dynamics of senescent cell formation and retention · using the extended depth of focus function...

Molecular and Cellular Pathobiology

Dynamics of Senescent Cell Formation and RetentionRevealed by p14ARF Induction in the Epidermis

Ronit Tokarsky-Amiel1, Narmen Azazmeh1, Aharon Helman1, Yan Stein1, Alia Hassan1,Alexander Maly2, and Ittai Ben-Porath1

AbstractCellular senescence, a state of cell-cycle arrest accompanied by dramatic morphologic andmetabolic changes,

is a central means by which cells respond to physiologic stress and oncogene activity. Senescence is thought toplay important roles in aging and in tumor suppression, yet the dynamics by which senescent cells are formed,their effects on tissue function and their eventual fate are poorly understood. To study cellular senescence withinan adult tissue, we developed transgenic mice inducibly expressing p14ARF (human ortholog of murine p19ARF), acentral activator of senescence. Induction of p14ARF in the epidermis rapidly led towidespread apoptosis and cell-cycle arrest, a stage that was transient, and was followed by p53-dependent cellular senescence. The endogenousCdkn2a products p19ARF and p16Ink4a were activated by the transgenic p14ARF through p53, revealing asenescence-promoting feed-forward loop. Commitment of cells to senescence required continued p14ARF

expression, indicating that entry into this state depends on a persistent signal. However, once formed, senescentcells were retained in the epidermis, often forweeks after transgene silencing, indicating an absence of an efficientrapidly acting mechanism for their removal. Stem cells in the hair follicle bulge were largely protected fromapoptosis upon p14ARF induction, but irreversibly lost their ability to proliferate and initiate follicle growth.Interestingly, induction of epidermal hyperplasia prevented the appearance of senescent cells upon p14ARF

induction. Our findings provide basic insights into the dynamics of cellular senescence, a central tumor-suppressive mechanism, and reveal the potential for prolonged retention of senescent cells within tissues. CancerRes; 73(9); 2829–39. �2013 AACR.

IntroductionCellular senescence is a coordinated program activated by

cells in response to a variety of physiologic stresses, includingDNA damage and oncogenic signaling (1–3). Senescent cellshave been identified in early tumor lesions in mouse modelsand human samples (4), providing evidence that senescence isan important barrier to the cancer formation. Cells enteringsenescence undergo cell-cycle arrest, which is thought to beirreversible, accompanied by dramatic changes inmorphology,chromatin structure, and metabolism (1–3). The Rb and p53tumor suppressor proteins are central in the execution ofsenescence and are most often activated in this context bythe two products of the Ink4a (Cdkn2a) locus: p16Ink4a and

p19ARF (p14ARF in human; refs. 1, 2, 5). These genes are notexpressed in the largemajority of tissues in the embryo and theadult, but are transcriptionally activated during stress, tumor-igenesis, or aging (6–8).

Senescence has been detected in vivo in various contexts,including in tissues suffering from DNA damage or shortenedtelomeres (9, 10), in chemotherapy-treated tumors (11, 12), andin tissues undergoing wound healing, fibrosis, or inflammation(13–15). Senescence is also detected during aging (16–18) andis thought to contribute to this process, potentially through thedepletion of functional stem cells (1–3). Secretion of inflam-matory cytokines and ECM components, an important aspectof the senescent cell phenotype, could influence tissue agingand exert other non-cell-autonomous effects including stim-ulation of proliferation of neighboring cells (19).

Although senescent cells can remain viable in culture formonths and years, their fate in vivo is unclear. The detection ofsenescent cells in aged tissues suggests that they can accu-mulate and affect tissue physiology. Consistent with this,experimental removal of senescent cells prevents aging-likephenotypes caused by genetically induced aneuploidy (20).However, recent studies have shown that immune cellsrecruited through cytokine secretion rapidly clear senescentcells from cancerous or fibrotic tissue (13, 21, 22). Whetheradditional mechanisms act to eliminate senescent cells is notknown, and a detailed analysis of the retention rates ofsenescent cells in tissues is lacking.

Authors' Affiliations: 1Department of Developmental Biology and CancerResearch, Institute for Medical Research – Israel–Canada, HadassahSchool ofMedicine, TheHebrewUniversity of Jerusalem; and 2Departmentof Pathology, Hadassah Medical Center, Jerusalem, Israel

Note: Supplementary data for this article are available at Cancer ResearchOnline (http://cancerres.aacrjournals.org/).

R. Tokarsky-Amiel and N. Azazmeh contributed equally to this work.

Corresponding Author: I. Ben-Porath, Hebrew University of Jerusalem,Hadassah School of Medicine, Jerusalem 91120, Israel. Phone: 972-2-6757006; Fax: 972-2-6757482; E-mail: [email protected]

doi: 10.1158/0008-5472.CAN-12-3730

�2013 American Association for Cancer Research.

CancerResearch

www.aacrjournals.org 2829

on January 1, 2021. © 2013 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst February 19, 2013; DOI: 10.1158/0008-5472.CAN-12-3730

http://cancerres.aacrjournals.org/

Basic study of the senescence program in vivo is essential forunderstanding its various physiologic roles including tumorsuppression. To allow the study of senescent cells in vivo, thedynamics of their formation and removal, and their effects onadult tissues, we sought to develop a system in which senes-cence could be activated in a direct and controlledmanner.Wegenerated transgenic mice carrying a tetracycline-induciblep14ARF gene, a central activator of senescence, whose productactivates p53 throughMdm2 sequestration (23).We focused onthe skin as a model epithelial tissue, in which senescence isthought to occur during normal aging (16, 24). Our findingsreveal basic aspects of the dynamics of senescent cell forma-tion and retention in the epidermis.

Materials and MethodsTransgenic mice

To generate tetracycline-inducible p14ARF (tet-p14)mice, werelied on Flp/FRT-mediated recombination into embryonicstem (ES) cells (25). We cloned the human p14ARF codingsequence into the pBS31 vector, which contains an FRTrecombination site and a tet-responsive element, and trans-fected this plasmid togetherwith a Flp/e expression vector intothe KH2 ES cells, in which an FRT site was placed in the Col1a1gene locus. Recombined cells were injected into tetraploidblastocysts for implantation. Tet-p14 mice (mixed C57Bl6 and129sv) were crossed with K5-rtTA mice (FVB; ref. 26) and withtet-shp53mice (mixedC57Bl6 and 129sv; ref. 27). For transgeneinduction, 2 mg/mL doxycycline (dox) was added to the drink-ing water of double-transgenic mice and sibling control single-transgenics at 3 weeks of age. For bromodeoxyuridine (BrdUrd)labeling mice were injected with 0.1 mg/g body-weight 2 hoursbefore sacrifice. For hyperplasia induction, mice were treated 3times a week with 6.5 mg 12-O-tetradecanoylphorbol-13-acetate(TPA) in 100mL acetone; untreated skin regions from the samemice were collected as control tissue.

