Histone H2AX promotes neuronal health by controllingmitochondrial homeostasisUrbain Weyemia, Bindu D. Paula, Deeya Bhattacharyaa, Adarsha P. Mallaa, Myriem Boufraqechb, Maged M. Harraza,William M. Bonnerc, and Solomon H. Snydera,d,e,1

aThe Solomon H. Snyder Department of Neuroscience, Johns Hopkins University School of Medicine, Baltimore, MD 21205; bEndocrine Oncology Branch,National Cancer Institute, National Institutes of Health, Bethesda, MD 20892; cDevelopmental Therapeutics Branch, Laboratory of Molecular Pharmacology,National Cancer Institute, National Institutes of Health, Bethesda, MD 20892; dDepartment of Psychiatry and Behavioral Sciences, Johns Hopkins UniversitySchool of Medicine, Baltimore, MD 21205; and eDepartment of Pharmacology and Molecular Sciences, Johns Hopkins University School of Medicine,Baltimore, MD 21205

Contributed by Solomon H. Snyder, February 27, 2019 (sent for review December 3, 2018; reviewed by Olga A. Martin and Robert Schwarcz)

Phosphorylation of histone H2AX is a major contributor to efficientDNA repair. We recently reported neurobehavioral deficits in micelacking H2AX. Here we establish that this neural failure stems fromimpairment of mitochondrial function and repression of the mito-chondrial biogenesis gene PGC-1α. H2AX loss leads to reduced levelsof the major subunits of the mitochondrial respiratory complexes inmouse embryonic fibroblasts and in the striatum, a brain regionparticularly vulnerable to mitochondrial damage. These defects aresubstantiated by disruption of the mitochondrial shape in H2AX mu-tant cells. Ectopic expression of PGC-1α restores mitochondrial oxida-tive phosphorylation complexes and mitigates cell death. H2AXknockout mice display increased neuronal death in the brain whenchallenged with 3-nitropronionic acid, which targets mitochondria.This study establishes a role for H2AX in mitochondrial homeostasisassociated with neuroprotection.

mitochondrial homeostasis | neuroprotection | oxidative stress |histone H2AX | DNA repair

Genome integrity is maintained by a number of tightly regu-lated pathways that lead to efficient DNA repair in response

to endogenous or exogenous genotoxic agents (1). Compromis-ing DNA repair or increasing mutation loads can lead to a broadrange of human diseases, including those related to immunedeficiency, cancer, inflammation, aging, and developmental dis-orders affecting the nervous system (2–5). During early devel-opment, especially when progenitor cells expand and differentiateinto mature neurons, efficient DNA repair machinery is crucial (1).Brain is uniquely vulnerable to genomic instability (1, 6, 7). Defectsin DNA repair genes often lead to age-associated neurologicaldisorders, but underlying mechanisms are still obscure (1, 7–9).Several neurological disorders originating from deficient DNA re-pair often involve oxidative stress (1, 6, 10). DNA repair proteinswhose deficiencies are associated with oxidative stress in the brain(3, 11) include ataxia-telangiectasia mutated (ATM) kinase, meioticrecombination 11 (MRE11), Nijmegen breakage syndrome 1(NBS1), aprataxin (APTX), and tyrosyl-DNA phosphodiesterase 1(TDP1) (1, 3). Murine ATM models facilitate clarification of linksbetween genomic instability and impaired redox homeostasis.When ATM is absent or mutated, the cell suffers from faulty DNArepair and genomic instability and harbors elevated levels of re-active oxygen species (ROS) originating from the impairment ofROS-sensing functions of ATM (12–14). These deficiencies lead toAtaxia telangiectasia (AT). We have shown that diminution of ROSreduces phenotypic damage associated with AT (15). ATM elicitsDNA repair by phosphorylating the histone variant H2AX fol-lowing double-stranded DNA breaks (16). This process is sub-stantiated by the well-established role of H2AX in DNA damagerepair (5). Similarities between ATM and H2AX are evident in thephenotypes of their knockout mouse models. In both instances,males are sterile, and there is increased genomic instability evi-denced by abnormalities in chromosome structure, immunodefi-ciency, and enhanced radiosensitivity (12, 17). These similarities

may reflect well their common roles in DNA repair. We recentlyprovided evidence that H2AX facilitates redox homeostasis bycontrolling an NRF2-regulated antioxidant response pathway (18).Mitochondrial defects are a major source of impaired redox ho-meostasis and are often associated with age-related neurologicaldiseases (19, 20). A major risk factor for several of these diseases isthe concomitant dysfunction of mitochondrial activity leading toincreased ROS and decreased DNA repair (21). Here we show thatH2AX loss leads to substantially diminished mitochondrial activityassociated with disturbed mitochondrial shape. These defects re-flect repression of the mitochondrial biogenesis gene peroxisomeproliferator-activated receptor gamma coactivator 1-aplha (PGC-1α) and the oxidative phosphorylation subunits. Ectopic expressionof PGC-1α partially rescued cells from mitochondrial defects andrestored toxicity associated with mitochondrial damage. We alsoreport that H2AX mutant mice were uniquely vulnerable to mito-chondrial damage in the brain. These findings identify histoneH2AX as a key regulator of mitochondrial homeostasis whichpromotes neuroprotection and clarify links between genomic in-stability and redox homeostasis in neurodegeneration and aging.

