Neurobiology of Disease 58 (2013) 169–178

Contents lists available at SciVerse ScienceDirect

Neurobiology of Disease
journal homepage: www.elsevier.com/locate/ynbdi

Diabetic neuropathic pain development in type 2 diabetic mouse model

and the prophylactic and therapeutic effects of coenzyme Q10
Yan Ping Zhang a,1, Chun Yu Song b,1, Yue Yuan a, Ariel Eber a, Yiliam Rodriguez a, Roy C. Levitt a,c,d,
Peter Takacs e, Zhe Yang a, Ronald Goldberg f, Keith A. Candiotti a,⁎
a
Department of Anesthesiology, Perioperative Medicine and Pain Management, University of Miami Miller School of Medicine, Miami, FL, USA
b
Department of Anesthesiology, the Second Affiliated Hospital of Harbin Medical University, Harbin, China
c
Hussman Institute of Human Genomics, University of Miami Miller School of Medicine, USA
d
Miami Veterans Healthcare System, Miami, FL, USA
e
Department of Obstetrics and Gynecology, University of Miami Miller School of Medicine, USA
f
Division of Endocrinology, Diabetes and Metabolism Diabetes Research Institute, University of Miami Miller School of Medicine, Miami, FL, USA

a r t i c l e i n f o a b s t r a c t

Article history: The early onset of type 2 diabetes mellitus (DM), driven by increasing obesity, is associated with peripheral
Received 24 October 2012 neuropathy. Here, we characterize diabetic neuropathic pain in New Zealand obese diabetic mice (NZO/HILtJ)
Revised 5 May 2013 as a polygenic model of obesity with type 2 diabetes and investigate the role of coenzyme Q10 (CoQ10) in the
Accepted 8 May 2013
prevention and treatment of diabetic neuropathic pain. Since the overexpression of mitogen-activated protein
Available online 16 May 2013
kinase (MAPK), nuclear factor-κB proteins (NF-Kb), toll-like receptor 4 (TLR4) and downstream cytokines
Keywords:
(such as CCL2, CXCL10) are considered important factors contributing to the development of neuropathic pain,
Type 2 diabetes the expression of these factors and the inhibitory effects of CoQ10 were evaluated.
Neuropathic pain NZO/HILtJ mice spontaneously developed type 2 DM and increased body mass with diabetic neuropathic pain.
Coenzyme Q10 CoQ10 treatment decreased pain hypersensitivity and long-term supplementation prevented the development
Oxidative stress of diabetic neuropathic pain but did not attenuate diabetes. Spinal cord, blood serum, liver tissue, and dorsal
Anti-oxidant root ganglia (DRG) from diabetic mice demonstrated increased lipid peroxidation, which was decreased by
CoQ10 treatment. The percentage of positive neurons of p65 (the activated marker of NF-KB) and MAPK in DRG
were significantly higher in DM mice compared to controls. However, CoQ10 treatment significantly decreased
p65 and MAPK positive neurons in the DRG of DM mice. RT-PCR demonstrated that elevated levels of mRNA
of CCL2, CXCL10 or TLR4 in the spinal cord of DM mice decreased significantly when DM mice were treated with
CoQ10.
Conclusion: This model may be useful in understanding the mechanisms of neuropathic pain in type 2 DM
induced neuropathic pain and may facilitate preclinical testing of therapies. CoQ10 may decrease oxidative stress
in the central and peripheral nervous system by acting as an anti-oxidant and free-radical scavenger. These
results suggest that CoQ10 might be a reasonable preventative strategy for long-term use and using
CoQ10 treatment may be a safe and effective long-term approach in the treatment of diabetic neuropathy.
Published by Elsevier Inc.

Introduction blood glucose control, which is often difficult to maintain (Harati,

Neuropathic pain is one of the most devastating complications
of diabetes mellitus (DM) and has a significant impact on quality
of life (Edwards et al., 2008). Frequently, neuropathy progresses over
time, which makes protecting the nervous system from diabetic damage
an important goal of healthcare practitioners. Currently, apart from tight
⁎ Corresponding author at: C301, JMH Central, Department of Anesthesiology, Peri-
operative Medicine and Pain Management, University of Miami, Miami, FL 33136, USA.
Fax: +1 305 585 5839.
E-mail address: KCandiot@med.miami.edu (K.A. Candiotti).
Available online on ScienceDirect (www.sciencedirect.com).
1
Both authors contributed equally.
2007), adequate approaches for prevention and treatment of this type
of neuropathic pain are lacking (Vinik, 2005). Recent work has provided
evidence that reactive oxygen species (ROS) play a major role in the
pathogenesis of diabetic neuropathy (Vincent et al., 2008). Additional
evidence suggests that the nervous system is greatly sensitized to ROS
damage due to the concomitant high levels of polyunsaturated lipids
(Friedman, 2011; Pajovic et al., 2003). Coenzyme Q10 (CoQ10) is an
endogenously synthesized compound that acts as an electron carrier
in the mitochondrial respiratory chain. In addition to its unique role in
the mitochondria, CoQ10 functions as an anti-oxidant, scavenging free
radicals and inhibiting lipid peroxidation (Battinoa et al., 1991; Zhou
et al., 1999). Recent studies have provided evidence of the potential
value of CoQ10 in prophylaxis and in therapies for various disorders

0969-9961/$ – see front matter. Published by Elsevier Inc.

