9
Telomerase-Independent Telomere Length Maintenance in the Absence of Alternative Lengthening of Telomeres–Associated Promyelocytic Leukemia Bodies Clare L. Fasching, Kylie Bower, and Roger R. Reddel Children’s Medical Research Institute, Westmead, Sydney, New South Wales, Australia Abstract Immortal tumor cells and cell lines employ a telomere main- tenance mechanism that allows them to escape the normal limits on proliferative potential. In the absence of telomerase, telomere length may be maintained by an alternative length- ening of telomeres (ALT) mechanism. All human ALT cell lines described thus far have nuclear domains of unknown function, termed ALT-associated promyelocytic leukemia bodies (APB), containing promyelocytic leukemia protein, telomeric DNA and telomere binding proteins. Here we describe telomerase- negative human cells with telomeres that contain a substantial proportion of nontelomeric DNA sequences (like telomerase- null Saccharomyces cerevisiae survivor type I cells) and that are maintained in the absence of APBs. In other respects, they resemble typical ALT cell lines: the telomeres are highly heterogeneous in length (ranging from very short to very long) and undergo rapid changes in length. In addition, these cells are capable of copying a targeted DNA tag from one telomere into other telomeres. These data show that APBs are not always essential for ALT-mediated telomere maintenance. (Cancer Res 2005; 65(7): 2722-9) Introduction The ends of linear eukaryotic chromosomes are protected by the presence of telomeric DNA (which in vertebrates consists of tandem repeats of the TTAGGG hexanucleotide) and the proteins that specifically bind to this DNA (1, 2). The proliferation of normal cells is accompanied by a progressive decrease in telomere length (3). This telomeric attrition eventually results in a permanent prolifer- ative arrest referred to as replicative senescence (4, 5), which is bypassed in all immortal cell lines and in the majority of tumors by activation of a telomere maintenance mechanism (6). There are at least two such mechanisms. The best known of these uses the ribonucleoprotein enzyme, telomerase (7), which lengthens telo- meres by reverse transcribing telomeric repeats from its intrinsic RNA primer moiety. Telomerase-independent telomere length maintenance is called alternative lengthening of telomeres (ALT; refs. 8, 9); in at least some cases, this involves synthesis of telomeric DNA by recombination-mediated DNA replication (10). There are some types of human cancer where a substantial subset of the tumors do not have evidence either of ALT or of telomerase activity (11, 12). Some tumors may not require any telomere maintenance mechanism because of specific features of their biology (13), but it will be very important to determine whether any of the apparently ALT-negative/telomerase-negative tumors use a currently unknown ALT mechanism. Although there are differences between the telomere biology of yeast and mammalian cells, the observation that there are two classes of Saccharomyces cerevisiae cells that survive mutations resulting in absence of telomerase activity (14–16), also raises the question whether there is more than one ALT mechanism in telomerase-negative human cells. Type I yeast telomerase-null survivors have undergone amplification of YV elements in their telomeres by a mechanism that is dependent on genes including RAD51 . In contrast, the telomeres of type II survivors contain telomeric sequence and use a mechanism that requires the RAD50 gene. All of the human ALT cell lines analyzed to date have characteristics in common. They lack significant levels of telomerase activity and have telomeres that are highly heteroge- neous in length, ranging from very long to very short (8). ALT telomeres can therefore be readily distinguished from telomeres of telomerase-positive cells by fluorescence in situ hybridization (FISH) using telomere-specific probes (17, 18), or by terminal restriction fragment (TRF) Southern analysis (8). This heterogeneity is generated by a combination of gradual telomere attrition and rapid lengthening or shortening events (19). ALT cells also have unique nuclear domains that contain telomeric DNA and the promyelocytic leukemia (PML) protein, together with the TRF1 and TRF2 telomere binding proteins (20). PML bodies with these contents have not been detected in mortal or telomerase-positive cells, and they are therefore referred to as ALT-associated PML bodies (APB; ref. 20). APBs have also been shown to contain proteins involved in recombination and repair, including BLM (21), BRCA1, hRAP1 (22), MRE11, NBS1, RAD50 (23, 24), RAD51, RAD52, replication protein A (RPA; ref. 20), RAD51D (25), WRN (26), ERCC1, XPF (27), hRAD1, hRAD9, hRAD17, and hHUS1 (28) consistent with the evidence that ALT is a recombination- mediated mechanism (10, 19). The function of APBs is unknown, but it has been suggested that they may be actively involved in the ALT mechanism (20, 24, 29). Here we describe for the first time variant ALT cells that lack APBs and have telomeres that contain nontelomeric as well as telomeric sequence. Their mechanism of telomere maintenance, however, had functional similarities to that of other ALT lines. The telomere lengths were highly heterogeneous and underwent rapid changes. A DNA tag that was targeted into the telomere of a chromosome, which was transferred into these cells by microcell- mediated chromosome transfer (MMCT), was copied into the telomeres of other chromosomes, as had previously been shown for other ALT cells (10). Whereas the data do not provide definitive evidence that these variant ALT cells use a novel ALT mechanism, they do indicate that APBs are not always essential for ALT. Requests for reprints: Roger Reddel, Children’s Medical Research Institute, 214 Hawkesbury Road, Westmead, New South Wales 2145, Australia. Phone: 61-2-9687- 2800; Fax: 61-2-9687-2120; E-mail: [email protected]. I2005 American Association for Cancer Research. Cancer Res 2005; 65: (7). April 1, 2005 2722 www.aacrjournals.org Research Article Research. on August 1, 2020. © 2005 American Association for Cancer cancerres.aacrjournals.org Downloaded from

Telomerase-Independent Telomere Length Maintenance in the ...€¦ · Telomerase-Independent Telomere Length Maintenance in the Absence of Alternative Lengthening of Telomeres–Associated

  • Upload
    others

  • View
    5

  • Download
    0

Embed Size (px)

Citation preview

Page 1: Telomerase-Independent Telomere Length Maintenance in the ...€¦ · Telomerase-Independent Telomere Length Maintenance in the Absence of Alternative Lengthening of Telomeres–Associated

Telomerase-Independent Telomere Length Maintenance in the

Absence of Alternative Lengthening of Telomeres–Associated

Promyelocytic Leukemia Bodies

Clare L. Fasching, Kylie Bower, and Roger R. Reddel

Children’s Medical Research Institute, Westmead, Sydney, New South Wales, Australia

