Available via license: CC BY 4.0
Content may be subject to copyright.
RESEARCH ARTICLE
Thymoquinone ameliorates diabetic
phenotype in Diet-Induced Obesity mice via
activation of SIRT-1-dependent pathways
Shpetim Karandrea
1
, Huquan Yin
1
, Xiaomei Liang
1
, Angela L. Slitt
2
, Emma A. Heart
1
*
1Department of Molecular Pharmacology and Physiology, University of South Florida, Tampa, Florida,
United States of America, 2Department of Pharmaceutical Sciences, University of Rhode Island, Kingston,
Rhode Island, United States of America
*eheart@health.usf.edu
Abstract
Thymoquinone, a natural occurring quinone and the main bioactive component of plant
Nigella sativa, undergoes intracellular redox cycling and re-oxidizes NADH to NAD
+
. TQ
administration (20 mg/kg/bw/day) to the Diet-Induced Obesity (DIO) mice reduced their
diabetic phenotype by decreasing fasting blood glucose and fasting insulin levels, and
improved glucose tolerance and insulin sensitivity as evaluated by oral glucose and insulin
tolerance tests (OGTT and ITT). Furthermore, TQ decreased serum cholesterol levels and
liver triglycerides, increased protein expression of phosphorylated Akt, decreased serum
levels of inflammatory markers resistin and MCP-1, and decreased NADH/NAD
+
ratio.
These changes were paralleled by an increase in phosphorylated SIRT-1 and AMPKαin
liver and phosphorylated SIRT-1 in skeletal muscle. TQ also increased insulin sensitivity in
insulin-resistant HepG2 cells via a SIRT-1-dependent mechanism. These findings are con-
sistent with the TQ-dependent re-oxidation of NADH to NAD
+
, which stimulates glucose and
fatty acid oxidation and activation of SIRT-1-dependent pathways. Taken together, these
results demonstrate that TQ ameliorates the diabetic phenotype in the DIO mouse model of
type 2 diabetes.
Introduction
Maintenance of glucose homeostasis involves insulin secretion from the pancreatic β-cells in
response to a rise in blood glucose, and insulin action in target tissues (predominantly liver,
muscle, and adipose tissue) to stimulate glucose entry and utilization, and inhibit hepatic glu-
cose production [1]. Development of type 2 diabetes (T2D) involves both peripheral insulin
resistance and pancreatic β-cell dysfunction. Insulin resistance, the inability of peripheral tis-
sues to properly respond to insulin, is initially compensated by a rise in insulin output in order
to maintain normoglycemia [1]. However, this compensatory mechanism is impaired in indi-
viduals predisposed to T2D, and later results in overt hyperglycemia [2,3].
Thymoquinone (TQ) is the main bioactive component of Nigella sativa, a spice plant of
Ranunculacea family, and a traditional medicine that has been used to treat diabetes symptoms
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 1 / 19
a1111111111
a1111111111
a1111111111
a1111111111
a1111111111
OPEN ACCESS
Citation: Karandrea S, Yin H, Liang X, Slitt AL,
Heart EA (2017) Thymoquinone ameliorates
diabetic phenotype in Diet-Induced Obesity mice
via activation of SIRT-1-dependent pathways. PLoS
ONE 12(9): e0185374. https://doi.org/10.1371/
journal.pone.0185374
Editor: Guillermo Lo
´pez Lluch, Universidad Pablo
de Olavide, SPAIN
Received: March 2, 2017
Accepted: September 12, 2017
Published: September 26, 2017
Copyright: ©2017 Karandrea et al. This is an open
access article distributed under the terms of the
Creative Commons Attribution License, which
permits unrestricted use, distribution, and
reproduction in any medium, provided the original
author and source are credited.
Data Availability Statement: All relevant data are
within the paper and its Supporting Information
files.
Funding: The work was supported by National
Institute of Diabetes and Digestive and Kidney
Diseases Grant R01-DK-098747 and American
Diabetes Association Grant No. 7-12- BS-073 (E. A.
Heart). Shpetim Karandrea was supported by the
Graduate Student Success Fellowship (University
of South Florida).
and to lower blood glucose [4]. Nigella sativa has been reported to increase both insulin secre-
tion and insulin sensitivity [5,6]. TQ has been shown to reduce hepatic glucose production [7]
and protect β-cells from oxidative stress following streptozotocin (STZ) treatment [8]. How-
ever, mechanistic studies and comprehensive evaluation of TQ action under physiological dia-
betic conditions and models is currently lacking.
TQ belongs to the family of quinones, naturally-derived compounds featuring a conjugated
double bond system, which is responsible for their reactivity and intracellular process known
as “redox cycling” [9]. Our laboratory has been instrumental in establishing the concept that
re-oxidation of NADH back to NAD
+
via quinone-dependent redox cycling lowers cellular
reductive poise and facilitates glucose and fatty acid oxidation, and is necessary for the overall
health of the cells [9,10]. Our group has previously shown that TQ supports redox cycling in
pancreatic β-cells, resulting in the reduction of NADH/NAD
+
ratio and normalization of
defective glucose-stimulated insulin secretion (GSIS) under chronically elevated glucose via
inhibition of acetyl CoA carboxylase (ACC) and enhanced oxidation of glucose and fatty acids
[11].
The oxidation status of nicotinamide adenine dinucleotide, represented by the ratio
between its reduced and oxidized forms (NADH/NAD
+
) is a critical determinant of the direc-
tion of metabolic flux [12,13], as NAD
+
promotes oxidative pathways via activation of TCA
cycle enzymes [14]. Furthermore, increased intracellular level of NAD
+
activates SIRT1-de-
pendent metabolic pathways, which stimulate energy metabolism, enhance life span, and can
positively regulate insulin secretion and insulin signaling [14,15].
Here we evaluated the capacity of TQ to ameliorate the diabetic phenotype in a physiologi-
cally relevant rodent model of obesity and diabetes, Diet-Induced Obesity (DIO) mice. We
hypothesized that sustained decrease in the NADH/NAD
+
ratio due to TQ-dependent redox
cycling will result in the enhanced fuel oxidation and amplification of NAD
+
-dependent SIRT-
1 pathway in metabolic tissues, leading to the enhanced insulin sensitivity and improved glu-
cose homeostasis.
Materials and methods
Chemicals
Human recombinant insulin, resveratrol, and AICAR were purchased from Tocris Bioscience
(Bristol, UK). Nicotinamide was purchased from Acros Organics (Geel, Belgium) and Com-
pound C was purchased from EMD Millipore (Billerica, MA). All other chemicals and reagents
were purchased from Sigma (St Louis, MO) unless specified otherwise. Stock solutions of thy-
moquinone, resveratrol, AICAR, Nicotinamide, and Compound C were prepared in DMSO
and added to culture medium to achieve the indicated concentrations.
Ethics statement
All procedures were performed in accordance with and approved by the Institutional Animal
Care and Use Committee (IACUC) of the University of South Florida.
Animals
Male C57BL/6J mice (6 weeks of age) were purchased from Jackson Laboratories (Bar Harbor,
ME) and housed (4 animals per cage) in a USF Animal Facility; room was maintained at a con-
stant temperature (25˚C) in a light:dark 12:12-h schedule. Food and water was available ad libi-
tum. Body weight was monitored on a weekly basis. Mice were pair fed either control low fat
diet, LFD (10% fat cal, Research Diets, New Brunswick, NJ) or high fat diet, HFD (45% fat cal,
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 2 / 19
Competing interests: The authors have declared
that no competing interests exist.
