Cardiovasc Toxicol
DOI 10.1007/s12012-015-9329-8
Biomarkers Associated with Ischemic Stroke in Diabetes Mellitus
Patients
Shuisheng Yang1 • Jingfeng Zhao2 • Yuxiang Chen2 • Minxiang Lei1
Ó Springer Science+Business Media New York 2015
Abstract Diabetes is an established risk factor for
ischemic stroke, but the associated molecular mechanisms
remain to be fully elucidated. This study investigated the
role of plasma and platelet microRNAs and their targeting
proteins in the activation of platelets and their association
with the occurrence of ischemic stroke in patients with type
2 diabetes mellitus (T2DM). Results showed that the
expressions of platelet and plasma miR-144 and miR-223
were significantly altered in T2DM patients with or with-
out ischemic stroke compared to that in healthy controls,
but these changes were more significant in T2DM patients
with ischemic stroke. The expressions of P2Y12 and IRS-1
as well as phosphorylation levels of IRS-1, PI3K, and Akt
in platelets were significantly altered in T2DM patients
with or without ischemic stroke. The expression of platelet
miR-144 and miR-223 significantly correlated with their
plasma levels, P2Y12 and IRS-1 expression, blood glucose
concentration, and platelet activation rate. High glucose
concentration significantly elevated P-selectin, miR-144
and P2Y12 expression and significantly reduced miR-223
and IRS-1 expression in UT-7 cells. Overexpression of
miR-223 and blocking of miR-144 expression significantly
normalized the effects of high glucose concentration in
& Yuxiang Chen
chenyx008@aliyun.com
& Minxiang Lei
leimingxiang@yahoo.com
1
Department of Endocrinology, Xiangya Hospital, Central
South University, Xiangya Road 87, Changsha 410008,
People’s Republic of China
2 antiplatelet therapy has been revealed to be effective in
Hepatobiliary Enteric Surgery Research Center, Xiangya
Hospital, Central South University,
Changsha 410008, Hunan, People’s Republic of China
UT-7 cells. In conclusion, hyperglycemia may activate
platelets through miR-144 and miR-223 to downregulate
IRS-1 and upregulate P2Y12 expression in the platelets of
T2DM patients through an IRS-1–PI3K–Akt signaling.
Low platelet and plasma miR-223 expression in addition to
high platelet and plasma miR-144 expression are risk fac-
tors for ischemic stroke in T2DM patients.
Keywords Ischemic stroke  Diabetes mellitus  Platelet 
miR-144  miR-223
Introduction
Type 2 diabetes mellitus (T2DM) patients are at 2–8 times
higher risk of developing cardiovascular disease than non-
diabetic individuals [1]. An alarming 80 % of the cases of
death in diabetes mellitus are caused by arterial thrombo-
sis-related complications [2]. Diabetes is an established
risk factor for stroke, and one of the earliest and most
influential studies on diabetes demonstrated that the rela-
tive risk of ischemic stroke is 2.45 times higher in diabetic
patients than in non-diabetic individuals [3]. The primary
causes of ischemic stroke in diabetes mellitus patients are
cerebrovascular atherosclerosis and thrombosis [4]. A
variety of mechanisms associated with atherosclerosis,
such as hyperglycemia, impaired endothelial function, and
platelet activation, may be risk factors for ischemic stroke
[5]. However, the associated molecular mechanisms have
yet to be fully elucidated.
Platelet activation and aggregation are thought to be
central to the pathophysiology of atherothrombosis, and
preventing ischemic stroke in patients with cardiovascular
diseases [6]. The initial interaction between a platelet and
123

