Recruitment of TRF2 to laser-induced DNA damage sites

Nazmul Huda a, Satoshi Abe a, Ling Gu b, Marc S. Mendonca a,c, Samarendra Mohanty b, David Gilley a,n

a Indiana University School of Medicine, Department of Medical and Molecular Genetics, Indianapolis, IN 46202 USAb University of Texas at Arlington, Biophysics and Physiology Group, Department of Physics, Arlington, TX, USAc Indiana University School of Medicine, Department of Radiation Oncology, Indianapolis, IN 46202, USA

a r t i c l e i n f o

Article history:

Received 11 June 2012

Received in revised form

10 July 2012

Accepted 17 July 2012Available online 27 July 2012

Keywords:

TRF2

Telomere

DNA damage response

Laser microirradiation

Phosphorylation

a b s t r a c t

Several lines of evidence suggest that the telomere-associated protein TRF2 plays critical roles in the

DNA damage response. TRF2 is rapidly and transiently phosphorylated by an ATM-dependent pathway

in response to DNA damage and this DNA damage-induced phosphoryation is essential for the DNA-PK-

dependent pathway of DNA double-strand break repair (DSB). However, the type of DNA damage that

induces TRF2 localization to the damage sites, the requirement for DNA damage-induced phosphoryla-

tion of TRF2 for its recruitment, as well as the detailed kinetics of TRF2 accumulation at DNA damage

sites have not been fully investigated. In order to address these questions, we used an ultrafast

femtosecond multiphoton laser and a continuous wave 405-nm single photon laser to induce DNA

damage at defined nuclear locations. Our results showed that DNA damage produced by a femtosecond

multiphoton laser was sufficient for localization of TRF2 to these DNA damage sites. We also

demonstrate that ectopically expressed TRF2 was recruited to DNA lesions created by a 405-nm laser.

Our data suggest that ATM and DNA-PKcs kinases are not required for TRF2 localization to DNA damage

sites. Furthermore, we found that phosphorylation of TRF2 at residue T188 was not essential for its

recruitment to laser-induced DNA damage sites. Thus, we provide further evidence that a protein

known to function in telomere maintenance, TRF2, is recruited to sites of DNA damage and plays critical

roles in the DNA damage response.

& 2012 Elsevier Inc. All rights reserved.

Introduction

Telomeres are nucleoprotein structures at chromosome ends,which act as protective caps to disguise chromosome ends frombeing recognized as a DNA double-strand break (DSB), therebypreventing chromosome end-to-end fusions [1]. Dysfunctional tel-omeres are known to cause genomic instability through chromo-some end fusion-induced breakage-fusion-bridge cycles [2]. There ismounting evidence that telomere dysfunction driven genomicinstability may lead to cancer and other age-related diseases [3,4].

A critical function of the telomere-associated protein TRF2 ismaintenance of telomere integrity [5]. However, TRF2 has nowbeen shown to be associated with a number of key DNA damageresponse proteins such as ATM [6,7], the MRN complex [8], Ku [9],DNA-PKcs [10], WRN protein [11], and BLM [12] and appears tocooperate with their DNA repair functions. These discoveriessuggest that the loss of TRF2 function may not only lead totelomere dysfunction but also hamper the DNA damage response[13]. Indeed, several studies now indicate that TRF2 is linked withthe DNA damage response not only at the telomere but in the

global DNA damage response as well [5,7,13–25]. The data sug-gests that TRF2 associates with DNA double-strand breaks rapidlyin response to DNA damage [7]. In addition, TRF2 appears to betransiently phosphorylated on DNA damage by an ATM-dependentpathway [24]. Furthermore, phosphorylation of TRF2 at residueT188 plays a critical role in the DNA-PK-dependent pathway ofDNA DSB repair [25]. Taken together, these data support theconcept that TRF2 functions in the DNA damage response, althoughthe precise role of TRF2 in this process is not yet clear.

Laser microirradiation generates DNA lesions at definednuclear DNA locations and has been successfully used to localizeDNA damage response proteins to these defined DNA lesions[18–21]. Importantly, the nature of damage produced depends onlaser intensity and wavelength [26–31]. Bradshaw et al. [7]reported that human TRF2 accumulated to the 390-nm pulsedlaser microbeam irradiation induced DNA damage sites. However,Williams et al. [32] reported that DNA DSBs were not sufficient todetect recruitment of TRF2 when a 405-nm continuous wave laserwas utilized to induce DNA damage. We therefore hypothesizedthat laser conditions and limits of assay detection levels mayinfluence the outcome of these experiments. Therefore, we testedseveral laser and molecular assay conditions to better understandthe role of TRF2 in the DNA damage response. We report here thatDNA damage produced by femtosecond multiphoton laser was

n Corresponding author.

E-mail address: dpgilley@iupui.edu (D. Gilley).

Reproduced from Free Radical Biology and Medicine 53: 1192-1197 (2012).

Nazmul Huda: Participant of the 25th UM, 1997-1998.