Abstract

Immortal tumor cells and cell lines employ a telomere main-tenance mechanism that allows them to escape the normallimits on proliferative potential. In the absence of telomerase,telomere length may be maintained by an alternative length-ening of telomeres (ALT) mechanism. All human ALT cell linesdescribed thus far have nuclear domains of unknown function,termed ALT-associated promyelocytic leukemia bodies (APB),containing promyelocytic leukemia protein, telomeric DNAand telomere binding proteins. Here we describe telomerase-negative human cells with telomeres that contain a substantialproportion of nontelomeric DNA sequences (like telomerase-null Saccharomyces cerevisiae survivor type I cells) and that aremaintained in the absence of APBs. In other respects, theyresemble typical ALT cell lines: the telomeres are highlyheterogeneous in length (ranging from very short to very long)and undergo rapid changes in length. In addition, these cellsare capable of copying a targeted DNA tag from one telomereinto other telomeres. These data show that APBs are not alwaysessential for ALT-mediated telomere maintenance. (Cancer Res2005; 65(7): 2722-9)

Introduction

The ends of linear eukaryotic chromosomes are protected by thepresence of telomeric DNA (which in vertebrates consists of tandemrepeats of the TTAGGG hexanucleotide) and the proteins thatspecifically bind to this DNA (1, 2). The proliferation of normal cellsis accompanied by a progressive decrease in telomere length (3).This telomeric attrition eventually results in a permanent prolifer-ative arrest referred to as replicative senescence (4, 5), which isbypassed in all immortal cell lines and in the majority of tumors byactivation of a telomere maintenance mechanism (6). There are atleast two such mechanisms. The best known of these uses theribonucleoprotein enzyme, telomerase (7), which lengthens telo-meres by reverse transcribing telomeric repeats from its intrinsicRNA primer moiety. Telomerase-independent telomere lengthmaintenance is called alternative lengthening of telomeres (ALT;refs. 8, 9); in at least some cases, this involves synthesis of telomericDNA by recombination-mediated DNA replication (10). There aresome types of human cancer where a substantial subset of thetumors do not have evidence either of ALT or of telomerase activity(11, 12). Some tumors may not require any telomere maintenancemechanism because of specific features of their biology (13), but it

will be very important to determine whether any of the apparentlyALT-negative/telomerase-negative tumors use a currently unknownALT mechanism.Although there are differences between the telomere biology of

yeast and mammalian cells, the observation that there are twoclasses of Saccharomyces cerevisiae cells that survive mutationsresulting in absence of telomerase activity (14–16), also raises thequestion whether there is more than one ALT mechanism intelomerase-negative human cells. Type I yeast telomerase-nullsurvivors have undergone amplification of YV elements in theirtelomeres by a mechanism that is dependent on genes includingRAD51 . In contrast, the telomeres of type II survivors containtelomeric sequence and use a mechanism that requires the RAD50gene.

All of the human ALT cell lines analyzed to date havecharacteristics in common. They lack significant levels oftelomerase activity and have telomeres that are highly heteroge-neous in length, ranging from very long to very short (8). ALTtelomeres can therefore be readily distinguished from telomeresof telomerase-positive cells by fluorescence in situ hybridization(FISH) using telomere-specific probes (17, 18), or by terminalrestriction fragment (TRF) Southern analysis (8). This heterogeneityis generated by a combination of gradual telomere attrition andrapid lengthening or shortening events (19). ALT cells also haveunique nuclear domains that contain telomeric DNA and thepromyelocytic leukemia (PML) protein, together with the TRF1and TRF2 telomere binding proteins (20). PML bodies with thesecontents have not been detected in mortal or telomerase-positivecells, and they are therefore referred to as ALT-associated PMLbodies (APB; ref. 20). APBs have also been shown to containproteins involved in recombination and repair, including BLM(21), BRCA1, hRAP1 (22), MRE11, NBS1, RAD50 (23, 24), RAD51,RAD52, replication protein A (RPA; ref. 20), RAD51D (25), WRN(26), ERCC1, XPF (27), hRAD1, hRAD9, hRAD17, and hHUS1 (28)consistent with the evidence that ALT is a recombination-mediated mechanism (10, 19). The function of APBs is unknown,but it has been suggested that they may be actively involved inthe ALT mechanism (20, 24, 29).

Here we describe for the first time variant ALT cells that lackAPBs and have telomeres that contain nontelomeric as well astelomeric sequence. Their mechanism of telomere maintenance,however, had functional similarities to that of other ALT lines. Thetelomere lengths were highly heterogeneous and underwent rapidchanges. A DNA tag that was targeted into the telomere of achromosome, which was transferred into these cells by microcell-mediated chromosome transfer (MMCT), was copied into thetelomeres of other chromosomes, as had previously been shown forother ALT cells (10). Whereas the data do not provide definitiveevidence that these variant ALT cells use a novel ALT mechanism,they do indicate that APBs are not always essential for ALT.

Requests for reprints: Roger Reddel, Children’s Medical Research Institute, 214Hawkesbury Road, Westmead, New South Wales 2145, Australia. Phone: 61-2-9687-2800; Fax: 61-2-9687-2120; E-mail: [email protected].

I2005 American Association for Cancer Research.

Cancer Res 2005; 65: (7). April 1, 2005 2722 www.aacrjournals.org

Research Article

Research. on August 1, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

Page 2: Telomerase-Independent Telomere Length Maintenance in the ...€¦ · Telomerase-Independent Telomere Length Maintenance in the Absence of Alternative Lengthening of Telomeres–Associated

Materials and Methods

Cell lines and microcell-mediated chromosome transfer. The SV40-

immortalized fibroblast cell lines, AG11395, GM847, and GM639 were

purchased from Coriell Cell Repositories (Camden, NJ), and W-v (30) was

provided by Dr. Lily Huschtscha (Children’s Medical Research Institute,

Westmead, Sydney, New South Wales, Australia). AG11395 and W-v were

derived from individuals with Werner syndrome. All four cell lines were

established either by SV40 infection, or by transfection with plasmid DNA

that contained the SV40 origin of replication and SV40 early region genes.

