The rapid onset of hyperglycaemia in ZDF ratswas associated with a widespread alterationof metabolic proteins implicated in glucosemetabolism in the heart

Claude Lajoie, Louise Beliveau, Francois Trudeau, Nathalie Lavoie,Guy Massicotte, Sylvain Gagnon, and Angelino Calderone

Abstract: The present study tested the hypothesis that the phosphorylation and regulation of metabolic proteins implicatedin glucose homeostasis were impaired in the heart of the type 2 diabetic Zucker-diabetic-fatty (ZDF) rat model. The onsetof hyperglycaemia in ZDF rats was not uniform, instead it either progressed rapidly (3–4 weeks) or was delayed (6–8 weeks). In both the early and late onset hyperglycaemic ZDF rats, AMPKa Thr172 phosphorylation in the heart was sig-nificantly decreased. In the early onset hyperglycaemic ZDF rats, PKB Ser473 phosphorylation was reduced, whereasThr308 phosphorylation was significantly increased. In the late onset hyperglycaemic ZDF rats, PKB Ser473 phosphorylationwas unchanged, but Thr308 phosphorylation remained elevated. Cardiac GLUT4 protein and mRNA expression were signif-icantly reduced in the early onset hyperglycaemic ZDF rats, whereas increased protein expression was observed in the lateonset hyperglycaemic ZDF rats. In conclusion, the present study has demonstrated that following a more rapid onset of hy-perglycaemia, the type 2 diabetic heart is more prone to alterations in the signaling proteins implicated in glucose metabo-lism.

Key words: type 2 diabetes, heart, PKB, AMPK, glycogen.

Resume : La present etude a verifie l’hypothese que la phosphorylation et la regulation des proteines metaboliques partici-pant a l’homeostasie du glucose sont alterees dans le coeur du modele de rat Zucker obese et diabetique (ZDF) de type 2.L’apparition de l’hyperglycemie chez les rats ZDF n’a pas ete uniforme; elle a progresse rapidement (3–4 semaines) ou aete retardee (6–8 semaines). Tant chez les rats ZDF presentant une hyperglycemie precoce que tardive, la phosphorylationde l’AMPKa sur le residu Thr172 a diminue de maniere significative dans le coeur. Chez les rats ZDF presentant une hy-perglycemie precoce, la phosphorylation de la PKB sur le residu Ser473 a diminue, alors que la phosphorylation de Thr308

a augmente significativement. Chez les rats ZDF presentant une hyperglycemie tardive, la phosphorylation de la PKB surle residu Ser473 est demeuree stable, mais la phosphorylation de Thr308 est demeuree elevee. L’expression de l’ARNm etde la proteine GLUT4 dans le coeur a ete reduite de maniere significative chez les rats presentant une hyperglycemie pre-coce, alors qu’une augmentation de l’expression de la proteine a ete observee chez les rats ZDF presentant une hypergly-cemie tardive. En conclusion, cette etude a demontre que le cœur diabetique de type 2 est plus sensible a desmodifications au niveau des proteines de signalisation impliquees dans le metabolisme du glucose apres une apparitionplus precoce de l’hyperglycemie.

Mots cles : diabete de type 2, coeur, PKB, AMPK, glycogene.

[Traduit par la Redaction]

Received 16 December 2005. Published on the NRC Research Press Web site at http://cjpp.nrc.ca on 21 December 2006.

C. Lajoie.1 Department of Human Kinetics, Laurentian University, Ramsey Lake Road, ON P3E 2C6, Canada; Department ofKinesiology, University of Montreal, C.P. 6128, Succ. Centre-Ville, Montreal, QC H3C 3J7, Canada; Departement des scienceshumaines, 555, boulevarde de l’Universite, Chicoutimi, QC G7H 2B1, Canada.L. Beliveau. Department of Kinesiology, University of Montreal, C.P. 6128, Succ. Centre-Ville, Montreal, QC H3C 3J7, Canada.F. Trudeau and S. Gagnon. Neuroscience Research Group, Universite du Quebec, Trois-Rivieres, QC G9A 5H7, Canada.N. Lavoie and G. Massicotte. Neuroscience Research Group, Universite du Quebec, Trois-Rivieres, QC G9A 5H7, Canada;Departement de Chimie-Biologie, Universite du Quebec a Trois-Rivieres, QC G9A 5H7, Canada.A. Calderone. Department of Physiology, University of Montreal, C.P. 6128, Succ. Centre-Ville, Montreal, QC H3C 3J7, Canada;Montreal Heart Institute, 5000, Belanger est, Montreal, QC H1T 1C8, Canada.