Research Diets, New Brunswick, NJ). Mice were separated in the following groups: LFD, LFD
+TQ, HFD, HFD+TQ. TQ (dissolved in canola oil) was administered daily by oral gavage at
20 mg/kg body weight for the duration of the study. Vehicle only (canola oil) was administered
to control groups (LFD and HFD). The dose of TQ was chosen because it was shown to lower
blood glucose [16], albeit in a non-physiological rodent model of diabetes. The chosen dose is
well below toxic doses established for oral administration in mice [17]. As expected, TQ was
well tolerated, and TQ administration did not affect the overall health of the animals in the
study. After 24 weeks, animals were euthanized with isoflurane, tissues and serum collected,
and either used immediately or were snap frozen in liquid nitrogen and stored in -80˚C until
further use.
Cell culture
HepG2 human hepatoma cell line was purchased from American Type Culture Collection
(ATCC, Manassas, VA) and cultured in DMEM medium supplemented with 10% FBS, 100
units of penicillin, and 100 μg/mL streptomycin at 37˚C in a humified incubator with 5% CO
2
.
Cells were made insulin resistant by treatment with 20mM glucose for 18 hours, as previously
described [18,19]. Following high glucose treatment, cells were starved for 2 hours in serum-
free medium, prior to treatment with the respective compounds for 24 hours. For inhibitor
treatment, cells were pre-incubated with the inhibitors for 30 mins, and the inhibitors were
also present during the 24-hour incubation period. To measure insulin signaling, insulin was
added during the last 30 minutes. Vehicle-treated cells (0.5% DMSO) in normal (5.5 mM) and
high (20 mM) glucose conditions served as controls.
OGTT and ITT
For in vivo studies, animals were anesthetized with ketamine (80 mg/kg body weight). Oral
glucose and insulin tolerance tests were performed following a 6 hr fast. Mice were oral
gavaged with 2 mg/kg/bw glucose (OGTT), or injected intraperitoneally with 0.5 IU insulin/
kg/bw (ITT). Blood glucose, obtained at 0, 15, 30, 60, 90, 120 and 180 minutes from the tail
vein was measured with a glucometer (Bayer Contour).
Cholesterol content
Total cholesterol, HDL, and LDL/VLDL content was determined from serum samples using
the HDL and LDL/VLDL Cholesterol Assay Kit (abcam, Cambridge, MA) according to the
manufacturer’s protocols.
Serum profile
Serum levels of insulin, resistin and MCP-1 were determined by Ocean Ridge Biosciences
(Deerfield Beach, FL) using a Luminex multiplex protein profiling assay (Luminex Corp., Aus-
tin, TX) according to the manufacturer’s protocols.
Western blot analysis
Liver and soleus muscle tissues were solubilized in RIPA lysis buffer (Pierce, Rockford, IL)
using Fast Prep 24G system (MP Biosciences, Santa Ana, CA). After exposure, HepG2 cells
were solubilized in RIPA lysis buffer. Protein content was determined using a BCA Protein
Assay Kit (Pierce, Rockford, IL) and SDS samples were prepared. Equal amount of protein
(100 μg per lane) were electrophoretically separated on SDS-polyacrylamide gel, followed by
blotting onto PVDF membrane. Following the transfer, membranes were blocked with TBST
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 3 / 19
(10 mmol/l Tris-HCl pH 7.4, 150 mmol/l NaCl, and 0.1% Tween 20) containing 5% nonfat dry
milk (blocking buffer) and incubated with the primary antibodies (diluted in blocking buffer
overnight at 4˚C) against SIRT-1 (Cell Signaling, cat. #9475), p-SIRT-1 (Cell Signaling, cat.
#2314), Akt (Cell Signaling, cat. #9272), p-Akt (Cell Signaling, cat. #9271), AMPKα(Cell Sig-
naling, cat. #5831), p-AMPKα(Cell Signaling, cat. #2535), NQO1 (Santa Cruz, cat. #sc-16464),
β-actin (Cell Signaling, cat. #4970), and β-tubulin (Cell Signaling, cat. #2146). Membranes
were incubated with goat anti-rabbit immunoglobulin (IgG) secondary antibody (Santa Cruz,
cat. #sc-2030) for 1 h at room temperature, and washed 5 times. Proteins were detected by
using enhanced chemiluminescence. Semiquantitative analysis of Western blot images were
performed using ImageJ.
Triglyceride content
Triglyceride content was determined in liver and soleus muscle RIPA buffer lysates (lysates as
described above) using the Triglyceride kit (Pointe Scientific, Canton, MI) according to the
manufacturer’s protocols.
Metabolomics analysis
Serum levels of glycerol, palmitic acid, oleic acid, and stearic acid were measured by gas chro-
matography—mass spectrometry (GC/MS) analysis. The GC/MS experiments were performed
by the University of Utah Metabolomics Core.
Determination of nucleotides
NADH/NAD
+
ratio was determined in liver and soleus muscle using the NAD/NADH assay
kit as per the manufacturer’s protocol (Abcam, Cat #65348, Cambridge, UK).
Quantitative real time RT-PCR
The tissue samples stored in RNAlater (Invitrogen, Carlsbad, CA) were homogenized by using
the Fast Prep 24G instrument (MP Biosciences, Santa Ana, CA). Total RNA was prepared
using the TRIzol reagent according to the manufacturer’s protocol (Invitrogen, Carlsbad, CA)
and single-strand cDNA was synthesized from the RNA in a reaction mixture containing opti-
mum blend of oligo(dT) primers and iScript reverse transcriptase (Bio-Rad, Richmond, CA).
qRT-PCR amplifications were performed using rEVAlution 2x qPCR Master Mix (Empirical
Bioscience, Grand Rapids, MI) in an MyIQ2 Real-Time PCR Detection System (Bio-Rad,
Richmond, CA) following manufacturer’s protocol. To determine the specificity of amplifica-
tion, melting curve analysis was applied to all final PCR products. The relative amount of tar-
get mRNA was calculated by the comparative threshold cycle method by normalizing target
mRNA threshold cycle to those for glyceraldehyde-3-phosphate dehydrogenase (GAPDH).
The primers used for analysis were as follows: NQO1: sense primer, 5’-AGGATGGGAGGTAC
TCGAATC-3’, anti-sense primer, 5’-AGGCGTCCTTCCTTATATGCTA-3’; GAPDH: sense
primer, 5’-CTTCACCACCATGGAGAAGGC-3’, anti-sense primer, 5’-GGCATGGACTGTGG
TCATGAG-3’.
Statistical analysis
Data are expressed as means ±SEM. Significance was determined for multiple comparisons
using one-way or two-way analysis of variance (ANOVA) followed by Sidak or Holm-Sidak
multiple comparisons tests [20,21] for planned comparisons (as mentioned in each figure) or
independent t-test as indicated. A p-value of 0.05 was considered significant.
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 4 / 19
Results
The Diet Induced Obesity (DIO) mice develop obesity, hyperinsulinemia, glucose intolerance
and insulin resistance when fed a high fat diet, making them a suitable model to study type 2
diabetes pathophysiology [22,23]. This is confirmed in our study, where after high fat feeding,
mice developed a diabetic phenotype as shown by the weight gain (Fig 1A), elevated fasting
blood glucose (BG) and insulin levels (Fig 1B and 1C), and impaired oral glucose and insulin
tolerance tests (OGTT an ITT) (Fig 2A and 2B). TQ administration was effective in ameliorat-
ing these parameters: TQ lowered body weight (Fig 1A), fasting blood glucose and insulin (Fig
1B and 1C, respectively), and improved glucose tolerance and insulin sensitivity, evaluated by
OGTT and ITT (Fig 2A and 2B).