Results and DiscussionHistone H2AX Deficiency Leads to Repression of Mitochondrial BiogenesisGenes and Impairment of Oxidative Phosphorylation. Deficits of DNArepair genes are associated with diverse forms of neurodegeneration(1). Neurologic disorders arising from deficient DNA repair involveoxidative stress as well as the inability to repair oxidative DNA

Significance

Histone H2AX elicits proper DNA repair through its phosphory-lation by ataxia-telangiectasia mutated (ATM) kinase. While ATMsenses reactive oxygen species, the role of H2AX in the mainte-nance of redox homeostasis remains unknown. Here we establishthat H2AX deletion leads to impairment of mitochondrial func-tion and repression of the mitochondrial biogenesis gene PGC-1α.Restoring PGC-1α abrogates mitochondrial deficits and mitigatescell death. This study unveils a role for H2AX in mitochondrialhomeostasis associated with neuroprotection.

Author contributions: U.W. and S.H.S. designed research; U.W., B.D.P., D.B., A.P.M., M.B.,and M.M.H. performed research; W.M.B. contributed new reagents/analytic tools; U.W.and S.H.S. analyzed data; and U.W. and S.H.S. wrote the paper.

Reviewers: O.A.M., Peter MacCallum Cancer Centre; and R.S., University of MarylandSchool of Medicine.

Conflict of interest statement: U.W. and O.A.M. are coauthors on a 2016 article; they didnot collaborate directly on the paper. The authors declare no competing financialinterests.

Published under the PNAS license.1To whom correspondence should be addressed. Email: ssnyder@jhmi.edu.

This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.1073/pnas.1820245116/-/DCSupplemental.

Published online March 25, 2019.

www.pnas.org/cgi/doi/10.1073/pnas.1820245116 PNAS | April 9, 2019 | vol. 116 | no. 15 | 7471–7476

NEU

ROSC

IENCE

Dow

nloa

ded

by g

uest

on

Aug

ust 2

4, 2

020

H2AX WT

H2AX KO

0.00

0.05

0.10

0.15

Rel

ativ

e pr

otei

n am

ount

**

POLRMT

TFB2MTFA

M

PPARGC1B0.000

0.002

0.004

0.006

0.008

0.010

mR

NA

leve

ls

H2AX KOH2AX WT

***

***

***

A

F

H2AX

H2AX WT H2AX KO

GAPDH

PGC1⍺

15 kDa

130 kDa

38 kDa

MEFs

H2AX WT H2AX KO

H2AX15 kDa

CV-ATP5A

CII-SDHB

CIII-UQCRC2CIV-MTCO1

CI-NDUFB8

55 kDa

37 kDa

48 kDa

25 kDa15 kDa

GAPDH38 kDa

MEFs

H

CIII-UQCRC2

H2AX WT H2AX KO

H2AX15 kDa

CV-ATP5A

CII-SDHB

CIV-MTCO1

CI-NDUFB8

55 kDa

37 kDa

48 kDa

25 kDa

15 kDa

GAPDH38 kDa

C

E

H2AX WT H2AX KO

PGC1⍺

H2AX

GAPDH

BRAIN: STRIATUM

15 kDa

130 kDa

38 kDa

H2AX WT

H2AX KO

0.0

0.2

0.4

0.6

Rel

ativ

e pr

otei

n am

ount

***

ComplexV

Complex IV

Complex III

Complex II

Complex I

0.0

0.5

1.0

1.5

2.0

Rel

ativ

e pr

otei

n am

ount

H2AX KOH2AX WT

**

* *

*****

Complex V

ComplexIV

Complex III

Complex II

ComplexI

0.0

0.5

1.0

1.5

Rel

ativ

e pr

otei

n am

ount

H2AX KOH2AX WT*

*

*

****

B

D

BRAIN: STRIATUM

PGC1⍺ (MEFs)

PGC1⍺ (Striatum)