The AG11395 cells were received at passage 89 with an unspecified

population doubling level. We passaged these cells for an additional 180

population doublings to confirm that they are immortalized, but all other

analyses were done within 40 population doublings of the population doub-

ling level at which they were obtained. CMC3c2 is a hybrid cell line gen-

erated by transferring a human chromosome with a DNA tag targeted into

one of its telomeres into the mouse A9 cell line by MMCT (10). All cell lines

were maintained in DMEM containing 10% fetal bovine serum; for CMC3c2

cells, the medium was supplemented with 300 Ag/mL G418. MMCT was

done as previously described (10, 31). Colonies were isolated and analyzed

by neo-specific FISH to confirm that they were microcell hybrid cell lines.

Antibodies and indirect immunofluorescence. The primary antibodies

and the dilutions used were rabbit anti-MRE11 Ab-1, 1:300 (Oncogene Re-

search Products, San Diego, CA); mouse anti-RAD50 cl13B3, 1:300 (GeneTex,

San Antonio, TX); rabbit anti-NBS1 Ab-1, 1:300 (CalBiochem, San Diego, CA);

rabbit anti-RAD51 Ab-1, 1:300 (Oncogene Research Products); rabbit anti-

RAD52 AB3221, 1:300 (Chemicon, Temecula, CA); mouse anti-RPA Ab-3, 1:300

(Oncogene Research Products); rabbit anti-PML AB1370, 1:500 (Chemicon);

rabbit anti-SP100, 1:300 (Chemicon); mouse anti-SV40 large T antigen, 1:500;

and mouse anti-TRF2, 1:200 (Upstate Biotechnology, Waltham, MA). The

secondary antibodies and the dilutions used were goat anti-rabbit Alexa 488,

1:500; goat anti-mouse Alexa 594, 1:500 and goat anti-rabbit Alexa 594, 1:500

(Molecular Probes, Eugene, OR). To detect RPA, PML, and SP100, the cells

were washed thrice with 1� PBS, then fixed and incubated with the

appropriate primary antibody for 60 minutes at 37jC. The cells were then

washed thrice in 1� PBS/0.01%Tween 20 and incubatedwith the appropriate

secondary antibody for 30minutes at 37jC, followed by three washes with 1�PBS/0.01% Tween 20. To detect MRE11, RAD50, NBS1, RAD51, RAD52, or

SV40 large T antigen, we used an extraction procedure (23).

Telomerase and telomere length assays. Telomerase activity was as-

sayed by the telomere repeat amplification protocol (TRAP; ref. 32), using 2 Agof CHAPS cellular lysate and a myogenin internal control primer (33). The

products were separated on a 10% acrylamide gel, stained with SYBR Green

(Molecular Probes) and visualized on a STORM 860 imager (Molecular

Dynamics, Sunnyvale, CA). TRF lengths were determined by pulsed field

gel electrophoresis of genomic DNA digested with restriction enzymes

and hybridization of dried gels to telomere-specific oligo probe (TTAGGG)3essentially as described (8, 9).

Isolation of low molecular weight DNA and in-gel hybridization.

Low molecular weight DNA was isolated as described (34), separated using

1.0% agarose/TAE [0.4 mol/L Tris-acetate and 0.001 mol/L EDTA (pH 8.0)]

gel electrophoresis and dried using a vacuum gel drier. The dried gel was

denatured and neutralized, and hybridized overnight with oligonucleotide

probes for telomere sequence, (TTAGGG)3, or the SV40 origin of replication

(gggcggagttaggggcgg). Probes were 5’-end labeled using T4 polynucleo-

tide kinase (New England Biolabs, Beverly, MA). The gel was washed and

signal detected using an Amersham phosphor-screen and STORM 860

(Molecular Dynamics).

Fluorescence in situ hybridization. For metaphase FISH slides wereprepared as described (10), and for fiber FISH cell suspensions were dropped

across the top of slides pretreatedwith 2% silane. These slides were immersed

in lysis buffer [0.5% SDS, 50 mmol/L EDTA, 200 mmol/L Tris (pH 7.4)] for

10 minutes, then overlaid with an equal volume of ethanol. Slides weretransferred to 70% ethanol for 30 minutes and air-dried. DNA probes were

produced using either plasmid DNA (pBR322-SV40) or PCR amplification of a

neoR gene sequence ( forward primer gctatgactgggcacaacag and reverse

primer ccaccatgatattcggcaag; template: pSXneo plasmid; ref. 10) then

labeled with bio-16-dUTP using the Biotin-Nick Translation Mix (Roche,Nutley, NJ). Signals were detected as described (10).

Visualization of telomeric DNA with a peptide nucleic acid probe.After indirect immunofluorescence or FISH with DNA probes, slides were

cross-linked in 4% formaldehyde, rinsed in 1� PBS, and dehydrated in 70%,90%, and 100% ethanol. Denaturation, hybridization, and detection with

(CCCTAA)3-Cy3 Tel-PNA probe were as described (18).

Figure 1. AG11395 is an ALT cell line containing both SV40 and TTAGGGsequences within its telomeres. A, AG11395 does not have detectable TRAPactivity. B, AG11395 has short telomeres when digested with Hin fI and Rsa I(TRF lanes ). The telomeres are heterogeneous ranging from very short to verylong when digested with enzymes that do not restrict the SV40 sequences (SV40lanes ). C, Tel-PNA signals reveal the heterogeneous nature of the AG11395telomeres (left ). A similar pattern is seen when the telomeres are visualized withFISH using an SV40 specific probe (middle ). Merge of Tel-Cy3 and SV40-FITCreveals an intricate pattern of signals (right ). D, visualizing the signal patternmore closely using fiber FISH revealed differing patterns of SV40 FISH andTel-PNA. Some contained more Tel-PNA signal (i), more SV40 FISH signal (ii ),adjacent signals (iii ), or alternating signals (iv ).

Variant Alternative Lengthening of Telomeres

www.aacrjournals.org 2723 Cancer Res 2005; 65: (7). April 1, 2005

Research. on August 1, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

Page 3: Telomerase-Independent Telomere Length Maintenance in the ...€¦ · Telomerase-Independent Telomere Length Maintenance in the Absence of Alternative Lengthening of Telomeres–Associated

Telomere length quantitation. SV40 or telomeric DNA was visualized onmetaphase spreads as described above. The signals at each end of a single

unpaired distinctive marker chromosome were quantitated essentially as

described (18). Image bitmap pixel values ranged from 0 (empty scale =

black) to 4,095 ( full scale = white) following a linear function of measuredintensity with increasing exposure time. Typically, maximum intensity values

of the short (p) and the long (q) arm of the marker chromosome were

recorded for up to 24 randomly selected metaphase spreads. The maximal

values (<4,095) were corrected for average intensity values of backgroundfluorescence.