Type 2 diabetes is associated with increased inflammation, which can contribute to insulin
resistance and is shown to be detrimental to many tissues including pancreatic β-cells [24,25].
Resistin, a hormone secreted by adipocytes, impairs glucose tolerance and insulin sensitivity in
mice [26] and has been associated with insulin resistance in humans [27,28]. Monocyte che-
moattractant protein-1 (MCP-1) is a pro-inflammatory chemokine that can induce insulin
resistance [29] and circulating levels of this chemokine are increased in patients with type 2
diabetes [30–32]. TQ lowered serum levels of resistin in DIO mice (Fig 3A). There was a trend
to lower the MCP-1 levels, however, this didn’t reach statistical significance in HFD animals
(p = 0.06), although TQ decreased MCP-1 in LFD animals (Fig 3B). These results demonstrate
the potential of TQ to alleviate tissue inflammation in diabetes and obesity.
Elevated levels of triglycerides, together with decreased HDL and increased LDL cholesterol
levels are the key identifiers of diabetic dyslipidemia, which can exacerbate insulin resistance
[33]. Consistent with our previously reported data demonstrating TQ-dependent increase in
fatty acid oxidation [11], and observed increased peripheral insulin sensitivity in this study (as
shown by the improvement of the ITT in DIO mice, Fig 2B), TQ ameliorated HFD-dependent
increase in liver triglyceride levels (Fig 4A). There was a trend to lower HFD-dependent mus-
cle triglyceride content, however this did not reach statistical significance (Fig 4B). We saw
similar trends when analyzed serum glycerol and three relevant fatty acids: palmitic acid, oleic
acid, and stearic acid. GC/MS analysis of serum levels of these metabolites were decreased
compared to HFD alone (Table 1), however this didn’t reach statistical significance.
There was also a trend to decrease serum cholesterol level, albeit statistically not significant
(Fig 5A). However, TQ significantly decreased the levels of LDL cholesterol in the serum of
HFD animals (Fig 5C), with no effect on the HDL levels (Fig 5B). This effect was selective to
the HFD diet, as LFD animals did not demonstrate changes in their HDL or LDL/VLDL cho-
lesterol in response to TQ regimen.
The lowered tissue triglyceride levels following TQ administration argues for the TQ-
dependent activation of the oxidative pathways (and consequent oxidation, rather than deposi-
tion of metabolic substrates). NADH/NAD
+
ratio is important determinant of metabolic flux
[14], and our group previously reported that TQ lowers NADH/NAD
+
ratio in pancreatic β-
cells exposed to glucose overload [11]. To confirm that TQ exerts this effect in vivo, we mea-
sured NADH/NAD
+
ratio in liver and skeletal muscle. In liver, there was an increase in this
ratio in HFD mice (Fig 6A), which is in agreement with prior studies suggesting an increase in
NADH in diabetes and obesity [14]. However, we did not observe this change in skeletal mus-
cle (Fig 6B). In both liver and soleus muscle, TQ lowered the NADH/NAD+ ratio in the HFD
group compared to HFD alone (Fig 6A and 6B).
Since NADH/NAD
+
ratio is known to regulate SIRT-1 pathway, we analyzed effect of TQ
feeding on this pathway in the liver and soleus muscle of TQ-treated compared to control
mice. Liver and soleus muscle from mice treated with TQ had enhanced phosphorylated
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 5 / 19
Fig 1. TQ ameliorates weight gain, lowers fasting blood glucose and insulin in DIO mice. (A) Effect of
TQ on body weight (B) Effect of TQ treatment on fasting blood glucose after a 6 hour fast. (C) Effect of TQ on
serum insulin. Total body weight was measured weekly for the duration of the study. p<0.05 when comparing
HFD and LFD (+), and HFD and HFD+TQ (*), using a one-way ANOVA followed by Sidak post-test (A and B)
or independent t-test (C). Results are means ±SEM (n = 10–12 mice per treatment group). LFD: low fat diet,
HFD: high fat diet, TQ: thymoquinone.
https://doi.org/10.1371/journal.pone.0185374.g001
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 6 / 19
Fig 2. TQ normalizes glucose tolerance and insulin sensitivity. (A) Blood glucose levels in response to
oral glucose tolerance test (OGTT). (B) Blood glucose levels in response to insulin tolerance test (ITT).
p<0.05 when comparing HFD and LFD (+), and HFD and HFD+TQ (*), using a two-way ANOVA followed by
Holm-Sidak post-test. Results are means ±SEM (n = 10–12 mice per treatment group). LFD: low fat diet,
HFD: high fat diet, TQ: thymoquinone.
https://doi.org/10.1371/journal.pone.0185374.g002
Fig 3. Effects of TQ on serum resistin and MCP-1. (A) Resistin serum concentration. (B) MCP-1 serum
concentration. p0.05 when comparing (+) HFD and LFD, (*) HFD + TQ and HFD, and (#) LFD and LFD +
TQ using independent t-tests. Results are means ±SEM (n = 10–12 mice per treatment group). LFD: low fat
diet, HFD: high fat diet, TQ: thymoquinone, MCP-1: monocyte chemotactic protein 1.
https://doi.org/10.1371/journal.pone.0185374.g003
Fig 4. Effects of TQ on triglyceride content in liver and muscle. (A) Triglyceride concentration in liver. (B)
Triglyceride concentration in soleus muscle. (*) p<0.05 when comparing HFD + TQ and HFD using a one-way
ANOVA followed by Sidak post-test. Results are means ±SEM (n = 8–12 mice per treatment group). LFD: low
fat diet, HFD: high fat diet, TQ: thymoquinone.
https://doi.org/10.1371/journal.pone.0185374.g004
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 7 / 19
(activated) SIRT-1 in both LFD and HFD groups (Fig 7A–7D). We also analyzed the protein
expression levels of other SIRT proteins in the liver, and did not see a difference with TQ treat-
ment across groups for SIRT-2, SIRT-3, SIRT-5, SIRT-6, and SIRT-7 (S1 Fig). This could be
due to SIRTs 2–7 having a lower deacetylase activity, as SIRT-6 and SIRT-7 have been previ-
ously shown to have a lower NAD
+
-deacetylase activity compared to SIRT-1 [34]. In the liver,
TQ enhanced AMPKαphosphorylation as well as phosphorylation of Akt (protein kinase B), a
key member of insulin signaling pathway [35,36] (Fig 8A and 8B).