G

I

Fig. 1. Histone H2AX depletion elicits pronounced diminution of mitochondrial biogenesis genes and reduced expression of OXPHOS complexes. (A–D)H2AX deletion reduces PGC-1α, a major mitochondrial biogenesis gene. Protein lysates from wild-type and H2AX knockout MEFs, as well as lysates from thestriatum, were processed for immunoblot detection of H2AX and PGC-1α. Actin was used as a loading control. (A) Representative image of PGC-1α expressionin MEFs and (B) quantification. (C) Representative image of PGC-1α expression in the striatum and (D) quantification. Data are means ± SEM (n = 3). Statisticalsignificance was determined by a two-tailed, unpaired Student’s t test. (E) Analysis of POLRMT, TFB2M, TFAM, and PPARGC1B transcript levels in wild-typeand H2AX knockout MEFs by RT-PCR. Expression values are relative fold changes for gene transcripts normalized to GAPDH. Error bars represent SEM (n = 3).(F and G) Western blot analysis of OXPHOS subunits in wild-type and H2AX knockout MEFs. We performed the immunoblot detection using total OXPHOSHuman WB Antibody Mixture, enabling concomitant analysis of main proteins of each complex in the electron transport chain, including complex I subunitNDUFB8, complex II subunit 30 kDa, complex III subunit Core 2, complex IV subunit I, and ATP synthase subunit alpha; (F) representative image; (G) quan-tification. Data are means ± SEM (n = 3). (H and I) Western blot analysis of the OXPHOS subunits in the striatum of 4-mo-old wild-type and H2AX knockoutmice; (H) representative image; (I) quantification. Data are means ± SEM (n = 3; *P < 0.05; **P < 0.001; ***P < 0.0001).

7472 | www.pnas.org/cgi/doi/10.1073/pnas.1820245116 Weyemi et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 2

4, 2

020

lesions (6, 8). These features are exemplified in AT, a form of pro-gressive cerebellar degeneration associated with oxidative stress (12,14). We have shown that diminution of ROS reduces phenotypicdamage associated with AT (15). ATM kinase elicits DNA repair byphosphorylating the histone variant H2AX following double-stranded DNA breaks, which is consistent with the well-establishedrole of H2AX in DNA damage repair (5). We observed neuro-behavioral defects and impaired redox disposition in mice with ge-netic deletion of H2AX (18). In the present study, we report a rolefor H2AX in promoting mitochondrial homeostasis and mediatingneuronal health.Most mitochondrial proteins are encoded in the nucleus (22).

Examples include basal transcription factors such as TFAM,TFB2M, TFB1M, and POLRMT, as well as the mitochondrialbiogenesis gene PGC-1α (22, 23). PGC-1α is a transcriptioncoactivator that regulates the expression of TFAM, TFB2M,TFB1M, and POLRMT, as well as the oxidative phosphorylationcomplex subunits (OXPHOS) through interaction with thetranscription factor nuclear respiratory factor 1 (NRF1) (24, 25).It is well established that mitochondrial biogenesis and respira-tion are stimulated by PGC-1α through induction of NRF1 andNRF2 gene expression (26). Changes in chromatin configurationor alteration of chromatin-based DNA repair pathways are fac-tors in regulating transcription factors (27, 28), but regulatorymechanisms are often unclear. As histone H2AX is a key

component of chromatin and an important player in DNA repairpathways, we speculated that loss of H2AX can alter chromatinand DNA damage responses, thereby impacting mitochondrialbiogenesis and eliciting oxidative stress. Deletion of H2AX led topronounced diminution of PGC-1α protein in mouse embryonicfibroblasts (MEFs) as well as the brain (Fig. 1 A–D). Similardecreases occurred for the PGC-1α targets, TFAM, TFB2M, andPOLRMT, as well as for PPARGC1B, another homolog of PGC-1α (Fig. 1E). Western blot analysis revealed substantial decreasesof the key subunits of the five OXPHOS complexes in bothH2AX-deficient MEFs and the brains of mutant mice (Fig. 1 F–I). The reduction in the OXPHOS complexes expression seemsmore pronounced in the striatum, since only a partial depletionof the OXPHOS complexes I and II was observed in the cortex(SI Appendix, Fig. S1A). Similar pattern of OXPHOS expressionwas detected in peripheral tissue such as the liver (SI Appendix,Fig. S1B), suggesting an increased vulnerability of the striatum toH2AX loss-induced mitochondrial damage. Taken together,H2AX deletion impairs expression of PGC-1α, the master reg-ulator of mitochondrial biogenesis, leading to alteration of theprincipal components of mitochondrial function.H2AX knockout cells displayed disorganized mitochondrial

cristae and slightly enlarged mitochondria (Fig. 2 A and B).Cristae are functionally dynamic compartments, serving as sitesfor OXPHOS complexes. A substantial amount of the main

