Molecular Basis for Vulnerability to Mitochondrial and
Oxidative Stress in a Neuroendocrine CRI-G1 Cell Line
Natasha Chandiramani, Xianhong Wang, Marta Margeta*
Department of Pathology, University of California San Francisco, San Francisco, California, United States of America
Background: Many age-associated disorders (including diabetes, cancer, and neurodegenerative diseases) are linked to
mitochondrial dysfunction, which leads to impaired cellular bioenergetics and increased oxidative stress. However, it is not
known what genetic and molecular pathways underlie differential vulnerability to mitochondrial dysfunction observed
among different cell types.
Methodology/Principal Findings: Starting with an insulinoma cell line as a model for a neuronal/endocrine cell type, we
isolated a novel subclonal line (named CRI-G1-RS) that was more susceptible to cell death induced by mitochondrial
respiratory chain inhibitors than the parental CRI-G1 line (renamed CRI-G1-RR for clarity). Compared to parental RR cells, RS
cells were also more vulnerable to direct oxidative stress, but equally vulnerable to mitochondrial uncoupling and less
vulnerable to protein kinase inhibition-induced apoptosis. Thus, differential vulnerability to mitochondrial toxins between
these two cell types likely reflects differences in their ability to handle metabolically generated reactive oxygen species
rather than differences in ATP production/utilization or in downstream apoptotic machinery. Genome-wide gene expression
analysis and follow-up biochemical studies revealed that, in this experimental system, increased vulnerability to
mitochondrial and oxidative stress was associated with (1) inhibition of ARE/Nrf2/Keap1 antioxidant pathway; (2) decreased
expression of antioxidant and phase I/II conjugation enzymes, most of which are Nrf2 transcriptional targets; (3) increased
expression of molecular chaperones, many of which are also considered Nrf2 transcriptional targets; (4) increased expression
of b cell-specific genes and transcription factors that specify/maintain b cell fate; and (5) reconstitution of glucose-
stimulated insulin secretion.
Conclusions/Significance: The molecular profile presented here will enable identification of individual genes or gene
clusters that shape vulnerability to mitochondrial dysfunction and thus represent potential therapeutic targets for diabetes
and neurodegenerative diseases. In addition, the newly identified CRI-G1-RS cell line represents a new experimental model
for investigating how endogenous antioxidants affect glucose sensing and insulin release by pancreatic b cells.
Citation: Chandiramani N, Wang X, Margeta M (2011) Molecular Basis for Vulnerability to Mitochondrial and Oxidative Stress in a Neuroendocrine CRI-G1 Cell
Line. PLoS ONE 6(1): e14485. doi:10.1371/journal.pone.0014485
Editor: Kathrin Maedler, University of Bremen, Germany
Received January 21, 2010; Accepted August 6, 2010; Published January 4, 2011
Copyright: ? 2011 Chandiramani et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the National Institutes of Health grant NS054113 (http://www.ninds.nih.gov/). The funders had no role in study design,
data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: Marta.Margeta@ucsf.edu
Mitochondrial dysfunction has multifactorial etiology. In rare
but severe inherited mitochondrial disorders (which typically
present in childhood or early adulthood and affect metabolically
active organs such as the brain, heart, liver, and skeletal muscle),
mitochondrial dysfunction is a result of germline mutations in
nuclear or mitochondrial DNA. In contrast, accumulation of
somatic mitochondrial DNA mutations that accompanies aging or
an impairment in mitochondrial function caused by metabolic and
environmental factors are thought to contribute to the pathogen-
esis of many common age-associated disorders including diabetes,
cancer, and neurodegeneration . In diabetes, impaired
mitochondrial metabolism contributes to insulin resistance ob-
served in peripheral tissues [2,3], in part through increase in the
level of reactive oxygen species (ROS) . There is also
accumulating evidence that mitochondrial dysfunction blunts
glucose-stimulated insulin secretion (GSIS) in pancreatic b cells.
In a b cell line, for example, GSIS is inhibited following depletion
of native mitochondrial DNA and can be restored by repopulation
of cybrid cells with foreign mitochondrial DNA . Similarly,
GSIS impairment caused by a partial loss of the pancreatic
transcription factor PDX-1 (heterozygosity of which leads to a
form of the maturity-onset diabetes of the young) is mediated by
changes in the mitochondrial gene expression [6,7]. While it is
currently not understood how mitochondrial dysfunction leads to
the GSIS impairment, changes in the ROS metabolism are an
important candidate because mitochondrially produced ROS act
as a signal both in the hypothalamic glucose sensing  and the b
cell GSIS .