To evaluate the mechanistic actions behind TQ-induced insulin sensitivity, we used the
HepG2 cell line as an in vitro model of insulin resistance to assess whether this action is SIRT-
1-dependent. HepG2 cells were made insulin resistant as previously described [18,19], which
was confirmed by decreased p-Akt protein after high glucose treatment (Fig 9A and 9B). TQ
increased p-Akt in high-glucose treated cells, restoring these levels to that of the control cells
(Fig 9A and 9B). This shows that TQ improves insulin resistance in similar fashion to what we
see in livers of DIO mice. This action showed to be SIRT-1-dependent, as pre-treatment of
insulin resistant cells with SIRT-1 inhibitor nicotinamide in the presence of TQ, significantly
decreases p-Akt protein and TQ-induced insulin sensitivity (Fig 9A and 9B). Furthermore,
treatment with SIRT-1 activator resveratrol and AMPKαactivator AICAR increased insulin
sensitivity, although this trend was not statistically significant (Fig 9A and 9B). Pre-treatment
with compound C (AMPKαinhibitor) or compound C and nicotinamide in the presence of
TQ decreased insulin sensitivity compared to TQ treatment alone, albeit statistically insignifi-
cant (Fig 9A and 9B). TQ treatment showed similar trends to the in vivo experiments in
increasing phosphorylation of SIRT-1 and AMPKαin insulin-resistant cells (S2A–S2D Fig).
Trends were also observed in increased p-SIRT-1 and p-AMPKαwith resveratrol and AICAR
in the presence of TQ (S2A–S2D Fig), as well as a decrease in phosphorylation of SIRT-1 with
nicotinamide or compound C in the presence of TQ after high glucose treatment (S2A and
S2B Fig). Pre-treatment with compound C or with compound C and nicotinamide signifi-
cantly decreased p-AMPKαin the presence of TQ compared to TQ treatment alone (S2C and
S2D Fig). Taken together, these results provide additional support about the role of TQ in
improving insulin resistance, as well as show that this action is likely mediated by SIRT-1
activation.
TQ applied in this study was within the physiologically relevant diet-derived levels. How-
ever, non-physiologically high and toxic levels of quinones is known to generate excessive lev-
els of reactive oxygen intermediates via quinone-dependent redox cycling, and this causes
induction of the NAD(P)H-dependent Quinone Oxidoreductase 1 (NQO1). NQO1 is a phase
2 detoxification enzyme induced in response to oxidative stress, which expression is regulated
by the Keap1/Nrf2/ARE pathway [10,37], and NQO1 alone has been show to regulate NADH/
Table 1. Effect of TQ on serum glycerol and fatty acids.
Metabolite Treatment
LFD LFD + TQ HFD HFD + TQ
Glycerol 844.7 ±70.1
a
1282 ±101.9
c
1348 ±124.2
b
1186 ±47.1
Palmitic Acid 820.3 ±26.1 970.4 ±61.3 862.7 ±53.8 798.7 ±23.4
Oleic Acid 2851 ±179.8 3335 ±195.8 2807 ±345.9 2597 ±114.5
Stearic Acid 381.9 ±15.0 371.9 ±22.3 471.2 ±25.8 437.5 ±21.4
Results expressed as means ±SEM. n = 10–12 mice/group. Means within the same row with different superscripts differ, p 0.05 as determined by using a
one-way ANOVA followed by Sidak post-test. a, b = LFD vs. HFD only; a, c = LFD vs. LFD + TQ only. TQ = Thymoquinone, LFD = low fat diet, HFD = high
fat diet.
https://doi.org/10.1371/journal.pone.0185374.t001
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 8 / 19
Fig 5. Effects of TQ on serum cholesterol. (A) Total cholesterol serum concentration. (B) HDL cholesterol
serum concentration. (C) LDL/VLDL cholesterol serum concentration. p0.05 when comparing (+) HFD and
LFD, (*) HFD + TQ and HFD using independent t-tests. Results are means ±SEM (n = 6–7 mice per
treatment group). LFD: low fat diet, HFD: high fat diet, TQ: thymoquinone, LDL: low-density lipoprotein, HDL:
high-density lipoprotein, VLDL: very-low-density lipoprotein.
https://doi.org/10.1371/journal.pone.0185374.g005
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 9 / 19
NAD
+
ratio [10,38]. To ascertain that applied doses of TQ were physiologically low and not
inductive of NQO1 and/or oxidative stress, mRNA and protein levels of NQO1 were measured
in liver and muscle. Levels of NQO1 were not elevated in any of these tissues, confirming that
applied doses, while effective in regulating the cellular redox, do not activate the Keap1/Nrf2/
ARE pathway and do not increase oxidative stress (Fig 10). This further supports our
Fig 6. Effects of TQ on NADH/NAD
+
ratio in liver and soleus muscle. (A) NADH/NAD
+
ratio in liver. (B)
NADH/NAD
+
ratio in soleus muscle. p 0.05 when comparing (+) HFD and LFD, (*) HFD + TQ and HFD, and
(#) LFD and LFD + TQ using independent t-tests. Results are means ±SEM (n = 8–10 mice per treatment
group). LFD: low fat diet, HFD: high fat diet, TQ: thymoquinone.
https://doi.org/10.1371/journal.pone.0185374.g006
Fig 7. Effects of TQ on SIRT-1 protein expression. (A) Western blot images of SIRT-1 and p-SIRT-1
protein in liver. β-actin was used as a loading control. (B) Western blot images of SIRT-1 and p-SIRT-1 protein
in soleus muscle. β-tubulin was used as a loading control. Western blot images are representative of
combined liver and soleus muscle lysates from n = 10–12 mice per treatment group. (C and D) Protein band
quantification using densitometry from three independent experiments. p 0.05 when comparing (+) HFD
and LFD and (*) HFD + TQ and HFD using independent t-tests LFD: low fat diet, HFD: high fat diet, TQ:
thymoquinone.
https://doi.org/10.1371/journal.pone.0185374.g007
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 10 / 19
hypothesis that TQ-dependent re-oxidation of NADH and consequent decrease of the
NADH/NAD
+
ratio is the main mechanism to activate SIRT-1/AMPK pathway and promote
fuel oxidation rather than deposition, which leads to the observed changes in normalization of
glucose homeostasis in DIO mice following TQ administration.
Discussion and conclusions
This is the first in vivo study aimed to comprehensively evaluate the effect of thymoquinone
(TQ), a bioactive component of the Nigella sativa plant, on whole body glucose homeostasis
using a physiologically-relevant mouse model of type 2 diabetes. In our published in vitro
study, we have reported that both Nigella sativa extract (NSE) of high thymoquinone (TQ)
content, as well as TQ alone, decreased NADH/NAD
+
ratio and stimulated glucose and fatty
acid oxidation in pancreatic β-cells, and this action was accompanied by the restoration of the
glucose-stimulated insulin secretion (GSIS) in cells exposed to glucose overload [11]. Here we
have expanded our studies to an in vivo model with focus on the TQ effect on the insulin sensi-
tive peripheral tissues, and evaluated the action of TQ on glucose homeostasis in Diet Induced
Obesity (DIO) mice.
