www.elsevier.com/locate/yexmp

Experimental and Molecular Pathology 76 (2004) 57–65

Contradictory effects of short- and long-term hyperglycemias on ischemic

injury of myocardium via intracellular signaling pathway

Guang Xu, En Takashi,* Mitsuhiro Kudo, Toshiyuki Ishiwata, and Zenya Naito

Department of Pathology II, Nippon Medical School, 1-1-5 Sendagi, Bunkyo-ku, Tokyo 113-8602, Japan

Received 19 July 2003

Abstract

Although clinical diabetes mellitus is obviously a high risk factor for myocardial infarction, there is disagreement about the sensitivity of

ischemic injury of an infarcted myocardium in experimental studies. The present study evaluated the influences of different durations of

hyperglycemia on ischemic and reperfusion injuries of the myocardium, and focused on extracellular signal-regulated kinase 1/2 (ERK1/2),

which plays an important role in the intracellular signaling pathway and is reported to be associated with myocardial protection against heart

injury. Short- and long-term hyperglycemias were induced in rats by streptozotocin (STZ) injection and the rats were examined 4 (4WDM)

and 20 weeks (20WDM) after the treatment. Ischemia and reperfusion were induced by occlusion and reperfusion (I/R) of the left coronary

artery (LCA). I/R-induced infarct size was determined using triphenyltetrazolium chloride (TTC) staining. After 20 weeks of STZ treatment

(20WDM+ I/R), the infarct size in the rat heart increased by 65.2F 4.3%, whereas after 4 weeks of STZ treatment (4WDM+ I/R), the infarct

size decreased compared with the time-matched I/R group (43.1F 3.6% and 59.5F 5.6%, respectively). The number of dead myocytes

including necrotic and apoptotic cells was determined using horseradish peroxidase (HRP) and terminal deoxynucleotide nick-end labeling

(TUNEL) methods. The number of dead myocytes decreased in the 4WDM+ I/R group, while the number of dead myocytes increased

markedly in the 20WDM+ I/R group, compared with the time-matched I/R group. The increment of ERK1/2 phosphorylation in the 4WDM

group and the slight enhancement of this phosphorylation by I/R treatment were observed by western blotting. However, in the 20WDM

group, the level of ERK1/2 phosphorylation reduced by approximately 1/3 compared with the time-matched control group; moreover, I/R

treatment did not enhance the phosphorylation level. This study demonstrated that short- and long-term hyperglycemias exert opposite

influences on ischemic myocardial injury, and these contradictory influences may depend on an ERK1/2-mediated intracellular signaling

pathway.

D 2003 Elsevier Inc. All rights reserved.

Keywords: Hyperglycemia; Myocardial infarction; Sensitivity of ischemic injury; ERK1/2; Apoptosis

Introduction studies using animal models of diabetes show contradictory

Cardiomyopathy may occur in diabetic patients even in

the absence of coronary artery diseases. It is a major

complication of diabetes that correlates with high morbidity

and mortality after myocardial infarction (Hamby et al.,

1974; Fein and Sonnenblick, 1985; McGuire and Granger,

1999; George and Linda, 2001). Despite hyperglycemia,

diabetes is the primary etiologic factor in the pathogenesis

of cardiomyopathy, and a controversy has arisen concerning

the sensitivity of an ischemic heart to hyperglycemia in an

experimental study (Paulson, 1997). Many experimental

0014-4800/$ - see front matter D 2003 Elsevier Inc. All rights reserved.

doi:10.1016/j.yexmp.2003.08.003

* Corresponding author. Fax: +81-3-5685-3067.

E-mail address: y-qiao@nms.ac.jp (E. Takashi).

results such as no change, increased or decreased sensitivity

of ischemia with or without reperfusion injury (Vogel and

Apstein, 1988; Forrat et al., 1993; Liu et al., 1993). The

disparity of these findings may stem from the duration and

severity of diabetes, even in an animal model (Paulson,

1997). Therefore, we considered it necessary to clarify

ischemic injury for different durations of hyperglycemia in

the same severe diabetes model.

Since clinical evidence indicates that a diabetic condition

leads to cardiomyopathy (Hamby et al., 1974), several

studies have been performed to examine the mechanism

underlying this etiology. Biochemical and morphological

disorders in hyperglycemia were reported to induce de-

creased Na/K-ATPase activity (Katori et al., 1999), synthe-

sis of diacylglycerol with the activation of protein kinase C

G. Xu et al. / Experimental and Molecular Pathology 76 (2004) 57–6558

(PKC) (Liu et al., 1999), mitogen-activated protein kinase

(MAPK) (Awazu et al., 1999) and mitochondrial KATP

(mitoKATP) channel (Ghosh et al., 2001). ERK1/2 is an

important component of intracellular signal transduction

pathways leading to cell proliferation and protein synthesis

(Davis, 1993; Pearson et al., 2001). Recent studies have

revealed that ERK1/2 is associated with a cardiac function

and is part of a ‘‘survival’’ pathway during some steps of

conditioning stress such as ischemia (Naito et al., 2000; Yue

et al., 2000; Fryer et al., 2001). Our recent study indicated

that short- and long-term hyperglycemias in the rat heart

have different influences on ERK1/2 in the myocardium

(Naito et al., 2002). The causal relationship between dia-

betic cardiomyopathy and ERK1/2 dysfunction during is-

chemic and reperfusion injuries remains to be elucidated.