1Corresponding author (e-mail: Claude_Lajoie@UQAC.ca).

1205

Can. J. Physiol. Pharmacol. 84: 1205–1213 (2006) doi:10.1139/Y06-070 # 2006 NRC Canada

IntroductionType 2 diabetes is a prevalent cause of morbidity and

mortality, mainly due to cardiovascular complications(Sowers et al. 2001). Moreover, hyperglycaemia worsensthe prognosis of myocardial infarction in diabetic patients(Capes et al. 2001). A major cellular abnormality in the dia-betic heart is an alteration in the control of carbohydratemetabolism shortly after the development of hyperglycae-mia in the absence of overt changes in systolic function(Chatham and Seymour 2002). Resolving the molecularevents that impair the metabolic behavior of the diabeticmyocardium could provide the requisite data for a therapeu-tic approach in the treatment of type 2 diabetes. Several sig-naling protein kinases coupled to the insulin pathway andenergy metabolism may represent therapeutic targets in thetreatment of type 2 diabetes. These kinases include proteinkinase B (PKB), glycogen synthase kinase 3 (GSK-3), andAMP-activated protein kinase a (AMPKa) isoforms.

AMPK is characterized as a major regulator of cardiac en-ergy substrate acting as a metabolic sensor following an en-ergetic imbalance (Winder and Hardie 1999). When ATPlevels are compromised, AMPK responds by either switch-ing off anabolic energy requiring processes, such as lipid bi-osynthesis pathways, or promoting ATP-producing catabolicprocesses via phosphorylation of metabolic enzymes. AMPKis activated by metabolic inhibitors, hypoglycaemia, hypo-xia, and ischemia (Russell et al 2004) and inactivated by in-sulin during normoxic conditions (Kovacic et al. 2003;Witters and Kemp 1992). Furthermore, pharmacologicalstimulation of AMPK increased GLUT 4 regulation in skel-etal muscle (Pold et al. 2005). Likewise, the recruitment ofPKB via an insulin-dependent pathway also promoted glu-cose uptake via GLUT4 (Kandel and Hay 1999). Interest-ingly, PKB contributes to the complexity of AMPKregulation by decreasing Thr172 phosphorylation leading toenzyme inhibition. Furthermore, PKB can also promote gly-cogen synthesis and cardiac hypertrophy (Cross et al. 1995;Haq et al. 2001) via the phosphorylation and inactivation ofGSK-3, a ubiquitously expressed serine/threonine kinasewith 2 related isoforms, GSK-3a and GSK-3b (Cross et al.1995; Woodgett 1990).

The Zucker diabetic fatty (ZDF) rat is an insulin-resistantanimal model, which genetically manifests numerous char-acteristics of type 2 diabetes similar to those observed in hu-mans (Clark et al. 1983), as the early progression of thedisease is associated with progressive hyperglycaemia, ele-vated myocardial glycogen content, and plasma insulin lev-els (Lajoie et al. 2004). Thus, employing the ZDF rat model,the present study tested the hypothesis that the phosphoryla-tion and regulation of metabolic proteins implicated in glu-cose homeostasis in the heart were impaired.

Materials and methods

Experimental proceduresAll protocols were approved by the Animal Care Commit-

tee of University of Quebec in Trois-Rivieres and followedthe Principles of Laboratory Animal Care (NIH publication85-23, 1985). Ten male ZDF rats and ten of their lean wild-type controls were studied. ZDF rats were obtained fromGenetic Models (GmiTM, Indianapolis, Ind.). Animals were

6 weeks old at reception. A standard rat diet was used andthe animals were fed ad libitum. The light/dark cycle was12 h / 12 h.

