild-type and mutant hTERT proteins (supplementary material

Fig. S5B).

To examine whether the expression of exogenous hTERTreconstitutes telomerase activity, cytoplasmic and nuclear

extracts were separately collected from cells stably expressing

the empty vector, hTERT or S227A/7A and subjected to a TRAPassay. Telomerase activity was exclusively detected in the

nuclear extracts but not in the cytoplasmic extracts (Fig. 7C).

Whereas the empty-vector-expressing cells did not exhibittelomerase activity, both the wild-type and mutant-hTERT-

expressing cells reconstituted telomerase activity (Fig. 7C,D),indicating that mutations in S227A/7A did not affect telomerase

activity in HFF cells. To determine whether telomerase activitychanges between early passages after infection and later passages

as cells near senescence, we performed the TRAP assay on HFFcells expressing wild-type or mutant hTERT at various PDs

(supplementary material Fig. S6). Although TRAP activity wasslightly decreased at later PDs, it was still detectable in both cell

lines. We also found that the majority of the S227A/7A signal

was detected in the cytoplasm at both early and late PDs, whereaswild-type hTERT showed predominant localization in the

nucleus (Fig. 7E). Digital images of hTERT fluorescence werequantitatively analyzed by confocal microscopy, and wild-type

hTERT was found to be predominantly in the nucleus (81.7%nuclear at PD 5 and 80.1% nuclear at PD 78). By contrast, the

majority of S227A/7A signal was detected in the cytoplasm(Fig. 7F). We note that nuclear localization of S227A/7A was

substantially reduced at later passages (10.2% nuclear at PD 70)

compared with early passages (19.1% nuclear at PD 5).

We next examined whether the growth arrest of S227A/7A-expressing cells at later PDs was accompanied by cellular

senescence. Approximately 91% of empty-vector-expressing

cells were stained intensely for senescence-associated b-galactosidase (SA-b-Gal) activity and had a flattened and

Fig. 5. Inhibition of Akt activity attenuates nuclear

translocation of endogenous hTERT. (A) H1299 and

HT1080 cells were transfected with HA-tagged

dominant negative Akt (DN-Akt) and analyzed by

immunofluorescence staining with anti-hTERT (red) or

anti-HA (green) antibodies. The nuclei were stained

with DAPI (blue). (B) After transfection, cells were

analyzed quantitatively for the cellular location of

endogenous hTERT. (C) H1299 and HT1080 cells were

treated with 1 mM wortmannin, or left untreated, for

1 hour and then subjected to immunofluorescence

staining with anti-hTERT antibody (green). (D) H1299

and HT1080 cells were treated with 1 mM wortmannin

for 1 hour and analyzed quantitatively for the cellular

location of endogenous hTERT.

Journal of Cell Science 125 (11)2690

Journ

alof

Cell

Scie

nce

enlarged morphology at PD 12 after antibiotic selection

(Fig. 7G,H). However, no such changes were found in cells

expressing wild-type hTERT throughout the duration of the

experiments. Approximately 40% of S227A/7A-expressing cells

were SA-b-Gal positive and exhibited a senescent phenotype at

PD 70. These results suggest that both the bipartite NLS and

serine 227 are necessary for optimal cellular immortalization.

To determine whether mutations in S227A/7A interfere with

the ability of telomerase to maintain telomeres in vivo, we

performed a terminal restriction fragment (TRF) analysis on HFF

cells expressing wild-type or mutant hTERT at various PDs. The

average telomere length in empty-vector-expressing cells became

shorter as the populations reached senescence (Fig. 7I, lane 2).

However, the average telomere length of hTERT-expressing cells

was stably maintained up to PD 78 (lanes 3–5). Expression of

S227A/7A maintained stable telomere length for the first ,47

PDs but there was a clear reduction after PD 70 when the

populations encountered senescence (lanes 6–8). These results

indicate that mutations in S227A/7A render telomerase unable to

maintain telomere length in later passage cells.

Fig. 6. Both the bipartite NLS and Akt-mediated phosphorylation are required for nuclear localization of hTERT. (A) Several mutant forms of the region

surrounding S227A were produced, in which the basic amino acids in the first cluster, second cluster, or both clusters were mutated to alanines. (B) H1299 cells

were transfected with the various mutant constructs and subjected to immunofluorescence staining with anti-FLAG (red) antibody. The nuclei were stained with

DAPI (blue). (C) After transfection, H1299 cells were analyzed quantitatively for the cellular location of the hTERT mutants. (D) Several mutant forms of the

region surrounding S227E were also examined. (E,F) As for B and C, but using the S227E mutants. (G) H1299 cells were transfected with wild-type hTERT or

various point mutant constructs and subjected to immunoprecipitation with anti-FLAG, anti-Akt or anti-HSP90 antibodies, followed by immunoblotting

as indicated.