The objectives of the present study were to clarify the

influence of hyperglycemia of different duration in an

animal model and to examine the infarct size and myocar-

dial cell death, such as necrosis and apoptosis, after ische-

mia and reperfusion. Another objective was to ascertain the

alterations in ERK1/2 activation in such an injury.

Materials and methods

Animal and experimental protocols

Adult male Wistar rats (250–300 g) were randomly

assigned to different experimental groups: 4- and 20-week

streptozotocin (STZ)-treated groups (4WDM and 20WDM,

n = 5 for each group); 4- and 20-week STZ groups treated

with I/R (4WDM+ I/R and 20WDM+ I/R, n = 5 for each

group); and time-matched control groups (4WC, 20WC,

4WC+ I/R and 20WC+ I/R, n = 5 for each group). Hyper-

glycemia was induced by an intraperitoneal injection of STZ

(60 mg/kg body weight). The time-matched control groups

received the vehicle (1 ml of normal saline) only. The present

study was performed in accordance with the guidelines of the

Animal Ethics Board of Nippon Medical School, Japan.

General surgical procedures for ischemia induction and

reperfusion

The rats were anesthetized by a peritoneal injection of 30

mg/kg body weight pentobarbital. The chest was incised and

ventilated using a small Harvard Rodent ventilator (Model

683). After left thoracotomy, the left coronary artery (LCA)

was identified. A 5–0 nylon ligature was placed around the

LCA 2–3 mm away from its origin. Heparin (300 IU/kg

body weight) was then administered intravenously 10 min

before the ligature was tightened. After 30 min of LCA

occlusion, the ligature was released for 120 min and

myocardial reperfusion was visually confirmed. Some of

the STZ-treated and time-matched control groups were

sham-operated and an only suture was placed loosely

around the LCA. In each group, 100 mg/kg body weight

horseradish peroxidase (HRP, Type II, Sigma, St. Louis,

MO) was administered intravenously 10 min before killing

the animal with an overdose of pentobarbital. After all

procedures of STZ treatment and surgery, blood samples

were collected from the tail vein for the measurement of

biochemical parameters, such as levels of blood glucose,

fructosamine and thiobarbituric acid reactive substances

(TBARS). The general parameters, such as body and heart

weights, were measured before the animals were anesthe-

tized and after the hearts were removed.

Morphological examination

Triphenyltetrazolium chloride (TTC) staining

For the measurement of the infarct size, the hearts were

removed immediately and perfused retrogradely through the

aorta using the Langendorff apparatus with Krebs–Hense-

leit (KH) buffer as previously described (Takashi and

Ashraf, 2000). The hearts in each group were perfused for

1 min to wash out blood from the coronary circulation. Two

percent TTC (Sigma) in cacodylate buffer was perfused

retrogradely at 37jC for 5 min and perfusion-fixed with 4%

paraformaldehyde in 0.1 M sodium cacodylate buffer (pH

7.4) for 10 min. To define the area at risk, the LCA was

reoccluded and 0.1 mg of fluorescent zinc cadmium-sulfate

particles (Duke Scientific, Palo Alto, CA) suspended in 1 ml

of normal saline was infused into the aortic stump. The

hearts were sliced transversely into 3–4 slices (2–3 mm

thick), and the slices were immersed in the fixative for an

additional 4 h at 4jC. The 100-Am-thick sections were cut

using a Vibratome (Vibratome, St. Louis, MO) for gross

morphological inspection. The area at risk and ischemic

region in the left ventricle (LV) in the image of 100-Am-

thick sections were magnified and measured based on an

NIH image. The area at risk was calculated as % of LVarea.

The infarct size was calculated as % of area at risk.

HRP reaction

To examine the dead myocytes, the 100-Am-thick sec-

tions were washed with cacodylate buffer, then incubated

for 15 min in 100 ml of 0.05 mg/l Tris (hydroxymethyl)

aminomethane (Tris)-hydrochloride buffer (pH 7.6) contain-

ing 100 mg of 3,3V-diaminobenzidine tetrahydrochloride

(DAB, Sigma). They were then allowed to react for 30

min with 33.3 Al of 30% H2O2. The specimens were washed

with 0.1 M phosphate-buffered saline (PBS). These thick

sections were postfixed with 2% buffered osmium tetroxide

for 1 h, dehydrated with a graded ethanol series and

embedded in Epon 812. A semiquantitative estimate of

dead myocytes defined as HRP-positive areas (% of TTC-

negative area) was carried out on 2-Am-thick sections

(Takashi and Ashraf, 2000).