Results

Lack of detectable telomerase activity in AG11395 cells.Although we confirmed by extensive passaging that the AG11395cell line is immortal, telomerase activity was not detected in thesecells by the TRAP assay (Fig. 1A). It is therefore likely that they usean ALT mechanism for telomere maintenance. To confirm this, wenext evaluated whether the telomere lengths were maintained inAG11395 cells.Terminal restriction fragment analyses of AG11395 cells. We

determined telomere lengths by TRF Southern analysis using thestandard restriction enzymes, RsaI and HinfI that do not digesttelomeric sequence. The telomeres seemed short and relativelyhomogeneous in length, resembling those of a telomerase-positivecell line (Fig. 1B). Because investigations described below confirmedthe finding of others that the telomeres of these cells contained SV40

sequences1 and the SV40 genome contains 10 HinfI and 12 RsaIrestriction sites, we repeated the TRF Southern analysis usingenzymes MnlI, MscI, SacI, and XbaI that do not restrict SV40 DNA.This revealed that the telomeres have the characteristic heteroge-neous length distribution, ranging from very short to very long, ofALTcells (Fig. 1B). The results suggested that AG11395 is an ALTcellline with telomeres that contain SV40 sequences.Dispersal of SV40 sequences within the telomeres. We

visualized the telomeres in metaphase spreads using a (CCCTAA)3-Cy3 conjugated peptide nucleic acid (Tel-PNA) probe. Consistentwith the TRF results, the Tel-PNA signal was highly heterogeneous inintensity, ranging from absent to very strong, as is characteristic ofALT cells (Fig. 1C). FISH analysis using an SV40-specific probe alsoproduced signals at most chromosome ends, with heterogeneity ofsignal intensity resembling that of the Tel-PNA probe (Fig. 1C).Although the SV40-FISH and Tel-PNA signals generally coincid-

ed, there was some variation at individual chromosome ends. Wedid FISH on chromatin fibers to resolve the pattern of SV40 andTel-PNA signals. In all cases, SV40 and Tel-PNA signals were bothfound on the same chromatin fibers. Some fibers contained agreater amount of Tel-PNA signal interspersed with a few SV40signals (Fig. 1D , i), and others vice versa (Fig. 1D , ii). Some fibershad adjacent blocks of Tel-PNA and SV40-FISH signals (Fig. 1D , iii),

1 B. Johnson et al., personal communication.

Figure 2. Both SV40 and telomericsequences participate in ALT activity. Theq/p arm telomere length ratio wasdetermined by quantitation oftelomere-PNA FISH signals on a uniquemarker chromosome (A) that was found inall metaphase spreads of AG11395 cells. Aratio of 1 (center line ), indicates equalsignal intensity on both arms; >1, moreintense q arm signal; and <1, more intensep arm signal. B, ratio of q/p arm signalshowed great variability when assayed withthe Tel-PNA probe. C, SV40 FISH showeda similar variability. D, determining the p/qarm length ratios by hybridizing telomereand SV40 probes to the same telomeresshowed that the ratios variedindependently.

Cancer Research

Cancer Res 2005; 65: (7). April 1, 2005 2724 www.aacrjournals.org

Research. on August 1, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

Page 4: Telomerase-Independent Telomere Length Maintenance in the ...€¦ · Telomerase-Independent Telomere Length Maintenance in the Absence of Alternative Lengthening of Telomeres–Associated

whereas some had alternating SV40 and telomeric signals (Fig. 1D, iv).Thus, the ratio of SV40 and (TTAGGG)n sequences varied fromtelomere to telomere.AG11395 shows alternative lengthening of telomeres activity

with both telomeric and SV40 DNA. The lengths of individualtelomeres within ALT cells change rapidly with time. This wasshown in a previous study by examining the ratio of telomere signalon the q and p arms of a single chromosome: ALT cells showed ahigh level of variability in the p/q arm ratio whereas the level ofvariability in telomerase-positive cells was very low (18). To examineAG11395 cells for variation in telomere length, a unique markerchromosome was chosen (Fig. 2A) and 23 individual metaphaseswere evaluated with the Tel-PNA probe (Fig. 2B and D) and 24 withan SV40 specific probe (Fig. 2C and D). The q/p arm telomere lengthratios ranged from 0.4 to 8.99 (a 22.5-fold variation) for the Tel-PNA(Fig. 2B and D) and 0.25 to 18.7 for the SV40 probe (a 74.8-foldvariation; Fig. 2C and D), in contrast to the <2-fold ratio variationfound in telomerase-positive cells (18). Both the telomeric DNA andSV40 DNA showed a range of ratios similar to that of other ALT celllines, when evaluated independently (18). To confirm that the rangeof Tel-PNA ratios reflected ALT activity and not heterogeneoussubpopulations that had accumulated within the cell line, wegenerated a subclone of AG11395 and found that the q/p armratios of the same marker chromosome ranged from 0.28 to 4.47(16-fold variation; data not shown). Interestingly, when both SV40and Tel-PNA ratios were evaluated on the same marker chromo-some in 12 metaphase spreads, the ratios for the Tel-PNA did notalways correlate with the SV40 values (Fig. 2D). These data areconsistent with both the SV40 and (TTAGGG)n sequencesparticipating in recombination-mediated length changes. Thus,the telomere length dynamics of these cells resemble those of ALTcell lines examined previously.Recombination-mediated replication of AG11395 telomeres.