the subendothelium causes the release of platelet agonists
such as adenosine diphosphate (ADP) [7]. ADP elicits its
effects on the platelet through the P2Y purinoceptor 1
(P2Y1) and P2Y12 receptors [8]. The P2Y12 receptor also
plays a crucial role in ADP-mediated generation of
thromboxane A2, another important platelet activator [9].
In normal platelets, insulin interferes with the P2Y12
pathway through binding and phosphorylating the insulin
receptor substrate (IRS)-1. Moreover, IRS-1 has been
demonstrated to be associated with PI3K activity, which is
necessary for insulin-induced Akt activation [10]. Also,
insulin stimulation can phosphorylate multiple tyrosine
phosphorylation sites in IRS-1, which subsequently pro-
vides docking sites for multiple SH2-containing proteins
including PI3K [11]. In patients with T2DM, the sensitivity
to insulin is lost due to a possible defect in IRS-1 [1].
However, how the expression of P2Y12 and IRS-1 is reg-
ulated at the molecular level remains to be fully elucidated.
MicroRNAs (miRNAs) are small, noncoding RNA
molecules which have been widely demonstrated to
degrade messenger RNA by binding to the 30 untranslated
region of mRNAs [12, 13]. Recent studies have observed
abnormal expression of microRNAs in human platelets,
and these miRNAs regulate genes involved in platelet
activation [14], platelet shape changes [15], and circadian
platelet reactivity [16]. A recent study in animals demon-
strated that systemic release of miR-144 is an important
protective factor in the remote ischemic preconditioning-
induced cardioprotection [17]. However, the expression of
miR-144 in patients with ischemic stroke has not been
reported. In contrast, circulating miR-223 levels has
recently been demonstrated to be a risk factor for ischemic
stroke in diabetes mellitus patients [18]. However, the
targets of miR-223 have not been identified in patients with
ischemic stroke. Importantly, previous studies demon-
strated that P2Y12 is the target of miR-223, while IRS-1 is
the target of miR-144 [19, 20].
In this study, we investigated the association of glucose
concentration, platelet activation, and changes in P2Y12,
IRS-1, miR-144, and miR-223 levels with ischemic stroke
in T2DM patients.
Materials and Methods
Participants
This study was pre-approved by the Ethics Committee for
Human Research, Central South University. Written
informed consent forms were obtained from all partici-
pants. Fifty-six patients with T2DM, 58 T2DM patients
with acute ischemic stroke, and 30 age- and gender-mat-
ched healthy controls (HC) were recruited at the
123
Cardiovasc Toxicol
Department of Endocrinology, Xiangya Hospital, from
January 2012 to December 2013. Acute ischemic stroke
was diagnosed for the first time in patients. No patient had
a history of using antiplatelet treatment. Acute ischemic
stroke was diagnosed according to a patient’s medical
history, neurological examination, magnetic resonance
imaging, angiographic findings, and laboratory examina-
tions. Acute ischemic stroke was classified by etiologic
diagnosis, and only stroke patients with large artery
atherosclerosis were selected for this study. Diabetes
mellitus was diagnosed according to World Health Orga-
nization criteria. All patients with a history of hyperten-
sion, atrial fibrillation, myocardial infarction, cancer, acute
infectious diseases, immune system disorders, blood dis-
eases, renal or liver failure, and hemorrhagic stroke or
recurrent stroke were excluded from the study.
Blood Sample Collection
Twenty milliliters of fasting blood was collected into tubes
containing 1.47 % glucose, 1.32 % sodium citrate, and
0.48 % citric acid. The platelet-rich plasma was acquired
from the blood by centrifugation at 200 g for 20 min at
room temperature (RT). The platelet-poor plasma was
acquired from the blood by centrifugation at 800 g for
8 min at RT. The cell pellet was collected, and red blood
cells were lysed using red blood cell lysis buffer (Sigma-
Aldrich, St. Louis, USA). To assess platelet activation,
2 mL of blood was collected into CTAD (a mixture of
citrate, theophylline, adenosine, and dipyridamole) tubes
containing 0.11 mmol/L sodium citrate, 15 mmol/L theo-
phylline, 3.7 mmol/L adenosine, and 0.198 mmo1/L
dipyridamole.
Blood Plasma Biochemistry Measurement
The total cholesterol, low-density lipoprotein, high-density
lipoprotein, and triglyceride levels were measured directly
with reagents from Roche Diagnostics (Indianapolis, IN,
USA) using spectrophotometric methods.
Leukocyte-Depleted Platelet Preparation
The leukocyte-free platelets were prepared using magnetic
separation columns (Miltenyi Biotec, Germany). Briefly,
platelet-rich plasma was diluted with 3 mL beads buffer
provided by the manufacturer and then incubated with
70 lL of human CD45 (cluster of differentiation 45)
antigen microbeads for 1 h at RT under gentle rotation.
The contamination of leukocytes was determined by the
transcription level of CD45 mRNA using real-time poly-
merase chain reaction (PCR). Leukocyte-depleted platelets
Cardiovasc Toxicol
with contamination lower than 1 leukocyte per 5 9 106
platelets were used for further experiments.
Platelet Activation Assay
Platelet activation was measured by flow cytometry anal-
ysis of P-selectin (CD62P) expression on the cellular sur-
face using an established protocol [21]. Specific
monoclonal anti-P-selectin antibody was purchased from
R&D systems (Minneapolis, MN, USA). Platelet activation
rate was calculated using the percentage of platelets with
positive P-selectin expression out of total platelets.
Cell Culture
Human UT-7, a cell line established from the bone marrow
of a patient with acute megakaryoblastic leukemia, was
purchased from Creative Bioarray (Shirley, NY, USA).
UT-7 cells were cultured in alpha-MEM medium contain-
ing 5 ng/mL of GM-CSF (granulocyte–macrophage col-
ony-stimulating factor) and 20 % fetal bovine serum
(FBS). Cells were maintained in a humidified incubator
with 5 % CO2 at 37 °C.
RNA Extractions and Reverse Transcription
Total RNA in leukocyte-free platelets, platelet-poor plasma,
leukocytes, and UT-7 cells was isolated using TRIzol reagent
(Invitrogen, Carlsbad, USA) by following the user manual. To
control for real-time PCR amplification of miRNA, cel-miR-
39 (Caenorhabditis elegans miRNA-39) was added during
total RNA isolation [22]. Briefly, 1000 lL of TRIzol reagent,
5 nmol of cel-miR-39 (RiboBio, Guangzhou, China), and
200 lL of leukocyte-free platelets or platelet-poor plasma
were incubated at RT for 5 min before RNA isolation. The
total RNA from leukocytes was also extracted using TRIzol
reagent according to the user manual. The concentration of
total RNA was measured using a NanoDrop spectropho-
tometer (Thermo Scientific, Utah, USA). Reverse transcrip-
tion was performed using TaqMan miRNA RT Kit (Applied
Biosystems, Foster City, USA) for miRNA expression assay.
Reverse Transcription and Real-Time Quantitative
PCR
The expression of miRNA was analyzed by real-time PCR
using a TaqManÒ miRNA assay kit (Step plus One, ABI,
USA). The primers for amplifying miR-144 and miR-223
or cel-miR-39 were designed as previously described [23].
The expression of miRNAs was analyzed using the 2DDCt
method. Spiked-in cel-miR-39 was used for internal con-
trol. The amplification of each sample was triplicated.
Western Blot
Western blot was performed as previously described [24].
The anti-P2Y12 antibody was purchased from Abcam
(Cambridge, MA, USA). The anti-IRS-1, anti-phospho-
IRS-1 Tyr895, anti-phospho-PI3K p85 Tyr458/p55
Tyr199, anti-phospho-Akt Ser473, and anti-phospho-Akt
Thr308 antibodies were purchased from Cell Signaling
Technology (Danvers, MA, USA). Briefly, cells were
homogenized in lysate buffer [1 % sodium dodecyl sul-
fate (SDS), 10 mM tris (hydroxymethyl) aminomethane
(Tris) pH 8.0, 130 mM sodium chloride (NaCl), 5 mM
ethylenediaminetetraacetate (EDTA)] containing protein
inhibitors. To detect protein expression in platelets, pla-
telets were pelleted by centrifugation at 1,500 g for
20 min at RT and then homogenized in lysis buffer.
Twenty micrograms of total protein was separated on
12 % SDS-PAGE gels and transferred onto polyvinyli-
dene difluoride membranes. After blocking with PBS
buffer containing 0.1 % Tween-20 (PBST) and 5 % non-
fat milk for 45 min, the membranes were incubated with
primary antibody at RT for 2 h. After washing with PBST,
membranes were incubated with secondary antibody for
2 h at RT. Immunoreactive proteins were detected using a
chemiluminescence reagent (Pierce, Rockford, IL, USA).
To control for loading efficiency, the blots were stripped
and re-probed with beta-actin antibody.
Statistical Analysis
Data were presented as mean ± standard error of the mean
(SEM) and analyzed with SPSS 18.0 software. Chi-square
statistics (v2 test) were used for analyzing quantitative data.
T test was used for the comparison between two groups,
and one-way ANOVA (LSD-t) was used for the compar-
ison among three groups. A P \ 0.05 was considered sta-
tistically significant.
Results
Clinical and Laboratory Characteristics
The general characteristics of the patients and HC are
presented in Table 1. No significant differences in age,
diastolic blood pressure, total cholesterol, low-density
lipoprotein, high-density lipoprotein, and triglyceride level
were observed between patients with T2DM, T2DM with
ischemic stroke, and HC. However, significantly higher
systolic blood pressure, fasting blood glucose level, and
body mass index (BMI) were observed between the two
groups of patients and the HC. A significantly higher
123
 Cardiovasc Toxicol