Given that mitochondrial dysfunction plays an important role in
the pathogenesis of both diabetes and neurodegeneration, it is
perhaps not surprising that epidemiologic studies have shown a
link between the two . For example, it is well established that
patients with diabetes have an increased risk of dementia and/or
Alzheimer’s disease (reviewed in ). To further dissect the
PLoS ONE | www.plosone.org1 January 2011 | Volume 6 | Issue 1 | e14485
relationship between these disorders, recent studies have stratified
patients into subcategories depending on the severity and duration
of diabetes on one hand and the subtype of dementia on the other.
In one such prospective cohort study, diabetes overall was
associated with increased risk of vascular dementia, while
borderline and undiagnosed diabetes were associated with
increased risk of Alzheimer’s disease . In another prospective
study, higher risk of Alzheimer’s disease was associated not with
low insulin sensitivity but with low early insulin response to oral
glucose challenge, a measure of insulin release ; this finding, in
particular, raises the possibility that the link between the two
diseases reflects an intracellular signaling defect in a pathway
common to neurons and pancreatic b cells. A link between
diabetes and Parkinson’s disease is also suggested by the
epidemiologic literature, but its nature is currently unclear: case
control studies performed thus far mostly showed that diabetes was
associated with a decreased risk of Parkinson’s disease, while
prospective cohort studies either showed that diabetes was
associated with an increased risk of Parkinson’s disease or that
there was no association between the two diseases (;  and
references therein). The discrepancy between these results raises
the possibility that – like Alzheimer’s disease – Parkinson’s disease
is not associated with the diabetic state per se, but with some
underlying metabolic or signaling abnormality that is shared
between neurons and b cells. While additional epidemiologic
studies with better patient stratification are required to clarify the
connections between diabetes and Parkinson’s disease, together
these studies underscore the underlying biological similarities
between neurons and b cells and highlight the importance of
identifying shared signaling pathways that modulate susceptibility
to mitochondrial dysfunction in neuroendocrine cells.
Here, two subclones of an insulinoma cell line that differ in the
susceptibility to mitochondrial and oxidative stressors were used to
identify gene expression changes associated with vulnerability to
mitochondrial dysfunction in the neuronal/endocrine cell type.
Isolation of a CRI-G1 cell line subclone that shows
increased vulnerability to mitochondrial and oxidative
CRI-G1 is one of four cell lines isolated from a transplantable
Cambridge Rat Islet cell tumor in 1985 . CRI-G1 cells are
rounded, have slender processes, and grow in clumps and ribbons
rather than forming an epithelial monolayer ( and Fig. 1A).
When maintained in culture for extended periods of time,
however, CRI-G1 cells flatten and become more epithelioid in
appearance (Fig. 1B); this transition is stochastic in nature, but can
be facilitated by growing cells at high densities. The differences
between the two CRI-G1 cell subtypes are not restricted to
morphology, however: compared to the parental clone, which is
resistant to cell death induced by mitochondrial complex I
inhibitor rotenone, the novel CRI-G1 cell subclone (which we
cryopreserved as a separate cell line after one of the stochastic
transition events) is highly susceptible to this mitochondrial toxin,
with 51.463.6% loss of viability and ,3 fold increase in release of
an intracellular protease (marker of cell death) following an
overnight treatment with a 1 mM rotenone (Figs. 2A and B). [In
this and all subsequent figures, the original clone is termed CRI-
G1-RR (for Rotenone-Resistant) and the novel subclone CRI-G1-
RS (for Rotenone-Susceptible).] Similar results were seen with
longer treatments (up to 5 days) and higher doses of rotenone (up
to 10 mM; not shown).
To determine whether CRI-G1-RS cell line is more susceptible
to cell death in general, we tested two additional mitochondrial
toxins (complex III inhibitor antimycin and complex V inhibitor
oligomycin) as well as staurosporine, a protein kinase inhibitor and
potent apoptosis inducer. Interestingly, CRI-G1-RS cells were
more susceptible to both antimycin- (Figs. 2C and D) and
oligomycin-induced cell death (Figs. 2E and 2F), but less
susceptible to staurosporine-induced apoptosis (Figs. 2G and H).
