doi:10.

1093/jmcb/mjs025 Journal of Molecular Cell Biology (2012), 4, 221– 230 | 221

Published online May 19, 2012

Article
Age- and diet-dependent requirement of DJ-1
for glucose homeostasis in mice with implications
for human type 2 diabetes
Deepak Jain1,† Ruchi Jain1,†, Daniel Eberhard1, Jan Eglinger1,2, Marco Bugliani3, Lorenzo Piemonti4,
Piero Marchetti3, and Eckhard Lammert1,2, *
1
Institute of Metabolic Physiology, Heinrich-Heine University, D-40225 Düsseldorf, Germany

Downloaded from http://jmcb.oxfordjournals.org/ at Lund University Libraries, Head Office on August 18, 2015
2
Paul Langerhans Group for Beta Cell Biology, German Diabetes Center (DDZ), D-40225 Düsseldorf, Germany
3
Metabolic Unit, Department of Endocrinology and Metabolism, Cisanello Hospital, 56100 Pisa, Italy
4
Laboratory of Experimental Surgery, Surgical Department, San Raffaele Scientific Institute, 20132 Milano, Italy
†
These authors contributed equally to this work.
* Correspondence to: Eckhard Lammert, E-mail: lammert@uni-duesseldorf.de

Elderly patients often suffer from multiple age-related diseases. Here we show that the expression of DJ-1, an antioxidant protein
with reduced expression in the central nervous system of patients with Parkinson’s disease, is reduced in pancreatic islets of
patients with type 2 diabetes mellitus (T2DM). In contrast, under non-diabetic conditions, DJ-1 expression increases in mouse
and human islets during aging. In mouse islets, we show that DJ-1 prevents an increase in reactive oxygen species levels as the
mice age. This antioxidant function preserves mitochondrial integrity and physiology, prerequisites for glucose-stimulated
insulin secretion. Accordingly, DJ-1-deficient mice develop glucose intolerance and reduced b cell area as they age or gain
weight. Our data suggest that DJ-1 is more generally involved in age- and lifestyle-related human diseases and show for the first
time that DJ-1 plays a key role in glucose homeostasis and might serve as a novel drug target for T2DM.

Keywords: diabetes, pancreatic islets, DJ-1, mitochondria

Introduction
DJ-1 is a multifunctional, highly conserved antioxidant protein
encoded by Parkinson’s disease (PD) gene park7 (Bonifati
et al., 2003; Dawson and Dawson, 2003; Guzman et al., 2010;
Hong et al., 2010). In neurons, DJ-1 is upregulated upon oxidative
stress and protects neurons from mitochondrial dysfunction (Hao
et al., 2010; McCoy and Cookson, 2012). Interestingly, in mouse
pancreatic islets and mouse insulinoma cells (MIN6), which
have a low expression of antioxidant proteins (Tiedge et al.,
1997; Robertson and Harmon, 2006; Wiederkehr and Wollheim,
2006; Lenzen, 2008), DJ-1 was recently shown to be specifically
upregulated under hyperglycemic conditions in vitro and in vivo
(Waanders et al., 2009; Inberg and Linial, 2010).
Pancreatic islets are aggregates of endocrine cells of which b
cells form the majority (Maedler, 2008; Collombat et al., 2010).
An increase in blood glucose concentration stimulates the b
cells to secrete insulin, which lowers blood glucose to normal
levels. Pancreatic islets and neurons have many features in
common (van Arensbergen et al., 2010; Soltani et al., 2011),
Received February 9, 2012. Revised April 13, 2012. Accepted April 19, 2012.
# The Author (2012). Published by Oxford University Press on behalf of Journal of
Molecular Cell Biology, IBCB, SIBS, CAS. All rights reserved.
and recently it was shown that diabetics have a 40% increased
risk of developing PD (Xu et al., 2011), suggesting a possible
link between type 2 diabetes mellitus (T2DM) and PD. Here, we
studied the role of DJ-1 in pancreatic islet function, glucose
homeostasis, and T2DM.
Results
Age-dependent upregulation of DJ-1 in human pancreatic islets
and reduced DJ-1 expression in islets from T2DM patients
We first showed that, in the human pancreas, DJ-1 is strongly
expressed in islets compared with the surrounding exocrine
tissue (Figure 1A – D). In addition, DJ-1 protein expression
almost doubled in human pancreatic islets under hyperglycemic
conditions (Figure 1E). When comparing islets from 38 to 56
years old and 68 to 81 years old, we also discovered that the
mRNA expression of DJ-1 significantly increased in the islets of
non-diabetic humans during aging (Figure 1F and Supple-
mentary Table S1). In contrast, DJ-1 expression was significantly
reduced on both the mRNA and protein levels in the islets from
T2DM patients compared with islets from non-diabetic controls
(Figure 1G and H, and Supplementary Tables S1 and S2). The
decline in DJ-1 protein expression might have been even larger
when age-matched T2DM patients and non-diabetic controls
222 | Journal of Molecular Cell Biology Jain et al.

Downloaded from http://jmcb.oxfordjournals.org/ at Lund University Libraries, Head Office on August 18, 2015
Figure 1 Age-dependent upregulation of DJ-1 in human pancreatic islets and reduced DJ-1 expression in islets from T2DM patients. (A– D) LSM
images of a section through human pancreas showing an islet (*) surrounded by exocrine pancreatic tissue, stained for cell nuclei (DAPI) (A),
DJ-1 (B), insulin (C), and both DJ-1 and insulin (D). Scale bar, 100 mm. (E) DJ-1 protein expression (normalized to b-actin) in human pancreatic
islets cultured at low (5.5 mM) and high (16.5 mM) glucose concentrations for 6 h as determined by ELISA. The islets were isolated from a male
donor (65 years, BMI 25.2, ischemic stroke). n ¼ 3. (F) Relative DJ-1 mRNA expression in human islets from non-diabetic individuals as deter-
mined by real-time RT– PCR. n ¼ 3– 4. (G) Comparison of relative DJ-1 mRNA expression in pancreatic islets from non-diabetic controls and
T2DM patients using real-time RT –PCR. n ¼ 3 –4. (H) DJ-1 protein expression (normalized to b-actin) in human pancreatic islets obtained
from non-diabetic controls and T2DM patients as determined by ELISA. n ¼ 3. **P , 0.01, *P , 0.05 (two-tailed Student’s t-test). All values
are means + SD. For further information on islet donors, see Supplementary Tables S1 and S2.