Terminal deoxynucleotide nick-end labeling (TUNEL) assay

The TUNEL method (MEBSTAIN Apoptosis kit II,

MBL, Nagoya, Japan) was employed. The slices were

G. Xu et al. / Experimental and Molecular Pathology 76 (2004) 57–65 59

embedded in the OCTcompound and rapidly frozen in liquid

nitrogen. Eight-Am-thick sections were washed in distilled

water after fixation. Terminal deoxynucleotidyl transferase

(TdT) buffer was applied directly on the specimens, which

were then placed in a humidified chamber for 10 min. The

specimens were then treated for DNA nick-end labeling

using biotinylated dUTP mediated by TdT at 37jC for 1

h in a humidified chamber. The sections were washed in PBS

and placed in a blocking solution for 10 min. Fluorescence

staining was carried out with avidin-conjugated fluorescein

5-isothiocyanate (FITC) at 37jC for 30 min and the sections

were then stained with propidium iodide (PI) for 2 min at 4

jC to visualize the nuclei. The specimens were examined

under a confocal laser scanning microscope (CLSM, TCS-

SP; Leica Lasertechnik, Heidelberg, Germany) based on an

upright microscope (DMRB; Leica Lasertechnik) equipped

with krypton and argon laser sources. Green FITC emission

covered a wavelength range from 500 to 550 nm, whereas

red PI emission was selected and recorded using a 560–600

nm wavelength range. The end field was also observed using

Nomarski optics. A semiquantitative estimate of apoptotic

cells defined as TUNEL-positive cells (% of PI-positive

nuclei in the TTC-negative area) was carried out.

Electron microscopy observation

The heart samples were cut into 1-mm3-thick slices that

were then fixed with 2.5% glutaraldehyde, followed with

2% buffered osmium tetroxide. The sections were dehy-

drated with a graded ethanol series and embedded in Epon

812. Ultrathin sections of the samples were cut using an

ultramicrotome (Dupont, Newtown, CT) and stained with

uranyl acetate and lead citrate. Sections were observed

under a Hitachi H-7000 electron microscope (Hitachi,

Tokyo, Japan).

Western blotting and analysis of total and phosphorylated

ERK1/2

Total protein was extracted from the rat heart tissue of

each group as previously described (Yue et al., 2000). The

heart tissues were homogenized on ice in a homogenate

buffer (20 mM Tris–HCl (pH 7.4), 150 mM NaCl, 2 mM

Table 1

Characteristics of the experimental model

Groups n Weight B

Body (g) Heart (g) G

4WC 5 362.2F 12.2 0.67F 0.03 1

4WDM 5 258.8F 10.8 0.60F 0.03 5

20WC 5 487.0F 19.5 1.14F 0.05 1

20WDM 5 213.0F 8.5y 0.57F 0.02y 5

Values are meanF S.E.M. W: Duration (weeks), DM: STZ-induced diabetic, C:y p< 0.05 vs. time-matched groups.z p< 0.001 vs. time-matched groups.

EDTA, 1% NP-40, 0.1% SDS, 1% DOC, 50 mM NaF, 1

mM Pefablock SC, 1 mg/ml pepstatin A, 10 Ag/ml leupep-

tion, 10 Ag/ml aprotinin, 1 mM Na3VO4 and 50 mM NaF).

The total protein concentrations in all the samples were

measured using a BCA protein assay kit (Pierce, Rockford,

IL, USA). The samples (20 Ag/lane) were subjected to 10%

sodium dodecyl sulfate-polyacrylamide gel electrophoresis

(SDS-PAGE), and the fractionated products were electro-

phoretically transferred onto an immobilon PVDF mem-

brane (Millipore, Bedford, MA, USA). The membrane was

incubated for 2 h with 5% skim milk in Tris-buffered saline

(TBS) (0.2 M Tris–HCl (pH 7.5) and 0.15 M NaCl)

containing 0.05% Tween 20 (TBST) in order to block

nonspecific reactions, and then reacted with the following:

(1) mouse monoclonal anti-phosphorylated ERK1/2 anti-

body at 1: 3000 dilution and (2) rabbit polyclonal anti-total

ERK1/2 antibody at 1:3000 dilution for 2 h. The membranes

were then sequentially washed and incubated with

corresponding secondary antibodies conjugated to horserad-

ish peroxidase for 1 h. After washing, the immunoreactions

were visualized by enhanced chemiluminescence (Pierce,

Rockforkford, IL, USA). An optical scanner (Epson, GT-

8700F) was used for the digitization of Western blotting

results to measure the level of proteins. The densitometry of

each band with digitized images in the same gel was carried

out using the public domain NIH image program.

Statistical analyses

All values were expressed as meansF S.E.M. Statistical

comparisons were performed using one-way ANOVA and

Student’s t-test. Differences were considered significant at

P < 0.05.

Results

Animal data

Table 1 summarizes the animal data including body and

heart weights, and the levels of blood glucose, fructosamine

and TBARS in each group. In the 4WDM and 20WDM

groups, the blood glucose, fructosamine and TBARS levels

lood

lucose (mg/dl) Fructosamine (AM/l) TBARS (nM/ml)

32.4F 7.9 138.0F 7.6 1.64F 0.14

59.6F 32.6z 242.6F 12.9z 3.22F 0.29y

61.0F 9.3 160.8F 8.5 1.82F 0.24

47.0F 39.3z 279.2F 16.1z 5.24F 0.46z

Without STZ treatment.