A

2.37

34.77

67.38

99.98

132.6

0

165.0

00

50

100

150

200

250

Time (min)

OC

R (p

mol

/min

)/ C

ell n

umbe

r

H2AX WTH2AX KO

Oligomycin

FCCPAnt A / Rot

C

H2A

X K

O

500 nm

H2A

X W

T

c1

d1

STRIATUM

500 nm

MEFa1

b1500 nm

500 nm

B

Fig. 2. H2AX deletion leads to disorganized cristae, enlarged mitochondria, and impairment of oxidative phosphorylation. (A) Illustrative images of mi-tochondrial shape of wild-type and H2AX knockout cells in MEFs (a1, b1) and in the striatum (c1, d1) using TEM. Arrows in a1 and b1 indicate mitochondriacristae. Images were taken at 33,000×. (B) Diameter of 30 individual mitochondria in both wild-type and H2AX knockout striatum were quantified. H2AX KOcells display 20% increase in the size of their mitochondrial diameter; **P < 0.001. (C) Significantly higher baseline mitochondrial respiration (0 min to 34min), ATP production (34.77 min to 70 min), and maximal respiration and sparse respiratory capacity (70 min to 120 min) are observed in H2AX wild-type MEFscompared with H2AX knockout cells. These measurements were assessed using Seahorse MitoStress Kit (n = 5).

Weyemi et al. PNAS | April 9, 2019 | vol. 116 | no. 15 | 7473

NEU

ROSC

IENCE

Dow

nloa

ded

by g

uest

on

Aug

ust 2

4, 2

020

subunits of the complex III and ATP synthase, and the cyto-chrome c, are stored in the cristae (29). Shapes of cristae andOXPHOS function are closely linked with direct impact on cel-lular metabolism (29). The loss of cristae in H2AX mutants fitswith the diminished levels of OXPHOS subunits in mutant cells.We monitored OXPHOS function, employing the Seahorsetechnique. We observed diminished baseline oxygen consump-tion rate (OCR) in H2AX mutant cells, as well as reduced ATPproduction and impaired spare respiratory capacity (Fig. 2C).

Ectopic Expression of PGC-1α Restores OXPHOS Complexes andReverses Mitochondria Damage-Induced Cytotoxicity. OverexpressingPGC-1α in the H2AX knockout cells partially restored OXPHOSsubunit levels (Fig. 3A) and diminished cytotoxicity elicited by themitochondrial complex II inhibitor 3-nitropropionic acid (3-NP)(Fig. 3B). Enhanced mitochondrial functions, especially those in-volving the respiratory oxidative phosphorylation complexes, areassociated with PGC-1α. PGC-1α is a master integrator ofcellular signals that govern mitochondrial biogenesis, oxidative

A

BGAPDH

CV-ATP5A

CIII-UQCRC2

55 kDa

48 kDa

CII-SDHB25 kDa

CI-NDUFB815 kDa

CIV-MTCO137 kDa

WT KO+PGC1KO

PGC1⍺

38 kDa

130 kDa

D

C

E

Fig. 3. Ectopic expression of PGC-1α restores OXPHOS subunits and protects from the cytotoxicity induced by mitochondrial damage. (A) Immunoblot analysis ofOXPHOS subunits in H2AX wild-type cells (H2AX WT), H2AX knockout cells (H2AX KO), and H2AX knockout cells in which PGC-1α expression was ectopically andtransiently restored for 48 h. Actin was used as a loading control. Both wild-type and knockout cells were transfected with control vector. Most of the subunits werepartially restored, except for complex IV and complex V. (Left) Representative image. (Right) Quantification. (B) Cells deficient in H2AX are vulnerable to3-nitropropionic acid (3-NP), an inhibitor of OXPHOS complex II. This cytotoxicity was reversed by ectopic expression of PGC-1α in H2AX knockout cells. Viable cellswere quantified using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay. Cells were treated with increasing doses of 0, 0.1, 1, and 2 mM 3-NPfor 48 h before MTT was added. Data are means ± SD; n = 3. Statistical significance was determined by a two-tailed, unpaired Student’s t test. (C–E) Mitochondrialcomplex inhibitors induce lower survival in H2AX mutant cells, and H2AX ectopic expression mitigates these cytotoxic effects. (C and D) Parental cells (H2AX WT),H2AX knockout cells (H2AX KO), and H2AX knockout cells in which H2AX expression was restored (REV) were treated with increasing concentrations of mitochondrialcomplex I inhibitors 1-methyl-4-phenylpyridinium (MPP+) and rotenone. Cell survival was estimated 48 h posttreatment using MTT assay. Cells were treated withincreasing doses of 0, 10, 50, and 100 μM MPP+, or with 0, 0.1, 1, and 10 nM rotenone for 48 h before MTT was added. (E) Cell vulnerability to 3-nitropropionic acid(3-NP) was analyzed usingMTT assay. Cells were treated with increasing doses of 0, 0.1, 1, and 2mM 3-NP for 48 h beforeMTTwas added. Data are means ± SD; n = 3.Statistical significance was determined by a two-tailed, unpaired Student’s t test; *P < 0.05; **P < 0.001; ns, nonsignificant.

