α-lipoic acid inhibits high glucose-induced apoptosis in HIT-T15 cells

Original Article

a-lipoic acid inhibits high glucose-induced apoptosis inHIT-T15 cells

Yi Yang,1,2† Weiping Wang,1† Yinan Liu,1 Ting Guo,1 Ping Chen,1 Kangtao Ma1 andChunyan Zhou1*1Department of Biochemistry and Molecular Biology, School of Basic Medical Sciences, Peking University, 38 Xueyuan

Road, Beijing, 100191; and 2Department of Biochemistry and Molecular Biology, School of Basic Medical Sciences,Ningxia Medical University, 1160 Sheng Li South Road, Yinchuan, Ningxia Hui Autonomous Region, 750004, China

High blood glucose plays an important role in the pathogenesis of diabetes. a-lipoic acid (LA) has been used toprevent and treat diabetes, and is thought to act by increasing insulin sensitivity in many tissues. However,whether LA also has a cytoprotective effect on pancreatic islet beta cells remains unclear. In this study, weassessed whether LA could inhibit apoptosis in beta cells exposed to high glucose concentrations. HIT-T15pancreatic beta cells were treated with 30 mmol/L glucose in the presence or absence of 0.5 mmol/L LA for8 days. LA significantly reduced the numbers of apoptotic HIT-T15 cells and inhibited the cell overgrowth nor-mally induced by high glucose treatment. Additionally, LA inhibited insulin expression and secretion in HIT-T15cells induced by high glucose. Further study demonstrated that LA upregulated Pdx1 and Bcl2 gene expres-sion, reduced Bax gene expression, and promoted phosphorylation of Akt in HIT-T15 cells treated with highglucose. Intriguingly, knockdown of Pdx1 expression partially offset the anti-apoptotic effect of LA. However,inhibition of Akt by PI3K/AKT antagonist LY294002 only slightly reversed the anti-apoptosis effect of LA andmildly decreased the gene expression level of Pdx1 (P > 0.05). Moreover, LA only slightly attenuated reactiveoxygen species (ROS) production and augmented mitochondrial membrane potential. Therefore, our data sug-gest that a-lipoic acid can effectively attenuate high glucose-induced HIT-T15 cell apoptosis probably byincreasing Pdx1 expression. These findings provide a new interpretation on the role of LA in the treatment ofdiabetes.

Key words: a-lipoic acid, apoptosis, diabetes, islet beta cells, Pdx1.

Introduction

a-lipoic acid (LA), isolated from bovine liver in 1950, is

a naturally-occurring cofactor found in a number of

multienzyme complexes. It is a powerful antioxidant

and can scavenge reactive oxygen species (ROS) (Bal-

kis et al. 2009). LA has a protective effect on oxidative

stress-induced apoptosis in some cell types. It has

been reported to inhibit high glucose-induced apopto-

sis of human umbilical vein endothelial cells (Menget al. 2008), reduce dopaminergic neuron loss in a rat

model of Parkinson’s disease (Abdin & Sarhan 2011)

and reduce the apoptosis of hepatocytes and bone

marrow stromal cells induced by TNF-a (Byun et al.

2005; Diesel et al. 2007). This cytoprotective potential

is not only due to its antioxidative properties but also

involves the activation of specific cellular signaling

pathways (Salinthone et al. 2010). LA has also been

reported to induce apoptosis in some cancer cells at

higher concentrations (over 1–2 mmol/L). It induces

apoptosis in human hepatoma cells by activating

PTEN and inhibiting Akt (Shi et al. 2008). In lung can-cer cells, LA induces apoptosis through both caspase-

independent and caspase-dependent pathways, which

are mediated by intracellular Ca2+ (Choi et al. 2009).

These data suggest that LA can act as an anti-apop-

totic or a pro-apoptotic agent depending on its con-

centration and the cell type. Recent evidence suggests

that LA has a wide range of benefits in the treatment

of diabetes. It is thought to ameliorate insulin resis-tance in type 2 diabetes mellitus (Lee et al. 2005;

Kamenova 2006; Gupte et al. 2009; Muellenbach

*Author to whom all correspondence should be addressed.Email: [email protected]†These authors contributed equally to this work.Received 20 November 2011; revised 27 March 2012;

accepted 10 April 2012.ª 2012 The AuthorsDevelopment, Growth & Differentiation ª 2012 Japanese

Society of Developmental Biologists

Develop. Growth Differ. (2012) 54, 557–565 doi: 10.1111/j.1440-169X.2012.01356.x

The Japanese Society of Developmental Biologists

et al. 2009; Wang et al. 2010). Another importantpathogenetic mechanism in diabetes is the apoptosis

of beta cells caused by high blood glucose (Wang

et al. 2011a). The purpose of this study was to investi-

gate whether LA confers a cytoprotective effect on

beta cells in the presence of high glucose concentra-

tions, and the possible mechanisms involved in this

effect. Pancreatic duodenal homeobox 1 (Pdx1) has

been reported to play central roles in pancreatic betacell function and survival (Johnson et al. 2003; Li et al.