Early and late onset of hyperglycaemia in ZDF ratsPlasma glucose levels were measured with a glucometer

(Bayer, Toronto, Ont.) from blood samples taken from thetail. Once the diabetic rat reached a glycemic level of15 mmol/L, animals were monitored for an additional6 weeks and sacrificed. Based on the latter approach, eachZDF rat was exposed to a hyperglycaemic (plasmaglucose >15 mmol/L) environment for the same duration. In5 ZDF rats, plasma glucose levels rose rapidly within3 weeks and progressively increased for the remaining6 weeks just prior to sacrifice (14.5 weeks old) (Fig. 1). Bycontrast, the rise of plasma glucose in 4 ZDF rats was mark-edly slower, as compared with the latter group (Fig. 1).Lastly, 1 ZDF rat failed to develop hyperglycaemia (Fig. 1).Based on these data, ZDF animals were characterized ashaving either early or late onset of hyperglycaemia.

At the end of the study, rats were anesthetized withisoflurane (Janssen, Toronto, Ont.) and sacrificed by decapi-tation. Plasma insulin levels were measured by radio-immunoassay (Linco Research, St-Charles, Mo.) in bloodsamples collected in heparinized glass test tubes. The heartwas rapidly excised and the atria and both ventricles wereweighed, frozen in liquid nitrogen, and stored at –80 8Cuntil analyzed. Glycogen concentration in the left ventriclewas determined spectrophotometrically using sulfuric acid(Lo et al. 1970).

Homogenization and sample preparationThe left ventricle was pulverized in a mortar cooled with

liquid nitrogen and subsequently transferred to a lysis buffercontaining 150 mmol/L NaCl, 10 mmol/L Tris, 1 mmol/LEDTA, 1 mmol/L EGTA, pH 7.4, 1% Triton X-100, 0.5%Nonidet P-40, protease inhibitors (0.5 mmol/L PMSF,

Fig. 1. The temporal pattern of hyperglycaemia in Zucker-diabetic-fatty (ZDF) rats. ZDF rats developed hyperglycaemia either early(ZDF-E; n = 5) or late (ZDF-L; n = 4) as compared with ZDF-leanrats. Plasma glucose levels remained normal in ZDF-lean rats dur-ing the course of the study. One-way ANOVA revealed a signifi-cant difference (p < 0.05 value) between the early and late onsethyperglycaemic ZDF rats and between ZDF-diabetic (regardless ofthe group) and ZDF-lean rats.

1206 Can. J. Physiol. Pharmacol. Vol. 84, 2006

# 2006 NRC Canada

1 mg/mL aprotinin, 1 mg/mL leupeptin), and phosphataseinhibitors (100 mmol/L sodium orthovanadate, 10 mmol/Lsodium fluoride), and re-homogenized. Lysates were subse-quently centrifuged at 12 000g for 10 min at 4 8C to re-move insoluble material. The supernatant was removed,and the protein content was determined by the Bradfordprotein assay (Bio-Rad Laboratories, Hercules, Calif.).

Protein analysis by Western blottingThe phosphorylation and regulation of metabolic proteins