Fig. 1. TTC stain showing the infarct size. A clear infarct zone devoid of TTC stain is observed in 4- and 20-week STZ-treated rat hearts. The infarct size

significantly decreased in 4-week STZ-treated rat hearts (A) compared with 20-week STZ-treated rat hearts (B). C and D are magnified light photomicrographs

of the central infarct area that is marked by squares in A and B. HRP reaction is observed in 4- and 20-week STZ-treated rat hearts. The dead myocytes were

HRP-positive (indicated by arrows). In 4-week STZ-treated group, a few remaining myocytes without HRP reaction (*) were observed within the ischemic area

(C). In 20-week STZ-treated group (D), the number of HRP-positive myocytes significantly increased. Magnification: C and D, �300.

G. Xu et al. / Experimental and Molecular Pathology 76 (2004) 57–6560

increased. Whereas, the body and heart weights decreased.

The blood glucose levels in 4WDM and 20WDM groups

were 559.6F 32.6 and 547F 39.3 mg/dl, respectively;

however, those in time-matched control groups at 4 and

20 weeks were 132.4F 7.9 and 161F 9.2 mg/dl, respec-

tively. The levels of fructosamine, a parameter of glycated

protein in blood, increased in the 4WDM and 20WDM

groups (242.6F 12.9 and 279.2F 16.1 AM, respectively)

compared with those in time-matched control groups

(138F 7.6 and 160.8F 8.3 AM). TBARS levels also in-

creased in the 4WDM and 20WDM groups (3.22F 0.29 and

5.24F 0.46 nM, respectively).

Morphological assessment of myocardial injury

Infarct size and myocardial death

To determine the infarct size, TCC staining was per-

formed. The infarct size in the 20WDM rat hearts was

significantly larger than that in the 4WDM rat hearts (Figs.

1A and B). A larger infarct size was observed in the

20WDM rat hearts at 65.2F 4.3% than in the time-matched

control group (56.0F 6.2%) and was significantly attenuat-

ed to 43.1F 3.6% in 4WDM rat hearts as compared with the

time-matched group (59.5F 5.6%, p < 0.05) (Table 2). The

dead myocytes was delineated by HRP. HRP was accurate

in determining the dead myocytes as well as assessing the

myocyte viability. In 20WDM rat hearts treated with I/R

(20WDM+ I/R), the HRP-positive area markedly increased

to 37.2F 3.1% compared with rat hearts treated with only I/

R (20WC+ I/R) (28.6F 2.7%, p < 0.05). In 4WDM rat

hearts treated with I/R (4WDM+ I/R), the HRP-positive

area was attenuated to 20.2F 2.9%. Some viable myocytes

were still present even at the center of the infarct zone

(Fig. 1C).

Apoptosis

TUNEL-positive myocytes were observed in both STZ-

treated and time-matched control hearts treated with I/R. In

the 20WDM+ I/R group, the number of TUNEL-positive

cells significantly increased compared with the 20WC+ I/R

group (20.4F 3.1% and 12.3F 1.4%, p < 0.05, respective-

ly), but decreased in the 4WDM+ I/R group compared with

Table 2

Infarct size and semiquantitative estimate of myocytes death

Groups Area at risk

(% of LV area)

Infarct size

(% of area at risk)

Myocytes death area

(% of TTC negative area)

TUNEL-positive myocytes

(% of PI-positive nuclei)

Non-infarct area Infarct area Non-infarct area Infarct area

4WC+ I/R 47.2F 4.0 59.5F 5.6 � 28.4F 2.4 � 11.5F 1.4

4WDM+ I/R 46.8F 5.2 43.1F 3.6y � 20.2F 2.9y � 6.4F 0.6y

20WC+ I/R 49.2F 4.5 56.0F 6.2 � 28.6F 2.7 � 12.3F 1.4

20WDM+ I/R 51.6F 4.8 65.2F 4.3 � 37.2F 3.1y + 20.4F 3.1y

W: Duration (weeks), DM: STZ-induced diabetes, C: time-matched control, � : negative for HRP or TUNEL, +: positive for HRP or TUNEL.y p< 0.05 vs. time-matched groups.

G. Xu et al. / Experimental and Molecular Pathology 76 (2004) 57–65 61

the 4WC + I/R group (6.4F 0.6% and 11.5F 1.4%,

p < 0.05, respectively). In addition, a few TUNEL-positive

myocytes were noted occasionally even in the nonischemic

area in the 20WDM+ I/R group, but not observed in the

4WDM+ I/R group (Fig. 2 and Table 2). TUNEL-positive

myocytes were not observed in 4WDM rat hearts, and only

a few myocytes were TUNEL-positive in 20WDM rat hearts

not treated with I/R (data not shown).

