Published OnlineFirst May 29, 2014; DOI: 10.1158/0008-5472.

CAN-13-2038

Cancer
Tumor and Stem Cell Biology Research

miR-155 Drives Telomere Fragility in Human Breast Cancer

by Targeting TRF1
Roberto Dinami1,4,5, Cristiana Ercolani7, Eleonora Petti1,4,5, Silvano Piazza2, Yari Ciani2, Rosanna Sestito5,
Andrea Sacconi6, Francesca Biagioni6, Carlos le Sage9, Reuven Agami9, Roberta Benetti3,8,
Marcella Mottolese7, Claudio Schneider2,8, Giovanni Blandino6, and Stefan Schoeftner1,5

Abstract
Telomeres consist of DNA tandem repeats that recruit the multiprotein complex shelterin to build a chromatin
structure that protects chromosome ends. Although cancer formation is linked to alterations in telomere
homeostasis, there is little understanding of how shelterin function is limited in cancer cells. Using a small-scale
screening approach, we identified miR-155 as a key regulator in breast cancer cell expression of the shelterin
component TERF1 (TRF1). miR-155 targeted a conserved sequence motif in the 30 UTR of TRF1, resulting in its
translational repression. miR-155 was upregulated commonly in breast cancer specimens, as associated with
reduced TRF1 protein expression, metastasis-free survival, and relapse-free survival in estrogen receptor–positive
cases. Modulating miR-155 expression in cells altered TRF1 levels and TRF1 abundance at telomeres. Compromis-
ing TRF1 expression by elevating miR-155 increased telomere fragility and altered the structure of metaphase
chromosomes. In contrast, reducing miR-155 levels improved telomere function and genomic stability. These
results implied that miR-155 upregulation antagonizes telomere integrity in breast cancer cells, increasing genomic
instability linked to poor clinical outcome in estrogen receptor–positive disease. Our work argued that miRNA-
dependent regulation of shelterin function has a clinically significant impact on telomere function, suggesting the
existence of "telo-miRNAs" that have an impact on cancer and aging. Cancer Res; 74(15); 4145–56.
2014 AACR.

Introduction telomeric 30 overhang and, together with TRF1 and TRF2,

Vertebrate telomeres are composed of TTAGGG tandem ensures the protection of chromosome ends by repressing
repeats that recruit the shelterin complex, which protects DNA damage signaling by the ATR and ATM kinases (5–7).
chromosome ends, regulates telomere length, recombination, RAP1 binds to telomeres via its interaction with TRF2 (8). TRF1
and DNA damage checkpoints (1, 2). Shelterin is composed of has been shown to contribute to telomere length regulation
TRF1, TRF2, POT1, TPP1, TIN2, and RAP1 (2). TRF1 and its and suppresses DNA breakage at TTAGGG repeats under
paralog TRF2 bind to double-stranded TTAGGG repeats and replicative stress, a phenomenon described as telomere fra-
interact with POT1 via interaction with TIN2 and TPP1 (2–4). gility (5, 6, 9, 10). Consistent with this, loss of TRF1 results in
POT1 interacts with single-stranded TTAGGG repeats in the telomere replication errors and the activation of ATR signaling
in S-phase (5, 6). Recently, evidence is accumulating that links
altered shelterin function with human cancer. POT1 mutations
have been linked to telomeric and chromosomal abnormalities
Authors' Affiliations: 1Telomeres in Cancer and Aging Unit, 2Bioinformat-
ics and Functional Genomics Unit, and 3Epigenetics Unit, Laboratorio in chronic lymphocytic leukemia (11). Wnt/b-catenin signaling
Nazionale Consorzio Interuniversitario Biotecnologie (LNCIB); 4Diparti- was shown to drive TRF2 expression, improving telomere
mento di Scienze della Vita, SDBM School of Molecular Biomedicine

degli Studi di Trieste, Trieste; 5Telomeres in Cancer
function in cancer cells (12). In addition, TRF2 protein levels
(SDBM), Universita
and Aging Unit, 6Translational Oncogenomics Unit, and 7Department of were shown to be controlled by the p53-inducible E3 ubiquitin
Pathology, Regina Elena National Cancer Institute, Rome; 8Dipartimento di ligase Siah1, linking shelterin function with the p53 tumor
Scienze Mediche e Biologiche, Universita degli Studi di Udine, Udine, Italy;
and 9Division of Gene Regulation, The Netherlands Cancer Institute,
suppressor pathway (13). Importantly, mice deficient for TRF1
Amsterdam, the Netherlands display telomere fragility and increased cancer formation in
Note: Supplementary data for this article are available at Cancer Research the absence of p53. This indicates that reduced TRF1 expres-
Online (http://cancerres.aacrjournals.org/). sion can enhance cancer formation by driving telomere fra-
Current address for F. Biagioni: European Institute of Oncology (IEO) Via
gility and genomic instability (6). Together, this suggests that
Adamello, 16, Milan 20139, Italy. shelterin function is intertwined with central pathways in
Corresponding Author: Stefan Schoeftner, Telomeres in Cancer and tumorigenesis and tumor suppression. miRNAs are 20–23
Aging Unit, Laboratorio Nazionale Consorzio Interuniversitario Biotecno- nucleotide, single-stranded RNA molecules that control gene
logie (LNCIB), Padriciano 99, Trieste 34149, Italy. Phone: 390-4037-56805; expression by reducing the translation or stability of target
Fax: 390-4039-8990; E-mail: [email protected]
mRNAs (14). Extensive studies demonstrated that miRNAs
doi: 10.1158/0008-5472.CAN-13-2038
control the expression of crucial tumor suppressors or onco-

2014 American Association for Cancer Research. genes and propagate essential features of cancer progression

www.aacrjournals.org 4145

Downloaded from cancerres.aacrjournals.org on August 8, 2019. © 2014 American Association for Cancer Research.
 Published OnlineFirst May 29, 2014; DOI: 10.1158/0008-5472.CAN-13-2038

Dinami et al.

(15). Importantly, miRNA expression signatures have been deoxycholate, 2.5 mmol/L sodium pyrophosphate, 1 mmol/L
established as efficient prognostic and predictive biomarkers, b-glycerophosphate, 1 mmol/L Na3VO4, 1 mg/mL leupeptin).
underlining the clinical relevance of miRNAs (15). Although Preparation of nuclear extracts: cells were resuspended in
telomere function is a central aspect in cancer and aging, Buffer 1 [20 mmol/L Hepes-KOH (pH 7.9), 10 mmol/L KCl,
miRNAs that modulate the expression of shelterin components 1.5 mmol/L MgCl2, 1 mmol/L EDTA, 1 mmol/L EGTA, 0.2%
are to date not known. Of interest, a recent study links miR-138 NP40] and incubated for 10 minutes on ice. Nuclei were
with telomerase (TERT) expression; however, the impact of pelleted and lysed in Buffer 2 [20 mmol/L Hepes-KOH (pH
miR-138 on telomere homeostasis is not known (16). 7.9), 350 mmol/L NaCl, 1.5 mmol/L MgCl2, 10 mmol/L KCl,
Here, we performed a small-scale screen to identify clinically 10% glycerol, 1 mmol/L dithiothreitol). All samples contained
relevant miRNAs that modulate telomere function in human complete protease inhibitor (Roche) and were sonicated fol-
breast cancer by targeting the expression of TRF1. We found lowed by centrifugation. Supernatants were recovered and
that the oncomiR miR-155 targets a conserved sequence motif used for Western blotting according to standard procedures.
in the 30 UTR of TRF1, resulting in reduced protein expression. Used primary antibodies are listed in Supplementary Material
miR-155 is efficiently upregulated in human breast cancer and Methods.
specimen, correlates with reduced TRF1 protein levels, and is
linked with poor prognosis in estrogen receptor (ER)–positive Immunofluorescence
breast cancer. On the mechanistic level, we demonstrate that Cells were fixed in 4% paraformaldehyde (PFA), followed by
miR-155-dependent reduction of TRF1 expression results in treatment with citrate buffer [0.1% (w/v), 0.05% Triton X-100]
telomere elongation, increased telomere damage, increased for 10 minutes at room temperature. Cells were blocked for 30
telomere fragility, and chromosome instability. In contrast, minutes in 3% BSA (1 PBS) and incubated with mouse anti-
reducing miR-155 expression improves telomere function and TRF1 [Santa Cruz (N-19) sc-6165-R], mouse anti-TRF2
genomic stability. Together, our data identify miR-155 as clin- [Millipore (4A794) 05-521], or rabbit anti-POT1 (Epitomics
ically relevant, novel telomere regulator that drives telomere 5334-1) antibodies in 3% BSA (1 PBS), 0.1% Tween-20 for 1
fragility and genomic instability by repressing TRF1 expression. hour at room temperature. Cells were washed in 0.3% BSA (1
Our work introduces miRNAs as modulators of shelterin func- PBS), 0.1% Tween-20 and incubated with secondary antibodies.
tion, anticipating the existence of additional "telo-miRNAs" that At least 20 DAPI (Vector Laboratories) stained nuclei were
link telomere function with fundamental processes in telomere- used for spot IOD analysis (TFL-TELO Software). A Student
related diseases such as cancer and organismal aging. t test was used to calculate statistical significance.