2005; Fujimoto & Polonsky 2009; Fujimoto et al. 2009;

Gauthier et al. 2009). Phosphatidylinositol 3-kinase

(PI3K) and its effector protein kinase B (PKB/Akt) have

also been implicated as critical mediators of pancreatic

beta cell survival (Artwohl et al. 2007; Wang et al.

2011b). We used hamster insulinoma cells (HIT-T15

cell line) to investigate whether these mechanisms areinvolved in the protection of pancreatic beta cells by

LA. Our data reveal that LA can prevent HIT-T15 cells

from undergoing apoptosis caused by long term cul-

ture in high glucose medium. The anti-apoptotic effect

might be mainly mediated by increasing expression of

Pdx1.

Materials and methods

Cell culture

Hamster pancreatic beta cells, HIT-T15 (ATCC num-

ber: CRL-1777), were cultured in RPMI-1640 medium

containing 11.5 mmol/L glucose with 10% heat-inacti-

vated fetal bovine serum (FBS), 100 U/mL penicillin

and 100 mg/mL streptomycin. Most experiments werecarried out with at least three groups: One group

remained in medium with a normal glucose concentra-

tion (11.5 mmol/L, the NG group); one group was cul-

tured in medium with high glucose concentration

(30 mmol/L, the HG group) and one group was cul-

tured in medium with high glucose concentration

(30 mmol/L) supplemented with 0.5 mmol/L lipoic acid

(the HG + LA group).

Evaluation of apoptosis

Apoptosis of HIT-T15 cells was induced by culturing

the cells in medium supplemented with 30 mmol/L

glucose with or without 0.5 mmol/L LA for up to

8 days (the HG + LA and HG groups, respectively).

Cells were washed twice with phosphate-buffered sal-ine (PBS) and fixed in 1% paraformaldehyde in PBS at

4°C for 30 min followed by 70% ethanol for 30 min.

The effects of LA were evaluated by Annexin-V and PI

staining according to the manufacturer’s protocol (Vig-

orous Biotechnology, Beijing, China). Samples were

analyzed on a Becton Dickinson FACS Calibur (SanJose, CA, USA).

Cell proliferation assay

HIT-T15 cells were seeded at a density of 1 9 103

cells/well in 96-well plates in normal glucose medium

(11.5 mmol/L glucose). After 24 h, cells were cultured

in NG, HG or HG + LA media for a further 4 days. Cellviability was measured using a CCK-8 cell proliferation

kit (Dojindo Laboratories, Kumamoto, Japan) accord-

ing to the manufacturer’s instructions. Briefly, 10 lLCCK-8 solution was added to each well at different

time-points. Plates were incubated at 37°C for 2 h

and the absorbance at 450 nm was measured with a

Microplate Reader (Bio-Rad, La Jolla, CA, USA).

Real-time RT-PCR

RNA extraction was performed using Trizol Reagent

(Invitrogen) based on the manufacturer’s instructions.

For real-time reverse transcription–polymerase chain

reaction (RT–-PCR), ABI Prism 7700 sequence detec-

tor and SYBR® Green Real-Time Master Mix (TOY-

OBO, Tokyo, Japan) were used. Primers are listed inTable 1. All annealing temperatures were 60°C. Tran-scription levels of each gene were normalized to 18S

rRNA level.

Western blotting

Western blot analysis was performed as described

previously (Guo et al. 2011). The antibodies includegoat polyclonal antibody against b-actin; mouse

monoclonal antibody against LaminB; rabbit polyclonal

antibodies against Bcl2, Bax, Pdx1 and caspase-12

(Santa Cruz), Akt and phosphorylated Akt (Cell Signal-

ling). PI3K/Akt antagonist LY294002 was obtained

from Sigma and used at 5 lmol/L for 24 h prior to

protein extraction.

Table 1. The sequences of real-time reverse transcription–poly-

merase chain reaction (RT–PCR)primers

Genes Primer Sequences (5′-3′)

Bcl2 F, ATAGCCCGTGTTTGTAATR, TTCCTGATAGGGTAGGTG

Bax F, AGAGGCAGCGGCAGTGATR, CGATCCTGGATGAAACCCT

Pdx1 F, CGCGTCCAGCTCCCTTTR, TGCCCACTGGCCTTTCC

Insulin F, AGGACCCACAAGTGGAACAACTR, CAACGCCAAGGTCTGAAGGT

ª 2012 The Authors

Development, Growth & Differentiation ª 2012 Japanese Society of Developmental Biologists

558 Y. Yang et al.

RNA interference (RNAi)

HIT-T15 cells were transfected with Pdx1-siRNA (siP-

dx1) or non-silencer siRNA (siNo), respectively. Either

siPdx1 or siNo duplexes were synthesized by Shang-hai GeneChem, China. The sequences of siPdx1 are:

sense, 5′-GAAAGAGGAAGAUAAGAAAtt-3′; antisense,

5′-UUUCUUAUCUUCCUCU UUCtt-3′. The sequences

of non-silencer siRNA are: sense, 5′-UUCUCCGAAC-GUG UCACGUtt-3′; antisense, 5′-ACGUGACACG-UUCGGAGAAtt-3′. HIT-T15 cells were cultured with

HG or HG + LA for 4 days and then siRNA was

transfected into the cells using LipofectamineRNAiMAX (Invitrogen). After 48 h the cells were

harvested for apoptosis assay, real-time RT–PCR and

western blotting analysis.