Nuclear translocation of hTERT 2691

Journ

alof

Cell

Scie

nce

hTERT expression reduces basal intracellular ROS levels

The above results show that, although S227A/7A was mostly

found in the cytoplasm, the S227A/7A-expressing cells did not

senesce at the same rate as parental or empty-vector-transfected

cells (Fig. 7A). This observation suggests that a non-telomeric

function of cytoplasmic S227A/7A contributes to the delayed

senescent phenotype. There is accumulating evidence that

hTERT is localized to mitochondria because of the presence of

a mitochondrial targeting signal (Santos et al., 2004; Santos et al.,

2006), and that hTERT exerts a protective function against

Fig. 7. Both the bipartite NLS and serine 227 are required for cellular immortalization. (A) Cell growth curves of stable HFF cell lines expressing the empty

vector, hTERT or S227A/7A. HFF cells were infected at ,40 PDs with the lentivirus particles to establish stable expression cell lines. Stable cells were replated

every 3–4 days to maintain log-phase growth and calculate the growth rate, with the first day after blasticidin selection regarded as day 0. (B) Lysates obtained

from HFF cells stably expressing the empty vector, hTERT or S227A/7A were immunoblotted with anti-FLAG antibody. (C) Cytoplasmic and nuclear extracts

were separately collected from HFF cells stably expressing the empty vector, hTERT or S227A/7A at ,40 PDs after selection, and analyzed for telomerase

activity by the TRAP assay with twofold dilutions. To test RNA-dependent extension, RNase A (0.25 mg/ml) was added to the extracts before the primer

extension reaction. ITAS, internal telomerase assay standard. (D) Telomerase products in the nuclear extracts were determined semiquantitatively; values are

means and standard deviations from three experiments. (E) HFF cells stably expressing hTERT or S227A/7A at the indicated PDs were subjected to

immunofluorescence staining with anti-FLAG (red) antibody. Digital images of hTERT fluorescence were captured using a confocal microscope. (F) The

fluorescence intensities of cytoplasmic and nuclear hTERT were calculated using MetaMorph software and averaged over the ten images as in E. (G) HFF cells

stably expressing the empty vector, hTERT or S227A/7A at the indicated PDs were photographed after staining for SA-b-Gal activity. (H) Percentages of cells

that were positive for SA-b-Gal activity in three HFF cell lines. Statistical analyses were performed using a two-tailed Student’s t-test (*P,0.001). (I) Genomic

blot of telomere restriction fragments in stable HFF clones expressing the empty vector, hTERT or S227A/7A. Genomic DNA was isolated at the indicated PDs,

digested with RsaI and HinfI, and analyzed by Southern blotting using a telomere repeat probe.

Journal of Cell Science 125 (11)2692

Journ

alof

Cell

Scie

nce

oxidative stress (Indran et al., 2011; Ahmed et al., 2008;

Haendeler et al., 2009). To examine the impact of hTERT

expression on reactive oxygen species (ROS) production, we first

determined whether S227A/7A localizes at the mitochondria.

Strong S227A/7A signals were found associated with

mitochondria, as indicated by immunofluorescence staining

with anti-Tom20 antibody (Fig. 8A). S227A/7A was also found

in cytoplasmic foci that did not correspond with mitochondria

although the nature of these localization sites was not

determined. Next, HFF cells stably expressing hTERT and

S227A/7A at ,70 PDs were analyzed for basal intracellular ROS

levels by examining intracellular peroxide and mitochondrial

superoxide levels using 29,79-dichlorodihydrofluorescein

diacetate (DCFH-DA) and MitoSOX RED, respectively.

Intracellular ROS levels were lower in cells expressing hTERT

and S227A/7A compared with cells expressing the control vector

(Fig. 8B,C). However, the levels were similar in wild-type and

mutant-hTERT-expressing cells. These findings were further

supported by fluorescence detection using confocal microscopy.

Both the DCF and MitoSOX RED signals were lower in cells

expressing hTERT and S227A/7A (Fig. 8D,E). Identical results

were obtained in cells expressing hTERT and S227A/7A at

earlier passages (,40 PDs; supplementary material Fig. S7).

Given that oxidative stress increases the basal rate of telomere

shortening by induction of telomeric DNA damage (Petersen et

al., 1998), cytoplasmic S227A/7A might contribute, at least in

part, to telomere length maintenance by reducing the basal level

of oxidative stress (see below for further discussion).

DiscussionTo ensure proper cellular function of telomerase, hTERT should

be correctly assembled with the associated proteins and

Fig. 8. hTERT expression alleviates basal intracellular ROS levels. (A) HFF cells stably expressing hTERT or S227A/7A at ,40 PDs after selection were

analyzed by indirect immunofluorescence with anti-FLAG (green) and anti-Tom20 (red) antibodies. DNA was stained with DAPI (blue). (B,C) For detection of

the intracellular ROS levels, HFF cells stably expressing the empty vector, hTERT or S227A/7A at ,70 PDs were incubated with either 10 mM CM-DCFH-DA

(B) or 5 mM MitoSOX RED (C) and analyzed by flow cytometry, as described in the Materials and Methods. The average G-values were computed, and statistical

analyses were performed using a two-tailed Student’s t-test (*P,0.005). (D,E) For fluorescence detection of the intracellular ROS levels, HFF cells stably

expressing the empty vector, hTERT or S227A/7A at ,70 PDs were incubated with 10 mM DCFH-DA and 5 mM MitoSOX RED. Digital images of DCF (D) and

MitoSOX (E) fluorescence were obtained by confocal microscopy and overlaid on the DAPI images. The mean fluorescence intensity per image was calculated

using MetaMorph software and averaged for ten images. Statistical analyses were performed using a two-tailed Student’s t-test (*P,0.01).