ImmunohistologyUpon sacrifice, mice were shaved and back skins were

dissected, formalin-fixed and paraffin-embedded. Immunohis-tologywas carried out according to standard procedures, usingPeroxidase Subtrate Kits (Vector) or fluorescently labeledsecondary antibodies (Jackson). Antibodies used: p14ARF

(ab3642, Abcam), p21 (sc-6246, Santa Cruz), p53 (CM5p, Novo-castra), p19ARF (ab80, Abcam), Ki67 (RM-9106, Labvision),BrdUrd (MS-1058, Labvision), CC3 (#9661, Cell Signaling),K14 (GP-CK14, Progen), K5 (GP-CK5, Progen), and K15 (sc-56520, SantaCruz). In SituCell DeathDetectionKit (Roche)wasused for terminal deoxynucleotidyl transferase–mediated dUTPnick end labeling (TUNEL) analysis. Imageswere collected usingan Olympus BH2 upright or a Nikon Eclipse Ti inverted fluo-rescent microscope, using DS-Qi1Mc and DS-Fi1 cameras, andprocessed using NIS Elements software (Nikon), in some casesusing the Extended Depth of Focus function for Z-stacking.Where noted, we used a Zeiss LSM710 confocal microscope.

Senescence-associated b-galactosidase stainsTen- to 12-mm cryosections of OCT-embedded mouse skins

were fixed in 0.5% glutaraldehyde for 15 minutes, stained

overnight at 37�C with 40 mmol/L phosphate buffer pH ¼ 6with 5 mmol/L K4Fe(CN)6, 5 mmol/L K3Fe(CN)6, 150 mmol/LNaCl, 2 mmol/L MgCl2, and 1 mg/mL X-gal, washed in PBS,fixed in 95% ethanol for 15 minutes, counterstained withnuclear fast red, dehydrated, and mounted. The numbers ofsenescence-associated b-galactosidase (SAbGal)-positive cellsamong interfollicular epidermal cells were scored from morethan 10 microscopic fields.

RNA extraction and qRT-PCRBack skin samples were minced and RNA was extracted

using the RNeasy Fibrous Tissue kit (Qiagen). Quantitativereverse transcribed PCR (qRT-PCR) and TaqMan reactionswere conducted using standard methods.

Resultsp14ARF induction in the skin

To develop a system for the induction of senescence in theadult skin, we generated tetracycline-inducible p14ARF mice(tet-p14). We used the human gene to facilitate the distinctionbetween the endogenous p19ARF and the transgene. Tet-p14mice were crossed with K5-rtTA transgenic mice (Fig. 1A),which carry the tetracycline-dependent transcriptional acti-vator under the control of the Keratin 5 promoter (26), direct-ing expression to the basal layer of the epidermis. Mice treatedwith doxycycline expressed the transgenic p14ARF in themajority of basal epidermal cells, as detected with a human-specific antibody (Fig. 1B and Supplementary Fig. S1). p14ARF

appeared in subnuclear puncta, consistent with localization inthe nucleoli, the location of the endogenous protein (Fig. 1B).

Double-transgenic mice treated with doxycycline for 2 daysshowed high levels of nuclear p53 expression in the epidermis,and colocalized expression of its target, p21Cip1 (Fig. 1C).Expression levels of additional known p53 transcriptionaltargets were increased (Fig. 1D), indicating that p53 wasactivated. The transgenic p14ARF was thus properly localizedand molecularly active.

p14ARF induces cell-cycle arrest and apoptosisp53 activation can lead to several cellular outcomes, includ-

ing reversible cell-cycle arrest, apoptosis, and senescence (28).We examined which of these occurred upon p14ARF induction.Two days of transgene activation in 3-week-old mice led to adramatic decrease in proliferation of basal epidermal cells, asindicated by Ki67 staining (Fig. 1E). However, this effect wastransient, and surprisingly, after 1 week we observed hyper-proliferation of basal cells, which was driven by cells thatdid not express the p14ARF transgene; the degree of hyperpro-liferation decreased after longer induction times (Fig. 1E andF). We also observed large numbers of cells undergoingapoptosis after 2 days of transgene activation, involvingapproximately 15% of basal cells (Fig. 1G and H). This alsowas transient: by one week of induction apoptosis levelsreturned nearly to normal (Fig. 1H), and at 2weeks of inductionthe expression of apoptosis-mediating p53 targets (Noxa andBax) was similar to that in control mice (Fig. 1D).

These findings indicate that p14ARF activation in vivo issufficient to induce twoof the central p53 responses: apoptosis,

Tokarsky-Amiel et al.

Cancer Res; 73(9) May 1, 2013 Cancer Research2830




and cell-cycle arrest. This initial synchronous response istransient, and is counteracted by hyperproliferation of neigh-boring transgene-nonexpressing cells, most likely acting tomaintain tissue homeostasis.

p14ARF induces p53-dependent cellular senescence andactivates the endogenous Ink4a productsBecause p14/19ARF activation often occurs in situations

where senescence is the cellular outcome, we tested whether

senescent cells appear upon transgene induction. Two or fourdays after p14ARF activation, we did not observe an increase inthe numbers of cells showing SAbGal activity, a robust markerof senescence (refs. 16, 29; Fig. 2A and B and data not shown).However, starting at 1 week of transgene activation, wedetected increased SAbGal-positive cell numbers in the epi-dermis (Fig. 2A and B). A small number of SAbGal-positive cellswas found also in controlmice; we determined that 4% to 8% ofepidermal cells underwent senescence in response to p14ARF

0

2

4

6

8

10

12

NoxaBaxMdm2 Dram1

Control

p14

0

2

4

6

8

10

12

NoxaBax

A

B tet-p14 K5-rtTA / tet-p14p14

K14 p14

Cp53

p21

p53 p21

tet-p14ARFK5-rtTA / tet-p14ARF

2d

2d

50 μm

25 μm

rtTAK5

tet-O

Col1a1 locus

p14ARF

D

EK5-rtTA / tet-p14tet-p14

2d

7d

Ki67

F

G

H

2d

Re

lati

ve

ex

pre

ss

ion

*

*

*

n=3

n=4

n=3

n=3

K14 p14 BrdUrd 20 µm

7d

50 μm

50 μm

tet-p14 K5-rtTA / tet-p14

TUNEL

CC3

2d

2d

50 μm

50 μm

n=4n=4n=3 n=5

7d2d

TU

NE

L-p

osit

ive f

racti

on

Ki6

7-p

os

itiv

e f

rac

tio

n

n=5n=4 n=6n=4n=9n=12n=3n=3

Control

p14

Control

p14

14d2d

6 wk7d4d2d

** * ** *

**

Figure 1. p14ARF induces p53 activation, cell-cycle arrest, and apoptosis. A, diagram of transgenic mouse lines crossed to generate inducible p14ARF