Table 1 Clinical and

HC (n = 30) DM (n = 58) DM with IS (n = 56)
laboratory data of patients and
control subjects Mean age (years) 48.2 ± 8.9 49.8 ± 9.1 53.3 ± 9.94
SBP(mmHg) 121.8 ± 6.9 129.7 ± 7.1* 134.1 ± 6.5*
DBP(mmHg) 75.13 ± 6.6 78.29 ± 7.6 79.5 ± 9.9
FBG(mmol/L) 4.8 ± 0.47 7.9 ± 0.91* 8.9 ± 1.2*
TG(mmol/L) 1.2 ± 0.19 1.8 ± 0.7 1.9 ± 0.9
TC(mmol/L) 4.2 ± 0.75 4.6 ± 1.1 4.7 ± 0.7
HDL(mmol/L) 1.41 ± 0.14 1.48 ± 0.49 1.52 ± 0.6
LDL(mmol/L) 2.18 ± 0.62 2.24 ± 0.72 2.26 ± 0.4
Current smoking (%) 27.2 % 28.8 % 33.5 %*
Current drinking (%) 13.4 % 14.5 % 17.2 %*
BMI (kg/m2) 24.4 ± 3.2 30.2 ± 3.7* 33.7 ± 4.1*
SBP systolic blood pressure, DBP diastolic blood pressure, FBG fasting blood glucose, TG triglyceride, TC
total cholesterol, HDL high-density lipoprotein, LDL low-density lipoprotein, BMI body mass index, DM
with IS patients with diabetes mellitus and ischemic stroke
#
* P \ 0.05 compared with the controls; P \ 0.05 compared to DM (diabetes mellitus only) group

percentage of T2DM patients with ischemic stroke con- activation rate in T2DM patients with ischemic stroke was
sumed alcohol and smoked compared to patients with significantly higher than that in patients with T2DM only
T2DM only. (P \ 0.01) (Fig. 1).