Nuclear translocation of hTERT 2693

Journ

alof

Cell

Scie

nce

translocated to the biologically functional nuclear compartment.A number of telomerase-associated proteins have been identified

in vertebrates and appear to regulate telomerase assembly (Leeet al., 2010), localization (Akiyama et al., 2003; Liu et al., 2001)and enzymatic function (Lee et al., 2004; Zhou and Lu, 2001).Nevertheless, the precise molecular changes that allow nuclear

import of hTERT remain poorly defined. In this study, wecharacterized the molecular mechanism regulating nuclear importof hTERT. We identified a bipartite NLS in hTERT (with the

amino acid sequence R222RRGGSASRSLPLPKRPRR240).Alanine substitutions of the basic amino acids in these motifsabolished the nuclear localization of hTERT, suggesting that this

bipartite NLS plays an essential role in nuclear import.Subcellular localization of hTERT is also regulated by aCRM1-dependent, leucine-rich canonical NES located in itsC-terminus (Seimiya et al., 2000; Kovalenko et al., 2010).

Consistently, we found that treatment of cells with LMBincreased the nuclear fluorescence signals of hTERT. Takentogether, these results suggest that hTERT undergoes active

nucleocytoplasmic shuttling by both active nuclear import andexport mechanisms.

The majority of the hTERT signal was detected in the nucleus

in all cell types examined, including H1299, HT1080 and MCF7cells. Although our proposed mechanism is based on experimentsusing transfected hTERT in H1299 cells, essentially the sameresults were obtained with Saos-2 cells, which lack endogenous

telomerase activity. Furthermore, endogenous hTERT was foundto localize mostly in the nucleus in H1299 and HT1080 cells.Therefore, nuclear localization would be true for both

endogenous and ectopically expressed hTERT and is notrelated to the cellular background in which hTERT isexpressed. These results suggest that NLS activity is stronger

than nuclear export activity in intact telomerase-positive cancercells. Although several proteins have been shown to facilitatenuclear localization of hTERT (Seimiya et al., 2000; Haendeler

et al., 2003; Jakob et al., 2008), it is also found to localize in thecytoplasm and to be involved in cellular proliferation andapoptosis apart from telomerase activity on telomeres (Xi et al.,2006; Massard et al., 2006; Del Bufalo et al., 2005). However, it

is still uncertain exactly how hTERT translocates betweensubcellular compartments.

There is accumulating evidence that telomerase activity is

regulated by phosphorylation of hTERT (Autexier and Lue,2006; Cong et al., 2002). It was reported that phosphorylation ofhTERT is induced by estrogen through a PI3K–Akt cascade in

human cancer cells, leading to telomerase activation (Kimuraet al., 2004; Kawagoe et al., 2003). Furthermore, regulation oftelomerase activity through phosphorylation of hTERT wasreported to correlate with nuclear localization of hTERT

(Akiyama et al., 2004; Liu et al., 2001). In resting CD4+ cells,hTERT is localized to the cytoplasm but upon activation it isphosphorylated and translocated to the nucleus (Liu et al., 2001).

Thus, phosphorylation of hTERT could be the initial step of thenuclear translocation of hTERT. Our data reveal that Akt-mediated phosphorylation at serine 227, but not at serine 824,

plays an important role in nuclear import of hTERT althoughboth sites are phosphorylated. Interestingly, serine 227 is locatedbetween two clusters of basic amino acids in the bipartite NLS.

Whereas wild-type hTERT and the S227E mutant predominantlylocalized to the nucleus, the S227A mutant was largely excludedfrom the nucleus. In addition, S227A was localized to the

cytoplasm even when NES activity was inhibited by LMBtreatment. These results suggest that the cytoplasmic retention of

S227A is not due to enhancement of nuclear export but to failureof nuclear import.

We postulate that both the bipartite NLS and phosphorylationof serine 227 contribute to nuclear import of hTERT in several

possible ways. First, phosphorylation might induce aconformational change to facilitate nuclear translocation,similar to other nucleocytoplasmic shuttling proteins. In the

case of ERK5, phosphorylation of the N-terminal half by MEK5results in the disruption of intramolecular interaction, resulting innuclear import (Kondoh et al., 2006). In hTERT, the activating

phosphorylation at the bipartite NLS might lead to aconformational switch that unmasks the NLS. Second, theconformational switch might subsequently induce a dockingsite for binding partners. For instance, hTERT phosphorylation

by Akt is required for its binding to NF-kB p65, and hTERTbound to NF-kB p65 dominantly translocates into the nucleus(Akiyama et al., 2003). Therefore, phosphorylation is crucial to

maintain hTERT in the correct conformation to facilitateinteraction with NF-kB p65. In addition, phosphorylation byAkt might also contribute to a conformation enabling the NLS

motif to interact with common nuclear import machinerycomponents. It has been shown that karyophilic proteins areimported into the nucleus through the NLS in a manner mediatedby importin-a/b and Ran (Chook and Blobel, 2001). Once

translocation through the nuclear pore complex is complete, thebinding of Ran–GTP to importin-b results in the dissociation ofimportin b from the importin a–NLS complex, promoting the

dissociation of importin-a and the cargo protein within thenucleus (Weis, 2003; Kosugi et al., 2009). Indeed, transfection ofRanQ69L caused the nuclear exclusion of hTERT, suggesting

that hTERT belongs to the class of proteins with nuclear importthrough an importin-a/b- and Ran-dependent pathway. Inagreement with the previous reports showing that RanQ69L

promotes nuclear export of proteins carrying a NES (Kehlenbachet al., 1998; Kehlenbach et al., 1999), we observed that the pre-existing hTERT complexes were not detected in the nucleus aftertransfection of RanQ69L.

