the nrf2 activation and antioxidative response are not impaired

Hindawi Publishing CorporationOxidative Medicine and Cellular LongevityVolume 2013 Article ID 798401 11 pageshttpdxdoiorg1011552013798401

Research ArticleThe NRF2 Activation and Antioxidative ResponseAre Not Impaired Overall duringHyperoxia-Induced Lung Epithelial Cell Death

Haranatha R Potteti1 Narsa M Reddy1 Tom K Hei2

Dhananjaya V Kalvakolanu3 and Sekhar P Reddy1

1 Department of Pediatrics College of Medicine University of Illinois at ChicagoChicago IL 60612 USA

2Mailman School of Public Health Columbia University New York NY 10032 USA3Department of Microbiology and Immunology and Greenebaum Cancer CenterUniversity of Maryland School of Medicine Baltimore MD 21201 USA

Correspondence should be addressed to Sekhar P Reddy sreddy03uicedu

Received 31 January 2013 Accepted 25 March 2013

Academic Editor Hye-Youn Cho

Copyright copy 2013 Haranatha R Potteti et al This is an open access article distributed under the Creative Commons AttributionLicense which permits unrestricted use distribution and reproduction in any medium provided the original work is properlycited

Lung epithelial and endothelial cell death caused by pro-oxidant insults is a cardinal feature of acute lung injuryacute respiratorydistress syndrome (ALIARDSs) patients The NF-E2-related factor 2 (NRF2) activation in response to oxidant exposure is crucialto the induction of several antioxidative and cytoprotective enzymes that mitigate cellular stress Since prolonged exposure tohyperoxia causes cell death we hypothesized that chronic hyperoxia impairs NRF2 activation resulting in cell death To test thishypothesis we exposed nonmalignant small airway epithelial cells (AECs) to acute (1ndash12 h) and chronic (36ndash48 h) hyperoxia andevaluated cell death NRF2 nuclear accumulation and target gene expression and NRF2 recruitment to the endogenous HMOX1and NQO1 promoters As expected hyperoxia gradually induced death in AECs noticeably and significantly by 36 h sim60 ofcells were dead by 48 h However we unexpectedly found increased expression levels of NRF2-regulated antioxidative genes andnuclear NRF2 in AECs exposed to chronic hyperoxia as compared to acute hyperoxia Chromatin Immunoprecipitation (ChIP)assays revealed an increased recruitment of NRF2 to the endogenous HMOX1 and NQO1 promoters in AECs exposed to acute orchronic hyperoxia Thus our findings demonstrate that NRF2 activation and antioxidant gene expression are functional duringhyperoxia-induced lung epithelial cell death and that chronic hyperoxia does not impair NRF2 signaling overall

1 Introduction

The induction of antioxidant gene expression in lung-residentand infiltrated inflammatory cells in response to oxidativestress plays a significant role in pulmonary defense mech-anisms [1 2] However disequilibrium between prooxidantload and antioxidant defenses leading to redox imbalancecould potentially enhance tissue susceptibility to oxidativestress thereby contributing to the lung pathogenesis ofmany acute and chronic airway diseases These diseases

include idiopathic pulmonary fibrosis emphysema bron-chopulmonary dysplasia acute lung injury (ALI)acute res-piratory distress syndrome (ARDS) and lung cancer [3ndash6]Supplemental oxygen (hyperoxia) is used as therapy totreat ALIARDS patients In mice chronic exposure tohyperoxia results in endothelial and alveolar epithelial celldeath accompanied by pulmonary edema and respiratoryimpairment these pathologic features are similar to thoseobserved in ALIARDS patients [7] Thus understandingthe mechanisms by which hyperoxia contributes to lung

2 Oxidative Medicine and Cellular Longevity

pathogenesis is crucial to limiting the potentially harmfuleffects of oxygen toxicity in the clinical setting

We have previously shown that NF-E2-related factor 2(Nfe2l2 also known as Nrf2) a bZIP transcription factoris crucial for the induction of several antioxidant and cyto-protective genes in response to various pro-oxidant stimuliincluding hyperoxia [8 9] Nrf2-deficient mice are moresusceptible than wild-type mice to inflammatory and hyper-permeability responses to hyperoxic insult this response hasbeen generally attributed to a diminished or low expressionlevel of several antioxidant enzymes (AOEs) including geneencoding NQO1 HMOX1 GCLC GCLM and GPXs [8 9]which detoxify reactive oxygen species (ROS) andor nitro-gen (RNS) species We have also shown that a loss of Nrf2impairs the resolution of hyperoxia-induced acute lung injuryand inflammation and also exacerbates bacterial infection inadult mice following hyperoxic insult [10]

One-day oldNrf2-deficient pups when exposed to hyper-oxia for 72 h develop greater levels of alveolar simplification(septal growth arrest) at the 4th day [11] and 14th day [12] thando Nrf2-sufficient pups Nrf2 deficiency enhances cellularstress and susceptibility to oxidant-induced lung epithelialcell death [13] and its overexpression confers cellular pro-tection against hyperoxia in lung epithelial cells [14] as wellagainst proapoptotic stimuli in nonlung epithelial cells [1516] These observations suggest an important role for theNrf2-driven transcriptional response in mitigating cellularstress induced by prooxidants

Since chronic exposure to hyperoxia causes the deathof lung epithelial cells despite the presence of NRF2 wehypothesized that dysfunctional NRF2 signaling may con-tribute to this cell death To test this hypothesis we have nowanalyzed the nature of NRF2 activation (nuclear accumula-tion) and recruitment to the antioxidant gene (HMOX1 andNQO1) promoters in human nonmalignant lung small airwayepithelial cells during acute and chronic hyperoxia exposureHere we report that chronic hyperoxia does not impair NRF2nuclear accumulation or antioxidant gene expression duringthe hyperoxia-induced death of lung epithelial cells

2 Materials and Methods

21 Human Lung Epithelial Cell Culture and Hyperoxia Expo-sure The human normal small airway epithelial cell line(hereafter referred to as AECs) was established by theectopic expression of human telomerase reverse transcriptase(hTERT) Cells were grown in Dulbeccorsquos Modified EagleMedium with Hamrsquos F12 nutrient mixture (DMEMF12) inthe presence of 10 FBS and antibiotics [17] Cells were platedin equal number grown to 70ndash80 confluence and thenexposed to hyperoxia in complete medium For generatinghyperoxic condition cells were kept in modular incubatorchamber (Billups-Rothenberg Del Mar CA) and filled withgas mixture containing 95 O

2and 5 CO

2and chambers

placed in 37∘C incubator The chambers were refilled withthe gas mixture every 12 h As room air control group cellswere placed in the regular cell culture incubator with room

air and 5 CO2at 37∘C Culture medium was changed every

24 h during the exposure period

22 Gene Expression Analysis Cells were exposed to eitherroom air (RA) or hyperoxia (95 O

2and 5 CO

2) for

the indicated time periods Total RNA was isolated usingTrizol reagent (Gibco-BRLLife Technologies Grand IslandNY) and reverse transcribed using the qScript cDNA Super-Mix (Quanta BioSciences Gaithersburg MD) Target geneexpression was assessed by quantitative RT-PCR (qRT-PCR)using TaqMan gene expression assays (Applied BiosystemsFoster City CA) For Immunoblot analyses total proteinwas extracted in lysis buffer consisting of 20mM Tris (pH75) 150mM NaCl 1mM EDTA 1mM EGTA 1 Triton X-100 25mM sodium pyrophosphate 1mM Na

3VO4 5mM

120573-glycerophosphate and 1 120583gmL leupeptin Comparableamount of total protein (sim40120583g) from each sample wasseparated on a 10 sodium dodecyl sulfate polyacrylamidegel electrophoresis (SDS-PAGE) and the membranes wereprobed with antibodies specific for NRF2 (Santa CruzBiotech Santa Cruz CA) HMOX1 (Santa Cruz BiotechSanta Cruz CA) NQO1 (Abcam Cambridge UK) GCLC(kindly provided by Dr Terrance Kavanagh University ofWashington Seattle WA) 120573-actin (Sigma St Louis MO)antibody was used as the loading control The blots weredeveloped using an ECL kit (HyGlo Denville Scientific IncMetuchen NJ)

23 Cell Viability Assays Cells in equal number were platedand exposed to hyperoxia as described above Cell viabilitywas quantified by CellTiter-Glo kit (Promega Madison WI)and MTT assay LDH release was measured by CytoTox 96NonRadioactive Cytotoxicity assay kit (Promega MadisonWI) Viability and LDH release was calculated as a percentageof increase or decrease over their respective roomair controls

