DNA damage within

chromatin: break it to shape it

Philipp OberdoerfferNCI, Bethesda, MD

DNA Repair Interest Group – 2017

Damage RNA

DNA repair: guarding genome and epigenome integrity?

DNA repair

SIRT1:NAD+ dependent HDAC.

DNA damage-induced gene silencing at DSBsEctopic gene expression

Oberdoerffer et al, Cell 2008Singh…Oberdoerffer, J Exp Med 2013

Gene A Gene B Gene C

Kruhlak et al., Nature 2007, Shanbhag et al, Cell 2010

Impact on epigenomic integrity

(Epi)genomic stem cell maintenance.

DNA replicationResistance to DNA damage

O H

::. :

Chromosomal defects Cell death

DSB-induced chromatin reorganization

Repressive chromatin domains

Active chromatin domains

ATRATRATMATM

Chromatin remodeling

DNA damageReplication stress

RNAPol

(Transient) epigenetic change

DNAPol

Impact of DNA damage on chromatin organization

Epigenome roadmap

Higher level chromatin organizationdetermines cell identity and function

DNA damage as means to shape and/or modulate the epigenome?

Age

Oberdoerffer and Sinclair, “Reorganization of Chromatin Modifiers Hypothesis” Nature Reviews 2007

Environmental stressReplication

DSBs

DNA damage as a driver of epigenomic change in aging and cancer?

Epigenetic changes/noiseLinked to both aging and cancer

Chronic damage exposureRepair defects

Age

Environmental stressReplication

DSBs

DNA damage as a driver of epigenomic change in aging and cancer?

Chronic damage exposureRepair defects

Consequences of DSBs for transcriptional integrity in vivo?

Chronic DNA damage response as a means to shape chromatin?

IPpoIER

A mouse model for temporally and spatially controlled DSB induction

Kim et al., Nucleic Acids Res 2016

Transcriptional consequences upon T cell-specific DSB induction?

I-PpoI: Intron-encoded endonuclease from Physarum polycephalum

Cuts mouse genome at ~ 150 defined sites

ER GFPI-PpoISTOP IRES

Inducible, tissue-specific I-PpoI mouse model

Cre recombinase

ER GFPI-PpoI IRES

Tamoxifen IPpoIER

• I-PpoI causes gene repression specifically in DSB-containing genes.

• DSB-induced gene repression is transient and dependent on DNA damage signaling.

The mammalian transcriptome is sufficiently robust to accommodate

short term DSB exposure.

However…

Data from labs of David Sinclair, Shelley Berger and Rafa de Cabo

Cre control (16 mo old) I-Ppo/Cre (16 mo old)

• Mice exposed to several weeks of DSB induction early in life

• Evidence for epigenetic memory of DSB induction…

Can we test the impact of chronic DNA damage under physiological conditions?

Replication stress: a source of chronic, HR-associated DNA damage

Sources of replication stress (fragile sites)

Adapted from Gaillard…Aguilera, Nat Rev Cancer 2015

Nucleotide limitation

Stalled DNA PolymeraseHydroxyurea

G4s Transcription (long genes)

RNA:DNA hybrids (R-loops)

Chromatin structureReplisome

Aphidicolin

macroH2A1.2?

Impact on chromatin?Replication stress

ssDNAATRATR

DSBs ATMATM

BRCABRCA

MacroH2A1 – a unique yet underappreciated histone variant

Discovered by John Pehrson

Promotes repressive chromatin

1.1 variant binds ADP-ribose derivatives

1.1 variant linked to DDR/NHEJ

Interferes with melanoma progression via CDK8

Interferes with reprogramming

1.2 variant linked to HR/BRCA11.1 variant promotes

senescence

macroH2A KO without major phenotype

Two splice variants: 1.1 and 1.2

0

5

10

15

20

25

30

1990 1995 2000 2005 2010 2015 2020

Publ

icat

ions

/yea

r Failed attempts to demonstrate

role in Xi

0

20

40

60

80

100

1980 1990 2000 2010 2020

H2AZCENPAmacroH2A

Publ

icat

ions

/yea

r

Papers on histone variants

Pehrson, Luger, Bernstein, Ladurner, Gamble, Price, Buschbeck and others…

Variant-specific analyses! Macro-domain

H2A core

Enriched on Xi

The macroH2A1.2 variant promotes BRCA1-dependent DSB repair

Khurana et al., Cell Reports 2014

Implications for (epi)genome maintenance?

• RNAi screen identified macroH2A1.2-specific role in homology-directed DSB repair

Macro-domain

H2A core

Facilitates BRCA1 accumulation, end resection and HR

MacroH2A1.2 promotes dynamic chromatin condensation at DSBs

BRCA1BRCA1

macroH2A1.2

DSB

Andy Trana.k.a.

The Biologist

Jeongkyu Kima.k.a.

The Postdoc

Dave Sturgilla.k.a.