http://dx.doi.org/10.1016/j.nbd.2013.05.003
170 Y.P. Zhang et al. / Neurobiology of Disease 58 (2013) 169–178
associated with oxidative stress induced neurodegeneration (Mancuso
et al., 2006; Wadsworth et al., 2008; Zhang et al., 2013).
Due to the great unmet need and medical importance, diabetic
neuropathy is under intensive study. Streptozotocin-induced type 1
diabetic rat and mouse models are most commonly employed to
study the mechanisms of painful diabetic neuropathy and to evaluate
potential therapies (Morrow, 2004). These are currently the only an-
imal models that possess clear sensory abnormalities that mimic
human neuropathy onset, such as tactile and cold allodynia, early
phase thermal hyperalgesia and late phase hypoalgesia (Talbot and
Couture, 2012). Type 2 DM makes up approximately 90% of the
cases of diabetes and thus its study is important. Although some stud-
ies have been conducted on type 2 diabetic animal models (Liao et al.,
2011; Otto et al., 2011; and Talbot and Couture, 2012), polygenic
models of obesity with diabetes, such as New Zealand obese (NZO/
HILtJ) mice, may provide more insight to the human condition
(Haskell et al., 2002; Leiter et al., 1998). In this study, we evaluate
the onset of neuropathic pain associated with type 2 diabetes in NZO/
HILtJ mice. This study reports on the evaluation of the relationship
between the development of diabetes and the onset of neuropathic
pain and examines the possibility that CoQ10 is protective, and possibly
therapeutic, for neuropathic pain in type 2 DM.
Many researchers have demonstrated that oxidative stress medi-
ates activation of mitogen-activated protein kinase (MAPK), nuclear
factor-κB proteins (NF-κB) and TLR4 signaling pathways (Chang and
Karin, 2001; Runchel et al., 2011). Overexpression of activated
MAPK and NF-κB is considered an important factor contributing to
the development of neuropathic pain (Ji and Suter, 2007), and accu-
mulating evidence also suggests that microglial TLR4 in the spinal
cord is a key receptor in initiating and maintaining microglial activation
and chronic pain (Ji and Suter, 2007; Tanga et al., 2005). For all these
reasons we evaluated the expression of activated MAPK, NF-κB, TLR4
and related downstream cytokines such as monocyte chemoattractant
protein 1 (CCL2) and C-X-C motif chemokine 10 (CXCL10) in diabetic
induced neuropathic pain and evaluated the inhibitory effects of
CoQ10 on these risk factors.
Materials and methods
Animal preparation
The study was approved by the Animal Care and Use Committee of
the University of Miami. Fifty-six mice from the NZO/HILtJ strain were
used in the experiments. Two female and two male ten-week-old
NZO/HILtJ mice were purchased from Jackson Laboratory (Bar Harbor,
Maine) and used as breeders of new offspring (16 males and 22 fe-
males) which were monitored for the development of type 2 diabetes
and the onset of associated diabetic neuropathic pain for up to six
months. Animal mechanical sensitivity and body weights were mea-
sured weekly, along with blood glucose (blood was drawn from the
tail veins) from four weeks of age until the end of the experiment.
Mice were considered to have diabetes if their blood glucose exceeded
250 mg/dl on one or more occasions. To investigate the analgesic effects
of CoQ10 on diabetic neuropathic pain, eighteen 10-week-old NZO/HILtJ
male mice were purchased from Jackson Lab. During the experimental
period (8 weeks), blood glucose levels and mechanical sensitivity were
monitored weekly. All of the mice were maintained or bred in viral-
and antigen-free facilities. All studies were conducted during the light
cycle, typically between 8:00 AM and 4:00 PM.
Determination of diabetic neuropathic pain
The presence of neuropathic pain was confirmed by testing for
mechanical allodynia and thermal hyperalgesia. The mechanical and
thermal tests were performed as described previously (Fu et al.,
2010).
Mechanical allodynia test
Briefly, the mechanical allodynia test was conducted with a Touch-
Test Sensory Evaluator (von Frey filaments, North Coast Medical, Inc.,
Wood Dale, IL). Filaments were used to assess the degree of allodynia
on each day of testing, conducted once a week, beginning at four
weeks of age. For each assessment, the mouse was placed on a wire
mesh platform and was covered with a transparent glass container; a
period of 30 min was allowed for habituation. Five measurements
were taken for each animal, randomly starting at the left or right paw.
The observation of a positive response (paw lifting, “shaking” or licking)
within 5 s of the application of the filament was then followed by the
application of the thinner filament (or the thicker one if the response
was negative). The paw withdrawal threshold was measured five
times and was expressed as the tolerance level in grams.
Thermal hyperalgesia test
A Ugo Basile Plantar Test apparatus (Biological Research Appara-
tus, Comerio, Italy) was used to measure paw withdrawal latency
time in unrestrained mice. To determine paw withdrawal latency to
radiant heat, the mouse was placed in an acrylic restrainer on a ther-
mal stimulator. The hindpaws made contact with a 1/4 in. thick glass
plate that was maintained at room temperature. A light source that
produced radiant heat was focused below the glass onto the plantar
surface of one hindpaw. The paw withdrawal latency was determined
for the left and right hindpaws, with a 5-minute inter-trial interval. A
cutoff of 20 s was used. Four trials were performed in each animal to
establish the mean paw withdrawal latencies.
Chemically induced inflammatory pain (formalin test)
This experiment was performed according to our previously de-
scribed method with slight modifications (Fu et al., 2007). Mice were
adapted to a Plexiglass chamber (14 × 14 × 14 cm3) for 10 min. Thirty
microliters of 0.5% formalin solution was injected into the plantar sur-
face of the left hindpaw with a 27-gauge needle on the day of testing;
the time spent on pain-like behaviors (licking/biting) was recorded
using a timer in two phases: Phase 1: 0–10 min immediately after for-
malin injection, and Phase II: 11–40 min immediately after formalin
injection.
Animal treatment protocols
For our experiments investigating the prophylactic effect of CoQ10
on the development of type 2 DM and diabetic neuropathic pain, 14
mice bred in-house were supplied with 50 mg/kg/day of CoQ10
(Sigma St. Louis, MO) in their regular standard diet for 2 months, be-
ginning at the age of 16 weeks (before the onset of DM). The content
of CoQ10 in the diet was controlled by dissolving CoQ10 in olive oil
(Sigma) which was then added to the standard diet according to the
body weight of the mouse to a total of 50 mg/kg/day; the volume of
olive oil was determined as 2 ml/kg. For example, a 50 g body weight
mouse would be supplied a diet with 2.5 mg of CoQ10 dissolved
in 100 μl of olive oil, which was placed in the cage of the mouse.
The control group (n = 10) received a standard mouse diet (3.1 kcal/g,
6.2 g% fat, 18.6 g% protein, 44.2 g% carbohydrate; Teklad Diets, Madison,
Wisconsin), which also had olive oil added according to mouse body
weight (2 ml/kg). Mice diets were adjusted twice a week according to
changing body weights.
For our experiments evaluating the analgesic effect of CoQ10,
eighteen 10-week-old NZO/HILtJ male mice purchased from Jackson
were used. At the age of 18 weeks old, 10 of the mice developed
DM and neuropathic pain, while 8 mice did not appear to develop
DM. CoQ10 or olive oil (control vehicle) was then dosed for three con-
secutive days and the animals were then evaluated by the mechanical
sensory test, thermal sensory test or formalin induced inflammation
pain test. CoQ10 was dissolved in olive oil to produce concentrations
 Y.P. Zhang et al. / Neurobiology of Disease 58 (2013) 169–178 171