Platelets Activation Rates

Platelet P2Y12 and Insulin Receptor Substrate-1
(IRS-1) Protein Expression as well as IRS-1, PI3K,
The platelet activation rates in T2DM patients with
and Akt Phosphorylation
ischemic stroke (29.89 ± 2.9 %) and patients with T2DM
only (21.5 ± 1.9 %) were significantly higher than that in
The protein levels of P2Y12 and IRS-1 as well as the
the HC (13.9 ± 1.5 %) (P \ 0.001). The platelet
phosphorylation levels of IRS-1 tyrosine 895, PI3K p85
tyrosine 458, PI3K p55 tyrosine 199, Akt serine 473, and
Akt threonine 308 in platelets were detected by Western
blot. The platelet P2Y12 protein levels were significantly
increased, while the platelet IRS-1 protein levels and the
levels of phospho-IRS-1 Tyr895, phospho-PI3K p85
Tyr458, and phospho-PI3K p55 Tyr199 were significantly
decreased in T2DM patients with ischemic stroke and
patients with T2DM only compared to the HC (P \ 0.01 or
P \ 0.001). The changes in these molecules were more
significant in T2DM patients with ischemic stroke than in
patients with T2DM only (P \ 0.01, or P \ 0.05). In
contrast, the levels of phospho-Akt Ser473 (P \ 0.001) and
phospho-Akt Thr308 (P \ 0.05) were significantly
decreased in T2DM patients with ischemic stroke, but not
in patients with T2DM only compared to the HC (Fig. 2).

The Expression of miRNAs

The expressions of miR-144 and miR-223 in plasma, pla-

Fig. 1 Comparison of activation rates of platelets. HC: healthy
controls (n = 30); DM: patients with type 2 diabetes mellitus
(n = 56); DM ? IS: type 2 diabetes patients with ischemic stroke
(n = 58). **P \ 0.001 compared with the HC; *P \ 0.01 compared
with DM
telets, and leukocytes were measured by real-time PCR.
The miR-144 levels in plasma and platelets, but not in
leukocytes, were significantly higher in T2DM patients
with ischemic stroke (P \ 0.001) and T2DM only

123
Cardiovasc Toxicol
Fig. 2 P2Y12 and IRS-1 protein expression and IRS-1, PI3K, and
Akt phosphorylation in platelets. a Representative Western blots.
b Semi-quantitative analysis of P2Y12 protein levels. c Semi-
quantitative analysis of IRS-1 protein levels. d Semi-quantitative
analysis of phospho-IRS-1 tyrosine 895 (Tyr895) levels in platelets.
e Semi-quantitative analysis of phospho-PI3K p85 tyrosine 458
(Try458) levels in platelets. f Semi-quantitative analysis of phospho-
PI3K p55 tyrosine 199 (Tyr199) levels in platelets. g Semi-quantitative
analysis of phospho-Akt serine 473 (S473) levels in platelets. h Semi-
quantitative analysis of phospho-Akt threonine 308 (T308) levels in
platelets. HC: healthy controls (n = 30); DM: patients with type 2
diabetes mellitus (n = 56); DM ? IS: type 2 diabetes patients with
ischemic stroke (n = 58). **P \ 0.001 versus the HC; *P \ 0.01
versus DM group
123
 Cardiovasc Toxicol

Fig. 3 miRNA-144 and miR-

223 expression. miR-144 levels
in platelets (a), leukocytes (c),
and plasma (e), and miR-223
levels in platelet (b), leukocytes
(d), and plasma (f) were
measured by real-time PCR.
HC: healthy controls (n = 30);
DM: patients with type 2
diabetes mellitus (n = 56);
DM ? IS: type 2 diabetes
patients with ischemic stroke
(n = 58).**P \ 0.001 versus
the HC; *P \ 0.05 or P \ 0.01
versus DM group

(P \ 0.01) than in HC (Fig. 3). In contrast, miR-223 levels
in plasma and platelets, but not in leukocytes, were sig-
nificantly lower in T2DM patients with ischemic stroke
(P \ 0.001) and patients with T2DM only (P \ 0.01) than
in HC (Fig. 3). The changes in miR-144 and miR-223
levels in plasma and platelets were more significant in
T2DM patients with ischemic stroke than in patients with
T2DM only (P \ 0.05) (Fig. 3). After considering that
smoking and drinking may have played a role in micro-
RNA regulation, we re-analyzed the miRNA expression
with exclusion of patients who consumed alcohol and
smoked from the three groups. Results showed no signifi-
cant differences in miR-144 and miR-223 expression
between analyses containing or not containing patients who
consumed alcohol and smoked (data not shown).
Correlation Between the Levels of Platelet
and Plasma miRNAs, Blood Glucose Concentration,
and Platelet Activation Rate
The data for correlation analysis are presented in Table 2.
The levels of platelet miR-144 and platelet miR-223