expression in the epidermis. B, top, immunohistochemical stain of humanp14ARF (brown) in the skins of 3-week-old double-transgenicmice (K5-rtTA/tet-p14)and control single-transgene mice (tet-p14) treated with doxycycline (Dox) for 2 days (2d). Bottom, immunofluorescent stain of p14ARF (white)in the same tissues. K14 (green) labels the basal epidermis. DNA [40, 6-diamidino-2-phenylindole (DAPI)] is labeled blue in all images. C, costain of p53 (green)and p21 (red) in the same tissues. Merged images are shown on bottom. D, expression levels of indicated p53 target genes as assessed by qRT-PCRon skins of mice in which p14 was induced for 2 days; graph on right shows expression of apoptosis-associated p53 targets following 14 days ofinduction. Values indicate mean across individual mice � SEM. E, stain of the proliferation marker Ki67 in skins of indicated mice treated withdoxycycline for 2or 7days. Bottom images showcostain ofBrdUrd (red) andp14ARF (white) upon7daysof activation; absenceof costainedcells indicates thatproliferating cells do not express the transgene. F, fraction of Ki67-positive basal epidermal cells following p14ARF activation for the indicated times.Dots indicate individual mice, bars indicate mean across mice � SEM. G, staining of apoptotic cells (arrows) in skins of mice treated with doxycycline for2 days, detected by cleaved caspase-3 (CC3, top, brown) or TUNEL (bottom, red). H, fraction of TUNEL-positive epidermal cells in mice treated withdoxycycline for the indicated times. �, P < 0.05; ��, P < 0.005; Student t test. NS, nonsignificant.

Senescent Cell Formation and Retention in the Epidermis

www.aacrjournals.org Cancer Res; 73(9) May 1, 2013 2831




activation above this basal level (Fig. 2B and SupplementaryFig. S2).

We examined the expression of senescence-associatedmarker genes in mRNA extracted fromwhole skins. Dcr2 levelswere mildly increased after 2 weeks of transgene induction,whereas Pai1 was not significantly induced (Fig. 2C). Interest-ingly, the levels of the two products of the endogenous Ink4alocus, p16Ink4a and p19ARF, were dramatically increased (Fig.2C). This revealed a positive-feedback loop bywhich p14/19ARF

reinforces its own expression as well as of its partner, p16Ink4a,promoting both p53 and Rb pathway activity. The endogenousp19ARF protein was detectable in a subset of highly expressingepidermal cells using a mouse-specific antibody (Fig. 2D).An increase in the mRNA levels of p16Ink4a and p19ARF wasdetected in some mice already at 2 days of induction (Fig. 2C),

suggesting that initiation of senescence occurs early aftertransgene expression.

p19ARF has been shown to execute p53-independent func-tions, but whether these pertain to senescence is unclear (30,31). We found that when an inducible transgenic short hairpinRNA (shRNA) against p53 (tet-shp53; ref. 27) was coexpressedwith the p14ARF transgene, eliminating p53 expression, noSAbGal-positive cells appeared (Fig. 2E–G). This indicates thatp14ARF-induced epidermal cell senescence is dependent onp53. The endogenous p16Ink4a and p19ARF genes were notactivated in shp53-expressing mice (Fig. 2H), revealing thatthe self-amplification of p14/19ARF expression occurredthrough p53.

Consistent with the activation of the cyclin/cyclin-depen-dent kinase (CDK) inhibitors p16Ink4a and p21Cip1, we observed

0

2

4

6

8

10

Control p14 p14 +

shp53

p16p19

0

5

10

15

20

25

Pai1Dcr2p16p19ARF

0

5

10

15

20

25

Pai1Dcr2p16p19ARF

Re

lati

ve

ex

pre

ss

ion

Control

p14

6wk

A tet-p14 K5-rtTA / tet-p14

4d

1wk

4wk

tet-p14

K5-rtTA / tet-p14

K14 p19 2wk2d

n=3

n=3 *

*

*

C D

2wk

K5-rtTA / tet-p14 / tet-shp53

K14 p14

K14 p53

K5-rtTA / tet-p14 GE

F

B

p14+p14

shp53Control

SA

βGal-

po

sit

ive f

racti

on

25 µm

SAβGal

K5-rtTA / tet-p14 / tet-shp53K5-rtTA / tet-p14

25 µm

H

2wk

25 µm

Rela

tive e

xp

ressio

n

SAβGal

SA

βGa

l-p

os

itiv

e fra

cti

on

ove

r m

atc

hin

g c

on

tro

lsS

AβG

al-

po

sit

ive

fra

cti

on

6wk2wk1wk

Control

p14

***

25 µm

n=3

n=4

4d

6wk2wk1wk4d

**

Figure 2. p14ARF induces p53-dependent senescence andactivates the endogenous Ink4aproducts. A, SAbGal stain of skinsections from double-transgenicand control mice treated withdoxycycline for the indicated timesstarting at 3 weeks of age (d, day;wk, week). B, fraction of epidermalSAbGal-positive cells in controland p14ARF-induced mice treatedwith doxycycline for the indicatedtimes. Bottom graph shows valuesafter subtraction of SAbGal-positive fractions in matchedcontrol mice (gray dots in topgraph), to correct for batch andnormal aging effects. C,expression levels of indicatedsenescence-associated genes inmice induced for p14ARF for theindicated times, assessed byqRT-PCR. D, confocal image ofendogenous p19ARF expression(red, punctate) in mice expressingp14ARF for 2 weeks. Red in topimage is autofluorescence. E, stainof p14ARF (top) and p53 (bottom)in double-transgenic miceexpressing the transgene for 2weeks (left) and in sibling triple-transgenic mice coexpressing aninducible p53 shRNA (tet-shp53).F, SAbGal stain of skin sectionsfrom the same mice. G, fractionof SAbGal-positive cells in theepidermis of control, double-,and triple-transgenic mice. H,expression levels of p19ARF andp16Ink4a in the skins of the samemice. *, P < 0.05; **, P < 0.005.






a dramatic decrease in the levels of phosphorylated Rb in theepidermis, particularly in transgene-expressing cells, after 2days and 2 weeks of activation (Supplementary Fig. S3). Inter-estingly, phosphorylation levels of the Rb homolog p130 werealso decreased in transgene-expressing cells, although this wasless apparent at 2 weeks of activation (Supplementary Fig. S3).These findings indicate that both Rb and p130 are initiallyactivated in response to p14ARF induction and may contributeto cell-cycle arrest and senescence.

