Buck Institute for Research on Aging

01 August, 2017

Research Advisory Committee on

Gulf War Veteran's Illnesses

Senescent cells and aging

Lawrence Berkeley National Laboratory

Disclosure:

I am a scientific co-founder of

UNITY Biotechnology

0

Aging = susceptibility to (chronic) disease

not a coincidence! caused by basic aging process(es)

memory loss

Neurodegeneration, Osteoporosis

Macular degeneration,

hearing loss

Heart disease

Vascular disease Sarcopenia,

frailty Diabetes,

metabolic syndrome Decreased

CANCER lung, kidney, etc function

Cellular senescence: a candidate basic aging process

What is cellular

senescence?

Cellular senescence, a complex stress

response

Irreversible

GROWTH

ARREST

Multi-faceted

SECRETORY

PHENOTYPE

Tri-

partite

pheno-

type

RESISTANCE to

APOPTOSIS

(epi)genomic

damage

metabolic

imbalances organelleoncogenic

stress mutations

Cellular senescence, a physiological

response

Irreversible

GROWTH

ARREST

Multi-faceted

SECRETORY

PHENOTYPE

Tri-

partite

pheno-

type

RESISTANCE to

APOPTOSIS

tissue repair embryonic wound healing development

Cellular senescence, a complex stress

response

Irreversible

GROWTH

ARREST

Multi-faceted

SECRETORY

PHENOTYPE

Tri-

partite

pheno-

type

RESISTANCE to

APOPTOSIS

Cellular senescence, an evolutionary

balancing act

Aging

phenotypes

Tumor

suppression

Age-related

disease (including

cancer)

Tissue

remodeling/

repair

Irreversible

GROWTH

ARREST

Multi-faceted

SECRETORY

PHENOTYPE

Tri-

partite

pheno-

type

RESISTANCE to

APOPTOSIS

How are senescent cells

defined?

-• • -~

What defines a senescent cell? cytokines, chemokines,

SA-Bgal

p16INK4a

persistent DNA damage foci

(DNA-SCARS/TIF)

heterochromatin

foci (SAHF)

SASP GROWTH ARREST

lamin B1 loss

HMGB1 loss/

secretion

0

GATA4 stabilization

growth factors, proteases'

eicosanoids

RIP

apoptosis

resistance

DAMPs

ROS

ps – there are no senescence-SPECIFIC markers

Dimri, PNAS, 1995; Beausejour, EMBO J, 2003; Narita, Cell, 2003; Rodier, Nature Cell, 2009;

Rodier, J Cell Sci, 2011; Freund, Mol Cell Biol, 2012; Kang, Science, 2015; Wiley, unpublished

When and where do senescent

cells occur?

Old, epidermis

Senescent cells increase with age in

many tissues

Human, non-human primates, rodents, zebrafish

skin, retina, liver, spleen, aorta, kidney, lung, etc.

young

SA-Bgal staining, old

human skin

old

Dimri et al., PNAS, 1995

-Q)

Senescent cells are present at sites of many

age-related pathologies

Venous ulcers, atherosclerotic plaques, arthritic joints,

COPD, visceral fat, AD brain, etc

Benign prostatic hyperplasia, pre-neoplastic lesions

pulmonary

artery SMCs astrocytes

AD brain

Noureddine et al., Circulation, 2011 Bhat et al, 2012, PLoS One 7:e45069.

Senescent cells ….

Present at the right time and place to drive

aging and multiple age-related diseases

How do senescent cells drive aging?

DO senescent cells drive aging?

--------­, • ', ~~ 0 • ~~ ,, • •• 0 . , I Q . 0 \ o o 0 1 • • oo• 0 0

oe oo• •

How might they do it?

SASP* DAMPs HMGB1

loss/ secretion

0

*SASP = senescence-associated secretory phenotype

destroys disrupts normal prevents stem promotes

tissues cell/tissue functions cell functions cancer

100 SCp2 cells alone

0

200

+ Presenescent 100 Fibroblasts

0

200

+ Senescent Fibroblasts

100

0

40 80 120 Days

Senescent cells have potent paracrine activities

on normal, premalignant and malignant cells

non-SEN cells

BrdU DAPI Merge

!25

0 !

M

100

!M

Veh

icle

"

non-SEN CM

SEN Non-SEN No Fb

promote malignant

phenotypes

SEN cells

BrdU DAPI Merge

!25

0 !