Analysis of ROS production

HIT-T15 cells (5 9 104/well in six-well plates) were cul-

tured in NG, HG or HG + LA media for 6 days. Cellswere washed with PBS and incubated with 2.5 lg/mL

dihydrorhodamine 123 (DHR123; Sigma). Fluorescence

was excited at 450–490 nm and emission was moni-

tored at 515–565 nm. Fluorescence intensities were

analyzed by recording the Rhodamine 123 relative flu-

orescence unit (RFU) by flow cytometry. The data were

collected using a FACscan fluorescence-activated cell

scanner with the data acquisition program, QCELLQuest (both from Becton Dickinson).

Mitochondrial membrane potential (ΔΨm) analysis

HIT-T15 cells (5 9 104/well in 6-well plates) were cul-

tured in NG, HG or HG + LA media for 6 days.

Freshly harvested HIT-T15 cells were washed with

PBS and incubated with rhodamine 123 (800 ng/mL,Sigma) for 30 min. DΨm was monitored by observing

RFU using a FluoroCount plate reader (Packard

Instruments) at excitation/emission wavelengths of

530/590 nm.

Insulin secretion assay

HIT-T15 cells (5 9 104/well in 6-well plates) were cul-tured in NG, HG or HG + LA media for 6 days. Cells

were washed in glucose-free Krebs-Ringer bicarbon-

ate (KRB) buffer three times and then re-incubated in

NG, HG or HG + LA media, respectively, at 37°C for

60 min. The media were collected after gentle centri-

fugation and stored at �20°C for measurement of

insulin secretion by radio immunoassay (RIA) kit (The

China Atomic Energy Diagnostics) normalized to totalprotein.

Statistical analysis

The data are expressed as mean ± standard deviation

(SD). Comparisons between groups were analyzed

using analysis of variance (ANOVA), and the Student–Newman–Kleuss method was used to estimate the

level of significance. Differences were considered to be

statistically significant at P < 0.05.

Results

LA protects HIT-T15 cells against high

glucose-induced apoptosis

The apoptosis of pancreatic islet beta cells is thought

to play an important role in the pathogenesis of diabe-

tes. Chronic exposure to high glucose levels (17–27 mmol/L for 48–72 h) may induce apoptosis of beta

cells (Wang et al. 2011a). To investigate whether LA

could protect beta cells from high glucose-induced

apoptosis, a hamster pancreatic beta cell line, HIT-T15, was cultured in the presence of 30 mmol/L of

glucose with or without 0.5 mmol/L LA from 0 to

8 days (the HG + LA and HG groups, respectively). LA

significantly reduced the number of apoptotic cells

induced by high glucose at day 4, and became more

evident from day 6 onwards (Fig. 1A). The percentage

of apoptotic cells was 38.7 ± 9% and 48.8 ± 8% in

the HG group, and 7.9 ± 2% and 9.4 ± 2% in theHG + LA group at day 6 and day 8, respectively

(n = 3; P < 0.01). LA could also inhibit the overprolifer-

ation of HIT-T15 cells induced by high glucose culture.

CCK-8 cell proliferation assay showed a time-depen-

dent elevation in cell viability from day 1 to day 4 in

the HG group. In the HG + LA group, cell growth

declined from day 2 to day 4 (n = 3, P < 0.05 com-

pared with the HG group) and was similar to growth inthe normal glucose control group (Fig. 1B). These

results demonstrate that LA can prevent the apoptosis

and overgrowth of HIT-T15 cells cultured in high

glucose medium.

LA protects HIT-T15 cells from apoptosis mainly by

upregulating Pdx1 gene expression

To explore the anti-apoptotic mechanisms of LA, we

examined expression of the Pdx1 gene that has been

reported to play important anti-apoptotic and cytopro-

tective roles in islet beta cells (Fujimoto & Polonsky

2009). RT–PCR and western blot analysis showed that

culture in high glucose media only slightly reduced the

expression of Pdx1, but that addition of LA increased

the expression of Pdx1 at both mRNA (5.12 ± 0.6-fold)and protein levels on day 6 (Fig. 2A,B, P < 0.01,

ª 2012 The Authors


LA inhibits HIT-T15 cells apoptosis 559

(A) (B)

Fig. 1. Effects of a-lipoic acid (LA) on hamster insulinoma (HIT) -T15 cell apoptosis and proliferation. (A) Cells were cultured in 30 mmol/

L glucose with or without 0.5 mmol/L LA (high glucose [HG] + LA and HG, respectively), for 8 days. The percentage of apoptotic cells

was determined by flow cytometry with Annexin-V and PI double staining. Each sample was in triplicate and the experiments were per-

formed three times. (B) Cells were cultured in 11.5 mmol/L glucose (normal glucose [NG]) or 30 mmol/L glucose with or without

0.5 mmol/L LA (HG + LA and HG, respectively), for 4 days. Cell proliferation rates were determined by CCK-8 cell proliferation assay

(n = 3). Each bar represents mean ± SD. *P < 0.05, **P < 0.01 versus HG.