7474 | www.pnas.org/cgi/doi/10.1073/pnas.1820245116 Weyemi et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 2

4, 2

020

phosphorylation, adaptive thermoregulation, and fatty acid bio-synthesis/degradation (22, 30).To further assess the link between the loss of H2AX and mi-

tochondrial damage, we explored effects of additional inhibitorsof mitochondrial complexes on the viability of cells deficient forH2AX. H2AX knockout cells were vulnerable to inhibitors ofmitochondrial OXPHOS complexes I and II with toxicity re-versed by overexpressing H2AX (Fig. 3 C–E). These observa-tions show that the deficits observed in H2AX mutants are notthe result of other unrelated differences between wild-type andmutant cells. Taken together, these findings establish that themitochondrial defects induced by H2AX deletion lead to in-creased cytotoxicity which can be reversed by restoring markersof mitochondrial biogenesis. Histone H2AX is a key componentof a cluster of proteins involved in genome stability and chro-matin remodeling following DNA double-strand breaks. Exam-ples include proteins such as ATM, NBS1, RAD50, and 53BP1,

among others. How changes in chromatin dynamic affect themetabolism of mitochondria via regulation of PGC-1α and itstranscriptional targets is unclear.

H2AX Deficiency Increases Vulnerability to 3-Nitropropionic Acid ThatTargets Mitochondria in the Brain: Role in Neuroprotection. Wechallenged wild-type and H2AX mutant mice with 3-NP as il-lustrated in Fig. 4A. In H2AX mutants, acute 3-NP treatmentresulted in a 50% loss of striatal dopaminergic nerve terminals asmeasured by immunostaining and immunoblot analyses of tyro-sine hydroxylase (TH) expression (Fig. 4 B–D). These observa-tions were substantiated by signs of systemic mitochondrialimpairment in H2AXmutant mice and in mice treated with 3-NPas revealed by measurement of body temperature (Fig. 4E). Thedamages seem limited to DA neurons, since other populationsremain relatively unaffected, as indicated by unchanged NeuNimmunoreactivity (SI Appendix, Fig. S2). Several models ofneurodegeneration employ drugs that target mitochondrialfunction (31, 32). For instance, PGC-1α−deficient mice areuniquely vulnerable to 1-methyl-4-phenyl-1,2,3,6-tetrahydropyr-idine−induced degeneration of dopamine neurons in the sub-stantia nigra and striatum (25). Overexpressing PGC-1α protectsneuronal cells from oxidative stress and cell death (25). Ourfindings that H2AX knockout mice exhibit increased vulnerabilityto 3-NP in the striatum suggest enhanced neurodegeneration inmice lacking H2AX following damage to mitochondria, furthersubstantiating evidence that H2AX promotes neuronal healthvia control of mitochondrial homeostasis.In summary, the present study demonstrates that H2AX, a

histone variant and classic DNA repair protein, promotes mi-tochondrial biogenesis and mediates responses to neurotoxicdrugs targeting mitochondria. In addition, we show that cellslacking H2AX have impaired oxygen consumption, as well asdisturbances in mitochondrial shape. Histone variant H2AX hasbeen studied primarily as a mediator of DNA repair (5). Thepresent study extends the role of this protein to the arena ofmitochondrial function and neuronal health. Our findings revealhow alterations in the nuclear genome influence mitochondrialactivity and cell metabolism. Other reports describe concomitantdefects in DNA repair and mitochondrial activity in neurode-generative diseases such as AT (7, 33). Several features of agingand neurodegeneration in AT are suppressed by complementationwith antiaging compounds or selective activators of mitochondrialbiogenesis (7). One of the hallmarks of neurodegeneration is theelevated risk of impaired mitochondrial function combined with thecell’s inability to properly handle DNA lesions. These phenomenaoccur in diverse diseases as well as during aging (21, 34). Ourfinding that deficiency of the DNA repair gene H2AX leads toimpaired mitochondrial homeostasis and increased neuronal dam-age reflects a key role for histone H2AX and chromatin-basedDNA repair in neurodegenerative diseases and aging. Accord-ingly, global genomic instability resulting from mutations or de-ficiencies in DNA repair genes may impact mitochondrialfunction leading to neurodegeneration.