Owing to the lack of detectable telomerase activity, HFF cellscan only be passaged for a limited number of generations (,50–60 PDs in our culture conditions). HFF cells stably expressing

wild-type hTERT have been passaged for .80 PDs afterantibiotic selection and were still growing normally.Surprisingly, the S227A/7A mutant substantially extended the

replicative lifespan of HFF cells despite having a dramaticallyreduced localization to the nucleus. Moreover, this extension wasaccompanied by an apparent complete maintenance of telomerelength for the first ,47 PDs after expression of S227A/7A. Thus,

it is intriguing how S227A/7A maintained telomere length atearly passages and delayed the senescent phenotype. Onepossible explanation is that, because about 19% of the S227A/

7A signals were found in the nucleus (Fig. 7F), the residual levelsof nuclear S227A/7A might be sufficient to maintain theproliferation capacity by extending telomeres in early passage

cells. However, although the underlying molecular mechanism isunclear, S227A/7A did not maintain telomere length in laterpassage cells, resulting in a senescent phenotype at PD 70. It has

been documented that telomerase elongates telomeric DNApreferentially during S phase of the cell cycle (Marcand et al.,2000). We therefore examined whether nuclear localization of

Journal of Cell Science 125 (11)2694

Journ

alof

Cell

Scie

nce

S227A/7A occurs preferentially during a particular phase of the

cell cycle. When the synchronized HFF cells were subjected to

the immunofluorescence staining, the majority of the S227A/7A

signal was still detected in the cytoplasm throughout the cell

cycle (data not shown). In addition, given that oxidative stress

accelerates telomere shortening in human fibroblasts (von

Zglinicki, 2002) and in human endothelial cells (Kurz et al.,

2004), the decrease in ROS production by the expression of

S227A/7A might contribute, at least in part, to telomere length

maintenance in early passage cells but not in later passage cells.

It is also possible that improved telomere maintenance in early

passage cells by the expression of S227A/7A might prevent

mitochondrial dysfunction and ROS production (Passos et al.,

2006; Sahin et al., 2011).

In this study, we have identified a bipartite NLS that is

responsible for nuclear localization of hTERT. Our data indicate

that Akt-mediated phosphorylation of serine 227 is important for

translocation of hTERT into the nucleus. Because the activation

of telomerase is a crucial step in human oncogenesis, the specific

inhibition of hTERT phosphorylation at the bipartite NLS could

be a new target for inhibiting telomerase activity in cancer.

Materials and MethodsPlasmid constructs

The FLAG–hTERT expression vector was constructed by inserting the EcoRI andSalI fragment from the full-length hTERT cDNA into the pCMV-tag2B vector(Agilent Technologies, Santa Clara, CA). To construct deletion mutant forms ofhTERT, DNA fragments with nested 59- and 39-deletions of the full-length hTERTcDNA were generated by PCR with appropriately designed primers and subclonedinto the EcoRI–SalI sites of the pCMV-tag2B vector. To create the Del101–200,Del201–300 and Del193–199 constructs, the overlap extension PCR method wasused to generate the deletions from the FLAG–hTERT vector (Ho et al., 1989), andthe PCR products were inserted into the EcoRI–SalI sites of the pCMV-tag2Bvector. The mutagenesis of hTERT was performed using the QuickChange site-directed mutagenesis kit according to the manufacturer’s instructions (AgilentTechnologies). The cDNAs were subcloned into pCMV-tag2B, and all constructswere verified by DNA sequencing.

Cell culture

The human lung carcinoma cell line H1299 was cultured in RPMI-1640 medium,and the human breast cancer cell line MCF-7 and fibrosarcoma cell line HT1080were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplementedwith 10% fetal bovine serum, 100 units/ml penicillin and 100 mg/ml streptomycinin 5% CO2 at 37 C. The human osteosarcoma cell line Saos-2 was maintained inMcCoy’s 5A medium containing 15% fetal bovine serum in 5% CO2 at 37 C.Primary cultures of human fibroblasts were established from adult foreskin inDMEM supplemented with 10% fetal calf serum as described elsewhere (Shinet al., 2005).

Immunofluorescence analysis

Cells grown on glass coverslips were fixed with 2% formaldehyde prepared inphosphate-buffered saline (PBS) for 10 minutes and permeabilized with 0.2%Triton X-100 in PBS for 20 minutes. Cells were then blocked in PBS containing5% bovine serum albumin and incubated with mouse anti-FLAG antibody (Sigma,St. Louis, MO) overnight at 4 C. After washing with PBS, cells were incubatedwith Rhodamine-conjugated anti-mouse immunoglobulin (Santa CruzBiotechnology, Santa Cruz, CA). To detect endogenous hTERT, cells wereincubated with rabbit anti-hTERT antibody (Rockland Immunochemicals,Gilbertsville, PA) overnight at 4 C, followed by incubation with either FITC-conjugated anti-rabbit immunoglobulin or Rhodamine-conjugated anti-rabbitimmunoglobulin (Invitrogen, Carlsbad, CA). DNA was stained with 4,6-diamino-2-phenylindole (DAPI; Vectashield; Vector Laboratories, Burlingame,CA). Fluorescence images were captured using an Olympus BX61 fluorescencemicroscopy. To quantitatively evaluate subcellular localization of the expressedproteins, the relative staining intensities in the nucleus and cytoplasm weremonitored for more than 100 cells per transfection and scored according to whetherthe staining in the nucleus exceeded that in the cytoplasm; the staining in thenucleus and the cytoplasm was equivalent; or staining in the cytoplasm exceededthat in the nucleus. The relative number of the cells in each category per totalnumber of the cells examined were calculated and expressed as a percentage. Dataare the means of three independent experiments.