Of the three mitochondrial inhibitors tested, antimycin was most
toxic to CRI-G1-RS cells (maximal viability loss of 59.963.7%),
but no mitochondrial toxin led to 100% CRI-G1-RS cell loss
regardless of the length of the treatment (up to 5 days) or the dose
used (data not shown). In contrast, the maximal dose of
staurosporine (1 mM) was completely toxic to both cell types, with
lower susceptibility of CRI-G1-RS cells apparent only in the mid-
range part of the dose-response curve (Fig. 2G). Consistent with
the cell viability data, the increase in caspase 3/7 activity was
greater in CRI-G1-RR than in CRI-G1-RS cells at these mid-
range doses (100 and 300 nM; Fig. 2H). The three mitochondrial
inhibitors led to only minimal activation of caspase 3/7 in both cell
types (data not shown), suggesting that in this experimental system
cell death induced by mitochondrial inhibition is largely non-
apoptotic in nature.
In general, mitochondrial inhibition leads to (1) a decrease in
ATP synthesis, eventually resulting in ATP depletion and (2) an
Figure 1. Morphology of CRI-G1 cell lines. A. Parental CRI-G1 cells are rounded, have slender processes, and grow in clumps and ribbons rather
than forming an epithelial monolayer. In subsequent text and figures, this phenotype is termed CRI-G1-RR (for explanation of the nomenclature, see
Results section.) B. Prolonged culturing at high density conditions leads to a preponderance of cells with surface-adherent, epithelioid appearance
that grow in clusters or islands. In subsequent text and figures, this phenotype is termed CRI-G1-RS. Images were acquired by differential interference
contrast (DIC) microscopy of living cultures; scale bar, 50 mM.
Metabolic Stress Vulnerability
PLoS ONE | www.plosone.org2 January 2011 | Volume 6 | Issue 1 | e14485
Figure 2. Differential vulnerability to mitochondrial inhibition and staurosporine-induced apoptosis. CRI-G1-RS cells are more
vulnerable to cell death induced by mitochondrial respiratory chain inhibitors rotenone (panels A and B), antimycin (panels C and D), and oligomycin
(panels E and F), but less vulnerable to apoptosis induced by protein kinase inhibitor staurosporine (panels G and H). Cell viability (panels A, C, E and
G) was measured by CellTiter 96 AQueousOne Solution Cell Proliferation Assay (‘‘MTS assay’’), a variant of the classic MTT assay; averaged data from 3–
6 repeat experiments are shown in each panel. Cell death (panels B, D, and F) was determined by CytoTox-GloTMAssay, which measures the activity of
a ‘‘dead-cell’’ protease released into the media from membrane-compromised cells; a representative of at least 3 repeat experiments is shown in each
panel (n=4 in each experiment). Caspase 3/7 activity, an indicator of apoptosis, was measured by Caspase-GloTM3/7 Assay (panel H); a representative
of 3 repeat experiments is shown (n=4 in each experiment). Although MTS assay is based on mitochondrial metabolism and thus could be directly
affected by mitochondrial inhibitors, there was a good correspondence between viability and cell death assay results for all compounds tested.
Statistical significance was determined by two-way ANOVA, which showed highly significant effect of both treatment and cell type in all conditions
tested. Data are plotted as mean 6 S.E.M.; ***, p,0.001.
Metabolic Stress Vulnerability
PLoS ONE | www.plosone.org3 January 2011 | Volume 6 | Issue 1 | e14485
at RT; and washed for 45 min at RT. Following a final wash in
TBS with 0.1% Tween for 15 min at RT and a water rinse,
protein-antibody complexes were detected using an ECL chemi-
luminescent kit (Pierce Biotechnology) and CL-XPosure Film
(Thermo Scientific) with a Konica SRX-101A developer.