Recombination-mediated replication of ALT telomeres was detectedpreviously by targeting a DNA tag containing the neoR gene into ALTtelomeres and demonstrating that the tag was copied into othertelomeres while also remaining at its original location (10). To de-termine whether the AG11395 cells also use a recombination-mediated replication mechanism, we transferred a chromosomecontaining a neoR gene–tagged telomere into AG11395 cells fromthe CMC3c2 donor cell line by MMCT as described previously (10).The majority (96-100%) of the cells in three clones analyzedcontained the donor neoR-tagged chromosome from CMC3c2(Fig. 3A and C). Many of the cells also contained a variety ofadditional neo-tagged telomeres at late population doublings (Fig.3B and C), demonstrating that the tag had been copied fromtelomere to telomere, as would be expected if these cells lengthentheir telomeres by recombination-mediated replication.Lack of alternative lengthening of telomeres–associated

promyelocytic leukemia bodies in AG11395 cells. One of thephenotypic markers of ALT cells is the presence of APBs, which arecharacterized by colocalization of PML to telomeric DNA and/ortelomere binding proteins. These have been detected in every ALTcell line examined to date (reviewed in ref. 35). We examinedAG11395 cells for the presence of APBs by indirect immunofluo-rescence with antibodies specific for the PML protein and thetelomere binding protein, TRF2. We also analyzed the cells forcolocalization of PML and SP100 to telomeric DNA using the Tel-PNA probe. PML (and SP100; data not shown) did not colocalizewith either TRF2 or telomeric DNA (Fig. 4A). Thus, AG11395 differfrom other ALT cells in that they do not contain APBs.

AG11395 contains distinct nuclear aggregations. Althoughthese cells do not contain APBs, we observed telomere aggrega-tions using Tel-PNA. We analyzed these aggregates with SV40 FISHand either TRF2 immunofluorescence or telomere-specific FISH toconfirm the presence of both SV40 and telomeric sequences. Weobserved f100% colocalization of the Tel-PNA probe with bothTRF2 and the SV40 sequences (Fig. 4B). The SV40/telomeric DNAaggregates were present in 5% to 10% of the cells in an asyn-chronous population (data not shown).To further characterize these SV40/telomeric DNA aggregates,

we did indirect immunofluorescence using antibodies against SV40large T antigen and proteins that have previously been shown to beinvolved in telomere biology such as MRE11, NBS1, RAD50, RAD51,RAD52, and RPA. The SV40/telomeric DNA aggregates colocalizedwith all of these proteins (Fig. 5).AG11395 contains extrachromosomal telomere and SV40

sequences. ALTcell lines contain a greater amount of lowmolecularweight extrachromosomal telomeric DNA when compared withtelomerase-positive cell lines (36, 37).2 To compare the extrachro-mosomal DNA in AG11395 to other ALTand telomerase-positive celllines, we analyzed the low molecular weight DNA extracted as Hirtsupernatants from AG11395, the ALTcell lines, GM847 and W-v, andthe telomerase-positive cell line GM639 using SV40- and telomere-specific probes. There was a significant amount of low molecularweight extrachromosomal telomeric DNA present in AG11395 cells,which had a larger molecular weight than the extrachromosomal

2 T.R. Yeager and R.R. Reddel, unpublished data.

Figure 3. AG11395 maintains its telomeres by recombination-mediated DNAreplication. A telomere-tagged chromosome was transferred into AG11395 andthe neoR gene tag in three individual clones (c1-3) was monitored by FISH overa number of population doublings (PD ). A, at early PD the transferredchromosome can be readily identified (cyan arrowhead ). B, At later PD theoriginal transferred chromosome can still be identified (cyan arrowhead), butother telomeres now contain tags (yellow arrowheads ). C , Percentage of cellscontaining additional tagged telomeres at early and late population doublings.Total number of additional tagged telomeres refers to chromosome ends that aretagged at least once within the cell population.

Variant Alternative Lengthening of Telomeres

www.aacrjournals.org 2725 Cancer Res 2005; 65: (7). April 1, 2005

Research. on August 1, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

Page 5: Telomerase-Independent Telomere Length Maintenance in the ...€¦ · Telomerase-Independent Telomere Length Maintenance in the Absence of Alternative Lengthening of Telomeres–Associated

telomeric DNA in the ALTcell lines, W-v and GM487 (Fig. 6A). Therewas no telomeric DNA recovered in the telomerase-positive cell line,GM639 (Fig. 6A). To investigate whether the SV40 origin ofreplication was present in the low molecular weight DNA, wehybridized it with an SV40 origin of replication–specific oligo. Thisdetected a significant amount of extrachromosomal DNA inAG11395 cells (Fig. 6B), with molecular weights corresponding tothose of the extrachromosomal DNA that hybridized to the(TTAGGG)3 probe. We also detected the SV40 origin of replicationin the telomeres of AG11395 cells (data not shown). No extrachro-mosomal SV40 DNA was recovered from the SV40-immortalized celllines W-v, GM847, or GM639 (Fig. 6B).

Discussion

We have investigated an immortalized cell line that maintains itstelomeres in the absence of telomerase and has features thatdistinguish it from other ALTcell lines. These cells contain extensiveamounts of SV40 sequence in their telomeres, together with(TTAGGG)n DNA. Furthermore, these cells differ from all othertelomerase-negative immortalized cell lines described to date in that

they lack APBs. In other ways, however, the telomere maintenancemechanism used by these cells resembles that of other ALTcells. TRFanalyses showed that these cells have the highly heterogeneoustelomere length distribution that is a hallmark of ALT cells. Further-more, the telomere lengths of a single chromosome varied widelywithin cells of both an uncloned population and a recently derivedsubclone, demonstrating that the telomere length heterogeneity isgenerated rapidly, as has been observed for other ALT cell lines (18).They lengthen their telomeres by recombination-mediated DNAreplication like other ALT cells (10), as evidenced by copying of anintratelomeric neoR gene tag from one telomere to others.FISH analyses showed a complex pattern of interspersion of

SV40 and telomeric sequences within the AG11395 telomeres, withSV40 and telomeric sequences alternating in some telomeres. DNAsequence analysis has shown that the telomeres of these cellscontain tandem arrays of SV40 and telomeric sequences (38). Thepresence of SV40 sequences within the telomeres of AG11395 cellsresembles the situation in yeast type I telomerase-null survivorsthat contain amplified YV subtelomeric elements interspersed withtelomeric sequence (14–16, 39), as well as immortalized telomerase-null mouse embryonic stem cells that contain tandem arrays of

Figure 4. AG11395 cells do not containAPBs. SV40-specific aggregates colocalized withboth TRF2 and telomeric DNA, but not withPML. A, monolayers of AG11395 cells werecostained with PML antibody and either theTRF2 antibody or the Tel-PNA probe.Twenty-five to 50 cells were evaluated for eachcostaining combination and nocolocalization was seen. B, cells were costainedwith the Tel-PNA probe and either TRF2antibody or an SV40 FISH probe.