had been available for the analysis of human islet lysates. In
summary, our data show that DJ-1 expression normally increases
in pancreatic islets during aging, but that, in the islets of patients
with T2DM, its expression decreases.
Age-dependent requirement of DJ-1 for reducing reactive oxygen
species in pancreatic islets
To investigate the role of DJ-1 in vivo, we turned to the mouse
and showed that in the mouse pancreas as in the human pan-
creas, DJ-1 is more strongly expressed in islets compared with
the surrounding exocrine tissue (Figure 2A –D). As in humans,
mouse DJ-1 expression was significantly higher in islets at both
the mRNA and protein levels during aging, as shown by compar-
ing islets taken from mice at 8 and 12– 13 weeks of age (Figure 2E
and F). Although DJ-1 mRNA expression remained unchanged up
to the age of 8 weeks, the expression of DJ-1 increased from 8 to
20 weeks of age (Supplementary Figure S1). In contrast, there was
no age-dependent change in the mRNA level for any other antioxi-
dant protein we studied, i.e. superoxide dismutase 1 and 2 (sod1,
sod2), glutathione peroxidase 1 (gpx1), and uncoupling protein 2
(ucp2) (Supplementary Figure S2).
Since the loss of DJ-1 was previously shown to increase mito-
chondrial reactive oxygen species (ROS) production in neurons
(Andres-Mateos et al., 2007; Guzman et al., 2010), we next
studied whether DJ-1 deficiency increased ROS in mouse pancre-
atic islets. Surprisingly, no significant increase in ROS levels was
observed in islets isolated from DJ-1-deficient mice at the age of 8
weeks (Figure 2G), showing that DJ-1 was not constitutively
required for lowering ROS levels in islets. In contrast, DJ-1 was ex-
plicitly required for lowering ROS levels in mouse islets when the
mice were 12 – 13 weeks of age (Figure 2G, I, and J), suggesting
that DJ-1 keeps ROS at constant levels as islets age. Moreover,
the levels of ROS could be normalized by culturing
DJ-1-deficient islets in the presence of the redox modulator
N-acetyl-L-cysteine (NAC) (Figure 2H and K).
Reduced expression of DJ-1 increases mitochondrial ROS
production in mouse insulinoma cells
Next, to determine the cellular compartment where ROS accu-
mulated in the absence of DJ-1, we transfected glucose-
responsive mouse insulinoma cells (MIN6), a model cell line for
b cells, with a redox-sensitive variant of green fluorescent
protein (roGFP2) containing a mitochondrial matrix targeting se-
quence, or with a cytoplasmic roGFP2 (Hanson et al., 2004).
Comparison of untreated MIN6 cells with fully reduced [dithio-
threitol (DTT)-treated] and fully oxidized (Aldrithiol-treated) cells
suggested that MIN6 cells normally contain low levels of ROS in
both mitochondria and cytosol (Figure 3A and B). We then
lowered DJ-1 protein expression in these cells using two different
small interfering RNAs (siRNA) (Supplementary Figure S3).
Importantly, a reduced expression of DJ-1 protein in MIN6 cells
significantly increased ROS levels in mitochondria as determined
in living cells using Laser Scanning Microscopy (LSM) (Figure 3C,
and Figure 3E – G vs H – J and K– M). We confirmed the specificity of
the knockdown by showing that normal ROS levels could be
restored in DJ-1-silenced cells by transfection with a DJ-1 cDNA
Role of DJ-1 in type 2 diabetes Journal of Molecular Cell Biology | 223

Downloaded from http://jmcb.oxfordjournals.org/ at Lund University Libraries, Head Office on August 18, 2015
Figure 2 Age-dependent requirement of DJ-1 for reducing ROS in mouse pancreatic islets. (A –D) LSM images of sections through mouse pan-
creas showing an islet (*) surrounded by exocrine pancreatic tissue, stained for cell nuclei (DAPI) (A), DJ-1 (B), insulin (C), and both DJ-1 and
insulin (D). Scale bar, 100 mm. (E) Relative DJ-1 mRNA expression measured by real-time RT –PCR of islets isolated from C57BL/6J mice at 8
and 12– 13 weeks of age. n ¼ 3. (F) Densitometric western blot analysis of DJ-1 protein expression in islets isolated from C57BL/6J mice at
8 and 12– 13 weeks of age. Representative protein bands of western blots probed with anti-DJ-1 and -GAPDH antibodies are shown below.
n ¼ 6. (G and H) Relative levels of ROS measured by CM-H2DCFDA fluorescence intensities in islets isolated from control and DJ-1-deficient
mice at 8 and 12 –13 weeks of age in the absence (G) or presence (H) of NAC. n ¼ 3. (I – K) LSM images of optical sections through pancreatic
islets (*) isolated from control mice (I), DJ-1-deficient mice (J), and DJ-1-deficient mice after treatment of the islets with NAC for 3 days (K). ROS
were stained with CM-H2DCFDA. Scale bar, 100 mm. **P , 0.01 (two-tailed Student’s t-test). All values are means + SD.

(Figure 3C, and Figure 3K – M vs N – P). In contrast, cytoplasmic
ROS were not significantly increased (Figure 3D). The data
suggest that DJ-1 is specifically required for reducing ROS in
the mitochondria of insulin-secreting b cells.
Age-dependent requirement of DJ-1 for mitochondrial integrity,
function, and glucose-stimulated insulin secretion from
pancreatic islets
Since elevated ROS levels can result in mitochondrial fragmenta-
tion (Mai et al., 2010; Nikic et al., 2011), we studied whether DJ-1
was required for normal mitochondrial morphology in pancreatic
b cells. To this end, we compared the mitochondria of DJ-1-
deficient b cells with those of controls in situ (Figure 4A–H).
A marked fragmentation was found to accompany the increased
ROS production in pancreatic b cells (Figure 4B vs F). While the
majority of wild-type mitochondria formed networks in the
primary b cells, the mitochondria in DJ-1-deficient b cells were
often fragmented (Figure 4A –D vs E– H). Moreover, significantly
more MIN6 cells displayed fragmented mitochondria when DJ-1
expression was reduced using siRNA (Supplementary Figure
S4A – D and I). However, in the presence of the redox modulator
NAC, mitochondrial morphology was rescued (Supplementary
Figure S4E – H and J), suggesting that mitochondrial fragmenta-
tion arose from defects induced by ROS accumulation in b cells
silenced for DJ-1.
As a measure of mitochondrial function, we then determined
adenosine triphosphate (ATP) levels in DJ-1-deficient islets as
well as in MIN6 cells silenced for DJ-1. In b cells, ATP was
shown to be a key metabolic signal from mitochondria that trans-
lates elevated nutrient levels into increased insulin secretion
(Maechler and Wollheim, 2001; Ashcroft, 2005). Interestingly,
ATP levels were not altered in DJ-1-deficient islets under low-
glucose conditions (Figure 4I, white columns). In contrast, ATP
224 | Journal of Molecular Cell Biology Jain et al.