Materials and MethodsCell Culture.MEF from both wild-type and H2AX mutant mice were obtainedfrom the Developmental Therapeutics Branch, Laboratory of MolecularPharmacology, National Cancer Institute, National Institutes of Health. Cellswere grown at 37 °C with 5% CO2 in DMEM (Invitrogen), supplemented with10% FBS (Atlanta Biologicals). All media were supplemented with 2 mMglutamine, penicillin, and streptomycin (Invitrogen).

Animals. Animals were housed on a 12-h light−dark schedule and receivedfood and water ad libitum. William Bonner and Andre Nussenzweig (Na-tional Cancer Institute/National Institutes of Health) kindly provided theH2AX heterozygote mice. All animals were treated in accordance withthe recommendations of the National Institutes of Health and approved bythe Johns Hopkins University Committee on Animal Care.

52 kDa

38 kDa

15kDa

BRAIN: STRIATUM

H2AX WT H2AX KOCTRL 3-NP CTRL 3-NP

TH

H2AX

GAPDH

A

3-NP

H2AX WT H2AX KO

Control

Sections and Tissues collection

Control 3-NPH

2AX

WT

H2A

X K

O

Control 3-NP

CORONAL SECTIONB

CD

E

THDAPI

WT CTRL

WT 3-

NP

KOCTRL

KO 3-NP

30

32

34

36

38

40

Rec

tal T

empe

ratu

re (°

C) *

ns *WT CTRL

WT 3-

NP

KO CTRL

KO 3-NP

0

50

100

150

Rel

ativ

e pr

otei

n am

ount

nsns

*

Fig. 4. H2AX deficiency leads to increased neuronal damage followingmitochondrial damage. (A) Wild-type and H2AX knockout mice were in-jected with either 3-nitropropionic acid or PBS following instructions de-scribed in Materials and Methods. After 24 h, mice were killed, and braintissues were either fixed in NBF or collected for biochemical analyses. (B)Coronal sections of the brain were taken through the striatum and analyzedfor TH-positive cells by immunohistochemistry. Purple, TH; blue, DAPI. (Scalebars, 100 μm.) (C and D) Western blot analysis of TH in the striatum. (C)Representative image; (D) quantification. Data are means ± SD; n = 4 miceper group. Statistical significance was determined by a two-tailed, unpairedStudent’s t test. (E) H2AX loss and 3-nitropionic acid treatment cause reductionof body temperature. We measured the rectal temperature in wild-type andH2AX knockout mice. H2AX mutant mice display a significant reduction inbody temperature, and this reduction progresses with 3-nitropropionic acidtreatment. These changes are reflective of lower mitochondrial activity. Dataare means ± SD; n = 4 mice per group. Statistical significance was determinedby a two-tailed, unpaired Student’s t test; *P < 0.05; ns, nonsignificant.

Weyemi et al. PNAS | April 9, 2019 | vol. 116 | no. 15 | 7475

NEU

ROSC

IENCE

Dow

nloa

ded

by g

uest

on

Aug

ust 2

4, 2

020

Four to five-month-old animals were used to analyze effects of 3-Nitropropionic acid (3-NP). Injections were based on protocols previouslyreported for mice (32). In brief, 3-NP (Sigma) was dissolved at 10 mg/mL insterile 0.1 ml PBS and adjusted to pH 7.4 with sodium hydroxide. For THdetection in the striatum, mice received i.p. injections of 60 mg/kg of 3-NPtwice, with 2 h between injections, and tissues were collected 24 hpostinjection.

Real-Time PCR. Total RNA was extracted from cells using RNeasy Mini Kit(Qiagen) according to the manufacturer’s instructions. Quality of RNApreparation, based on the 28S/18S ribosomal RNAs ratio, was assessed usingthe RNA 6000 Nano Lab-On-chip (Agilent Technologies). Reverse transcrip-tion and real-time PCR (RT-PCR) were performed as previously described (35).Oligonucleotides were predesigned and validated, and were considered tobe proprietary information by Thermo Fisher Scientific. However, the assaysIDs are available and are referenced as follow: PPARGC1B (Mm00504730_m1),TFAM (Mm00447485_m1), TFB2M (Mm01620397_s1), and POLRMT(Mm00553272_m1)

