Key words miR-21 - insulin resistance - PTEN - AKT
Introduction
miRNAs are a novel class of highly conserved, non-coding small RNAs that negatively
regulate gene expression via degradation or translational repression of their target
mRNAs [1 ]
[2 ]. It has been proved that miRNAs play important roles in energy homeostasis [3 ], sugar and lipid metabolism [4 ], insulin secretion [5 ], pancreatic β-cell development [6 ], and adipocyte differentiation [7 ]. Recent · ndings showed that some miRNA played a role in the formation of IR. For
instance, miR-320 was up-regulated in IR adipocytes, and antisense oligonucleotides
against miR-320 ameliorated IR [8 ]. Over-expression of miR-29 aggravated IR [9 ]. However, it is unknown whether miR-21 plays a role in the development of IR. Recently,
Kim et al. reported miR-21 regulated adipogenic differentiation through the modulation
of TGF-β signaling[10 ], Dey et al. identified miR-21 as the molecular link between high glucose and phosphatase
tensin homologue (PTEN) suppression [11 ], and Zhang et al. found miR-21 protected from mesangial cell proliferation induced
by diabetic nephropathy in db/db mice via its target gene PTEN [12 ]. PTEN is a dual-function lipid and phosphoinositide phosphatase that was reported
as a direct target of miR-21 in some diseases [13 ]
[14 ]. Studies displayed PTEN negatively regulated insulin signaling through the PI3K
pathway in 3T3-L1 adipocytes [15 ]
[16 ]. Based on these observations, we infer that miR-21 can influence IR via regulation
of PTEN pathway.
In this study, we investigated the potential role and mechanism of miR-21 in the formation
of IR in 3T3-L1 adipocytes. Our data validated that miR-21 markedly down-regulated
in IR adipocytes compared with normal adipocytes. Over-expression of miR-21 significantly
increased insulin-stimulated glucose uptake in IR-adipocytes. Consistent with its function, over-expression of miR-21 decreased the expression
of PTEN, increased insulin-induced phosphorylation of Akt (Ser473) and GSK3β (Ser9),
and promoted GLUT4 translocation in IR-adipocytes. Taken together, the current result
firstly demonstrated that miR-21 improved IR in 3T3-L1 adipocytes, possibly through
modulating PTEN-AKT pathway.
Materials and Methods
Materials
3T3-L1 pre-adipocytes were purchased from the American Type Culture Collection (Manassas,
VA), 3-isobutyl-1-methylxanthine (IBMX), insulin and dexamethasone were obtained Sigma
(St. Louis, MO); Lipofectamine 2000, Trizol reagent and SYBR Green I dye were obtained
from Invitrogen (Carlsbad CA), p Silencer ™ 3.1-H1 expression vectors were purchased from Ambion (Austin, TX), polyclonal rabbit
against PTEN, IRβ, Akt, phospho-Akt (Ser473), GSK3β and phospho-GSK3β (Ser9) antibodies
were purchased from Cell Signaling Technology (Beverly, MA, USA); polyclonal goat
anti-GLUT4 and mouse monoclonal anti-β-actin were purchased from Santa Cruz Biotechnology
(Santa Cruz, CA).
Cell culture and induction of insulin resistance
3T3-L1 pre-adipocytes were propagated and induced to differentiation as described
previously [17 ]. Typically, mature adipocytes appeared within 9 days of differentiation. To induce
IR, 3T3-L1 adipocytes were preincubated for 24 h at 37 °C with DMEM (10 % FBS) containing
5 mmol/L glucose with or without 1 μmol/L insulin, or 25 mmol/L glucose with or without
1 μmol/L insulin.