After 24 weeks of HFD, C57/BLJ mice became obese and diabetic, as demonstrated by their
increased body weight (Fig 1A), elevated fasting blood glucose (Fig 1B), insulin (Fig 1C) and
impaired OGTT and ITT (Fig 2). While TQ treatment improved all these parameters in HFD
animals, TQ had no significant effect on weight, fasting blood glucose and insulin, or OGTT
/ITT in animals treated with LFD, suggesting that TQ primarily affects DIO metabolism by
increasing oxidation of diet-derived fatty acid surplus. However, it is still possible that TQ
treatment beyond the 24 weeks could lead to observed changes in physiological parameters in
the LFD group as well, and further studies are required to address this issue. Bioavailability of
TQ after an oral administration can be a limiting factor on TQ actions. Although such studies
have been very limited in mice, studies with other animal models have shown that TQ is rap-
idly eliminated and slowly absorbed [39,40]. Therefore, further studies are required to address
the bioavailability of TQ after oral administration in mice to properly determine a relevant
Fig 8. Effects of TQ on Akt and AMPKαprotein expression in liver. (A) Western blot images of Akt, p-Akt,
AMPKαand p-AMPKαprotein in liver. β-actin was used as a loading control. Western blot images are
representative of combined liver lysates from n = 10–12 mice per treatment group. (B) Protein band
quantification using densitometry from three independent experiments. p0.05 when comparing (+) HFD and
LFD, (*) HFD + TQ and HFD, and (#) LFD and LFD + TQ using independent t-tests. LFD: low fat diet, HFD:
high fat diet, TQ: thymoquinone.
https://doi.org/10.1371/journal.pone.0185374.g008
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 11 / 19
Fig 9. TQ improves insulin sensitivity in HepG2 cells via a SIRT-1 dependent mechanism. HepG2 cells
were cultured in high (20 mM) glucose or in growth media containing 5.5 mM glucose for 18 hours, starved
with serum-free media for 2 hours, then pre-incubated with vehicle control (0.5% DMSO), nicotinamide (0.5
mM), compound C (20 μM), or with nicotinamide and compound C together for 30 mins, followed by
incubation with TQ (10 μM) in the presence or absence of nicotinamide and compound C; or with TQ,
resveratrol (50 μM), or AICAR (2 mM) alone for 24 hours in 20mM glucose media. Vehicle-treated cells in 5.5
mM glucose served as control. Insulin (100 nM) was added during the last 30 min. (A) Western blot images of
p-Akt, Akt, and β-actin. (B) Protein band quantification using densitometry from three independent
experiments. p0.05 where (*) is significantly different from 5.5G, (#) is significantly different from 20G, and
(Δ) is significantly different from 20G + TQ using independent t-tests. 5.5 G: 5.5 mM glucose, 20G: 20 mM
glucose, TQ: thymoquinone, R: resveratrol, AIC: AICAR, NIC: nicotinamide, C: compound C.
https://doi.org/10.1371/journal.pone.0185374.g009
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 12 / 19
dose and exposure window, particularly in physiologically relevant mouse models of type 2
diabetes.
Metabolism is governed by the oxidation status of nicotinamide adenine dinucleotide, rep-
resented by the ratio between its reduced and oxidized forms (NADH/NAD
+
) [14]. During
glycolysis NAD
+
is reduced to NADH, which needs to be re-oxidized back to NAD
+
[14]. In
chronic hyperglycemic conditions, such as in type 2 diabetes, there can be NADH overproduc-
tion due to the fact that mitochondrial shuttles are unable to efficiently re-oxidize NADH,
which leads to the condition known as reductive stress [14,41]. This leads to increased pres-
sure on mitochondrial complex I, the primary site of NADH recycling, which in turn causes
the formation of superoxide [14,42] and enhanced oxidative stress, known to be detrimental
to insulin sensitivity and insulin secretion and exacerbate the diabetic phenotype [43]. NADH
excess inhibits glycolytic and TCA cycle enzymes (glyceraldehyde 3-phosphate dehydrogenase,
pyruvate dehydrogenase, isocitrate dehydrogenase, α-ketoglutarate dehydrogenase, malate
dehydrogenase), leading to the impairment of glucose oxidation and TCA cycle oxidative path-
ways [14,43]. TQ has been shown to regulate oxidation level of adenine nucleotides [11]. Due
to its conjugated double bond system, TQ is able to re-oxidize NADH in the process of NAD
(P)-dependent redox cycling [44], and thus decrease the NADH/NAD
+
ratio, as shown by our
group [11]. Furthermore, in this study we also demonstrate that TQ treatment leads to a
decrease in the NADH/NAD
+
ratio in liver and skeletal muscle in HFD mice (Fig 6). Regener-
ation of NAD
+
from TQ can thus increase glucose and fatty acid oxidation and ameliorate
diabetic dyslipidemia. Diabetic dyslipidemia is characterized by high plasma triglyceride
Fig 10. Effects of TQ on NQO1 expression. NQO1 mRNA expression in liver (A) and soleus muscle (B). (C)
Western blot images of NQO1 and β-actin protein in liver (D) Western blot images of NQO1 protein in soleus
muscle. β-tubulin was used as a loading control. Statistical analysis (A and B): one-way ANOVA followed by
Sidak post-test (p0.05). qPCR results are means ±SEM (n = 8–12 mice per treatment group). Western blot
images are representative of combined liver and soleus muscle lysates from n = 10–12 mice per treatment
group. LFD: low fat diet, HFD: high fat diet, TQ: thymoquinone.
https://doi.org/10.1371/journal.pone.0185374.g010
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 13 / 19
concentration, low HDL cholesterol and elevated non-HDL cholesterol [33]. The main cause
of this phenotype is the increased free fatty acid release from insulin-resistant adipose tissue in
type 2 diabetes [45,46]. The influx of free fatty acids in the liver can promote triglyceride syn-
thesis, which increases the production of non-HDL (LDL, VLDL) cholesterol to transfer lipids
to tissues and decreases HDL cholesterol levels, which transfers lipids back to liver for degrada-
tion [33]. Indeed, our data demonstrating that TQ treatment decreased serum LDL/VLD levels
(while not affecting HDL levels) and tissue level of triglycerides (Figs 4and 5) in HFD mice
are consistent with TQ antidiabetic action and effect on lipid homeostasis. TQ-dependent
decrease in triglyceride and LDL/VLDL levels correlated with improved insulin signaling and
insulin sensitivity judged by enhanced phosphorylation of Akt (Fig 8). Akt activation is consis-
tent with the observed improvement in insulin sensitivity seen with the insulin tolerance test
(Fig 2B). These results are in accordance with our previously reported in vitro results [11] that
TQ increases glucose and fatty acid oxidation, which can lead to enhanced fuel oxidation by
peripheral tissues, weight loss and increased insulin sensitivity.
In addition to serving as a regulator of metabolic flux and substrate for metabolic processes,
NAD
+
can activate sirtuin 1 (SIRT-1) and consequently SIRT-1-dependent pathways [15].
SIRT-1 is a class III histone deacetylase, where NAD
+
functions as a substrate for SIRT-1 dea-
cetylation of target proteins [15]. SIRT-1 has been implicated directly in critical aspects of glu-
cose homeostasis, such as increasing insulin secretion and insulin sensitivity, and lowering the
inflammation and oxidative stress associated with diabetes and obesity [15,47–49]. Enhanced
production of NAD
+
via TQ-dependent redox cycling is consistent with increased level of
SIRT-1 phosphorylation in liver and muscle (Fig 7A–7D). It has been previously shown that
SIRT-1 can activate AMPK (AMP-activated protein kinase) by de-acetylating and activating
serine-threonine liver kinase B1 (LBK1), an upstream activator of AMPK [50]. AMPK is acti-
vated when cellular energy levels are low (e.g. high AMP/ATP ratio), and has been shown to
enhance fatty acid oxidation, glycolysis, stimulate glucose uptake in skeletal muscle, and
inhibit cholesterol synthesis [51]. We saw increased phosphorylated AMPKαprotein in the
liver of both LFD and HFD animals treated with TQ (Fig 8), suggesting that TQ can activate
AMPK-dependent pathways. Due to similarities in their action on different processes, such as
cellular metabolism and inflammation, it has been suggested that AMPK and SIRT-1 are
involved in a cycle where they regulate each other [50]. Whether TQ administration activates
AMPK indirectly via SIRT-1, or directly via alteration of parameters different from NADH/
NAD
+
ratio, warrants further investigation. To mechanistically explore whether the increase
in insulin sensitivity with TQ treatment is SIRT-1-dependent, we used the HepG2 cell line as a
model of insulin resistance. TQ treatment reversed insulin resistance after 24 hours, shown by
the increase in phosphorylated Akt (Fig 9). Pre-treatment with SIRT-1 inhibitor nicotinamide
suppressed this TQ effect on insulin signaling, suggesting that it is likely SIRT-1-dependent.