Cancer Research

Cancer Res 2005; 65: (7). April 1, 2005 2726 www.aacrjournals.org

Research. on August 1, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

Page 6: Telomerase-Independent Telomere Length Maintenance in the ...€¦ · Telomerase-Independent Telomere Length Maintenance in the Absence of Alternative Lengthening of Telomeres–Associated

nontelomeric and telomeric sequences at most chromosome ends(40). The yeast type I and II mechanisms are dependent on differentgenes, so more information regarding the molecular genetics ofALT and the variant ALT cells described here will be required beforeit is clear whether the variant ALT cells are the human equivalentof yeast type I survivors.Regarding the mechanism whereby SV40 DNA became inserted

into the telomeres of these cells, it is possible that a telomere wasthe original site of integration of SV40 DNA. Alternatively, SV40sequences initially may have integrated elsewhere in the genome andsubsequently inserted into a telomere. In either case, once SV40sequences became inserted into the telomere they could be copied toother telomeres when the ALT mechanism became activated atimmortalization.The differing contributions of SV40 and (TTAGGG)n sequences to

some telomeres may be explained by the nature of the telomere

maintenance mechanism. We have previously suggested that theremay be a variety of templates for recombination-mediated DNAreplication of telomeres in ALT cells: in some cases it may be othertelomeres, in others a telomeremay be able to use itself as a templatefor DNA synthesis via t-loop formation, or it may be able to use linearor circular extrachromosomal telomeric DNA as a copy template(41, 42). Studies in the yeast Kluyveromyces lactis showed thatexogenous plasmid sequences were added to telomeres by rollingcircle gene conversion and then ‘‘spread’’ to the other telomeres byrecombination (43). SV40 DNA containing the origin of replicationhas the capacity to become amplified in situ , to form circularextrachromosomal DNA, and to reintegrate into new genomiclocations (44–46) which could lead to complex patterns within thetelomeres. Thus, there may have been integration of SV40 followedby an excision event, which included telomeric DNA andmaintenance of this SV40/telomere unit as an episome. TheAG11395 cells were found to contain abundant extrachromosomalDNA ranging in size from f2.8 to 8.5 kb that hybridized to(TTAGGG)n and SV40 probes. Rolling circle replication of episomalDNA would result in repeating units of SV40 and telomeric DNA. Incontrast, recombination with telomeres could result in addition ofvariable proportions of SV40 and (TTAGGG)n sequences dependingon the composition of the template telomere and the point ofrecombination within the telomere. Also, given the presence of theSV40 origin of replication, it is possible that SV40 sequences mayhave been amplified in situ within some telomeres.The presence of substantial amounts of SV40 DNA in the telomeres

of these cells raises the question of how telomere capping function isaccomplished. It is interesting to note that t-loop formation has beenshown in in vitro experiments to be tolerant of nontelomeric sequenceat the terminus (47). Moreover, it has been shown that Schizosacchar-omyces pombe telomere chromatin structure and function may bestably maintained in the absence of telomeric repeats (48).Despite the abundant presence of the extrachromosomal

telomeric DNA, the AG11395 cells do not contain APBs. Thesenuclear domains contain PML and other constituents of PML bodiestogether with telomeric DNA, telomere binding proteins, and other

Figure 6. AG11395 extrachromosomal DNA includes both SV40 origin ofreplication and telomeric sequences. Hirt lysates were hybridized with probesspecific for telomeric DNA or the SV40 origin of replication. A, telomere-specificprobe hybridizes with low molecular weight DNA in all three ALT cell lines,GM847, W-v and AG11395, but not in the telomerase-positive GM639. The bulkof extra-chromosomal DNA found in AG11395 has a higher molecular weightthan that in GM847 and W-v. B, SV40 origin of replication–specific probehybridizes only with the extrachromosomal DNA in AG11395 and the pattern issimilar to that of the telomeric probe.

Figure 5. AG11395 cells contain aggregates of SV40/telomeric DNA andspecific proteins. Monolayers of AG11395 were costained with antibodiesagainst MRE11, RAD50, NBS1, RAD51, RAD52, RPA, and SV40 large T antigen(left) and the PNA probe, Tel-FITC (middle ). Merged images (right ) ofcolocalization of the proteins with the SV40/telomere aggregates. Fifteen to 20cells were analyzed with each antibody.

Variant Alternative Lengthening of Telomeres

www.aacrjournals.org 2727 Cancer Res 2005; 65: (7). April 1, 2005

Research. on August 1, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

Page 7: Telomerase-Independent Telomere Length Maintenance in the ...€¦ · Telomerase-Independent Telomere Length Maintenance in the Absence of Alternative Lengthening of Telomeres–Associated

proteins involved in DNA recombination and replication, includingWRN, and have been found in 2% to 10% of asynchronously dividingcells within every ALT cell line population that has been examinedpreviously (20, 35). The absence of APBs does not seem to be due tothe absence of wild-type WRN protein in AG11395 cells, becauseAPBs are present in another Werner Syndrome cell line, W-v.Although it has been suggested that APBs participate in the ALTmechanism (20) in a cell cycle–dependent manner (24, 29), and APBsdisappear from cells after the ALT mechanism is repressed bysomatic cell hybridization (49), the relationship of APBs to the ALTmechanism is still unclear. Most or all of the proteins required forALTmay be assembled within APBs, which could therefore functionas domains in which the ALT process is facilitated.Although APBs per se are clearly not required for telomere

maintenance in the AG11395 cell line, there are nuclear aggregatespresent in these cells containing SV40 large T antigen and many ofthe same components as APBs including MRE11, NBS1, RAD50,RAD51, RAD52, RPA, and TRF2 (but not PML or SP100). Large Tantigen may accumulate within these aggregates by binding to theSV40 origin of replication within the telomeric DNA in AG11395cells. SV40 large T antigen interacts with many proteins and hasbeen shown to interact with the nuclear matrix (50). It is possible,

therefore, that large T antigen may be responsible for the unusuallocalization of the DNA/protein complexes in AG11395 cells thatwould otherwise aggregate in PML bodies and thus form APBs,by both binding to the SV40 origin of replication sequences inthe extrachromosomal mixed SV40/telomeric DNA and deter-mining its localization within the nucleus by its other bindingactivities. Regardless of the mechanism of this localization, theobservation that there are nuclear domains containing concen-trations of telomeric DNA and proteins involved in DNAprocessing in these cells, leaves open the possibility that inALT cells such aggregates are an important aspect of thetelomere maintenance mechanism.