Table 2 Correlation analysis

Platelet miR-223 Platelet miR-144

Plasma miR-223 r = 0.932, P \ 0.0001 NA

Plasma miR-144
Leukocyte miR-223
Leukocyte miR-144
Blood glucose
Platelet activation rate
NA
r = 0.227, P [ 0.05
NA
r = -0.814, P \ 0.001
r = -0.888, P \ 0.001
r = 0.944, P \ 0.0001
NA
r = 0.305, P [ 0.05
r = 0.897, P \ 0.001
r = 0.846, P \ 0.001

123
Cardiovasc Toxicol
significantly correlated with their levels in the plasma
(P \ 0.0001). In contrast, no significant correlation was
observed between the expressions of platelet and plasma
miR-144 and miR-223 and their levels in leukocytes. The
platelet miR-144 and miR-223 levels negatively correlated
with IRS-1 and P2Y12 protein levels in platelets of all
T2DM patients, respectively. We further analyzed the
correlations between platelet miR-144 and miR-223
expressions, blood glucose concentrations, and platelet
activation rates in all patients with T2DM. Significant
correlations between platelet miR-144 and miR-223 levels
and blood glucose concentrations (P \ 0.001) and platelet
activation rates (P \ 0.001) were observed.
High Glucose Concentration Increased miR-144
and P2Y12 Expression, but Decreased miR-223
and IRS-1 Expression in Human UT-7 Cells
UT-7 cells were treated with 5 mmol/L (as a control),
10 mmol/L, and 15 mmol/L glucose for 24 h. Western blot
showed that 15 mmol/L glucose significantly increased
P2Y12 and P-selectin expression, but decreased IRS-1
protein (Fig. 4a) expression in UT-7 cells compared to
cells treated with 5 mmol/L glucose. Interestingly, treat-
ment with 15 mmol/L glucose for 24 h significantly
decreased miR-223 (Fig. 4b) level and increased miR-144
level (Fig. 4c).
The Effects of miR-144 Antisense Sequence
and miR-223 Mimics on P-selectin, P2Y12 and IRS-1
Protein Expression in High Concentration Glucose-
Treated Human UT-7 Cells
UT-7 cells were transfected with cel-miR-39 as a miRNA
control, anti-miR-144 (a mRNA with completive sequence
of matured miR-144), or miR-223 mimics (a mRNA with
Fig. 4 Gene expression in high glucose-treated UT-7 cells. UT-7
cells were treated with 5 mmol/L (as a control), 10 and 15 mmol/L
glucose (Glu) for 24 h. a Representative Western blots. Treatment
with 15 mmol/L glucose obviously increased P2Y12, P-selectin, but
same sequence of matured miR-223) using Lipofectamine
2000 by following the user manual (Invitrogen, Grand
Island, NY, USA) or without transfection. Twenty-four
hours later, the transfected or non-transfected UT-7 cells
were treated with 15 mmol/L glucose for 24 h. Western
blot showed that transfection of miR-223 mimics signifi-
cantly decreased P2Y12 and P-selectin protein levels in UT-
7 cells treated with 15 mmol/L glucose (Fig. 5a, b, d).
Transfection of anti-miR-144 significantly increased IRS-
1, but decreased P-selectin protein levels (Fig. 5a, c, d).
Transfection of miR-144 antisense sequence and miR-223
mimics normalized glucose-induced increase in P-selectin
and P2Y12 protein expression and decrease in IRS-1 pro-
tein expression in human UT-7 cells.
Discussion
In this study, significant activation of platelets was
observed in patients with type 2 diabetes compared to HC.
However, platelet activation was more significant in dia-
betes patients who also exhibited acute ischemic stroke.
The platelet activation was associated with the elevation of
P2Y12 expression and reduction in IRS-1 expression, IRS-1
tyrosine phosphorylation, PI3K tyrosine phosphorylation,
and Akt serine/threonine phosphorylation in platelets. The
elevated P2Y12 expression and reduced IRS-1 expression
in platelets correlated with low miR-223 and high miR-144
levels in plasma and platelets, but not in leukocytes, of
diabetic patients. The in vitro studies further revealed the
association between high glucose concentration and plate-
let production, P2Y12 and IRS-1 protein expression, and
changes in miR-223 and miR-144 levels. Our study sug-
gests that hyperglycemia-induced activation of platelets is
regulated by miRNAs in platelets, which may be a risk
factor for ischemic stroke in DM patients.
decreased IRS-1 protein expression in UT-7 cells compared to cell
incubated with 5 mmol/L glucose. Treatment with 15 mmol/L
glucose significantly decreased miR-223 (b) and increased miR-144
(c) levels. **P \ 0.001, *P \ 0.01 versus Glu 5 mM. N = 4
123
 Cardiovasc Toxicol