Microscopy
Materials and Methods Cells subjected to immunofluorescence, telomere DNA
Cell lines and culture FISH, or immunofluorescence combined with telomere DNA
Cell lines used were obtained from ATCC and have not been FISH were analyzed using a Zeiss Axiovert 200 M microscope,
cultured for longer than 6 months. MCF-7 (Michigan Cancer equipped with a Zeiss AxioCam MRm digital camera. Images
Foundation-7, breast adenocarcinoma), SK-BR-3 (breast ade- were captured using AxioVision Rel. 4.8 imaging software.
nocarcinoma derived), and HCT-116 (colorectal carcinoma)
cells were cultured in RPMI-1640, 10% heat-inactivated FBS Telomere DNA FISH
(Gibco). MDA-MB-468 (breast adenocarcinoma), HeLa (cervix Preparation of metaphases and telomere DNA FISH was
adenocarcinoma), U-2 OS (osteosarcoma), and H1299 (carci- performed as previously described (17). At least 20 metaphases
noma; non–small cell lung cancer) cells were cultured in were analyzed for metaphase chromosome aberrations (TFL-
DMEM, 10% heat-inactivated FBS. Stable MCF-7 and SK-BR- TELO software). For quantitative telomere DNA FISH analysis,
3 cells overexpressing miR-155 were selected with blasticidin (5 at least 20 interphase nuclei were analyzed using spot IOD
mg/mL). Treatments: Aphidicolin (Sigma) for 20 hours, 0.4 analysis (TFL-TELO) software. The Student t test was used to
mmol/L; MG-132 (Millipore) for 8 hours, 12.5 mmol/L. calculate statistical significance.

Luciferase reporter screening Immunofluorescence combined with telomere DNA FISH

HeLa cells were cotransfected with 30 UTR luciferase reporter For immunofluorescence, cells were fixed in 4% PFA, fol-
plasmids (18 ng) and miRNA expression vectors (80 ng) using lowed by treatment with 0.1% Triton X-100 in 1 PBS for 10
Lipofectamine 2000 (Life Technologies). Seventy-two hours minutes at room temperature. Cells were blocked for 30
posttransfection, Renilla/Firefly luciferase reporter activity was minutes in 3%BSA (1 PBS) and incubated with a mouse
assayed using a Dual Luciferase Reporter Assay System (Pro- anti-phospho-gH2A.X (S139) antibody (clone JBW301, Milli-
mega) and a GloMax 96 Microplate Luminometer (Promega). A pore, 06-570) in 3% BSA, 0.1% Tween-20 at room temperature
Student t test was used to calculate statistical significance. for 1 hour. After incubation with secondary antibodies, cells
were fixed in 1% PFA and subjected to standard telomere
Protein extracts and Western blotting DNA FISH (18). Nuclei showing at least three telomere-gH2AX
Whole-cell lysates were prepared using a modified RIPA colocalization events were considered for the quantification
buffer (20 mmol/L Tris-HCl (pH 7.5) 350 mmol/L NaCl, 1 (minimum 40 nuclei). The Student t test was used to calculate
mmol/L Na2EDTA, 1 mmol/L EGTA, 1% NP-40, 1% sodium statistical significance.

4146 Cancer Res; 74(15) August 1, 2014 Cancer Research

Downloaded from cancerres.aacrjournals.org on August 8, 2019. © 2014 American Association for Cancer Research.
 Published OnlineFirst May 29, 2014; DOI: 10.1158/0008-5472.CAN-13-2038

Telomere Regulation by miR-155

Breast cancer specimen collection and levels (Fig. 1H). We conclude that miR-155 is a novel regulator
immunohistochemistry of TRF1 expression in human cells.
TRF1 expression was analyzed by immunohistochemistry
(IHC) on formalin-fixed paraffin-embedded tissues using miR-155 targets a conserved sequence motif in the 30 UTR
the monoclonal antibody (TRF-78, #ab10579, Abcam). Two of TRF1
micron–thick sections were stained with a streptavidin- The predicted targeting region for miR-155 is located at
enhanced immunoperoxidase technique (Supersensitive Mul- position 93-115 in the human TRF1 30 UTR and is conserved in
tilink, Menarini) using a pH 6 citrate buffer antigen retrieval chimpanzee, cow, and rabbit (Fig. 2A). Mouse and rat 30 UTRs
protocol. TRF1 was considered as positive when at least 20% of do not show miR-155 target site conservation, suggesting
neoplastic cells showed nuclear immunoreactivity. Staining that miR-155 could contribute to differences in telomere reg-
was classified as negative, score 0; low, score 1þ; medium, ulation in rodents. To directly demonstrate interaction between
score 2þ, and high, score 3þ. miR-155 and its predicted target site, we generated a human
TRF1 30 UTR reporter construct carrying a deletion of the miR-
Bioinformatics on clinical data from breast cancer 155 target site. Synthetic mimic-miR-155 siRNAs efficiently
patients reduced wild-type TRF1 30 UTR luciferase reporter activity;
Raw data were retrieved from the gene expression omni- however, deleting the miR-155 target site rendered the luciferase
bus (GEO) public gene expression database (GSE7390, reporter construct resistant to targeting by miR-155 (DmiR-155-
GSE3494, GSE1456, GSE2034, GSE2603, GSE6532, GSE4922, TRF1-30 UTR; Fig. 2B). In addition, transfection of antago-miR-
GSE12093, GSE5327, GSE11121, and Chin dataset). To vali- 155 molecules that target endogenous miR-155 resulted in
date the correlation of TRF1 expression and the mir-155 increased TRF1 30 UTR luciferase reporter activity (Fig. 2C). In
target gene expression signature and breast cancer clinical line with luciferase reporter assays, we found that ectopic
data, a Mantel–Haenszel test was applied to the meta-data- introduction of antago-miR-155 increases TRF1 expression
set. Kaplan–Meier survival curve of distant metastasis-free levels in H1299 human non–small lung cancer cells and U-2
survival and relapse-free survival (DMFS_mixed) was obtain- OS osteosarcoma cells (Fig. 2D and E). Mimic miR-155 siRNAs
ed using the GOBO tool. decreased TRF1 protein levels without impacting on TRF1
mRNA levels, indicating that miR-155 induces translational
repression of TRF1 (Fig. 2D and E and Supplementary Fig.
Results S2A). Alterations of miR-155 levels do not affect protein levels
A candidate approach to identify miRNAs with targeting of shelterin components TRF2 and POT1 (Fig. 2D and E). Non-
specificity for TRF1 telomere bound TRF1 is degraded via the ubiquitin–protea-
TRF1 is a telomere repeat binding protein that acts as some pathway (21). To exclude that miR-155 reduces TRF1
negative regulator of telomere length, contributes to mitotic levels by promoting proteasome-dependent TRF1 degradation,
stability, and protects from replication-associated DNA dam- we cotransfected SK-BR-3 cells with control or mimic-miR-155
age (5, 6, 9, 10, 19). Using a panel of human cancer cell lines, we siRNAs and flag-tagged TRF1 lacking the corresponding 30 UTR.
found evidence for divergent posttranscriptional regulation of Treatment of cells with the proteasome inhibitor MG-132
gene expression of TRF1 and TRF2: in contrast to TRF2, TRF1 augmented both endogenous TRF1 and flag-tagged TRF1 in
protein levels significantly diverge from mRNA levels (Fig. 1A– cells transfected with control miRNAs (Supplementary Fig. S2B
D). This indicates that TRF1 is subjected to efficient posttran- and S2C). Ectopic introduction of mimic-miR-155 reduced
scriptional gene regulation. To identify miRNAs that control endogenous TRF1 but did not interfere with the stabilization
TRF1 expression on the posttranscriptional level, we per- of flag-tagged TRF1, indicating that miR-155 does not modify
formed computational target prediction analysis to identify proteasome-dependent degradation of TRF1. Moreover, target
miRNAs with in silico target specificity for the 30 UTR of TRF1 prediction analysis (Targetscan) did not reveal a link between
(Supplementary Tables S1–S3). To validate targeting, luciferase miR-155 and regulators that stabilize TRF1 on the posttrans-
reporter vectors were generated by fusing a Renilla luciferase lational level (Supplementary Table S4A).
cassette to the 30 UTR of human TRF1. HeLa cells were cotrans- Altogether, our data demonstrate that miR-155 directly
fected with the TRF1-30 UTR luciferase reporter and 23 candi- controls TRF1 expression levels by specifically targeting a
date miRNA expression vectors, obtained from a miRNA partially conserved sequence motif in the 30 UTR of TRF1,
expression vector library (Fig. 1E; ref. 20). Luminometry mea- leading to translational repression.
surements 3 days posttransfection revealed that miR-155,
miR-125b1, miR-296, and miR-330 significantly reduced Clinical relevance of TRF1 targeting by miR-155 in breast
TRF1-30 UTR luciferase reporter activity (Fig. 1F). miR-125b1, cancer
miR-296, and miR-330 also reduced luciferase activity of an miR-155 is a reported "oncomiR" that targets multiple cancer
unrelated TRF2-30 UTR reporter construct, indicating nonspe- relevant genes to promote neoplastic disease, including breast
cific miRNA–luciferase reporter interaction (Supplementary cancer (22, 23). We consequently were interested in demon-
Fig. S1A–S1C). In contrast, ectopic expression of miR-155 does strating that miR-155-dependent regulation of TRF1 is clinically
not impact on TRF2-30 UTR or TERT-30 UTR reporter activity relevant in human breast cancer. Consistent with previous
(Fig. 1G). Finally, transient transfection of HeLa cells with the reports, we found a significant upregulation of miR-155 expres-
miR-155 expression vector resulted in reduced TRF1 protein sion in not only luminal but also HER2þ and triple-negative