MiR-21 plasmid construction and transient transfection
Oligonucleotides corresponding to the murine precursor sequence of miR-21 were introduced
into pSilencer™ 3.1-H1 vector to obtain pmiR-21 plasmid. The introduced sequences
were as follows: sense 5′-GATCCTAGCTTATCAGACT GATG TTGATTCAAGAGATCAACATCAGTCTGATAAGCTATTTTTGGAAA-3′;
and antisense 5′-AGCTTTTCCAAAAATCAACATCAGTCTGATAAGCTATCTCTTGAATAGCTTATCAGACTGATGTTGAG-3′.
pmiR-21 plasmid was confirmed by DNA sequencing. Empty plasmid pSilencer™ 3.1-H1 was
used as a negative control (pNeg). For transient transfection, mature adipocytes in
six-well plates were transfected in triplicate with pmiR-21 or pNeg plasmid using
Lipofectamine 2000 for 4 h (4 μg oligonucleotide was used in each well).
Quantitative real-time PCR (qRT-PCR) analysis for miR-21 and PTEN mRNA expression
For analysis of miR-21 and PTEN mRNA expression, the qRT-PCR was performed as described
previously [17 ]. Briefly, miRNA-enriched total RNA was extracted from cells using an RNeasy mini
kit, and miR-21 expression was determined using a TaqMan MicroRNA Assay kit (ABI,
USA) according to the manufacturer’s instruction. The highly conserved snRNA U6 was
used as an internal normalizing control. For analysis of PTEN mRNA expression, total
RNA was extracted and the mRNA was quantified using SYBR green PCR master mix and
a LightCycler Real Time PCR system (Bio-Rad, Hercules, CA, USA). The sequences of
PCR the primers were as follows: (i) PTEN forward 5′-CGGCAGCATCAAATGTTT CAG-3′ and
reverse 5′-AACTGGCAGGTAGAAG GCAACTC-3′ and (ii) β-actin forward 5′-TGTCCACCTTCCAGCAGATGT-3′
and reverse 5′-AGCTCAGTAACA GTCCGCCTAGA-3′. The relative expression ratio of miR-21
and PTEN were calculated using the 2 − ΔΔCt method [18 ].
2-Deoxyglucose transport assay
Adipocytes in six-well plates were incubated in serum-free medium for 2 h. Then, cells
were washed 3 times with Krebs-Ringer phosphate buffer (KRPB, consisting of 128 mmol/L
NaCl, 4.7 mmol/L KCl, 5 mmol/L NaH2 PO4 , 5 mmol/L Na2 HPO4 , 1.25 mmol/L MgSO4 , 1.25 mmol/L CaCl2 , pH 7.4) before being incubated for 30 min at 37 °C in the presence or absence of
100 nmol/L insulin. [3 H]-2-Deoxyglucose (50 μmol/L; 9.25 KBq) and 2-deoxyglucose (final concentration 0.1 mmol/L) were added to each well for
10 min and cells were then washed quickly in ice-cold PBS. 2-Deoxyglucose uptake was
assayed by scintillation counting. At the same time, ice-cold containing PBS 10 μmol/L
containing cytochalasin B was added to each well and the cells were washed 3 times
with ice-cold PBS for the measurement of nonspecific uptake. Specific uptake, nonspecific
uptake subtracted from total uptake, was determined.
Bioinformatics analysis
3 programs, miRanda, TargetScan and PicTar, were used to predict the targets of miR-21.
Western blot analysis of total cellular lysates
Cells were washed twice in ice-cold PBS, and lysed in a buffer containing 10 mmol/L
HEPES (pH7.9), 5 mmol/L MgCl2 , 10 mmol/L KCl and 0.5 % NP-40. Cell lysates were collected by centrifuging at 13 000 g
for 15 min at 4 °C. Protein concentrations in the cell lysates were determined by
BCA assay. Briefly, sample proteins (30–50 μg) were separated by 10 % sodium dodecyl
sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred to PVDF
membranes. The membranes were blocked for 60 min in a buffer containing 0.1 % Tween-20
and 5 % milk. Antibodies against PTEN, Akt, phospho-Akt (Ser473), IRβ, GSK3β, phospho-GSK3β
(Ser9) and GLUT4 were used to identify specific proteins, which were then visualized
by the ECL method. The intensity of a protein band of interest was quantified by densitometry.