of: (1) 30 mg/ml for the 100 mg/kg dose, and (2) 60 mg/ml for the
200 mg/kg dose. These doses represent the equivalent human doses
of 8 mg/kg and 16 mg/kg, respectively, based on body surface area
(Reagan-Shaw et al., 2008). The CoQ10 solution was administered
to the mice via intraperitoneal injection once a day, at a volume of
100 μl/30 g of body weight. Control diabetic mice were treated with
the same volume of olive oil.
Lipid peroxidation measurement
The thiobarbituric reactive substances (TBARS) assay is a well-
established method for monitoring tissue lipid peroxidation due to free
radical generation associated with oxidative stress. In this study, the
TBARS assay was used to quantify oxidative stress associated with the
onset of diabetes and the anti-oxidative (lipid peroxidation) effects of
CoQ10 in liver, serum and spinal cord. The detailed experimental proce-
dure followed the instructions in the TBARS assay kit (Cayman Chemical,
Ann Arbor, Michigan). TBARS values were measured at an absorbance of
532 nm. Five microliters of homogenate from each sample was assayed
for protein, using the Bradford dye-binding method. The concentration
of lipid peroxides was normalized and expressed as a percentage of pro-
tein concentration.
Immunohistochemistry and image quantification
Normal mice, mice with diabetic pain and diabetic mice treated
with CoQ10 were sacrificed via an over-dose of nembutal and were
then decapitated. L4–L5 DRG were removed, fixed in 4% paraformalde-
hyde in phosphate buffered saline (pH 7.4) overnight, cryoprotected in
0.1 M phosphate buffered saline containing 20% sucrose and sectioned
by cryostat into 15 μm thick sections. Sections were incubated over-
night at 4 °C with the primary antibodies: anti-4-Hydroxy-2-nonenal
(HNE, Abcam, Catalog #: ab46545; Cambridge, MA), anti-CoQ10
(Abcam, Catalog #: ab41997), anti-active MAPK (Promega, Catalog #:
V803A; WI) or anti-p65 (a marker of active NF-κB; Millipore, Catalog #:
MAB3026; Billerica, MA), followed by secondary species-specific fluores-
cent antibodies (Jackson ImmunoResearch Lab, Inc. West Grove, PA)
or biotinylated secondary antibody (VECTOR Lab, Burlingame, CA) for
1 h at 22 °C. To ensure the specificity of each primary antibody, the
primary antibody was replaced by the diluent of the antibody in one
section in each set of stains so as to exclude non-specific background
staining. For image quantification for the DRG sections, two sections
were randomly selected from each sample and images were taken with
a high-resolution digital camera and analyzed with NIH Image J software
(U.S. National Institutes of Health, Bethesda, Maryland). Tissue sections
from 4 to 6 animals per experimental condition were analyzed. The pro-
portion of positive cells of MAPK or p65 was determined by positive
neuron profiles divided by the total neuronal profiles on each section
of DRG. The researcher conducting the image analysis was blinded to
which group the tissue was from to avoid any possible observation bias.
RNA isolation and RT-PCR of TLR4, CCL2 and CXCL10 message RNA
(mRNA) of the spinal cord
CCL2: 5′-ccccactcacctgctgctac-3′ and 5′-cctgctgctggtgattctctt-3′; CXCL10:
5′-gccgtcattttctgcctcatcct-3′ and 5′-ctcattctcactggcccgtcatc-3′; TLR4: 5′-
gctttcacctctgccttcac-3′ and 5′-aggcgatacaattccacctg-3′.
Statistical analysis
Data were presented as mean ± SEM and analyzed using Prism 4
software (GraphPad Software Inc., San Diego CA). The time-course of
parameters was statistically analyzed by repeated measure analysis of
variance and post-hoc test of Tukey's Multiple Comparison Test. Data
for the mechanical allodynia test, at different time points among dif-
ferent treatments, were analyzed using two-way ANOVA followed by
post-hoc Tukey's Multiple Comparison Test. Unpaired comparisons
between different groups were assessed by two-tailed Student t-test.
The Pearson correlation coefficient analysis was performed for the cor-
relation of blood glucose levels with mechanical sensitivities. P values
of less than b0.05 were designated as statistically significant.
Results
The development of type 2 DM and the onset of neuropathic pain
NZO/HILtJ mice developed spontaneous obesity at an early age and
their body weights increased throughout the experiment. Tables 1A
and 1B summarize the 38 inbred mice in terms of body weight, blood
glucose level, and mechanical sensitivity throughout the 24 week
experimentation period. DM in this inbred strain is well known to be
heritable but displays heterogeneity, presumably due to epigenetic
causes (Haskell et al., 2002; Kanasaki and Koya, 2011; Leiter et al.,
1998). All of the animals' blood glucose levels were monitored weekly.
Animals were sub-grouped as those developing early onset DM and those
with delayed-onset disease. Table 1A shows a subgroup of mice with
early onset high blood glucose levels (as early as four weeks of age)
(n = 14, only males, 37% of all mice bred in-house), and Table 1B
shows the subgroup of mice with gradually increasing blood glucose
levels after the age of 16 weeks (n = 24, 63% of all mice bred in-
house). If blood glucose exceeded 250 mg/dl the mice were considered
to have DM. All 38 of the inbred mice (16 males and 22 females) even-
tually developed DM. In other words, our mice bred in-house demon-
strated 100% penetrance of type 2 DM in both males and females.
Figs. 1A and B show the significant correlation between the in-
crease in diabetic neuropathic pain and increasing blood glucose
levels. The onset of mechanical hypersensitivity, measured with
von Frey filaments, was closely associated with the onset of high blood
glucose levels. Mice appeared to respond in one of two patterns: high
blood glucose appearing at 4 weeks (Pearson's r = −0.67, P = 0.017)
(Fig. 1A), or after 16 weeks of age (Pearson's r = −0.93, P = 0.007)
(Fig. 1B). These data confirm that hyperglycemic mice demonstrate
significantly lower thresholds for mechanical neuropathic pain.

The mRNA levels of CCL2, CXCL10 or toll like receptor 4 (TLR4) in

the spinal cord tissue were evaluated by RT-PCR. Extraction of total Table 1A
RNA was carried out with TRIzol (Invitrogen, Carlsbad, CA) according Subgroup of NZO/HILtJ mice that developed high blood glucose levels at four weeks of
to the manufacturer's instructions. Total RNA (0.5 μg) was reverse- age (early onset of DM).
transcribed with 200 U/sample SuperScript II (Invitrogen) and 250 ng/ Age of mice Body weight Blood glucose Mechanical thresholds
reaction of random primers (Promega). Real-time PCR was performed (weeks) (grams) (mg/dl) (grams)
using the TAQurate green real-time PCR master mix (Epicentre Biotech- 4 32.17 ± 1.25 272 ± 14.18 0.49 ± 0.12
nologies, Madison) with specific primers following the manufacturer's 8 44.10 ± 1.56 251 ± 50.99 0.46 ± 0.12
instructions and data were collected on the Rotor-Gene 2000 12 52.00 ± 1.48 265 ± 26.58 0.41 ± 0.14
(Corbett Research, Mortlake, Australia). The measurements revealed 16 60.92 ± 1.36 355 ± 58.57 0.43 ± 0.07
20 67.33 ± 2.58 386 ± 53.33 0.36 ± 0.10
the mRNA activity for each of these factors in control and experimen-
24 70.83 ± 3.82 432 ± 19.14 0.31 ± 0.08
tal mice tissues. The oligonucleotide sequences of the primers are:
172 Y.P. Zhang et al. / Neurobiology of Disease 58 (2013) 169–178

Table 1B levels and mechanical allodynia. Oral CoQ10 treatment did not
Subgroup of NZO/HILtJ mice that developed high blood glucose levels after 16 weeks of attenuate the blood glucose level but did successfully inhibit the
age (late onset of DM).
development of mechanical allodynia. Two-way ANOVA analysis
Age of mice Body weight Blood glucose Mechanical thresholds demonstrated a significant interaction between group (F = 70.150,
(weeks) (grams) (mg/dl) (grams) P b 0.001) and time (F = 2.868, P b 0.01) and significant main effect
4 36.67 ± 2.28 157 ± 4.89 1.11 ± 0.10 for group and time (F = 4.351, P b 0.001).
8 42.13 ± 0.94 130 ± 13.40 0.91 ± 0.07
12 52.22 ± 1.66 139 ± 16.44 0.88 ± 0.12
16 59.32 ± 1.49 190 ± 18.04 0.86 ± 0.11
20 70.33 ± 2.59 384 ± 34.72** 0.45 ± 0.09* The analgesic effects of CoQ10 treatment on mechanical allodynia and
24 76.17 ± 1.76 376 ± 50.39*** 0.43 ± 0.06** thermal hyperalgesia
In Tables 1 A and B data are expressed as means ± SEM. Blood glucose levels higher
than 250 mg/dl were considered demonstrative of DM. Data in Tables 1A and B were Ten-week-old NZO/HILtJ male mice were purchased from Jackson Lab.
analyzed by repeated measures analysis of variance and post-hoc test of Tukey's At 16 weeks of age, 10 mice developed DM (blood glucose >250 mg/dl),
Multiple Comparison. *, ** or *** represents P b 0.05, 0.01 or 0.001 comparing data at however 8 mice had normal blood glucose levels similar to non-DM
a given time point with the same type of data at 4 weeks of age.
controls. The average mechanical withdrawal threshold for normal
controls was 0.88 ± 0.05 g (n = 8). In contrast, the average withdraw-
al threshold of DM mice was significantly reduced to 0.22 ± 0.04 g,
The prophylactic effects of CoQ10 on diabetic neuropathic pain (n = 10), (P b 0.001, DM vs. non-DM). To examine the analgesic effects
development of CoQ10, DM mice with diabetic neuropathic pain were injected
intraperitoneally once daily at the same time for three consecutive
Mice bred in-house that developed DM after 16 weeks old were days with either 100 mg/kg or 200 mg/kg of CoQ10, or with olive
divided into two groups (G). G1: n = 14, received a diet containing oil alone which was the vehicle for CoQ10. Both doses of CoQ10
CoQ10 at a calculated dose of 50 mg/kg/day for 8 weeks (from the significantly increased the mechanical thresholds (compared to vehicle
age of 16 weeks to the age of 23 weeks). G2: n = 10, received a controls, P b 0.001). These effects were short-lived, as mechanical thres-
regular daily diet containing only the vehicle of CoQ10 through the holds returned to pretreatment levels within 72 h of the last CoQ10
entire experimental period. Figs. 2A and B show the effect of daily dose. This data supports the idea that repeated doses of CoQ10
oral CoQ10 treatment on the development of high blood glucose provide an anti-nociceptive effect in mice with diabetic neuropathic
pain (Fig. 3A). Two-way ANOVA analysis showed a significant inter-
action between group (F = 12.62, P b 0.0001) and time (F = 8.34,
P b 0.0001), and significant main effect for group and time (F = 2.105,
P = 0.041).