Fig. 5 The effect of

interference with miR-144 and
miR-223 expression on gene
expression. a Representative
Western blots of P2Y12, IRS-1,
and P-selectin protein levels in
UT-7 cells. Cells were
transfected with cel-miR-39
(Glu ? miR-39), miR-144
antisense RNA (Glu ? anti-
MiR-144), or miR-223 mimics
(Glu ? miR-223 m) for 24 h
and then treated with 15 mmol/
L glucose (Glu) for 24 h.
b Semi-quantitative analysis of
P2Y12 levels in a. c Semi-
quantitative analysis of IRS-1
levels in a. d Semi-quantitative
analysis of P-selectin levels in
a. **P \ 0.001 versus cells
treated with 15 mM glucose
alone. N = 4

Although the roles of P2Y12 and IRS-1 in platelet acti-
vation have been widely reported [25, 26], their expression
and the regulatory mechanisms they are associated with in
diabetes have rarely been reported. In this study, increased
P2Y12 and decreased IRS-1 expression were observed in
the platelets in both groups of diabetic patients. Moreover,
lower miR-223 and higher miR-144 expression were
observed in both plasma and platelets of diabetic patients.
The platelet miR-223 and miR-144 levels negatively cor-
related with platelet P2Y12 and IRS-1 levels, respectively.
As stated above, P2Y12 is the target of miR-223, while
IRS-1 is the target of miR-144 [19, 20]. As expected,
transfection of miR-144 antisense RNA and miR-223
mimics normalized the alterations in P2Y12, IRS-1, and
P-selectin levels in high glucose concentration-treated UT-
7 cells. In addition, pre-miRNA and miRNA processing
proteins, such as Dicer and Argonaute 2, have already been
identified in platelets [19]. These findings suggest that
miRNA level can be regulated in platelets in response to
stimulation. It has been widely demonstrated that platelets
synthesize numerous proteins from their mRNA pool [27,
28]. Therefore, it is reasonable to believe that miR-144 and
miR-223 regulates IRS-1 and P2Y12 expression in the
platelets of type 2 diabetes patients, respectively. Inter-
estingly, the changes in miR-144, miR-223, P2Y12, and
IRS-1 expression were more significant in type 2 diabetes
patients with ischemic stroke than in type 2 diabetes
patients without ischemic stroke. Thus, dysregulation of
miR-144, miR-223, P2Y12, and IRS-1 expression is
involved in the progression of type 2 diabetes. However,
how the miRNAs were regulated bidirectionally in platelets
of diabetic patients requires further studies.
Notably, our study observed significant elevation of
blood glucose in type 2 diabetes patients with and without
ischemic stroke. In these patients, the glucose concentra-
tions significantly correlated with the platelet miR-223 and
miR-144 levels, suggesting that hyperglycemia may regu-
late the expression of miRNAs. Consistent with a previous
study [29], miR-223 expression was significantly down-
regulated in megakaryocytes exposed to high glucose
concentrations. In contrast, the expression of miR-144 was
significantly upregulated in megakaryoblastic UT-7 cells.
The downregulation of miR-223 was accompanied by
upregulation of P2Y12 expression, while the upregulation
of miR-144 expression was accompanied by downregula-
tion of IRS-1 expression in high glucose-stimulated UT-7.
Moreover, blocking miR-144 activity and increasing miR-
223 levels normalized the alterations in P2Y12, IRS-1, and
P-selectin levels induced by high glucose concentrations in
UT-7 cells. These observations further suggest that high
glucose concentrations directionally regulate miR-144 and
miR-223 expression which subsequently regulates IRS-1
and P2Y12 expression, respectively. It is widely accepted
that the ADP receptor P2Y12 plays a crucial role in
thrombus formation and stabilization [14]. In normal pla-
telets, insulin interferes with ADP- and thrombin-induced
platelet functions through interference with the P2Y12-
mediated regulation of Gi pathway through