Downloaded from http://jmcb.oxfordjournals.org/ at Lund University Libraries, Head Office on August 18, 2015
Figure 3 Reduction in DJ-1 expression increases mitochondrial ROS production in mouse insulinoma cells. (A – D) Degree of oxidation of roGFP2
measured in mouse insulinoma cells (MIN6) transfected with mitochondrial roGFP2 (A), or cytoplasmic roGFP2 (B), under reducing (DTT) or
oxidizing (Aldrithiol-2) conditions. The cells were also transfected with control siRNA, DJ-1 siRNA-1, DJ-1 siRNA-2, and DJ-1 siRNA-2 + DJ-1
cDNA expression vector (rescue) (C and D). *P , 0.05, **P , 0.01 (Non-parametic test by Kruskal– Wallis analysis of variance with Dunnet’s
post hoc test for comparison of multiple groups). n ¼ 15 cells each. (E –P) LSM live cell images of MIN6 cells taken at 530 nm after excitation
at 405 and 488 nm. The ratios of the emitted fluorescence intensities (405/488 nm) are also shown, since a higher ratio (405/488 nm) indicates
an increased amount of ROS in mitochondria. MIN6 cells were transfected as indicated. Scale bar, 10 mm.

levels in response to stimulatory high-glucose concentrations
were significantly lower in DJ-1-deficient islets compared with
controls (Figure 4I, black columns), and addition of NAC partially
rescued the ATP production under high-glucose conditions
(Figure 4J). Similar results were obtained using MIN6 cells
silenced for DJ-1 (Supplementary Figure S4K and L).
Since ATP is an important signal for glucose-stimulated insulin
secretion (GSIS; Lowell and Shulman, 2005; Leibiger and
Berggren, 2006; Muoio and Newgard, 2008), we studied
whether the reduction of ATP production led to reduced GSIS
from DJ-1-deficient islets. As shown in Figure 4K, GSIS was not
significantly changed in islets isolated from DJ-1-deficient mice
at the age of 8 weeks, showing that DJ-1 was not constitutively
required for GSIS. In contrast, DJ-1-deficient islets from 12 to 13
weeks old mice showed a significant reduction in GSIS compared
with controls (Figure 4K), which was not due to lower insulin
content (Figure 4L).
To demonstrate that DJ-1 is required for GSIS at this age
because of its effect on mitochondria and the glucose-stimulated
increase in ATP levels, we tested glibenclamide-induced insulin
secretion from DJ-1-deficient islets taken from 12 to 13 weeks
old mice (Figure 4M). Glibenclamide, a sulphonylurea, is one of
the two oral anti-diabetic drugs in the WHO list of essential med-
icines for treatment of T2DM. Like other sulphonylureas, it closes
the KATP channels of b cells independent of an increase in ATP
levels (Ashcroft, 2005). Importantly, glibenclamide was found to
induce insulin secretion from pancreatic islets regardless of the
presence of DJ-1 (Figure 4M, grey columns), suggesting that
DJ-1 acts upstream of KATP channels in pancreatic b cells. Since
islets harbor endocrine cells other than b cells (Thorel et al.,
2010), we also used MIN6 cells to corroborate our results
obtained with pancreatic islets (Supplementary Figure S5). In
these cells, DJ-1 was similarly required for glucose-stimulated
rather than glibenclamide-induced insulin secretion
(Supplementary Figure S5A and B).
Age- and diet-dependent glucose intolerance and reduced b cell
area in the absence of DJ-1
Glucose intolerance can be a consequence of b cell dysfunction
and/or reduced b cell area (Ueki et al., 2006; Poy et al., 2009;
Sachdeva et al., 2009; Liew et al., 2010; Esterhazy et al., 2011).
Since we observed reduced b cell function in DJ-1-deficient
animals, we evaluated glucose tolerance and b cell area in
Role of DJ-1 in type 2 diabetes Journal of Molecular Cell Biology | 225

Downloaded from http://jmcb.oxfordjournals.org/ at Lund University Libraries, Head Office on August 18, 2015
Figure 4 Age-dependent requirement of DJ-1 for mitochondrial integrity, function and GSIS from mouse pancreatic islets. (A – H) LSM images of
islets in the pancreas of control (A – D) and DJ-1-deficient mice (E – H) stained with DAPI (A and E), anti-VDAC1 antibodies (B and F), and anti-
insulin antibodies (C, G). Merged images are shown in D and H. Scale bars, 10 mm. (I and J) ATP production in islets from control and
DJ-1-deficient mice. Data are presented as percentage of basal control in the absence (I) or presence of NAC (J). n ¼ 3. (K) Insulin secretion
(normalized to total insulin content) from mouse islets at a low (2 mM) and a high (25 mM) glucose concentration. Islets were isolated from
DJ-1-deficient mice and controls at 8 and 12– 13 weeks of age. n ¼ 3. (L) Total insulin content (normalized to total protein content) of islets
isolated from DJ-1-deficient mice and controls at 8 and 12– 13 weeks of age. n ¼ 3. (M) Insulin secretion (normalized to total insulin
content) from mouse islets at a low-glucose concentration (2 mM), a high-glucose concentration (25 mM), and at a low-glucose concentration
(2 mM) in the presence of 1 mM glibenclamide. n ¼ 3. *P , 0.05 (two-tailed Student’s t-test). All values are means + SD.

these animals (Figure 5). Consistent with the age-dependent
changes in islet physiology, DJ-1-deficient mice did not show
any defect in glucose-stimulated plasma insulin (Figure 5A),
glucose tolerance (Figure 5B), and b cell area (Figure 5C) at the
age of 8 weeks. At 12– 13 weeks of age, however, DJ-1-deficient
mice had significantly lower levels of plasma insulin after
glucose challenge (Figure 5D), displayed glucose intolerance
(Figure 5E), and had a small but significant reduction in b cell
area (Figure 5F).
To investigate whether glucose intolerance could also be
induced in younger animals, we placed 6 weeks old mice on a
high-fat diet for 2 weeks. DJ-1 mRNA and protein levels were ele-
vated in the islets from these mice compared with islets from mice
kept on a chow diet (Supplementary Figure S6). In contrast, there
was no diet-dependent change in mRNA level for any other anti-
oxidant protein we studied (Supplementary Figure S6A).
Importantly, plasma insulin was significantly reduced upon
glucose challenge in the 8 weeks old mice that had been kept
226 | Journal of Molecular Cell Biology Jain et al.