(A) (B)

(C)

(D)

Fig. 2. Effects of pancreatic duodenal homeobox 1 (Pdx1) on the anti-apoptosis of a-lipoic acid (LA) in hamster insulinoma (HIT)-T15

cells. Cells were cultured in 11.5 mmol/L glucose (normal glucose [NG]) or 30 mmol/L glucose with or without 0.5 mmol/L LA (high

glucose [HG] + LA and HG, respectively), for 6 days. (A) The expression of Pdx1, Bax and Bcl2 were determined by real-time reverse

transcription–polymerase chain reaction (RT–PCR). Each bar represents mean ± SD. (n = 3) *P < 0.05, **P < 0.01. (B) The expression of

Pdx1, Bax, Bcl2 and Caspase12 were determined by western blot analysis. b-actin was used as a loading control. (C) Real-time RT–

PCR (top) and western blotting (bottom) showed the expression of Pdx1 in Pdx1 knockdown (siPdx1) HIT-T15 cells; non-silencer siRNA

(siNo) was used as a siRNA control (**P < 0.01). (D) The level of apoptosis was measured by flow cytometry analysis in Pdx1 knock-

down HIT-T15 cells on day 6. The percentage of apoptotic cells in HG + LA group compared to NG, HG, HG + LA + siPdx1 and

HG + LA + siNo groups. A representative figure is shown on the left and the statistical analysis data are shown on the right. Each bar

represents mean ± SD from three samples (**P < 0.01). The experiments were repeated three times and a representative figure is

shown.

ª 2012 The Authors


560 Y. Yang et al.

n = 3), suggesting that Pdx1 might be involved in theprotective effect of LA. To further confirm the anti-

apoptotic role of Pdx1, Pdx1-specific siRNA (si-Pdx1)

was transfected into HIT-T15 cells cultured with HG

and LA. The expression of Pdx1 at both mRNA and

protein levels were significantly reduced by Pdx1-siR-

NA (Fig. 2C). Flow cytometry indicated that knock-

down of Pdx1 could increase apoptosis of HIT-T15

cells over threefold (Fig. 2D, P < 0.01, n = 3), sug-gesting that Pdx1 can partially contribute to the anti-

apoptotic effect of LA in these cells.

We also noted expression changes in the anti-apop-

totic gene Bcl2 as well as the pro-apoptotic gene Bax.

The gene expression level of Bax was increased over

fourfold in the HG group compared with the NG con-

trol, while Bcl2 was not affected significantly. However,

LA could partially reverse the expression levels ofthese two genes (Fig. 2A,B). In contrast, the activation

of caspase-12, a representative caspase involved in

response to endoplasmic reticulum stress, was

enhanced by HG culture but not significantly affected

by treatment with LA. These results provide further

evidence that LA could play an anti-apoptosis role in

HIT-T15 cells treated with high glucose.

Since the PI3-K/Akt pathway is known to participatein anti-apoptotic signaling cascades in different cells

(Artwohl et al. 2007; Wang et al. 2011b), and Akt can

regulate Pdx1 function directly or indirectly (Fujimoto &

Polonsky 2009), we also examined whether Akt activa-

tion was involved in the protective effect of LA on high

glucose-induced apoptosis. The phosphorylation at

serine 473 of Akt was increased when HIT-T15 cells

were treated with high glucose, while Akt phosphoryla-tion was further enhanced in the HG + LA group

(Fig. 3A). However, inhibition of Akt by PI3K/AKT

antagonist LY294002 only slightly reversed the anti-

apoptosis effect of LA and mildly decreased the gene

expression level of Pdx1 (Fig. 3A,B). The results sug-

gest that LA could induce the activation of Akt, but the

Akt activation may not be the main mechanism

involved in the preventive effects of LA against highglucose-induced apoptosis in HIT-T15 cells.

LA inhibits insulin secretion of HIT-T15 cells induced

by high glucose

LA acts as an insulin sensitizer in skeletal muscle

(Gupte et al. 2009; Wang et al. 2010). We evaluated

(A) (B)

Fig. 3. Effects of activating Akt on the pancreatic duodenal homeobox 1 (Pdx1) expression and anti-apoptosis of a-lipoic acid (LA) in

hamster insulinoma (HIT)-T15 cells. Cells were cultured in 11.5 mmol/L glucose (normal glucose [NG]) or 30 mmol/L glucose with or

without 0.5 mmol/L LA (high glucose [HG] + LA and HG, respectively) for 6 days. PI3K/Akt antagonist LY294002 (5 lmol/L) was incu-

bated with cells for 24 h prior to protein extraction or apoptosis rate determination in the HG + LA + LY group. (A) Real-time reverse

transcription–polymerase chain reaction (RT–PCR) (top) and western blotting (bottom) showed the expression of Pdx1 and Akt phos-

phorylation in HIT-T15 cells (compared with HG group, **P < 0.01). (B) The level of apoptosis was measured by flow cytometry analysis

in the indicated group (compared with HG group, **P < 0.01). A representative figure is shown on the top and the statistical analysis

data are shown at the bottom.