Commitment to senescence requires continuous p14ARF

expressionWe next assessed the time required for cells to commit to

senescence after p14ARF activation. Conceivably, once initiated,the senescence program could be executed independently ofthe inducing signal. To test whether this is the case, weactivated p14ARF for 4 days and then resilenced the transgene

by doxycycline removal for another 3 days (Fig. 3A and B).While SAbGal-positive cells appeared in mice in which thetransgene was continuously expressed for 7 days, we did notobserve the appearance of senescent cells in mice in which thetransgene was silenced (Fig. 3C and D). The numbers of cellsexpressing stable p53 were also decreased in the epidermis ofthesemice (Fig. 3E and F). Three days after transgene silencing,the mRNA levels of p19ARF and p16Ink4a were decreased, yetremained higher than those observed in control mice (Fig. 3G),reflecting a lag in response to p14ARF silencing. Together theseresults indicate that cells do not commit to senescence until atleast 5 days of continuous high p14ARF expression levels,despite activating p53, p16Ink4a, and p19ARF, and are able torevert to a nonsenescent state or be eliminated from the tissue.This suggests that in a physiologic setting only persistent stressleading to continuous p14/19ARF expression will cause cells tocommit to senescence.

0

10

20

30

40

50

4d on +7dCont3d off

Rela

tive e

xp

ressio

n

0

2

4

6

8

10

4d on +7dCont3d off

Rela

tive e

xp

ressio

n

0

1

2

3

4

5

4d on +7dCont3d off

Rela

tive e

xp

ressio

n

D

F

B

n=6n=5n=6

n=7n=5n=7

A

C

E

4d on + 3d off7d

4d on + 3d off7d

25 µm

25 μm

4d on + 3d off7d

25 µm

* NS

p16p19

***

***

n=3n=4n=4

NS

****

p5

3 p

os

itiv

e f

rac

tio

nS

AβG

al

po

sit

ive

fra

cti

on

4d on +

3d off

7dCont

G

NS

****

4d on +

3d off

7dCont

****

NS

n=3n=4n=4

K14 p14

SAβGal

p53

Cont

Cont

Cont

n=3n=4n=4

p14

p53

SAβGal

Figure 3. Commitment to senescence requires continuous p14ARF expression. A, stain of p14ARF (white) in skins ofmice inwhich it was activated for 7 days (7d)or for 4 days followed by 3 days in which doxycycline was removed (4d on þ 3d off). B, relative p14ARF expression levels in indicated mice, showingtransgene shutoff. C, SAbGal stain of skins of indicated mice. D, fraction of SAbGal-positive cells in the epidermis. E, stain of p53 (brown) in skins ofthe same mice. F, fraction of p53-positive cells in the epidermis. G, p19ARF and p16Ink4a expression levels in the indicated samples. �, P < 0.05; ��, P < 0.005;NS, nonsignificant.






Senescent cells are retained in the epidermisWe next assessed whether, once formed, senescent cells are

retained in the epidermis or are rapidly eliminated. We acti-vated p14ARF for 2 weeks to generate senescent cells, and thenremoved doxycycline to resilence the transgene for an addi-tional 4 weeks (Fig. 4A and B). Strikingly, we found that highnumbers of SAbGal-positive cells remained in the epidermis4 weeks after transgene silencing (Fig. 4C andD). Furthermore,p53-positive cells were also detected in the epidermis aftertransgene silencing, indicating that, indeed, senescent cellswere retained in the tissue (Fig. 4E and F). Many of the retainedSAbGal-positive and p53-positive cells were located in thebasal layer, despite the high cell turnover in this compartment(Fig. 4C andE). Thesefindings indicate that, once formed in theepidermis, senescent cells are capable of retaining their phe-notype and position for weeks, independently of the initiatingsignal.

The numbers of SAbGal-positive and p53-positive cells,were, however, reduced in the transgene-silencedmice relativeto mice continuously expressing p14ARF for 6 weeks (Fig. 4Dand F). This could be a result of the reduced formation of newsenescent cells upon transgene silencing, but also due to

gradual clearance of senescent cells and/or a partial loss ofthe senescent phenotype. Surprisingly, the levels of p19ARF andp16Ink4a were greatly reduced in transgene-silenced mice (Fig.4G), indicating that their expression depended on a continuoussignal, and that independent mechanisms must support p53activity and senescence observed in the retained cells.

p14ARF induces hair follicle stem cell dysfunctionDespite the various effects of transgene induction, skin

morphology remained largely intact (Supplementary Fig. S4).Epidermal thickness was increased in some mice, a potentialresult of the observed hyperproliferation; however this increasewas limited in significance and was variable between mice andskin regions (Supplementary Fig. S4).

The most striking change in the skin was a dramaticdecrease in the numbers of developed hair follicles, andprogressive hair loss (alopecia), which was noticeable 10days after induction and pronounced by 6 weeks (Fig. 5A andB). The growth stage of hair follicles, anagen, is initiated byentry into the cell cycle of hair follicle stem cells residing inthe bulge region; this occurs synchronously in all hairfollicles at 3 weeks of age (32). We found that the transgenic

0

2

4

6

8

10

12

14

2wk on +

4wk off

6wkCont

Rela

tive e

xp

ressio

n

0

5

10

15

20

25

30

35

2wk on +

4wk off

6wkCont

Re

lati

ve

ex

pre

ss

ion

0

5

10

15

20

25

30

35

2wk on +

4wk off

6wkCont

Re

lati

ve

ex

pre

ss

ion

G

A

C2wk on + 4wk off6wk2wkCont

E2wk on + 4wk off6wk2wkCont

p53

2wk on + 4wk off2wk

25 µm

25 µm

25 µm

K14 p14

SAβGal

p16p19 NS

p14NS

***

B

D

F

*********

*

p53-p

osit

ive f

racti

on

rela

tive to

co

ntr

ol

p53

SA

βGal-

po

sit

ive f

racti

on

2wk on +

4wk off

6wkCont

SAβGal

2wk on +

4wk off

6wkCont

******

6wkCont

*

n=8n=2n=4

n=8n=2n=3

n=9

n=6

n=3

n=10

n=4

n=3

n=9n=3n=5

*

Figure 4. Senescent cells are retained in the epidermis. A, stain of p14ARF in skins of mice in which it was activated for 2 or 6 weeks (2, 6wk) and in mice inwhich it was activated for 2 weeks and then silenced for 4 weeks (2wk on þ 4wk off). Control (Cont) mice received doxycycline either for 6 weeks or for2weeks followedby 4weekswithout treatment. B, relativep14ARF expression levels in indicatedmice. C, SAbGal stain of skins of indicatedmice. D, fraction ofSAbGal-positive cells in the epidermis. E, stain of p53 in indicated mice. F, fraction of p53-positive cells in the epidermis, normalized to the meanvalues of control mice in two independent experiments. G, expression levels of p19ARF and p16Ink4a in the indicated samples. �, P < 0.05; ��, P < 0.005;NS, nonsignificant.