Downloaded from http://jmcb.oxfordjournals.org/ at Lund University Libraries, Head Office on August 18, 2015
Figure 5 Age- and diet-dependent glucose intolerance and reduced b cell area in the absence of DJ-1. (A, D, and G) Plasma insulin levels of male
DJ-1-deficient mice and controls before (0 min), and 40 min after intraperitoneal administration of glucose (2 g/kg) at 8 weeks of age (A), 12– 13
weeks of age (D) and at 8 weeks of age after 2 weeks on high-fat diet (HFD; G). n ¼ 7 –9. (B, E, and H) Glucose tolerance tests of male mice
having the indicated genotype after intraperitoneal administration of glucose (2 g/kg) at 8 weeks of age (B), 12 –13 weeks of age (E), and at 8
weeks of age after 2 weeks on HFD (H). n ¼ 6 – 10. (C, F, and I) Morphometric analyses of relative b cell mass from DJ-1-deficient mice and
controls quantified and calculated as insulin-positive area in total area of evenly spaced pancreatic sections at 8 weeks of age (C), 12– 13
weeks of age (F) and at 8 weeks of age after 2 weeks on HFD (I). n ¼ 3. *P , 0.05, **P , 0.01 (two-tailed Student’s t-test). All values are
means + SD. (J) Model. During aging and upon weight gain, DJ-1 is upregulated in pancreatic b cells to maintain ROS at constant low
levels, so they do not disturb mitochondrial morphology and function. Thus, elevated blood glucose levels accurately increase the intracellular
ATP concentration, a prerequisite for GSIS and maintenance of normal glucose tolerance (right green panel). Following a failure to increase DJ-1
expression, b cells do not efficiently reduce ROS levels upon aging and weight gain. The increased amount of ROS not only affects the morph-
ology of the mitochondria, but also compromises their function. This lowers the ability of b cells to produce ATP in response to glucose stimu-
lation and results in glucose intolerance, a first step in the development of T2DM (left red panel).