www.aacrjournals.org Cancer Res; 74(15) August 1, 2014 4147

Downloaded from cancerres.aacrjournals.org on August 8, 2019. © 2014 American Association for Cancer Research.
 Published OnlineFirst May 29, 2014; DOI: 10.1158/0008-5472.CAN-13-2038

Dinami et al.

A B

HCT‐116 p53–/–
MDA-MB-468

HCT-116
SK-BR-3

U-2 OS
MCF-7

H1299

TRF1
TRF1 (L.E.)
Actin

C D
p53–/–
MDA-MB-468

HCT-116

HCT-116
SK-BR-3

U-2 OS
MCF-7

H1299

TRF2
TRF2 (L.E.)
Actin

E F
n = 3–6

firefly
AAAA
Cap
miR- renilla TRF1
vector 3′UTR
library
HeLa
Luciferase reporter activity

G P = 0.134 P = 0.133 H
175
n=3
(control set 100)

n=3
150 P = 0.0001
n=3
125
miR-155

n=3 n=3
Control

100
75
50
n=3
25 TRF1
0 Actin
Control

miR-155

Control

miR-155

Control

miR-155

TRF1- TRF2- TERT-

3′UTR 3′UTR 3′UTR

Figure 1. A luciferase reporter screen identified miR-155 to target the 30 UTR of TRF1. A, Western blotting for TRF1 (left) and quantification (right) in a panel of
human cancer cell lines. B, quantitative real-time PCR for TRF1. C, Western blotting for TRF2 (left) and quantification (right) in a panel of human cancer cell
lines. D, quantitative real-time PCR for TRF2, normalized to actin. E, strategy of luciferase reporter screen. F, luciferase reporter assay. Renilla: Firefly
luciferase ratios <100 indicate miRNA-dependent targeting of the TRF1 reporter. Mouse Pax9 30 UTR and miR206 were used as positive controls. G, luciferase
reporter assays involving miR-155 and TRF1, TRF2, or TERT 30 UTR reporter constructs. H, TRF1 Western blotting of HeLa cells transiently transfected with
the miR-155 expression vector. Arrowheads, specific band for TRF1. L.E., long exposure; n, number of independent experiments. A Student t test was used to
calculate statistical significance ( , P < 0.05;   , P < 0.01;   , P < 0.001).

4148 Cancer Res; 74(15) August 1, 2014 Cancer Research

Downloaded from cancerres.aacrjournals.org on August 8, 2019. © 2014 American Association for Cancer Research.
 Published OnlineFirst May 29, 2014; DOI: 10.1158/0008-5472.CAN-13-2038

Telomere Regulation by miR-155

P = 0.0001
B 125
P = 0.532 C P > 0.001
n=3
Relative luciferase reporter

n=3
activity (control set “100”)

n=3 P > 0.001 n = 8

Luciferase reporter activity

100
n=8

(control set “100”)

75

50 n=3 n=8

25

0
Control

miR-155

Control

miR-155

Control

miR-155
Antago-
miR-155
wt-TRF1 ΔmiR-155
3′UTR TRF1-3′UTR

D P = 0.0002
E P = 0.009
miR-155

miR-155

miR-155

miR-155
Antago-

Antago-
Control

Control

P = 0.0001
n=3 n=3
P = 0.0004 n=3
TRF1 n=3 TRF1
Actin Actin
n=3 n=3

TRF2 TRF2
Actin POT1

Control

miR-155

Antago-miR-155
Control
miR-155

Antago-miR-155

Actin
POT1 U-2 OS
Actin

H1299

Figure 2. miR-155 targets a partially conserved sequence motif in the TRF1 30 UTR. A, sequence conservation of miR-155–TRF1-30 UTR target site interaction in
human, chimpanzee, cow, and rabbit. B, TRF1 luciferase reporter assays using a reporter construct carrying a deletion of the miR-155 target site. C, luciferase
reporter assays using mimic-miR-155 or antago-miR-155 siRNAs. D and E, Western blotting for TRF1, TRF2, and POT1 expression in miR-155 in gain-
and loss-of-function experiments using H1299 (D) and U-2 OS (E) cells; right, quantification of TRF1 levels. n, number of independent experiments. A Student
t test was used to calculate statistical significance. P values are shown.

breast cancer specimen when compared with their respective to standard clinical parameters such as lymph node status,
peritumoral tissue (Fig. 3A; Supplementary Table S6; refs. 24–26). tumor size, or age (Fig. 3C). This underlines that low TRF1
To obtain information on a possible clinical relevance of miR- expression predicts poor clinical outcome in ERþ breast cancer.
155-dependent regulation of TRF1 expression, we used breast Importantly, we found that poor prognosis of patients with
cancer gene expression datasets obtained from a total of 1,881 breast cancer characterized by low TRF1 expression was reca-
patients (27). Kaplan–Meier survival analysis revealed that low pitulated by panel of 20 experimentally validated miR-155 target
TRF1 mRNA expression is linked to reduced distant metastasis- genes (Supplementary Table S5; Supplementary Fig. S3; ref. 23).
free survival and relapse-free survival (P ¼ 0.0055) of patients This suggests that TRF1 is part of a miR-155 signature that
with ER-positive (ERþ) breast cancer (Fig. 3B). Furthermore, predicts poor survival in ERþ breast cancer. Of note, TRF1 does
mulitvariate analysis revealed that TRF1 acts as independent not affect distant metastasis-free survival and relapse-free sur-
predictor of clinical outcome in ERþ breast cancer with respect vival of other breast cancer subtypes (data not shown).

www.aacrjournals.org Cancer Res; 74(15) August 1, 2014 4149

Downloaded from cancerres.aacrjournals.org on August 8, 2019. © 2014 American Association for Cancer Research.
 Published OnlineFirst May 29, 2014; DOI: 10.1158/0008-5472.CAN-13-2038

Dinami et al.