411

sufficient for the recruitment of endogenous TRF2 to DSB sites.In addition, we found that DNA damage induced by a 405-nmlaser caused the recruitment of ectopically expressed TRF2 toDNA damage sites. Additionally, we demonstrate that transientDNA damage-induced phosphorylation of TRF2 was not essentialfor its recruitment to DNA damage sites. Taken together, ourresults provide further positive evidence that TRF2 is recruited toa variety of laser-induced DNA damage sites and likely playsimportant roles in the early DNA damage response.

Materials and methods

Cells and cell culture

Human Tet-Off HT1080 cell line was purchased from Clontech(CA, USA). Human BJ fibroblast cells were generously provided byWoodring Wright, University of Texas Southwestern MedicalCenter, Dallas. Human MO59K and MO59J cells were kind giftsfrom Janice M. Pluth, Lawrence Berkeley National Laboratory,Berkeley, Callifornia. AT primary fibroblast were obtained fromCoriell Cell Repository (Camden, NJ). The culture condition andmethod of expression of wild-type and mutant TRF2 in Tet-OffHT1080 have been described elsewhere [25]; all other cells werecultured in Advanced DMEM (Invitrogen Corp., CA, USA) supple-mented with 10% fetal bovine serum. Cells were cultured in cellculture CO2 incubator at 37 1C in an atmosphere of 5% CO2.

Vector construction

Tet-Off-inducible c-myc-tagged wild-type and mutant TRF2expression vectors were constructed by the cloning of PCR pro-ducts into pTRE2hyg vector (Clontech, USA). PCR products wereamplified by using Phusion High-Fidelity polymerase (NEB, USA),c-myc tag sequence-containing primer, and previously reportedwild-type and mutant TRF2 expression vectors as templates.

Induction of localized DNA damage by lasers

Two-photon 800-nm laser microbeam irradiation was carriedout using an Olympus FV1000-MPE confocal/multiphoton micro-scope equipped with a Newport-SpectraPhysics MaiTai Deep Seelaser by modifying a previously described method [33]. Briefly,cells were grown in appropriate media on gridded glass-bottomculture dishes (MatTek Co., MA, USA) until nearly 50% confluent.Cells were treated with 10 ng/ml Hoechst 33258 in medium for30 min at 37 1C in 5% CO2. The cells were washed and resus-pended in microscopy medium containing 25 mM Hepes inphenol red-free DMEM (HyClone, UT, USA) supplemented with10% FBS. The two-photon 800-nm laser beam with pulse width of200 fs and repetition rate of 80 MHz was focused through a 63x,1.42 NA objective. The laser energy at the irradiated site wasmeasured with a laser power meter (Nova, Ophir Optronics,Israel) and the optimum laser output (0.15 nJ/pulse) at whichfemtosecond multiphoton laser irradiation is performed wasdetermined. In parallel experiments, a 50 mW, 405-nm diodelaser coupled with an Olympus FV1000-MPE confocal microscopewas used to induce local nuclear damage. In order to create a linepattern inside the nucleus, laser irradiation was carried out inpredefined locations using a 63x, 1.42 NA objective with ascanning rate of 40 ms/pixel and a power output of 65%. Wetitrated laser output at variable scanning rates to determine theoptimum level of DSBs in the cell nucleus. At 65% power, the405 nm laser beam generated 1.25 mW to create a single line scan.All laser irradiations were performed at 37 1C in a stage warmer.

Immunostaining

After laser treatment, cells were incubated, fixed, permeabi-lized, and blocked according to the method described byBradshaw et al. [7]. DNA breaks in irradiated cells were labeledwith fluorescein-dUTP following the manufacturer’s recommen-dations (Roche, IN, USA). Rabbit polyclonal anti-g-H2AX(Millipore, USA) antibody was also used as a DNA double-strandbreak marker. TRF2 was stained with mouse anti-TRF2 antibody(Millipore, MA, USA). Anti-Rap1 and anti-hTert antibodies werepurchased from Millipore. Ectopically expressed c-myc-taggedTRF2 was stained with mouse anti-c-myc antibody (Roche).Primary antibodies for g-H2AX, TRF2, and c-myc were diluted to1:100 in 1X PBS containing 1% BSA. Secondary antibodies usedwere goat anti-mouse rhodamine (Molecular Probes, OR, USA),goat anti-rabbit Alexa-488 (Molecular Probes), and goat anti-rabbit Texas Red (Molecular Probes). After rinsing, cells weremounted in Vectashield antifade with DAPI (Vector Lab, CA, USA).