on a high-fat diet for 2 weeks (Figure 5A vs G). Additionally,
glucose intolerance was detected (Figure 5B vs H), and the b
cell area of the fat mice was reduced to a small but significant
extent compared with that of the age-matched controls that had
been kept on a chow diet (Figure 5C vs I). This metabolic pheno-
type was observed in male mice (Figure 5) and, to a lesser but sig-
nificant extent, in female mice (Supplementary Figure S7). The
data show that DJ-1 is specifically required for GSIS, normal b
cell area and glucose tolerance during aging and on a high-fat
diet.
Discussion
DJ-1 is an antioxidant protein, and its loss leads to an early
onset of PD (Bonifati et al., 2003). In neurons, DJ-1 deficiency
was shown to increase ROS as well as mitochondrial fragmenta-
tion and dysfunction (Surmeier et al., 2011). DJ-1 is a highly
conserved protein that is expressed in bacteria, plants, and
animals (Lin et al., 2011). A few different scenarios have been pro-
posed for how DJ-1 reduces high ROS levels or oxidative stress in
cells. For example, DJ-1 has a conserved cysteine residue
(Cys106), which allows DJ-1 to act both as an ROS scavenger
and sensor of cellular redox homeostasis (Canet-Aviles et al.,
2004). It has also been suggested that DJ-1 acts as a transcrip-
tional co-activator (Inberg and Linial, 2010), stabilizes the
NF-E2-related factor (Nrf2) (Clements et al., 2006), and is an inte-
gral mitochondrial protein that maintains the activity of mitochon-
drial complex I in mammalian cells (Hayashi et al., 2009). In plant
cells, it was recently shown that DJ-1 interacts with glutathione
peroxidase (gpx) and superoxide dismutase 1 (SOD1), and acti-
vates the latter (Xu et al., 2010).
We have shown previously that the DJ-1 protein is significantly
upregulated in the islets of mice when they are fed with a high-fat
Role of DJ-1 in type 2 diabetes
diet and become hyperglycemic (Waanders et al., 2009). In add-
ition, DJ-1 was found to be upregulated in mouse islets cultured
in vitro under hyperglycemic conditions (Waanders et al., 2009;
Inberg and Linial, 2010). However, the role of DJ-1 in maintaining
pancreatic islet function and glucose homeostasis in vivo
remained unknown. Furthermore, no information was available
on the role of DJ-1 in islets during aging and on a high-fat diet.
The data of this study now suggest that DJ-1 is required to main-
tain physiological ROS levels in an age- and diet-dependent
manner (Figure 5J).
Physiological levels of ROS are important to maintain glucose
homeostasis, whereas prolonged exposure to high ROS levels
results in chronic oxidative stress and impaired glucose homeo-
stasis (Robertson, 2004; Wang and Hai, 2011). Importantly, pan-
creatic islets are particularly vulnerable to oxidative stress as they
express low levels of antioxidant proteins (Lenzen et al., 1996;
Tiedge et al., 1997). In addition, increased ROS levels are
observed in animal models of T2DM (i.e. db/db mice, GK rats,
and ZDF rats) and human T2DM patients (Ihara et al., 1999;
Kaneto et al., 1999; Tanaka et al., 1999).
Mitochondria are a major source of ROS, and mutations in mito-
chondrial DNA in humans and b-cell-specific deletion of mito-
chondrial genes in animal models cause T2DM in humans and
mice, respectively (Silva et al., 2000; Maassen et al., 2004). Our
data point to mitochondria as a main site of elevated ROS produc-
tion in pancreatic b cells during aging, and we show that DJ-1 is
required to reduce mitochondrial fragmentation in pancreatic b
cells during aging and on a high-fat diet. This fragmentation
might contribute to the reduced ATP levels and impaired GSIS
observed in DJ-1-deficient islets, as mitochondrial fission proteins
have been shown to influence ATP levels and GSIS in pancreatic
islets and b cell lines (Park et al., 2008; Twig et al., 2008).
Oxidative stress and mitochondrial dysfunction are also the
hallmarks of PD (Surmeier et al., 2011), and a recent report indi-
cates an 40% increase in prevalence of PD in T2DM patients (Xu
et al., 2011). In turn, several studies reveal glucose intolerance in
patients with PD (Aviles-Olmos et al., 2012). The link between PD
and T2DM might lie in the disruption of common molecular
mechanisms. Alternatively, PD may simply result from hypergly-
cemia, hyperinsulinemia, or both. In support of the first scenario,
we showed that DJ-1 is not only expressed in the substantia nigra,
which gets functionally impaired during PD (Surmeier et al.,
2011), but is also expressed in human pancreatic islets, which
become dysfunctional during T2DM. Similar to the situation in
PD where the concentration of DJ-1 decreases in the cerebrospinal
fluid (Hong et al., 2010), we found that in the islets of elderly
T2DM patients, DJ-1 expression is decreased compared with age-
matched controls. A possible reason for a lower expression of DJ-1
during T2DM might be an epigenetic silencing, since several
genes have been shown to be epigenetically downregulated in
the islets from T2DM patients (Ling et al., 2008; Yang et al., 2011).
Since DJ-1 regulates the androgen receptor (Niki et al., 2003)
that binds to the tyrosine hydroxylase (TH) promoter, active in
both the neurons of the substantia nigra and the beta cells of pan-
creatic islets, we also studied whether gender-specific differences
in DJ-1 gene expression could be detected. Interestingly, 74 + 7
years old females expressed significantly more DJ-1 in their
Journal of Molecular Cell Biology | 227
pancreatic islets than 79 + 3 years old males (Supplementary
Table S3), a finding that might help us to explain gender-specific
differences in the incidence and phenotype of PD and T2DM, re-
spectively (Ding et al., 2006; Wirdefeldt et al., 2011).
Based on our findings, we now propose that DJ-1 expression
increases in pancreatic b cells upon aging and a high-fat diet to
prevent a pathological increase in mitochondrial ROS levels
(Figure 5J, right panel). This function helps in preserving mito-
chondrial morphology and physiology during aging and weight
gain, thus maintaining normal ATP-dependent GSIS and glucose
tolerance (Figure 5J, right panel). In contrast, when DJ-1 expres-
sion fails to increase, ROS levels begin to increase beyond a
normal level, thereby altering mitochondrial morphology and
function in pancreatic islets. This results in glucose intolerance,
Downloaded from http://jmcb.oxfordjournals.org/ at Lund University Libraries, Head Office on August 18, 2015
a first step in the development of T2DM (Figure 5J, left panel).
Since we identified T2DM as a human disease where DJ-1 expres-