p14ARF was expressed in bulge cells, identified by Keratin15 (K15) expression, and that p53 was concomitantly sta-bilized in these cells (Fig. 5C). Activation of p14ARF at 3weeks of age blocked the entry of bulge stem cells into thecell cycle and the initiation of anagen, and follicles remainedin a telogen-like (resting) state (Fig. 5B and D). Silencing ofp53 at the time of p14ARF induction allowed normal entryinto anagen (Supplementary Fig. S5).Wenoted that at 2 days of activation, despite their high levels

of p53 expression, much fewer hair follicle bulge cells under-went apoptosis thandid other epidermal cells (Fig. 5E andF); incontrast, after 2 weeks of inductionmany bulge cells expressedhigh levels of endogenous p19ARF (Fig. 5G). These stem cellsthus display a qualitatively distinct response to p53 from thatof other epidermal cells, primarily activating the circuitry ofsenescence rather than of apoptosis. Together, these findingsindicate that p14ARF activation prevents hair follicle stem cells

from initiating growth through blockage of their proliferativepotential.

p14ARF resilencing does not allow hair follicle growthWe next assessed whether the resilencing of p14ARF would

allow hair follicle stem cells to regain function, and initiateanagen. We found that in mice in which p14ARF was activatedfor 2 weeks and then resilenced for 4 weeks, only 18% of hairfollicles contained proliferating cells in the bulge or the hairgerm region below it, in contrast with 66% of follicles in controlmice (Fig. 6A and B). This indicates that the dysfunction of hairfollicle stem cells is largely irreversible. As in the interfollicularepidermis, p53 was detected in bulge cells of transgene-silenced mice (Fig. 6C). p19ARF was detectible up to 2 weeksafter transgene shutoff (Fig. 6D).

To directly test whether transgene resilencing wouldallow hair growth, we activated p14ARF for 6 weeks to induce

BK5-rtTA / tet-p14tet-p14

8d

A tet-p14 K5-rtTA / tet-p14

6wk

D

4d

tet-p14

K5 K15 Ki67

K5-rtTA / tet-p14

K5 K15 Ki67

25 μm

C K5-rtTA / tet-p14

2d

K15 p53 K15 p14

25 μm

K15 p14 TUNEL

E

2d

K5-rtTA / tet-p14

F

25 μm

n=4n=4n=5n=4

0.007

7d2d

TU

NE

L-p

osit

ive f

racti

on

Control

p14

**

*NS

bulgebulge epi

0.15

n=5

K5 p19 K15 p53

2wk

K5-rtTA / tet-p14

G

25 μm

100 μm

Figure 5. p14ARF causes dysfunction of hair follicle bulge stem cells. A, control and double-transgenic mice following 6 weeks of p14ARF induction. Shown aretwo induced mice with different degrees of alopecia. B, skins of control and double-transgenic mice (hematoxylin and eosin–stained) treated withdoxycycline for 8 days starting at 3 weeks of age, the time of synchronous entry into anagen. Control mice show large follicles in anagen. C, left, stainof p14ARF (white) in the hair follicle bulge labeled with K15 (red) 2 days after induction. Right, expression of p53 in the bulge (different follicle). D, stainof Ki67 (white) in 3-week-old mice treated with doxycycline for 4 days. K5 (green) labels the basal epidermis. Follicles of control mice showproliferating bulge cells (arrowhead) and progenitor cells (arrow); p14ARF-induced hair follicle stem cells do not express Ki67 and do not give rise tothe hair germ. Insets show region of bulge cell proliferation. E, TUNEL stain (red) of mice 2 days after p14ARF induction, showing lack of apoptosis inthe bulge (K15, green) and apoptosis in the epidermis. Insets magnify the bulge and the epidermis. F, fraction of TUNEL-positive cells in K15-positivebulge cells 2 and 7 days after p14ARF activation; gray-shaded area shows the apoptosis levels in the epidermis (epi) of the same 2d-induced mice,as shown inFig. 1H.G, left, expression of endogenousp19ARF (red) in thebulge inmice after 2weeksof p14ARF induction; right, p53expression (white) andK15expression (red) in a consecutive section of the same follicle. *, P < 0.05; **, P < 0.005; NS, nonsignificant.






widespread alopecia, and then resilenced the transgene foranother 4 weeks; this did not result in hair regrowth (Fig. 6E).These findings indicate that the majority of bulge cells attain anonproliferative state that is not reversed after transgenesilencing.

p14ARF activation in skin hyperplasiaTo assess the effects of p14ARF on cells in a preneoplastic

state, we induced epidermal hyperplasia by treatment with thephorbol ester TPA (33, 34). We hypothesized that the compet-itive advantage of transgene-nonexpressing cells may allowhyperplasia to develop also in p14ARF-induced mice, but thatsenescence formation could be affected by this different tissuecontext. We treated mice with TPA for 4 days and thenactivated p14ARF for 2 weeks, maintaining TPA treatment.Indeed, hyperplasia occurred in both control and p14ARF

-induced mice, manifested by epidermal thickening and highbasal layer proliferation rates (Fig. 7A–C). Strikingly, in TPA-treated mice, p14ARF induction did not increase the fraction of

epidermal SAbGal-positive cells, nor the absolute number ofSAbGal-positive cells per field (Fig. 7D and E and Supplemen-tary Fig. S6); this, despite the presence of transgene expressingcells (Fig. 7F and Supplementary Fig. S6).We also did not detectan increase in SAbGal-positive cells after one week of p14ARF

induction in TPA-treated mice (Supplementary Fig. S6). Thesefindings indicate that upon p14ARF induction within TPA-induced hyperplasia, senescent cells are either very rapidlyeliminated or are not formed at all.

TPA is known to induce hair follicle stem cell activation (34);indeed, control TPA-treated mice displayed a dramaticincrease in hair follicle growth (Fig. 7A). In contrast, hairfollicles of p14ARF-induced mice did not enter anagen butinstead developed into thickened epidermal growths contain-ing proliferating cells (Fig. 7G and H). K15-positive cells wereno longer detected in most follicles, and where present weresurrounded by these growths (Fig. 7I). Thus, TPA cannotreverse the inactivation of hair follicle stem cells by p14ARF

induction, and these cells are mostly lost in this context.