Induction of DNA damage by ultrafast laser microbeam and live

imaging

For live imaging of the accumulation of TRF2 at DNA damagesites, the tunable Ti:Sapphire laser beam (MaiTai HP, Newport-SpectraPhysics, CA, USA) was coupled via the laser port of aninverted Nikon microscope using a dichroic mirror reflecting NIRlaser and transmitting visible light. The femtosecond laser beamwas expanded using a beam expander, and focused to a diffrac-tion-limited spot by a 100x (1.3 NA) microscope objective.A shutter was utilized to control the duration of exposure perfocused spot. For induction of DNA damage, spot and line lasermicroirradiation patterns were created in nuclear sites of HT1080cells by piezo-scanning of the stage. A polarizer was used tocontrol laser beam power (pulse energy). A halogen lamp andfluorescence lamp were used for bright-field and fluorescenceimaging, respectively. Suitable excitation (490–520 nm) andemission (530–560 nm) band-pass filters were used for the liveimaging of recruitment of specific proteins. For quantification ofaccumulation of TRF2 at DNA lesion sites in ectopic TRF2-expres-sing cells, a cooled CCD camera (Coolsnap, Photometrics Inc.) wasused in the side port of the inverted Nikon laser microscope.

Western blotting

Western blotting was carried out as described elsewhere[24,25]. Mouse monoclonal anti-TRF2 (1:1000; 4A794, Millipore)was used as a primary antibody and horseradish peroxidase-conjugated goat anti-mouse antibody (Santa Cruz Biotechnology)was used as a secondary antibody in immunoblot analyses.

Results

The ultrafast femtosecond lasers generate high peak powers,making them ideal tools for directing submicrometer size radiantenergy within a particular region of a live cell nuclei to activateionizing and photochemically driven processes that cause DNAdamage [34]. We first employed an 800-nm ultrafast femtosecondmultiphoton laser beam to create DSBs in a predefined regionacross the cell nucleus and studied TRF2 recruitment to these DSBsites using immunofluorescence. The presence of free DNA ends atthe laser-irradiated locations was confirmed by TUNEL assay(Fig. 1A, B, D–F). HT1080 cells were incubated for 5 min afterlaser irradiation and subsequently fixed and analyzed for TRF2 tolaser-induced DNA damage sites by immunofluorescence. Wefound that endogenous TRF2 was localized to the DNA damaged

412

sites induced by an 800-nm ultrafast femtosecond multiphotonlaser beam (Fig. 1A, B, and F).

The specific type of DNA damage induced by the 800-nmultrafast femtosecond multiphoton laser beam depends on wave-length, power density, and pulse duration [28]. We thereforecalculated the fluorescence intensity of the immunostained TRF2at the DSB sites by ImageJ (Fig. 1C). We found an approximatedouble amount of TRF2 localization with 0.3 nJ/pulse energycompared to that of 0.15 nJ/pulse energy (Fig. 1C). Therefore,higher levels of laser energy were found to cause increased levelsof TRF2 recruitment to damage sites (Fig. 1A, B, and C), suggestingthat the localization of TRF2 to the DNA damage sites was laser-dose dependent. In addition, cells were immunostained with anti-Rap1 and anti-hTERT antibodies to determine if these proteinslocalize to DNA damage sites. However, we did not see anydetectable localization of these proteins to the laser-inducedDNA damage sites under these conditions (Fig. 1D and E).In order to investigate if TRF2 accumulation to DNA damage siteswas specific to HT1080 cells, we also tested BJ primary fibroblastcells for TRF2 localization to damage sites. Indeed, we observedsimilar accumulation of endogenous TRF2 to the DNA damage sitein BJ cells as observed in the HT1080 cell line suggesting thatTRF2 recruitment to laser-induced DNA sites is not cell specific(Fig. 1F).

Next, we performed live imaging of HT1080 cells expressingYFP-tagged TRF2 to study the kinetics of TRF2 localizationto laser-induced DNA damage sites. In Fig. 2, we show livecell imaging kinetics of the recruitment of TRF2-YFP to ultrafast

laser-induced DNA damage lesions in HT1080 cells. Cells expres-sing TRF2-YFP led to accumulation of tagged-TRF2 at spotsmicroirradiated sites within 5 s using a 200-fs laser beam(marked by arrow in Fig. 2A, ii; 810 nm, energy/pulse 0.8 nJ, andexposure of 40 ms). The fluorescence intensity of TRF2 at theselaser-induced DNA lesions was found to decrease after severalminutes (Fig. 2A, iii–v). Laser microirradiation at multiple sites inthe same nucleus led to TRF2-YFP accumulation at those lesionsites (Fig. 2A, v). For Fig. 2A, v, an 810-nm laser with single fspulse energy of 0.7 nJ and exposure of 20 ms was used. Fig. 2Bshows the kinetics of TRF2-YFP accumulation at femtosecondlaser-irradiated spot DNA damage sites at two different laserenergies. The increase in normalized TRF2-YFP fluorescenceintensity (I) was fitted to the equation I¼ I0