Ultrastructural observation

By transmission electron microscopy, necrotic and apo-

ptotic myocytes were observed in the STZ-treated hearts

within the infarcted zone. Necrotic myocytes exhibited

highly swollen cytoplasm and mitochondria, clumped and

Fig. 2. Photomicrographs showing TUNEL staining (green, indicated by arrows). I

noninfarcted area, and some TUNEL-positive myocytes were occasionally observe

few TUNEL-positive cells were observed in 20-week STZ-treated rat hearts outsid

following I/R treatment (G). B, D, F and H show a combination of myocardial nuc

stainings. Magnification, �250.

marginated nuclear chromatin and electron-dense deposits

in the mitochondria. The sarcolemma was ruptured. Apo-

ptotic myocytes exhibited nuclei with deep invaginations of

the nucleolonema and contained large sharply delineated

electron-dense chromatin masses around the nuclear enve-

lope. The cytoplasm demonstrated shrinkage but the sarco-

lemma appeared intact (Fig. 3).

ERK1/2 phosphorylation by ischemia and reperfusion in

hyperglycemia

Western blotting revealed two 44- and 42-kDa bands

corresponding to phosphorylated and total ERK1/2, respec-

tively (Figs. 4A and B). After different durations of STZ

treatment, opposite expression patterns of phosphorylated

n 4-week STZ-treated rat hearts, TUNEL-positive cells were not observed in

d within an infarcted area following I/R treatment (A and C, respectively). A

e an infarcted area (E), the number of TUNEL-positive myocytes increased

lei (indicated by arrows) identified by overlaying with TUNEL and PI (red)

Fig. 3. Transmission electron microscopy of 20-week STZ-treated rat hearts

with I/R. Necrotic myocyte (A): The cytoplasm and mitochondria are

highly swollen (*), and clumped and marginated nuclear chromatin (Nu)

and electron-dense deposits are observed in the mitochondria (indicated by

arrowhead). The sarcolemma is ruptured (indicated by arrow). Apoptotic

myocyte (B): The nucleus shows deep invaginations of the nucleolenoma

and contains large sharply delineated electron-dense chromatin masses

(Nu). The myofilament (Mf) and mitochondria (Mi) exhibit electron-dense

deposits and shrinkage. The sarcolemma (indicated by arrow) appears

intact. Magnification, �7000.

Fig. 4. A: Phosphorylated ERK1/2 (P-ERK) (left panel) and total ERK1/2 (T-ER

hyperglycemic groups with I/R as determined by Western blotting. B: Expression

quantified by Western blot analysis. The mean value of activities at 20-week STZ

G. Xu et al. / Experimental and Molecular Pathology 76 (2004) 57–6562

ERK1/2 in rat hearts were observed. Phosphorylated ERK1/

2 increased threefold in the 4WDM group compared with

the 4WC group. However, the phosphorylated ERK1/2 in

the 20WDM group decreased to about 1/3 compared with

the 20WC group. In the I/R-treated groups, phosphorylated

ERK1/2 level was elevated in the two time-matched groups

(4WC+ I/R and 20WC+ I/R), and phosphorylated ERK1/2

level also increased in the 4WDM+ I/R group. However, I/

R did not enhance the phosphorylation of ERK1/2 in the

20WDM+ I/R group (Figs. 4A and B, left panels). The level

of total ERK1/2 almost did not change in either group (Figs.

4A and B, right panels).

Discussion

In a previous study of hyperglycemic rats for different

durations, the cardiac functions were noted to be inhibited

and the myocardial ultrastructure was altered at 12 and 24

weeks, but no changes were observed at 6 weeks (Jackson

et al., 1985). We also confirmed the ultrastructural alter-

ation in the myocardium after long-term hyperglycemia

and less change after short-term hyperglycemia (Naito et

al., 2002). Tosaki et al. (1995) also demonstrated that the

diabetic heart was more resistant to ischemia/reperfusion

after short-term hyperglycemia, but this resistance was

absent in long-term hyperglycemia. In the present study,

we used two morphological and molecular markers, name-

ly TTC staining and HRP reaction to determine ischemic

myocardial injury and showed that as a previous study

(Takashi and Ashraf, 2000), HRP reaction was more

superior in confirmation of the dead myocytes in ischemic

lesion than TTC staining. The results of HRP reaction

confirmed that the ischemic and reperfused injuries were

enhanced after long-term hyperglycemia, whereas short-

term hyperglycemia appeared to decrease the size of infarct

lesion and the number of dead myocytes.

K) (right panel) were expressed in the myocardium of time-matched and

s of phosphorylated ERK1/2 (left panel) and total ERK1/2 (right panel) are

-treated group is 1. *P< 0.05, **P < 0.001 vs. time-matched control groups.

G. Xu et al. / Experimental and Molecular Pathology 76 (2004) 57–65 63

Biochemical and mechanical abnormalities of the myo-

cardium have been identified in short-term hyperglycemia

(Hofmann et al., 1995; Malhotra et al., 1997; Ishikawa et

al., 1999). It has been reported that metabolic changes

occurring during the early phase of diabetes may chemi-

cally precondition the myocardium, rendering it more

resistant to I/R injury (Tosaki et al., 1995). In this study,

we focused on ERK1/2, the intracellular protein kinases.