A Expression (log base 2 scale)

10
B
9

5 P = 0.00552

4
tumor peri- tumor peri- tumor peri- TRF1 low: n = 593
tumoral tumoral tumoral TRF1 high: n = 606

luminal triple negative HER2+

n = 42 n=4 n=6
P = 8.14E–05 P = 0.002 P = 0.02 Time (y)

C D E
TRF1 expression (IHC)
TRF1 high

Multivariate analysis for n = 876 High (3)

28%
39% Medium-low (2)
Low (1) –negative (0)
P= 33%
n = 37
P=
miR-155 medium
TRF1 medium

P= F
P=
miR-155–
TRF1 low P= miR-155+/++
n = 37
P = 0.034
miR-155 high
TRF1 low

G
miR-155

miR-155
miR-155

miR-155

Antago-
miR-155

miR-155

Antago-

Control

siTRF1
Antago-

Control

siTRF1
Control

siTRF1

TRF1 TRF1 TRF1

Actin Actin Actin

MBA-MD-468 SK-BR-3 MCF-7

Figure 3. Clinical relevance of miR-155 and TRF1 in human breast cancer. A, box plots showing miR-155 upregulation in breast cancer compared with
peritumoral tissues, as determined by miRNA expression array analysis described in ref. 28. P values were calculated using the Fisher method. B, Kaplan–
þ
Meier survival curve of time considering both distant metastasis-free survival and relapse-free survival (DMSF_mixed) of patients with ER breast
cancer classified according to the expression of TRF1 (Supplementary Material and Methods). Red line, high TRF1 mRNA expression; gray line, low expression
of TRF1. n, number of patients. C, multivariate analysis showing that low TRF1 expression behaves as independent predictor of poor clinical outcome.
In the panel for each clinical variable and TRF1 expression, HRs and corresponding P values calculated by Cox regression analysis are shown. D,
Immunohistochemistry for TRF1 on luminal breast cancer samples (high TRF1, top panel) used for miRNA expression studies in A. Scale bar, 30 mm. E,
proportion of TRF1 staining intensities in samples analyzed in D. F, luminal cancer samples with significantly increased miR-155 expression show reduced
2
TRF1 staining in IHC performed in D. A Pearson c test was used to calculate statistical significance. G, TRF1 expression is miR-155–dosage sensitive in
SK-BR-3, MCF-7, and MDA-MB-468 breast cancer cell lines, as determined by Western blotting. n, number of patients.

4150 Cancer Res; 74(15) August 1, 2014 Cancer Research

Downloaded from cancerres.aacrjournals.org on August 8, 2019. © 2014 American Association for Cancer Research.
 Published OnlineFirst May 29, 2014; DOI: 10.1158/0008-5472.CAN-13-2038

Telomere Regulation by miR-155

To obtain biologic evidence that confirms in silico data, we reduced by the stable overexpression of miR-155 in long-term
performed IHC for TRF1 on four normal breast tissue samples experiments using SK-BR-3 cells (Supplementary Fig. S5A–
and 37 ERþ luminal cancer samples previously assayed for S5E). This indicates that miR-155 promotes telomere elonga-
miR-155 expression (Supplementary Table S6; Fig. 3A; ref. 28). tion by reducing abundance of TRF1 at telomeres. Loss of TRF1
All control and 28% of breast cancer samples were classified as is associated with impaired telomere function, leading to the
TRF1 high. Seventy-two percent of cancer samples were clas- elicitation of a DNA damage response at telomeres (5, 6). In line
sified medium-low or low-negative for TRF1 (33% and 39%, with this, we found by Western blotting that transient trans-
respectively), indicating that downregulation of TRF1 is a fection of luminal MCF-7 and SK-BR-3 breast cancer cell lines
frequent event in luminal breast cancer (Fig. 3D and E). with mimic-miR-155 siRNAs result in elevated levels of gH2AX,
Importantly, 83% of specimens with reduced TRF1 protein indicative for the activation of a DNA damage response (Fig. 4C
expression demonstrate miR-155 upregulation (P ¼ 0.034), and D, left). DNA damage response was associated with the
supporting a role for miR-155 in controlling TRF1 expression activation of wild-type p53 in MCF-7 cells, as indicated by
in human breast cancer (Fig. 3F). Finally, ectopic introduction increased levels of phospho-p53 (Ser15) (Fig. 4C and D, left).
of miR-155 in human luminal SK-BR-3 and MCF-7 and basal In SK-BR-3 cells, which express mutant p53, introduction of
MDA-MB-468 breast cancer cells that express similar levels of miR-155 was not able to significantly increase phosphorylation
endogenous miR-155 resulted in reduced TRF1 expression (Fig. of p53. This is presumably due to the high basal levels of
3G and Supplementary Fig. S4A). In contrast, introduction of phospho-p53 (Ser15) present in this cell line (Fig. 4D). In line
antago-miR-155 increases TRF1 expression in all breast cancer with results from TRF1 knockout mice, we found that miR-155-
lines tested (Fig. 3G). Our data indicate that TRF1 expression is dependent reduction of TRF1 expression leads to the phos-
miR-155 dosage sensitive in breast cancer cells. Reverse cor- phorylation of ATM and ATR and the downstream DNA
relation of miR-155 and TRF1 expression and poor prognosis of damage checkpoint kinases Chk1 and Chk2 (Fig. 4C and D,
patients with low expression of TRF1 or a miR-155 target gene right). To test whether telomere dysfunction contributes to the
signature underline the clinical relevance of miR-155-depen- increased DNA damage load in the context of ectopic miR-155,
dent regulation of TRF1 in ERþ breast cancer. we performed gH2AX immunofluorescence combined with
telomere DNA FISH. Consistent with a proposed role in
miR-155 alters telomere structure and function controlling telomere protection, we found that ectopic intro-
We next wished to link miR-155 to molecular pathways that duction of miR-155 increased the abundance of gH2AX local-
modulate TRF1-related aspects of telomere function. To test ized at telomeres in MCF-7 and SK-BR-3 cells (Fig. 4E and F).
whether alteration of miR-155 expression levels does not This effect was recapitulated by RNAi-mediated depletion of
only impact on global TRF1 expression levels but also alter TRF1 and in long-term experiments using MCF-7 cells stably
TRF1 abundance at telomeres, we transiently transfected overexpressing miR-155 (Supplementary Fig. S6A–S6D).
SK-BR-3 luminal breast cancer cells with mimic-miR-155, Importantly, introduction of antago-miR-155 reduces basal
antago-miR-155 siRNAs, or TRF1-specific siRNAs and per- levels of telomere dysfunction in MCF-7 and SK-BR-3, indicat-
formed quantitative immunofluorescence analysis using ing that chromosome end protection is miR-155 dosage sen-
anti-TRF1 antibodies. We found that ectopically introduced sitive (Fig. 4E and F). Together, these data indicate that miR-
miR-155 significantly reduced fluorescence signal intensity, 155 affects telomere length regulation and telomere capping
indicative for a reduced abundance of TRF1 at telomeres (Fig. by modulating the expression of TRF1.
4A). This result is also reflected by the appearance of an
increased proportion of telomeres with low TRF1 loading; this miR-155 drives telomere fragility and chromosome
effect was recapitulated by transfecting a TRF1-specific siRNA instability in human breast cancer
pool (Fig. 4A). In contrast, treatment with antago-miR-155 Studies using genetic mouse models revealed that TRF1
increased TRF1 abundance at telomeres (Fig. 4A). Same results suppresses telomere fragility under replicative stress condi-
were obtained when modulating miR-155 levels in H1299 non– tions (5, 6). Fragile sites are enriched in a sequence context that
small lung carcinoma cells, that express elevated levels of TRF1 challenges DNA replication. Fragility can lead to chromosome
and miR-155 compared with breast cancer cell lines used in breakage and was demonstrated to drive genomic instability at
this study (Fig. 1A and Supplementary Fig. S4A and S4B). In line sites that are subjected to amplifications or deletions, suggest-
with Western blotting results we found that alteration of miR- ing an important link between chromosome fragility and
155 levels does not impact on the abundance of TRF2 or POT1 cancer formation (29). The potential of telomeric TTAGGG
at telomeres (Fig. 2D and E and Supplementary Fig. S4C and tandem repeats to generate guanine quadruplex (G4) DNA
S4D). Our results indicate that miR-155-dependent regulation structures and form the so-called T-Loop structure by invasion
of TRF1 expression is an efficient mechanism to control the of the 30 single-stranded telomeric DNA into the telomere
abundance of TRF1 at telomeres. TRF1 acts as negative reg- duplex imposes stress to DNA replication, leading to telomere
ulator of telomere length, presumably by restricting the access fragility (30, 31). Telomere fragility results in DNA breaks and
of telomerase to chromosome ends (9, 10). In line with this, we loss of telomere DNA, leading to multiple telomere signals at
found a significant increase in telomere length after 3 cycles of the end of metaphase chromosomes (Fig. 5A). In line with a role
transient transfection of telomerase-positive SK-BR-3 cells for TRF1 in suppressing telomere fragility, we found that
with mimic-miR-155 siRNAs or TRF1-specific siRNAs (Fig. transient transfection of mimic-miR-155 siRNAs increased
4B). This result was recapitulated when TRF1 expression was telomere fragility in SK-BR-3 cells (Fig. 5B). Targeting