was determined by Western blot analysis. Briefly, 100–200 mg of left ventricular lysates was subjected to SDS –polyacrylamide gel (10%–15%) electrophoresis, and trans-ferred to a polyvinylidene difluoride (PVDF) membrane(Millipore, Billerica, Mass.). Equal protein loading and suc-cessful transfer was confirmed by Ponceau S staining.PVDF membranes were incubated in PBS–Tween 20 (0.1%v/v) supplemented with 5% skim milk. By contrast, mem-branes were incubated in PBS–Tween 20 (0.1%) supple-mented with 5% bovine serum albumin (BSA) to assessphosphorylated proteins. The following antibodies wereused: a rabbit polyclonal directed against either PKB(1:500; Cell Signaling), phosphorylated serine473 (Ser473;1:1000), or threonine308 (Thr308; 1:1000) residue of PKBa(Cell Signaling); a rabbit polyclonal antibody (1:500) di-rected against either GSK-3b (Cell Signaling) or the phos-phorylated serine21 (Ser21) residue of glycogen synthasekinase-3a (GSK-3a; ~51 kDa; Cell Signaling) and phos-phorylated Ser9 residue of GSK-3b (~46 kDa); a rabbitpolyclonal directed against either AMPKa-pan (1:1000;Upstate, Charlottesville, Va.) or phosphorylated threo-nine172 (AMPKa thr172; 1:1000) of AMPKa (Cell SignalingTech., Beverly, Mass.); and a rabbit polyclonal antibodydirected against GLUT4 (1:1000; Santa Cruz Biotechnol-ogy, Santa Cruz, Calif). Following overnight antibody in-cubation at 4 8C, the membrane was washed and re-probed with an anti-rabbit conjugated horseradish peroxi-dase antibody in PBS–Tween 20 containing 3% skimmilk (1:10 000; Santa Cruz Biotechnology) for 1–2 h atroom temperature. The bands were subsequently detectedby autoradiography with the ECL detection kit (AmershamLtd, Piscataway, N.J.). Films were quantified using a flatbedscanner and Scion image (Scion Corp., Frederick, Md.).

Real time PCRRT-PCR was performed by standard methodology on total

RNA isolated from the left ventricle, as previously described(Nguyen et al. 2003). Real time PCR was performed accord-

ing to the manufacturer’s instructions employing SYBRGreen (Applied BioSystems, Foster City, Calif.). Primers foreach gene were obtained from distinct exons that span an in-tron employing the program Ensembl Genome Browser(http://www.ensembl.org). The sequence specificity of eachprimer was verified with the program Blast derived from theNational Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov). The primers used were the following: ratAMPK a2, forward 5’-GTGATCAGCACTCCAACAGACT-3’ and reverse 5’-GGCGAGCTTCCACCTCTTCAA-3’; ratGLUT4, 5’-GTTTCCAGCAGATCGGCTCTGA-3’ and re-verse 5’-GCAAGGACCAGTGTCCCAGTCA-3’; rat ANF,forward 5’-AGAGCGGACTAGGCTGCAACA-3’ and re-verse 5’ATTTGGCTGTTATCTTCGGTA-3’; and rat b-actinforward 5’-CCCTAAGGCCAACCGTGAA-3’ and reverse 5’-GAGGCATACAGGGACAACACAG-3’. Appropriate nega-tive controls were used for each experiment.

StatisticsIn the Western blot experiments, a densitometric approach

was used to assess either the phosphorylation status or pro-tein content in ZDF-diabetic and lean rats. Semi-quantitativeanalysis of the data was determined by the phosphorylationlevel normalized to total protein content. The data was rep-resented as the mean ± SE and in ZDF-lean rats the valuewas arbitrarily set at 1. Correlations were assessed by aPearson product-moment test. Data were analyzed by eithera one-way or two-way ANOVA followed by Newman–Keuls post-hoc test. The level of statistical significance wasset at p < 0.05.

Results

Early and late ZDF grouping according to thedevelopment of hyperglycaemia

Following their reception, 5 ZDF rats developed hyper-glycaemia within 3–4 weeks, and their plasma glucose lev-els progressively rose until sacrifice 6 weeks later (Fig. 1).By contrast, a delayed increase in plasma glucose was ob-served in 4 additional ZDF rats, and a hyperglycaemic statewas observed 6–8 weeks following reception (Fig. 1). De-spite the temporal delay in the onset of hyperglycaemia, allZDF rats were exposed to a hyperglycaemic environment(plasma glucose >15 mmol/L) for an identical duration(6 weeks). An additional ZDF rat failed to develop hyper-glycaemia and was not included in the study. Based on thesedata, ZDF rats were classified as having either early or lateonset of hyperglycaemia (Fig. 1). Plasma glucose levels re-

Table 1. Morphologic data and insulin and glycogen concentrations in Zucker diabetic fatty rats.