24 Chromatin Immunoprecipitation (ChIP) Assays ChIPassays were performed using the EZ-ChIP assay kit (Milli-pore Billerica MA) Briefly AECs were exposed to eitherroom air or hyperoxia for indicated time points cross-linked with formaldehyde and chromatin fragmentationwas carried out as detailed in the kit procedure Dilutedsoluble chromatin solution was incubated with rabbit anti-NRF2 (Santa Cruz Biotechnology Santa Cruz CA) for18 h at 4∘C with rotation Nonimmune rabbit IgG wasused as a negative control to determine the binding speci-ficity Following incubation with protein AG agarose beadsthe bound products were washed and DNA was elutedDNA was subjected to PCR with primers encompass-ing the functional antioxidant response elements (AREs)located upstream of transcriptional start site of HMOX1(F 51015840-CCCTGCTGAGTAATCCTTTCCCGA-31015840 and R 51015840-ATGTCCCGACTCCAGACTCCA-31015840) and NQO1 (F 51015840-GTGGAAGTCGTCCCAAGAGA-31015840 and R 51015840-TGTCTCCCCAGGACTCTCTCAG-31015840) to determine the binding of NRF2in ChIP assays

Oxidative Medicine and Cellular Longevity 3

100

89

100

58

100

45

0

20

40

60

80

100

120

NS

24 h 36 h 48 h

RAHyp

Cel

l via

bilit

y (

)(C

ellT

iter-

Glo

assa

y)

lowast

lowastsect

sect

(a)

100

97 100

63

100

51

0

20

40

60

80

100

120

Cel

l via

bilit

y (

)(M

TT as

say)

NS

RAHyp

lowast

lowast

sect

sect

24 h 36 h 48 h

(b)

100

141

100

208

100

234

0

50

100

150

200

250

300

LDH

rele

ase (

)

RAHyp

lowast

lowast

lowastsect

sect

24 h 36 h 48 h

(c)

Figure 1 Hyperoxia induced cell death in AECs Cells were exposed to hyperoxia (hyp) or room air (RA) for up to 48 h and cell viability wasmeasured at the indicated time points Cell viability assay was evaluated with CellTiter Glo Reagent (a) MTT reagent (b) and LDH releaseinto the culture medium (c) Results are represented as the percent viability of cells (a and b) or LDH release (c) in comparison to their timematched RA control groups The graph represents mean plusmn SD of 4 independent experiments performed in triplicate 119875 values are calculatedusing two-way ANOVA by Prism software lowast119875 le 005 RA versus hyp sect 119875 le 005 24 h hyp versus 36 h hyp or 36 h hyp versus 48 h hyp

25 Transfections and Reporter Gene Analyses Cells weretransfected with the NQO1 (NADPH quinone oxidasereductase-1) promoter reporter (luciferase) construct [18](kindly provided by Jeffrey Johnson University of Wiscon-sin) To normalize transfection efficiency between wells thecells were cotransfected with 5 ng of the Renilla luciferaseplasmid pRL-TK (Promega Corp Madison WI) At 24 hafter transfection cells were exposed to either room airor hyperoxia and extracts were assayed for firefly andthe Renilla luciferase activities using a dual luciferase kit(Promega Corp MadisonWI) Firefly luciferase activity wasnormalized to that of Renilla

26 Statistical Analyses Data were expressed as the mean plusmnSD (119899 = 3ndash9) as indicated in the legends The significancebetween the exposures was calculated by one-way or two-way(for cell viability and LDH release) analysis of variance fol-lowed by the Bonferroni post hoc tests by GraphPad PRISM4 Software A 119875 value of le005 is considered statisticallysignificant We also performed Studentrsquos 119905-test to confirmone-way analysis

3 Results

Malignant human lung epithelial cells express higher levelsof NRF2 than do nonmalignant lung epithelial cells mainlybecause of mutations in the inhibitor of NRF2 KEAP1which is known to facilitate NRF2 degradation under basalconditions [19ndash21] To overcome this problem and to deter-mine whether chronic hyperoxia promotes lung epithelialcell death by suppressing the NRF2-mediated transcriptionalresponse we have utilized a nononcogenic human lung(small airway) epithelial cell line immortalized by telomerase[17] To assess the effect of hyperoxia on airway epithelialcells (AECs) we subjected the cells to hyperoxia for 24 to

48 h and measured cell viability by using CellTiter-Glo andMTT reagents LDH release was evaluated by Cytotox Nosignificant difference in cell viability was observed betweenroom air and hyperoxia during the first 24 h of exposure(Figure 1(a)) However after 36 h hyperoxia had produced asignificant decline (sim42) in cell viability (as determined byCellTiter-Glo) which fell tosim55at 48 h To verify this resultwe evaluated cell viability using MTT assay (Figure 1(b))Hyperoxia caused a significant decline (sim37 loss) in cellviability after 36 h and the loss of viability increased sim49at 48 h in contrast hyperoxia exposure for 24 h had nosignificant effect on cell viability when compared to room aircontrols

We also analyzed hyperoxia-induced cellular toxicityas measured by LDH release into the culture medium Asignificant increase in LDH release was detected in cellsexposed to hyperoxia for 24 h (Figure 1(c)) when comparedto their room air-exposed counterparts However the LDHreleased by the cells exposed to hyperoxia for 36 h to 48 hwas markedly higher than the amount in the correspondingroom air controls (Figure 1(c)) The discordance in resultsbetween the cell viability (Figures 1(a) and 1(b)) and LDHmeasurement (Figure 1(c)) at the 24 h time point may berelated to the differences in the sensitivity of the assays

To determine the impact of chronic hyperoxia on the reg-ulation of the NRF2-dependent antioxidant transcriptionalresponse we subjected cells to either acute (1ndash12 h) or chronic(36ndash48 h) hyperoxia and determined the expression levels ofHMOX1 GCLC and NQO1 by qRT-PCR and immunoblotanalysis We selected these genes because they are upregu-lated by cellular stress and are putative transcriptional targetsof NRF2 and because they are known to play key rolesin cellular detoxification processes (reviewed in [22]) Asanticipated acute hyperoxiamarkedly stimulatedGCLC (23-fold) and NQO1 (27-fold) mRNA expression by as early as


1 1533 39

121

0

5

10

15

20

Relat

ive m

RNA

expr

essio

n

Relat

ive m

RNA

expr

essio

n

Relat

ive m

RNA

expr

essio

n

1 11 09

23

36

0

2

4

6

1 17 1827

58

0

2

4

6

8

10

RAHyperoxia

RAHyperoxia

RAHyperoxia

RA 1 h 3 h 6 h 12 h RA 1 h 3 h 6 h 12 h RA 1 h 3 h 6 h 12 h

lowastlowast

lowast

lowast

lowast

lowast

lowast

HMOX1 GCLC NQO1

(a)

1 1 1 18

17

09 1

8 23

10 1

91

0 14

0

2

4

6

8

10

12

14Re

lativ

e ban

d in

tens

ity3 h 6 h 12 hRA

120573-actin

120573-actin

120573-actin

RA 3 h 6 h 12 h

lowast

lowastlowast

lowastlowastlowast

HMOX1

GCLCNQO1

HMOX1

GCLC

NQO1

(b)

Figure 2 NRF2 target gene expression in AECs exposed to acute hyperoxia Cells were exposed to either room air or hyperoxia for 1 to12 h and then harvested at indicated time points for total RNA and protein extraction (a) cDNA was prepared from total RNA and themRNA expression levels of HMOX1 GCLC and NQO1 were quantified using TaqMan assay probes The values represent the mean plusmn SDof three independent experiments (119899 = 3) lowast119875 lt 005 RA versus hyperoxia (b) A comparable amount of whole cell extracts (sim40120583g) wasseparated on SDS-PAGE gel and immunoblotted with anti-HMOX1 anti-NQO1 or anti-GCLC antibodies 120573-actin was used a as referenceGraph represents the relative band intensities of the HMOX1 NQO1 and GCLC of three independent experiments done in duplicate (119899 = 6)The band intensities were quantified using Image J software and normalized with their respective 120573-actin band intensities Band intensitiesof room air exposed samples were considered as one arbitrary unit lowast119875 lt 005 RA versus hyperoxia

6 h and the levels remained high up to 12 h (36-fold and 58-fold increases for GCLC and NQO1 resp) (Figure 2(a)) Incontrast a significant increase in HMOX1mRNA expressionwas noticed during the acute phase as early as 3 h (33-fold)and its expression remained high (121-fold) up to 12 h Toverify that the changes in mRNA expression also occurredat the protein level we performed western blot analyses

using anti-HMOX1 -GCLC and -NQO1 antibodies on totalprotein extracts prepared from cells exposed to hyperoxiaor normoxia In agreement with the RT-PCR data westernblot analyses showed an increased expression of these genes(Figure 2(b)) HMOX1 expression was increased by 18-18- and 19-fold at 3 h 6 h and 12 h respectively whencompared to the room air control group Likewise NQO1


RA 36 h 48 h

1

63

113

0

5

10

15

20

1

75

155

0

5

10

15

20

25

Relat

ive m

RNA

expr

essio

n

1 575

246

0

50

100

150

200

250

300

350

RA 36 h 48 hRA 36 h 48 h

Relat

ive m

RNA

expr

essio

n

Relat

ive m

RNA

expr

essio

n

lowast

lowast

lowast

lowast

lowast

lowast

HMOX1 GCLC NQO1

(a)