www.aacrjournals.org Cancer Res; 74(15) August 1, 2014 4151

Downloaded from cancerres.aacrjournals.org on August 8, 2019. © 2014 American Association for Cancer Research.
 Published OnlineFirst May 29, 2014; DOI: 10.1158/0008-5472.CAN-13-2038

Dinami et al.

A B

Control miR-155 Antago-miR-155 siTRF1 Control miR-155 siTRF1

P < 0.0001
8,000 P < 0.0001
TRF1 fluorescence intensity (a.f.u.)

P < 0.0001 C: >1251 P < 0.0001 C: >1251

Telomere a.f.u intensity categories (%)

P = 0.004 B: 801-1251 670 663 B: 801-1251
A: 0-800 ± 88 ± 22 A: 0-800

Telomere fluorescence units (a.f.u.)

1,306 1,447 N=3 N=3

TRF1 a.f.u. intensity categories (%)

± 36.17 ± 33.15 n = 2,043 n = 2,213 100
N=3 N=3
100
n = 454 595
4,000 1,024 n = 494 ± 64
± 26.37
C 3,000 N=3 80
N=3 907.1
80 n = 2,143
n = 416 ± 24.00 C
N=3
60
n = 409 60
2,000
2,000 40
40
1,000 B
B 20 20

A A
0
Control
miR-155
Antago-
miR-155
siTRF1

Control
miR-155
siTRF1
miR-155
Antago-
miR-155
siTRF1
Control

Control

miR-155

siTRF1
C E γH2AX
miR-155

miR-155

miR-155

miR-155

Telomere Merge
Antago-

Antago-
Control

Control

P = 0.04
Control

TRF1 P-ATM (S1981) P = 0.04

TIF frequency (>3 TIFs; %)

40
Actin ATM N=3
P-Chk2 (T68) n = 46
γH2AX 30
Chk2
P-p53 (S15) P-ATR (S428)
miR-155

Actin 20 N=3
ATR
n = 44
MCF-7 P-Chk1 (S345) N=3
10 n = 43
Chk1
0
miR-155

Lamin A/C
Antago-

Control

miR-155
Antago-
miR-155
MCF-7
miR-155
miR-155

miR-155

miR-155
Antago-

Antago-
Control

Control

D F Telomere γH2AX Merge

P = 0.002
TRF1 P-ATM (S1981)
Control

P = 0.001
TIF frequency (>3 TIFs; %)

Actin ATM 80 N=3

n = 91
P-Chk2 (T68)
γH2AX Chk2 60 N=3
n = 81
P-p53 (S15) N=3
P-ATR (S428) n = 86
miR-155

Actin 40
ATR
SK-BR-3 P-Chk1 (S345)
20
Chk1
0
miR-155

Lamin A/C
Antago-

Control

miR-155
Antago-
miR-155

SK-BR-3

4152 Cancer Res; 74(15) August 1, 2014 Cancer Research

Downloaded from cancerres.aacrjournals.org on August 8, 2019. © 2014 American Association for Cancer Research.
 Published OnlineFirst May 29, 2014; DOI: 10.1158/0008-5472.CAN-13-2038

Telomere Regulation by miR-155

endogenous miR-155 by transfecting antago-miR-155 resulted sister chromatid fusions (P ¼ 0.01) and also protected from
in a significant suppression of telomere fragility. Increased telomere sister chromatid fusions driven by Aphidicolin (Fig.
telomere fragility was recapitulated by classic RNAi-mediated 5H). This indicates that telomere-related genomic instability is
depletion of TRF1 and by stable overexpression of a miR-155 miR-155 dosage sensitive.
mini gene in SK-BR-3 cells (Supplementary Fig. S7A and S7B). Altogether, we show that miR-155-dependent regulation of
Together, this indicates that telomere fragility is miR-155 TRF1 is an efficient mechanism to control telomere fragility
dosage sensitive. We next wished to exclude that alternative and genomic stability in human breast cancer cells. The
miR-155 targets drive TRF1-independent telomere fragility. consistent upregulation of miR-155 in breast cancer suggests
Cotransfection of SK-BR-3 cells with a mix of mimic-miR- that impaired telomere function is a central aspect of the
155 and TRF1-specific siRNAs did not augment telomere oncogenic function of miR-155 that promotes genomic insta-
fragility induced by mimic-miR-155 siRNAs, suggesting that bility in human breast cancer and contributes to reduced
miR-155 and TRF1 siRNA act on the same pathway controlling distant metastasis-free survival and relapse-free survival in
telomere fragility (Fig. 5C). Furthermore, we were able to ERþ breast cancer.
rescue increased telomere fragility induced by ectopic miR-
155 in SK-BR-3 cells when miR-155 was cotransfected with an
expression vector encoding Flag-tagged TRF1 lacking the Discussion
respective 30 UTR (Fig. 5D and E). Finally, miR-155 target Telomere dysfunction is a major type of chromosome
prediction analysis (Targetscan) did not reveal target sites in damage in cancer and interfering with regulators of telomere
the 30 UTR of reported suppressors of telomere fragility (Sup- function drives central features of cancer such as loss of
plementary Table S4B). Together, this demonstrates that miR- heterozygosity, chromosomal rearrangements, aneuploidy,
155 drives telomere fragility by specifically targeting TRF1. and the repression of DNA damage response checkpoints
Next, we aimed to demonstrate that antagonizing miR-155- (33, 34). The importance of telomere regulation in human
dependent regulation of TRF1 expression could mediate resis- cancer is underlined by the upregulation of telomerase
tance to telomere fragility induced by Aphidicolin, a selective activity in 90% of human cancers and frequent alteration of
inhibitor of DNA Polymerase a (5, 6, 32). As expected, telomere the expression levels of shelterin components (35–43).
fragility was reduced by antago-miR-155 and increased by Increased cancer formation upon loss of TRF1 in the context
Aphidicolin treatment (Fig. 5F; Supplementary Fig. S7C). of compromised tumor suppression suggests that alteration
Importantly, we found that introduction of antago-miR-155 of TRF1 expression drives genomic instability, a hallmark
was able to protect from Aphidicolin-induced telomere fragil- feature of human cancer (6, 11). Using a small-scale miRNA
ity, thus reducing telomere fragility to levels observed in screening approach, we show that miR-155 efficiently reg-
untreated, control miRNA-transfected cells (Fig. 5F). We con- ulates TRF1 expression by targeting a partially conserved
clude that miR-155 efficiently modulates telomere fragility by sequence motif in the TRF1 30 UTR. miR-155 is a classic onco-
controlling the expression of TRF1. We next were interested miRNA that is processed from a noncoding transcript
whether miR-155-driven reduction of TRF1 leads to increased encoded by the BIC locus and has been linked to lympho-
genomic instability by quantifying telomere sister chromatid magenesis (44, 45). However, more recent studies indicate a
fusions, a reported consequence of TRF1 loss-of-function (6). general role for miR-155 in human cancer, as demonstrated
We found that ectopic miR-155 increases the frequency of by a robust upregulation in breast, colon, and lung cancer
telomeric sister chromatid fusions in metaphase spreads that (46). Ectopic miR-155 was shown to drive the proliferation of
were protease treated before telomere-specific DNA FISH (Fig. breast cancer cell lines in vitro but also, when xenografted
5G; ref. 17). Augmented telomere sister chromatid fusion was into nude mice, to underline the role of miR-155 as onco-
rescued when miR-155 was cotransfected with Flag-tagged miRNA in breast cancer (47, 48). In addition to its role in
TRF1. This result indicates that miR-155 drives telomere sister promoting cell proliferation, a large panel of targets has been
chromatid fusions by targeting TRF1 and excludes miR-155 off- identified that link miR-155 to cancer relevant pathways,
target effects (Fig. 5D and G). Aphidicolin-treated SK-BR-3 cells including apoptosis, inflammation, or DNA mismatch repair
display significantly increased frequency of telomere sister (22, 23, 49). miRNA expression signatures have been estab-
chromatid fusions (Supplementary Fig. S7D). Remarkably, lished as efficient prognostic and predictive biomarkers,
transient transfection of antago-miR-155 reduced telomere underlining the clinical relevance of miRNAs (15).