ª 2012 The Authors



the influence of HG medium and LA on insulin secre-tion in HIT-T15 cells for up to 6 days. Insulin secretion

increased in the first 2 days in the HG group (Fig. 4A,

P < 0.01, n = 3), then became close to the levels of

NG group in days 3–6. LA inhibited insulin secretion

on days 1 and 2 compared to the HG group (Fig. 4A,

P < 0.05, n = 3). The upregulation of insulin gene

expression in the HG group was also inhibited by addi-

tion of LA (Fig. 4B, P < 0.01, n = 3). LA thus inhibitsinsulin over-secretion by HIT-T15 cells induced by high

glucose.

LA influences ROS production and mitochondrial

membrane potential slightly in HIT-T15 cells cultured

with high glucose

Hyperglycaemia has been reported to enhance reac-tive oxygen species (ROS) production through the

mitochondrial electron transport chain, which in turn

activates cellular apoptotic pathways (Piconi et al.

2006). LA is an important cofactor of mitochondrial de-

hydrogenases and has been used in the treatment ofseveral oxidative injury conditions, such as neural

degeneration in diabetes (Varkonyi & Kempler 2008).

However, in our study, ROS production and mitochon-

drial membrane potential (ΔΨm) only showed slight

changes, a 14.3% increase in ROS production and a

marginal decrease of ΔΨm, in the HG group compared

to the NG group after 6 days culture (Fig. 5A,B,

P > 0.05, n = 3). When incubated with LA, ROS levelwas reduced 26% and ΔΨm was increased 17%,

respectively, but there was no significant difference

compared with high glucose culture alone (P > 0.05).

Together, these results suggest that LA can slightly

attenuate ROS production and augment ΔΨm. The

antioxidative effect might not be excluded from LA

anti-apoptosis mechanism.

Discussion

a-lipoic acid, a potent antioxidant, has been success-

fully used as a dietary supplement to prevent and treat

(A) (B)

Fig. 4. Effects of a-lipoic acid (LA) on insulin secretion and expression in hamster insulinoma (HIT)-T15 cells. (A) Cells were cultured in

11.5 mmol/L glucose (normal glucose [NG]) or 30 mmol/L glucose with or without 0.5 mmol/L LA (high glucose [HG] + LA and HG,

respectively) for 6 days and assayed for insulin secretion by radio immunoassay (RIA) (n = 3). (B) The expression of insulin in the cells

cultured for 2 days was determined by real-time reverse transcription–polymerase chain reaction (RT–PCR). Each bar represents

mean ± SD (n = 3). *P < 0.05, **P < 0.01.

(A) (B)

Fig. 5. Effects of a-lipoic acid (LA) on reactive oxygen species (ROS) production and mitochondrial membrane potential level in hamster

insulinoma (HIT)-T15 cells. Cells were cultured in 11.5 mmol/L glucose (normal glucose [NG]) or 30 mmol/L glucose with or without

0.5 mmol/L LA (high glucose [HG] + LA and HG, respectively) for 6 days. (A) Mitochondria ROS and (B) mitochondrial membrane

potential were measured by recording the Rhodamine 123 fluorescence intensities (relative fluorescence unit) by flow cytometry. Each

bar represents mean ± SD, (n = 3).

ª 2012 The Authors


562 Y. Yang et al.

many diseases, including neurodegeneration, hepaticdisorders and diabetes (Lee et al. 2005; Diesel et al.

2007; Abdin & Sarhan 2011). The evidence suggests

that LA has a wide range of benefits in the treatment

of diabetes. LA could partly ameliorate insulin resis-

tance in type 2 diabetic rats (Lee et al. 2005), and

additional antidiabetic effects have been reported in

earlier rodent and human studies (Kamenova 2006;

Gupte et al. 2009). LA has also been shown toenhance glucose disposal in skeletal muscle, liver and

adipocytes, to suppress hepatic gluconeogenesis (Lee

et al. 2005), and even to be a direct binding activator

of the insulin receptor to prevent hepatocyte apoptosis

(Diesel et al. 2007). Pancreatic beta cell apoptosis is

thought to play an important role in the pathogenesis

of diabetes. However, whether LA could protect beta

cells from apoptosis has not been fully investigated. Inthis study, we demonstrated that LA could protect

HIT-T15 cells from high glucose-induced apoptosis. Its

cytoprotective effects are mainly mediated by upregu-

lating Pdx1 expression.

The protective action of LA is usually explained by its

antioxidative potential (Varkonyi & Kempler 2008; Abdin

& Sarhan 2011). However, in our study, ROS production

and ΔΨm did not seem to be influenced by LA treatmentsignificantly. We should point out that no significant

change of ROS production was observed between the

NG and HG groups in our study. It has been reported

that glucose toxicity could increase the formation of

ROS in a variety of cell lines including HIT-T15 (Tanaka

et al. 2002). One possible reason why we did not

observe such change in our high glucose group could

be that the glucose concentration in the control (NG)group (11.5 mmol/L) is already the super-physiological

concentration (5.6 mmol/L), which might increase the

baseline of ROS (Robertson et al. 1992). It has been

reported that LA is a powerful antioxidant (Balkis et al.