Hyp RAHyp

36 h

RA

48 h

1 1 1 239

157

137 1 1 1

834

257

192

0

2

4

6

8

10

12

14

RA RAHyp Hyp

Rela

tive b

and

inte

nsity

36 h 48 h

lowastlowast

lowast

lowastlowastlowast

HMOX1GCLCNQO1

HMOX1

GCLC

NQO1

120573-actin

120573-actin

120573-actin

(b)

Figure 3The effects of chronic hyperoxia on NRF2 target gene expression in AECs Cells were exposed to either room air or hyperoxia for 36or 48 h and then harvested for total RNA and protein (a) HMOX1 GCLC and NQOmRNA expressionThe values represent the mean plusmn SDof four independent experiments (119899 = 6) lowast119875 lt 005 RA versus hyperoxia (b)Western blot analysis of whole-cell lysates using anti-HMOX1anti-NQO1 or anti-GCLC antibodies Membranes were stripped and reprobed with 120573-actin Graph represents the relative band intensitiesof HMOX1 NQO or GCLC as quantified in Figure 2 Values represent the mean plusmn SD of four independent experiments (119899 = 4) lowast119875 lt 005RA versus hyperoxia

expression was increased by 17-fold and 23-fold at 3 h and6 h respectively No increasewas found at the 12-h time pointIn the case of GCLC the protein expression was increased by14-fold at the 12-h time point only

We next analyzed the expression levels of these genesin cells exposed to chronic hyperoxia in order to determinewhether their lack of induction by NRF2 could be attributedto cell death Intriguingly we found increased levels ofHMOX1 (575-fold) GCLC (75-fold) and NQO1 (63-fold)

mRNA expression in cells exposed to chronic hyperoxia(36 h) when compared to those exposed to room air(Figure 3(a)) The induction of HMOX1 (246-fold) GCLC(155-fold) andNQO1 (113-fold)mRNAexpression remainedhigh up to 48 h The results of the western blot analyseswere correlated with a significantly increased expression ofHMOX1 (239-fold at 36 h and 834-fold at 48 h) GCLC (137-fold at 36 h and 192-fold at 48 h) and NQO1 (157-fold at 36 hand 257-fold at 48 h) over the corresponding roomair control


NRF2

Lamin B

RA 1 h 3 h 12 h

1

19

37

44

0

2

4

6

8

Rela

tive b

and

inte

nsity

RA 1 h 3 h 12 h

lowast

lowast

lowast

(a)

RA 36 h 48 h

NRF2

Lamin B

1

46

53

0

2

4

6

8

Rela

tive b

and

inte

nsity

RA 36 h 48 h

lowast

lowast

(b)

Figure 4The effects of acute and chronic hyperoxia on nuclear NRF2 levels Cells were exposed to room air (RA) or hyperoxia for indicatedtime points and nuclear extracts were prepared Equal amount of protein (sim15120583g) was separated on SDS-PAGE and probed with anti-NRF2antibody Membrane was stripped and re-probed with anti-lamin B antibody as loading control (a) Nuclear NRF2 levels in cells exposedto hyperoxia for 1 to 12 h (b) Nuclear NRF2 levels in cells exposed to hyperoxia for 36 h or 48 h Graph represents the relative NRF2 bandintensity of three independent experimentsplusmn SD (119899 = 3)The band intensities of NRF2were quantified using Image J software and normalizedwith their respective lamin B band intensities NRF2 band intensities of room air-exposed samples considered as one arbitrary unit

groups (Figure 3(b)) but the levels of each protein were notreflected by a corresponding mRNA abundance

In order to examine the effects of acute and chronichyperoxia on NRF2 nuclear accumulation we exposed AECsto hyperoxia then prepared nuclear extracts and performedwestern blotting using an anti-NRF2 antibody Hyperoxiaincreased the levels of nuclear NRF2 (Figure 4(a)) as early as1 h (19-fold) after exposure and the levels remained higherthan those of the corresponding room air control group at3 h (37-fold) and up to 12 h (44-fold) In response to chronichyperoxia NRF2 nuclear accumulation was higher at 36 h(46-fold) and 48 h (53-fold) than to room air (Figure 4(b))These results demonstrate that NRF2 accumulates in thenucleus in response to chronic hyperoxia in a manner com-parable to that observed in cells exposed to acute hyperoxia

To determine whether the increased levels of HMOX1and NQO1 expression are due to NRF2 binding we per-formed ChIP assays analyzing the recruitment of NRF2 toendogenousHMOX1 andNQO1 promoters in cells exposed toacute and chronic hyperoxia We used gene-specific primersflanking space the critical AREs in each case (Figure 5(a))These experiments revealed the recruitment of NRF2 to

the HMOX1 promoter as early as 3 h the levels returnedto baseline at 6 h during acute hyperoxia (Figure 5(b) leftpanel) NRF2 binding to the NQO1 promoter was very lowor undetectable under conditions of room air and it rapidlyincreased as early as 3 h (135-fold) but returned to basallevels at 6 h (Figure 5(b) right panel) However the bindingofNRF2 to theNQO1 promoter rose again at 12 h of hyperoxia(224-fold) (Figure 5(b) left panel) An increased enrichmentof NRF2 at theNQO1 promoter at the 12 h time point was alsoreflected in higher mRNA levels but western blot analysisrevealed no increase in NQO1 protein expression at the 12 htime point perhaps because of a lag in mRNA translation

During chronic hyperoxia we found that the binding ofNRF2 to theHMOX1 enhancerwas significantly higher at 36 h(18-fold) and 48 h (24-fold) than in cells exposed to room air(Figure 5(c) left panel) Likewise during chronic hyperoxiathe binding of NRF2 to the NQO1 promoter although nothigher at the 36 h time point was significantly higher at48 h than in cells exposed to room air (Figure 5(c) rightpanel) However the level of NRF2 enrichment at the NQO1promoter was considerably lower than in cells exposed toacute hyperoxia (Figure 5(b) right panel)


ARE AREminus618

minus449

minus9204

minus8963

HMOX1NQO1

(a)

113

27

1113

0

1

2

3

4

5

Rela

tive N

RF2

bind

ing

RA 12 h6 h3 h1 hRA 12 h6 h3 h1 h

IP NRF2

Input

1 22

135

12

224

0

5

10

15

20

25

30

Relat

ive N

RF2

bind

ing

IP NRF2

Input

RA 12 h6 h3 h1 hRA 12 h6 h3 h1 h

lowast

lowast

lowast

(b)

Input

IP NRF2

RA Hyp RA Hyp IgG Hist

one H

3

Input

IP NRF2

RA Hyp RA Hyp IgG

Hist

one H

3

Relat

ive N

RF2

bind

ing

Relat

ive N

RF2

bind

ing

36 h 48 h 36 h 48 h

NQO1promoterHMOX1promoter

1 11 1 170

5

10

15

20

25

30

RA HypRAHyp

36 h 48 h

lowast1

18

1

24

0

1

2

3

4

5

RA HypRAHyp36 h 48 h

lowast

lowast

(c)

Figure 5 The effects of acute and chronic hyperoxia on the binding of NRF2 to HMOX1 and NQO1 promoters Cells were exposed to roomair or hyperoxia for 1 to 12 h or for 36 to 48 h chromatin was cross-linked and immunoprecipitated with IgG or anti-NRF2 antibodies andDNA was amplified using gene specific primers (a) Scheme showing the positions of the ARE sites forward and reverse primers of HMOX1andNQO1 promoters used in ChIP assays (b) NRF2 binding to theHMOX1 andNQO1 promoters in cells exposed to acute hyperoxia (1ndash12 h)(c) NRF2 binding to the HMOX1 and NQO1 promoters in cells exposed to chronic hyperoxia (36 h or 48 h) PCR products were analyzed on2 agarose gel Band intensities were quantified with Image J software Input DNA was used as a control Graph represents mean plusmn SD of fiveindependent experiments (119899 = 5) Fold increase was calculated over their respective room air controls lowast119875 lt 005 RA versus hyperoxia


Rela

tive l

ucife

rase

activ

ity

1 11

21

69

0

2

4

6

8

RA 12 h6 h3 h

lowast

lowast

(a)

115

0

2

4

6

8

RA 36 h

lowast

(b)

Figure 6 NQO1 promoter activity in AECs exposed to hyperoxia acutely and chronically To determine the effects of hyperoxia on NQO1promoter activity luciferase reporter construct bearingNQO1 promoter (100 ng) and pRL-TK coding for theRenilla luciferasewere transientlytransfected into AECs exposed to hyperoxia or room air and luciferase activity was analyzed using dual luciferase assay kit The promoteractivity of AECs exposed to room air was set as on unit Values represent the average of 3 independent experiments done in triplicate (119899 = 9)

We next determined the transcriptional activity of theNQO1 promoter in cells exposed to acute and chronic hyper-oxia Cells were transiently transfected with the NQO1-Lucreporter construct and then exposed to hyperoxia As shownin Figure 6 hyperoxia enhanced NQO1-Luc expression by21- and 69-fold at 6 h and 12 h of exposure (Figure 6(a))respectively when compared to room air controls We foundthat the Luc activity was also significantly higher (15-fold)in cells exposed to chronic hyperoxia for 36 h than in thecorresponding room air control group (Figure 6(b)) thisresult is comparable to the 21-fold increase found in cellsexposed to hyperoxia for 6 h However the magnitude of thepromoter inducibility was considerably lower than the 69-fold increase observed in cells exposed to hyperoxia for 12 h