Western blot analysis of total cellular membrane GLUT4 [19 ]
Cells were washed twice in ice-cold HES buffer (20 mmol/L HEPES, pH 7.5; 1 mmol/L
EDTA; and 250 mmol/L sucrose) and scraped in HES buffer in the presence of protease
inhibitors. Samples were homogenized (30 strokes in a glass Dounce homogenizer) at
4 °C and centrifuged at 1 000 × g for 5 min to remove unbroken cells, supernatant
was further centrifuged at 16 000 × g for 15 min to produce a crude plasma membrane
fraction. The resulting pellet was resuspended in a 1.15 mol/L sucrose cushion (HES
buffer containing 1.15 mol/L sucrose), and centrifuged at 100 000 × g in swing rotor
for 60 min. The white fluffy band at the interface was collected, diluted in HES,
and centrifuged again at 48 000 × g for 60 min, yielding a pellet of the plasma membrane.
The amount of GLUT4 in the plasma membrane was determined by immunoblotting using
anti-GLUT4 antibody.
Statistical analysis
Data are presented as the mean ± SD. 2 groups were compared by unpaired Student’s
t test and multiple groups were analyzed by one-way analysis of variance (ANOVA) followed
by Dunnett's test. P < 0.05 was considered significant.
Results
Establishment of insulin-resistant adipocytes
To evaluate the insulin-resistant 3T3-L1 adipocyte model, 2-deoxyglucose (2-DG) uptake
was measured by radioimmunoassay to determine the insulin sensitivity of the cells.
As expected after differentiation, 3T3-L1 adipocytes exhibited a marked response to
an acute maximal dose of insulin (100 nmol/L) with signi· cant increases in glucose
transport above basal under all conditions studied. When cells were preincubated for
24 h in media containing 5 mmol/L glucose (normal adipocytes), acute insulin stimulation
(100 nmol/L) induced a 4.4-fold increase in 2-DG transport. The addition of 1 μmol/L
insulin during preincubation did not signi· cantly affect either basal or insulin-stimulated
2-DG transport. Also, when cells were preincubated without insulin in 25 mmol/L glucose,
the acute insulin response of glucose transport was not signi· cantly different when
compared with cells preincubated in 5 mmol/ L glucose. However, 24 h of preincubation
in media containing high glucose in the presence of 1μmol/L insulin decreased acute
insulin-stimulated glucose transport by 41.4 %, when compared with cells preincubated
with 25 mmol/L glucose. The data indicated that treatment of 3T3-L1 adipocytes with
high glucose (25 mmol/L) and high insulin (1 μmol/L) resulted in the induction of
IR ([Fig. 1 ]).
Fig. 1 The establishment of insulin-resistant adipocytes. 3T3-L1 adipocytes were preincubated
for 24 h in media containing 5 mmol/L glucose or 25 mmol/L glucose with (+) or without
( − ) 1 μmol/L insulin, respectively. Cells were washed and acutely stimulated with
or without 100 nmol/L insulin for 30 min. 2-DG uptake was measured as described in
methods. The degree of basal glucose uptake in adipocytes treated with 5 mmol/L glucose
(normal adipocytes) was set to 100 %. Data are the mean ± SD (n = 3): **P < 0.01.
Reduced expression of miR-21 in insulin-resistant adipocytes
We have previously shown that there was a differential expression profile of miRNAs
between normal adipocytes (control group) and insulin-resistant adipocytes (IR-adipocytes)
(IR group). Among them, miR-21 was the most significantly down-regulated miRNA, which
was 16.5-fold reduction in the IR group compared with control group [8 ]. To further confirm miR-21 expression, qRT-PCR analysis was used. As shown in [Fig. 2a ], miR-21 was significantly decreased (about 5.3 folds) in the IR group compared to
control group, which was consistent with the microarray data. Together, it suggested
that miR-21 expression was inversely related to IR.