M

1

00

!M

Veh

icle

"

SEN CM

disrupt stem

cell functions

disrupt normal

structures, functions

Parrinello et al, J Cell Sci, 2005; Chintar et al, unpublished collaboration with

Andersen lab; Krtolica et al, PNAS, 2001; Coppe et al, PLoS One, 2010

•

aged tissue degenerating tissue

young tissue

senescent cell

dysfunctional cell

SASP

aged tissue

senescent cell

premalignant cell

neoplastic tissue

SASP

Are you depressed yet?

Strategies to combat aging phenotypes

and pathologies fueled by senescent cells

SASP*HMGB1 loss/

secretion

0

ILs

MMPs

MCPs

GFs

etc

Suppress secretory phenotype

What are the pathways and molecules

that drive the secretory phenotype?

(three pathways relevant to cancer and aging)

The DNA Damage Response (DDR) pathway

The p38MAPK-NF-kB pathway

The mTOR pathway

These are important pathways that are

required for tissue homeostasis

Drugs that suppress the SASP require

continuous dosing (a safe drug?)

Strategies to combat aging phenotypes

and pathologies fueled by senescent cells

SASP*HMGB1 loss/

secretion

0

ILs

MMPs

MCPs

GFs

etc

Suppress secretory phenotype

Kill/eliminate senescent cells

( 11--11---r-1---i-1 -1 )

p16-3MR (tri-modal reporter) mice

BAC containing murine INK4a locus inserted into mouse genome

3MR knock-in: downstream of p16INK4a promoter + inactivation of p14ARF

Mice have normal (diploid) copies of p16 and p14 genes

p16 Promoter renLuc mRFP HSV-tk

3MR: renilla luciferase; modified

red fluorescent protein; herpes

simplex virus thymidine kinase GCV = gancyclovir

GCV

P

HSV-Tk

GCV

P

phosphorylation

Low affinity for cellular TK

High affinity for viral TK

DNA CHAIN

TERMINATOR RIP

Demaria et al, Dev Cell, 2014; Laberge et al, CDD

PBS

OLD+ GCV

YOUNG+ PBS

YOUNG+ GCV

t

Se

ne

sce

nt

ce

lls c

an

be

elim

ina

ted

fro

m n

atu

rally

ag

ed

mic

e

Luminescence (A.U.) 10

00

00

Lu

min

es

ce

nc

e a

nd

GC

Vtr

ea

tme

nt

in a

gin

g p

16

-3M

R

80

00

0

60

00

0

40

00

0

20

00

0 0

mic

e

12

1

8

20

2

32

3 +

GC

VM

on

ths

pa

ralle

l a

ge-r

ela

ted

incre

ase

in

en

do

gen

ous p

16IN

K4a

, 3

MR

, IL

-6,

etc

; a

ll re

du

ce

d b

y G

CV

tre

atm

en

t

Senescent cells ….. Alzheimer's and Parkinson's disease

Atherosclerosis

Cardiovascular dysfunction

Cancer metastasis and recurrence

Chemotherapy (HAART) cardiotoxicity, fatigue

Cognitive decline/loss of neurogenesis

Diabetes

Myeloidlymphoid skewing

Osteoarthritis

Sarcopenia/frailty

Wound healing, tissue regeneration * published; * unpublished

Cellular senescence

Adverse effects of chemotherapy

Parkinson's disease and brain aging

Injury-induced osteoarthritis

Wound healing

Many cancer + other therapies

DNA damage

DNA damage senescence/SASP

DNA damaging therapies

long- and short-term adverse side effects

"Among adult survivors of childhood cancer, the prevalence of

adverse health outcomes was high ….. medical assessment

identified a substantial number of previously undiagnosed

problems that are more prevalent in an older population."

Clinical ascertainment of health outcomes among adults treated for childhood cancer

Hudson et al, JAMA, 2013

Ctrl ctrl doxorubicin paclitaxel temozolomidecisplatin

0

20000

40000

60000

80000

ctrl

doxorubicin

Paclitaxel

Cisplatin

Temozolomide

Lum

inesc

ence

(A

.U.)

p160

2

4

6

8

ctrl

doxorubicin

Paclitaxel

Cisplatin

Temozolomide

Targ

et

gene/A

ctin

(A

.U.)