The Data-Pro

291.8 88.6147.1

AphidicolinUntreated

278.4 56.1

7.7

229.0 131.689.1

γH2AXmH2A1.2

ChIP peak coverage (in Mb)A role for macroH2A1.2 in replication stress? 40.3% overlap with mH2A1.2 (p < 0.001)

DNA polymerasestalling

K562 (CML)

Aphidicolin ChIP-SeqmH2A1.2γ-H2AX

1 Mb

–

+

–

+

mH

2A1.

2γ-

H2A

X

Aph:

Genes:

FRA9I

Relationship to sites of replication stress?

Common fragile sites (CFS)

FRA7FFRA9I

Aphidicolin

Perc

ent p

eak

cove

rage

25

20

15

10

0

5

NFS 1 2 >2

mH2A1.2

γ-H2AX

30

Break frequency at CFS

Replication stress drives macroH2A1.2 accumulation at fragile sites genome-wide

A system for locally defined replication fork arrest

macroH2A1.2 accumulates at artificial replication blocks

Replication

256 x LacO

DNAPol

mCherry-LacR

macroH2A1.2 accumulates at artificial replication stops

***

Enric

hmen

t at L

ac a

rray

EdU–EdU+

1

2

3

4

5

mH

2A1.

2+

48% 17%LacR

EdU–

EdU+

mH2A1.2

Beuzer…Almouzni, Cell Cycle 2014

LacR BRCA1

EdU+

(S phase)

EdU–

Link to the replication stress response?

RS-induced mH2A1.2 accumulation is DDR-dependent

– Aph+ Aph WT

– Aph+ Aph

mH2A1.2 KO

Peak

cov

erag

e [%

]

>2

Break frequency at CFS

20

15

10

0

5

NFS 1 2

γ-H2AX peak coverage FANCD2

DAPI

Aph/CDK1i+ 1h release

WT

1.2

KO

***

20

30

FAN

CD

2 fo

ci p

er c

ell

0

10

WT KO

% D2+: 62 70

0

0.5

1

1.5

2

2.5

3

3.5

NFS-1

NFS-2

FRA3B

FRA3He

FRA9I

FRA11G

–

Enric

hmen

t ove

r unt

reat

ed

**

Aph AphATRi

Role of macroH2A1.2 during replication stress

MacroH2A1.2 loss causes increased DNA damage at CFSs

Also by comet assay…

ATMATM

ATRATR

Can macroH2A1.2 protect from replication stress-induced damage?

70% 38%

LacR LacR-macro

***

γ-H

2AX+

γ-H

2AX

inte

nsity

at L

ac a

rray

[AU

]

5

10

15

20

LacR

γ-H

2AX

mer

ge

LacRLacRmacro

DNA damage response

?

LacR-macro

macromacro

256 x LacO

DNAPol

The macro domain protects from DNA damage at replication blocks

Molecular basis for macroH2A1.2-dependent protection from RS?

macroH2A1.2 facilitates BRCA1 accumulation upon replication stress

NS

0

10

20

30

40

50

si-ctrlsi-mH1.2

**

BRC

A1+

LacR

-foci

[%]

EdU+ EdU–

si-mH2A1.2

mCherryLacR

BRCA1

EdU+

si-control

macroH2A1.2 promotes BRCA1 accumulation at replication blocks

Fork specific effect?

ATRATRATMATM

RPARPA

RPARPA

RPARPA

mH2A1.2

BRCA1BRCA1?

Fragile region

Fragile region

–cl

ick Nascent

ForkStalledFork

inpu

tIP

BRCA1

PCNA

BRCA1

γ-H2AX

PCNA

H2AX

γ-H2AX

H2AX

WT WT KO WT KO

HU

EdU-label replication forks

Isolation of proteins on nascent DNA (iPOND)

macroH2A1.2 promotes BRCA1 recruitment to replication forks

GSTfragments

BRCA1Exon 11 ORF1 1863

A B C D E F

B E

v

CPT – + – + – + – +

5% input

GST

F-H2A

F-H2AZ

-Flag

GST-BRCA1

F-mH1.2

BRCA1 interacts with macroH2A1.2

ATRATRATMATM

RPARPA

RPARPA

RPARPABRCA1BRCA1

What mediates macroH2A1.2 accumulation upon replication stress?

mH2A1.2

No known macroH2A1.2 chaperones!

Buschbeck & Hake, Nat. Reviews, 2017

FACT remodels H2A/H2B dimers at replication forks

mH2A1.2

RPARPARPARPA

RPARPA

FACT

FACT subunits promote RS-associated mH2A1.2 accumulation at Lac arrays…

… and common fragile sites.

?

Reinberg, Diffley, Aguilera, Formosa, Stillman, others…

LacR

mH

2A1.

2

sh-RFP sh-SPT16 sh-SSRP1

0

10

20

30

40

50

60

70sh-RFPsh-SPT16sh-SSRP

EdU+ EdU–

% c

ells

with

mH

2A1.