4 Discussion

Exposure to hyperoxia for prolonged periods (chronic expo-sure) is known to induce oxidative stress mainly as a resultof a redox imbalance caused by excessive accumulation ofreactive oxygen species which initially causes cellular dam-age and ultimately cell death [23] NRF2-induced expressionof antioxidative and cytoprotective genes is crucial to main-taining redox homeostasis during exposure to toxicants andinjurious insultsWe have previously shown thatmice lackingNrf2 are highly susceptible to hyperoxia-induced lung injuryand lung epithelial cell death in vivo and in vitro [8 14] sug-gesting that prolonged exposure to hyperoxia promotes lungepithelial cell injury and death by impairing NRF2 activationand subsequently inhibiting its induction of downstreamtarget genes However in the present study we demonstrate

that the enhanced expression levels of putative NRF2 targetgenes in lung epithelial cells exposed to chronic hyperoxiaare of a higher magnitude than those observed under acutehyperoxia Moreover we observed that the recruitment ofnuclear NRF2 to the promoters of antioxidant genes (egHMOX1 and NQO1) in cells exposed to chronic hyperoxiaThus it appears that NRF2-mediated gene expression is notglobally or largely compromised but it induced to mounta cytoprotective response to preserve redox homeostasisthereby helping to maintain cell survival or preventing lungepithelial cell death during chronic hyperoxia

NRF2 is mainly localized to the cytoplasm in its nativestate and its nuclear accumulation in response to stressfulinsults is critical for antioxidant gene induction [24 25]The nucleocytoplasmic shuttling of NRF2 appears to varyaccording to the inducer andor cell type and is regulatedby multiple complex mechanisms [24 25] For examplevarious protein kinases such as PKC [26 27] ERK12 [28]and AKT [29] activated by pro-oxidant exposure are alsoknown to facilitate and enhance the accumulation of NRF2in the nucleus After the signal-induced dissociation from itscytoplasmic inhibitor Keap1 importins facilitate the nucleartranslocation of NRF2 [30] Phosphorylation of NRF2 inthe nucleus promotes its nuclear exclusion and subsequentKeap1-mediated degradation [31 32] Thus deregulation ofNRF2 nuclear accumulation following chronic stressful stim-uli such as hyperoxia is generally connected with cellularinjury and death However our present results indicate thatthis is not the case We found higher levels of NRF2 in thenucleus of cells during chronic hyperoxia than in room air-exposed cells and the levels were comparable to those of


cells exposed to acute hyperoxia suggesting that the variouseffector pathways that facilitate nuclear accumulation ofNRF2 and its DNA binding are functional during conditionsof hyperoxia-induced lung epithelial cell death

NRF2 alone is insufficient to bind to the ARE Its het-erodimerization with other b-ZIP transcription factors suchas JUN and the small MAF family of proteins is also required[33ndash36] The magnitude and duration of the antioxidantgene expression can be dictated by both cooperative andcombinatorial interactions that occur between NRF2 and theMAFJUN family of proteins Thus the impairment of NRF2binding to the DNAmight affect its transactivational activitydespite elevated levels of NRF2 in the nucleus Because thelevels of NRF2-dependent target gene expression are almosthigher under chronic hyperoxia than during exposure toacute hyperoxia it appears that hyperoxia does not alterthese interactions or diminish the lung antioxidant capacityleading to a dysfunctional cellular response Members of theJun family of proteins c-Jun and Jun-D can dimerize withNrf2 and upregulate antioxidant gene expression [37 38] Forexample mouse embryonic fibroblasts lacking c-Jun or Jun-D demonstrate decreased levels of antioxidant enzymes andenhanced oxidative stress [33ndash36] Likewise c-Myc upregu-lates the expression of cytoprotective genes in response tostressful stimuli [39] Thus it is possible that these proteinsact in a cooperative or synergistic manner with NRF2 topotentiate ARE-mediated gene transcription during chronichyperoxia

We have previously reported the binding of Nrf2 to AREsin murine alveolar type-II like epithelial cells following acutehyperoxia [14 28] In the present study ChIP assays revealedthe binding of NRF2 to its target gene promoters (NQO1 andHMOX1) in cells exposed to chronic hyperoxia (48 h) sug-gesting that chronic hyperoxia does not compromise NRF2binding to antioxidant promoters Although the magnitudeof the NRF2 binding at the HMOX1 promoter was similarunder acute (at 3 h) and chronic (at 36 h and 48 h) hyperoxicconditions (Figure 5) the enrichment of NRF2 at the NQO1promoter appeared to be variable and biphasic in responseto hyperoxia There was a 13-fold increase in the bindingof NRF2 to the NQO1 promoter at 3 h and the bindingreturned to basal levels at 6 h after hyperoxia However theNRF2 binding increased significantly to greater levels at 12 hthan at 3 h after hyperoxia During chronic hyperoxia asignificant increase in NRF2 binding at the NQO1 promoterwas detected at 48 h but not at 36 h when compared toroom air controls It is possible that the association of NRF2with its partners is dynamic or that its negative regulatorssuch as BACH1 [40] and Fra-1 [41] limit the availability ofNRF2 or its partners to form complexes with NRF2 makingit differentially bind to the AREs in a gene-promoter context-dependent manner during acute and chronic hyperoxia Therelevance andmechanisms of such differential NRF2 bindingto the NQO1 promoter during hyperoxia-induced cell deathwarrants a separate investigation

It is noteworthy that the lower NRF2 binding at theendogenous NQO1 promoter was reflected in diminishedpromoter activation by chronic hyperoxia in our transient

transfection assays however NQO1mRNA expression levelsin cells exposed to chronic hyperoxia were nearly comparableto those observed in cells exposed to acute hyperoxia Weassume that the posttranscriptional regulation of NQO1explains the discordance between diminished promoteractivity and increased mRNA expression ofNQO1 in chronichyperoxia It should also be noted that reporter analysesutilize a short fragment of the promoter unlike the nativegene promoter and this difference may also explain some ofthe differences Also we would like to point out the lack ofa direct correlation between mRNA and protein expressionFor example the increased protein abundance ofHMOX1 andother genes analyzed by western blot analysis was sim10-foldless than the corresponding mRNA expression levels in cellsexposed to either acute or chronic hyperoxia It is unclearwhether this result reflects error and noise in the real-timePCR and immunoblot analysis andor variations in proteinsynthesis and degradation [42]

Previously it has been shown that the exposure tooxidants (such as high doses of H

2O2) leads to decreased

NRF2 levels and a consequent suppression of the antioxidantresponse pathway culminating in cellular injury and death[43 44] However we found that nuclear NRF2 proteinlevels in cells exposed to chronic hyperoxia are not reducedwhen compared to room air-exposed cells or to cells exposedto acute hyperoxia (Figure 4) ruling out such a possibilityThe ability of hyperoxia to induce cell death despite thepresence of high levels of antioxidant gene expression issomewhat surprising It is important to note that we haveonly analyzed the expression of a subset of the NRF2-targetgenes (GCLC HMOX1 and NQO1) (data not shown forTRX1 and GCLM) that are markedly induced during chronichyperoxia Previously by global gene expression profiling andprimary lung type alveolar II epithelial cultures from Nrf2-null and wild-type mice we have shown that Nrf2 regulatesthe expression of several genes involved in antioxidativeand cytoprotective responses as well as cell proliferation andsurvival [14 45] Global mapping ChIP-binding assays andmRNA expression profiling in Keap1-null or Nrf2-null mouseembryonic fibroblasts revealed that Nrf2 binds and regulatesthe expression of sim500 genes involved in cell proliferationand stress response [46] It is possible that some of thesegene products that dampen the initiation and executionof cell death pathways are not being induced by NRF2during chronic hyperoxia despite its nuclear presence wecannot rule out this possibility Recently Taguchi et al havedemonstrated that an increased level of NRF2 accumulationpromotes liver damage in autophagy-deficient mice [47]Whether high levels of nuclear NRF2 have any role inmediating cellular injury during chronic hyperoxic settingremains to be investigated

In summary we have demonstrated that both NRF2nuclear enrichment and upregulation of antioxidative geneexpression occur in cells exposed to chronic hyperoxia ChIPassays revealed that themagnitude of theNRF2 binding at theantioxidant gene promoters is dynamic variable and bipha-sic in response to hyperoxia exposure suggesting that NRF2-mediated signaling is not globally or largely compromised


Rather it continues to be functional during chronic hyperoxiathat causes the death of lung epithelial cells

Conflict of Interests

All the authors declare no conflict of interests

Acknowledgments

This work was funded by the National Institute of HealthGrants HL66109 ES11863 and ES18998 (to Sekhar P Reddy)and CA78282 and CA105005 (to Dhananjaya V Kalvak-olanu)

References

[1] Y S Ho R Vincent M S Dey J W Slot and J D CrapoldquoTransgenic models for the study of lung antioxidant defenseenhanced manganese-containing superoxide dismutase activ-ity gives partial protection to b6c3 hybrid mice exposed tohyperoxiardquo American Journal of Respiratory Cell and MolecularBiology vol 18 no 4 pp 538ndash547 1998