Early ZDF-diabetic ZDF-lean Late ZDF-diabetic ZDF-lean

Body mass (g) 379±27*,{ 303±21{ 457±18* 344.5±23Heart (g) 1.15±0.10 1.05±0.05{ 1.27±0.04 1.29±0.21HM/BM (mg/g) 3.05±0.03* 3.47±0.01 2.79±0.08* 3.70±0.45LV (g) 0.88±0.06 0.76±0.05{ 0.93±0.03 0.95±0.17LV/BM (mg/g) 2.32±0.19 2.52±0.15 2.05±0.12* 2.76±0.35Atria (mg) 56±4* 41±4{ 68±1 65±18Plasma insulin (ng/mL) 3.01±0.3*,{ 1.28±0.2 4.86±1.3* 0.99±0.3Glycogen (mg/mg tissue) 2.95±0.16* 1.06±0.11 3.04±0.21* 0.63±0.11

Note: *p < 0.05 vs. respective control; {p < 0.05 vs. late control; and {p < 0.05 vs. late ZDF.

Lajoie et al. 1207

# 2006 NRC Canada

mained constant in ZDF lean rats and were significantlylower as compared with either early or late onset hypergly-caemic ZDF rats (Fig. 1).

Morphologic dataThe body mass of the early and late onset hyperglycaemic

ZDF rats was significantly higher compared with ZDF leanrats (Table 1). Furthermore, the body mass of late hypergly-caemic ZDF rats was higher than the early hyperglycaemicZDF rats. The latter result was partially attributed to theolder age of the rats in the late hyperglycaemic group(Table 1). Regardless, the heart mass of the ZDF diabeticgroup was similar to ZDF lean rats (Table 1). The HM/BMratio was significantly lower in both the early and late onsethyperglycaemic ZDF rats compared with ZDF lean rats(Table 1). The latter result was partially attributed to thehigher body mass of the ZDF diabetic rats. Lastly, the ex-pression of ANF mRNA (normalized to b-actin mRNA), amarker of cardiac hypertrophy, was similar in the left ven-tricle of ZDF diabetic rats (6.85 ± 1.5; n = 6), regardless

the onset of hyperglcaemia, as compared with ZDF leanrats (5.97 ± 1.69; n = 6).

Plasma insulin levels and cardiac glycogen contentPlasma insulin in the ZDF diabetic group was signifi-

cantly increased compared with ZDF lean rats (Table 1).Furthermore, in the late onset hyperglycaemic ZDF rats,plasma insulin levels were significantly greater comparedwith early onset hyperglycaemic ZDF rats (Table 1). Lastly,the glycogen content in the left ventricle of the ZDF diabeticgroup was significantly higher compared with ZDF lean rats(Table 1).

PKB regulationRegardless of the ZDF diabetic group, Thr308 phosphory-

lation of PKB in the left ventricle was significantly in-creased in the absence of a change in total PKB proteincontent, as compared with ZDF lean rats (Fig. 2). By con-trast, Ser473 phosphorylation of PKB in the left ventricle ofearly onset hyperglycaemic ZDF rats was significantly re-

Fig. 2. PKB phosphorylation in ZDF-diabetic rats. (Panel A) PKB Thr308 phosphorylation in ZDF-diabetic early (ZDF-diabetic-E) andZDF-diabetic late (ZDF-diabetic-L) hyperglycaemic rats compared with respective ZDF-lean rats. (Panel B) PKB Ser473 phosphorylation inZDF-diabetic-E and ZDF-diabetic late hyperglycaemic rats compared with respective ZDF-lean rats. (Panel C) Total PKB protein content inZDF-diabetic-E, ZDF-diabetic-L hyperglycaemic rats, and ZDF-lean rats. (Panel D) Semi-quantitative analysis of PKB Ser473 phosphoryla-tion normalized to total PKB protein. In ZDF-lean rats, the value was arbitrarily set at 1. (Panel E) Semi-quantitative analysis of PKB Thr308

phosphorylation normalized to total PKB protein. In ZDF-lean rats, the value was arbitrarily set at 1. ap < 0.05, ZDF-diabetic vs. respectiveZDF-lean; bp < 0.05, ZDF-diabetic-E vs. ZDF-diabetic-L.