Immunoprecipitation and immunoblotting

The expression vectors were transiently transfected into cells using Lipofectamine-PLUS reagent according to the manufacturer’s protocol (Invitrogen). Cells weresuspended in lysis buffer (1% NP-40, 50 mM Tris-HCl, pH 7.5, 137 mM NaCl,2 mM EDTA and 10% glycerol) with protease inhibitors (2 mMphenylmethylsulfonyl fluoride, 100 mM sodium orthovanadate, 100 mM sodiumfluoride, 10 mg/ml aprotinin and 10 mg/ml leupeptin). Lysates were thencentrifuged at 15,000 g for 15 minutes to remove insoluble material. Forimmunoprecipitation, lysates were preincubated with protein-G–Sepharose andincubated with primary antibodies precoupled to protein-G–Sepharose beads for2 hours at 4 C. The precipitated proteins were washed extensively and subjected toimmunoblot analysis. Immunoprecipitation and immunoblot analyses wereperformed using anti-hTERT (Rockland Immunochemicals), anti-FLAG (Sigma),anti-phosphorylated Akt substrate, anti-Akt (Cell Signaling Technology, Danvers,MA), anti-HSP90, anti-Myc, anti-HA (Santa Cruz Biotechnology) and anti-actin(Sigma) antibodies.

Establishment of stable cell lines

Stable cell lines expressing the empty vector, hTERT and the S227A/7A mutantwere established using the ViraPower lentiviral expression system according to themanufacturer’s instructions (Invitrogen). Briefly, the full-length cDNA wassubcloned into a pLenti6/V5-DEST lentiviral vector and co-transfected into293FT cells with packaging vectors (ViraPower packaging Mix) usingLipofectamine 2000 (Invitrogen) to generate a lentiviral stock. The culturesupernatants containing viral vector particles were harvested 72 hours aftertransfection and filtered through a 0.45 mm membrane filter. Human foreskinfibroblasts were infected with the different lentiviruses at the same multiplicity ofinfection (m.o.i.51) and selected with blasticidin (Invitrogen) for 2 weeks as masscultures. We did not observe any significant difference in the times to confluencebetween the empty vector-, hTERT- and S227A/7A-expressing cell lines. Weperformed independent infections at least twice for each type of virus and obtainedthe similar results.

Telomerase assay

The telomeric repeat amplification protocol (TRAP) was used as previouslydescribed (Lee et al., 2004). Briefly, cell extracts (200 ng of protein) were added tothe telomerase extension reactions and incubated for 20 minutes at 37 C. PCR wasperformed using the HTS primer and HACX primer for 30 cycles (denaturation at94 C for 30 seconds, annealing at 62 C for 30 seconds and extension at 72 C for30 seconds). As an internal telomerase assay standard, NT and TSNT primers wereadded to the PCR mixture as previously described (Kim and Wu, 1997).Telomerase products were resolved by electrophoresis on a 10% nondenaturingpolyacrylamide gel. Bands were then visualized by staining with SYBR Green(Invitrogen), and the signal intensity was quantified with a LAS-4000 Plus Imageanalyzer (Fujifilm, Tokyo, Japan).

Senescence-associated b-galactosidase (SA-b-Gal) assay

Stable cell lines expressing the empty vector, hTERT or the S227A/7A mutantwere plated in side culture chambers at the indicated population doublings (PDs)and washed with PBS, followed by fixation with 2% formaldehyde/0.2%glutaraldehyde for 5 minutes. Cells were washed with PBS twice and thenincubated for 12 hours with SA-b-Gal staining solution (1 mg/ml X-Gal, 40 mMcitric acid/sodium phosphate, pH 6, 5 mM potassium ferrocyanide, 5 mMferricyanide, 150 mM NaCl and 2 mM MgCl2. For quantification of senescentcells, a total of 200 cells were counted in each field and four fields were examined.

Terminal restriction fragment (TRF) analysis

To measure the telomere length, genomic DNA was digested with RsaI and HinfIand separated on a 0.7% agarose gel. DNA samples were transferred to a nylonmembrane (Amersham Biosciences) and hybridized with a 32P-labeled probe(TTAGGG)20. Signals were detected with a phosphoimage analyzer (Fujifilm).

Flow cytometric analysis of intracellular ROS

MitoSOX RED (Invitrogen) and 5-(and-6)-chloromethyl-29,79-dichlorodihydrofluorescein diacetate (CM-DCFH-DA; Invitrogen) were used todetect mitochondrial superoxide and intracellular peroxides, respectively. Briefly,cells were harvested and washed with PBS, followed by incubation with either10 mM CM-DCFH-DA or 5 mM MitoSOX RED for 30 minutes at 37 C in thedark. After washing with PBS, cells were analyzed by flow cytometry (BDBiosciences, Franklin Lakes, NJ). At least 10,000 events were analyzed. Forfluorescence detection of intracellular ROS levels, cells were incubated with either10 mM CM-DCFH-DA or 5 mM MitoSOX RED for 30 minutes in the incubator.Digital images of DCF and MitoSOX fluorescence were captured using a confocallaser-scanning microscope (Carl Zeiss, Jena, Germany) and overlaid using LSMsoftware. The mean fluorescence intensity per image was calculated and averagedover the ten images, using MetaMorph software (Molecular Devices, Sunnyvale,CA).