The following primary antibodies were used: rabbit polyclonal
anti-PARK7/DJ-1, 1 mg/mL (GenWay Biotech); mouse mono-
clonal anti-Keap1, 1 mg/mL (ProteinTech Group); mouse mono-
clonal anti-COXIV 20E8, 0.1 mg/mL (Abcam); mouse monoclo-
nal anti-cytochrome c, 0.5 mg/mL (BD Pharmingen); rabbit
polyclonal anti-Tom20 FL-145, 0.07 mg/mL, and rabbit poly-
clonal Nrf2 H300, 0.5 mg/mL (Santa Cruz Biotechnology); rabbit
polyclonal anti-Aldh2, 1:10000 (gift of Dr. Henry Weiner, Purdue
University); rabbit polyclonal anti-Akt and anti-phospho-Akt,
1:1000 (Cell Signaling); mouse monoclonal anti-tubulin clone
DM1A (ascites fluid), 1 mg/mL, and rabbit polyclonal anti-lamin
A C-terminal, 1 mg/mL (Sigma). Horseradish peroxidase-conju-
gated anti-mouse and anti-rabbit secondary H+L IgG antibodies
were purchased from Jackson ImmunoResearch; anti-mouse
antibody was used at 0.8 mg/mL, while anti-rabbit antibody was
used at either 8 mg/mL (for anti-PARK7/DJ-1 and anti-Nrf2) or
0.8 mg/mL (for anti-Tom20, anti-Aldh2, anti-Akt, anti-P-Akt. and
ImageJ software (http://rsb.info.nih.gov/ij/) was used for band
quantification by densitometry. Data were analyzed with Graph-
Pad Prism statistical software using two-tailed t-test (for compar-
ison of baseline protein levels) or two-way ANOVA with ad hoc
Bonferroni post-test (for comparison of drug treatment effects
between the two cell types).
Cells were grown overnight on poly-L-lysine-coated glass
coverslips, incubated with hypoxanthine (0.5 mM)/xanthine
oxidase (16 U/L) for 1 hour, washed once with PBS, and fixed
in 4% formaldehyde 4% sucrose in PBS for 15 minutes at room
temperature; coverslips were stored in PBS at 4uC until
immunostaining. Cells were blocked in PBS with 0.1% Triton-X
and 5% normal goat serum for 1 hour at room temperature,
incubated in blocking solution with 4 mg/mL rabbit anti-4-HNE
(Alpha Diagnostic) at 4uC overnight, washed 3 times in PBS with
0.1% Triton-X, and incubated in blocking solution with 2 mg/mL
Alexa Fluor 488-conjugated goat anti-rabbit IgG (H+L; Invitro-
gen) for 2 hours at room temperature. Coverslips were mounted
on glass slides with Vectashield mounting medium containing
DAPI nuclear counter stain (Vector Laboratories). Images were
taken with Zeiss LSM 510 imaging software (version 4.2) using
206 and 636 objectives on Zeiss LSM 510 NLO confocal
microscope; they were edited with Adobe Photoshop CS3 Version
The concentration of reduced (GSH) and total (GSH+GSSG)
glutathione was determined using the GSH-GloTMGlutathione
Assay kit (Promega) according to the manufacturer’s instructions;
this luminescence-based assay is based on the conversion of a
luciferin derivative into luciferin, catalyzed by glutathione S-
transferase in the presence of glutathione. Frozen cell pellets were
resuspended in 200 mL ice-cold phosphate-buffered saline supple-
mented with CompleteTMProtease Inhibitor Cocktail (Roche).
10 mL aliquot of each cell suspension (3 technical replicates/
sample) was used to measure GSH. Total glutathione was
measured by adding reducing agent TCEP (Pierce; 1 mM final
concentration) to 3 separate aliquots of each suspension to convert
GSSG to GSH prior to GSH measurement; the amount of TCEP
added did not significantly alter the total volume. Readings were
then normalized to protein concentration measured by CBB kit
(Dojindo) in a lysate prepared from an aliquot of each cell
suspension (as described for whole cell lysate in the Western
blotting section). The amount of GSSG in the sample was
calculated by halving the difference between the total and reduced
glutathione; two RS cell samples that had negative/undetectable
GSSG levels were excluded from the GSH/GSSG ratio analysis.
Data were analyzed with GraphPad Prism statistical software
using two-way ANOVA followed by ad-hoc Bonferroni post-tests.
Cells were plated in a 12-well dish (56104cells/well). The next
day, cells were washed twice with KRBB 2% BSA (Krebs-Ringer
Bicarbonate Buffer – 130 mM NaCl, 5 mM KCl, 1.25 mM
KH2PO4, 1.25 mM MgSO4, 2.68 mM CaCl2, 5.26 mM NaH-
CO3,10 mM HEPES; pH=7.4), incubated in KRBB 2% BSA at
37uC for 3 h, and then incubated in KRBB 2% BSA with or
without 20 mM glucose for 30 min; the supernatant from 30 min
incubation, which contained the secreted insulin, was collected,
spun down to remove debris, and stored in aliquots at 280uC. To
measure residual intracellular insulin, cells that remained in the
dish were scraped, pelleted, and lysed in 200 mL acid ethanol
(0.2 M HCl in 75% ethanol) overnight at 4uC. The lysate was
spun down to remove debris, neutralized with 0.2 M NaOH,
aliquoted, and stored at 280uC. Insulin was measured using the
Rat/Mouse Insulin ELISA kit (Millipore); values for both secreted
and intracellular insulin were normalized to the total DNA content
(measured by DNA spectrometry) in each lysate. Data were
analyzed with GraphPad Prism statistical software using two-way
ANOVA followed by ad-hoc Bonferroni post-tests.