Mitochondrial Stress Assay Using Seahorse. Mitochondrial function was de-termined by measuring OCR of each cell line using XF Cell Mito Stress Test Kit(Agilent Technologies). Wild-type and H2AX knockout MEFs were seeded inan XF96 cell culture microplate. Media were prepared by adding 1 mmol/L ofpyruvate, 2 mmol/L of glutamine, and 10 mmol/L of glucose and stored as perthe manufacturer’s instructions. Seahorse assay was run in XF96 ExtracellularFlux Analyzer (Agilent Technologies). Following three baseline OCR mea-surements, cells were exposed sequentially to oligomycin (0.5 μmol/L), car-bonyl cyanide-4 (trifluoromethoxy) phenylhydrazone (FCCP; 1 μmol/L), androtenone/antimycin A (0.5 μmol/L). Oligomycin inhibits ATP synthase (com-plex V), and the decrease in OCR following injection of oligomycin correlateswith the mitochondrial respiration associated with cellular ATP production.FCCP is an uncoupling agent that collapses the proton gradient and disruptsthe mitochondrial membrane potential, allowing cells to achieve maximalOCR. As a result, electron flow through the electron transfer chain is uninhib-ited, and oxygen is maximally consumed by complex IV. The FCCP-stimulatedOCR can then be used to calculate spare respiratory capacity, defined asthe difference between maximal respiration and basal respiration. Sparerespiratory capacity is a measure of the ability of the cell to respond to

increased energy demand. The third injection is a mix of rotenone, a com-plex I inhibitor, and antimycin A, a complex III inhibitor. This combinationshuts down mitochondrial respiration and enables the calculation ofnonmitochondrial respiration driven by processes outside the mitochondria.Data were normalized to total cell survival. The assay results were analyzedusing Wave program 2.3.0 (Seahorse Bioscience), and data were exported inGraphPad Prism 7.

Brain Tissue Processing and Immunostaining. Mice were anesthetized by i.p.injection of sodium pentobarbital (80 mg/kg), then the heart was exposed,and transcardiac perfusion was performed using PBS for 5 min followed byice-cold 10%neutral buffered formalin (NBF; vol/vol) for 30min (using 3mL/minrate of perfusion) and maintained in NBF for 24 h. Tissues were transferred to30% sucrose and maintained at 4 °C for 24 h. Brains were sectioned on afreezing stage sliding microtome into a series of 35-μm sections. Sections werepermeabilizedwith 0.5% Triton X-100 in Tris buffered saline, then blocked with5% normal goat serum. Mouse anti-TH was used as a primary antibody at thedilution of 1:1,000. Anti-mouse IgG (H+L) Alexa Fluor 647 conjugate was usedas a secondary antibody. ZEISS LSM 800 confocal microscopy was used to imageimmunostained sections.

Statistical Analysis. Statistical analyses were performed using GraphPad Prism7 software (GraphPad Software) and Microsoft Excel 2010. Parametric datawere analyzed using a two-tailed t test. A value of P < 0.05 was consideredstatistically significant. Data are presented as mean ± SD or as mean ± SEM.

Generation of cells expressing H2AX-WT [H2AX-rescued cells or revertant(REV)], Western blots, cell viability assay, Transmission Electron Microscopy(TEM), Mitochondrial extraction, and immunoblot can be found in SI Ap-pendix, SI Materials and Methods.

ACKNOWLEDGMENTS. We thank Roxanne Barrow and Lauren Albacarys ofThe Solomon H. Snyder Department of Neuroscience (Johns Hopkins Univer-sity) for help with experiments. We thank Barbara Smith (TEM Specialist atJohns Hopkins School of Medicine) for her help in preparation and imagingof TEM samples. This work was supported by the National Institutes ofHealth and United States Public Health Service (USPHS) Grants MH18501and DA000266.

1. McKinnon PJ (2009) DNA repair deficiency and neurological disease. Nat Rev Neurosci10:100–112.

2. Bossy-Wetzel E, Schwarzenbacher R, Lipton SA (2004) Molecular pathways to neu-rodegeneration. Nat Med 10:S2–S9.

3. McKinnon PJ (2013) Maintaining genome stability in the nervous system. Nat Neurosci16:1523–1529.

4. Hoeijmakers JH (2001) Genome maintenance mechanisms for preventing cancer.Nature 411:366–374.

5. Bonner WM, et al. (2008) GammaH2AX and cancer. Nat Rev Cancer 8:957–967.6. Subba Rao K (2007) Mechanisms of disease: DNA repair defects and neurological

disease. Nat Clin Pract Neurol 3:162–172.7. Fang EF, et al. (2016) NAD+ replenishment improves lifespan and healthspan in ataxia

telangiectasia models via mitophagy and DNA repair. Cell Metab 24:566–581.8. Canugovi C, Misiak M, Ferrarelli LK, Croteau DL, Bohr VA (2013) The role of DNA

repair in brain related disease pathology. DNA Repair (Amst) 12:578–587.9. Chan SF, et al. (2014) ATM-dependent phosphorylation of MEF2D promotes neuronal

survival after DNA damage. J Neurosci 34:4640–4653.10. D’Souza AD, Parish IA, Krause DS, Kaech SM, Shadel GS (2013) Reducing mitochon-

drial ROS improves disease-related pathology in a mouse model of ataxia-telangiec-tasia. Mol Ther 21:42–48.