Fig. 2 Over-expression of miR-21 increased insulin-induced glucose uptake in IR-adipocytes.
Normal adipocytes (control), miR-21-transfected adipocytes (pmiR-21) and empty plasmid
pSilencer™ 3.1-H1-transfected adipocytes (pNeg) were induced to insulin-resistant
adipocytes with high glucose and high insulin for 24 h as the IR group, pmiR-21+IR
group and pNeg+IR group, respectively. a Relative expression ratio of miR-21 was measured by quantitative real-time reverse
transcription polymerase chain reaction (qRT-PCR). The expression levels of miR-21
in the Control group were considered as 1, its relative expression levels in other
groups were presented as a ratio with that of Control group. b Insulin-induced glucose uptake was determined by 2-Deoxyglucose transport assay.
The degree of basal glucose uptake in control group was set to 100 %. Data are the
mean ± SD (n = 3). **P < 0.01.
Over-expression of miR-21 increased insulin-induced glucose uptake in IR-adipocytes
Results of qRT-PCR showed that miR-21 was signi· cantly increased in the miR-21-transfected
adipocytes (pmiR-21) compared with empty plasmid pSilencer™ 3.1-H1-transfected adipocytes
(pNeg), suggesting miR-21 can be over-expressed by Lipofectamine 2000 transfection
in 3T3-L1 adipocytes ([Fig. 2a ]). To investigate the impact of miR-21 on the formation of IR, glucose uptake was
determined after pmiR-21 or pNeg were treated with high glucose and high insulin for
24 h. As shown in [Fig. 2b ], in the absence of insulin stimulation, there was no significant difference in glucose
uptake in all groups. However, following insulin stimulation, IR adipocytes exhibited
a marked reduction in insulin-induced glucose uptake (an approximate 1.7-fold reduction
in the absence or presence of pNeg treatment), which was restored (46.5 % recovery)
following the treatment of cells with pmiR-21. The data indicated that a negative
role of miR-21 in the regulation of IR, over-expression of miR-21 significantly promoted
insulin-induced glucose uptake in IR-adipocytes.
PTEN is a target of miR-21 in 3T3-L1 adipocytes
PTEN has been reported as a direct target of miR-21 in some diseases [13 ]
[14 ]. Recent studies demonstrate that nearly 25 % of miRNA target sites in the 3’UTR
are conserved in humans and mice. Analysis of human, mouse and rat PTEN 3’UTR showed
the presence of the highly conserved miR-21 recognition element ([Fig. 3a ]). Therefore we predicted PTEN might be a potential target gene of miR-21 in 3T3-L1
adipocytes. To confirm this hypothesis, we examined the mRNA and protein expression
of PTEN by qRT-PCR and Western blot analysis, respectively. Our data showed PTEN mRNA
and protein levels were significantly increased in IR adipocytes compared with normal
adipocytes, over-expression of miR-21 significantly decreased PTEN protein level,
whereas it had no significant effect on PTEN mRNA expression in IR-adipocytes. These data indicated that PTEN was a target gene
of miR-21 in 3T3-L1 adipocytes ([Fig. 3b, c ]).
Fig. 3 PTEN is a target of miR-21 in 3T3-L1 adipocytes. 3T3-L1 adipocytes of control, pmiR-21
and pNeg were induced to IR-adipocytes with high glucose and high insulin for 24 h
as the IR group, pmiR-21+IR group and pNeg+IR group, respectively. a Sequence complementarity between miR-21 and its target sites in the 3′ UTR of human
(hsa), mouse (mmu) and rat (rno) PTEN mRNAs. b PTEN protein levels were measured by Western blot analysis. Representative immunoblots
(top) and densitometric analysis (bottom) on the PTEN protein levels normalized to
internal β-actin level were reported. c PTEN mRNA expression was measured using qRT-PCR, relative mRNA levels were calculated
as relative change from the Control group level. Data are the mean ± SD (n = 3). **P < 0.01.