1208 Can. J. Physiol. Pharmacol. Vol. 84, 2006

# 2006 NRC Canada

duced compared with ZDF lean rats (Fig. 2). In the late on-set hyperglycaemic ZDF rats, PKB Ser473 phosphorylationwas similar to ZDF lean rats.

AMPK regulationAMPKa2 mRNA expression was similar in both early

(1.03 ± 0.2) and late onset (1.24 ± 0.3) hyperglycaemicZDF rats compared with ZDF lean rats. Furthermore, totalAMPK protein content was comparable between ZDF dia-betic rats, regardless of the onset as compared with ZDFlean rats (Fig. 3). By contrast, Thr172 phosphorylation ofAMPKa was significantly reduced in both the early and lateonset hyperglycaemic ZDF rats compared with ZDF leanrats (Fig. 3). In addition, the decrease of AMPKa Thr172

phosphorylation was greater in the early versus late onsethyperglycaemic ZDF rats (Fig. 3).

GSK-3 regulationGSK-3a phosphorylation was significantly decreased by

48% and 45% in both the early and late onset hyperglycae-mic ZDF rats, respectively, compared with ZDF lean rats(Fig. 4). By contrast, GSK-3b phosphorylation was modestlyincreased (+21%; p < 0.05) in the early onset hyperglycae-mic ZDF rats and reduced (–41%; p < 0.05) in the late onset

hyperglycaemic ZDF rats, as compared with ZDF lean rats(Fig. 4). GSK-3b total protein content was similar betweenZDF diabetic rats, regardless of the onset, as compared withZDF lean rats (Fig. 4). ZDF diabetic rats from both the earlyand late onset groups were pooled and GSK-3a and GSK-3bphosphorylation correlated (r = –0.50 and –0.78, respec-tively; p < 0.05) with PKB Ser473 phosphorylation. By con-trast, only GSK-3a phosphorylation correlated with PKBThr308 phosphorylation (r = –0.78; p < 0.05) in ZDF diabeticrats, regardless of the onset of the disease.

GLUT4 protein and gene expressionIn the myocardium of early onset ZDF-diabetic rats,

GLUT4 protein and mRNA expression were significantlydecreased compared with ZDF lean rats (Fig. 5). In contrast,in the late onset ZDF-diabetic rats, GLUT4 protein levelswere elevated in the myocardium but was not attributed toincreased mRNA transcription (Fig. 5).

Discussion

As observed in the present study, a previous report dem-onstrated a temporal delay in the development of hypergly-caemia in ZDF rats (Slieker et al. 1992). Furthermore, the

Fig. 3. AMPK phosphorylation in ZDF-diabetic rats. (Panel A) AMPK a2 Thr172 phosphorylation in ZDF-diabetic-early (ZDF-diabetic-E)and ZDF-diabetic-late (ZDF-diabetic-L) hyperglycaemic rats compared with respective ZDF-lean rats. (Panel B) Total AMPK protein con-tent in ZDF-diabetic-E, ZDF-diabetic-L hyperglycaemic rats, and ZDF-lean rats. (Panel C) Semi-quantitative analysis of AMPK a2 Thr172

phosphorylation normalized to total AMPK protein. In ZDF-lean rats, the value was arbitrarily set at 1. ap < 0.05, ZDF-diabetic vs. respec-tive ZDF-lean rats; bp < 0.05, ZDF-diabetic-E vs. ZDF-diabetic-L.