Fig. 2. The prophylactic effects of CoQ10 on diabetic neuropathic pain development.

Fig. 1. The correlation of blood glucose level with mechanical sensitivity. Fig. 1A repre-
sents data from mice with high blood glucose levels at four weeks of age (early onset of
DM). Pearson's r = −0.67, P b 0.05. Fig. 1B represents data from the subgroup of mice
after 16 weeks of age with gradual elevation in blood glucose levels (late onset of DM).
Pearson's r = −0.93, P b 0.01. Both groups show significant correlation of high blood
glucose level with mechanical hypersensitivities.
This figure shows oral daily CoQ10 treatment at a dose of 50 mg/kg/day for 8 weeks
(from the age of 16 weeks to 23 weeks) on the development of DM and mechanical
allodynia. Fig. 2A demonstrates that CoQ10 did not change the trend of increasing
blood glucose. Fig. 2B demonstrates that CoQ10 treatment effectively inhibited the de-
velopment of mechanical allodynia (* or ** presents P b 0.05 or 0.01 in comparing the
data at the same time point in non-CoQ10 and CoQ10 treatment groups).
 Y.P. Zhang et al. / Neurobiology of Disease 58 (2013) 169–178 173

Fig. 3B shows the effects of CoQ10 treatment on thermal hyperalgesia.
The mean latency to thermal stimulation of non-diabetic age-matched
(18 weeks of age) NZO/HILtJ male mice (n = 8) was 8.54 ± 0.66 and
7.49 ± 0.95 s on the left and right paws, respectively. Latencies were sig-
nificantly decreased, indicating neuropathic pain, in DM mice (n = 8),
measuring 3.21 ± 0.28 s on the left paw and 3.86 ± 0.92 s on the
right paw (normal vs. DM: P = 0.008 in the left paws, or P = 0.022
in the right paws by Student t-test). Five hours after treatment with
200 mg/kg of CoQ10, the DM mice showed significantly increased
latencies, with both paws returning to normal levels as compared to
vehicle-treated DM mice (DM vs. DM treated with COQ10: P = 0.008
in the left paws and P = 0.043 in the right paws by Student t-test).
CoQ10 treatment prevents formalin-evoked hyperalgesia in diabetic
mice
The average licking behavior in DM mice (n = 8) was significantly
longer than in normal mice (n = 8) in Phase I (0–10 min) (normal
vs. DM in vehicle treated: P = 0.033, Student t-test) and in Phase II
(11–40 min) (normal vs. DM in vehicle treated, P = 0.007, Student
t-test). The results demonstrate that diabetic mice have a marked
increase in the frequency of paw licking/biting after paw formalin
injection, illustrating a response of nociceptive hyperalgesia to this
inflammatory chemical stimulus. Furthermore, the administration
of 200 mg/kg CoQ10 4–5 h prior to formalin injection significantly
reduced formalin-induced licking/biting behaviors during Phase I and
Phase II in DM mice (n = 8) (DM treated with vehicle vs. DM treated
with CoQ10 in Phase I, P = 0.023; in Phase II, P = 0.004 by Student
t-test). CoQ10 did not have any effects on normal mice (Fig. 4).
Hyperglycemia-induced lipid peroxidation and the anti-oxidant effect
of CoQ10
With the TBARS assay we measured lipid peroxidation in spinal cord
tissue, blood serum and liver tissue as an indicator of ROS production
and investigated the anti-oxidant activity of CoQ10 in diabetic mice.
Hyperglycemia induced lipid peroxidation and 200 mg/kg of CoQ10
treatment significantly decreased the levels of lipid peroxidation in
the spinal cord tissue, blood serum and liver tissue (Fig. 5A).
The production of HNE, a most reactive lipid peroxidation product,
was measured in DRG neurons. Fig. 5B shows the staining of HNE in
DRG sections and Fig. 5C demonstrates the quantification of HNE pro-
duction and the anti-oxidative effects of CoQ10 (200 mg/kg) through
an analysis of staining intensity of HNE. The staining intensity of HNE
in DM mice was significantly greater than in normal mice (P b 0.001)
or DM mice treated with CoQ10 (P = 0.002). The results confirmed
that diabetes resulted in an increase of lipid peroxidation product in
the DRG and that treatment with CoQ10 decreased it. These data sug-
gest that systemic CoQ10 treatment is capable of reducing or preventing
hyperglycemia-induced increases in lipid peroxidation in peripheral
nerve tissues.
Figs. 5B and C also show the staining of CoQ10 in DRG sections and
the quantification of CoQ10 levels. CoQ10 levels in the DRG of DM
mice were significantly lower than in the DRG of normal controls
(P b 0.001) or DM mice treated with CoQ10 (P b 0.001). These results
suggest that DM is associated with lower levels of neuronal CoQ10
and this may represent a mechanism whereby the production of super-
oxide in DM mice leads to pathology. Furthermore, we demonstrated
that exogenous systemic administration of CoQ10 overcomes the ap-
parent shortage of CoQ10 in DRG cells.

Fig. 3. Analgesic effect of systemic CoQ10 on the mechanical allodynia and thermal hyperalgesia. In Fig. 3A, “normal mice” represents the mean mechanical threshold from healthy
untreated NZO/HILtJ mice. Mechanical thresholds are also presented for mice that were pretreated, or those that were treated on days 1, 2, or 3 with vehicle, and with 100 or 200 mg/day of
CoQ10 in DM mice with neuropathic pain. These data show that hindpaw withdrawal thresholds are reduced with the development of diabetes, indicating neuropathic pain (### P b 0.001,
pretreated or vehicle-treated vs. normal mice). These data also show improvement with CoQ10 treatment, compared to vehicle-treated DM mice (*P b 0.05, **P b 0.01). Fig. 3B shows that DM
mice develop thermal hyperalgesia. CoQ10 (200 mg/kg i.p. once after treatment 5 h) treatment completely reversed thermal hypersensitivity (#P b 0.05, ##P b 0.01 compares data in DM
mice vs. normal mice, *P b 0.05, **P b 0.01 compares data in DM mice treated with CoQ10 vs. DM mice treated with vehicle).
174 Y.P. Zhang et al. / Neurobiology of Disease 58 (2013) 169–178

Fig. 4. The effect of CoQ10 on preventing formalin-evoked hyperalgesia in DM mice.
The Y-axis shows the duration of pain-related behaviors (hindpaw licking/biting) in
seconds. The time spent on pain-related behaviors in DM mice was significantly great-
er than in normal mice in both Phase I (* P b 0.05, t-test) and Phase II (**P b 0.01,
t-test) in vehicle controls. CoQ10 (200 mg/kg, i.p. 4–5 h prior to formalin injection)
significantly reduced pain-related behaviors during both Phase I and Phase II in DM
mice (DM treated with vehicle vs. DM treated with CoQ10 in Phase I, # P b 0.05; in
Phase II, ##P b 0.01).
Increased signaling pathways of MAPK, NF-KB and TLR4 activation in
DRG and spinal cord of DM mice, the inhibit effect of COQ10 treatment
To obtain further insights into the mechanisms of DM-induced
neuropathic pain and the therapeutic effects of CoQ10, we examined
the expression of MAPK, NF-κB, TLR4 and related factors—CCL2, CXCL10
in the DRG and spinal cord. As shown in Fig. 6, few neurons stain pos-
itive for MAPK and p65 (the activated marker of NF-κB) in normal
mice DRG. Additionally, p65 positive cells are mostly glial cells that sur-
round neurons in normal mice DRG. However, in DM mice, there was a
significant increase in the positive neurons for MAPK and p65 (Figs. 6A
and B), and quantification analysis by Student t-test demonstrated that
the percentage of positive neurons of MAPK and p65 significantly in-
creased in DM mice compared to normal controls (normal vs. DM:
MAPK, P = 0.015; p65, P b 0.001). In contrast, in DM mice treated with
CoQ10, the percentage of MAPK and p65 positive neurons decreased
significantly (DM vs. DM treated with CoQ10: MAPK, P = 0.029: and
p65, P = 0.021). RT-PCR results demonstrated significantly elevated
expression levels of mRNA of CCL2, CXCL10 and TLR4 in the spinal cord
of DM mice (normal vs. DM: CCL2, P = 0.002; TLR4, P = 0.002, and
CXCL10, P b 0.001). In tissue from DM mice treated with COQ10 the
expression levels of these three factors decreased significantly compared
to tissue from untreated DM mice (DM vs. DM treated with COQ10: CCL2,
P = 0.004; TLR4, P = 0.005; CXCL10, P = 0.025). These findings suggest
that the effect of CoQ10 in improving diabetic neuropathic pain may be a
result of the down regulation of CCL2, CXCL10 and TRP4 (and possibly
other related factors).
Discussion
Neuropathic pain is a devastating chronic condition and a common
complication of diabetes (Edwards et al., 2008; Harati, 2007; Vinik,