Over-expression of miR-21 led to increased insulin signaling in IR-adipocytes
PTEN is a negative regulator of PI3K-dependent signaling. Activation of PI3-K results
in the activation of Akt and downstream mediators involved in insulin signaling such
as GSK3. To investigate the potential involvement of PTEN in the regulation of insulin
signaling by miR-21, Akt phosphorylation at Ser474 (Akt-PSer474 ) was determined by Western blot analysis. As shown in [Fig. 4 ], IR-adipocytes exhibited a significant 54.8 % reduction in insulin-stimulated Akt-PSer474 compared with normal adipocytes. This decrement was significantly recovered (58.9 %
recovery) by treatment of IR-adipocytes with pmiR-21. To further explore whether the
Akt pathway is activated, we determined the level of phospho-GSK3β (Ser9), a substrate
of Akt kinase activity in adipocytes. There were about 1.8-fold decreases of phospho-GSK3β
protein level in IR-adipocytes compared to normal adipocytes. Furthermore, IR-adipocytes
over-expressing miR-21 resulted in an increase in insulin-induced phosphorylation
of GSK3β by ~1.6-fold, relative to untreated or pNeg-treated IR-adipocytes, whereas
the Akt protein levels remained the same. No difference was observed with PTEN upstream
signaling molecules such as IR-β ([Fig. 4 ]). Taken together, these results suggested the inhibition of PTEN expression by miR-21
led to activation of the PI3K/Akt pathway without perturbing PTEN upstream signaling
molecules.
Fig. 4 Over-expression of miR-21 led to increased Akt signaling in IR-adipocytes. 3T3-L1
adipocytes of control, pmiR-21 and pNeg were induced to IR-adipocytes with high glucose
and high insulin for 24 h as the IR group, pmiR-21+IR group and pNeg+IR group, respectively.
Cell lysates were collected and subjected to western blot analysis using antibodies
speci· c for P-Akt, P-GSK3β and IRβ a . Densitometry data of Akt phosphorylation, GSK3β phosphorylation and IRβ protein
levels in 3T3-L1 cells were showed, respectively b , c and d . Data are expressed as the mean ± SD (n = 3). **P < 0.01.
Over-expression of miR-21 promoted insulin-stimulated GLUT4 translocation in IR-adipocytes
Akt, a serine/threonine protein kinase, is one of the downstream target molecules
of PI3K and has been shown to mediate the metabolic actions of insulin. Activated
Akt plays a significant role in promoting GLUT4 translocation. To assess GLUT4 translocation,
we performed SDS-PAGE and Western blot analysis using anti-GLUT4 antibody in total
cell lysates and plasma membrane lawns ([Fig. 5 ]). We found that high glucose and high insulin decreased insulin-stimulated GLUT4
translocation to plasma membrane by 61.7 %, over-expression of miR-21 exhibited a
significant 36.9 % increase in GLUT4 protein level compared with untreated or pNeg-treated
IR-adipocytes, which agreed well with the glucose uptake results. However, over-expression
of miR-21 had no significant effect on total GLUT4 protein expression in IR-adipocytes
(data not shown). Taken together, these observations implicated that over-expression
of miR-21 reversed insulin resistance in 3T3-L1 adipocytes through PTEN-Akt pathway.
Fig. 5 Over-expression of miR-21 promoted insulin-stimulated GLUT4 translocation in IR-adipocytes.
3T3-L1 adipocytes of control, pmiR-21 and pNeg were induced to IR-adipocytes with
high glucose and high insulin for 24 h as the IR group, pmiR-21+IR group and pNeg+IR
group, respectively. The plasma membrane of 3T3-L1 cells was fractionated, and the
GLUT4 protein level was determined by Western blot analysis. Representative immunoblots
and densitometric analysis were reported. The amounts of GLUT4 were expressed as percentage
of that in the Control group. Data are expressed as the mean ± SD (n = 3). **P < 0.01.