ERK1/2 is one of the subfamilies of MAPK involved in

ischemic preconditioning (PC), which is an effective en-

dogenous cardioprotective mechanism. The PC protects the

heart from ischemic injury that occurs via a signal trans-

duction pathway that includes ERK1/2 activations. ERK1/

2 would be activated in various tissues after short-term

hyperglycemia and in cells cultured under high-glucose

conditions (Awazu et al., 1999; Hua et al., 2001). In our

recent report, short-term hyperglycemia elevated ERK1/2

phosphorylation in the heart (Naito et al., 2002). This

study further suggests that the activation of ERK1/2 is

little altered under ischemic and reperfused stimulation

after short-term hyperglycemia. This may be because

ERK1/2 has been already sufficiently activated. The atten-

uated infarct size in this condition is considered to explain

this observation, an enhanced resistance to ischemia

through already activated ERK1/2 which may ‘mask’ the

effect. It seems that both short-term hyperglycemia and PC

share some common mechanism of myocardial protection.

More importantly, diabetes is a long-term disease. The

alteration of ERK1/2 with long-term hyperglycemia has

not been extensively investigated. In addition, little is

known about the role of long-term hyperglycemia in the

intracellular signaling pathway (Naito et al., 2002). Poten-

tial mechanisms for the development of larger infarcts in

diabetics may be related to biological abnormalities, such

as hyperglycemia, glycation of proteins, increased platelet

aggregation, alteration in lipid distribution and composi-

tion, reduced antioxidant defenses, and anomalies of myo-

cardial metabolism during ischemia (Forrat et al., 1993).

In our recent report, although the amount of total ERK1/

2 did not change, phosphorylated ERK1/2 level was

significantly reduced after long-term hyperglycemia. It

depends on ERK1/2-advanced glycation end products

(AGEs) formation, which alters its structure and function

(Naito et al., 2002). This study reconfirmed that ERK1/2

has significantly low level of activation and is not related

to I/R stimulation in long-term hyperglycemia. Attenua-

tion of this signal transduction pathway during long-term

hyperglycemia may also explain the failure of ischemic

PC to protect the myocardium. This result supports the

findings of Tosaki et al. (1995) that PC may be a ‘‘heart

health phenomenon’’. It may be related to the evolution

of myocardial injury as increased infarct size and myo-

cardial death, that is, necrosis and apoptosis in long-term

hyperglycemia.

Recently, apoptosis, a form of programmed cell death

has been reported to occur in infarcted myocardium by

Takashi and Ashraf (2000). Additionally, clinical and

experimental results have described apoptosis in diabetic

hearts (Fiordaliso et al., 2000). A direct correlation exists

between hyperglycemia and oxidative stress. The oxidative

damage may be the initial onset of apoptosis in a diabetic

heart (Fiers et al., 1999). A recent study has identified

some of the underlying biochemical mechanisms by which

intracellular Ca2 + overload can trigger apoptosis. Hyper-

glycemia with I/R was demonstrated to accelerate the

occurrence of apoptosis of renal cells (Melin et al.,

2001). It is an intrinsic mechanism that mediates Ca2+

overload in response to a disturbance of the redox state in

apoptosis. In the present study, we suggest that the

occurrence of TUNEL-positive myocytes in the hypergly-

cemic heart depends on the accumulation of superoxides

produced such as TBARS in the serum after hyperglyce-

mia of the duration of diabetes. More importantly, the

number of TUNEL-positive myocytes increased signifi-

cantly in long-term hyperglycemia with I/R. The findings

of this study show that myocyte death in the presence

hyperglycemia during infarction involves both apoptotic

and necrotic cell death. Increasing evidence suggests that

the activation of the Ras–Raf-1–MEK–ERK pathway is a

protective mechanism against apoptosis. This pathway has

been implicated in the phosphorylation of bcl-2, providing

a potential link between ERK1/2 activation and myocyte

survival. Furthermore, the blockade of ERK1/2 activation

augmented the apoptosis of cardiomyocytes, suggesting

that the ERK1/2 pathway may mediate antiapoptotic sig-

naling in myocytes (Aikawa et al., 1997). This study also

confirmed that the activation of ERK1/2 attenuated apo-

ptotic signaling in short-term hyperglycemia and enhanced

it in long-term hyperglycemia due to different reactions of

ERK1/2. In addition, previous studies suggested that the

activation of PKC and the opening of the mitoKATP

channel are also important pathways in myocardial protec-

tion (Takashi et al., 1999; Wang et al., 2001). PKC is

upstream of the ERK1/2 pathway and the end effector is

the mitoKATP channel in the myocardium (Ping et al.,

1999). Although we did not detect the alteration of PKC

activation and the mitoKATP channel in this study, previous

studies suggested that prolonged hyperglycemia leads to

the downregulation of PKC activity (Gabbay et al., 1990;

Cooper et al., 1993) and mitoKATP channel impairment

(Kersten et al., 2001). Therefore, PKC and the mitoKATP

channel may be also regulated by the ERK1/2 pathway

and related to the etiology of diabetic cardiomyopathy.

Furthermore, to recover the endogenous cardioprotective

function by applying substances, such as aminoguanidine,

antioxidants and vitamin E, restoration of ERK1/2 phos-

phorylation by inhibiting AGEs formation would be im-

portant. In addition, short-term glucose treatment may be

similar to an ischemic PC, which prevents protects against

myocardial injury due to ischemia.