Pre-treatment with AMPKαinhibitor compound C also inhibited the effect of TQ, albeit sta-
tistically insignificant. Furthermore, there was an improvement in insulin resistance after
treatment with SIRT-1 and AMPK activators, suggesting a positive role of these pathways in
insulin signaling.
Diabetes and obesity are associated with tissue inflammation, which can exacerbate insu-
lin resistance. Adipose-derived pro-inflammatory markers such as resistin and chemokines
(MCP-1) can exacerbate insulin resistance by activating c-Jun N-terminal (JNK) kinases
and NF-κB transcription factors, which can promote serine phosphorylation (inhibition) of
insulin receptor substrate-1 (IRS-1), a key component of insulin signaling [52]. SIRT-1 has
been shown to inhibit NF-κB activity, and therefore suppress the inflammatory process
[53]. Indeed, TQ treatment decreased serum levels of the pro-inflammatory marker resistin
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 14 / 19
(Fig 3A). Lower resistin levels are consistent with observed increase in the insulin sensitivity
in HFD animals (Fig 2B). Since resistin has been shown to increase LDL levels [54], lower-
ing of this marker is also consistent with the observed decreases in serum LDL cholesterol
(Fig 5C).
Taken together, our study shows that TQ administration improves glucose tolerance and
insulin sensitivity in the diet-induced obesity (DIO) mouse model of type 2 diabetes. Further-
more, TQ treatment has the potential to ameliorate inflammation, altered lipid profile, and
weight gain associated with the diabetic and obese state. These anti-diabetic effects of TQ may
be mediated by activating SIRT-1 and AMPK pathways, as shown from this study. Our results
add to the existing evidence supporting the role of TQ as a natural therapeutic for the treat-
ment of type 2 diabetes, however, further studies are necessary to establish the potential of TQ
to treat type 2 diabetes in humans.
Supporting information
S1 Fig. Effect of TQ on expression of other SIRT proteins. Western blot images showing
protein expression of SIRT-2, SIRT-3, SIRT-5, SIRT-6, SIRT-7, and β-actin in liver. LFD: low
fat diet, HFD: high fat diet, TQ: thymoquinone.
(TIF)
S2 Fig. Effect of TQ on SIRT-1 and AMPKαactivation in HepG2 cells. HepG2 cells were
cultured in high (20 mM) glucose or in growth media containing 5.5 mM glucose for 18 hours,
starved with serum-free media for 2 hours, then pre-incubated with vehicle control (0.5%
DMSO), nicotinamide (0.5 MM), compound C (20 μM), or with nicotinamide and compound
C together for 30 mins, followed by incubation with TQ (10 μM) in the presence or absence of
nicotinamide and compound C; or with TQ, resveratrol (50 μM), or AICAR (2 mM) alone for
24 hours in 20mM glucose media. Vehicle-treated cells in 5.5 mM glucose served as control.
Insulin (100 nM) was added during the last 30 min. (A) Western blot images of p-SIRT-1,
SIRT-1, and β-actin. (C) Western blot images of p-AMPKα, AMPKα, and β-actin. (B and D)
Protein band quantification using densitometry from three independent experiments. p0.05
where () is significantly different from 20G + TQ using independent t-tests. 5.5 G: 5.5 mM
glucose, 20G: 20 mM glucose, TQ: thymoquinone, R: resveratrol, AIC: AICAR, NIC: nicotin-
amide, C: compound C.
(TIF)
S1 File. Data for Fig 1.GraphPad file with corresponding data for Fig 1.
(PZFX)
S2 File. Data for Fig 2.GraphPad file with corresponding data for Fig 2.
(PZFX)
S3 File. Data for Fig 3.GraphPad file with corresponding data for Fig 3.
(PZFX)
S4 File. Data for Fig 4.GraphPad file with corresponding data for Fig 4.
(PZFX)
S5 File. Data for Fig 5.GraphPad file with corresponding data for Fig 5.
(PZFX)
S6 File. Data for Fig 6.GraphPad file with corresponding data for Fig 6.
(PZFX)
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 15 / 19
S7 File. Data for Figs 7and 8.GraphPad file with corresponding data for Fig 7 (panels C and
D) and Fig 8 (panel B).
(PZFX)
S8 File. Data for Fig 9.GraphPad file with corresponding data for Fig 9 (panel B).
(PZFX)
S9 File. Data for Fig 10.GraphPad file with corresponding data for Fig 10 (panels A and B).
(PZF)
Acknowledgments
We are forever indebted to the intellectual input of the late Prof. M. Meow, Prof. L. Dracek,
and Prof. K. Rocket for their relentless support in the preparation of this manuscript.
Author Contributions
Conceptualization: Shpetim Karandrea, Angela L. Slitt, Emma A. Heart.
Data curation: Shpetim Karandrea, Huquan Yin.
Formal analysis: Shpetim Karandrea, Huquan Yin, Emma A. Heart.
Funding acquisition: Emma A. Heart.
Investigation: Shpetim Karandrea, Emma A. Heart.
Methodology: Shpetim Karandrea, Huquan Yin, Xiaomei Liang.
Project administration: Shpetim Karandrea, Angela L. Slitt.
Resources: Shpetim Karandrea, Emma A. Heart.
Supervision: Shpetim Karandrea, Angela L. Slitt, Emma A. Heart.
Validation: Shpetim Karandrea.
Visualization: Shpetim Karandrea.
Writing – original draft: Shpetim Karandrea.
Writing – review & editing: Shpetim Karandrea, Emma A. Heart.
References
1. Kahn SE, Cooper ME, Del Prato S (2014) Pathophysiology and treatment of type 2 diabetes: perspec-
tives on the past, present, and future. Lancet 383: 1068–1083. https://doi.org/10.1016/S0140-6736(13)
62154-6 PMID: 24315620
2. Bergman RN (2005) Minimal model: perspective from 2005. Horm Res 64: 8–15. https://doi.org/10.
1159/000089312 PMID: 16439839
3. Fu Z, Gilbert ER, Liu D (2013) Regulation of insulin synthesis and secretion and pancreatic Beta-cell
dysfunction in diabetes. Curr Diabetes Rev 9: 25–53. PMID: 22974359
4. Ali BH, Blunden G (2003) Pharmacological and toxicological properties of Nigella sativa. Phytother Res
17: 299–305. https://doi.org/10.1002/ptr.1309 PMID: 12722128
5. Le PM, Benhaddou-Andaloussi A, Elimadi A, Settaf A, Cherrah Y, Haddad PS (2004) The petroleum
ether extract of Nigella sativa exerts lipid-lowering and insulin-sensitizing actions in the rat. J Ethnophar-
macol 94: 251–259. https://doi.org/10.1016/j.jep.2004.04.030 PMID: 15325727
6. Rchid H, Chevassus H, Nmila R, Guiral C, Petit P, Chokaïri M, et al. (2004) Nigella sativa seed extracts
enhance glucose-induced insulin release from rat-isolated Langerhans islets. Fundam Clin Pharmacol
18: 525–529. https://doi.org/10.1111/j.1472-8206.2004.00275.x PMID: 15482373
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 16 / 19
7. Fararh KM, Shimizu Y, Shiina T, Nikami H, Ghanem MM, Takewaki T (2005) Thymoquinone reduces
hepatic glucose production in diabetic hamsters. Res Vet Sci 79: 219–223. https://doi.org/10.1016/j.
rvsc.2005.01.001 PMID: 16054891
8. Sankaranarayanan C, Pari L (2011) Thymoquinone ameliorates chemical induced oxidative stress and
β-cell damage in experimental hyperglycemic rats. Chem Biol Interact 190: 148–154. https://doi.org/10.