sion is significantly reduced in elderly humans, similar to the situ-
ation observed in PD, our data point to a more general role of the
oxidative stress modulator DJ-1 in age- and lifestyle-related
human diseases.
Finally, our findings suggest that DJ-1 expression could be tar-
geted for the treatment of T2DM. Along this line, it has recently
been shown that the histone deacetylase inhibitor, phenylbuty-
rate increases the expression of DJ-1 by 300% in the N27 dopa-
mine cell line and that it is useful for the treatment of both PD
and glucose intolerance (Ozcan et al., 2006; Xiao et al., 2011;
Zhou et al., 2011). Besides phenylbutyrate, substances have
been developed to specifically increase the activity of DJ-1 by pre-
venting its hyperoxidation and inactivation (Inden et al., 2011;
Kitamura et al., 2011). Since we showed that upregulation of
DJ-1 in pancreatic islets is key to maintain GSIS and blood
glucose levels during aging and weight gain, our study now war-
rants investigations on compounds that increase DJ-1 expression
or activity for the treatment of T2DM, the most common metabolic
disease worldwide.
Materials and methods
Human islet isolation and analysis of DJ-1 expression
Human pancreatic islets were prepared with the approval of the
local ethics committees from the pancreases of multi-organ
donors by collagenase digestion followed by density gradient
purification. Then, the islets were maintained in M199 culture
medium, containing 5.5 mM glucose, supplemented with 10%
(v/v) serum and 100 U/ml penicillin, 100 mg/ml streptomycin,
50 mg/ml gentamicin, and 750 ng/ml amphotericin B
(Sigma-Aldrich). Total RNA was extracted, quantified, and its in-
tegrity was assessed using an Agilent 2100 Bioanalyzer (Agilent
Technologies). For semiquantitative RT– PCR experiments, 1 mg
total RNA was reverse transcribed with iScript cDNA Synthesis
Kit (Bio-Rad Laboratories), and the following primers were used
for PCR: DJ-qPCR_huRP1 (5′ -GACCACATCACGGCTACACT-3′ ) and
DJ-1-qPCR_huFP1 (5′ -GAGCTGGGATTAAGGTCACCG-3′ ). The level
of mRNA for DJ-1 was quantified using an Applied Biosystems
7700 Bioanalyzer (Applied Biosystems) and normalized to the
housekeeping gene hypoxanthine-guanine phosphoribosyltrans-
ferase. DJ-1 and b-actin protein levels were measured in human
islet lysates using specific enzyme-linked immunosorbent assay
228 | Journal of Molecular Cell Biology
(ELISA) kits (MBL international cooperation; Cell signaling).
Mouse model and glucose tolerance test
Control (C57BL/6J) and DJ-1-deficient (B6.Cg-Park7tm1Shn/J)
mice were purchased from Jackson Laboratory (Goldberg et al.,
2005), and fed with standard laboratory chow or a high-fat diet
(Research Diets INC), and drinking water ad libitum. All experi-
ments were approved by the local Animal Ethics Committee. If
not mentioned, the age of mice was always more than 12
weeks. Glucose tolerance tests were carried out after the mice
had been fasted overnight.
Cell culture and RNA interference
MIN6 cells (Miyazaki et al., 1990) were used from passage 29–
35 and were electroporated via nucleofection (Lonza) according
to the manufacturer’s instructions with either pEGFP-DJ-1 or
pc-DNA3-FLAG-DJ-1 plasmid (kindly provided by Dr Hiroyoshi
Ariga at the University of Hokkaido, Japan) and/or siRNA. siRNA
against DJ-1 was prepared as previously described (Nikolova
et al., 2006; Konstantinova et al., 2007), and knockdown efficien-
cies were tested using western blots.
Insulin secretion and content
Islets and MIN6 cells were maintained at 378C for 1 h in 2 mM
glucose, 25 mM glucose, or 2 mM glucose with 1 mM glibencla-
mide in Krebs-Ringer HEPES buffer. The amount of insulin
secreted into the buffer was measured using an ultrasensitive
rat insulin ELISA (Crystal Chem.). Total insulin content was deter-
mined in lysates of islets or MIN6 cells using the same ELISA kit.
Measurement of ROS levels
ROS levels were measured in islets using the ROS-sensitive
fluorescent dye CM-H2DCFDA (Invitrogen). After starvation in
2 mM glucose, islets were stimulated with 25 mM glucose con-
taining 10 mM CM-H2DCFDA. Islets were homogenized and after
a centrifugation step, the fluorescence was measured in the
supernatant using a microplate reader (Tecan, infinite M1000).
The fluorescence intensity was also detected in intact pancreatic
islets using an LSM 710 confocal microscope with an excitation
wavelength of 488 nm and emission wavelength at 520 nm. In
addition, ROS levels were measured in MIN6 cells co-transfected
with roGFP2 plasmid (kindly provided by the University of Oregon,
Eugene, OR, USA) and siRNA against DJ-1 or control siRNA. Cells
were starved for 1 h in KRH buffer containing 2 mM glucose and
then treated with 25 mM glucose, 25 mM glucose containing
10 mM DTT to reduce or 500 mM Aldrithiol-2 (Ald) to oxidize the
samples. The fluorescence intensities (I) of cells were measured
at excitation wavelengths of 405 and 488 nm and emission wave-
lengths of 530 nm using an LSM 710 confocal microscope. The
fluorescence intensities were determined using ImageJ, and the
degree of oxidation of roGFP2 was determined using fluorescence
measurements after fully reducing the samples with DTT and then
fully oxidizing with Aldrithiol-2 using
OxDroGFP2 = R − Rred /[I488 ox /I488 red (Rox − R) + (R − Rred )]
(Meyer and Dick, 2010), where R is the ratio of the fluorescence
intensities of the sample at the excitation wavelengths of 405
and 488 nm, Rred is the ratio of the intensities of the fully
reduced sample at the excitation wavelengths of 405 and
488 nm, and Rox is the ratio of the intensities of the fully oxidized
Jain et al.
sample at the excitation wavelengths of 405 and 488 nm. I488ox
and I488red are the fluorescence intensities at the excitation
wavelength of 488 nm of the fully oxidized and fully reduced
sample, respectively.
Immunohistochemistry, immunocytochemistry, morphometric,
and western blot analyses
Immunofluorescence studies were performed with antibodies
against DJ-1 (Santa Cruz) and insulin (DAKO) on 4%
paraformaldehyde fixed pancreas sections. Secondary antibodies
were conjugated with AlexaFluor488 (Molecular Probes) and with
Cy5 (Dianova). DAPI (Sigma) was used to stain cell nuclei.
Mitochondria of cells in culture were stained using MitoTracker
Deep Red 633 (Molecular Probes). In addition, the mitochondria
of b cells in pancreatic sections were stained using an antibody
Downloaded from http://jmcb.oxfordjournals.org/ at Lund University Libraries, Head Office on August 18, 2015
against VDAC 1 (Abcam; voltage-dependent anion channel).