Acknowledgments

Received 8/9/2004; revised 11/8/2004; accepted 1/14/2005.Grant support: Judith Hyam Memorial Trust Fund for Cancer Research,

Carcinogenesis Fellowship of the Cancer Council New South Wales, and NationalHealth and Medical Research Council of Australia postgraduate scholarship andproject grant.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Sheila Stewart for helpful discussions and Brad Johnson, Len Guarente,and Robert Marciniak for providing unpublished data.

References1. Moyzis RK, Buckingham JM, Cram LS, et al. A highlyconserved repetitive DNA sequence, (TTAGGG)n ,present at the telomeres of human chromosomes.Proc Natl Acad Sci U S A 1988;85:6622–6.

2. Blackburn EH. Structure and function of telomeres.Nature 1991;350:569–73.

3. Hastie ND, Dempster M, Dunlop MG, Thompson AM,Green DK, Allshire RC. Telomere reduction in humancolorectal carcinoma and with ageing. Nature1990;346:866–8.

4. Olovnikov AM. A theory of marginotomy. Theincomplete copying of template margin in enzymicsynthesis of polynucleotides and biological significanceof the phenomenon. J Theor Biol 1973;41:181–90.

5. Levy MZ, Allsopp RC, Futcher AB, Greider CW, HarleyCB. Telomere end-replication problem and cell aging.J Mol Biol 1992;225:951–60.

6. Cech TR. Beginning to understand the end of thechromosome. Cell 2004;116:273–9.

7. Greider CW, Blackburn EH. Identification of a specifictelomere terminal transferase activity in Tetrahymenaextracts. Cell 1985;43:405–13.

8. Bryan TM, Englezou A, Gupta J, Bacchetti S, Reddel RR.Telomere elongation in immortal human cells withoutdetectable telomerase activity. EMBO J 1995;14:4240–8.

9. Bryan TM, Reddel RR. Telomere dynamics andtelomerase activity in in vitro immortalised humancells. Eur J Cancer 1997;33:767–73.

10. Dunham MA, Neumann AA, Fasching CL, Reddel RR.Telomere maintenance by recombination in humancells. Nat Genet 2000;26:447–50.

11. Gupta J, Han L-P, Wang P, Gallie BL, Bacchetti S.Development of retinoblastoma in the absence oftelomerase activity. J Natl Cancer Inst 1996;88:1152–7.

12. Hakin-Smith V, Jellinek DA, Levy D, et al. Alternativelengthening of telomeres and survival in patients withglioblastoma multiforme. Lancet 2003;361:836–8.

13. Reddel RR. The role of senescence and immortaliza-tion in carcinogenesis. Carcinogenesis 2000;21:477–84.

14. Lundblad V, Blackburn EH. An alternative pathwayfor yeast telomere maintenance rescues est1� senes-cence. Cell 1993;73:347–60.

15. Teng S-C, Zakian VA. Telomere-telomere recombina-tion is an efficient bypass pathway for telomere main-enance in Saccharomyces cerevisiae . Mol Cell Biol 1999;19:8083–93.

16. Le S, Moore JK, Haber JE, Greider CW. RAD50 andRAD51 define two pathways that collaborate tomaintaintelomeres in the absence of telomerase. Genetics 1999;152:143–52.

17. Lansdorp PM, Poon S, Chavez E, et al. Telomeres inthe hematopoietic system. Ciba Found Symp 1997;211:209–18.

18. Perrem K, Colgin LM, Neumann AA, Yeager TR,Reddel RR. Coexistence of alternative lengthening oftelomeres and telomerase in hTERT-transfected GM847cells. Mol Cell Biol 2001;21:3862–75.

19. Murnane JP, Sabatier L, Marder BA, Morgan WF.Telomere dynamics in an immortal human cell line.EMBO J 1994;13:4953–62.

20. Yeager TR, Neumann AA, Englezou A, HuschtschaLI, Noble JR, Reddel RR. Telomerase-negative immor-talized human cells contain a novel type of promye-locytic leukemia (PML) body. Cancer Res 1999;59:4175–9.

21. Yankiwski V, Marciniak RA, Guarente L, Neff NF.Nuclear structure in normal and Bloom syndrome cells.Proc Natl Acad Sci U S A 2000;97:5214–9.

22. Wu G, Jiang X, Lee WH, Chen PL. Assembly offunctional ALT-associated promyelocytic leukemiabodies requires Nijmegen breakage syndrome 1.Cancer Res 2003;63:2589–95.

23. Zhu X-D, Kuster B, Mann M, Petrini JHJ, de Lange T.Cell-cycle-regulated association of RAD50/MRE11/NBS1 with TRF2 and human telomeres. Nat Genet2000;25:347–52.

24. Wu G, Lee W-H, Chen PL. NBS1 and TRF1 colocalizeat promyelocytic leukemia bodies bodies during late S/G2 phases in immortalized telomerase-negative cells:implication of NBS1 in alternative lengthening oftelomeres. J Biol Chem 2000;275:30618–22.

25. Tarsounas M, Munoz P, Claas A, et al. Telomeremaintenance requires the RAD51D recombination/repair protein. Cell 2004;117:337–47.

26. Johnson FB, Marciniak RA, McVey M, Stewart SA,Hahn WC, Guarente L. The Saccharomyces cerevisiaeWRN homolog Sgs1p participates in telomere mainte-nance in cells lacking telomerase. EMBO J 2001;20:905–13.

27. Zhu XD, Niedernhofer L, Kuster B, Mann M,Hoeijmakers JH, de Lange T. ERCC1/XPF removesthe 3V overhang from uncapped telomeres andrepresses formation of telomeric DNA-containingdouble minute chromosomes. Mol Cell 2003;12:1489–98.

28. Nabetani A, Yokoyama O, Ishikawa F. Localiza-tion of hRad9, hHus1, hRad1 and hRad17, andcaffeine-sensitive DNA replication at ALT (alterna-tive lengthening of telomeres)-associated promyelo-cytic leukemia body. J Biol Chem 2004;279:25849–57.