Lajoie et al. 1209

# 2006 NRC Canada

progression of diabetes, regardless of the stage of onset, wasnot associated with a hypertrophic response of the heart, asANP mRNA expression was similar between lean and dia-betic groups. A seminal finding of the present study wasthat in the early onset hyperglycaemic ZDF rats, cardiacPKB Ser473 and AMPKa Thr172 phosphorylation, were sig-nificantly reduced. Furthermore, a significant reduction inboth GLUT4 protein and mRNA expression was observedin the early onset hyperglycaemic ZDF rats. In the late onsethyperglycaemic ZDF rats, however, only AMPKa Thr172

phosphorylation remained significantly reduced.Laviola and al. (2001) observed in a rat model of type 1

diabetes that PKB Ser473 phosphorylation in the heart wasreduced. In the present study, myocardial PKB Ser473 phos-phorylation was likewise reduced in the early onset hyper-glycaemic ZDF rats, whereas phosphorylation wasunchanged in the late onset group. By contrast, PKB Thr308

phosphorylation was markedly elevated in both early andlate onset hyperglycaemic ZDF rats. Collectively, these data

suggest that the upstream kinases implicated in PKB Ser473

and Thr308 are disparate and differentially influenced by hy-perglycaemia and (or) hyperinsulinemia. A similar observa-tion was made in type 1 diabetic rats and ZDF ratsfollowing exercise as PKB Ser473 and Thr308 were differen-tially regulated in the heart (Lajoie et al. 2004; Laviola etal. 2001). The disparate regulation of PKB residues couldbe related to the differential recruitment of various stimuliand (or) signaling mechanisms (Lynch et al. 1999; Persad etal. 2000; Schubert et al. 2000). At least with regard to thespecific regulation of PKB Ser473 phosphorylation, hypergly-caemia may be directly implicated. In skeletal musclebathed in a high glucose medium, PKB Ser473 phosphoryla-tion was significantly diminished (Oku and Asano 2001).Conversely, a hypoglycaemic environment increased PKBSer473 phosphorylation in the brain cortex (Ouyang et al.2000). Furthermore, reduced phosphorylation of PKB Ser473

may represent an early phenotypic event in the type 2 dia-betic heart. The rationale for the latter statement is sup-

Fig. 4. GSK-3a/b phosphorylation in ZDF-diabetic rats. (Panel A) GSK-3a and GSK-3b phosphorylation in ZDF-diabetic-early (ZDF-diabetic-E) and ZDF-diabetic-late (ZDF-diabetic-L) hyperglycaemic rats compared with respective ZDF-lean rats. (Panel B) Total GSK-3bprotein content in ZDF-diabetic-E, ZDF-diabetic-L hyperglycaemic rats, and ZDF-lean rats. (Panel C) Semi-quantitative analysis of GSK-3b phosphorylation normalized to total GSK-3b protein. (Panel D) Semi-quantitative analysis of GSK-3a phosphorylation normalized tototal GSK-3b protein. In ZDF-lean rats, the value was arbitrarily set at 1. ap < 0.05, ZDF-diabetic vs. respective ZDF-lean; bp < 0.05,ZDF-diabetic-E vs. ZDF-diabetic-L.

1210 Can. J. Physiol. Pharmacol. Vol. 84, 2006

# 2006 NRC Canada

ported by a previous study published by our group demon-strating that older ZDF rats with long-standing hyperglycae-mia were associated with the dephosphorylation of bothPKB residues (Lajoie et al. 2004). Indirectly supporting thelatter observation, a significant negative correlation was ob-served between plasma glucose levels and PKB Ser473 phos-phorylation (r = –0.51, p < 0.05), regardless of the onset ofhyperglycaemia.

Following insulin stimulation, PKB activation promotesthe phosphorylation and inactivation of GSK-3 leading toglycogen synthesis. In the present study, regardless of theonset of hyperglycaemia, GSK-3a phosphorylation was de-creased in the heart of ZDF rats, thereby supporting in-creased kinase activity. The decreased phosphorylation ofGSK-3a may be directly related to the incomplete activationof PKB, as increased Thr308 phosphorylation was associatedwith either reduced or unchanged Ser473 phosphorylation. Asreported in the literature, the dual phosphorylation of Thr308

and Ser473 residues are critical to achieve a high level of ac-tivity (Alessi et al. 1996). Likewise, GSK-3b phosphoryla-

tion was decreased in the myocardium of late onsethyperglycaemic ZDF rats that may be related to the incom-plete activation of PKB. By contrast, GSK-3b phosphorylationwas modestly increased in the early onset hyperglycaemicZDF rats. The latter data suggest that the GSK-3a andGSK-3b isoforms can also be differentially regulated. In-deed, in response to voluntary running, increased GSK-3aphosphorylation in the heart was associated with no changein the phosphorylation state of GSK-3b (Gosselin et al.2006).