n[1–e(-cnt)] to obtain

the recruitment time constant. For a 0.5 nJ/fs pulse and exposuretime of 20 ms (Fig. 2B, i), TRF2 recruitment time to laser-inducedDNA damage sites was 22.5 s. The TRF2 recruitment time forpulse energy of 0.8 nJ and single exposure (20 ms) was slightlyless at 20.5 s (Fig. 2B, ii). The values of ‘c’ for the YFP-control are0.52704 71.7311 and 0.19369 7 0.39058, respectively for 0.5(Fig. 2B, i) and 0.8 nJ/pulse (Fig. 2B, ii). However, the recruitmenttime of TRF2 to DNA damage sites was found to decreasesignificantly with repeat exposures. For 0.8 nJ pulses, the accu-mulation time for single 20 ms exposure (20.5 s, Fig. 2B, ii)decreased to 2.1 s for a second exposure of 20 ms within thesame nucleus. With a third 20 ms exposure within the samenucleus at the same laser pulse energy (0.8 nJ), the accumulationtime of TRF2 at DNA damage sites decreased to 1.8 s. This impliesthat with repetitive exposures within the same nucleus, the two-photon efficacy of femtosecond-laser-induced DNA damageincreases, leading to a significant reduction in TRF2 recruitmenttime to secondary and tertiary DNA damage sites within the samenucleus. While nonlinear effects are more intensity dependent(i.e., the increase in exposures has a more profound effect thanthe laser energy), it is hypothesized that the observed increase intwo-photon absorption after multiple short exposures may bedue to a highly positioned and primed DNA damage responsewithin these cells after laser-induced DNA damage.

In response to DNA damage, ATM kinase activates many DNAdamage response pathways [35,36]. It has been demonstratedthat TRF2 binds to ATM [37], suggesting that ATM may modulateTRF2 recruitment to DNA damage sites. In addition, it has beenshown that on DNA damage TRF2 is phosphorylated transientlyby an ATM-dependent pathway [24]. However, when we inves-tigated TRF2 recruitment to DNA damage sites in the ATM-deficient cells, A-T (ataxia telangiectasia) human primary fibro-blasts [38], we clearly observed TRF2 recruitment to laser-induced DNA damage sites. This strongly suggests that ATM isnot required for the recruitment of TRF2 to laser-induced DNAdamage sites (Fig. 3A). Additionally, to determine if there is anassociation between the TRF2 accumulation at DNA damage sitesand the DNA-dependent protein kinase catalytic subunit (DNA-PKcs) [39] we used the human glioma cell lines, M059J (deficientin DNA-PKcs) and M059K (wild-type for DNA-PKcs). Despite theabsence of DNA-PKcs in MO59J cells, we observed TRF2 accumu-lation at the laser-irradiated sites in these cells (Figs. 3B and C).Therefore, these results indicate that neither ATM nor DNA-PKcskinases are essential for TRF2 recruitment to laser-induced DNAdamage sites.

Previously, Bradshaw et al. [7] showed rapid accumulation ofTRF2 to DNA lesions induced by 390-nm pulsed laser microbeamirradiation. Later, Williams et al. [32] investigated TRF2 localiza-tion to DNA lesions utilizing a 405-nm continuous wave laser, butcould not detect TRF2 at DNA damage sites under their condi-tions. We therefore investigated whether laser type and energycould influence the outcome of these experiments. Interestingly,

Fig. 1. Recruitment of endogenous TRF2 to the DNA DSB sites induced by an

ultrafast femtosecond (fs) multiphoton 800-nm laser in HT1080 and BJ fibroblast

cells. (A) A defined area in the nucleus of an HT1080 cell was irradiated with the

multiphoton fs laser (wavelength, 800 nm; pulse energy, 0.15 nJ; repetition rate,

80 MHz) using an Olympus FV1000-MPE confocal microscope equipped with a

Spectra Physics MaiTai Deep See laser system. Fluorescein-labeled dUTP (green)

formed colocalizing tracks in DAPI-stained nucleus of the cell fixed after 5 min of

laser microirradiation. Note: cells induced with this laser generally did not display

typical punctuate TRF2 foci patterns at telomeres. It is possible that a portion of

TRF2 from telomeres were recruited to these laser-induced DNA lesions.

(B) Similar experiment was performed with 0.30 nJ pulse energy keeping all other

parameters identical to those in panel A. As noted above, a typical punctuate TRF2

foci pattern at telomere was not visible in this microirradiated cell.

(C) Quantification of fluorescence intensities of TRF2 in panels A and B was

performed by ImageJ. Error bars indicate standard error. (D and E) After micro-

irradiation HT1080 as described in panel A, cells were immunostained for Rap1

(red) or hTERT (red), respectively. (F) BJ fibroblasts were irradiated under the same

conditions as those in panel A.

413

we also found undetectable levels of endogenous TRF2 at DNAlesions generated by a 405-nm continuous wave laser aftersensitization of DNA with Hoechest 33342 (Fig. 4A and B).We speculated that the number of intrinsic unbound TRF2recruited to DNA lesions created by a 405-nm laser may not besufficient to be detectable by immunofluorescence. In eachimmunofluorescence focus, hundreds of target molecules arerequired for immunofluorescence detection [40,41]. Therefore,to test this possibility, we conditionally and ectopically expressed

TRF2 in HT1080 cells under the control of the Tet-Off system [19].Our goal was to increase the amount of TRF2 molecules availablefor potential recruitment and possible immunofluorescencevisualization at DNA lesions generated by 405 nm laser exposure.Fig. 4D shows the expression levels of Tet-induced untaggedectopically expressed TRF2 compared to uninduced and parentalcells. We irradiated the cells after sensitization with Hoechest33342 and examined the presence of DSBs either by immuno-fluorescence labeling of g-H2AX or by TUNEL assay (Fig. 4).Indeed, on 405 nm laser exposure, ectopically expressed TRF2was observed at DNA lesions and colocalized with g-H2AX andTUNEL-labeled DSBs using immunofluorescence (Fig. 4C). Immu-nostaining with an antibody against c-myc also demonstratedthat c-myc-tagged TRF2 accumulates at the 405-nm laser-inducedDNA DSB damage sites (Fig. 5). Taken together, these resultssuggest that TRF2 is indeed recruited to 405-nm laser-inducedDNA damage sites but at levels that may not be sufficient to bedetectable by immunofluorescence alone.