29. Grobelny JV, Godwin AK, Broccoli D. ALT-associatedPML bodies are present in viable cells and are enrichedin cells in the G2/M phase of the cell cycle. J Cell Sci2000;113:4577–85.

30. Huschtscha LI, Thompson KVA, Holliday R. Thesusceptibility of Werner’s syndrome and other humanskin fibroblasts to SV40-induced transformation andimmortalization. Proc R Soc Lond B Biol Sci 1986;229:1–12.

31. Saxon PJ, Srivatsan ES, Leipzig GV, Sameshima JH,Stanbridge EJ. Selective transfer of individual humanchromosomes to recipient cells. Mol Cell Biol1985;5:140–6.

32. Kim NW, Piatyszek MA, Prowse KR, et al. Specificassociation of human telomerase activity with immortalcells and cancer. Science 1994;266:2011–5.

33. Wright WE, Shay JW, Piatyszek MA. Modifications ofa telomeric repeat amplification protocol (TRAP) resultin increased reliability, linearity and sensitivity. NucleicAcids Res 1995;23:3794–5.

34. Cole SL, Tevethia MJ. Simian Virus 40 Large T antigenand two independent T-antigen segments sensitize cellsto apoptosis following genotoxic damage. J Virol2002;76:8420–32.

35. Henson JD, Neumann AA, Yeager TR, Reddel RR.Alternative lengthening of telomeres in mammaliancells. Oncogene 2002;21:598–610.

36. Ogino H, Nakabayashi K, Suzuki M, et al. Release oftelomeric DNA from chromosomes in immortal humancells lacking telomerase activity. Biochem Biophys ResCommun 1998;248:223–7.

37. Tokutake Y, Matsumoto T, Watanabe T, et al. Extra-chromosome telomere repeat DNA in telomerase-negative immortalized cell lines. Biochem Biophys ResCommun 1998;247:765–72.

38. Marciniak RA, Cavazos D, Montellano R, Johnson FB.A novel form of ALT telomere maintenance in humancells. Cancer Res. Under revision 2004.

39. Chen Q, Ijpma A, Greider CW. Two survivorpathways that allow growth in the absence oftelomerase are generated by distinct telomere recom-bination events. Mol Cell Biol 2001;21:1819–27.

Cancer Research

Cancer Res 2005; 65: (7). April 1, 2005 2728 www.aacrjournals.org

Research. on August 1, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

Page 8: Telomerase-Independent Telomere Length Maintenance in the ...€¦ · Telomerase-Independent Telomere Length Maintenance in the Absence of Alternative Lengthening of Telomeres–Associated

40. Niida H, Shinkai Y, Hande MP, et al. Telomeremaintenance in telomerase-deficient mouse embryonicstem cells: characterization of an amplified telomericDNA. Mol Cell Biol 2000;20:4115–27.

41. Neumann AA, Reddel RR. Opinion: Telomere main-tenance and cancer-look, no telomerase. Nat RevCancer 2002;2:879–84.

42. Fasching CL, Reddel RR. Telomere maintenancein human cell lines and tumors without telomer-ase. In: Krupp G, Parwaresch R, editors. Telo-merases, telomeres and cancer. Georgetown (TX):Landes Bioscience/Eurekah.com and New York(NY): Kluwer Academic/Plenum Publishers; 2002.p. 359–73.

43. Natarajan S, Groff-Vindman C, McEachern MJ.Factors influencing the recombinational expansionand spread of telomeric tandem arrays in Kluyveromyceslactis . Eukaryot Cell 2003;2:1115–27.

44. Molecular biology of tumor viruses, part 2. DNAtumor viruses. 2nd ed. Cold Spring Harbor: Cold SpringHarbor Laboratory Press; 1981.

45. Zouzias D, Jha KK, Mulder C, Basilico C, Ozer HL.Human fibroblasts transformed by the early region ofSV40 DNA: analysis of ‘‘free’’ viral DNA sequences.Virology 1980;104:439–53.

46. Stary A, Sarasin A. Molecular analysis of DNAjunctions produced by illegitimate recombination inhuman cells. Nucleic Acids Res 1992;20:4269–74.

47. Stansel RM, de Lange T, Griffith JD. T-loop assemblyin vitro involves binding of TRF2 near the 3V telomericoverhang. EMBO J 2001;20:5532–40.

48. Sadaie M, Naito T, Ishikawa F. Stable inheritance oftelomere chromatin structure and function in theabsence of telomeric repeats.GenesDev 2003;17:2271–82.

49. Perrem K, Bryan TM, Englezou A, Hackl T, Moy EL,Reddel RR. Repression of an alternative mechanism forlengthening of telomeres in somatic cell hybrids.Oncogene 1999;18:3383–90.

50. Deppert W, Von der Weth A. Functional interactionof nuclear transport-defective simian virus 40 large Tantigen with chromatin and nuclear matrix. J Virol1990;64:838–46.

Variant Alternative Lengthening of Telomeres

www.aacrjournals.org 2729 Cancer Res 2005; 65: (7). April 1, 2005

Research. on August 1, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from

Page 9: Telomerase-Independent Telomere Length Maintenance in the ...€¦ · Telomerase-Independent Telomere Length Maintenance in the Absence of Alternative Lengthening of Telomeres–Associated

2005;65:2722-2729. Cancer Res   Clare L. Fasching, Kylie Bower and Roger R. Reddel  Associated Promyelocytic Leukemia Bodies

−the Absence of Alternative Lengthening of Telomeres Telomerase-Independent Telomere Length Maintenance in

  Updated version

  http://cancerres.aacrjournals.org/content/65/7/2722

Access the most recent version of this article at:

   

   

  Cited articles

  http://cancerres.aacrjournals.org/content/65/7/2722.full#ref-list-1

This article cites 46 articles, 20 of which you can access for free at:

  Citing articles

  http://cancerres.aacrjournals.org/content/65/7/2722.full#related-urls

This article has been cited by 17 HighWire-hosted articles. Access the articles at:

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  [email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. (CCC)Click on "Request Permissions" which will take you to the Copyright Clearance Center's

.http://cancerres.aacrjournals.org/content/65/7/2722To request permission to re-use all or part of this article, use this link

Research. on August 1, 2020. © 2005 American Association for Cancercancerres.aacrjournals.org Downloaded from