2009) and has a protective effect on oxidative stress-

induced apoptosis in a number of cell types (Lee et al.

2005; Goraca et al. 2011). In the present study, we also

observe that LA slightly attenuates ROS production(26%) and augments ΔΨm (17%), although the

difference is not significant (P > 0.05). Therefore, the

antioxidative effect might not be excluded from LA

anti-apoptosis mechanism. Since 0.5 mmol/L LA could

almost completely prevent apoptosis in our study, other

molecular mechanisms have to be considered.

Pdx1 is a transcription factor that plays a central role

in pancreatic beta cell function and survival (Fujimoto& Polonsky 2009). Pdx1 knockout increased beta cell

apoptosis (Johnson et al. 2003; Gauthier et al. 2009).

Chronic hyperglycaemia can reduce Pdx1 expression

and cause beta cell dysfunction (Fujimoto & Polonsky

2009). We also observed a slightly decreased expres-

sion of Pdx1 in HIT-T15 cells with high glucose culture.However, LA could significantly elevate expression of

Pdx1. Furthermore, knockdown of Pdx1 expression

could partially offset the anti-apoptotic effect of LA.

The expressions of Isl1 and Beta2, which have been

reported as anti-apoptosis effectors in islet cells (Guo

et al. 2011), were not influenced by LA treatment (data

not shown). This suggests that it is Pdx1, not Isl1 or

Beta2, that is involved in the cytoprotective effect ofLA.

PI3-kinase/Akt signal pathway is critical to the con-

trol of beta cell growth and survival (Wang et al.

2011b). It has been reported that the activation of the

Akt pathway could increase Pdx1 expression in beta

cells through direct and indirect pathways (by inactiva-

tion of GSK-3b and Foxo1) (Fujimoto & Polonsky

2009). In this study, we also found that Akt phosphor-ylation was enhanced in HIT-T15 cells cultured in HG

with LA at 0.5 mmol/L. However, inhibition of Akt by

PI3K/AKT antagonist LY294002 only slightly decrease

the gene expression level of Pdx1, mildly reversed the

anti-apoptosis effect of LA. Therefore, it is reasonable

to propose that LA could induce the activation of Akt,

but the Akt activation may not be the main mechanism

involved in the preventive effects of LA against highglucose-induced apoptosis in HIT-T15 cells.

It is well known that PI3K activation/Akt phosphoryla-

tion is also a downstream effect of insulin signaling. In

the present study, LA inhibited HG-induced insulin

secretion, but also increased Akt phosphorylation. This

is not controversial. It has been reported that LA could

activate Akt in certain circumstances (Lee et al. 2011;

Shay & Hagen 2009). Furthermore, the chronic activa-tion of Akt in beta cells could inhibit ERK1/2 activation,

which in turn could have adverse effects on the beta

cell function, such as downregulating insulin gene

expression (Dickson & Rhodes 2004). We propose that

the activation of Akt might be involved in the inhibited

insulin secretion in HIT-T15 cells treated with LA.

The beneficial property of LA as a beta cell anti-

apoptotic agent makes it a potentially useful agent fordiabetes treatment. However, we have shown here

that LA inhibits insulin expression and secretion, as

well HIT-T15 cells proliferation with high glucose cul-

ture. Similar findings are also reported by Targonsky

et al. They found that acute or chronic exposure to LA

can cause a reduction in insulin secretion and inhibit

cells growth in isolated rat islets and MIN6 beta cells

(Targonsky et al. 2006). Therefore, it is a coordinativeeffect of LA to inhibit HIT-T15 cells apoptosis and

insulin secretion to protect and maintain beta cell func-

tion under hyperglycemia.

In our study, we observed apoptosis rate of HIT-T15

cells was increased significantly, meanwhile, we also

ª 2012 The Authors



found cells proliferation rate was increased with highglucose culture. This phenomenon seems to conflict

but could be interpreted as a consequence of islets

cells in response to the hyperglycemia. It has been

reported that in the compensation stage of diabetes,

the numbers of beta cells will be increased and islets

hypertrophy will occur to adapt insulin over-secretion.

However, long-term or chronic elevated glucose con-

centrations will result in beta cell apoptosis, dys-function and ultimately death, a state called

decompensation during the progress of diabetes (Weir

et al. 2001).

In conclusion, our data suggest that a-lipoic acid

can effectively attenuate high glucose-induced HIT-

T15 cell apoptosis mainly by increasing Pdx1 expres-

sion.

Acknowledgments

This work was supported by the National Natural Sci-

ence Foundation of China (30470402, 81071675,

81170713, 81160103), Specialized Research Fund for

the New Teacher in Doctoral Program of Higher

Education (20070001798), Specialized Research Fund

for the Doctoral Program of Higher Education(20060001107), Science Research and International

Cooperation Project of Ningxia Hui Autonomous

Region (NXIC2011010) and the National Natural

Science Foundation of Ningxia (NZ1097). We thank Dr

Jason Wong, University of Cambridge, UK for his kind

help in the preparation of this manuscript.