11. Fakouri NB, et al. (September 20, 2018) Toward understanding genomic instability,mitochondrial dysfunction and aging. FEBS J, 10.1111/febs.14663.

12. Barlow C, et al. (1996) Atm-deficient mice: A paradigm of ataxia telangiectasia. Cell86:159–171.

13. Barlow C, et al. (1999) Loss of the ataxia-telangiectasia gene product causes oxidativedamage in target organs. Proc Natl Acad Sci USA 96:9915–9919.

14. Guo Z, Kozlov S, Lavin MF, Person MD, Paull TT (2010) ATM activation by oxidativestress. Science 330:517–521.

15. Weyemi U, et al. (2015) NADPH oxidase 4 is a critical mediator in ataxia telangiectasiadisease. Proc Natl Acad Sci USA 112:2121–2126.

16. Rogakou EP, Pilch DR, Orr AH, Ivanova VS, Bonner WM (1998) DNA double-strandedbreaks induce histone H2AX phosphorylation on serine 139. J Biol Chem 273:5858–5868.

17. Celeste A, et al. (2002) Genomic instability in mice lacking histone H2AX. Science 296:922–927.

18. Weyemi U, et al. (2018) Histone H2AX deficiency causes neurobehavioral deficits andimpaired redox homeostasis. Nat Commun 9:1526.

19. Nakamura T, Cho DH, Lipton SA (2012) Redox regulation of protein misfolding, mi-tochondrial dysfunction, synaptic damage, and cell death in neurodegenerative dis-eases. Exp Neurol 238:12–21.

20. Nakamura T, Lipton SA (2017) ‘SNO’-storms compromise protein activity and mito-chondrial metabolism in neurodegenerative disorders. Trends Endocrinol Metab 28:879–892.

21. Lin MT, Beal MF (2006) Mitochondrial dysfunction and oxidative stress in neurode-generative diseases. Nature 443:787–795.

22. Lin J, Handschin C, Spiegelman BM (2005) Metabolic control through the PGC-1 familyof transcription coactivators. Cell Metab 1:361–370.

23. Leigh-Brown S, Enriquez JA, Odom DT (2010) Nuclear transcription factors in mam-malian mitochondria. Genome Biol 11:215.

24. Ruas JL, et al. (2012) A PGC-1α isoform induced by resistance training regulatesskeletal muscle hypertrophy. Cell 151:1319–1331.

25. St-Pierre J, et al. (2006) Suppression of reactive oxygen species and neuro-degeneration by the PGC-1 transcriptional coactivators. Cell 127:397–408.

26. Wu Z, et al. (1999) Mechanisms controlling mitochondrial biogenesis and respirationthrough the thermogenic coactivator PGC-1. Cell 98:115–124.

27. Tam WL, Weinberg RA (2013) The epigenetics of epithelial-mesenchymal plasticity incancer. Nat Med 19:1438–1449.

28. Weyemi U, et al. (2016) The histone variant H2A.X is a regulator of the epithelial-mesenchymal transition. Nat Commun 7:10711.

29. Cogliati S, Enriquez JA, Scorrano L (2016) Mitochondrial cristae: Where beauty meetsfunctionality. Trends Biochem Sci 41:261–273.

30. LeBleu VS, et al. (2014) PGC-1α mediates mitochondrial biogenesis and oxidativephosphorylation in cancer cells to promote metastasis. Nat Cell Biol 16:992–1003.

31. Jackson-Lewis V, Przedborski S (2007) Protocol for the MPTP mouse model of Par-kinson’s disease. Nat Protoc 2:141–151.

32. Mealer RG, Subramaniam S, Snyder SH (2013) Rhes deletion is neuroprotective in the3-nitropropionic acid model of Huntington’s disease. J Neurosci 33:4206–4210.

33. Browne SE, et al. (2004) Treatment with a catalytic antioxidant corrects the neuro-behavioral defect in ataxia-telangiectasia mice. Free Radic Biol Med 36:938–942.

34. Sedelnikova OA, et al. (2008) Delayed kinetics of DNA double-strand break processingin normal and pathological aging. Aging Cell 7:89–100.

35. Weyemi U, et al. (2012) ROS-generating NADPH oxidase NOX4 is a critical mediator inoncogenic H-Ras-induced DNA damage and subsequent senescence. Oncogene 31:1117–1129.

7476 | www.pnas.org/cgi/doi/10.1073/pnas.1820245116 Weyemi et al.

Dow

nloa

ded

by g

uest

on

Aug

ust 2

4, 2

020