Found at: doi:10.1371/journal.pone.0014485.s001 (0.13 MB PPT)
Found at: doi:10.1371/journal.pone.0014485.s002 (0.25 MB PPT)
Found at: doi:10.1371/journal.pone.0014485.s003 (0.10 MB PPT)
Found at: doi:10.1371/journal.pone.0014485.s004 (2.07 MB
Found at: doi:10.1371/journal.pone.0014485.s005 (0.83 MB PPT)
Found at: doi:10.1371/journal.pone.0014485.s006 (0.09 MB PPT)
Found at: doi:10.1371/journal.pone.0014485.s007 (2.13 MB
Found at: doi:10.1371/journal.pone.0014485.s008 (3.27 MB PPT)
Found at: doi:10.1371/journal.pone.0014485.s009 (2.41 MB PPT)
Found at: doi:10.1371/journal.pone.0014485.s010 (0.08 MB PPT)
Found at: doi:10.1371/journal.pone.0014485.s011 (0.04 MB
Metabolic Stress Vulnerability
PLoS ONE | www.plosone.org 16January 2011 | Volume 6 | Issue 1 | e14485
Found at: doi:10.1371/journal.pone.0014485.s012 (0.03 MB
We thank Ms. Linda Ta (Gladstone Institutes Genomic Core Facility) for
assistance with microarray hybridization, Dr. Jane Fridlyand and Ms. Ritu
Roy (UCSF Cancer Center Biostatistics Core) for statistical analysis of
microarray data, and Ms. Christine Lin for help with figure design. In
addition, we are grateful to Dr. Lily Jan (HHMI and UCSF Department of
Physiology) for support of the initial phase of the project, Dr. Henry
Weiner (Purdue University) for the gift of the anti-Aldh2 antiserum, Dr. Ha
Il Kim from the German lab (UCSF Diabetes Center) for help with insulin
ELISA, Dr. Angela Brennan from the Swanson lab (UCSF Department of
Neurology and San Francisco VAMC) for help with 4-HNE immuno-
chemistry, and to Dr. Igor Mitrovic and members of the Margeta lab for
valuable comments on the manuscript.
Conceived and designed the experiments: MM. Performed the experi-
ments: NC XW MM. Analyzed the data: NC XW MM. Wrote the paper:
1. Wallace DC (2005) A mitochondrial paradigm of metabolic and degenerative
diseases, aging, and cancer: a dawn for evolutionary medicine. Annu Rev Genet
2. Petersen KF, Befroy D, Dufour S, Dziura J, Ariyan C, et al. (2003)
Mitochondrial dysfunction in the elderly: possible role in insulin resistance.
Science 300: 1140–1142.
3. Petersen KF, Dufour S, Befroy D, Garcia R, Shulman GI (2004) Impaired
mitochondrial activity in the insulin-resistant offspring of patients with type 2
diabetes. N Engl J Med 350: 664–671.
4. Houstis N, Rosen ED, Lander ES (2006) Reactive oxygen species have a causal
role in multiple forms of insulin resistance. Nature 440: 944–948.
5. Soejima A, Inoue K, Takai D, Kaneko M, Ishihara H, et al. (1996)
Mitochondrial DNA is required for regulation of glucose-stimulated insulin
secretion in a mouse pancreatic beta cell line, MIN6. J Biol Chem 271:
6. Brissova M, Shiota M, Nicholson WE, Gannon M, Knobel SM, et al. (2002)
Reduction in pancreatic transcription factor PDX-1 impairs glucose-stimulated
insulin secretion. J Biol Chem 277: 11225–11232.
7. Gauthier BR, Brun T, Sarret EJ, Ishihara H, Schaad O, et al. (2004)
Oligonucleotide microarray analysis reveals PDX1 as an essential regulator of
mitochondrial metabolism in rat islets. J Biol Chem 279: 31121–31130.