In order to investigate if DNA damage-induced phosphoryla-tion of TRF2 is required for its localization to DNA damagesites, we irradiated HT1080 cells harboring ectopically expressedphosphorylation T188A null mutant c-myc-tagged-TRF2 (c-myc-TRF2T188A) and the phospho-mimick T188E) mutant c-myc-tagged-TRF2 (c-myc-TRF2T188E ) with a 405-nm laser [19]. Likethe wild-type ectopic expression TRF2 used above, mutant TRF2proteins were conditionally expressed under the control of theTet-Off system [19]. We checked the expression levels of thesemutants before irradiation and the Western blot result is shownin Fig. 5D. We observed that both mutant TRF2 proteins wererecruited to DNA damage sites as wild-type ectopically expressedTRF2 (Fig. 5B and C). These results suggest that phosphorylationof TRF2 at residue 188 is not required for its recruitment to DNAdamage sites.

Fig. 2. Live cell imaging of YFP-tagged TRF2 protein at DNA lesions induced by an ultrafast femtosecond multiphoton 800-nm laser. (A) Live imaging of TRF2 recruitment

to sites of laser-induced DNA damage in HT1080 cells using a tunable Ti:Sapphire laser beam (MaiTai HP, Newport-SpectraPhysics). (i) Fluorescence before irradiation in a

TRF2-YFP cell and YFP-control cell. (ii–vii) Time-lapse images of TRF2-YFP fluorescence accumulation and decay in laser-induced DNA damage site(s). The arrows indicate

the spot where the laser beam was directed to create damage. (B) Kinetics of TRF2-YFP accumulation (fluorescence intensity) at fs laser-irradiated spot at laser pulse

energy of (i) 0.5 nJ and (ii) 0.8 nJ. The rise in TRF2-YFP fluorescence is fitted to equation I¼ I0n[1–e(-c

nt)].

Fig. 3. Recruitment of endogenous TRF2 to the ultrafast femtosecond (fs) multi-

photon 800-nm laser-induced DNA lesion sites in ATM and DNA PKcs-deficient

cells. (A) AT primary fibroblast (ATM deficient), (B) human glioma cell lines

MO59K, and (C) MO59J (devoid of DNA-PKcs and also exhibit reduced expression

of ATM kinase, isogenic clone) cells were treated with a multiphoton femtosecond

laser under the conditions described in Fig. 1. Endogenous TRF2 (red) colocalized

with fluorescein-dUTP-labeled DSBs (green) which is independent of ATM or DNA-

PKcs.

414

Discussion

The work presented here provides direct evidence that thetelomeric protein TRF2 is recruited to DNA damage sites inducedby two wavelengths of laser microirradiation. We induced DNAdamage and monitored TRF2 recruitment within live cells using anultrafast femtosecond laser (800 nm) and with a continuous wave405-nm laser. The ultrafast femtosecond lasers generate high peakpowers making them ideal tools for depositing submicrometer sizeradiant energy within a particular region of a live cell nuclei toactivate ionizing and photochemically driven processes that causeDNA damage [34]. We used the ultrafast femtosecond multiphotonlaser beam because it eliminates nonspecific UV absorption bymolecules present within the cell, penetrates efficiently throughmammalian cells, and minimally affects adjacent cells. Recently,several groups reported that the ultrafast femtosecond multi-photon laser beam induces DNA damage response that recruitsDNA double-strand break repair proteins with minimal activationof other DNA repair pathways [42,43].

Williams et al. [24] did not observe detectable levels of endogen-ous TRF2 accumulation using immunofluorescence at DNA lesionsinduced by a 405-nm continuous wave diode laser. By contrast,Bradshaw et al. [7] induced DNA lesions with a 390-nm pulsedmicrobeam laser and observed a clear localization of endogenousTRF2 at these laser-induced DNA damage sites. We hypothesized thatDNA lesions created by the 405-nm laser may not induce sufficientlevels of endogenous TRF2 to be detectable by conventional immu-nofluorescence. As stated previously, several hundred moleculesof a specific target protein are generally required for a detectablefoci using immunofluorescence, thereby making interpretation of

negative results highly problematic for this experimental approach[40,41]. Therefore, to test our hypotheses, we ectopically expressedwild-type and mutant TRF2 and then irradiated with a 405-nm laser.We detected clear localization of ectopically expressed TRF2 at 405-nm laser-induced DNA lesions (Figs. 4 and 5), supporting ourhypothesis that a 405-nm laser is unable to recruit sufficient levelsof endogenous TRF2 to these DNA lesions to be visualized byimmunofluorescence. Taken together, our result presented here andthose of several other researchers support the concept that TRF2plays a critical role in the DNA damage response [7,19,24,25,44,45].