Fig. 5. The anti-oxidant effect of CoQ10 treatment in diabetic mice. Fig. 5A shows anti-lipid peroxidation in TBARS assay. The Y-axis shows the lipid peroxidation level. These data
demonstrate an increase in lipid peroxidation with DM as compared to normal control (#, ##, or ###P b 0.05, 0.01 or 0.001 compared data on normal mice with DM mice). CoQ10
treatment (200 mg/kg, i.p. 4–5 h prior to sample collection) also decreased lipid peroxidation in spinal cord, blood serum and liver tissue of DM mice (* or ***P b 0.05 or 0.001
compared data on DM + CoQ10 with DM within the same type of sample). Fig. 5B represents immunohistochemistry staining of HNE and CoQ10 in DRG tissues. HNE expression
is weak in the normal mice DRG, but expression is strong in the DRG of DM mice, and weak in the DRG of DM mice treated with CoQ10 (200 mg/kg). However, CoQ10 staining is
significantly weaker in the DRG of DM mice than in the DRGs of normal or DM mice treated with COQ10. Bar = 100 μm. Fig. 5C represents the relative quantification of HNE product
and CoQ10 (normal vs. DM, ***P b 0.001; DM vs. DM treated with COQ10, ## or ###P b 0.01 or 0.001).
 Y.P. Zhang et al. / Neurobiology of Disease 58 (2013) 169–178 175

Fig. 6. Expression of neuropathic pain related factors and the inhibit effects of CoQ10. Fig. 6A shows photomicrographs of the immunohistological staining of mitogen-activated
protein kinase (MAPK) and p65, the activated marker of nuclear factor-κB proteins (NF-κB) in DRGs of normal control, DM and DM treated with CoQ10 mice. There was an increase
in the number of positive MAPK and p65 neurons in the DRG of DM mice, the inserts show the magnified positive cells, black arrows point the positive neuron in MAPK or p65, red
arrow points the positive glial cell of p65. In the DRG of DM treated with CoQ10, the number of positive neurons of MAPK and p65 decreased compared to DM. Scale bar = 100 μm.
Fig. 6B shows the quantification of positive cells of MAPK and p65. (Normal vs. DM, * or ***P b 0.05 or 0.001; DM vs. DM treated with COQ10, #P b 0.05). Fig. 6C shows the mRNA
expression of CCL2, CXCL10 and TLR4 in the spinal cord by RT-PCR. (Normal vs. DM, *, **or ***P b 0.05, 0.01 or 0.001; DM vs. DM treated with COQ10, # or ##P b 0.05 or 0.01).