[2] P J Lee and A M K Choi ldquoPathways of cell signaling inhyperoxiardquo Free Radical Biology andMedicine vol 35 no 4 pp341ndash350 2003

[3] OD Saugstad ldquoBronchopulmonary dysplasiamdashoxidative stressand antioxidantsrdquo Seminars in Neonatology vol 8 no 1 pp 39ndash49 2003

[4] I Fridovich ldquoOxygen toxicity a radical explanationrdquo Journal ofExperimental Biology vol 201 no 8 pp 1203ndash1209 1998

[5] B Halliwell J M C Gutteridge and C E Cross ldquoFree radicalsantioxidants and human disease where are we nowrdquo Journalof Laboratory and Clinical Medicine vol 119 no 6 pp 598ndash6201992

[6] W MacNee and I Rahman ldquoIs oxidative stress central tothe pathogenesis of chronic obstructive pulmonary diseaserdquoTrends in Molecular Medicine vol 7 no 2 pp 55ndash62 2000

[7] J D Crapo ldquoMorphologic changes in pulmonary oxygentoxicityrdquoAnnual Review of Physiology vol 48 pp 721ndash731 1986

[8] H Y Cho A E Jedlicka S P M Reddy et al ldquoRole of NRF2in protection against hyperoxic lung injury in micerdquo AmericanJournal of Respiratory Cell and Molecular Biology vol 26 no 2pp 175ndash182 2002

[9] H Y Cho A E Jedlicka S P M Reddy L Y Zhang T WKensler and S R Kleeberger ldquoLinkage analysis of susceptibilityto hyperoxia Nrf2 is a candidate generdquo American Journal ofRespiratory Cell and Molecular Biology vol 26 no 1 pp 42ndash512002

[10] N M Reddy V Suryanarayana D V Kalvakolanu et alldquoInnate immunity against bacterial infection following hyper-oxia exposure is impaired in NRF2-deficient micerdquo Journal ofImmunology vol 183 no 7 pp 4601ndash4608 2009

[11] H Y Cho B van Houten X Wang et al ldquoTargeted deletion ofnrf2 impairs lung development and oxidant injury in neonatalmicerdquoAntioxidants and Redox Signaling vol 17 no 8 pp 1066ndash1082 2012

[12] S McGrath-Morrow T Lauer M Yee et al ldquoNrf2 increasessurvival and attenuates alveolar growth inhibition in neonatalmice exposed to hyperoxiardquoAmerican Journal of Physiology vol296 no 4 pp L565ndashL573 2009

[13] N M Reddy S R Kleeberger H Y Cho et al ldquoDeficiency inNrf2 GSH signaling impairs type II cell growth and enhancessensitivity to oxidantsrdquoAmerican Journal of Respiratory Cell andMolecular Biology vol 37 no 1 pp 3ndash8 2007

[14] S Papaiahgari S R Kleeberger H Y Cho D V Kalvakolanuand S P Reddy ldquoNADPH oxidase and ERK signaling regulateshyperoxia-induced Nrf2-ARE transcriptional response in pul-monary epithelial cellsrdquo Journal of Biological Chemistry vol 279no 40 pp 42302ndash42312 2004

[15] N Morito K Yoh K Itoh et al ldquoNrf2 regulates the sensitivityof death receptor signals by affecting intracellular glutathionelevelsrdquo Oncogene vol 22 no 58 pp 9275ndash9281 2003

[16] S B Cullinan D Zhang M Hannink E Arvisais R JKaufman and J A Diehl ldquoNrf2 is a direct PERK substrateand effector of PERK-dependent cell survivalrdquo Molecular andCellular Biology vol 23 no 20 pp 7198ndash7209 2003

[17] C Q Piao L Liu Y L Zhao A S Balajee M Suzuki and T KHei ldquoImmortalization of human small airway epithelial cells byectopic expression of telomeraserdquo Carcinogenesis vol 26 no 4pp 725ndash731 2005

[18] J D Moehlenkamp and J A Johnson ldquoActivation of anti-oxidantelectrophile-responsive elements in IMR-32 humanneuroblastoma cellsrdquo Archives of Biochemistry and Biophysicsvol 363 no 1 pp 98ndash106 1999

[19] A Singh V Misra R K Thimmulappa et al ldquoDysfunctionalKEAP1-NRF2 interaction in non-small-cell lung cancerrdquo PLoSMedicine vol 3 no 10 pp 1865ndash1876 2006

[20] B Padmanabhan K I Tong T Ohta et al ldquoStructural basis fordefects of Keap1 activity provoked by its pointmutations in lungcancerrdquoMolecular Cell vol 21 no 5 pp 689ndash700 2006

[21] T Suzuki J Maher and M Yamamoto ldquoSelect heterozygousKeap1 mutations have a dominant-negative effect on wild-typeKeap1 in vivordquo Cancer Research vol 71 no 5 pp 1700ndash17092011

[22] T W Kensler N Wakabayashi and S Biswal ldquoCell survivalresponses to environmental stresses via the Keap1-Nrf2-AREpathwayrdquo Annual Review of Pharmacology and Toxicology vol47 pp 89ndash116 2007

[23] C Barazzone S Horowitz Y R Donati I Rodriguez and P FPiguet ldquoOxygen toxicity in mouse lung pathways to cell deathrdquoAmerican Journal of Respiratory Cell andMolecular Biology vol19 no 4 pp 573ndash581 1998

[24] M Kobayashi andM Yamamoto ldquoMolecular mechanisms acti-vating the Nrf2-Keap1 pathway of antioxidant gene regulationrdquoAntioxidants and Redox Signaling vol 7 no 3-4 pp 385ndash3942005

[25] H Y Cho S P Reddy and S R Kleeberger ldquoNrf2 defends thelung from oxidative stressrdquoAntioxidants amp Redox Signaling vol8 no 1-2 pp 76ndash87 2006 Review

[26] H C Huang T Nguyen and C B Pickett ldquoRegulation of theantioxidant response element by protein kinase C-mediatedphosphorylation of NF-E2-related factor 2rdquo Proceedings of theNational Academy of Sciences of the United States of Americavol 97 no 23 pp 12475ndash12480 2000

[27] H C Huang T Nguyen and C B Pickett ldquoPhosphorylationof Nrf2 at Ser-40 by protein kinase C regulates antioxidantresponse element-mediated transcriptionrdquo Journal of BiologicalChemistry vol 277 no 45 pp 42769ndash42774 2002

[28] S Papaiahgari Q Zhang S R Kleeberger H Y Cho and SP Reddy ldquoHyperoxia stimulates an Nrf2-ARE transcriptionalresponse via ROS-EGFR-P13K-AktERKMAP kinase signaling


in pulmonary epithelial cellsrdquoAntioxidants andRedox Signalingvol 8 no 1-2 pp 43ndash52 2006

[29] J H Lim K M Kim S W Kim O Hwang and H J ChoildquoBromocriptine activates NQO1 via Nrf2-PI3KAkt signalingnovel cytoprotective mechanism against oxidative damagerdquoPharmacological Research vol 57 no 5 pp 325ndash331 2008

[30] MTheodore Y Kawai J Yang et al ldquoMultiple nuclear localiza-tion signals function in the nuclear import of the transcriptionfactor Nrf2rdquo Journal of Biological Chemistry vol 283 no 14 pp8984ndash8994 2008

[31] J W Kaspar and A K Jaiswal ldquoAntioxidant-induced phospho-rylation of tyrosine 486 leads to rapid nuclear export of Bach1that allows Nrf2 to bind to the antioxidant response elementand activate defensive gene expressionrdquo Journal of BiologicalChemistry vol 285 no 1 pp 153ndash162 2010

[32] R N Karapetian A G Evstafieva I S Abaeva et al ldquoNuclearoncoprotein prothymosin 120572 is a partner of Keap1 implicationsfor expression of oxidative stress-protecting genesrdquo Molecularand Cellular Biology vol 25 no 3 pp 1089ndash1099 2005

[33] D Gerald E Berra Y M Frapart et al ldquoJunD reduces tumorangiogenesis by protecting cells from oxidative stressrdquo Cell vol118 no 6 pp 781ndash794 2004

[34] A MacLaren E J Black W Clark and D A F Gillespie ldquoC-Jun-deficient cells undergo premature senescence as a resultof spontaneous DNA damage accumulationrdquo Molecular andCellular Biology vol 24 no 20 pp 9006ndash9018 2004

[35] A Meixner F Karreth L Kenner J M Penninger and E FWagner ldquoJun and JunD-dependent functions in cell prolifera-tion and stress responserdquo Cell Death and Differentiation vol 17no 9 pp 1409ndash1419 2010

[36] J B Weitzman L Fiette K Matsuo and M Yaniv ldquoJunDprotects cells from p53-dependent senescence and apoptosisrdquoMolecular Cell vol 6 no 5 pp 1109ndash1119 2000