123
Cardiovasc Toxicol
phosphorylating IRS-1 receptors [30]. The IRS-1 phos-
phorylation has been associated with PI3K–Akt signaling
[12]. In this study, the levels of IRS-1 and PI3K tyrosine
phosphorylation and Akt phosphorylation were signifi-
cantly lowered in the platelets of T2DM patients with
ischemic stroke compared to HC. However, this study
provided no further evidence regarding whether the
decreased IRS-1 phosphorylation is a direct result of
decreased IRS-1 expression or a result of the decreased
PI3K–Akt activation. The UT-7 cell exhibits the typical
characteristics of megakaryocytes, which is the ability of
the cell to produce platelets. Interestingly, a significant
increase in P-selectin expression was observed in high
glucose concentration-treated UT-7 cells. Thus, this study
provided more direct evidence for the role of hyper-
glycemia in platelet activation through regulating miRNAs
expression and IRS-1–PI3K–Akt–P2Y12 signaling, which
increases the risk of ischemic stroke in patients with type 2
diabetes.
Previous studies demonstrated that plasma miRNAs can
be produced from a variety of cells, such as leukocytes,
endothelial cells, and platelets [31, 32]. This study first
demonstrated that platelet miR-223 and miR-144 levels
significantly correlated with their levels in plasma, but not
in leukocytes. This implicates that plasma miR-223 and
miR-144 may mainly originate from platelets. However,
our study provided no direct evidence. In this study, we
also observed significant elevation of systolic blood pres-
sure and body mass index in both groups of diabetic
patients. Although the percentage of patients that smoke
and drink regularly were significantly higher in patients
with type 2 diabetes and ischemic stroke, further analysis
of miR-144 and miR-223 expression with exclusion of
these drinking and smoking patients showed no significant
differences from that of drinking and smoking patients.
This suggests that smoking and drinking are risk factors for
ischemic stroke in type 2 diabetes. We also acknowledge
some limitations to this study. For example, although these
patients had no history of using antiplatelet treatments, the
potential pharmacological interactions between antidia-
betic agents and platelet function were not considered.
In conclusion, low platelet and plasma miR-223 and
high platelet and plasma miR-144 levels may be a risk
factor for ischemic stroke in T2DM. Our study also sug-
gests that hyperglycemia-induced activation of platelets
through an IRS-1–PI3K–Akt–P2Y12 pathway plays a cen-
tral role in the occurrence of ischemic stroke in T2DM
patients.
Acknowledgments This study was supported by Changsha Science
and Technology Plan (K0902033-31), Hunan Science and Technol-
ogy Plan (2009FJ3209), National Hi-tech Project of China
(2007AA021809), Key Project of International Cooperation, Ministry
of Science and Technology of China (2010DFB30300), and National
Natural Science Foundation of China (No: 81272193).
Conflict of interest The authors declared no conflict of interest.
References
1. Gerrits, A. J., Koekman, C. A., van Haeften, T. W., & Akkerman,
J. W. (2010). Platelet tissue factor synthesis in type 2 diabetic
patients is resistant to inhibition by insulin. Diabetes, 59,
1487–1495.
2. Nakagami, H., Kaneda, Y., Ogihara, T., & Morishita, R. (2005).
Endothelial dysfunction in hyperglycemia as a trigger of
atherosclerosis. Current Diabetes Review, 1, 59–63.
3. Burchfiel, C. M., Curb, J. D., Rodriguez, B. L., Abbott, R. D.,
Chiu, D., & Yano, K. (1994). Glucose intolerance and 22-year
stroke incidence. The Honolulu Heart Program. Stroke, 25,
951–957.
4. Ferreiro, J. L., Gomez-Hospital, J. A., & Angiolillo, D. J. (2010).
Platelet abnormalities in diabetes mellitus. Diabetes and Vascular
Disease Research, 7, 251–259.
5. Todd Cade, W. (2008). Diabetes-related microvascular and
macrovascular diseases in the physical therapy setting. Physical
Therapy, 88, 1322–1335.
6. Kirshner, H. S. (2009). Differentiating ischemic stroke subtypes:
Risk factors and secondary prevention. Journal of the Neuro-
logical Sciences, 279, 1–8.
7. Jackson, S. P., Nesbitt, W. S., & Kulkarni, S. (2003). Signaling
events underlying thrombus formation. Journal of Thrombosis
and Haemostasis, 1, 1602–1612.
8. Dorsam, R. T., & Kunapuli, S. P. (2004). Central role of the
P2Y12 receptor in platelet activation. Journal of Clinical Inves-
tigation, 113, 340–345.
9. Jin, J., Quinton, T. M., Zhang, J., Rittenhouse, S. E., & Kunapuli,
S. P. (2002). Adenosine diphosphate (ADP)-induced thrombox-
ane A(2) generation in human platelets requires coordinated
signaling through integrin alpha(IIb)beta(3) and ADP receptors.
Blood, 99, 193–198.
10. Brozinick, J. T, Jr, Roberts, B. R., & Dohm, G. L. (2003).
Defective signaling through Akt-2 and -3 but not Akt-1 in insulin-
resistant human skeletal muscle: Potential role in insulin resis-
tance. Diabetes, 52, 935–941.
11. Mardilovich, K., Pankratz, S. L., & Shaw, L. M. (2009).
Expression and function of the insulin receptor substrate proteins
in cancer. Cell Communication and Signaling, 7, 14.
12. Bruchova, H., Merkerova, M., & Prchal, J. T. (2008). Aberrant
expression of microRNA in polycythemia vera. Haematologica,
93, 1009–1016.
13. Bartel, D. P. (2009). MicroRNAs: target recognition and regu-
latory functions. Cell, 136, 215–233.
14. Evans, D. J., Jackman, L. E., Chamberlain, J., Crosdale, D. J.,
Judge, H. M., Jetha, K., et al. (1992). Platelet P2Y(12) receptor
influences the vessel wall response to arterial injury and throm-
bosis. Circulation, 119, 116–122.
15. Hartwig, J. H. (1992). Mechanisms of actin rearrangements
mediating platelet activation. Journal of Cell Biology, 118,
1421–1442.
16. Kondratov, R. V., Chernov, M. V., Kondratova, A. A., Gor-
bacheva, V. Y., Gudkov, A. V., & Antoch, M. P. (2003).
BMAL1-dependent circadian oscillation of nuclear CLOCK:
Posttranslational events induced by dimerization of transcrip-
tional activators of the mammalian clock system. Genes &
Development, 17, 1921–1932.
123