2005). Our present study demonstrated that NZO/HILtJ inbred mice
developed severe neuropathic pain as a result of spontaneously devel-
oped obesity and type 2 diabetes. This represents a readily available an-
imal model for future research on type 2 diabetic peripheral neuropathy
and neuropathic pain.
Models of obesity with type 2 diabetes are classified into two
categories: 1) those containing a mutation in the leptin or leptin re-
ceptor gene, such as ob/ob mouse model (obese gene mutation) or
the db/db mouse model (leptin receptor gene mutation) (Chen et al.,
1996; Kanasaki and Koya, 2011) and 2) polygenic models (Hirayama
et al., 1999; Jürgens et al., 2006). Leptin abnormalities, however, only
comprise a minority of obesity/diabetic cases in humans (Clément et
al., 1998; Matsuoka et al., 1997) which may not be the same condition
as the type 2 diabetes, that has become a worldwide epidemic. It
appears that polygenic models of obesity resulting in diabetes may
provide better insight into the human condition. Therefore, models
like NZO/HILtJ inbred mice, that spontaneously develop both obesity
and type 2 DM, could be useful tools to better understand the biology/
pathology underlying human obesity and related type 2 DM (Haskell
et al., 2002; Leiter et al., 1998). This study was designed to evaluate
critical physiologic and metabolic endpoints to assess the development
of diabetes and neuropathic pain in NZO/HILtJ inbred mice and to
describe a new model for the study of type 2 diabetic peripheral neuro-
pathic pain. Our results show that NZO/HILtJ mice developed obesity
and type 2 diabetes spontaneously, which is consistent with previous
reports (Haskell et al., 2002; Leiter et al., 1998). In addition, after devel-
oping type 2 diabetes, these animals demonstrated chronic mechanical
and thermal hyperalgesia, as well as chemically induced hyperalgesia.
Interestingly, the literature indicates that about 50% of male NZO/
HILtJ mice develop type 2 DM, however in our study, mice bred
in-house, demonstrated 100% occurrence of type 2 DM, in both males
and females. In contrast, 56% of male mice we purchased from Jackson
Labs (n = 18) developed DM. One possible explanation for this finding
is that differing environmental or epigenetic factors may influence the
occurrence and onset of type 2 DM in this strain. These factors remain
to be established, but may include differences in bedding and diet. For
example, the diet we used for breeding and maintaining our mice (Teklad
rodent diet from Harlan Lab) contains 6.2 g% fat, however, the diet which
Jackson Lab uses for breeding contains only 4% fat, therefore, increased fat
content may potentially alter the onset of diabetes. Future work may
examine environmental, epigenetic, and genetic factors that may contrib-
ute to the high incidence of type 2 DM in NZO/HILtJ mice bred in our
facility.
The formalin test was introduced as a model of inflammatory pain
in 1977, and has since been used extensively in rats and mice (Tjolsen
et al., 1992; Wheeler-Aceto and Cowan, 1991). In the majority of
these studies researchers assess two distinct phases of formalin pain
behavior. The first phase starts immediately after injection, and lasts
up to 10 min. This phase appears to be mediated by C-fiber activation
in response to direct chemical stimulation of nociceptors. The second
phase, starting at 10 min and lasting for the next 30–45 min, seems to
represent inflammation-induced changes in peripheral nerves and
central sensitization of noxious stimuli through the alteration of
dorsal horn signaling (Tjolsen et al., 1992). In our present study, we
176 Y.P. Zhang et al. / Neurobiology of Disease 58 (2013) 169–178
observed significantly more pain-related activities in Phase I and
Phase II in type 2 diabetic mice than in non-diabetic controls. These
findings are similar to those reported by Calcutt et al. in diabetic
rats (Calcutt et al., 2000), and are consistent with increased local
oxidative stress involving the peripheral afferent nerves and the dorsal
horn of the spinal cord and are associated with increased pain-related
behaviors in diabetic mice. We also demonstrated that systemic CoQ10
pretreatment can effectively eliminate pain-related behaviors in Phases
I and II, perhaps by decreasing local oxidative stress at these anatomic
sites.
Painful diabetic neuropathy is a common cause of considerable
morbidity. Many agents have been tested for their ability to rapidly
alleviate allodynia or hyperalgesia in diabetics (Heubeck, 2006; Smith
and Nicholson, 2007; Veves et al., 2008). Local analgesics have dominat-
ed treatment and have been a focus of research in neuropathic pain in
diabetes (Smith and Nicholson, 2007). The evolving role of oxidative
stress in neuropathic pain, however, suggests that local analgesics
may only address the symptoms of progressive nerve injury in type 2
diabetes. The natural history of diabetic neuropathic pain suggests a
progressive functional change. Ultimately, these metabolic alterations
over time can lead to structural changes, including mitochondrial
damage and neurodegeneration, which can be very difficult to treat or
reverse (Joyce et al., 2010; Stanzione and Tropepi, 2011). Our long-
term study demonstrates that diabetic neuropathic pain can occur in
the early stages of type 2 diabetes, suggesting that both chronic and
acute hyperglycemia-induced oxidative stress in the central and pe-
ripheral nervous system can promote the development of diabetic
neuropathy. Moreover, the occurrence of neuropathic pain coincided
with the onset of diabetes in our study, and this may imply that timely
treatments at the earliest stages of the disease may restore function and
prevent neuronal damage and neuropathic pain. Our data further
suggests that interventions may be beneficial even in the stages of
prediabetes (mild glucose intolerance).
CoQ10 has been tested in clinical trials and pre-clinical studies and
has been shown to be beneficial in several medical conditions that have
relevance for type 1 diabetes (Dzugkoev et al., 2012; Zhang et al., 2013),
and type 2 diabetes (Mezawa et al., 2012; Persson et al., 2012; Sourris
et al., 2012). In an animal model of DM induced cardiomyopathy,
CoQ10 was found to be effective in reducing superoxide generation,
ameliorating diastolic dysfunction and reducing cardiomyocyte
hypertrophy and fibrosis in db/db mice (Huynh et al., 2012). In
alcohol-induced chronic neuropathy, CoQ10 treatment prevented the
biochemical and molecular neurotoxic effects of alcohol via inhibition of
oxido-nitrosative stress and the release of proinflammatory cytokines
(Kandhare et al., 2012). Beal et al. (1994) showed in vivo, neuroprotective
effects of CoQ10 via attenuation of malonate-induced ATP depletions.
Other research has demonstrated that combination therapy with
CoQ10 and creatine produces neuroprotective effects in Parkinson's
and Huntington's diseases (Yang et al., 2009). Consistent with these
diverse findings our current study demonstrated the therapeutic
effect of long-term administration of CoQ10 on type 2 diabetes in-
duced neuropathic pain.
The neuroprotective effect of CoQ10 was demonstrated by decreas-
ing HNE level in the DRG of DM mice. HNE is an α, β-unsaturated alde-
hyde and is considered to be the most reactive lipid peroxidation
product and the most damaging to the tissue (Esterbauer et al., 1991).
Despite the increasing use of HNE detection as an index of lipid perox-
idation and the knowledge that increased HNE formation has been
detected in the brain of patients with Alzheimer's disease (Montine et
al., 1997) and in the spinal cord following traumatic injury (Springer
et al., 1997), little information is available about HNE production in
the nervous system associated with hyperglycemia and diabetes. Our
study reports that HNE is significantly increased in the DRG of DM
mice. Since HNE exerts broad pathologic effects, including inhibition
of DNA and protein synthesis, modification of low-density lipoprotein
and modulation of gene expression (Esterbauer et al., 1991), increased
production of HNE in the DRG of DM mice maybe implicated as the
mechanism of neural dysfunction which could represent the root
cause of hypersensitivity in diabetic neuropathic pain. CoQ10
administration effectively decreased HNE production in the DRG,