Figure 4. miR-155 drives alterations in telomere function in breast cancer cell lines. A, quantitative immunofluorescence for TRF1 on SK-BR-3 cells
transfected with the indicated mimic-miRNAs or siRNAs; top, representative images. Bottom left, TRF1 fluorescence intensity analyzed for each telomere.
Bottom right, individual telomere signal intensities were categorized in three subgroups. B, telomere length measurements by quantitative telomere
DNA FISH after repeated transfection of SK-BR-3 cells with the indicated siRNAs; top, representative images. Bottom left, telomere fluorescence intensity
analyzed for each telomere. Bottom right, individual telomere signal intensities were categorized in three subgroups. P values and SDs are indicated. N,
number of independent experiments; n, total signals analyzed; at least 20 nuclei were analyzed. a.f.u, arbitrary fluorescence units. C and D, ectopic
introduction of miR-155 induces DNA damage signaling in SK-BR-3 and MCF-7 cells. Representative images are shown; actin/lamin A/C was used as loading
control. Arrowheads, specific bands. E and F, ectopic introduction of miR-155 in MCF-7 (E) and SK-BR-3 (F) cells induces DNA damage as demonstrated
by colocalization of gH2AX with telomeres in immunotelomere FISH experiments. N, number of independent experiments; n, number of analyzed nuclei; A–F, a
Student t test was used to calculate statistical significance; P values are shown.

www.aacrjournals.org Cancer Res; 74(15) August 1, 2014 4153

Downloaded from cancerres.aacrjournals.org on August 8, 2019. © 2014 American Association for Cancer Research.
 Published OnlineFirst May 29, 2014; DOI: 10.1158/0008-5472.CAN-13-2038

Dinami et al.

A B C P < 0.002
P = 0.03 60
n.s.
P = 0.005
50 n = 1,456

Fragile telomeres (%)

60 n = 1,396 Figure 5. miR-155 drives alterations
N = 24 N = 21

Fragile telomeres (%)

n = 1,874 n = 1,416 N = 21 in telomere function in breast
50 40 N = 20
N = 22 n = 1,453 cancer cell lines. A, exemplary
n = 1,598 N = 22
40 n = 1,654 N = 21 images of fragile telomeres,
30
30 characterized by multiple telomere
20 signals; for enlargement see
20
images 1 and 2. B, frequency of
10 10 telomere fragility in SK-BR-3 cells
0 transfected with mimic-miR-155 or
0 antago-miR-155 siRNAs. C,
Control

miR-155

Antago-miR-155
+ − + − con. miRNA
+ + − − con. siRNA telomere fragility frequency after
− − + + siTRF1 combined transfection of SK-BR-3
− + − + miR-155 cells with mimic-miR-155 and
TRF1 siRNAs (top). Bottom, TRF1
TRF1
levels in experimental samples.
Actin Actin was used as a loading
control. D, anti-TRF1 Western
D E P = 0.3 F blotting of SK-BR-3 cells
P = 0.34
P = 0.002 P = 0.02 cotransfected with Flag-tagged
P = 0.01
70 N = 26 TRF1 and mimic-miRNAs
n = 2,111
N = 26
molecules; actin was used as
− + 60 N = 25
Fragile telomeres (%)

+ miR-155 N = 25 50 n = 1,974 n = 1,988 loading control. E, telomere fragility

N = 25 n = 1,992
− − Flag-TRF1
Fragile telomeres (%)

+ 50 N = 25 in cells described in D; fragility

n = 1,890
40 n = 1,885
Flag-TRF1 40 induced by miR-155 is rescued by
TRF1 30 cointroduction of Flag-tagged
Actin 30 TRF1. F, telomere fragility in
20 20 antago-miR-155-transfected SK-
BR-3 cells treated or untreated with
10 10 Aphidicolin. G, increased
0 frequency of telomeric sister
miR-155 – + + 0
+ − − con.miRNA chromatid fusions induced by miR-
Flag-TRF1 – – + − + + Antago-miR-155 155 is rescued by cointroduction of
Flag-tagged TRF1 in SK-BR-3 cells
− − + Aphidicolin
P = 0.2 (cells are described in D).
G P = 0.004 P = 0.039
H Exemplary images are shown. H,
telomere sister chromatid fusions
N = 26 10 in antago-miR-155–transfected
5 P = 0.39
n = 2,111 SK-BR-3 cells treated or untreated
chromatid fusions (%)

8 P = 0.01 with Aphidicolin. N, number of

chromatid fusions (%)

N = 25 N = 25
Telomeric sister

4 N = 25
n = 1,890 n = 1,992 metaphases spreads analyzed;
Telomeric sister

n = 1,913
6 nN==2,134
26
n, number of analyzed
3 N = 26
n = 2,181 chromosomes. At least three
4
2 experimental replicas were
analyzed. A Student t test was used
2
1 to calculate statistical significance.
0 P values are shown.
0
– + + + − − con.miRNA
miR-155
Flag-TRF1 – – + − + + Antago-miR-155
− − + Aphidicolin

Here, we found that elevated miR-155 levels reduce TRF1 previous results from TRF1 gain- and loss-of-function
abundance at telomeres and competing endogenous miR- experiments (10, 50). Of note, miR138 has been proposed
155 expression increases TRF1 at telomeres. This indicates to regulate the expression of human telomerase; however,
that alterations in miR-155 expression do not only affect the the relevance of this interaction for telomere homeostasis
nuclear pool of TRF1, but directly impact on TRF1 abun- remains unaddressed (16). In line with reports demonstrat-
dance at telomeres. miR-155-dependent modulation of the ing a central role of TRF1 in controlling the replication of
stoichiometry of shelterin complex components resulted in telomeres, we show that elevated miR-155 levels promote
impaired telomere function related to reduced TRF1 expres- telomere fragility coupled with the recruitment of the DNA
sion. We show that miR-155 is a positive regulator of damage marker gH2AX at telomeres (5, 6, 32). Importantly,
telomere length in telomerase-positive SK-BR-3 a luminal miR-155-dependent telomere fragility resulted in increased
breast adenocarcinoma-derived cell line. This is in line with genomic instability exemplified by an increased telomere