[37] K Igarashi K Kataoka K Itoh N Hayashi M Nishizawa andM Yamamoto ldquoRegulation of transcription by dimerization oferythroid factor NF-E2 p45 with small Maf proteinsrdquo Naturevol 367 no 6463 pp 568ndash572 1994

[38] R Venugopal and A K Jaiswal ldquoNrf2 and Nrf1 in associa-tion with Jun proteins regulate antioxidant response element-mediated expression and coordinated induction of genesencoding detoxifying enzymesrdquo Oncogene vol 17 no 24 pp3145ndash3156 1998

[39] C V Dang K A OrsquoDonnell K I Zeller T Nguyen R COsthus and F Li ldquoThe c-Myc target gene networkrdquo Seminarsin Cancer Biology vol 16 no 4 pp 253ndash264 2006

[40] S Dhakshinamoorthy A K Jain D A Bloom and A KJaiswal ldquoBach1 competes with Nrf2 leading to negative reg-ulation of the antioxidant response element (ARE)-mediatedNAD(P)Hquinone oxidoreductase 1 gene expression andinduction in response to antioxidantsrdquo Journal of BiologicalChemistry vol 280 no 17 pp 16891ndash16900 2005

[41] M Vaz N Machireddy A Irving et al ldquoOxidant-inducedcell death and Nrf2-dependent antioxidative response are con-trolled by Fra-1AP-1rdquo Molecular and Cellular Biology vol 32no 9 pp 1694ndash1709 2012

[42] D Greenbaum C Colangelo K Williams and M GersteinldquoComparing protein abundance and mRNA expression levelson a genomic scalerdquo Genome Biology vol 4 no 9 article 1172003

[43] J L Ning L W Mo and X N Lai ldquoLow-and high-dosehydrogen peroxide regulation of transcription factor NF-E2-related factor 2rdquo Chinese Medical Journal vol 123 no 8 pp1063ndash1069 2010

[44] N F Villeneuve Z Sun W Chen and D D Zhang ldquoNrf2and p21 regulate the fine balance between life and death bycontrolling ROS levelsrdquo Cell Cycle vol 8 no 20 pp 3255ndash32562009

[45] N M Reddy S R Kleeberger M Yamamoto et al ldquoGeneticdissection of the Nrf2-dependent redox signaling-regulatedtranscriptional programs of cell proliferation and cytoprotec-tionrdquo Physiological Genomics vol 32 no 1 pp 74ndash81 2007

[46] D Malhotra E Portales-Casamar A Singh et al ldquoGlobalmapping of binding sites for Nrf2 identifies novel targets incell survival response through chip-seq profiling and networkanalysisrdquo Nucleic Acids Research vol 38 no 17 Article IDgkq212 pp 5718ndash5734 2010

[47] K Taguchi N Fujikawa M Komatsu et al ldquoKeap1 degradationby autophagy for the maintenance of redox homeostasisrdquoProceedings of the National Academy of Sciences of the UnitedStates of America vol 109 no 34 pp 13561ndash13566 2010

Submit your manuscripts athttpwwwhindawicom

Stem CellsInternational

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


MEDIATORSINFLAMMATION

of


Behavioural Neurology

EndocrinologyInternational Journal of



Disease Markers


BioMed Research International

OncologyJournal of



Oxidative Medicine and Cellular Longevity


PPAR Research

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

Immunology ResearchHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

Journal of

ObesityJournal of



Computational and Mathematical Methods in Medicine

OphthalmologyJournal of


Diabetes ResearchJournal of



Research and TreatmentAIDS


Gastroenterology Research and Practice


Parkinsonrsquos Disease

Evidence-Based Complementary and Alternative Medicine

Volume 2014Hindawi Publishing Corporationhttpwwwhindawicom


pathogenesis is crucial to limiting the potentially harmfuleffects of oxygen toxicity in the clinical setting

We have previously shown that NF-E2-related factor 2(Nfe2l2 also known as Nrf2) a bZIP transcription factoris crucial for the induction of several antioxidant and cyto-protective genes in response to various pro-oxidant stimuliincluding hyperoxia [8 9] Nrf2-deficient mice are moresusceptible than wild-type mice to inflammatory and hyper-permeability responses to hyperoxic insult this response hasbeen generally attributed to a diminished or low expressionlevel of several antioxidant enzymes (AOEs) including geneencoding NQO1 HMOX1 GCLC GCLM and GPXs [8 9]which detoxify reactive oxygen species (ROS) andor nitro-gen (RNS) species We have also shown that a loss of Nrf2impairs the resolution of hyperoxia-induced acute lung injuryand inflammation and also exacerbates bacterial infection inadult mice following hyperoxic insult [10]

One-day oldNrf2-deficient pups when exposed to hyper-oxia for 72 h develop greater levels of alveolar simplification(septal growth arrest) at the 4th day [11] and 14th day [12] thando Nrf2-sufficient pups Nrf2 deficiency enhances cellularstress and susceptibility to oxidant-induced lung epithelialcell death [13] and its overexpression confers cellular pro-tection against hyperoxia in lung epithelial cells [14] as wellagainst proapoptotic stimuli in nonlung epithelial cells [1516] These observations suggest an important role for theNrf2-driven transcriptional response in mitigating cellularstress induced by prooxidants

Since chronic exposure to hyperoxia causes the deathof lung epithelial cells despite the presence of NRF2 wehypothesized that dysfunctional NRF2 signaling may con-tribute to this cell death To test this hypothesis we have nowanalyzed the nature of NRF2 activation (nuclear accumula-tion) and recruitment to the antioxidant gene (HMOX1 andNQO1) promoters in human nonmalignant lung small airwayepithelial cells during acute and chronic hyperoxia exposureHere we report that chronic hyperoxia does not impair NRF2nuclear accumulation or antioxidant gene expression duringthe hyperoxia-induced death of lung epithelial cells

2 Materials and Methods

21 Human Lung Epithelial Cell Culture and Hyperoxia Expo-sure The human normal small airway epithelial cell line(hereafter referred to as AECs) was established by theectopic expression of human telomerase reverse transcriptase(hTERT) Cells were grown in Dulbeccorsquos Modified EagleMedium with Hamrsquos F12 nutrient mixture (DMEMF12) inthe presence of 10 FBS and antibiotics [17] Cells were platedin equal number grown to 70ndash80 confluence and thenexposed to hyperoxia in complete medium For generatinghyperoxic condition cells were kept in modular incubatorchamber (Billups-Rothenberg Del Mar CA) and filled withgas mixture containing 95 O

2and 5 CO

2and chambers

placed in 37∘C incubator The chambers were refilled withthe gas mixture every 12 h As room air control group cellswere placed in the regular cell culture incubator with room

air and 5 CO2at 37∘C Culture medium was changed every

24 h during the exposure period

22 Gene Expression Analysis Cells were exposed to eitherroom air (RA) or hyperoxia (95 O

2and 5 CO

2) for

the indicated time periods Total RNA was isolated usingTrizol reagent (Gibco-BRLLife Technologies Grand IslandNY) and reverse transcribed using the qScript cDNA Super-Mix (Quanta BioSciences Gaithersburg MD) Target geneexpression was assessed by quantitative RT-PCR (qRT-PCR)using TaqMan gene expression assays (Applied BiosystemsFoster City CA) For Immunoblot analyses total proteinwas extracted in lysis buffer consisting of 20mM Tris (pH75) 150mM NaCl 1mM EDTA 1mM EGTA 1 Triton X-100 25mM sodium pyrophosphate 1mM Na

3VO4 5mM

120573-glycerophosphate and 1 120583gmL leupeptin Comparableamount of total protein (sim40120583g) from each sample wasseparated on a 10 sodium dodecyl sulfate polyacrylamidegel electrophoresis (SDS-PAGE) and the membranes wereprobed with antibodies specific for NRF2 (Santa CruzBiotech Santa Cruz CA) HMOX1 (Santa Cruz BiotechSanta Cruz CA) NQO1 (Abcam Cambridge UK) GCLC(kindly provided by Dr Terrance Kavanagh University ofWashington Seattle WA) 120573-actin (Sigma St Louis MO)antibody was used as the loading control The blots weredeveloped using an ECL kit (HyGlo Denville Scientific IncMetuchen NJ)

23 Cell Viability Assays Cells in equal number were platedand exposed to hyperoxia as described above Cell viabilitywas quantified by CellTiter-Glo kit (Promega Madison WI)and MTT assay LDH release was measured by CytoTox 96NonRadioactive Cytotoxicity assay kit (Promega MadisonWI) Viability and LDH release was calculated as a percentageof increase or decrease over their respective roomair controls