along with mechanical, thermal, and chemical inflammatory hyper-
sensitivities associated with DM. Moreover, as shown previously in
diabetes, mitochondrial oxidative phosphorylation is significantly
reduced; thus ATP production is reduced along with decreased
level of CoQ10 (Kucharska et al., 2000). Our study supports these
findings by showing that CoQ10 content is lower in the DRG of DM
mice with neuropathic pain. This reduction in endogenous CoQ10
may result from an increased production of superoxide, and also
may be associated with disturbances of lipid metabolism in the
DRG of DM mice. These results further suggest that lower levels of
neuronal CoQ10 may represent a mechanism whereby the production
of superoxide in DM mice leads to pathology. Fortunately, exogenous
systemic administration of CoQ10 overcomes the apparent shortage of
CoQ10 in DRG cells and inhibits increased DM associated lipid perox-
idation. Additional research concerning mitochondrial dysfunction as
it relates to CoQ10 deficiency in the DM nervous system is required.
Oxidative stress can activate some stress-sensitive signaling
pathways such as MAPK, NF-κB and TLR4 and hence regulate the
expression of downstream factors (Allen and Tresini., 2000; Ko et
al., 2011). Increases in levels of MAPK and NF-κB have been reported
in nerve tissue from models of type 1 and type 2 diabetes (Du et al.,
2010; Kumar et al., 2012). Our previous studies in peripheral nerve
injury-induced neuropathic pain have demonstrated that NF-κB and
its related cytokines CCL2, CXCL10 are involved in the pathogenesis of
neuropathic pain (Fu et al., 2007, 2010). Additionally, accumulating
evidence suggests that microglial TLR4 in the spinal cord is the key
receptor in initiating microglial activation and maintaining chronic pain
after peripheral nerve injury (Bettoni et al., 2008; Tanga et al., 2005).
The present study supports the concept that the signaling
pathways of MAPK, NF-κB and TLR4 are activated in DM. Further-
more, CoQ10 treatment not only decreases lipid peroxidation but also
downregulates the expression of these factors. Our results are consis-
tent with Kandhare et al. (2012) who reported that chronic administra-
tion of CoQ10 combined with vitamin E exerts neuroprotective effects
on chronic alcoholic-induced neuropathy through decreasing levels of
endogenous oxidative–nitrosative stress factors TNF-α, IL-1β, and IL-4.
Results suggest that these stress-sensitive and neuropathic pain-related
factors may also participate in the mechanisms of diabetic neuropathic
pain. The analgesic mechanism of CoQ10 treatment may come from its
function of anti-oxidative stress and inhibition of these factors.
Recently, Shi et al. (2013) reported that in the db/db mouse model
of type 2 DM, CoQ10 administrated starting at 7 weeks of age until
33 weeks of age, partially protected against neuron loss, counteracted
nerve conduction impairment and prevented the hypoalgesia seen in
the late stages of DM.
In conclusion, we present the inbred NZO/HILtJ strain of mice as a
useful model for the study of type 2 diabetes-associated neuropathy.
We further characterized this model, showing that spontaneous
diabetes and obesity are associated with increased lipid peroxidation
and neuropathic pain behaviors. Additionally, we demonstrated that
systemic CoQ10 suppresses neuropathic pain in this model of obesity
and type 2 diabetes. Finally, our data indicated that CoQ10 effectively
suppresses lipid peroxidation in the DRG, spinal cord, and blood
of diabetic mice, and can prevent or alleviate neuropathic pain-
related behaviors. The analgesic effect of CoQ10 may result from
anti-oxidative stress and further decreasing of stress-sensitive and
pain-related signaling pathways such as MAPK, NF-κB and TLR4.
Taken together, our data suggests that the ideal preventative and/or ther-
apeutic interventions would target the underlying causes of neuropathic
pain early on in the DM disease state, in addition to treating symptoms.
Thus, CoQ10 might represent a reasonable preventative strategy for
long-term use, and if used in combination with other analgesic therapies,
 Y.P. Zhang et al. / Neurobiology of Disease 58 (2013) 169–178
may be a safe and effective long-term approach for the treatment of
diabetic neuropathy.
Acknowledgments
No potential conflicts of interest relevant to this article were
reported. Y.P.Z. and C.Y.S. designed and conducted the trials; Y.P.Z.
wrote the manuscript. Y.Y., A. E., and Y.R. conducted the trial. R.C. L.,
P.T., R.G. and K. A. C. assisted in the designing of the trial, contributed
to the discussion, and edited the final version of the manuscript.
The authors are grateful to the support of Diabetes Action Research
and Education Foundation and partial support from the Department of
Anesthesiology, University of Miami Miller School of Medicine, and
thanks to Elizabeth Kelly, Department of Anesthesiology, University of
Miami Miller School of Medicine, for helping with the review of the
manuscript.
References
Allen, R.G., Tresini, M., 2000. Oxidative stress and gene regulation. Free Radic. Biol.
Med. 28, 463–499.
Battinoa, M., Ferrib, E., Girotti, S., Lenazc, G., 1991. Free radical scavenging activity of
coenzyme Q measured by a chemiluminescent assay. Anal. Chim. Acta 255 (2),
367–371.
Beal, M.F., Henshaw, D.R., Jenkins, B.G., Rosen, B.R., Schulz, J.B., 1994. Coenzyme Q10
and nicotinamide block striatal lesions produced by the mitochondrial toxin
malonate. Ann. Neurol. 36, 882–888.
Bettoni, I., Comelli, F., Rossini, C., Granucci, F., Giagnoni, G., Peri, F., Costa, B., 2008. Glial
TLR4 receptor as new target to treat neuropathic pain: efficacy of a new receptor
antagonist in a model of peripheral nerve injury in mice. Glia 56, 1312–1319.
Calcutt, N.A., Stiller, C., Gustafsson, H., Malmberg, A.B., 2000. Elevated substance-P-like
immunoreactivity levels in spinal dialysates during the formalin test in normal and
diabetic rats. Brain Res. 856, 20–27.
Chang, L., Karin, M., 2001. Mammalian MAP kinase signalling cascades. Nature 410,
37–40.
Chen, H., Charlat, O., Tartaglia, L.A., Woolf, E.A., Weng, X., Ellis, S.J., Lakey, N.D.,
Culpepper, J., Moore, K.J., Breitbart, R.E., Duyk, G.M., Tepper, R.I., Morgenstern,
J.P., 1996. Evidence that the diabetes gene encodes the leptin receptor: identification
of a mutation in the leptin receptor gene in db/db mice. Cell 84, 491–495.
Clement, K., Vaisse, C., Lahlou, N., Cabrol, S., Pelloux, V., Cassuto, D., Gourmelen, M.,
Dina, C., Chambaz, J., Lacorte, J.M., Basdevant, A., Bougneres, P., Lebouc, Y.,
Froguel, P., Guy-Grand, B., 1998. A mutation in the human leptin receptor gene
causes obesity and pituitary dysfunction. Nature 392, 398–401.
Du, Y., Tang, J., Li, G., Berti-Mattera, L., Lee, C.A., Bartkowski, D., Gale, D., Monahan, J.,
Niesman, M.R., Alton, G., Kern, T.S., 2010. Effects of p38 MAPK inhibition on early
stages of diabetic retinopathy and sensory nerve function. Invest. Ophthalmol.
Vis. Sci. 51, 2158–2164.
Dzugkoev, S.G., Kaloeva, M.B., Dzugkoeva, F.S., 2012. Effect of combination therapy
with coenzyme Q10 on functional and metabolic parameters in patients with
type 1 diabetes mellitus. Bull. Exp. Biol. Med. 152, 364–366.
Edwards, J.L., Vincent, A.M., Cheng, H.T., Feldman, E.L., 2008. Diabetic neuropathy:
mechanisms to management. Pharmacol. Ther. 120, 1–34.
Esterbauer, H., Schaur, R.J., Zollner, H., 1991. Chemistry and biochemistry of 4-
hydroxynonenal, malonaldehyde and related aldehydes. Free Radic. Biol. Med. 11,
81–128.
Friedman, J., 2011. Why is the nervous system vulnerable to oxidative stress? In: Gadoth, N.,
Gobel, H.H. (Eds.), Oxidative Stress and Free Radical Damage in Neurology, Oxidative
Stress in Applied Basic Research and Clinical Practice. Springer Science Business Media,
LLC, pp. 19–29.
Fu, E.S., Zhang, Y.P., Sagen, J., Yang, Z.Q., Bethea, J.R., 2007. Transgenic glial nuclear
factor-kappa B inhibition decreases formalin pain in mice. Neuroreport 18, 713–717.
Fu, E.S., Zhang, Y.P., Sagen, J., Candiotti, K.A., Morton, P.D., Liebl, D.J., Bethea, J.R.,
Brambilla, R., 2010. Transgenic inhibition of glial NF-kappa B reduces pain behavior
and inflammation after peripheral nerve injury. Pain 148, 509–518.
Harati, Y., 2007. Diabetic neuropathies: unanswered questions. Neurol. Clin. 25,
303–317.
Haskell, B.D., Flurkey, K., Duffy, T.M., Sargent, E.E., Leiter, E.H., 2002. The diabetes-prone