Author contributions

Yi Yang, Weiping Wang and Chunyan Zhou designed

research; Yi Yang, Weiping Wang, Yinan Liu, Ting Guo

and Ping Chen performed research; Yi Yang, Weiping

Wang and Kangtao Ma analyzed data; Weiping Wang

and Chunyan Zhou wrote the paper.

Competing interests

The funders had no role in study design, data collec-

tion and analysis, decision to publish, or preparation of

the manuscript. The authors declare no conflicts of

interest.

References

Abdin, A. & Sarhan, N. 2011. Intervention of mitochondrial dys-function-oxidative stress-dependent apoptosis as a possibleneuroprotective mechanism of a-lipoic acid against rote-none-induced parkinsonism and L-dopa toxicity. Neurosci.

Res. 71, 387–395.

Artwohl, M., Muth, K., Kosulin, K., de Martin, R., Holzenbein, T.,Rainer, G., Freudenthaler, A., Huttary, N., Schmetterer, L.,Waldhausl, W. K. & Baumgartner-Parzer, S. M. 2007. R-(+)-alpha-lipoic acid inhibits endothelial cell apoptosisand proliferation: involvement of Akt and retinoblastoma pro-tein/E2F-1. Am. J. Physiol. Endocrinol. Metab. 293, E681–E689.

Balkis, B. S., Othman, F., Louis, S. R., Abu, B. M., Radzi, M.,Osman, K., Das, S. & Mohamed, J. 2009. Effect of alpha li-poic acid on oxidative stress and vascular wall of diabeticrats. Rom. J. Morphol. Embryol. 50, 23–30.

Byun, C. H., Koh, J. M., Kim, D. K., Park, S. I., Lee, K. U. &Kim, G. S. 2005. Alpha-lipoic acid inhibits TNF-alpha-induced apoptosis in human bone marrow stromal cells.J. Bone Miner. Res. 20, 1125–1130.

Choi, S. Y., Yu, J. H. & Kim, H. 2009. Mechanism of alpha-lipoicacid-induced apoptosis of lung cancer cells. Ann. N. Y.

Acad. Sci. 1171, 149–155.Dickson, L. M. & Rhodes, C. J. 2004. Pancreatic beta-cell

growth and survival in the onset of type 2 diabetes: a rolefor protein kinase B in the Akt? Am. J. Physiol. Endocrinol.

Metab. 287, E192–E198.Diesel, B., Kulhanek-Heinze, S., Holtje, M., Brandt, B., Holtje, H.

D., Vollmar, A. M. & Kiemer, A. K. 2007. Alpha-lipoic acid asa directly binding activator of the insulin receptor: protectionfrom hepatocyte apoptosis. Biochemistry 46, 2146–2155.

Fujimoto, K. & Polonsky, K. S. 2009. Pdx1 and other factors thatregulate pancreatic beta-cell survival. Diabetes Obes. Metab.

11(Suppl 4), 30–37.Fujimoto, K., Hanson, P. T., Tran, H., Ford, E. L., Han, Z., John-

son, J. D., Schmidt, R. E., Green, K. G., Wice, B. M. &Polonsky, K. S. 2009. Autophagy regulates pancreatic betacell death in response to Pdx1 deficiency and nutrient depri-vation. J. Biol. Chem. 284, 27664–27673.

Gauthier, B. R., Wiederkehr, A., Baquie, M., Dai, C., Powers, A.C., Kerr-Conte, J., Pattou, F., MacDonald, R. J., Ferrer, J. &Wollheim, C. B. 2009. PDX1 deficiency causes mitochondrialdysfunction and defective insulin secretion through TFAMsuppression. Cell Metab. 10, 110–118.

Goraca, A., Huk-Kolega, H., Piechota, A., Kleniewska, P., Ciejka,E. & Skibska, B. 2011. Lipoic acid – biological activity andtherapeutic potential. Pharmacol. Rep. 63, 849–858.

Guo, T., Wang, W. P., Zhang, H., Liu, Y. N., Chen, P., Ma, K. T.& Zhou, C. Y. 2011. ISL1 promotes pancreatic islet cell pro-liferation. PLoS ONE 6, e22387.

Gupte, A. A., Bomhoff, G. L., Morris, J. K., Gorres, B. K. & Gei-ger, P. C. 2009. Lipoic acid increases heat shock proteinexpression and inhibits stress kinase activation to improveinsulin signaling in skeletal muscle from high-fat-fed rats.J. Appl. Physiol. 106, 1425–1434.

Johnson, J. D., Ahmed, N. T., Luciani, D. S., Han, Z., Tran, H.,Fujita, J., Misler, S., Edlund, H. & Polonsky, K. S. 2003.Increased islet apoptosis in Pdx1+/� mice. J. Clin. Invest.

111, 1147–1160.Kamenova, P. 2006. Improvement of insulin sensitivity in patients

with type 2 diabetes mellitus after oral administration ofalpha-lipoic acid. Hormones (Athens) 5, 251–258.

Lee, S. J., Kim, S. H., Kang, J. G., Kim, C. S., Ihm, S. H., Choi,M. G. & Yoo, H. J. 2011. Alpha-lipoic acid inhibits endoplas-mic reticulum stress-induced cell death through PI3K/Aktsignaling pathway in FRTL5 thyroid cells. Horm. Metab. Res.