24 Chromatin Immunoprecipitation (ChIP) Assays ChIPassays were performed using the EZ-ChIP assay kit (Milli-pore Billerica MA) Briefly AECs were exposed to eitherroom air or hyperoxia for indicated time points cross-linked with formaldehyde and chromatin fragmentationwas carried out as detailed in the kit procedure Dilutedsoluble chromatin solution was incubated with rabbit anti-NRF2 (Santa Cruz Biotechnology Santa Cruz CA) for18 h at 4∘C with rotation Nonimmune rabbit IgG wasused as a negative control to determine the binding speci-ficity Following incubation with protein AG agarose beadsthe bound products were washed and DNA was elutedDNA was subjected to PCR with primers encompass-ing the functional antioxidant response elements (AREs)located upstream of transcriptional start site of HMOX1(F 51015840-CCCTGCTGAGTAATCCTTTCCCGA-31015840 and R 51015840-ATGTCCCGACTCCAGACTCCA-31015840) and NQO1 (F 51015840-GTGGAAGTCGTCCCAAGAGA-31015840 and R 51015840-TGTCTCCCCAGGACTCTCTCAG-31015840) to determine the binding of NRF2in ChIP assays


100

89

100

58

100

45

0

20

40

60

80

100

120

NS

24 h 36 h 48 h

RAHyp

Cel

l via

bilit

y (

)(C

ellT

iter-

Glo

assa

y)

lowast

lowastsect

sect

(a)

100

97 100

63

100

51

0

20

40

60

80

100

120

Cel

l via

bilit

y (

)(M

TT as

say)

NS

RAHyp

lowast

lowast

sect

sect

24 h 36 h 48 h

(b)

100

141

100

208

100

234

0

50

100

150

200

250

300

LDH

rele

ase (

)

RAHyp

lowast

lowast

lowastsect

sect

24 h 36 h 48 h

(c)




3 Results






1 1533 39

121

0

5

10

15

20

Relat

ive m

RNA

expr

essio

n

Relat

ive m

RNA

expr

essio

n

Relat

ive m

RNA

expr

essio

n

1 11 09

23

36

0

2

4

6

1 17 1827

58

0

2

4

6

8

10

RAHyperoxia

RAHyperoxia

RAHyperoxia


lowastlowast

lowast

lowast

lowast

lowast

lowast

HMOX1 GCLC NQO1

(a)

1 1 1 18

17

09 1

8 23

10 1

91

0 14

0

2

4

6

8

10

12

14Re

lativ

e ban

d in

tens

ity3 h 6 h 12 hRA

120573-actin

120573-actin

120573-actin

RA 3 h 6 h 12 h

lowast

lowastlowast

lowastlowastlowast

HMOX1

GCLCNQO1

HMOX1

GCLC

NQO1

(b)





RA 36 h 48 h

1

63

113

0

5

10

15

20

1

75

155

0

5

10

15

20

25

Relat

ive m

RNA

expr

essio

n

1 575

246

0

50

100

150

200

250

300

350

RA 36 h 48 hRA 36 h 48 h

Relat

ive m

RNA

expr

essio

n

Relat

ive m

RNA

expr

essio

n

lowast

lowast

lowast

lowast

lowast

lowast

HMOX1 GCLC NQO1

(a)

Hyp RAHyp

36 h

RA

48 h

1 1 1 239

157

137 1 1 1

834

257

192

0

2

4

6

8

10

12

14

RA RAHyp Hyp

Rela

tive b

and

inte

nsity

36 h 48 h

lowastlowast

lowast

lowastlowastlowast

HMOX1GCLCNQO1

HMOX1

GCLC

NQO1

120573-actin

120573-actin

120573-actin

(b)






NRF2

Lamin B

RA 1 h 3 h 12 h

1

19

37

44

0

2

4

6

8

Rela

tive b

and

inte

nsity

RA 1 h 3 h 12 h

lowast

lowast

lowast

(a)

RA 36 h 48 h

NRF2

Lamin B

1

46

53

0

2

4

6

8

Rela

tive b

and

inte

nsity

RA 36 h 48 h

lowast

lowast

(b)








ARE AREminus618

minus449

minus9204

minus8963

HMOX1NQO1

(a)

113

27

1113

0

1

2

3

4

5

Rela

tive N

RF2

bind

ing

RA 12 h6 h3 h1 hRA 12 h6 h3 h1 h

IP NRF2

Input

1 22

135

12

224

0

5

10

15

20

25

30

Relat

ive N

RF2

bind

ing

IP NRF2

Input

RA 12 h6 h3 h1 hRA 12 h6 h3 h1 h

lowast

lowast

lowast

(b)

Input

IP NRF2


one H

3

Input

IP NRF2

RA Hyp RA Hyp IgG

Hist

one H

3

Relat

ive N

RF2

bind

ing

Relat

ive N

RF2

bind

ing

36 h 48 h 36 h 48 h


1 11 1 170

5

10

15

20

25

30

RA HypRAHyp

36 h 48 h

lowast1

18

1

24

0

1

2

3

4

5


lowast

lowast

(c)



Rela

tive l

ucife

rase

activ

ity

1 11

21

69

0

2

4

6

8

RA 12 h6 h3 h

lowast

lowast

(a)

115

0

2

4

6

8

RA 36 h

lowast

(b)



4 Discussion


















Acknowledgments


References























































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















100

89

100

58

100

45

0

20

40

60

80

100

120

NS

24 h 36 h 48 h

RAHyp

Cel

l via

bilit

y (

)(C

ellT

iter-

Glo

assa

y)

lowast

lowastsect

sect

(a)

100

97 100

63

100

51

0

20

40

60

80

100

120

Cel

l via

bilit

y (

)(M

TT as

say)

NS

RAHyp

lowast

lowast

sect

sect

24 h 36 h 48 h

(b)

100

141

100

208

100

234

0

50

100

150

200

250

300

LDH

rele

ase (

)

RAHyp

lowast

lowast

lowastsect

sect

24 h 36 h 48 h

(c)




3 Results






1 1533 39

121

0

5

10

15

20

Relat

ive m

RNA

expr

essio

n

Relat

ive m

RNA

expr

essio

n

Relat

ive m

RNA

expr

essio

n

1 11 09

23

36

0

2

4

6

1 17 1827

58

0

2

4

6

8

10

RAHyperoxia

RAHyperoxia

RAHyperoxia


lowastlowast

lowast

lowast

lowast

lowast

lowast

HMOX1 GCLC NQO1

(a)

1 1 1 18

17

09 1

8 23

10 1

91

0 14

0

2

4

6

8

10

12

14Re

lativ

e ban

d in

tens

ity3 h 6 h 12 hRA

120573-actin

120573-actin

120573-actin

RA 3 h 6 h 12 h

lowast

lowastlowast

lowastlowastlowast

HMOX1

GCLCNQO1

HMOX1

GCLC

NQO1

(b)





RA 36 h 48 h

1

63

113

0

5

10

15

20

1

75

155

0

5

10

15

20

25

Relat

ive m

RNA

expr

essio

n

1 575

246

0

50

100

150

200

250

300

350

RA 36 h 48 hRA 36 h 48 h

Relat

ive m

RNA

expr

essio

n

Relat

ive m

RNA

expr

essio

n

lowast

lowast

lowast

lowast

lowast

lowast

HMOX1 GCLC NQO1

(a)

Hyp RAHyp

36 h

RA

48 h

1 1 1 239

157

137 1 1 1

834

257

192

0

2

4

6

8

10

12

14

RA RAHyp Hyp

Rela

tive b

and

inte

nsity

36 h 48 h

lowastlowast

lowast

lowastlowastlowast

HMOX1GCLCNQO1

HMOX1

GCLC

NQO1

120573-actin

120573-actin

120573-actin

(b)






NRF2

Lamin B

RA 1 h 3 h 12 h

1

19

37

44

0

2

4

6

8

Rela

tive b

and

inte

nsity

RA 1 h 3 h 12 h

lowast

lowast

lowast

(a)

RA 36 h 48 h

NRF2

Lamin B

1

46

53

0

2

4

6

8

Rela

tive b

and

inte

nsity

RA 36 h 48 h

lowast

lowast

(b)








ARE AREminus618

minus449

minus9204

minus8963

HMOX1NQO1

(a)

113

27

1113

0

1

2

3

4

5

Rela

tive N

RF2

bind

ing

RA 12 h6 h3 h1 hRA 12 h6 h3 h1 h

IP NRF2

Input

1 22

135

12

224

0

5

10

15

20

25

30

Relat

ive N

RF2

bind

ing

IP NRF2

Input

RA 12 h6 h3 h1 hRA 12 h6 h3 h1 h

lowast

lowast

lowast

(b)

Input

IP NRF2


one H

3

Input

IP NRF2

RA Hyp RA Hyp IgG

Hist

one H

3

Relat

ive N

RF2

bind

ing

Relat

ive N

RF2

bind

ing

36 h 48 h 36 h 48 h


1 11 1 170

5

10

15

20

25

30

RA HypRAHyp

36 h 48 h

lowast1

18

1

24

0

1

2

3

4

5


lowast

lowast

(c)



Rela

tive l

ucife

rase

activ

ity

1 11

21

69

0

2

4

6

8

RA 12 h6 h3 h

lowast

lowast

(a)

115

0

2

4

6

8

RA 36 h

lowast

(b)



4 Discussion


















Acknowledgments


References























































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















1 1533 39

121

0

5

10

15

20

Relat

ive m

RNA

expr

essio

n

Relat

ive m

RNA

expr

essio

n

Relat

ive m

RNA

expr

essio

n

1 11 09

23

36

0

2

4

6

1 17 1827

58

0

2

4

6

8

10

RAHyperoxia

RAHyperoxia

RAHyperoxia


lowastlowast

lowast

lowast

lowast

lowast

lowast

HMOX1 GCLC NQO1

(a)