Nuclear translocation of hTERT 2695

Journ

alof

Cell

Scie

nce

FundingThis work was supported in part by a World Class University Fund fromthe Korean Ministry of Education, Science and Technology [grantnumber R31-2009-000-10086-0]; and by a Korea Research FoundationGrant from the Korean Ministry of Education, Science and Technology[grant number KRF-M1075604000107N560400110].

Supplementary material available online at

http://jcs.biologists.org/lookup/suppl/doi:10.1242/jcs.099267/-/DC1

ReferencesAhmed, S., Passos, J. F., Birket, M. J., Beckmann, T., Brings, S., Peters, H., Birch-

Machin, M. A., von Zglinicki, T. and Saretzki, G. (2008). Telomerase does notcounteract telomere shortening but protects mitochondrial function under oxidativestress. J. Cell Sci. 121, 1046-1053.

Akiyama, M., Hideshima, T., Hayashi, T., Tai, Y. T., Mitsiades, C. S., Mitsiades, N.,

Chauhan, D., Richardson, P., Munshi, N. C. and Anderson, K. C. (2003). Nuclear

factor-kappaB p65 mediates tumor necrosis factor alpha-induced nuclear translocationof telomerase reverse transcriptase protein. Cancer Res. 63, 18-21.

Akiyama, M., Yamada, O., Hideshima, T., Yanagisawa, T., Yokoi, K., Fujisawa, K.,

Eto, Y., Yamada, H. and Anderson, K. C. (2004). TNFalpha induces rapidactivation and nuclear translocation of telomerase in human lymphocytes. Biochem.

Biophys. Res. Commun. 316, 528-532.

Autexier, C. and Lue, N. F. (2006). The structure and function of telomerase reversetranscriptase. Annu. Rev. Biochem. 75, 493-517.

Blackburn, E. H. (2001). Switching and signaling at the telomere. Cell 106, 661-673.

Bodnar, A. G., Ouellette, M., Frolkis, M., Holt, S. E., Chiu, C. P., Morin, G. B.,

Harley, C. B., Shay, J. W., Lichtsteiner, S. and Wright, W. E. (1998). Extension oflife-span by introduction of telomerase into normal human cells. Science 279, 349-

352.

Chook, Y. M. and Blobel, G. (2001). Karyopherins and nuclear import. Curr. Opin.

Struct. Biol. 11, 703-715.

Cong, Y. S., Wright, W. E. and Shay, J. W. (2002). Human telomerase and itsregulation. Microbiol. Mol. Biol. Rev. 66, 407-425.

Del Bufalo, D., Rizzo, A., Trisciuoglio, D., Cardinali, G., Torrisi, M. R.,

Zangemeister-Wittke, U., Zupi, G. and Biroccio, A. (2005). Involvement of

hTERT in apoptosis induced by interference with Bcl-2 expression and function. Cell

Death Differ. 12, 1429-1438.

Dickmanns, A., Bischoff, F. R., Marshallsay, C., Luhrmann, R., Ponstingl, H. and

Fanning, E. (1996). The thermolability of nuclear protein import in tsBN2 cells issuppressed by microinjected Ran-GTP or Ran-GDP, but not by RanQ69L orRanT24N. J. Cell Sci. 109, 1449-1457.

Forsythe, H. L., Jarvis, J. L., Turner, J. W., Elmore, L. W. and Holt, S. E. (2001).Stable association of hsp90 and p23, but Not hsp70, with active human telomerase.

J. Biol. Chem. 276, 15571-15574.

Haendeler, J., Hoffmann, J., Brandes, R. P., Zeiher, A. M. and Dimmeler, S. (2003).

Hydrogen peroxide triggers nuclear export of telomerase reverse transcriptase via Srckinase family-dependent phosphorylation of tyrosine 707. Mol. Cell. Biol. 23, 4598-4610.

Haendeler, J., Drose, S., Buchner, N., Jakob, S., Altschmied, J., Goy, C.,

Spyridopoulos, I., Zeiher, A. M., Brandt, U. and Dimmeler, S. (2009).Mitochondrial telomerase reverse transcriptase binds to and protects mitochondrial

DNA and function from damage. Arterioscler. Thromb. Vasc. Biol. 29, 929-935.

Ho, S. N., Hunt, H. D., Horton, R. M., Pullen, J. K. and Pease, L. R. (1989). Site-

directed mutagenesis by overlap extension using the polymerase chain reaction. Gene

77, 51-59.

Holt, S. E., Aisner, D. L., Baur, J., Tesmer, V. M., Dy, M., Ouellette, M., Trager,

J. B., Morin, G. B., Toft, D. O., Shay, J. W. et al. (1999). Functional requirement ofp23 and Hsp90 in telomerase complexes. Genes Dev. 13, 817-826.

Indran, I. R., Hande, M. P. and Pervaiz, S. (2011). hTERT overexpression alleviatesintracellular ROS production, improves mitochondrial function, and inhibits ROS-mediated apoptosis in cancer cells. Cancer Res. 71, 266-276.

Jakob, S., Schroeder, P., Lukosz, M., Buchner, N., Spyridopoulos, I., Altschmied, J.

and Haendeler, J. (2008). Nuclear protein tyrosine phosphatase Shp-2 is oneimportant negative regulator of nuclear export of telomerase reverse transcriptase.

J. Biol. Chem. 283, 33155-33161.

Kalab, P., Weis, K. and Heald, R. (2002). Visualization of a Ran-GTP gradient in

interphase and mitotic Xenopus egg extracts. Science 295, 2452-2456.