We previously reported that TRF2 is rapidly and transientlyphosphorylated in response to DNA damage [24]. We alsoreported that the DNA damage-induced phosphorylated form ofTRF2 is not bound to telomeric DNA in telomerase-dependent celllines [18]. Additionally, we reported that DNA damage-inducedphosphorylation plays a critical role in DNA DSB repair within thefirst hour during the DNA-PK-dependent pathway of DSB repair[25]. In the studies presented here we found that phosphorylationof TRF2 is not required for the recruitment of TRF2 to the DNAdamage sites (Fig. 5). Therefore, our results now suggest that DNAdamage-induced phosphorylation of TRF2 is likely to function atDNA damage sites. Future studies will investigate factors involvedin the recruitment of TRF2 and the mechanistic role of TRF2 andphosphorylated TRF2 at DNA damage sites.

Acknowledgments

We thank Dr. Malgorzata Kamocka for assisting in the confocallaser microscopy. We thank LiRen Tu for his work on the preparation

Fig. 4. Association of TRF2 with g-H2AX or fluorescein-dUTP-labeled (TUNEL

assay) DSBs induced by a 405-nm diode laser coupled with an Olympus FV1000-

MPE confocal microscope in ectopic TRF2-expressing HT1080 cells. (A and B)

Defined nuclear regions were irradiated with a continuous wave diode laser

(wavelength, 405 nm; energy delivered, 1.2 mW; scan number, 10; pulse width,

40 mS/pixel) in parental HT1080 cells. Green signals represent either g-H2AX or

TUNEL-labeled DSBs and the red signals represent TRF2. (C) TRF2 was ectopically

expressed in HT1080 and irradiated with 405-nm laser at the specific portion of

nuclei. DNA double-strand breaks were labeled with g-H2AX (green) and ectopic

TRF2 foci with an anti-TRF2 antibody. (D) Expression level of ectopic wild-type

TRF2; the levels of endogenous and noninduced TRF2 are also shown and actin

was used as a loading control.

Fig. 5. Accumulation of ectopic TRF2 mutants at the damaged sites induced by a

405-nm diode laser coupled with an Olympus FV1000-MPE confocal microscope

(A) Anti-c-myc antibody was used to observe c-myc-tagged-TRF2 accumulation at

the DNA DSB sites. (B and C) c-myc-tagged-TRF2 null mutant T188A (red) and

compensatory mutant T188E (red), respectively, formed foci with g-H2AX (green).

(D) Expression level of ectopic wild-type and mutant TRF2 proteins in HT1080.

Both endogenous and exogenous TRF2 and its mutants were detected by anti-TRF2

antibody and anti-c-myc antibody from induced (þ) and noninduced (�) cell

extracts. The actin was used as a loading control.

415

of figures. This work was supported by grants to D.G. from theIndiana University Simon Cancer Center, the American CancerSociety, the Showalter Foundation, the Susan G. Komen Foundation,the Avon Foundation, the Flight Attendant Medical Research Insti-tute, and the Indiana Genomics Initiative (INGEN). INGEN of IndianaUniversity is supported in part by Lilly Endowment Inc.

References

[1] Blackburn, E. H.; Greider, C. W.; Szostak, J. W. Telomeres and telomerase: thepath from maize, Tetrahymena and yeast to human cancer and aging. Nat.Med 12:1133–1138; 2006.

[2] O’Sullivan, R. J.; Karlseder, J. Telomeres: protecting chromosomes againstgenome instability. Nat. Rev. Mol.Cell. Biol. 11:171–181; 2010.

[3] Gilley, D. P. A two way street: the telomere and the DNA damage response.Cell Cycle 8:2855–2856; 2009.

[4] Artandi, S. E.; DePinho, R. A. Telomeres and telomerase in cancer. Carcinogen-esis 31:9–18; 2010.

[5] Palm, W.; de Lange, T. How shelterin protects mammalian telomeres. Annu.Rev. Genet. 42:301–334; 2008.

[6] Karlseder, J.; Hoke, K.; Mirzoeva, O. K.; Bakkenist, C.; Kastan, M. B.; Petrini, J.H. J.; de Lange, T. The telomeric protein TRF2 binds the ATM kinase and caninhibit the ATM-dependent DNA damage response. Plos Biol. 2:1150–1156;2004.

[7] Bradshaw, P. S.; Stavropoulos, D. J.; Meyn, M. S. Human telomeric proteinTRF2 associates with genomic double-strand breaks as an early response toDNA damage. Nat. Genet. 37:193–197; 2005.