43, 445–451.Lee, W. J., Song, K. H., Koh, E. H., Won, J. C., Kim, H. S.,

Park, H. S., Kim, M. S., Kim, S. W., Lee, K. U. & Park, J. Y.

ª 2012 The Authors


564 Y. Yang et al.

2005. Alpha-lipoic acid increases insulin sensitivity by acti-vating AMPK in skeletal muscle. Biochem. Biophys. Res.

Commun. 332, 885–891.Li, Y., Cao, X. Li, X. L., Brubaker, P. L., Edlund, H. & Drucker, D.

J. 2005. beta-Cell Pdx1 expression is essential for the gluc-oregulatory, proliferative, and cytoprotective actions of gluca-gon-like peptide-1. Diabetes 54, 482–491.

Meng, X., Li, Z. M., Zhou, Y. J., Cao, Y. L. & Zhang, J. 2008.Effect of the antioxidant alpha-lipoic acid on apoptosis inhuman umbilical vein endothelial cells induced by high glu-cose. Clin. Exp. Med. 8, 43–49.

Muellenbach, E. M., Diehl, C. J., Teachey, M. K., Lindborg, K.A., Hasselwander, O., Matuschek, M. & Henriksen, E. J.2009. Metabolic interactions of AGE inhibitor pyridoxamineand antioxidant alpha-lipoic acid following 22 weeks of treat-ment in obese Zucker rats. Life Sci. 84, 563–568.

Piconi, L., Quagliaro, L., Assaloni, R., Da, R. R., Maier, A., Zuo-dar, G. & Ceriello, A. 2006. Constant and intermittent highglucose enhances endothelial cell apoptosis through mito-chondrial superoxide overproduction. Diabetes Metab. Res.

Rev. 22, 198–203.Robertson, R., Zhang, H., Pyzdrowski, K. & Walseth, T. 1992.

Preservation of insulin mRNA levels and insulin secretion inHIT cells by avoidance of chronic exposure to high glucoseconcentrations. J. Clin. Invest. 90, 320–325.

Salinthone, S., Yadav, V., Schillace, R. V., Bourdette, D. N. &Carr, D. W. 2010. Lipoic acid attenuates inflammation viacAMP and protein kinase A signaling. PLoS ONE 28, pii:e13058.

Shay, K. P. & Hagen, T. M. 2009. Age-associated impairment ofAkt phosphorylation in primary rat hepatocytes is remediatedby alpha-lipoic acid through PI3 kinase, PTEN, and PP2A.Biogerontology 10, 443–456.

Shi, D. Y., Liu, H. L., Stern, J. S., Yu, P. Z. & Liu, S. L. 2008.Alpha-lipoic acid induces apoptosis in hepatoma cells viathe PTEN/Akt pathway. FEBS Lett. 582, 1667–1671.

Tanaka, Y., Tran, P. O., Harmon, J. & Robertson, R. P. 2002. Arole for glutathione peroxidase in protecting pancreatic betacells against oxidative stress in a model of glucose toxicity.Proc. Natl Acad. Sci. USA 99, 12363–12368.

Targonsky, E., Dai, F., Koshkin, V., Karaman, G., Gyulkhanda-nyan, A., Zhang, Y., Chan, C. & Wheeler, M. 2006. Alpha-lipoic acid regulates AMP-activated protein kinase andinhibits insulin secretion from beta cells. Diabetologia 49,1587–1598.

Varkonyi, T. & Kempler, P. 2008. Diabetic neuropathy: new strat-egies for treatment. Diabetes Obes. Metab. 10, 99–108.

Wang, H. W., Mizuta, M., Saitoh, Y., Noma, K., Ueno, H. & Nak-azato, M. 2011a. Glucagon-like peptide-1 and candesartanadditively improve glucolipotoxicity in pancreatic b-cells.Metabolism 60, 1081–1089.

Wang, X., Liu, Y., Yang, Z., Zhang, Z., Zhou, W., Ye, Z., Zhang,W., Zhang, S., Yang, Z., Feng, X., Chen, F. & Hu, R. 2011b.Glucose metabolism-related protein 1 (GMRP1) regulatespancreatic beta cell proliferation and apoptosis via activationof Akt signalling pathway in rats and mice. Diabetologia 54,852–863.

Wang, Y., Li, X., Guo, Y., Chan, L. & Guan, X. 2010. alpha-Li-poic acid increases energy expenditure by enhancing adeno-sine monophosphate-activated protein kinase-peroxisomeproliferator-activated receptor-gamma coactivator-1alphasignaling in the skeletal muscle of aged mice. Metabolism

59, 967–976.Weir, G. C., Laybutt, D. R., Kaneto, H., Bonner-Weir, S. & Shar-

ma, A. 2001. Beta-cell adaptation and decompensation dur-ing the progression of diabetes. Diabetes 50, S154–S159.

ª 2012 The Authors



Documents

α-lipoic acid inhibits high glucose-induced apoptosis in HIT-T15 cells