4154 Cancer Res; 74(15) August 1, 2014 Cancer Research

Downloaded from cancerres.aacrjournals.org on August 8, 2019. © 2014 American Association for Cancer Research.
 Published OnlineFirst May 29, 2014; DOI: 10.1158/0008-5472.CAN-13-2038

Telomere Regulation by miR-155

sister chromatid fusions in SK-BR-3 cells that are charac- Our work also anticipates the existence of multiple miRNAs
terized by impaired tumor suppression due to the expression that affect telomere function and homeostasis. Identification
of a mutant form of p53. In line with this, experimental and functional characterization of "telo-miRNAs" is expected
reduction of endogenous miR-155 levels improves genomic to provide inroads into the understanding and potential ther-
stability at telomeres. These findings are in line with increas- apeutic treatment of telomere-related maladies such as cancer
ed cancer occurrence in TRF1 knockout mice lacking p53 (6). and aging.
Our results show that driving telomere fragility and genomic
instability is a central aspect in the repertoire of cancer-
Disclosure of Potential Conflicts of Interest
promoting functions of miR-155. This result is of special No potential conflicts of interest were disclosed.
relevance in the light of the recent finding that more than
50% of recurrent amplifications/deletions in human diffuse
large B-cell lymphoma map to fragile sites (29). We show Authors' Contributions
Conception and design: R. Dinami, R. Benetti, C. Schneider, G. Blandino,
that increased miR-155 expression correlates with reduced S. Schoeftner
TRF1 protein levels in specimen from luminal breast cancer Development of methodology: R. Sestito, F. Biagioni, M. Mottolese
that is typically ERþ. Importantly, low TRF1 expression is an Acquisition of data (provided animals, acquired and managed patients,
provided facilities, etc.): C. Ercolani, M. Mottolese
independent predictor of clinical outcome that is associated Analysis and interpretation of data (e.g., statistical analysis, biostatistics,
with reduced distant metastasis-free survival and relapse- computational analysis): E. Petti, S. Piazza, Y. Ciani, A. Sacconi
free survival in patients with ERþ breast cancer. This pattern Writing, review, and/or revision of the manuscript: R. Dinami, E. Petti,
S. Piazza, Y. Ciani, R. Benetti, C. Schneider, S. Schoeftner
is recapitulated by a panel of validated miR-155 targets, Administrative, technical, or material support (i.e., reporting or orga-
underlining the role of miR-155 as "oncomiR." However, this nizing data, constructing databases): C. le Sage, R. Agami
Study supervision: S. Schoeftner
also indicates that TRF1 expression is an integral compo-
nent of a miR-155-dependent signature that predicts poor
clinical outcome in ERþ breast cancer. Grant Support
In conclusion, our work identifies the "oncomiR" miR-155 as This work was supported by the Italian Association for Cancer Research
(AIRC) grant cod. 10299, a Fondazione Veronesi grant (S. Schoeftner), and a
the first miRNA that controls the expression of a shelterin Young Investigator Grant, Ministry of Health GR-2007-683407 (R. Benetti).
component to alter telomere function. Our finding that pro- The costs of publication of this article were defrayed in part by the payment of
motion of telomere-related genomic instability by miR-155 is page charges. This article must therefore be hereby marked advertisement in
linked with poor clinical outcome in an ERþ breast cancer
accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

underlines the relevance of mechanisms that control telomere Received July 24, 2013; revised April 22, 2014; accepted May 12, 2014;
function during cancer formation and/or progression. published OnlineFirst May 29, 2014.

References
1. Chan SRWL, Blackburn EH. Telomeres and telomerase. Philos Trans R 11. Ramsay AJ, Quesada V, Foronda M, Conde L, Martínez-Trillos A,
Soc Lond B Biol Sci 2004;359:109–21. Villamor N, et al. POT1 mutations cause telomere dysfunction in
2. Palm W, de Lange T. How shelterin protects mammalian telomeres. chronic lymphocytic leukemia. Nat Genet 2013;45:526–30.
Annu Rev Genet 2008;42:301–34. 12. Diala I, Wagner N, Magdinier F, Shkreli M, Sirakov M, Bauwens S, et al.
3. Houghtaling BR, Cuttonaro L, Chang W, Smith S. A dynamic molecular Telomere protection and TRF2 expression are enhanced by the
link between the telomere length regulator TRF1 and the chromosome canonical Wnt signalling pathway. EMBO Rep 2013;14:356–63.
end protector TRF2. Curr Biol 2004;14:1621–31. 13. Fujita K, Horikawa I, Mondal AM, Jenkins LMM, Appella E, Vojtesek B,
4. Liu D, Safari A, O'Connor MS, Chan DW, Laegeler A, Qin J, et al. PTOP et al. Positive feedback between p53 and TRF2 during telomere-
interacts with POT1 and regulates its localization to telomeres. Nat Cell damage signalling and cellular senescence. Nat Cell Biol 2010;12:
Biol 2004;6:673–80. 1205–12.
5. Sfeir A, Kosiyatrakul ST, Hockemeyer D, MacRae SL, Karlseder J, 14. Krol J, Loedige I, Filipowicz W. The widespread regulation of microRNA
Schildkraut CL, et al. Mammalian telomeres resemble fragile sites and biogenesis, function and decay. Nat Rev Genet 2010;11:597–610.
require TRF1 for efficient replication. Cell 2009;138:90–103. 15. Iorio M V, Croce CM. MicroRNA dysregulation in cancer: diagnostics,
6. Martínez P, Thanasoula M, Mun ~oz P, Liao C, Tejera A, McNees C, et al. monitoring and therapeutics. A comprehensive review. EMBO Mol
Increased telomere fragility and fusions resulting from TRF1 deficiency Med 2012;4:143–59.
lead to degenerative pathologies and increased cancer in mice. Genes 16. Mitomo S, Maesawa C, Ogasawara S, Iwaya T, Shibazaki M, Yashima-
Dev 2009;23:2060–75. Abo A, et al. Downregulation of miR-138 is associated with over-
7. Denchi EL, de Lange T. Protection of telomeres through independent expression of human telomerase reverse transcriptase protein in
control of ATM and ATR by TRF2 and POT1. Nature 2007;448: human anaplastic thyroid carcinoma cell lines. Cancer Sci 2008;99:
1068–71. 280–6.
8. Li B, Oestreich S, de Lange T. Identification of human Rap1: implica- 17. Canela A, Klatt P, Blasco MA. Telomere length analysis. Methods Mol
tions for telomere evolution. Cell 2000;101:471–83. Biol 2007;371:45–72.
9. Ancelin K, Brunori M, Bauwens S, Koering CE, Brun C, Ricoul M, 18. Samper E, Goytisolo FA, Slijepcevic P, van Buul PP, Blasco MA.
et al. Targeting assay to study the cis functions of human telomeric Mammalian Ku86 protein prevents telomeric fusions independently
proteins: evidence for inhibition of telomerase by TRF1 and for of the length of TTAGGG repeats and the G-strand overhang. EMBO
activation of telomere degradation by TRF2. Mol Cell Biol 2002; Rep 2000;1:244–52.
22:3474–87. 19. Mun ~oz P, Blanco R, de Carcer G, Schoeftner S, Benetti R, Flores JM,
10. Van Steensel B, de Lange T. Control of telomere length by the human et al. TRF1 controls telomere length and mitotic fidelity in epithelial
telomeric protein TRF1. Nature 1997;385:740–3. homeostasis. Mol Cell Biol 2009;29:1608–25.

www.aacrjournals.org Cancer Res; 74(15) August 1, 2014 4155

Downloaded from cancerres.aacrjournals.org on August 8, 2019. © 2014 American Association for Cancer Research.
 Published OnlineFirst May 29, 2014; DOI: 10.1158/0008-5472.CAN-13-2038

Dinami et al.