In conclusion, this study demonstrates that long-term

hyperglycemia enhances the ischemic injury and inhibits

G. Xu et al. / Experimental and Molecular Pathology 76 (2004) 57–6564

the rate of ERK1/2 phosphorylation, whereas short-term

hyperglycemia reduced the myocardial injury. The reverse

response of ERK1/2 may result in the different sensitiv-

ities to myocardial ischemia in short- and long-term

hyperglycemic myocardium. Short-term hyperglycemia

seems to be associated with an endogenous cardioprotec-

tive effect via ERK1/2 activation; the physiological adap-

tation may enhance myocardial survival after an ischemic

injury. In long-term hyperglycemia, the myocardium is

more sensitive to ischemic and reperfusion injuries result-

ing from ERK1/2 dysfunction. It seems that long-term

hyperglycemia is a more appropriate model of the path-

ologic alterations observed in clinical diabetes.

Acknowledgments

The authors thank Ms. Kawahara, Mr. Teduka and Ms.

Kawamoto (Department of Pathology II, Nippon Medical

School) for skillful technical assistance.

References

Aikawa, R., Komuro, I., Yamazaki, T., Zou, Y., Kudoh, S., Tanaka,

M., Shiojima, I., Hiroi, Y., Yazaki, Y., 1997. Oxidative stress

activates extracellular signal-regulated kinases through src and rats

in cultured cardiac myocytes of neonatal rats. J. Clin. Invest. 100,

1813–1821.

Awazu, M., Ishikura, K., Hida, M., Hoshiya, M., 1999. Mechanisms of

itogen-activated protein kinase activation in experimental diabetes.

J. Am. Soc. Nephrol. 10, 738–745.

Cooper, D.R., Watson, J.E., Dao, M.L., 1993. Decreased expression of

protein kinase-C alpha, beta, and epsilon in soleus muscle of Zucker

obese (fa/fa) rats. Endocrinology 133, 2241–2247.

Davis, R.J., 1993. The mitogen-activated protein kinase signal transduction

pathway. J. Biol. Chem. 268, 14553–14556.

Fein, F.S., Sonnenblick, E.H., 1985. Diabetic cardiomyopathy. Prog. Car-

diovasc. Dis. 27, 255–270.

Fiers, W., Beyaert, R., Declercq, W., Vandenabeele, P., 1999. More than

one way to die: apoptosis, necrosis and reactive oxygen damage. On-

cogene 18, 7719–7730.

Fiordaliso, F., Li, B., Latini, R., Sonnenblick, E.H., Anversa, P., Leri, A.,

Kajstura, J., 2000. Myocyte death in streptozotocin-induced diabetes in

rats is angiotensin II- dependent. Lab. Invest. 80, 513–527.

Forrat, R., Sebbag, L., Wiernsperger, N., Guidollet, J., Renaud, S., de

Lorgeril, M., 1993. Acute myocardial infarction in dogs with experi-

mental diabetes. Cardiovasc. Res. 27, 1908–1912.

Fryer, R.M., Pratt, P.F., Hsu, A.K., Gross, G.J., 2001. Differential acti-

vation of extracellular signal regulated kinase isoforms in precondi-

tioning and opioid-induced cardioprotection. J. Pharmacol. Exp. Ther.

296, 642–649.

Gabbay, R.A., Siconolfi-Baez, L., Lebovitz, H.E., 1990. Calcium-depend-

ent protein kinase activity is decreased in diabetic rat sciatic nerve.

Endocrinology 127, 1194–1198.

George, M., Linda, H., 2001. Diabetic cardiomyopathy: fact or fiction?

Heart 85, 247–248.

Ghosh, S., Standen, N.B., Galinianes, M., 2001. Failure to precondi-

tion pathological human myocardium. J. Am. Coll. Cardiol. 37,

711–718.

Hamby, R.I., Zoneraich, S., Sherman, L., 1974. Diabetic cardiomyopathy.

JAMA 229, 1749–1754.

Hofmann, P.A., Menon, V., Gannaway, K.F., 1995. Effects of diabetes on

isometric tension as a function of [Ca2 +] and pH in rat skinned cardiac

myocytes. Am. J. Physiol. 269, H1656–H1663.

Hua, H., Goldberg, H.J., Fantus, I.G., Whiteside, C.I., 2001. High glu-

cose-enhanced mesangial cell extracellular signal-regulated protein

kinase activation and alpha1 (IV) collagen expression in response to

endothelin-1: role of specific protein kinase C isozymes. Diabetes 50,

2376–2383.

Ishikawa, T., Kajiwara, H., Kurihara, S., 1999. Alterations in contractile

properties and Ca2 + handling in streptozotocin-induced diabetic rat

myocardium. Am. J. Physiol. 277, H2185–H2194.

Jackson, C.V., McGrath, G.M., Tahiliani, A.G., Vadlamudi, R.V., McNeill,

J.H., 1985. A functional and ultrastructural analysis of experimental

diabetic rat myocardium. Manifestation of a cardiomyopathy. Diabetes

34, 876–883.