Kang, S. S., Kwon, T., Kwon, D. Y. and Do, S. I. (1999). Akt protein kinase enhances

human telomerase activity through phosphorylation of telomerase reverse transcriptasesubunit. J. Biol. Chem. 274, 13085-13090.

Kawagoe, J., Ohmichi, M., Takahashi, T., Ohshima, C., Mabuchi, S., Takahashi, K.,

Igarashi, H., Mori-Abe, A., Saitoh, M., Du, B. et al. (2003). Raloxifene inhibitsestrogen-induced up-regulation of telomerase activity in a human breast cancer cellline. J. Biol. Chem. 278, 43363-43372.

Kawauchi, K., Ihjima, K. and Yamada, O. (2005). IL-2 increases human telomerasereverse transcriptase activity transcriptionally and posttranslationally through

phosphatidylinositol 39-kinase/Akt, heat shock protein 90, and mammalian target ofrapamycin in transformed NK cells. J. Immunol. 174, 5261-5269.

Kehlenbach, R. H., Dickmanns, A. and Gerace, L. (1998). Nucleocytoplasmic

shuttling factors including Ran and CRM1 mediate nuclear export of NFAT In vitro.J. Cell Biol. 141, 863-874.

Kehlenbach, R. H., Dickmanns, A., Kehlenbach, A., Guan, T. and Gerace, L. (1999).

A role for RanBP1 in the release of CRM1 from the nuclear pore complex in aterminal step of nuclear export. J. Cell Biol. 145, 645-657.

Kharbanda, S., Kumar, V., Dhar, S., Pandey, P., Chen, C., Majumder, P., Yuan,

Z. M., Whang, Y., Strauss, W., Pandita, T. K. et al. (2000). Regulation of the hTERTtelomerase catalytic subunit by the c-Abl tyrosine kinase. Curr. Biol. 10, 568-575.

Kim, J. H., Park, S. M., Kang, M. R., Oh, S. Y., Lee, T. H., Muller, M. T. and

Chung, I. K. (2005). Ubiquitin ligase MKRN1 modulates telomere lengthhomeostasis through a proteolysis of hTERT. Genes Dev. 19, 776-781.

Kim, N. W. and Wu, F. (1997). Advances in quantification and characterization oftelomerase activity by the telomeric repeat amplification protocol (TRAP). Nucleic

Acids Res. 25, 2595-2597.

Kim, N. W., Piatyszek, M. A., Prowse, K. R., Harley, C. B., West, M. D., Ho,

P. L. C., Coviello, G. M., Wright, W. E., Weinrich, S. L. and Shay, J. W. (1994).Specific association of human telomerase activity with immortal cells and cancer.Science 266, 2011-2015.

Kimura, A., Ohmichi, M., Kawagoe, J., Kyo, S., Mabuchi, S., Takahashi, T.,

Ohshima, C., Arimoto-Ishida, E., Nishio, Y., Inoue, M. et al. (2004). Induction ofhTERT expression and phosphorylation by estrogen via Akt cascade in human

ovarian cancer cell lines. Oncogene 23, 4505-4515.

Kondoh, K., Terasawa, K., Morimoto, H. and Nishida, E. (2006). Regulation of

nuclear translocation of extracellular signal-regulated kinase 5 by active nuclearimport and export mechanisms. Mol. Cell. Biol. 26, 1679-1690.

Kosugi, S., Hasebe, M., Matsumura, N., Takashima, H., Miyamoto-Sato, E.,

Tomita, M. and Yanagawa, H. (2009). Six classes of nuclear localization signalsspecific to different binding grooves of importin alpha. J. Biol. Chem. 284, 478-485.

Kovalenko, O. A., Caron, M. J., Ulema, P., Medrano, C., Thomas, A. P., Kimura,

M., Bonini, M. G., Herbig, U. and Santos, J. H. (2010). A mutant telomerasedefective in nuclear-cytoplasmic shuttling fails to immortalize cells and is associated

with mitochondrial dysfunction. Aging Cell 9, 203-219.

Kurz, D. J., Decary, S., Hong, Y., Trivier, E., Akhmedov, A. and Erusalimsky, J. D.

(2004). Chronic oxidative stress compromises telomere integrity and accelerates theonset of senescence in human endothelial cells. J. Cell Sci. 117, 2417-2426.

Lange, A., Mills, R. E., Lange, C. J., Stewart, M., Devine, S. E. and Corbett, A. H.

(2007). Classical nuclear localization signals: definition, function, and interactionwith importin alpha. J. Biol. Chem. 282, 5101-5105.

Lee, G. E., Yu, E. Y., Cho, C. H., Lee, J., Muller, M. T. and Chung, I. K. (2004).DNA-protein kinase catalytic subunit-interacting protein KIP binds telomerase byinteracting with human telomerase reverse transcriptase. J. Biol. Chem. 279, 34750-

34755.

Lee, J. H., Khadka, P., Baek, S. H. and Chung, I. K. (2010). CHIP promotes human

telomerase reverse transcriptase degradation and negatively regulates telomeraseactivity. J. Biol. Chem. 285, 42033-42045.

Li, H., Zhao, L., Yang, Z., Funder, J. W. and Liu, J. P. (1998). Telomerase iscontrolled by protein kinase Calpha in human breast cancer cells. J. Biol. Chem. 273,33436-33442.