0

20

40

60

80

100

p14

2wk on

4wk off

p14 Contp14Cont

% F

oll

icle

s w

ith

Ki6

7+

ce

lls

in b

ulg

e o

r b

elo

wCont (2+4wk)Cont (2wk)

6wk 6wk on + 4wk off

A

B

6wk2wk

E

K14 K15 Ki67

2wk 6wk

n=7n=4n=7n=4n=3

** 2wk on + 4wk off2wk on 6wk on

K5 p19

2wk + 4wk off+ 2wk off+ 1wk off

C

D

Cont

K14 K15 p53

2wk on + 4wk off

25 μm

25 μm

50μm

NS

**

Figure 6. Hair-follicle stem cells remain dysfunctional after p14ARF shutoff. A, Ki67 stain of hair follicles of mice in which p14ARF was activated for 2 or6 weeks (2wk, 6wk), or activated for 2 weeks and then resilenced for 4 weeks (2wk on þ 4wk off). Also shown are matched single-transgene controls. B,fraction of hair follicles containing Ki67-positive cells in the bulge or hair germ. C, stain of p53 (white) in the bulge region of indicated mice. D, stain ofendogenous p19ARF (red) in follicles of mice in which p14ARF was activated for 2 weeks and then silenced for 1, 2, or 4 weeks. E, images of mice in whichp14ARF was activated for 6 weeks (left) or activated for 6 weeks followed by 4 weeks in which doxycycline was removed (right). Matched control mice areshown on left in each image. Bottom images show close up magnification of skins (shaved). **, P < 0.005; NS, nonsignificant.






DiscussionRecent studies have convincingly established that cellular

senescence is a stress–response program functioning in vivo indiverse physiologic contexts, and acting prominently in tumorsuppression (2–4). Nevertheless, most basic aspects of thisprogram remain poorly characterized. Here, we describe anexperimental system that allows the generation and study ofsenescent cells in tissues of choice, relying on p14ARF, a centralactivator of this program.p14ARF induction in the adult epidermis led to p53 acti-

vation and caused apoptosis, growth arrest, and senescence,all seemingly initiated in parallel in different cell subsets.The rapidly occurring stage of apoptosis and proliferationarrest was transient, suggesting tissue adaptation. An above-basal level of apoptosis may persist also during longertransgene induction times, yet may be less apparent, as itdoes not occur synchronously as upon initial induction.Interestingly, proliferative inhibition was followed by epi-dermal hyperproliferation, most likely representing a non-cell-autonomous compensatory response to cell loss.

Whether this is induced by factors secreted by the p14ARF

-expressing cells remains to be studied.Senescent cells, as defined by SAbGal activity, were first

detected only one week after transgene activation. This isconsistent with previous reports, which showed that p19ARF

and p16Ink4a are activated during keratinocyte senescence andfunctionally contribute to this process (35–37).

Our finding that exogenous p14ARF expression activates theendogenous p16Ink4a and p19ARF genes reveals a novel positive-feedback loop promoting entry into senescence, and involvingboth the p53 and Rb pathways. The molecular circuitry medi-ating this loop, and the potential role played in it by knownregulators of the Ink4a locus, such as Polycomb, p63 and E2Fproteins (5), require further analysis. Our finding that p53mediates this positive loop stands in contrast to previousstudies reporting suppression of p16Ink4/p19ARF by p53 (38–40). Compensatory activation of these genes in situations ofp53 loss may occur through a distinct pathway.

The parameters determining the different outcomes ofp53 activation remain largely unknown (28). p14ARF and p53

TPA +TPA -

K5-rtTA / tet-p14TPA +TPA -

K5-rtTA / tet-p14tet-p14tet-p14

A

B

E

F

D

SAββGal

**

SA

βG

al-

po

sit

ive f

racti

on

TPA -

p14Cont

n=6n=3

NS

Cont p14

TPA +

n=6n=3

25 µm

K5-rtTA / tet-p14

C

Ki67

SAβGal

25 µm

Ki6

7-p

osit

ive f

racti

on

of

basal cell

s

TPA -

p14Cont Cont p14

TPA +

Ki67

**NS

n=6n=3n=5n=3

**

*

50 µm

G

I

H

0.5 mm

K14 K15 p14 50 μm

25 µm

Ki67

50 µm

K14 p14 50 μm

Figure 7. p14ARF activation during hyperplasia. A, H&E-stained skin sections of TPA-treated (TPAþ) or untreated (TPA�) control and p14-induced mice(2 weeks). Top, epidermis, bottom, whole skin. B, Ki67 expression in same skins. C, fraction of Ki67-positive cells in basal layer. D, SAbGal stain. E, fraction ofSAbGal-positive epidermal cells. F, p14ARF expression (white) in the epidermis of TPA-treated and untreated p14-induced mice. G, H&E stainshowing hair follicles in the same mice. H, Ki67 stain showing proliferation throughout the follicle of TPA-treated mice, contrasting with lack of Ki67 inthe bulge of untreated mice (arrow). I, K15-positive cells are eliminated from many follicles in TPA-treated mice; shown is an example of a follicle in whichremaining K15 cells are placed within a thickened follicle. Red stain in dermis is nonspecific. *, P < 0.05; **, P < 0.005; NS, nonsignificant.






protein levels varied among epidermal cells, and this variabilitymost likely contributed to outcome determination. However,the resistance of bulge cells to apoptosis indicates that cellidentity also plays a central role in choice of outcome.

Our detection of increased p16Ink4a and p19ARF levels in somemice early after transgene activation suggests that the initia-tion of the senescence program occurs rapidly in a subset ofcells. We found that continuous p14ARF expression wasrequired for these cells to complete the entry into senescence.This finding has interesting implications: when stress is lifted,damage is repaired, or potentially oncogenic signals aresilenced, cells that have already activated the senescenceprogram seem to be able to abort its completion and avoidthis irreversible fate.

Once formed, many senescent cells were retained in theepidermis for weeks after the silencing of p14ARF. This wassurprising due to the high cell turnover in the tissue. Senescentbasal cells are thus capable of "holding on" to their positionwithin the proliferating compartment. Furthermore, althoughthe numbers of SAbGal- and p53-positive cells did decreaseover time, the retention of many of these cells indicates thatsenescent epidermal cells do not trigger a rapidly actingefficient system for their removal. This stands in contrast toother previously described settings in vivo, including liverfibrosis and Ras-induced senescence, in which inflammatorycytokines recruit immune cells to rapidly remove senescentcells (13, 21, 22). We did not observe overt inflammation in theskins of p14ARF-activated mice (Supplementary Fig. S4), andonly small increases in cytokine expression were detected inthe tissue (data not shown). It has been shown that p53 in factsuppresses the senescence-associated secretory phenotype(SASP), and that p16Ink4a function does not influence it (41,42). Whether the retention of senescent cells is due to theabsence of a SASP or to a mechanism overriding it requires adetailed dissection of the signaling interactions in the tissue.