Pancreatic sections and MIN6 cells were imaged using a Carl
Zeiss LSM 710 confocal microscope and quantified using the soft-
ware ImageJ/Fiji. Rabbit anti-DJ-1 (Santa Cruz) and rabbit
anti-GAPDH antibodies (Abcam) were used for western blots.
The intensities of DJ-1 signal were normalized to the signal of
GAPDH and presented as percent of basal control.
Measurement of cellular ATP levels
Cellular ATP levels were measured using the ApoSENSOR ATP
assay kit (Biovision) according to the manufacturer’s instruc-
tions. Luminescence was determined using an LB9507 lumin-
ometer (Berthold Technologies) and normalized to total protein
levels.
Supplementary material
Supplementary material is available at Journal of Molecular Cell
Biology online.
Acknowledgements
We thank all our colleagues in Düsseldorf, in particular
B. Bartosinska for the islet isolation, and T. Dick in Heidelberg
for helpful advice and discussions.
Funding
This work was supported by the Collaborative Research Center
590 (SFB 590, B11 to R.J., D.E., E.L.), the European Foundation for
the Study of Diabetes and Lilly Research Fellowship (EFSD to D.J.),
and the German Center for Diabetes Research of the Federal
Ministry for Education and Research (DZD to J.E., E.L.).
Conflict of interest: none declared.
References
Andres-Mateos, E., Perier, C., Zhang, L., et al. (2007). DJ-1 gene deletion
reveals that DJ-1 is an atypical peroxiredoxin-like peroxidase. Proc. Natl
Acad. Sci. USA 104, 14807 – 14812.
Ashcroft, F.M. (2005). ATP-sensitive potassium channelopathies: focus on
insulin secretion. J. Clin. Invest. 115, 2047 – 2058.
Aviles-Olmos, I., Limousin, P., Lees, A., et al. (2012). Parkinson’s disease,
insulin resistance and novel agents of neuroprotection. Brain, doi:
10.1093/brain/aws009.
Bonifati, V., Rizzu, P., van Baren, M.J., et al. (2003). Mutations in the DJ-1 gene
associated with autosomal recessive early-onset parkinsonism. Science
299, 256 – 259.
Role of DJ-1 in type 2 diabetes
Canet-Aviles, R.M., Wilson, M.A., Miller, D.W., et al. (2004). The Parkinson’s
disease protein DJ-1 is neuroprotective due to cysteine-sulfinic acid-driven
mitochondrial localization. Proc. Natl Acad. Sci. USA 101, 9103 –9108.
Clements, C.M., McNally, R.S., Conti, B.J., et al. (2006). DJ-1, a cancer- and
Parkinson’s disease-associated protein, stabilizes the antioxidant transcrip-
tional master regulator Nrf2. Proc. Natl Acad. Sci. USA 103, 15091 –15096.
Collombat, P., Xu, X., Heimberg, H., et al. (2010). Pancreatic beta-cells: from
generation to regeneration. Semin. Cell Dev. Biol. 21, 838 – 844.
Dawson, T.M., and Dawson, V.L. (2003). Molecular pathways of neurodegen-
eration in Parkinson’s disease. Science 302, 819 – 822.
Ding, E.L., Song, Y., Malik, V.S., et al. (2006). Sex differences of endogenous
sex hormones and risk of type 2 diabetes: a systematic review and
meta-analysis. J. Am. Med. Assoc. 295, 1288– 1299.
Esterhazy, D., Stutzer, I., Wang, H., et al. (2011). Bace2 is a beta cell-enriched
protease that regulates pancreatic beta cell function and mass. Cell Metab.
14, 365 –377.
Goldberg, M.S., Pisani, A., Haburcak, M., et al. (2005). Nigrostriatal dopamin-
ergic deficits and hypokinesia caused by inactivation of the familial
Parkinsonism-linked gene DJ-1. Neuron 45, 489 –496.
Guzman, J.N., Sanchez-Padilla, J., Wokosin, D., et al. (2010). Oxidant stress
evoked by pacemaking in dopaminergic neurons is attenuated by DJ-1.
Nature 468, 696 – 700.
Hanson, G.T., Aggeler, R., Oglesbee, D., et al. (2004). Investigating mitochon-
drial redox potential with redox-sensitive green fluorescent protein indica-
tors. J. Biol. Chem. 279, 13044 –13053.
Hao, L.Y., Giasson, B.I., and Bonini, N.M. (2010). DJ-1 is critical for mitochon-
drial function and rescues PINK1 loss of function. Proc. Natl Acad. Sci.
USA 107, 9747 – 9752.
Hayashi, T., Ishimori, C., Takahashi-Niki, K., et al. (2009). DJ-1 binds to mito-
chondrial complex I and maintains its activity. Biochem. Biophys. Res.
Commun. 390, 667 – 672.
Hong, Z., Shi, M., Chung, K.A., et al. (2010). DJ-1 and alpha-synuclein in human
cerebrospinal fluid as biomarkers of Parkinson’s disease. Brain 133,
713 – 726.
Ihara, Y., Toyokuni, S., Uchida, K., et al. (1999). Hyperglycemia causes oxida-
tive stress in pancreatic beta-cells of GK rats, a model of type 2 diabetes.
Diabetes 48, 927 – 932.
Inberg, A., and Linial, M. (2010). Protection of pancreatic beta-cells from
various stress conditions is mediated by DJ-1. J. Biol. Chem. 285,
25686 – 25698.
Inden, M., Kitamura, Y., Takahashi, K., et al. (2011). Protection against dopa-
minergic neurodegeneration in Parkinson’s disease-model animals by a
modulator of the oxidized form of DJ-1, a wild-type of familial Parkinson’s
disease-linked PARK7. J. Pharmacol. Sci. 117, 189 – 203.
Kaneto, H., Kajimoto, Y., Miyagawa, J., et al. (1999). Beneficial effects of anti-
oxidants in diabetes: possible protection of pancreatic beta-cells against
glucose toxicity. Diabetes 48, 2398 –2406.
Kitamura, Y., Watanabe, S., Taguchi, M., et al. (2011). Neuroprotective effect of
a new DJ-1-binding compound against neurodegeneration in Parkinson’s
disease and stroke model rats. Mol. Neurodegener. 6, 1– 19.
Konstantinova, I., Nikolova, G., Ohara-Imaizumi, M., et al. (2007).
EphA-Ephrin-A-mediated beta cell communication regulates insulin secre-
tion from pancreatic islets. Cell 129, 359 – 370.
Leibiger, I.B., and Berggren, P.O. (2006). Sirt1: a metabolic master switch that
modulates lifespan. Nat. Med. 12, 34– 36.
Lenzen, S. (2008). Oxidative stress: the vulnerable beta-cell. Biochem. Soc.
Trans. 36, 343 – 347.
Lenzen, S., Drinkgern, J., and Tiedge, M. (1996). Low antioxidant enzyme gene
expression in pancreatic islets compared with various other mouse tissues.
Free Radic. Biol. Med. 20, 463 – 466.
Liew, C.W., Bochenski, J., Kawamori, D., et al. (2010). The pseudokinase trib-
bles homolog 3 interacts with ATF4 to negatively regulate insulin exocytosis
in human and mouse beta cells. J. Clin. Invest. 120, 2876 – 2888.
Lin, J., Nazarenus, T.J., Frey, J.L., et al. (2011). A plant DJ-1 homolog is essential
for Arabidopsis thaliana chloroplast development. PLoS One 6, e23731.
Ling, C., Del Guerra, S., Lupi, R., et al. (2008). Epigenetic regulation of
PPARGC1A in human type 2 diabetic islets and effect on insulin secretion.
Diabetologia 51, 615 – 622.
Journal of Molecular Cell Biology | 229
Lowell, B.B., and Shulman, G.I. (2005). Mitochondrial dysfunction and type 2
diabetes. Science 307, 384 – 387.
Maassen, J.A., ‘t Hart, L.M., Van Essen, E., et al. (2004). Mitochondrial dia-
betes: molecular mechanisms and clinical presentation. Diabetes