8. Leloup C, Magnan C, Benani A, Bonnet E, Alquier T, et al. (2006)
Mitochondrial reactive oxygen species are required for hypothalamic glucose
sensing. Diabetes 55: 2084–2090.
9. Pi J, Bai Y, Zhang Q, Wong V, Floering LM, et al. (2007) Reactive oxygen
species as a signal in glucose-stimulated insulin secretion. Diabetes 56:
10. Ristow M (2004) Neurodegenerative disorders associated with diabetes mellitus.
J Mol Med 82: 510–529.
11. Kopf D, Frolich L (2009) Risk of incident Alzheimer’s disease in diabetic
patients: a systematic review of prospective trials. J Alzheimers Dis 16: 677–685.
12. Xu WL, von Strauss E, Qiu CX, Winblad B, Fratiglioni L (2009) Uncontrolled
diabetes increases the risk of Alzheimer’s disease: a population-based cohort
study. Diabetologia 52: 1031–1039.
13. Ronnemaa E, Zethelius B, Sundelof J, Sundstrom J, Degerman-Gunnarsson M,
et al. (2009) Glucose metabolism and the risk of Alzheimer’s disease and
dementia: a population-based 12 year follow-up study in 71-year-old men.
Diabetologia 52: 1504–1510.
14. Hu G, Jousilahti P, Bidel S, Antikainen R, Tuomilehto J (2007) Type 2 diabetes
and the risk of Parkinson’s disease. Diabetes Care 30: 842–847.
15. D’Amelio M, Ragonese P, Callari G, Di Benedetto N, Palmeri B, et al. (2009)
Diabetes preceding Parkinson’s disease onset. A case-control study. Parkinson-
ism Relat Disord 15: 660–664.
16. Carrington CA, Rubery ED, Pearson EC, Hales CN (1986) Five new insulin-
producing cell lines with differing secretory properties. J Endocrinol 109:
17. Hoglinger GU, Carrard G, Michel PP, Medja F, Lombes A, et al. (2003)
Dysfunction of mitochondrial complex I and the proteasome: interactions
between two biochemical deficits in a cellular model of Parkinson’s disease.
J Neurochem 86: 1297–1307.
18. Moon Y, Lee KH, Park JH, Geum D, Kim K (2005) Mitochondrial membrane
depolarization and the selective death of dopaminergic neurons by rotenone:
protective effect of coenzyme Q10. J Neurochem 93: 1199–1208.
19. Fonck C, Baudry M (2003) Rapid reduction of ATP synthesis and lack of free
radical formation by MPP+in rat brain synaptosomes and mitochondria. Brain
Res 975: 214–221.
20. Moriscot C, Candel S, Sauret V, Kerr-Conte J, Richard MJ, et al. (2007)
MnTMPyP, a metalloporphyrin-based superoxide dismutase/catalase mimetic,
protects INS-1 cells and human pancreatic islets from an in vitro oxidative
challenge. Diabetes Metab 33: 44–53.
21. Brennan AM, Suh SW, Won SJ, Narasimhan P, Kauppinen TM, et al. (2009)
NADPH oxidase is the primary source of superoxide induced by NMDA
receptor activation. Nat Neurosci 12: 857–863.
22. Sies H, Akerboom TP (1984) Glutathione disulfide (GSSG) efflux from cells and
tissues. Methods Enzymol 105: 445–451.
23. Beissbarth T, Speed TP (2004) GOstat: find statistically overrepresented Gene
Ontologies within a group of genes. Bioinformatics 20: 1464–1465.
24. Kensler TW, Wakabayashi N, Biswal S (2007) Cell survival responses to
environmental stresses via the Keap1-Nrf2-ARE pathway. Annu Rev Pharmacol
Toxicol 47: 89–116.
25. Nguyen T, Yang CS, Pickett CB (2004) The pathways and molecular
mechanisms regulating Nrf2 activation in response to chemical stress. Free
Radic Biol Med 37: 433–441.
26. Clements CM, McNally RS, Conti BJ, Mak TW, Ting JP (2006) DJ-1, a cancer-
and Parkinson’s disease-associated protein, stabilizes the antioxidant transcrip-
tional master regulator Nrf2. Proc Natl Acad Sci U S A 103: 15091–15096.