20. Voorhoeve PM, le Sage C, Schrier M, Gillis AJM, Stoop H, Nagel R, 37. Miyachi K, Fujita M, Tanaka N, Sasaki K, Sunagawa M. Correlation
et al. A genetic screen implicates miRNA-372 and miRNA-373 as between telomerase activity and telomeric-repeat binding factors in
oncogenes in testicular germ cell tumors. Cell 2006;124:1169–81. gastric cancer. J Exp Clin Cancer Res 2002;21:269–75.
21. Peuscher MH, Jacobs JJL. Posttranslational control of telomere main- 38. Yamada M, Tsuji N, Nakamura M, Moriai R, Kobayashi D, Yagihashi A,
tenance and the telomere damage response. Cell Cycle 2012;11: et al. Down-regulation of TRF1, TRF2 and TIN2 genes is important to
1524–34. maintain telomeric DNA for gastric cancers. Anticancer Res 22:3303–7.
22. Faraoni I, Antonetti FR, Cardone J, Bonmassar E. miR-155 gene: a 39. Nakanishi K, Kawai T, Kumaki F, Hiroi S, Mukai M, Ikeda E, et al.
typical multifunctional microRNA. Biochim Biophys Acta 2009;1792: Expression of mRNAs for telomeric repeat binding factor (TRF)-1 and
497–505. TRF2 in atypical adenomatous hyperplasia and adenocarcinoma of the
23. Mattiske S, Suetani RJ, Neilsen PM, Callen DF. The oncogenic role of lung. Clin Cancer Res 2003;9:1105–11.
miR-155 in breast cancer. Cancer Epidemiol Biomarkers Prev 2012;21: 40. Kondo T, Oue N, Yoshida K, Mitani Y, Naka K, Nakayama H, et al.
1236–43. Expression of POT1 is associated with tumor stage and telomere
24. Yan LX, Huang XF, Shao Q, Huang MY, Deng L, Wu QL, et al. MicroRNA length in gastric carcinoma. Cancer Res 2004;64:523–9.
miR-21 overexpression in human breast cancer is associated with 41. Oh BK, Kim YJ, Park C, Park YN. Up-regulation of telomere-binding
advanced clinical stage, lymph node metastasis and patient poor proteins, TRF1, TRF2, and TIN2 is related to telomere shortening
prognosis. RNA 2008;14:2348–60. during human multistep hepatocarcinogenesis. Am J Pathol 2005;
25. Chen J, Wang BC, Tang JH. Clinical significance of microRNA-155 166:73–80.
expression in human breast cancer. J Surg Oncol 2012;106:260–6. 42. Matsutani N, Yokozaki H, Tahara E, Tahara H, Kuniyasu H, Haruma K,
26. Wang F, Zheng Z, Guo J, Ding X. Correlation and quantitation of et al. Expression of telomeric repeat binding factor 1 and 2 and TRF1-
microRNA aberrant expression in tissues and sera from patients with interacting nuclear protein 2 in human gastric carcinomas. Int J Oncol
breast tumor. Gynecol Oncol 2010;119:586–93. 2001;19:507–12.
27. Ringne r M, Fredlund E, Ha €kkinen J, Borg Å, Staaf J. GOBO: gene 43. Kishi S, Wulf G, Nakamura M, Lu KP. Telomeric protein Pin2/TRF1
expression-based outcome for breast cancer online. PLoS ONE 2011; induces mitotic entry and apoptosis in cells with short telomeres and
6:e17911. is down-regulated in human breast tumors. Oncogene 2001;20:
28. Biagioni F, Bossel Ben-Moshe N, Fontemaggi G, Canu V, Mori F, 1497–508.
Antoniani B, et al. miR-10b , a master inhibitor of the cell cycle, is 44. Clurman BE, Hayward WS. Multiple proto-oncogene activations in
down-regulated in human breast tumours. EMBO Mol Med 2012;4: avian leukosis virus-induced lymphomas: evidence for stage-specific
1214–29. events. Mol Cell Biol 1989;9:2657–64.
29. Barlow JH, Faryabi RB, Calle n E, Wong N, Malhowski A, Chen HT, et al. 45. Costinean S, Zanesi N, Pekarsky Y, Tili E, Volinia S, Heerema N, et al.
Identification of early replicating fragile sites that contribute to genome Pre-B cell proliferation and lymphoblastic leukemia/high-grade lym-
instability. Cell 2013;152:620–32. phoma in E(mu)-miR-155 transgenic mice. Proc Natl Acad Sci U S A
30. De Lange T. T-loops and the origin of telomeres. Nat Rev Mol Cell Biol 2006;103:7024–9.
2004;5:323–9. 46. Volinia S, Calin GA, Liu CG, Ambs S, Cimmino A, Petrocca F, et al. A
31. Sen D, Gilbert W. Guanine quartet structures. Methods Enzymol 1992; microRNA expression signature of human solid tumors defines cancer
211:191–9. gene targets. Proc Natl Acad Sci U S A 2006;103:2257–61.
32. Bosco N, de Lange T. A TRF1-controlled common fragile site contain- 47. Chang S, Wang RH, Akagi K, Kim KA, Martin BK, Cavallone L, et al.
ing interstitial telomeric sequences. Chromosoma 2012;121:465–74. Tumor suppressor BRCA1 epigenetically controls oncogenic micro-
33. Blasco MA. Telomeres and human disease: ageing, cancer and RNA-155. Nat Med 2011;17:1275–82.
beyond. Nat Rev Genet 2005;6:611–22. 48. Jiang S, Zhang HW, Lu MH, He XH, Li Y, Gu H, et al. MicroRNA-155
34. De Lange T. Telomere-related genome instability in cancer. Cold functions as an OncomiR in breast cancer by targeting the suppressor
Spring Harb Symp Quant Biol 2005;70:197–204. of cytokine signaling 1 gene. Cancer Res 2010;70:3119–27.
35. Shay JW, Bacchetti S. A survey of telomerase activity in human cancer. 49. Valeri N, Gasparini P, Fabbri M, Braconi C, Veronese A, Lovat F, et al.
Eur J Cancer 1997;33:787–91. Modulation of mismatch repair and genomic stability by miR-155. Proc
36. Garcia-Aranda C, de Juan C, Diaz-Lopez A, Sanchez-Pernaute A, Natl Acad Sci U S A 2010;107:6982–7.
Torres AJ, Diaz-Rubio E, et al. Correlations of telomere length, telo- 50. Lee TH, Tun-Kyi A, Shi R, Lim J, Soohoo C, Finn G, et al. Essential role
merase activity, and telomeric-repeat binding factor 1 expression in of Pin1 in the regulation of TRF1 stability and telomere maintenance.
colorectal carcinoma. Cancer 2006;106:541–51. Nat Cell Biol 2009;11:97–105.

4156 Cancer Res; 74(15) August 1, 2014 Cancer Research

Downloaded from cancerres.aacrjournals.org on August 8, 2019. © 2014 American Association for Cancer Research.
 Published OnlineFirst May 29, 2014; DOI: 10.1158/0008-5472.CAN-13-2038

miR-155 Drives Telomere Fragility in Human Breast Cancer by

Targeting TRF1
Roberto Dinami, Cristiana Ercolani, Eleonora Petti, et al.

Cancer Res 2014;74:4145-4156. Published OnlineFirst May 29, 2014.

Updated version Access the most recent version of this article at:
doi:10.1158/0008-5472.CAN-13-2038

Supplementary Access the most recent supplemental material at:

Material http://cancerres.aacrjournals.org/content/suppl/2014/07/21/0008-5472.CAN-13-2038.DC1

Cited articles This article cites 49 articles, 17 of which you can access for free at:
http://cancerres.aacrjournals.org/content/74/15/4145.full#ref-list-1

Citing articles This article has been cited by 1 HighWire-hosted articles. Access the articles at:
http://cancerres.aacrjournals.org/content/74/15/4145.full#related-urls

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

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at
Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link
http://cancerres.aacrjournals.org/content/74/15/4145.
Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)
Rightslink site.

Downloaded from cancerres.aacrjournals.org on August 8, 2019. © 2014 American Association for Cancer Research.