53(Suppl 1), S103 – S109.
Maechler, P., and Wollheim, C.B. (2001). Mitochondrial function in normal and
diabetic beta-cells. Nature 414, 807 –812.
Maedler, K. (2008). Beta cells in type 2 diabetes—a crucial contribution to
pathogenesis. Diabetes Obes. Metab. 10, 408– 420.
Mai, S., Klinkenberg, M., Auburger, G., et al. (2010). Decreased expression of
Drp1 and Fis1 mediates mitochondrial elongation in senescent cells and
enhances resistance to oxidative stress through PINK1. J. Cell Sci. 123,
917 – 926.
McCoy, M.K., and Cookson, M.R. (2012). Mitochondrial quality control and dy-
namics in Parkinson’s disease. Antioxid. Redox Signal. 16, 869 –882.
Meyer, A.J., and Dick, T.P. (2010). Fluorescent protein-based redox probes.
Downloaded from http://jmcb.oxfordjournals.org/ at Lund University Libraries, Head Office on August 18, 2015
Antioxid. Redox Signal. 13, 621 –650.
Miyazaki, J., Araki, K., Yamato, E., et al. (1990). Establishment of a pancreatic
beta cell line that retains glucose-inducible insulin secretion: special refer-
ence to expression of glucose transporter isoforms. Endocrinology 127,
126 – 132.
Muoio, D.M., and Newgard, C.B. (2008). Mechanisms of disease: molecular
and metabolic mechanisms of insulin resistance and beta-cell failure in
type 2 diabetes. Nat. Rev. Mol. Cell Biol. 9, 193 – 205.
Niki, T., Takahashi-Niki, K., Taira, T., et al. (2003). DJBP: a novel DJ-1-binding
protein, negatively regulates the androgen receptor by recruiting histone
deacetylase complex, and DJ-1 antagonizes this inhibition by abrogation
of this complex. Mol. Cancer Res. 1, 247– 261.
Nikic, I., Merkler, D., Sorbara, C., et al. (2011). A reversible form of axon
damage in experimental autoimmune encephalomyelitis and multiple scler-
osis. Nat. Med. 17, 495 –499.
Nikolova, G., Jabs, N., Konstantinova, I., et al. (2006). The vascular basement
membrane: a niche for insulin gene expression and beta cell proliferation.
Dev. Cell 10, 397 – 405.
Ozcan, U., Yilmaz, E., Ozcan, L., et al. (2006). Chemical chaperones reduce ER
stress and restore glucose homeostasis in a mouse model of type 2 dia-
betes. Science 313, 1137– 1140.
Park, K.S., Wiederkehr, A., Kirkpatrick, C., et al. (2008). Selective actions of
mitochondrial fission/fusion genes on metabolism-secretion coupling in
insulin-releasing cells. J. Biol. Chem. 283, 33347 – 33356.
Poy, M.N., Hausser, J., Trajkovski, M., et al. (2009). miR-375 maintains normal
pancreatic alpha- and beta-cell mass. Proc. Natl Acad. Sci. USA 106,
5813 –5818.
Robertson, R.P. (2004). Chronic oxidative stress as a central mechanism for
glucose toxicity in pancreatic islet beta cells in diabetes. J. Biol. Chem.
279, 42351 – 42354.
Robertson, R.P., and Harmon, J.S. (2006). Diabetes, glucose toxicity, and oxi-
dative stress: a case of double jeopardy for the pancreatic islet beta cell.
Free Radic. Biol. Med. 41, 177 – 184.
Sachdeva, M.M., Claiborn, K.C., Khoo, C., et al. (2009). Pdx1 (MODY4) regu-
lates pancreatic beta cell susceptibility to ER stress. Proc. Natl Acad. Sci.
USA 106, 19090 –19095.
Silva, J.P., Kohler, M., Graff, C., et al. (2000). Impaired insulin secretion and
beta-cell loss in tissue-specific knockout mice with mitochondrial diabetes.
Nat. Genet. 26, 336 – 340.
Soltani, N., Qiu, H., Aleksic, M., et al. (2011). GABA exerts protective and regen-
erative effects on islet beta cells and reverses diabetes. Proc. Natl Acad. Sci.
USA 108, 11692 –11697.
Surmeier, D.J., Guzman, J.N., Sanchez-Padilla, J., et al. (2011). The role of
calcium and mitochondrial oxidant stress in the loss of substantia nigra
pars compacta dopaminergic neurons in Parkinson’s disease.
Neuroscience 198, 221 –231.
Tanaka, Y., Gleason, C.E., Tran, P.O., et al. (1999). Prevention of glucose tox-
icity in HIT-T15 cells and Zucker diabetic fatty rats by antioxidants. Proc.
Natl Acad. Sci. USA 96, 10857 – 10862.
Thorel, F., Nepote, V., Avril, I., et al. (2010). Conversion of adult pancreatic
alpha-cells to beta-cells after extreme beta-cell loss. Nature 464,
1149 – 1154.
230 | Journal of Molecular Cell Biology
Tiedge, M., Lortz, S., Drinkgern, J., et al. (1997). Relation between antioxidant
enzyme gene expression and antioxidative defense status of insulin-
producing cells. Diabetes 46, 1733– 1742.
Twig, G., Elorza, A., Molina, A.J., et al. (2008). Fission and selective fusion
govern mitochondrial segregation and elimination by autophagy. EMBO J.
27, 433 – 446.
Ueki, K., Okada, T., Hu, J., et al. (2006). Total insulin and IGF-I resistance in pan-
creatic beta cells causes overt diabetes. Nat. Genet. 38, 583 – 588.
van Arensbergen, J., Garcia-Hurtado, J., Moran, I., et al. (2010). Derepression of
Polycomb targets during pancreatic organogenesis allows insulin-producing
beta-cells to adopt a neural gene activity program. Genome Res. 20, 722–732.
Waanders, L.F., Chwalek, K., Monetti, M., et al. (2009). Quantitative proteomic ana-
lysis of single pancreatic islets. Proc. Natl Acad. Sci. USA 106, 18902–18907.
Wang, X., and Hai, C.X. (2011). ROS acts as a double-edged sword in the patho-
genesis of type 2 diabetes mellitus: is Nrf2 a potential target for the treat-
ment? Mini Rev. Med. Chem. 11, 1082 – 1092.
Wiederkehr, A., and Wollheim, C.B. (2006). Minireview: implication of mito-
chondria in insulin secretion and action. Endocrinology 147, 2643 –2649.
Jain et al.
Wirdefeldt, K., Adami, H.O., Cole, P., et al. (2011). Epidemiology and etiology of
Parkinson’s disease: a review of the evidence. Eur. J. Epidemiol. 26(Suppl 1),
S1– S58.
Xiao, C., Giacca, A., and Lewis, G.F. (2011). Sodium phenylbutyrate, a drug with
known capacity to reduce endoplasmic reticulum stress, partially alleviates
lipid-induced insulin resistance and beta-cell dysfunction in humans.
Diabetes 60, 918 – 924.
Xu, X.M., Lin, H., Maple, J., et al. (2010). The Arabidopsis DJ-1a protein confers
stress protection through cytosolic SOD activation. J. Cell Sci. 123,
1644 – 1651.
Xu, Q., Park, Y., Huang, X., et al. (2011). Diabetes and risk of Parkinson’s
disease. Diabetes Care 34, 910 –915.
Yang, B.T., Dayeh, T.A., Kirkpatrick, C.L., et al. (2011). Insulin promoter DNA
methylation correlates negatively with insulin gene expression and positive-
ly with HbA(1c) levels in human pancreatic islets. Diabetologia 54, 360 – 367.
Zhou, W., Bercury, K., Cummiskey, J., et al. (2011). Phenylbutyrate up-regulates
Downloaded from http://jmcb.oxfordjournals.org/ at Lund University Libraries, Head Office on August 18, 2015
the DJ-1 protein and protects neurons in cell culture and in animal models of
Parkinson disease. J. Biol. Chem. 286, 14941 – 14951.