27. Malhotra D, Thimmulappa R, Navas-Acien A, Sandford A, Elliott M, et al.
(2008) Decline in NRF2-regulated antioxidants in chronic obstructive pulmo-
nary disease lungs due to loss of its positive regulator, DJ-1. Am J Respir Crit
Care Med 178: 592–604.
28. Gan L, Johnson DA, Johnson JA (2010) Keap1-Nrf2 activation in the presence
and absence of DJ-1. Eur J Neurosci 31: 967–977.
29. Andres-Mateos E, Perier C, Zhang L, Blanchard-Fillion B, Greco TM, et al.
(2007) DJ-1 gene deletion reveals that DJ-1 is an atypical peroxiredoxin-like
peroxidase. Proc Natl Acad Sci U S A 104: 14807–14812.
30. Shendelman S, Jonason A, Martinat C, Leete T, Abeliovich A (2004) DJ-1 is a
redox-dependent molecular chaperone that inhibits alpha-synuclein aggregate
formation. PLoS Biol 2: e362.
31. Bonifati V, Rizzu P, van Baren MJ, Schaap O, Breedveld GJ, et al. (2003)
Mutations in the DJ-1 gene associated with autosomal recessive early-onset
parkinsonism. Science 299: 256–259.
32. Inberg A, Linial M (2010) Protection of pancreatic b-cells from various stress
conditions is mediated by DJ-1. J Biol Chem Epub ahead of print (PMID:
33. Chen CH, Budas GR, Churchill EN, Disatnik MH, Hurley TD, et al. (2008)
Activation of aldehyde dehydrogenase-2 reduces ischemic damage to the heart.
Science 321: 1493–1495.
34. Wang B, Wang J, Zhou S, Tan S, He X, et al. (2008) The association of
mitochondrial aldehyde dehydrogenase gene (ALDH2) polymorphism with
susceptibility to late-onset Alzheimer’s disease in Chinese. J Neurol Sci 268:
35. Ohsawa I, Nishimaki K, Yasuda C, Kamino K, Ohta S (2003) Deficiency in a
mitochondrial aldehyde dehydrogenase increases vulnerability to oxidative stress
in PC12 cells. J Neurochem 84: 1110–1117.
36. Ohsawa I, Nishimaki K, Murakami Y, Suzuki Y, Ishikawa M, et al. (2008) Age-
dependent neurodegeneration accompanying memory loss in transgenic mice
defective in mitochondrial aldehyde dehydrogenase 2 activity. J Neurosci 28:
37. Cho HY, Reddy SP, Debiase A, Yamamoto M, Kleeberger SR (2005) Gene
expression profiling of NRF2-mediated protection against oxidative injury. Free
Radic Biol Med 38: 325–343.
38. Reisman SA, Yeager RL, Yamamoto M, Klaassen CD (2009) Increased Nrf2
activation in livers from Keap1-knockdown mice increases expression of
cytoprotective genes that detoxify electrophiles more than those that detoxify
reactive oxygen species. Toxicol Sci 108: 35–47.
39. Thimmulappa RK, Mai KH, Srisuma S, Kensler TW, Yamamoto M, et al.
(2002) Identification of Nrf2-regulated genes induced by the chemopreventive
agent sulforaphane by oligonucleotide microarray. Cancer Res 62: 5196–
40. Hu R, Xu C, Shen G, Jain MR, Khor TO, et al. (2006) Gene expression profiles
induced by cancer chemopreventive isothiocyanate sulforaphane in the liver of
C57BL/6J mice and C57BL/6J/Nrf2 (2/2) mice. Cancer Lett 243: 170–192.
41. Cerf ME (2006) Transcription factors regulating b-cell function. Eur J Endocrinol
42. Miyazaki S, Yamato E, Miyazaki J (2004) Regulated expression of pdx-1
promotes in vitro differentiation of insulin-producing cells from embryonic stem
cells. Diabetes 53: 1030–1037.
43. Pi J, Zhang Q, Fu J, Woods CG, Hou Y, et al. (2009) ROS signaling, oxidative
stress and Nrf2 in pancreatic beta-cell function. Toxicol Appl Pharmacol 244:
Metabolic Stress Vulnerability
PLoS ONE | www.plosone.org17 January 2011 | Volume 6 | Issue 1 | e14485
44. Tiedge M, Lortz S, Drinkgern J, Lenzen S (1997) Relation between antioxidant
enzyme gene expression and antioxidative defense status of insulin-producing
cells. Diabetes 46: 1733–1742.