NZO/HlLt strain. I. Immunophenotypic comparison to the related NZB/BlNJ and
NZW/LacJ strains. Lab. Invest. 82, 833–842.
Heubeck, E., 2006. New treatment options for diabetic neuropathy. Doc News 3, 8–9.
Hirayama, I., Yi, Z., Izumi, S., Arai, I., Suzuki, W., Nagamachi, Y., Kuwano, H., Takeuchi, T.,
Izumi, T., 1999. Genetic analysis of obese diabetes in the TSOD mouse. Diabetes 48,
1183–1191.
Huynh, K., Kiriazis, H., Du, X.J., Love, J.E., Jandeleit-Dahm, K.A., Forbes, J.M., McMullen, J.R.,
Ritchie, R.H., 2012. Coenzyme Q10 attenuates diastolic dysfunction, cardiomyocyte
hypertrophy and cardiac fibrosis in the db/db mouse model of type 2 diabetes.
Diabetologia 55, 1544–1553.
Ji, R.R., Suter, M.R., 2007. p38 MAPK, microglial signaling, and neuropathic pain. Mol.
Pain 3, 33.
177
Joyce, N., Annett, G., Wirthlin, L., Olson, S., Bauer, G., Nolta, J.A., 2010. Mesenchymal
stem cells for the treatment of neurodegenerative disease. Regen. Med. 5, 933–946.
Jurgens, H.S., Schurmann, A., Kluge, R., Ortmann, S., Klaus, S., Joost, H.G., Tschop, M.H.,
2006. Hyperphagia, lower body temperature, and reduced running wheel ac-
tivity precede development of morbid obesity in New Zealand obese mice.
Physiol. Genomics 25, 234–241.
Kanasaki, K., Koya, D., 2011. Biology of obesity: lessons from animal models of obesity.
Available from: J. Biomed. Biotechnol. http://dx.doi.org/10.1155/2011/197636
(Accessed 26th March 2013).
Kandhare, A.D., Ghosh, P., Ghule, A.E., Bodhankar, S.L., 2012. Elucidation of molecular
mechanism involved in neuroprotective effect of coenzyme Q10 in alcohol-induced
neuropathic pain. Fundam. Clin. Pharmacol. (Available from: doi: 10.1111. Accessed
26th March 2013).
Ko, M.K., Saraswathy, S., Parikh, J.G., Rao, N.A., 2011. The role of TLR4 activation in
photoreceptor mitochondrial oxidative stress. Invest. Ophthalmol. Vis. Sci. 52,
5824–5835.
Kucharska, J., Braunova, Z., Ulicna, O., Zlatos, L., Gvozdjakova, A., 2000. Deficit of coenzyme Q
in heart and liver mitochondria of rats with streptozotocin-induced diabetes. Physiol.
Res. 49, 411–418.
Kumar, A., Negi, G., Sharma, S.S., 2012. Suppression of NF-kappaB and NF-kappaB
regulated oxidative stress and neuroinflammation by BAY 11-7082 (IkappaB
phosphorylation inhibitor) in experimental diabetic neuropathy. Biochimie
94, 1158–1165.
Leiter, E.H., Reifsnyder, P.C., Flurkey, K., Partke, H.J., Junger, E., Herberg, L., 1998. NIDDM
genes in mice: deleterious synergism by both parental genomes contributes to
diabetogenic thresholds. Diabetes 47, 1287–1295.
Liao, Y.H., Zhang, G.H., Jia, D., Wang, P., Qian, N.S., He, F., Zeng, X.T., He, Y., Yang, Y.L.,
Cao, D.Y., Zhang, Y., Wang, D.S., Tao, K.S., Gao, C.J., Dou, K.F., 2011. Spinal astrocytic
activation contributes to mechanical allodynia in a mouse model of type 2 diabetes.
Brain Res. 1368, 324–335.
Mancuso, M., Coppede, F., Migliore, L., Siciliano, G., Murri, L., 2006. Mitochondrial
dysfunction, oxidative stress and neurodegeneration. J. Alzheimers Dis. 10, 59–73.
Matsuoka, N., Ogawa, Y., Hosoda, K., Matsuda, J., Masuzaki, H., Miyawaki, T., Azuma, N.,
Natsui, K., Nishimura, H., Yoshimasa, Y., Nishi, S., Thompson, D.B., Nakao, K., 1997.
Human leptin receptor gene in obese Japanese subjects: evidence against either
obesity-causing mutations or association of sequence variants with obesity.
Diabetologia 40, 1204–1210.
Mezawa, M., Takemoto, M., Onishi, S., Ishibashi, R., Ishikawa, T., Yamaga, M., Fujimoto,
M., Okabe, E., He, P., Kobayashi, K., Yokote, K., 2012. The reduced form of coenzyme
Q10 improves glycemic control in patients with type 2 diabetes: an open label pilot
study. Biofactors 38, 416–421.
Montine, K.S., Olson, S.J., Amarnath, V., Whetsell Jr., W.O., Graham, D.G., Montine, T.J.,
1997. Immunohistochemical detection of 4-hydroxy-2-nonenal adducts in
Alzheimer's disease is associated with inheritance of APOE4. Am. J. Pathol. 150,
437–443.
Morrow, T.J., 2004. Animal models of painful diabetic neuropathy: the STZ rat model.
Curr. Protoc. Neurosci. Supplement 29 (9.18), 1–11.
Otto, K.J., Wyse, B.D., Cabot, P.J., Smith, M.T., 2011. Longitudinal study of painful diabetic
neuropathy in the Zucker diabetic fatty rat model of type 2 diabetes: impaired basal
G-protein activity appears to underpin marked morphine hyposensitivity at 6 months.
Pain Med. 12, 437–450.
Pajovic, S.B., Saicic, Z.S., Spasic, M.B., Petrovic, V.M., 2003. The effect of ovarian hormones on
antioxidant enzyme activities in the brain of male rats. Physiol. Res. 52, 189–194.
Persson, M.F., Franzen, S., Catrina, S.B., Dallner, G., Hansell, P., Brismar, K., Palm, F.,
2012. Coenzyme Q10 prevents GDP-sensitive mitochondrial uncoupling, glomeru-
lar hyperfiltration and proteinuria in kidneys from db/db mice as a model of type 2
diabetes. Diabetologia 55, 1535–1543.
Reagan-Shaw, S., Nihal, M., Ahmad, N., 2008. Dose translation from animal to human
studies revisited. FASEB J. 22, 659–661.
Runchel, C., Matsuzawa, A., Ichijo, H., 2011. Mitogen-activated protein kinases in mammalian
oxidative stress responses. Antioxid. Redox Signal. 15, 205–218.
Shi, T.J., Zhang, M.D., Zeberg, H., Nilsson, J., Grunler, J., Liu, S.X., Xiang, Q., Persson, J.,
Fried, K.J., Catrina, S.B., Watanabe, M., Arhem, P., Brismar, K., Hokfelt, T.G., 2013.
Coenzyme Q10 prevents peripheral neuropathy and attenuates neuron loss in the
db−/db− mouse, a type 2 diabetes model. Proc. Natl. Acad. Sci. U. S. A. 110, 690–695.
Smith, T., Nicholson, R.A., 2007. Review of duloxetine in the management of diabetic
peripheral neuropathic pain. Vasc. Health Risk Manag. 3, 833–844.
Sourris, K.C., Harcourt, B.E., Tang, P.H., Morley, A.L., Huynh, K., Penfold, S.A., Coughlan, M.T.,
Cooper, M.E., Nguyen, T.V., Ritchie, R.H., Forbes, J.M., 2012. Ubiquinone (coenzyme
Q10) prevents renal mitochondrial dysfunction in an experimental model of type 2
diabetes. Free Radic. Biol. Med. 52, 716–723.
Springer, J.E., Azbill, R.D., Mark, R.J., Begley, J.G., Waeg, G., Mattson, M.P., 1997. 4-
hydroxynonenal, a lipid peroxidation product, rapidly accumulates following trau-
matic spinal cord injury and inhibits glutamate uptake. J. Neurochem. 68, 2469–2476.
Stanzione, P., Tropepi, D., 2011. Drugs and clinical trials in neurodegenerative diseases.
Ann. Ist. Super. Sanita 47, 49–54.
Talbot, S., Couture, R., 2012. Emerging role of microglial kinin B1 receptor in diabetic
pain neuropathy. Exp. Neurol. 234, 373–381.
Tanga, F.Y., Nutile-McMenemy, N., DeLeo, J.A., 2005. The CNS role of Toll-like receptor 4
in innate neuroimmunity and painful neuropathy. Proc. Natl. Acad. Sci. U. S. A. 102,
5856–5861.
Tjolsen, A., Berge, O.G., Hunskaar, S., Rosland, J.H., Hole, K., 1992. The formalin test: an
evaluation of the method. Pain 51, 5–17.
Veves, A., Backonja, M., Malik, R.A., 2008. Painful diabetic neuropathy: epidemi-
ology, natural history, early diagnosis, and treatment options. Pain Med. 9,
660–674.
178 Y.P. Zhang et al. / Neurobiology of Disease 58 (2013) 169–178
Vincent, A.M., Edwards, J.L., Sadidi, M., Feldman, E.L., 2008. The antioxidant response as
a drug target in diabetic neuropathy. Curr. Drug Targets 9, 94–100.
Vinik, A., 2005. Clinical review: use of antiepileptic drugs in the treatment of chronic
painful diabetic neuropathy. J. Clin. Endocrinol. Metab. 90, 4936–4945.
Wadsworth, T.L., Bishop, J.A., Pappu, A.S., Woltjer, R.L., Quinn, J.F., 2008. Evaluation of
coenzyme Q as an antioxidant strategy for Alzheimer's disease. J. Alzheimers Dis.
14, 225–234.
Wheeler-Aceto, H., Cowan, A., 1991. Standardization of the rat paw formalin test for
the evaluation of analgesics. Psychopharmacology (Berl) 104, 35–44.
Yang, L., Calingasan, N.Y., Wille, E.J., Cormier, K., Smith, K., Ferrante, R.J., Beal, M.F., 2009. Com-
bination therapy with coenzyme Q10 and creatine produces additive neuroprotective ef-
fects in models of Parkinson's and Huntington's diseases. J. Neurochem. 109, 1427–1439.
Zhang, Y.P., Eber, A., Yuan, Y., Yang, Z., Rodriguez, Y., Levitt, R.C., Takacs, P., Candiotti,
K.A., 2013. Prophylactic and antinociceptive effects of coenzyme Q10 on diabetic
neuropathic pain in a mouse model of type 1 diabetes. Anesthesiology 118, 945–954.
Zhou, M., Zhi, Q., Tang, Y., Yu, D., Han, J., 1999. Effects of coenzyme Q10 on myocardial
protection during cardiac valve replacement and scavenging free radical activity in
vitro. J. Cardiovasc. Surg. (Torino) 40, 355–361.