2 at

arra

y

****

NSNS

• FACT (SUPT16H/SSRP1) complex is the main H2A/H2B chaperone at replication forks.

• FACT loss causes HR and replication defects.

FACT mediates macroH2A1.2 accumulation in a replication stress-dependent manner.

FACT interacts with macroH2A1.2/H2B dimer in vitro and in vivo.

With help from Luger lab

00.20.40.60.8

11.21.41.61.8 vehicle

HU

macroH2A1.2 ChIP

** * *

Enric

hmen

t ove

r unt

reat

ed

FRA – 3B 3H 9I 11G – 3B 3H 9I 11G

sh-RFP sh-SUPT16H

-H2AX

?

MacroH2A1.2 incorporation also depends on DNA damage signaling.

MacroH2A1.2 accumulation at fragile sites requires H2AX phosphorylation

ATRATRATMATM

RPARPARPARPA

RPARPA

Urbain Weyemi / Bill Bonner

MacroH2A1.2 incorporation requires H2AX phosphorylation.

• FACT (SUPT16H/SSRP1) complex is the main H2A/H2B chaperone at replication forks.

• FACT loss causes HR and replication defects.H2AX

HCT116H2AX KO

x

H2AX-S139Aor 0.00.20.40.60.81.01.21.41.61.8

Enric

hmen

t ove

r unt

reat

ed

9I

*****

16D

**

γ-H2AX mH2A1.2ChIP

FRA 2I 4D 9I 16D2I 4D

**** *** *

FACT mediates macroH2A1.2 accumulation in a replication stress-dependent manner.

FACT

mH2A1.2

The Lego problem…

Potential to shape chromatin during continued replication stress?

Serum Starvation

Minimal replication

Continuous replication drives persistent macroH2A1.2 accumulation

Dependence on replication? ✔

***

Replicative age causes mH2A1.2 accumulation at CFSs

Young

Old

Common fragile site

Replication

Replicative aging

20 PDs

Primary fibroblast

0

0.5

1

1.5

2 NFS-1NFS-2FRA15EFRA17EFRA10AFRA1IFRA3BFRA20EFRA10CFRA9I

starvedmacroH2A1.2 ChIP

PD 50 PD 70

Enric

hmen

t old

/you

ng

DDR

Replication stress

Senescence

Bartek, d’Adda di Fagagna and others

Consequences of macroH2A1.2 loss in primary cells

BRCA1BRCA1

Barrier to malignant transformation

HR defectsmacroH2A loss?

ATRATRATMATM

RPARPARPARPA

RPARPA

mH2A1.2

macroH2A loss promotes senescence

0

1

2

3

4

5

6

0 5 10 15

si-ctrlsi-mH2A1.2HU

Time in culture [d]

si-control

si-mH1.2

SA-

-Gal

SA-

-Gal

Popu

latio

n do

ublin

gs

DDR is activated

0

1

2

3

4

5

p16 p21

si-ctrl

si-mH2A1.2

Rel

ativ

e m

RN

A

0

0.2

0.4

0.6

0.8

1

1.2

si-ctrlsi-mH2A1.2

EdU

+ce

lls re

lativ

e to

con

trol

siRNA 2: ctrl ATM p53

****

Cell cycle arrest is DDR-dependent

siRNA 1:

ATRATRATMATMRPARPA

RPARPARPARPA

-H2AXmH2A1.2

FACT

?

BRCA1

Cumulative cell divisions

Repair Incomplete chromatin restauration

macroH2A1.2 accumulation

Fragile region

Senescence(primary cells)

Genome instability(Tumor cells)

Fragile region

DDR

macroH2A1.2 loss

Replication stress

Replication stress shapes a protective chromatin environment at fragile regions

Kim et al, Mol Cell, in press

Implications for the epigenome

Fragile regions / macroH2A1.2 domains

• A subset of chromatin domains are the result of chronic DNA damage exposure.

• Implications for age-associated chromatin reorganization and epigenomic dysfunction.

A more general role for macroH2A1.2 in HR and/or replication stress-associated genome maintenance?

Replication (stress)DNA damage

Chromatinreorganization

Modulation of DNA repair

Non-homologous end joiningReplication stress

Senescence

Outcome BHomologous recombination/ALT

Replication fork protection(Tumor) cell growth

Outcome A

Cell type and ageCell cycle, metabolic state

Feedback between DNA repair and chromatin modulates cell function

Alternative splicingmacroH2A1.2 macroH2A1.1

Over-expressed in cancer

Acknowledgements

Jeongkyu KimRobin SebastianEri HosoganeWill HannonAlex KruswickDavid Sturgill

Former:Simran KhuranaJinping LiuAlex Kruswick

Collaborators:Lab members:

NIH National Cancer InstituteCenter for Cancer Research

Bill Bonner (NCI)Urbain Weyemi

Sandy Burkett (NCI)

Karolin Luger (UC Boulder)Garrett Edwards

Yie Liu (NIA)Chongkiu Sun

Pei-Ji Chin