1 1 1 18

17

09 1

8 23

10 1

91

0 14

0

2

4

6

8

10

12

14Re

lativ

e ban

d in

tens

ity3 h 6 h 12 hRA

120573-actin

120573-actin

120573-actin

RA 3 h 6 h 12 h

lowast

lowastlowast

lowastlowastlowast

HMOX1

GCLCNQO1

HMOX1

GCLC

NQO1

(b)





RA 36 h 48 h

1

63

113

0

5

10

15

20

1

75

155

0

5

10

15

20

25

Relat

ive m

RNA

expr

essio

n

1 575

246

0

50

100

150

200

250

300

350

RA 36 h 48 hRA 36 h 48 h

Relat

ive m

RNA

expr

essio

n

Relat

ive m

RNA

expr

essio

n

lowast

lowast

lowast

lowast

lowast

lowast

HMOX1 GCLC NQO1

(a)

Hyp RAHyp

36 h

RA

48 h

1 1 1 239

157

137 1 1 1

834

257

192

0

2

4

6

8

10

12

14

RA RAHyp Hyp

Rela

tive b

and

inte

nsity

36 h 48 h

lowastlowast

lowast

lowastlowastlowast

HMOX1GCLCNQO1

HMOX1

GCLC

NQO1

120573-actin

120573-actin

120573-actin

(b)






NRF2

Lamin B

RA 1 h 3 h 12 h

1

19

37

44

0

2

4

6

8

Rela

tive b

and

inte

nsity

RA 1 h 3 h 12 h

lowast

lowast

lowast

(a)

RA 36 h 48 h

NRF2

Lamin B

1

46

53

0

2

4

6

8

Rela

tive b

and

inte

nsity

RA 36 h 48 h

lowast

lowast

(b)








ARE AREminus618

minus449

minus9204

minus8963

HMOX1NQO1

(a)

113

27

1113

0

1

2

3

4

5

Rela

tive N

RF2

bind

ing

RA 12 h6 h3 h1 hRA 12 h6 h3 h1 h

IP NRF2

Input

1 22

135

12

224

0

5

10

15

20

25

30

Relat

ive N

RF2

bind

ing

IP NRF2

Input

RA 12 h6 h3 h1 hRA 12 h6 h3 h1 h

lowast

lowast

lowast

(b)

Input

IP NRF2


one H

3

Input

IP NRF2

RA Hyp RA Hyp IgG

Hist

one H

3

Relat

ive N

RF2

bind

ing

Relat

ive N

RF2

bind

ing

36 h 48 h 36 h 48 h


1 11 1 170

5

10

15

20

25

30

RA HypRAHyp

36 h 48 h

lowast1

18

1

24

0

1

2

3

4

5


lowast

lowast

(c)



Rela

tive l

ucife

rase

activ

ity

1 11

21

69

0

2

4

6

8

RA 12 h6 h3 h

lowast

lowast

(a)

115

0

2

4

6

8

RA 36 h

lowast

(b)



4 Discussion


















Acknowledgments


References























































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















RA 36 h 48 h

1

63

113

0

5

10

15

20

1

75

155

0

5

10

15

20

25

Relat

ive m

RNA

expr

essio

n

1 575

246

0

50

100

150

200

250

300

350

RA 36 h 48 hRA 36 h 48 h

Relat

ive m

RNA

expr

essio

n

Relat

ive m

RNA

expr

essio

n

lowast

lowast

lowast

lowast

lowast

lowast

HMOX1 GCLC NQO1

(a)

Hyp RAHyp

36 h

RA

48 h

1 1 1 239

157

137 1 1 1

834

257

192

0

2

4

6

8

10

12

14

RA RAHyp Hyp

Rela

tive b

and

inte

nsity

36 h 48 h

lowastlowast

lowast

lowastlowastlowast

HMOX1GCLCNQO1

HMOX1

GCLC

NQO1

120573-actin

120573-actin

120573-actin

(b)






NRF2

Lamin B

RA 1 h 3 h 12 h

1

19

37

44

0

2

4

6

8

Rela

tive b

and

inte

nsity

RA 1 h 3 h 12 h

lowast

lowast

lowast

(a)

RA 36 h 48 h

NRF2

Lamin B

1

46

53

0

2

4

6

8

Rela

tive b

and

inte

nsity

RA 36 h 48 h

lowast

lowast

(b)








ARE AREminus618

minus449

minus9204

minus8963

HMOX1NQO1

(a)

113

27

1113

0

1

2

3

4

5

Rela

tive N

RF2

bind

ing

RA 12 h6 h3 h1 hRA 12 h6 h3 h1 h

IP NRF2

Input

1 22

135

12

224

0

5

10

15

20

25

30

Relat

ive N

RF2

bind

ing

IP NRF2

Input

RA 12 h6 h3 h1 hRA 12 h6 h3 h1 h

lowast

lowast

lowast

(b)

Input

IP NRF2


one H

3

Input

IP NRF2

RA Hyp RA Hyp IgG

Hist

one H

3

Relat

ive N

RF2

bind

ing

Relat

ive N

RF2

bind

ing

36 h 48 h 36 h 48 h


1 11 1 170

5

10

15

20

25

30

RA HypRAHyp

36 h 48 h

lowast1

18

1

24

0

1

2

3

4

5


lowast

lowast

(c)



Rela

tive l

ucife

rase

activ

ity

1 11

21

69

0

2

4

6

8

RA 12 h6 h3 h

lowast

lowast

(a)

115

0

2

4

6

8

RA 36 h

lowast

(b)



4 Discussion


















Acknowledgments


References























































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















NRF2

Lamin B

RA 1 h 3 h 12 h

1

19

37

44

0

2

4

6

8

Rela

tive b

and

inte

nsity

RA 1 h 3 h 12 h

lowast

lowast

lowast

(a)

RA 36 h 48 h

NRF2

Lamin B

1

46

53

0

2

4

6

8

Rela

tive b

and

inte

nsity

RA 36 h 48 h

lowast

lowast

(b)








ARE AREminus618

minus449

minus9204

minus8963

HMOX1NQO1

(a)

113

27

1113

0

1

2

3

4

5

Rela

tive N

RF2

bind

ing

RA 12 h6 h3 h1 hRA 12 h6 h3 h1 h

IP NRF2

Input

1 22

135

12

224

0

5

10

15

20

25

30

Relat

ive N

RF2

bind

ing

IP NRF2

Input

RA 12 h6 h3 h1 hRA 12 h6 h3 h1 h

lowast

lowast

lowast

(b)

Input

IP NRF2


one H

3

Input

IP NRF2

RA Hyp RA Hyp IgG

Hist

one H

3

Relat

ive N

RF2

bind

ing

Relat

ive N

RF2

bind

ing

36 h 48 h 36 h 48 h


1 11 1 170

5

10

15

20

25

30

RA HypRAHyp

36 h 48 h

lowast1

18

1

24

0

1

2

3

4

5


lowast

lowast

(c)



Rela

tive l

ucife

rase

activ

ity

1 11

21

69

0

2

4

6

8

RA 12 h6 h3 h

lowast

lowast

(a)

115

0

2

4

6

8

RA 36 h

lowast

(b)



4 Discussion


















Acknowledgments


References























































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















ARE AREminus618

minus449

minus9204

minus8963

HMOX1NQO1

(a)

113

27

1113

0

1

2

3

4

5

Rela

tive N

RF2

bind

ing

RA 12 h6 h3 h1 hRA 12 h6 h3 h1 h

IP NRF2

Input

1 22

135

12

224

0

5

10

15

20

25

30

Relat

ive N

RF2

bind

ing

IP NRF2

Input

RA 12 h6 h3 h1 hRA 12 h6 h3 h1 h

lowast

lowast

lowast

(b)

Input

IP NRF2


one H

3

Input

IP NRF2

RA Hyp RA Hyp IgG

Hist

one H

3

Relat

ive N

RF2

bind

ing

Relat

ive N

RF2

bind

ing

36 h 48 h 36 h 48 h


1 11 1 170

5

10

15

20

25

30

RA HypRAHyp

36 h 48 h

lowast1

18

1

24

0

1

2

3

4

5


lowast

lowast

(c)



Rela

tive l

ucife

rase

activ

ity

1 11

21

69

0

2

4

6

8

RA 12 h6 h3 h

lowast

lowast

(a)

115

0

2

4

6

8

RA 36 h

lowast

(b)



4 Discussion


















Acknowledgments


References























































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of

















Rela

tive l

ucife

rase

activ

ity

1 11

21

69

0

2

4

6

8

RA 12 h6 h3 h

lowast

lowast

(a)

115

0

2

4

6

8

RA 36 h

lowast

(b)



4 Discussion


















Acknowledgments


References























































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of






























Acknowledgments


References























































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of










































of






Disease Markers



OncologyJournal of





PPAR Research



Journal of

ObesityJournal of
















the nrf2 activation and antioxidative response are not impaired

Documents