SKIN

Cl

Cl -Cl

Cl -

"'""' 01 - r.r-1 Ocl - x&M

DNA damaging therapies persistent senescent

cells

Senescent cells promote metastases

MMTV-PyMT breast cancer

MMTV-PyMT cells

expressing firefly

luciferase

Inject into inguinal mammary fat pad

multifocal mammary adenocarcinomas

+ lung/liver metastases

CJ + + +++++ + l.,___-......... 1 ==---+l-+-+11 ......... 1 1--1

I I I I

1000

800

600

400

Senescent cells promote metastases in mice with

breast cancer

breast cancer

10 days

GCV

5 days 14-21 days

IMAGE DOXO

3 days

Control GCV

Cardiotoxicity often limits

chemotherapy

80

60

"#, 40

20

600

400

E a. .c

200

*

Heart rate

Cl CTRL

- DOXO - DOXOGCV

Cl CTRL

- DOXO - DOXOGCV

40

Cl CTRL 30

- DOXO - DOXOGCV

"#, 20

10

0

21 days post-doxo No effect 7 days post

Senescent cells contribute to chemotherapy-

induced cardiotoxicity

Chemotherapy-induced

loss of activity (fatigue?)

D DID ____.) -D

D �

~

3 d

ays

5d

ays

GC

V

Mic

e w

ith

ou

t b

rea

st

can

cer

tre

ate

d w

ith

ch

em

oth

era

py (

do

xo

rub

icin

)

Sa

line

(no

ca

nce

r)

7 d

ays

DO

XO

(10

mg

/kg

)

Me

tab

olic

ca

ge

s

3 d

ays

Behavior: mice + chemotherapy (no cancer)

DOXO + PBS DOXO + GCV NT

10.04 ± 3.28 9.02 ± 2.05 8.3 ± 2.22

1.56 ± 0.33 2.29 ± 0.95 2.22 ± 0.54

0.99 ± 0.71 2.046 ± 2.18 3.06 ± 3.79

1.36 ± 2.01 1.59 ± 1.46 2.01 ± 1.81

9.58 ± 13.12 37.22 ± 36.771 ± 14.29* 18.13*

1.86 ± 1.66 2.64 ± 3.34 7.026 ± 6.05

1.15 ± 1.12 1.09 ± 1.58 1.91 ± 1.19

61.46 ± 18.11 37.67 ± 15.81 27.53 ± 18.81*

6.94 ± 2.51 9.52 ± 2.13 11.17 ± 1.96*

Interaction with food hopper A (no significant uptake)

Interaction with water dispenser (significant uptake found)

Interaction with water dispenser (no significant uptake)

Interaction with wheel (>= 1 revolution)

Entered habitat (stable mass reading)

EFODA

TFODA

DWART

TWART

WHEEL

IHOME

THOME

LLNGE

SLNGE

% Total time

p-value: *<0.05; **<0.01; ***<0.001

N=5

Measurements at night

Interaction with food hopper A (significant uptake found)

Interaction with habitat (no stable mass reading)

Long lounge (> 60 sec, no non-XY sensor interactions)

Short lounge (5 - 60 sec, no non-XY sensor interactions) Demaria et al, in progress

Cellular senescence

Adverse effects of chemotherapy

Parkinson's disease and brain aging

Injury-induced osteoarthritis

Wound healing

1.5 # #

4 -:- 3 - I ::::,

::i

I ~

i3 <f <t ti 2 z E a:: 2 C: E .:::

(,)

I: 1.0 C:

~ 1 .:: ·- :il 1 .c - co C. :::, iu :::! .... .c 0 :::! 0

:::, O' q_<> O' q_<> !'S !'S o<::-..., o<::' v ..... v (0

c. 0.5 IL-8 MMP3 8 - * .-.. & ·•

I ::i ,i

I -=I:: :;: 4 a::

0.0 Ii c::

Normal PD ~2 •!IJ ;; -' g,,, :!!!

0 :iii 1)-

q_() tQ•

Senescence marker p16INK4a increases in

brains of human PD patients

#

p16 mRNA

Chintar, Demaria, Andersen et al; unpublished

Paraquat (PQ) causes

Parkinson's disease in

mice and humans

PQ causes astrocytes to

undergo senescence

Chinta et al, submitted

c: 15 ·-E ~ (1'

c. 10 V, ~ n, (1'

0:::

0 5 ~

(1' .c E ::::s 0 z

Rearing behaviour

#

PQ reduces motor neuron function;

restored by GCV Fig. 5 B

Cellular senescence

Adverse effects of chemotherapy

Parkinson's disease and brain aging

Injury-induced osteoarthritis

Wound healing

Surgical cut in anterior cruciate ligament

Figure Legends

Figure 1 Clearance of SCs by GCV treatment prevents the development of the

post-traumatic OA. (a) Scheme of experiment for b-g. p16-3MR mice undergoing

the anterior cruciate ligament transection (ACLT) of one hind limb to induce OA

were injected intra-articular with vehicle (Veh) or gancyclovir (GCV, 2 mM) for 2

weeks starting 3 weeks post-surgery. (b) Left, representative luminescence images of