17. Li, J., Rohailla, S., Gelber, N., Rutka, J., Sabah, N., Gladstone, R.
A., et al. (2014). MicroRNA-144 is a circulating effector of
remote ischemic preconditioning. Basic Research in Cardiology,
109, 423.
18. Duan, X., Zhan, Q., Song, B., Zeng, S., Zhou, J., Long, Y., et al.
(2014). Detection of platelet microRNA expression in patients
with diabetes mellitus with or without ischemic stroke. Journal of
Diabetes and Its Complications, 28, 705–710.
19. Landry, P., Plante, I., Ouellet, D. L., Perron, M. P., Rousseau, G., &
Provost, P. (2009). Existence of a microRNA pathway in anucleate
platelets. Nature Structural & Molecular Biology, 16, 961–966.
20. Karolina, D. S., Armugam, A., Tavintharan, S., Wong, M. T.,
Lim, S. C., Sum, C. F., & Jeyaseelan, K. (2011). MicroRNA 144
impairs insulin signaling by inhibiting the expression of insulin
receptor substrate 1 in type 2 diabetes mellitus. PLoS One, 6,
e22839.
21. Michelson, A. D., Barnard, M. R., Krueger, L. A., Frelinger, A. L,
3rd, & Furman, M. I. (2000). Evaluation of platelet function by
flow cytometry. Methods, 21, 259–270.
22. Mitchell, P. S., Parkin, R. K., Kroh, E. M., Fritz, B. R., Wyman,
S. K., Pogosova-Agadjanyan, E. L., et al. (2008). Circulating
microRNAs as stable blood-based markers for cancer detection.
Proceedings of the National Academy of Science USA, 105,
10513–10518.
23. Bai, M., Zhu, X., Zhang, Y., Zhang, S., Zhang, L., Xue, L., et al.
(2012). Abnormal hippocampal BDNF and miR-16 expression is
associated with depression-like behaviors induced by stress dur-
ing early life. PLoS One, 7, e46921.
24. Zhang, X., Li, Y., Huang, Q., Wang, H., Yan, B., Dewhirst, M.
W., & Li, C. Y. (2003). Increased resistance of tumor cells to
hyperthermia mediated by integrin-linked kinase. Clinical Can-
cer Research, 9, 1155–1160.
123
Cardiovasc Toxicol
25. Ferreira, I. A., Mocking, A. I., Feijge, M. A., Gorter, G., van
Haeften, T. W., Heemskerk, J. W., & Akkerman, J. W. (2006).
Platelet inhibition by insulin is absent in type 2 diabetes mellitus.
Arteriosclerosis, Thrombosis, and Vascular Biology, 26, 417–422.
26. Angiolillo, D. J., Bernardo, E., Zanoni, M., Vivas, D., Capran-
zano, P., Malerba, G., et al. (2011). Impact of insulin receptor
substrate-1 genotypes on platelet reactivity and cardiovascular
outcomes in patients with type 2 diabetes mellitus and coronary
artery disease. Journal of the American College of Cardiology,
58, 30–39.
27. Dangwal, S., & Thum, T. (2012). MicroRNAs in platelet bio-
genesis and function. Thrombosis and Haemostasis, 108,
599–604.
28. Zimmerman, G. A., & Weyrich, A. S. (2008). Signal-dependent
protein synthesis by activated platelets: New pathways to altered
phenotype and function. Arteriosclerosis, Thrombosis, and Vas-
cular Biology, 28, s17–s24.
29. Zampetaki, A., Kiechl, S., Drozdov, I., Willeit, P., Mayr, U.,
Prokopi, M., et al. (2010). Plasma microRNA profiling reveals
loss of endothelial miR-126 and other microRNAs in type 2
diabetes. Circulation Research, 107, 810–817.
30. Ferreira, I. A., Mocking, A. I., Feijge, M. A., Gorter, G., van
Haeften, T. W., Heemskerk, J. W., & Akkerman, J. W. (2006).
Platelet inhibition by insulin is absent in type 2 diabetes mellitus.
Arteriosclerosis, Thrombosis, and Vascular Biology, 26, 417–422.
31. Shantsila, E., Kamphuisen, P. W., & Lip, G. Y. (2010). Circu-
lating microparticles in cardiovascular disease: Implications for
atherogenesis and atherothrombosis. Journal of Thrombosis and
Haemostasis, 8, 2358–2368.
32. Diehl, P., Fricke, A., Sander, L., Stamm, J., Bassler, N., Htun, N.,
et al. (2012). Microparticles: Major transport vehicles for distinct
microRNAs in circulation. Cardiovascular Research, 93, 633–644.