In the context of TPA-induced hyperplasia, despite thepresence of p14ARF-expressing cells, we did not detectincreased senescence. There are several possible explanationsfor this interesting finding: TPA could affect keratinocytes in amanner preventing their senescence, increased epidermal cellturnover during hyperplasia could lead to the rapid eliminationof transgene-expressing cells prior to their entry into senes-cence, or a rapid mechanism for directed senescent cellremoval could be activated in these conditions, potentiallymediated by TPA-induced inflammation (34). Additional studyis required to distinguish between these possibilities. Inter-estingly, we noted increased epidermal thickness in TPA-treated p14ARF-expressing mice due to basal cell expansion

(Supplementary Fig. S6), further emphasizing the complextissue dynamics induced by this transgene.

Hair follicle stem cells have been shown to be resistant toDNA damage-induced apoptosis due to reduced levels of p53activation in them (43). Here, we show that very littleapoptosis of these cells occurs even when p53 is highlyactive. The resistance to apoptosis thus seems to be aninherent trait of these stem cells. In contrast, p19ARF proteinlevels were abundant in the bulge, indicating that thepositive feedback loop activated by p14ARF is more power-fully active in this region than in the interfollicular epider-mis. Potentially, the nonproliferating status of these stemcells at telogen contributes to the apparent local accumu-lation of senescence. The persistence of nonfunctional hairfollicle stem cells in the tissue may mimic hair follicle aging.

Together, our findings provide novel insights into the work-ings of cellular senescence within the living tissue, and indicatethat senescent cells can be retained in organs for prolongedperiods, affecting their physiology.

Disclosure of Potential Conflicts of InterestNo potential conflicts of interest were disclosed.

Authors' ContributionsConception and design: R. Tokarsky-Amiel, N. Azazmeh, A. Helman, I. Ben-PorathDevelopment of methodology: R. Tokarsky-Amiel, N. Azazmeh, I. Ben-PorathAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): R. Tokarsky-Amiel, N. Azazmeh, A. Helman, Y. Stein,A. HassanAnalysis and interpretation of data (e.g., statistical analysis, biostatistics,computational analysis): R. Tokarsky-Amiel, N. Azazmeh, A. Helman, Y. Stein,A. Maly, I. Ben-PorathWriting, review, and/or revision of the manuscript: I. Ben-PorathAdministrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): A. HassanStudy supervision: I. Ben-Porath

AcknowledgmentsThe authors thank Yuval Dor and Eli Pikarsky for critical reviewing of the

article, members of the Ben-Porath laboratory for assistance in experimentation,and Norma E. Kidess-Bassir for histologic preparation. The authors also thankUri Gat and Valery Krizhanovsky for advice; Amir Eden, Konrad Hochedlinger,and Rodulf Jaenisch for assistance in generating the tet-p14 mice; Livnat Dotan,Joshua Riegelhaupt, and Boris Winterhoff for assistance in their early charac-terization; Stuart Yuspa for K5-rtTA mice; and Scott Lowe and Prem Premsrirutfor tet-shp53 mice.

Grant SupportThis study was supported by EU grant 205129 and an Israel Cancer Research

Fund RCDA grant.The costs of publication of this article were defrayed in part by the payment of

page charges. This article must therefore be hereby marked advertisement inaccordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Received September 24, 2012; revised January 30, 2013; accepted February 15,2013; published OnlineFirst February 19, 2013.

References1. Campisi J, d'Adda di Fagagna F. Cellular senescence: when bad

things happen to good cells. Nat Rev Mol Cell Biol 2007;8:729–40.

2. Kuilman T, Michaloglou C, Mooi WJ, Peeper DS. The essence ofsenescence. Genes Dev 2010;24:2463–79.

3. Rodier F, Campisi J. Four faces of cellular senescence. J Cell Biol2011;192:547–56.

4. Collado M, Serrano M. Senescence in tumours: evidence from miceand humans. Nat Rev Cancer 2009;10:51–7.

5. Kim WY, Sharpless NE. The regulation of INK4/ARF in cancer andaging. Cell 2006;127:265–75.

6. Zindy F, Quelle DE, RousselMF, Sherr CJ. Expression of the p16INK4atumor suppressor versus other INK4 family members during mousedevelopment and aging. Oncogene 1997;15:203–11.






7. Zindy F, Williams RT, Baudino TA, Rehg JE, Skapek SX, Cleveland JL,et al. Arf tumor suppressor promotermonitors latent oncogenic signalsin vivo. Proc Natl Acad Sci U S A 2003;100:15930–5.

8. Krishnamurthy J, Torrice C, RamseyMR, Kovalev GI, Al-Regaiey K, SuL, et al. Ink4a/Arf expression is a biomarker of aging. J Clin Invest2004;114:1299–307.

9. Cao L, Li W, Kim S, Brodie SG, Deng CX. Senescence, aging, andmalignant transformation mediated by p53 in mice lacking the Brca1full-length isoform. Genes Dev 2003;17:201–13.

10. Satyanarayana A, Wiemann SU, Buer J, Lauber J, Dittmar KE, Wus-tefeld T, et al. Telomere shortening impairs organ regeneration byinhibiting cell cycle re-entry of a subpopulation of cells. EMBO J2003;22:4003–13.

11. Schmitt CA, Fridman JS, YangM, LeeS, Baranov E, HoffmanRM, et al.A senescence programcontrolled byp53 andp16INK4a contributes tothe outcome of cancer therapy. Cell 2002;109:335–46.

12. Jackson JG, Pant V, Li Q, Chang LL, Quintas-Cardama A, Garza D,et al. p53-Mediated senescence impairs the apoptotic response tochemotherapy and clinical outcome in breast cancer. Cancer Cell2012;21:793–806.

13. Krizhanovsky V, Yon M, Dickins RA, Hearn S, Simon J, Miething C,et al. Senescence of activated stellate cells limits liver fibrosis. Cell2008;134:657–67.

14. Jun JI, Lau LF. The matricellular protein CCN1 induces fibroblastsenescence and restricts fibrosis in cutaneous wound healing. NatCell Biol 2010;12:676–85.

15. Lee KE, Bar-Sagi D. Oncogenic KRas suppresses inflammation-asso-ciated senescence of pancreatic ductal cells. Cancer Cell 2010;18:448–58.

16. Dimri GP, Lee X, Basile G, Acosta M, Scott G, Roskelley C, et al. Abiomarker that identifies senescent human cells in culture and in agingskin in vivo. Proc Natl Acad Sci U S A 1995;92:9363–7.

17. Paradis V, Youssef N, Dargere D, Ba N, Bonvoust F, Deschatrette J,et al. Replicative senescence in normal liver, chronic hepatitis C, andhepatocellular carcinomas. Hum Pathol 2001;32:327–32.

18. Wang C, Jurk D, Maddick M, Nelson G, Martin-Ruiz C, von Zglinicki T.DNA damage response and cellular senescence in tissues of agingmice. Aging Cell 2009;8:311–23.