ACLT mice after vehicle or GCV treatment. Right, quantification of the luminescence

(in arbitrary units, A.U.) at the indicated time (days) after ACLT surgery. (c-d)

Quantification of HMGB1-positive non-SCs in articular cartilage and the medial tibial

plateau joint score in p16-3MR mice, based on the OARSI scoring system (No

surgery, n=3; Sham surgery, n=5; Vehicle-treated, n=6; GCV-treated, n=7). (e)

Representative images of HMGB1 and p16INK3a expression (brown) by

immunohistochemistry and safranin O/methyl green from p16-3MR mice to evaluate

the pathological changes 4 weeks after ACLT, proteoglycan (red) and bone (blue). (f)

The percentage of weight placed on the operated limb versus contralateral control and

response time of mice after placement onto a 55°C hotplate on day 28 after ACLT

surgery. (g) Quantification of mRNA expression for p16INK4a, p21, MMP-13, IL-1! ,

Col II and Aggrecan in joints from no surgery, sham, and ACLT mice treated as

indicated (n=5 mice per group). In (b) to (g), data are averages ± SD. *p < 0.05, **p <

0.01, ***p < 0.001, as indicated. t test.

Formatted: Font:Times, Bold

Figure Legends

Figure 1 Clearance of SCs by GCV treatment prevents the development of the

post-traumatic OA. (a) Scheme of experiment for b-g. p16-3MR mice undergoing

the anterior cruciate ligament transection (ACLT) of one hind limb to induce OA

were injected intra-articular with vehicle (Veh) or gancyclovir (GCV, 2 mM) for 2

weeks starting 3 weeks post-surgery. (b) Left, representative luminescence images of

ACLT mice after vehicle or GCV treatment. Right, quantification of the luminescence

(in arbitrary units, A.U.) at the indicated time (days) after ACLT surgery. (c-d)

Quantification of HMGB1-positive non-SCs in articular cartilage and the medial tibial

plateau joint score in p16-3MR mice, based on the OARSI scoring system (No

surgery, n=3; Sham surgery, n=5; Vehicle-treated, n=6; GCV-treated, n=7). (e)

Representative images of HMGB1 and p16INK3a expression (brown) by

immunohistochemistry and safranin O/methyl green from p16-3MR mice to evaluate

the pathological changes 4 weeks after ACLT, proteoglycan (red) and bone (blue). (f) The percentage of weight placed on the operated limb versus contralateral control and

response time of mice after placement onto a 55°C hotplate on day 28 after ACLT

surgery. (g) Quantification of mRNA expression for p16INK4a, p21, MMP-13, IL-1! ,

Col II and Aggrecan in joints from no surgery, sham, and ACLT mice treated as

indicated (n=5 mice per group). In (b) to (g), data are averages ± SD. *p < 0.05, **p <

0.01, ***p < 0.001, as indicated. t test.

Formatted: Font:Times, Bold

weeks after surgery

1 2 3 4 5

vehicle or GCV

3 4 5 6

Figure Legends

Figure 1 Clearance of SCs by GCV treatment prevents the development of the

post-traumatic OA. (a) Scheme of experiment for b-g. p16-3MR mice undergoing

the anterior cruciate ligament transection (ACLT) of one hind limb to induce OA

were injected intra-articular with vehicle (Veh) or gancyclovir (GCV, 2 mM) for 2

weeks starting 3 weeks post-surgery. (b) Left, representative luminescence images of

ACLT mice after vehicle or GCV treatment. Right, quantification of the luminescence

(in arbitrary units, A.U.) at the indicated time (days) after ACLT surgery. (c-d)

Quantification of HMGB1-positive non-SCs in articular cartilage and the medial tibial

plateau joint score in p16-3MR mice, based on the OARSI scoring system (No

surgery, n=3; Sham surgery, n=5; Vehicle-treated, n=6; GCV-treated, n=7). (e)

Representative images of HMGB1 and p16INK3a expression (brown) by

immunohistochemistry and safranin O/methyl green from p16-3MR mice to evaluate

the pathological changes 4 weeks after ACLT, proteoglycan (red) and bone (blue). (f) The percentage of weight placed on the operated limb versus contralateral control and

response time of mice after placement onto a 55°C hotplate on day 28 after ACLT

surgery. (g) Quantification of mRNA expression for p16INK4a, p21, MMP-13, IL-1! ,

Col II and Aggrecan in joints from no surgery, sham, and ACLT mice treated as

indicated (n=5 mice per group). In (b) to (g), data are averages ± SD. *p < 0.05, **p <

0.01, ***p < 0.001, as indicated. t test.