[8] Zhu, X. D.; Kuster, B.; Mann, M.; Petrini, J. H.; de Lange, T. Cell-cycle-regulatedassociation of RAD50/MRE11/NBS1 with TRF2 and human telomeres. Nat.Genet. 25:347–352; 2000.

[9] Song, K.; Jung, D.; Jung, Y.; Lee, S. G.; Lee, I. Interaction of human Ku70 withTRF2. FEBS Lett 481:81–85; 2000.

[10] Bombarde, O.; Boby, C.; Gomez, D.; Frit, P.; Giraud-Panis, M. J.; Gilson, E.;Salles, B.; Calsou, P. TRF2/RAP1 and DNA-PK mediate a double protectionagainst joining at telomeric ends. EMBO J 29:1573–1584; 2010.

[11] Machwe, A.; Xiao, L. R.; Orren, D. K. TRF2 recruits the Werner syndrome(WRN) exonuclease for processing of telomeric DNA. Oncogene 23:149–156;2004.

[12] Lillard-Wetherell, K.; Machwe, A.; Langland, G. T.; Combs, K. A.; Behbehani, G.K.; Schonberg, S. A.; German, J.; Turchi, J. J.; Orren, D. K.; Groden, J.Association and regulation of the BLM helicase by the telomere proteinsTRF1 and TRF2. Hum. Mol. Genet 13:1919–1932; 2004.

[13] Zhang, Y. W.; Zhang, Z. W.; Miao, Z. H.; Ding, J. The telomeric protein TRF2 iscritical for the protection of A549 cells from both telomere erosion and DNAdouble-strand breaks driven by salvicine. Mol. Pharmacol 73; 2008. 1587–1587.

[14] De Boeck, G.; Forsyth, R. G.; Praet, M.; Hogendoom, P. C. W. Telomere-associated proteins: cross-talk between telomere maintenance and telomere-lengthening mechanisms. J. Pathol. 217:327–344; 2009.

[15] Pandita, T. Telomere metabolism and DNA damage response. In: Khanna, K.,Shiloh, Y., editors. The DNA damage response: implications on cancer formationand treatment. Springer; 2009.

[16] Celli, G. B. d. L. T. DNA processing is not required for ATM-mediated telomeredamage response after TRF2 deletion. Nat. Cell Biol 7:712–718; 2005.

[17] Guo, X.; Deng, Y.; Lin, Y.; Cosme-Blanco, W.; Chan, S.; He, H.; Yuan, G.; Brown,E. J.; Chang, S. Dysfunctional telomeres activate an ATM-ATR-dependent DNAdamage response to suppress tumorigenesis. EMBO J 26:4709–4719; 2007.

[18] Denchi, E. L.; de Lange, T. Protection of telomeres through independentcontrol of ATM and ATR by TRF2 and POT1. Nature 448:1068–1071; 2007.

[19] Kamranvar, S. A.; Masucci, M. G. The Epstein-Barr virus nuclear antigen-1promotes telomere dysfunction via induction of oxidative stress. Leukemia25:1017–1025; 2011.

[20] Opresko, P. L.; Fan, J. H.; Danzy, S.; Wilson, D. M.; Bohr, V. A. Oxidativedamage in telomeric DNA disrupts recognition by TRF1 and TRF2. NucleicAcids Res. 33:1230–1239; 2005.

[21] Muftuoglu, M.; Wong, H. K.; Imam, S. Z.; Wilson, D. M.; Bohr, V. A.; Opresko,P. L. Telomere repeat binding factor 2 interacts with base excision repairproteins and stimulates DNA synthesis by DNA polymerase beta. Cancer Res.66:113–124; 2006.

[22] Zhang, P. S.; Furukawa, K.; Opresko, P. L.; Xu, X. R.; Bohr, V. A.; Mattson, M. P.TRF2 dysfunction elicits DNA damage responses associated with senescence

in proliferating neural cells and differentiation of neurons. J. Neurochem.97:567–581; 2006.

[23] Miller, A. S.; Balakrishnan, L.; Buncher, N. A.; Opresko, P. L.; Bambara, R. A.Telomere proteins POT1, TRF1 and TRF2 augment long-patch base excisionrepair in vitro. Cell Cycle 11:998–1007; 2012.

[24] Tanaka, H.; Mendonca, M. S.; Bradshaw, P. S.; Hoelz, D. J.; Malkas, L. H.; Meyn,M. S.; Gilley, D. D. N. A. damage-induced phosphorylation of the humantelomere-associated protein TRF2. Proc. Natl. Acad. Sci. USA 102:15539–15544;2005.

[25] Huda, N.; Tanaka, H.; Mendonca, M. S.; Gilley, D. DNA Damage-inducedphosphorylation of TRF2 is required for the fast pathway of DNA double-strand break repair. Mol. Cell. Biol. 29:3597–3604; 2009.

[26] Botchway, S. W.; Reynolds, P.; Parker, A. W.; O’Neill, P. Laser-inducedradiation microbeam technology and simultaneous real-time fluorescenceimaging in live cells. Methods Enzymol. 504:3–28; 2012.