Katori, T., Bannai, C., Hayashi, Y., Yamashita, K., 1999. Decreased Na,K-

ATPase activity by glycation at the catalytic center. Horm. Metab. Res.

31, 5–7.

Kersten, J.R., Montgomery, M.W., Ghassemi, T., Gross, E.R., Toller, W.G.,

Pagel, P.S., Warltier, D.C., 2001. Diabetes and hyperglycemia impair

activation of mitochondrial KATP channels. Am. J. Physiol. 280,

1744–1750.

Liu, Y., Thornton, J.D., Cohen, M.V., Downey, J.M., Schaffer, S.W., 1993.

Streptozotocin-induced non-insulin-dependent diabetes protects the

heart from infarction. Circulation 88, 1273–1278.

Liu, X., Wang, J., Takeda, N., Binaglia, L., Panagia, V., Dhalla, N.S., 1999.

Changes in cardiac protein kinase C activities and isozymes in strepto-

zotocin-induced diabetes. Am. J. Physiol. 277, E798–E804.

Malhotra, A., Reich, D., Reich, D., Nakouzi, A., Sanghi, V., Geenen, D.L.,

Buttrick, P.M., 1997. Experimental diabetes is associated with func-

tional activation of protein kinase C epsilon and phosphorylation of

troponin I in the heart, which are prevented by angiotensin II receptor

blockade. Circ. Res. 81, 1027–1033.

McGuire, D.K., Granger, C.B., 1999. Diabetes and ischemic heart disease.

Am. Heart J. 138, S366–S375.

Melin, J., Hellberg, O., Larsson, E., Zezina, L., Fellstrom, B.C., 2001.

Protective effect of insulin on ischemic renal injury in diabetes mellitus.

Kidney Int. 61, 1383–1392.

Naito, Z., Kudo, M., Xu, G., Nishigaki, R., Yokoyama, M., Yamada, N.,

Asano, G., 2000. Immunohistochemical localization of mitogen-acti-

vated protein kinase (MAPK) family and morphological changes in

rat heart after ischemia-reperfusion injury. Med. Electron. Microsc.

33, 74–81.

Naito, Z., Takashi, E., Xu, G., Ishiwata, T., Teduka, K., Yokoyama, M.,

Yamada, N., Sugisaki, Y., Asano, G., 2002. Different influences of

hyperglycemic duration on phosphorylated extracellular signal-regu-

lated kinase 1/2 in rat heart. Exp. Mol. Pathol. 74, 23–32.

Paulson, D.J., 1997. The diabetic heart is more sensitive to ischemic injury.

Cardiovasc. Res. 34, 104–112.

Pearson, G., Robinson, F., Beers Gibson, T., Xu, B.E., Karandikar, M.,

Berman, K., Cobb, M.H., 2001. Mitogen-activated protein (MAP) kin-

ase pathways: regulation and physiological functions. Endocr. Rev. 22,

153–183.

Ping, P., Zhang, J., Cao, X., Li, R.C., Kong, D., Tang, X.L., Qiu, Y.,

Manchikalapudi, S., Auchampach, J.A., Black, R.G., Bolli, R., 1999.

PKC-dependent activation of p44/p42 MAPKs during myocardial

ischemia-reperfusion in conscious rabbits. Am. J. Physiol. 276,

H1468–H1481.

Takashi, E., Ashraf, M., 2000. Pathologic assessment of myocardial

cell necrosis and apoptosis after ischemia and reperfusion with

molecular and morphological markers. J. Mol. Cell. Cardiol. 32,

209–224.

Takashi, E., Wang, Y., Ashraf, M., 1999. Activation of mitochondrial

KATP channel elicits late preconditioning against myocardial infarc-

tion via protein kinase C signaling pathway. Circ. Res. 85,

1146–1153.

Tosaki, A., Engelman, D.T., Engelman, R.M., Das, D.K., 1995. Diabetes

G. Xu et al. / Experimental and Molecular Pathology 76 (2004) 57–65 65

and ATP-sensitive potassium channel openers and blockers in iso-

lated ischemic/reperfused hearts. J. Pharmacol. Exp. Ther. 275,

1115–1123.

Vogel, W.M., Apstein, C.S., 1988. Effects of alloxan-induced diabetes

on ischemia – reperfusion injury in rabbit hearts. Circ. Res. 62,

975–982.

Wang, Y., Takashi, E., Xu, M., Ayub, A., Ashraf, M., 2001. Downregula-

tion of protein kinase C inhibits activation of mitochondrial KATP chan-

nels by diazoxide. Circulation 104, 85–90.

Yue, T.L., Wang, C., Gu, J.L., Ma, X.L., Kumar, S., Lee, J.C., Feuerstein,

G.Z., Thomas, H., Maleeff, B., Ohlstein, E.H., 2000. Inhibition of ex-

tracellular signal-regulated kinase enhances Ischemia/Reoxygenation-

induced apoptosis in cultured cardiac myocytes and exaggerates reper-

fusion injury in isolated perfused heart. Circ. Res. 86, 692–699.