45. Donath MY, Ehses JA, Maedler K, Schumann DM, Ellingsgaard H, et al. (2005)
Mechanisms of b-cell death in type 2 diabetes. Diabetes 54 Suppl 2: S108–113.
46. Rolo AP, Palmeira CM (2006) Diabetes and mitochondrial function: role of
hyperglycemia and oxidative stress. Toxicol Appl Pharmacol 212: 167–178.
47. Mattson MP, Cheng A (2006) Neurohormetic phytochemicals: Low-dose toxins
that induce adaptive neuronal stress responses. Trends Neurosci 29: 632–639.
48. Gier B, Krippeit-Drews P, Sheiko T, Aguilar-Bryan L, Bryan J, et al. (2009)
Suppression of KATPchannel activity protects murine pancreatic b cells against
oxidative stress. J Clin Invest 119: 3246–3256.
49. Liss B, Haeckel O, Wildmann J, Miki T, Seino S, et al. (2005) K-ATP channels
promote the differential degeneration of dopaminergic midbrain neurons. Nat
Neurosci 8: 1742–1751.
50. Nakaso K, Yano H, Fukuhara Y, Takeshima T, Wada-Isoe K, et al. (2003) PI3K
is a key molecule in the Nrf2-mediated regulation of antioxidative proteins by
hemin in human neuroblastoma cells. FEBS Lett 546: 181–184.
51. Wang L, Chen Y, Sternberg P, Cai J (2008) Essential roles of the PI3 kinase/Akt
pathway in regulating Nrf2-dependent antioxidant functions in the RPE. Invest
Ophthalmol Vis Sci 49: 1671–1678.
52. Li X, Chen H, Epstein PN (2006) Metallothionein and catalase sensitize to
diabetes in nonobese diabetic mice: reactive oxygen species may have a
protective role in pancreatic b-cells. Diabetes 55: 1592–1604.
53. McClung JP, Roneker CA, Mu W, Lisk DJ, Langlais P, et al. (2004)
Development of insulin resistance and obesity in mice overexpressing cellular
glutathione peroxidase. Proc Natl Acad Sci U S A 101: 8852–8857.
54. Makar TK, Nedergaard M, Preuss A, Gelbard AS, Perumal AS, et al. (1994)
Vitamin E, ascorbate, glutathione, glutathione disulfide, and enzymes of
glutathione metabolism in cultures of chick astrocytes and neurons: evidence
that astrocytes play an important role in antioxidative processes in the brain.
J Neurochem 62: 45–53.
55. Raps SP, Lai JC, Hertz L, Cooper AJ (1989) Glutathione is present in high
concentrations in cultured astrocytes but not in cultured neurons. Brain Res 493:
56. Kraft AD, Johnson DA, Johnson JA (2004) Nuclear factor E2-related factor 2-
dependent antioxidant response element activation by tert-butylhydroquinone
and sulforaphane occurring preferentially in astrocytes conditions neurons
against oxidative insult. J Neurosci 24: 1101–1112.
57. R Development Core Team (2005) R: A language and environment for
statistical computing. R Foundation for Statistical Computing, Vienna, Austria.
58. Bolstad BM (2004) Low level analysis of high-density oligonucleotide array data:
Background, normalization and summarization. PhD Dissertation. University of
59. Irizarry RA, Hobbs B, Collin F, Beazer-Barclay YD, Antonellis KJ, et al. (2003)
Exploration, normalization, and summaries of high density oligonucleotide array
probe level data. Biostatistics 4: 249–264.
60. Smyth GK (2005) Limma: linear models for microarray data. In: Gentleman R,
Carey V, Dudoit S, Irizarry RA, Huber W, eds. Bioinformatics and
Computational Biology Solutions using R and Bioconductor. New York:
61. Smyth GK (2004) Linear models and empirical Bayes methods for assessing
differential expression in microarray experiments. Stat Appl Genet Mol Biol 3:
62. Gentleman RC, Carey VJ, Bates DM, Bolstad B, Dettling M, et al. (2004)
Bioconductor: open software development for computational biology and
bioinformatics. Genome Biol 5: R80.
63. Benjamini Y, Hochberg Y (1995) Controlling the false recovery rate: A practical
and powerful approach to multiple testing. J Royal Stat Soc B 57: 289–300.
Metabolic Stress Vulnerability
PLoS ONE | www.plosone.org 18 January 2011 | Volume 6 | Issue 1 | e14485