[27] Gomez-Godinez, V.; Wu, T.; Sherman, A. J.; Lee, C. S.; Liaw, L. H.; You, Z. S.;Yokomori, K.; Berns, M. W. Analysis of DNA double-strand break responseand chromatin structure in mitosis using laser microirradiation. Nucleic AcidsRes. 38; 2010.

[28] Kruhlak, M. J.; Celeste, A. Nussenzweig, A. Monitoring DNA breaks in opticallyhighlighted chromatin in living cells by laser scanning confocal microscopy.Methods Mol. Biol 523:125–140; 2009.

[29] Lan, L.; Nakajima, S.; Oohata, Y.; Takao, M.; Okano, S.; Masutani, M.; Wilson,S. H.; Yasui, A. In situ analysis of repair processes for oxidative DNA damagein mammalian cells. Proc. Natl. Acad. Sci. USA 101:13738–13743; 2004.

[30] Singh, D. K.; Karmakar, P.; Aamann, M.; Schurman, S. H.; May, A.; Croteau, D.L.; Burks, L.; Plon, S. E.; Bohr, V. A. The involvement of human RECQL4 in DNAdouble-strand break repair. Aging Cell 9:358–371; 2010.

[31] Lan, L.; Nakajima, S.; Komatsu, K.; Nussenzweig, A.; Shimamoto, A.; Oshima,J.; Yasui, A. Accumulation of Werner protein at DNA double-strand breaks inhuman cells. J. Cell Sci. 118:4153–4162; 2005.

[32] Williams, E. S.; Stap, J.; Essers, J.; Ponnaiya, B.; Luijsterburg, M. S.; Krawczyk,P. M.; Ullrich, R. L.; Aten, J. A.; Bailey, S. M. DNA double-strand breaks are notsufficient to initiate recruitment of TRF2. Nat. Genet. 39:696–698; 2007.

[33] Kong, X.; Mohanty, S. K.; Stephens, J.; Heale, J. T.; Gomez-Godinez, V.; Shi, L. Z.;Kim, J. S.; Yokomori, K.; Berns, M. W. Comparative analysis of different lasersystems to study cellular responses to DNA damage in mammalian cells. NucleicAcids Res. 37; 2009.

[34] Botchway, S. W.; Reynolds, P.; Parker, A. W.; O’Neill, P. Use of near infraredfemtosecond lasers as sub-micron radiation microbeam for cell DNA damageand repair studies. Mutat. Res. 704:38–44; 2010.

[35] Abraham, R. T. P. I. 3-kinase related kinases: ‘big’ players in stress-inducedsignaling pathways. DNA Repair 3:883–887; 2004.

[36] Bensimon, A.; Aebersold, R.; Shiloh, Y. Beyond ATM: the protein kinaselandscape of the DNA damage response. FEBS Lett 585:1625–1639; 2011.

[37] Karlseder, J.; Broccoli, D.; Dai, Y. M.; Hardy, S.; de Lange, T. p53- and ATM-dependent apoptosis induced by telomeres lacking TRF2. Science283:1321–1325; 1999.

[38] McKinnon, P. J. ATM and the molecular pathogenesis of ataxia telangiectasia.Annu. Rev. Pathol 7:303–321; 2012.

[39] Neal, J. A.; Meek, K. Choosing the right path: does DNA-PK help make thedecision? Mutat Res.-Fund. Mol. M 711:73–86; 2011.

[40] Bonner, W. M.; Redon, C. E.; Dickey, J. S.; Nakamura, A. J.; Sedelnikova, O. A.;Solier, S.; Pommier, Y. GammaH2AX and cancer. Nat. Rev. Cancer 8:957–967;2008.

[41] Laszlo, G.; Dickler, H. B. Detection of low numbers of lymphocyte surface-membrane molecules using concentration immunofluorescence analysis(Cia). Hybridoma 9:111–117; 1990.

[42] Gomez-Godinez, V.; Wakida, N. M.; Dvornikov, A. S.; Yokomori, K.; Berns, M.W. Recruitment of DNA damage recognition and repair pathway proteinsfollowing near-IR femtosecond laser irradiation of cells. J. Biomed. Opt. 12;2007.

[43] Ha, K.; Fiskus, W.; Rao, R.; Balusu, R.; Venkannagari, S.; Nalabothula, N. R.;Bhalla, K. N. Hsp90 inhibitor-mediated disruption of chaperone association ofATR with hsp90 sensitizes cancer cells to DNA damage. Mol. Cancer Ther10:1194–1206; 2011.

[44] Splinter, J.; Jakob, B.; Lang, M.; Yano, K.; Engelhardt, J.; Hell, S. W.; Chen, D. J.;Durante, M.; Taucher-Scholz, G. Biological dose estimation of UVA lasermicroirradiation utilizing charged particle-induced protein foci. Mutagenesis25:289–297; 2010.

[45] Mao, Z. Y.; Seluanov, A.; Jiang, Y.; Gorbunova, V. TRF2 is required for repair ofnontelomeric DNA double-strand breaks by homologous recombination. Proc.Natl. Acad. Sci. USA 104:13068–13073; 2007.

416