19. Coppe JP, Desprez PY, Krtolica A, Campisi J. The senescence-associated secretory phenotype: the dark side of tumor suppression.Annu Rev Pathol 2009;5:99–118.

20. Baker DJ, Wijshake T, Tchkonia T, LeBrasseur NK, Childs BG, van deSluis B, et al. Clearance of p16Ink4a-positive senescent cells delaysageing-associated disorders. Nature 2011;479:232–6.

21. XueW, Zender L,MiethingC,DickinsRA,HernandoE, KrizhanovskyV,et al. Senescence and tumour clearance is triggered by p53 restorationin murine liver carcinomas. Nature 2007;445:656–60.

22. Kang TW, Yevsa T,Woller N, Hoenicke L,Wuestefeld T, DauchD, et al.Senescence surveillance of pre-malignant hepatocytes limits livercancer development. Nature 2011;479:547–51.

23. Weber JD, Taylor LJ, Roussel MF, Sherr CJ, Bar-Sagi D. Nucleolar Arfsequesters Mdm2 and activates p53. Nat Cell Biol 1999;1:20–6.

24. Ressler S, Bartkova J,NiedereggerH, Bartek J, Scharffetter-KochanekK, Jansen-Durr P, et al. p16INK4A is a robust in vivo biomarker ofcellular aging in human skin. Aging Cell 2006;5:379–89.

25. Beard C, Hochedlinger K, Plath K, Wutz A, Jaenisch R. Efficientmethod to generate single-copy transgenic mice by site-specificintegration in embryonic stem cells. Genesis 2006;44:23–8.

26. Diamond I, Owolabi T, Marco M, Lam C, Glick A. Conditional geneexpression in the epidermis of transgenic mice using the tetracycline-regulated transactivators tTA and rTA linked to the keratin 5 promoter.J Invest Dermatol 2000;115:788–94.

27. Premsrirut PK, Dow LE, Kim SY, Camiolo M, Malone CD, Miething C,et al. A rapid and scalable system for studying gene function in miceusing conditional RNA interference. Cell 2011;145:145–58.

28. Junttila MR, Evan GI. p53–a Jack of all trades but master of none. NatRev Cancer 2009;9:821–9.

29. Caldwell ME, DeNicola GM, Martins CP, Jacobetz MA, Maitra A,Hruban RH, et al. Cellular features of senescence during the evolutionof human and murine ductal pancreatic cancer. Oncogene 2011;31:1599–608.

30. Sherr CJ. Divorcing ARF and p53: an unsettled case. Nat Rev Cancer2006;6:663–73.

31. Ha L, Ichikawa T, Anver M, Dickins R, Lowe S, Sharpless NE, et al.ARF functions as a melanoma tumor suppressor by inducing p53-independent senescence. Proc Natl Acad Sci U S A 2007;104:10968–73.

32. Blanpain C, Fuchs E. Epidermal homeostasis: a balancing act of stemcells in the skin. Nat Rev Mol Cell Biol 2009;10:207–17.

33. Oskarsson T, Essers MA, Dubois N, Offner S, Dubey C, Roger C, et al.Skin epidermis lacking the c-Myc gene is resistant to Ras-driventumorigenesis but can reacquire sensitivity upon additional loss ofthe p21Cip1 gene. Genes Dev 2006;20:2024–9.

34. Doles J, Storer M, Cozzuto L, Roma G, Keyes WM. Age-associatedinflammation inhibits epidermal stem cell function. Genes Dev 2012;26:2144–53.

35. Keyes WM, Wu Y, Vogel H, Guo X, Lowe SW, Mills AA. p63 Deficiencyactivates a program of cellular senescence and leads to acceleratedaging. Genes Dev 2005;19:1986–99.

36. Vijayachandra K, Higgins W, Lee J, Glick A. Induction of p16ink4a andp19ARF by TGFbeta1 contributes to growth arrest and senescenceresponse in mouse keratinocytes. Mol Carcinog 2009;48:181–6.

37. Su X, Cho MS, Gi YJ, Ayanga BA, Sherr CJ, Flores ER. Rescue of keyfeatures of the p63-null epithelial phenotype by inactivation of Ink4aand Arf. EMBO J 2009;28:1904–15.

38. Stott FJ, Bates S, James MC, McConnell BB, Starborg M, Brookes S,et al. The alternative product from the human CDKN2A locus, p14(ARF), participates in a regulatory feedback loop with p53 and MDM2.EMBO J 1998;17:5001–14.

39. Yamakoshi K, Takahashi A, Hirota F, NakayamaR, IshimaruN, KuboY,et al. Real-time in vivo imaging of p16Ink4a reveals cross talk with p53.J Cell Biol 2009;186:393–407.

40. Leong WF, Chau JF, Li B. p53 Deficiency leads to compensatory up-regulation of p16INK4a. Mol Cancer Res 2009;7:354–60.

41. Coppe JP, Patil CK, Rodier F, Sun Y, Munoz DP, Goldstein J, et al.Senescence-associated secretory phenotypes reveal cell-nonauton-omous functions of oncogenic RAS and the p53 tumor suppressor.PLoS Biol 2008;6:2853–68.

42. Coppe JP, Rodier F, Patil CK, FreundA, Desprez PY,Campisi J. Tumorsuppressor and aging biomarker p16(INK4a) induces cellular senes-cence without the associated inflammatory secretory phenotype.J Biol Chem 2011;286:36396–403.

43. Sotiropoulou PA, Candi A, Mascre G, De Clercq S, Youssef KK,Lapouge G, et al. Bcl-2 and accelerated DNA repair mediates resis-tance of hair follicle bulge stem cells to DNA-damage-induced celldeath. Nat Cell Biol 2010;12:572–82.






2013;73:2829-2839. Published OnlineFirst February 19, 2013.Cancer Res Ronit Tokarsky-Amiel, Narmen Azazmeh, Aharon Helman, et al.

Induction in the EpidermisARFp14Dynamics of Senescent Cell Formation and Retention Revealed by

Updated version

10.1158/0008-5472.CAN-12-3730doi:

Access the most recent version of this article at:

Material

Supplementary

http://cancerres.aacrjournals.org/content/suppl/2013/02/19/0008-5472.CAN-12-3730.DC1

Access the most recent supplemental material at:

Cited articles

http://cancerres.aacrjournals.org/content/73/9/2829.full#ref-list-1

This article cites 43 articles, 14 of which you can access for free at:

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

Subscriptions

Reprints and

[email protected]

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

Permissions

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

.http://cancerres.aacrjournals.org/content/73/9/2829To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-12-3730

http://cancerres.aacrjournals.org/content/suppl/2013/02/19/0008-5472.CAN-12-3730.DC1

http://cancerres.aacrjournals.org/content/73/9/2829.full#ref-list-1

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/73/9/2829


dynamics of senescent cell formation and retention · using the extended depth of focus function...

Documents