Discussion
Insulin resistance (IR) is the pathogenic hallmark of type 2 diabetes mellitus (T2DM).
Therefore, further exploration into the molecular mechanism of IR will contribute
to the prevention and treatment for T2DM and its complications. Emerging evidence
suggests that miRNAs play an important role in diabetes and its related complications
[12 ]
[20 ]
[21 ]. Some studies showed miR-21 might be associated with IR or diabetes [10 ]
[11 ]
[12 ]. However, the role of miR-21 in the development of IR was not clear. In the current
study, we further confirmed that the expression of miR-21 was reduced, and miR-21
significantly increased insulin-stimulated glucose uptake in IR-adipocytes ([Fig. 2 ]), suggesting that miR-21 enhanced insulin sensitivity and improved IR.
What are the underlying mechanisms that miR-21 improved IR? Physiologically, insulin
signals passes through a pathway involving protein kinases including, but not limited
to, PI3K, AKT or protein kinase B (PKB), and GSK-3β (the PI3K/AKT/GSK-3β pathway)
[22 ]. Emerging evidence suggests that IR can potentially be treated via modulation of
the PI3K/AKT pathway by targeting its up- or downstream modulators [23 ]. Studies also indicated PTEN could inhibit insulin signaling and antagonized PI3K-mediated
signaling, and the inhibition of PTEN may enhance insulin signaling [15 ]
[16 ]. Our data show that PTEN may be a target of miR-21 and miR-21 negatively regulated
the process of IR by targeting PTEN . First, a search with Targetscan revealed that miR-21 is complementary to sites in
the 3′ untranslated regions (3’UTR) of PTEN ([Fig. 3a ]). Second, the induction of IR significantly increased PTEN protein expression, which
occurred concomitant with a 5.3-fold reduction in miR-21 expression. Third, the enhancement
in PTEN protein expression was recovered following treatment of IR-adipocytes with
miR-21, while PTEN mRNA expression was not changed.
AKT is a downstream signal molecule of PTEN and AKT activation is a hallmark of PTEN
loss [24 ]. To further confirm the involvement of PTEN in miR-21-mediated effects, Akt phosphorylation
was determined. Our data showed that endogenous Akt protein level did not significantly
change, while the phosphorylation level of Akt significantly decreased in the IR group
compared with control group. Over-expression of miR-21 resulted in an increase in
Akt phosphorylation and a concomitant reduction in PTEN expression ([Fig. 4 ]), Similar results were obtained concerning GSK3β (Ser9) phosphorylation level. GSK3β,
a substrate of Akt kinase activity, plays an important role in the regulation of glycogen
synthesis. Phosphorylation of Ser 9 in GSK3β leads to the inhibition of its activity,
so, miR-21 may inhibit GSK-3β through the phosphorylation. Because the limit of experiment
condition, we did not detect the PI3K activity, but we observed there was no difference
that PTEN upstream signaling molecules such as IR-β ([Fig. 4 ]). Taken together, these data demonstrated the inhibition of PTEN expression led
to activation of AKT-GSK3β signal pathway without perturbing PTEN upstream signaling
molecules.
Akt is required for GLUT4 translocation to the cell surface following insulin stimulation,
which subsequently augments glucose transport [25 ]. We found insulin-stimulated glucose uptake and GLUT4 translocation were significantly
decreased in IR-adiocytes, and miR-21 reversed these changes ([Fig. 4 ]
[5 ]). These results reinforce the notion that miR-21 improved IR by PTEN-AKT signaling,
miR-21 could be a causal factor of the down-regulation of PTEN and activation of PI3K/AKT
pathway in IR-adipocytes.
In conclusion, our study indicated that miR-21 reversed high glucose and high insulin
induced IR in 3T3-L1 adipocytes through modulating the PTEN-AKT pathway, and miR-21
might be a novel potential target for prevention and therapy of IR and other metabolic
diseases.