1016/j.cbi.2011.02.029 PMID: 21382363
9. Heart E, Palo M, Womack T, Smith PJ, Gray JP (2012) The level of menadione redox-cycling in pancre-
atic beta-cells is proportional to the glucose concentration: role of NADH and consequences for insulin
secretion. Toxicol Appl Pharmacol 258: 216–25. https://doi.org/10.1016/j.taap.2011.11.002 PMID:
22115979
10. Gray JP, Karandrea S, Burgos DZ, Jaiswal AA, Heart EA (2016) NAD(P)H-dependent quinone oxidore-
ductase 1 (NQO1) and cytochrome P450 oxidoreductase (CYP450OR) differentially regulate menadi-
one-mediated alterations in redox status, survival and metabolism in pancreatic β-cells. Toxicol Lett
262: 1–11. https://doi.org/10.1016/j.toxlet.2016.08.021 PMID: 27558805
11. Gray JP, Burgos DZ, Yuan T, Seeram N, Rebar R, Follmer R, et al. (2016) Thymoquinone, a bioactive
component of Nigella sativa, normalizes insulin secretion from pancreatic β-cells under glucose over-
load via regulation of malonyl-CoA. Am J Physiol Endocrinol Metab 310: E394–E404. https://doi.org/
10.1152/ajpendo.00250.2015 PMID: 26786775
12. Berger F, Ramırez-Hernandez MH, Ziegler M (2004) The new life of a centenarian: signaling functions
of NAD (P). Trends Biochem Sci 29: 111–118. https://doi.org/10.1016/j.tibs.2004.01.007 PMID:
15003268
13. Pollak N, Do
¨lle C, Ziegler M (2007) The power to reduce: pyridine nucleotides—small molecules with a
multitude of functions. Biochem J 402: 205–218. https://doi.org/10.1042/BJ20061638 PMID: 17295611
14. Wu J, Jin Z, Zheng H, Yan LJ (2016) Sources and implications of NADH/NAD+ redox imbalance in dia-
betes and its complications. Diabetes Metab Syndr Obes 9: 145–153. https://doi.org/10.2147/DMSO.
S106087 PMID: 27274295
15. Kitada M, Koya D (2013) SIRT1 in type 2 diabetes: mechanisms and therapeutic potential. Diabetes
Metab J 37: 315–325. https://doi.org/10.4093/dmj.2013.37.5.315 PMID: 24199159
16. Pari L, Sankaranarayanan C (2009) Beneficial effects of thymoquinone on hepatic key enzymes in
streptozotocin—nicotinamide induced diabetic rats. Life Sci 85: 830–834. https://doi.org/10.1016/j.lfs.
2009.10.021 PMID: 19903489
17. Al-Ali A, Alkhawajah AA, Randhawa MA, Shaikh NA (2008) Oral and intraperitoneal LD50 of thymoqui-
none, an active principle of Nigella sativa, in mice and rats. J Ayub Med Coll Abbottabad, 20: 25–27.
PMID: 19385451
18. Zhu D, Wang Y, Du Q, Liu Z, Liu X (2015) Cichoric acid reverses insulin resistance and suppresses
inflammatory responses in the glucosamine-induced HepG2 cells. J Agric Food Chem 63: 10903–
10913. https://doi.org/10.1021/acs.jafc.5b04533 PMID: 26592089
19. Zhu D, Zhang N, Zhou X, Zhang M, Liu Z, Liu X (2017) Cichoric acid regulates the hepatic glucose
homeostasis via AMPK pathway and activates the antioxidant response in high glucose-induced hepa-
tocyte injury. RSC Adv, 7: 1363–1375.
20. Neter J, Kutner MH, Nachtsheim CJ, Wasserman W (1996) In: Applied linear statistical models, edited
by Neter J. Chicago, IL: Irwin; 1992.
21. Wright SP (1992) Adjusted p-values for simultaneous inference. Biometrics 48: 1005–1013. https://doi.
org/10.2307/2532694
22. Petro AE, Cotter J, Cooper DA, Peters JC, Surwit SJ, Surwit RS (2004) Fat, carbohydrate, and calories
in the development of diabetes and obesity in the C57BL/6J mouse. Metabolism 53: 454–457. PMID:
15045691
23. Srinivasan K, Ramarao P (2007) Animal models in type 2 diabetes research: an overview. Indian J Med
Res 125: 451–472. PMID: 17496368
24. Dula SB, Jecmenica M, Wu R, Jahanshahi P, Verrilli GM, Carter JD, et al. (2010) Evidence that low-
grade systemic inflammation can induce islet dysfunction as measured by impaired calcium handling.
Cell Calcium 48: 133–142. https://doi.org/10.1016/j.ceca.2010.07.007 PMID: 20800281
25. Lumeng CN, Saltiel AR (2011) Inflammatory links between obesity and metabolic disease. J Clin Invest
121: 2111–2117. https://doi.org/10.1172/JCI57132 PMID: 21633179
26. Steppan CM, Bailey ST, Bhat S, Brown EJ, Banerjee RR, Wright CM, et al. (2001) The hormone resistin
links obesity to diabetes. Nature 409: 307–312. https://doi.org/10.1038/35053000 PMID: 11201732
27. Rodrı
´guez IM, Garcı
´a JO, Sa
´nchez JJA, Gonza
´lez DA, Coello SD, Dı
´az BB, et al. (2016) Lipid and
inflammatory biomarker profiles in early insulin resistance. Acta Diabetol 1–9. https://doi.org/10.1007/
s00592-016-0885-6 PMID: 27432443
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 17 / 19
28. Santilli F, Liani R, Di Fulvio P, Formoso G, Simeone P, Tripaldi R, et al. (2016) Increased circulating
resistin is associated with insulin resistance, oxidative stress and platelet activation in type 2 diabetes
mellitus. Thromb Haemost 116: 1089–1099. https://doi.org/10.1160/TH16-06-0471 PMID: 27709225
29. Tateya S, Tamori Y, Kawaguchi T, Kanda H, Kasuga M (2010) An increase in the circulating concentra-
tion of monocyte chemoattractant protein-1 elicits systemic insulin resistance irrespective of adipose tis-
sue inflammation in mice. Endocrinology 151: 971–979. https://doi.org/10.1210/en.2009-0926 PMID:
20056828
30. Nomura S, Shouzu A, Omoto S, Nishikawa M, Fukuhara S (2000) Significance of chemokines and acti-
vated platelets in patients with diabetes. Clin Exp Immunol 121: 437–443. https://doi.org/10.1046/j.