Formatted: Font:Times, Bold

sham vehicle GCV

Jeon, Elisseeff; unpublished

Surgical cut in anterior cruciate ligament:

eliminating senescent cells restores function

% w

eig

ht 120% *** ***

100%

80%

60%

40%

no

surgery

sham veh GCV

ACLT

Cellular senescence, an evolutionary

balancing act (why did the SASP evolve?)

Aging

phenotypes

Tumor

suppression

Age-related

disease (including

cancer)

Tissue

remodeling/

repair

Irreversible

GROWTH

ARREST

Multi-faceted

SECRETORY

PHENOTYPE

Tri-

partite

pheno-

type

RESISTANCE to

APOPTOSIS

Cellular senescence

Adverse effects of chemotherapy

Parkinson's disease and brain aging

Injury-induced osteoarthritis

Wound healing

Cellular senescence is induced

during wound healing

3MR 3MR

WT

Wound healing: 4 days

after punch biopsy

Induction of p16INK4a, 3MR, IL-6 expression

Demaria et al,Dev Cell, 2014

+ ----/--~ ----

l

Senescent cells are present transiently

during wound healing

0

2000

4000

6000

8000

10000

12000

Lu

min

es

ce

nc

e (

A.U

.)

3MR

3MR GCV

d1 d3 d5 d7 d8

..... 3MR + GCV

100 I ..... 3MR + GCV

100 ..... 3MR ..... 3MR

~ ~ ..... WT+ GCV

WT Q) Q) N N (I) 50 (I) 50

"C "C C: C: ::J ::J

~ ~ 0 0

2 4 6 8 10 2 4 6 8 10

..... 3MR+ I • • • • GCV 1iOD ..... Wf + GCV -~ 0 -!Ill)

N (I) ,5D

"C C ::J 0 s:

0 2 4 16, 8 10

WOUND HEALING IS RETARDED BY ELIMINATING

SENESCENT CELLS

GCV 0-5 days after biopsy

female mice

100· ..... 3MR

- ---- 3,MR + GCV ~ 0 - ..... 3,MR + GCV + PDGF-M G) N en 50

"'C C: ::J 0

~

0 . 3 6 ·12.

Topical PDGF-AA topical rescues slow wound healing

in GCV-treated 3MR mice

Punch i.p. GCV / topical PDGF-AA

Day 0 1 2 3 4 5 6 7 8 9 10 11

Cellular senescence, an evolutionary

balancing act

Aging

phenotypes

Tumor

suppression

Age-related

disease (including

cancer)

Tissue

remodeling/

repair

Irreversible

GROWTH

ARREST

Multi-faceted

SECRETORY

PHENOTYPE

Tri-

partite

pheno-

type

RESISTANCE to

APOPTOSIS

Replace SnC

factors Kill SnCs

Cellular senescence, a complex stress

response

Irreversible

GROWTH

ARREST

Multi-faceted

SECRETORY

PHENOTYPE

Tri-

partite

pheno-

type

RESISTANCE to

APOPTOSIS

THANKS! Present lab members

Nick Aguirre Past lab members

Fatouma Alimirah Christian Beausejour (Montreal U)

Nate Basisty Marco Demaria (ERIBA)

Albert Davalos Pierre Desprez (CA Pacific Med Cntr)

Jose Domingo-Lopez Peter de Keizer (Erasmus U)

Chisaka Kuehnemann Francis Rodier (Montreal U)

Abhijit Kale

Clare Kim Collaborators Chandani Limbad PPG: Jan Vijg, Yousin Suh (Einstein); Jan

Su Liu Hoeijmakers (Erasmus); Paul Hasty (UTHSCSA) Kevin Perrott PPG: Jim Kirkland, Jan van Deursen, Tamara Chris Wiley Tchkonia, Darren Baker, Nathan LeBrasseur

Ying Zou (Mayo), Yuji Ikeno (UTHSCSA), Rich Miller (U MI)

Jennifer Elisseeff (Johns Hopkins U)

Pete Nelson (Fred Hutch)

Steve Yannone, Paul Yaswen, Cilla Cooper (LBNL)

Julie Andersen, Pankaj Kapahi, Simon Melov, Brad

Gibson, Arvind Ramachandran, Birgit Schilling (Buck Inst)

Eiji Hara, Naoko Ohtani (Osaka U, Tokyo U)