Lingner, J., Cooper, J. P. and Cech, T. R. (1995). Telomerase and DNA endreplication: no longer a lagging strand problem? Science 269, 1533-1534.

Liu, D., O’Connor, M. S., Qin, J. and Songyang, Z. (2004). Telosome, a mammaliantelomere-associated complex formed by multiple telomeric proteins. J. Biol. Chem.

279, 51338-51342.

Liu, K., Hodes, R. J. and Weng Np, (2001). Cutting edge: telomerase activation inhuman T lymphocytes does not require increase in telomerase reverse transcriptase

(hTERT) protein but is associated with hTERT phosphorylation and nucleartranslocation. J. Immunol. 166, 4826-4830.

Marcand, S., Brevet, V., Mann, C. and Gilson, E. (2000). Cell cycle restriction of

telomere elongation. Curr. Biol. 10, 487-490.

Massard, C., Zermati, Y., Pauleau, A. L., Larochette, N., Metivier, D., Sabatier, L.,

Kroemer, G. and Soria, J. C. (2006). hTERT: a novel endogenous inhibitor of themitochondrial cell death pathway. Oncogene 25, 4505-4514.

Oh, W., Lee, E. W., Lee, D., Yang, M. R., Ko, A., Yoon, C. H., Lee, H. W., Bae,

Y. S., Choi, C. Y. and Song, J. (2010). Hdm2 negatively regulates telomeraseactivity by functioning as an E3 ligase of hTERT. Oncogene 29, 4101-4112.

Palacios, I., Weis, K., Klebe, C., Mattaj, I. W. and Dingwall, C. (1996). RAN/TC4mutants identify a common requirement for snRNP and protein import into thenucleus. J. Cell Biol. 133, 485-494.

Palm, W. and de Lange, T. (2008). How shelterin protects mammalian telomeres.Annu. Rev. Genet. 42, 301-334.

Passos, J. F., von Zglinicki, T. and Saretzki, G. (2006). Mitochondrial dysfunction andcell senescence: cause or consequence? Rejuvenation Res. 9, 64-68.

Petersen, S., Saretzki, G. and von Zglinicki, T. (1998). Preferential accumulation ofsingle-stranded regions in telomeres of human fibroblasts. Exp. Cell Res. 239, 152-

160.

Richards, S. A., Carey, K. L. and Macara, I. G. (1997). Requirement of guanosinetriphosphate-bound ran for signal-mediated nuclear protein export. Science 276,

1842-1844.

Sahin, E., Colla, S., Liesa, M., Moslehi, J., Muller, F. L., Guo, M., Cooper, M.,

Kotton, D., Fabian, A. J., Walkey, C. et al. (2011). Telomere dysfunction inducesmetabolic and mitochondrial compromise. Nature 470, 359-365.

Journal of Cell Science 125 (11)2696

Journ

alof

Cell

Scie

nce

Santos, J. H., Meyer, J. N., Skorvaga, M., Annab, L. A. and Van Houten, B. (2004).Mitochondrial hTERT exacerbates free-radical-mediated mtDNA damage. Aging Cell

3, 399-411.

Santos, J. H., Meyer, J. N. and Van Houten, B. (2006). Mitochondrial localization oftelomerase as a determinant for hydrogen peroxide-induced mitochondrial DNAdamage and apoptosis. Hum. Mol. Genet. 15, 1757-1768.

Seimiya, H., Sawada, H., Muramatsu, Y., Shimizu, M., Ohko, K., Yamane, K. and

Tsuruo, T. (2000). Involvement of 14-3-3 proteins in nuclear localization oftelomerase. EMBO J. 19, 2652-2661.

Shin, M. H., Rhie, G. E., Park, C. H., Kim, K. H., Cho, K. H., Eun, H. C. and

Chung, J. H. (2005). Modulation of collagen metabolism by the topical application ofdehydroepiandrosterone to human skin. J. Invest. Dermatol. 124, 315-323.

Siomi, M. C., Eder, P. S., Kataoka, N., Wan, L., Liu, Q. and Dreyfuss, G. (1997).Transportin-mediated nuclear import of heterogeneous nuclear RNP proteins. J. Cell

Biol. 138, 1181-1192.

Stewart, S. A. and Weinberg, R. A. (2006). Telomeres: cancer to human aging. Annu.

Rev. Cell Dev. Biol. 22, 531-557.Suntharalingam, M. and Wente, S. R. (2003). Peering through the pore: nuclear pore

complex structure, assembly, and function. Dev. Cell 4, 775-789.Vivanco, I. and Sawyers, C. L. (2002). The phosphatidylinositol 3-kinase AKT

pathway in human cancer. Nat. Rev. Cancer 2, 489-501.von Zglinicki, T. (2002). Oxidative stress shortens telomeres. Trends Biochem. Sci. 27,

339-344.Weis, K. (2003). Regulating access to the genome: nucleocytoplasmic transport

throughout the cell cycle. Cell 112, 441-451.Xi, L., Chen, G., Zhou, J., Xu, G., Wang, S., Wu, P., Zhu, T., Zhang, A., Yang, W.,

Xu, Q. et al. (2006). Inhibition of telomerase enhances apoptosis induced by sodium

butyrate via mitochondrial pathway. Apoptosis 11, 789-798.Zhou, X. Z. and Lu, K. P. (2001). The Pin2/TRF1-interacting protein PinX1 is a potent

telomerase inhibitor. Cell 107, 347-359.

Nuclear translocation of hTERT 2697

Journ

alof

Cell

Scie

nce