1365-2249.2000.01324.x PMID: 10971508
31. Piemonti L, Calori G, Lattuada G, Mercalli A, Ragogna F, Garancini MP, et al. (2009) Association
between plasma monocyte chemoattractant protein-1 concentration and cardiovascular disease mortal-
ity in middle-aged diabetic and nondiabetic individuals. Diabetes Care 32: 2105–2110. https://doi.org/
10.2337/dc09-0763 PMID: 19641159
32. Zietz B, Bu¨chler C, Herfarth H, Mu¨ller-Ladner U, Spiegel D, Scho
¨lmerich J, Scha
¨ffler A (2005) Cauca-
sian patients with type 2 diabetes mellitus have elevated levels of monocyte chemoattractant protein-1
that are not influenced by the–2518 A!G promoter polymorphism. Diabetes Obes Metab 7: 570–578.
https://doi.org/10.1111/j.1463-1326.2004.00436.x PMID: 16050950
33. Mooradian AD (2009) Dyslipidemia in type 2 diabetes mellitus. Nat Clin Pract Endocrinol Metab 5: 150–
159. https://doi.org/10.1038/ncpendmet1066 PMID: 19229235
34. Michishita E, Park JY, Burneskis JM, Barrett JC, Horikawa I (2005) Evolutionarily conserved and non-
conserved cellular localizations and functions of human SIRT proteins. Mol Biol Cell 16: 4623–4635.
https://doi.org/10.1091/mbc.E05-01-0033 PMID: 16079181
35. Muoio DM, Newgard CB (2008) Molecular and metabolic mechanisms of insulin resistance and β-cell
failure in type 2 diabetes. Nat Rev Mol Cell Biol 9: 193–205. https://doi.org/10.1038/nrm2327
36. Tanti JF, Jager J (2009) Cellular mechanisms of insulin resistance: role of stress-regulated serine
kinases and insulin receptor substrates (IRS) serine phosphorylation. Curr Opin Pharmacol 9: 753–
762. https://doi.org/10.1016/j.coph.2009.07.004 PMID: 19683471
37. Dinkova-Kostova AT, Talalay P (2010) NAD(P)H:quinone acceptor oxidoreductase 1 (NQO1), a multi-
functional antioxidant enzyme and exceptionally versatile cytoprotector. Arch Biochem Biophys 501:
116–123. https://doi.org/10.1016/j.abb.2010.03.019 PMID: 20361926
38. Gaikwad A, Long DJ, Stringer JL, Jaiswal AK (2001) In vivo role of NAD(P)H: quinone oxidoreductase 1
(NQO1) in the regulation of intracellular redox state and accumulation of abdominal adipose tissue. J
Biol Chem 276: 22559–22564. https://doi.org/10.1074/jbc.M101053200 PMID: 11309386
39. Alkharfy KM, Ahmad A, Khan RM, Al-Shagha WM (2015) Pharmacokinetic plasma behaviors of intrave-
nous and oral bioavailability of thymoquinone in a rabbit model. Eur J Drug Metab Pharmacokinet 40:
319–323. https://doi.org/10.1007/s13318-014-0207-8 PMID: 24924310
40. Elmowafy M, Samy A, Raslan MA, Salama A, Said RA, Abdelaziz AE, El-Eraky W, El Awdan S, Viitala
T (2016) Enhancement of bioavailability and pharmacodynamic effects of thymoquinone via nanostruc-
tured lipid carrier (NLC) formulation. AAPS PharmSciTech, 17: 663–672. https://doi.org/10.1208/
s12249-015-0391-0 PMID: 26304932
41. Ido Y (2007) Pyridine nucleotide redox abnormalities in diabetes. Antioxid Redox Signal 9: 931–942.
https://doi.org/10.1089/ars.2007.1630 PMID: 17508915
42. Yan LJ (2014) Pathogenesis of chronic hyperglycemia: from reductive stress to oxidative stress. J Dia-
betes Res https://doi.org/10.1155/2014/137919 PMID: 25019091
43. Luo X, Li R, Yan LJ (2015) Roles of pyruvate, NADH, and mitochondrial complex I in redox balance and
imbalance in βcell function and dysfunction. J Diabetes Res. https://doi.org/10.1155/2015/512618
PMID: 26568959
44. Khader M, Bresgen N, Eckl PM (2009) In vitro toxicological properties of thymoquinone. Food Chem
Toxicol 47: 129–133. https://doi.org/10.1016/j.fct.2008.10.019 PMID: 19010375
45. Krauss RM, Siri PW (2004) Dyslipidemia in type 2 diabetes. Med Clin North Am 88: 897–909. https://
doi.org/10.1016/j.mcna.2004.04.004 PMID: 15308384
46. Taskinen MR (2003) Diabetic dyslipidaemia: from basic research to clinical practice. Diabetologia 46:
733–749. https://doi.org/10.1007/s00125-003-1111-y PMID: 12774165
47. Bordone L, Motta MC, Picard F, Robinson A, Jhala US, Apfeld J. et al. (2005) Sirt1 regulates insulin
secretion by repressing UCP2 in pancreatic βcells. PLoS Biol 4: e31. https://doi.org/10.1371/journal.
pbio.0040031 PMID: 16366736
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 18 / 19
48. Sun C, Zhang F, Ge X, Yan T, Chen X, Shi X, et al. (2007) SIRT1 improves insulin sensitivity under insu-
lin-resistant conditions by repressing PTP1B. Cell Metab 6: 307–319. https://doi.org/10.1016/j.cmet.
2007.08.014 PMID: 17908559
49. Zhang J (2007) The direct involvement of SirT1 in insulin-induced insulin receptor substrate-2 tyrosine
phosphorylation. J Biol Chem 282: 34356–34364. https://doi.org/10.1074/jbc.M706644200 PMID:
17901049
50. Ruderman NB, Xu XJ, Nelson L, Cacicedo JM, Saha AK, Lan F, et al. (2010) AMPK and SIRT1: a long-
standing partnership? Am J Physiol Endocrinol Metab 298: E751–E760. https://doi.org/10.1152/
ajpendo.00745.2009 PMID: 20103737
51. Coughlan KA, Valentine RJ, Ruderman NB, Saha AK (2014) AMPK activation: a therapeutic target for
type 2 diabetes? Diabetes Metab Syndr Obes 7: 241–253. https://doi.org/10.2147/DMSO.S43731
PMID: 25018645
52. Shoelson SE, Lee J, Goldfine AB (2006) Inflammation and insulin resistance. J Clin Invest 116: 1793–
1801. https://doi.org/10.1172/JCI29069 PMID: 16823477
53. Yoshizaki T, Schenk S, Imamura T, Babendure JL, Sonoda N, Bae EJ, et al. (2010) SIRT1 inhibits
inflammatory pathways in macrophages and modulates insulin sensitivity. Am J PhysiolEndocrinol
Metab 298: E419–E428. https://doi.org/10.1152/ajpendo.00417.2009 PMID: 19996381
54. Rashid S (2013) Mechanisms by which elevated resistin levels accelerate atherosclerotic cardiovascu-
lar disease. Rheumatol Curr Res 3: 115. https://doi.org/10.4172/2161-1149.1000115
TQ improves diabetes in DIO mice by activating SIRT-1
PLOS ONE | https://doi.org/10.1371/journal.pone.0185374 September 26, 2017 19 / 19