Previous studies have described a casual association be-tween elevated muscle glycogen content and diminishedPKB and AMPK activity (Derave et al. 2000; Kawanaka etal. 2000; Nielsen et al. 2002; Wojtaszewski et al. 2002). Ithas been well established that hypoglycaemia activatedAMPK and the kinase was inactivated by insulin (Russell etal 2004; Kovacic et al. 2003; Witters and Kemp 1992). Inthe present study, glycogen content was significantly in-creased in the heart of both early and late onset hypergly-caemic ZDF rats. Consistent with the paradigm observed in

Fig. 5. GLUT4 protein and mRNA expression in ZDF-diabetic rats. (Panel A) GLUT4 protein expression in ZDF-diabetic early hypergly-caemic and ZDF-lean rats. (Panel B) GLUT4 protein expression in ZDF-diabetic late hyperglycaemic and ZDF-lean rats. (Panel C) Semi-quantitative analysis of GLUT4 protein expression in ZDF-diabetic-early (ZDF-diabetic-E), ZDF-diabetic-late (ZDF-diabetic-L) hypergly-caemic rats, and ZDF-lean rats. In ZDF-lean rats, the value was arbitrarily set at 1. (Panel D) GLUT4 mRNA expression in ZDF-diabetic-E,ZDF-diabetic-L hyperglycaemic rats, and ZDF-lean rats was determined by real-time PCR. mRNA data normalized to b-actin mRNA. InZDF-lean rats, the value was arbitrarily set at 1. a p < 0.05, ZDF-diabetic vs. respective ZDF-lean rats; b p < 0.05, ZDF-diabetic-E vs. ZDF-diabetic-L.

Lajoie et al. 1211

# 2006 NRC Canada

skeletal muscle, AMPK a Thr172 phosphorylation was like-wise decreased, whereas PKB Ser473 and Thr308 phosphory-lation was discordant. Collectively, these data suggest thatelevated glycogen content and hyperinsulinemia in eitherskeletal or cardiac muscle may directly and (or) indirectlyinfluence AMPK activity, albeit the underlying mechanismremains to be resolved.

As previously demonstrated in experimental diabetes(Pelzer et al. 2005), GLUT4 protein and mRNA expressionin the myocardium were significantly decreased in the earlyonset hyperglycaemic ZDF rats. The latter findings are con-sistent with the premise that the downregulation of GLUT4protein content may represent an adaptive response to limitglucose uptake and limit further glycogen overload in a hy-perglycaemic environment. By contrast, in late onset hyper-glycaemic ZDF rats, GLUT4 mRNA expression wasunchanged compared with ZDF lean rats, whereas proteincontent was significantly increased. It is possible that the in-creased content of GLUT4 in the heart of late onset hyper-glycaemic ZDF rats may reflect decreased proteindegradation related in part to the delayed and slow develop-ment of hyperglycaemia.

The present study has demonstrated that there exists atemporal delay in the onset of hyperglycaemia in the ZDFgenetic model of type 2 diabetes. In the rapid onset hyper-glycaemic ZDF rats, the perturbations regarding the variouskinases implicated in glucose metabolism were widespread.By contrast, in the late onset hyperglycaemic ZDF rats, se-lective changes were observed including the persistent de-phosphorylation of AMPK and GSK-3a. Furthermore, thedisparate pattern of GLUT4 protein and mRNA may repre-sent a seminal feature highlighting the temporal delay asso-ciated with the early and late onset of hyperglycaemia inZDF rats.

ReferencesAlessi, D.R., Andjelkovic, M., Caudwell, B., Cron, P., Morrice,

N., Cohen, P., and Hemmings, B.A. 1996. Mechanism of acti-vation of protein kinase B by insulin and IGF-1. EMBO J. 15:6541–6551. PMID:8978681.

Capes, S.E., Hunt, D., Malmberg, K., Pathak, P., and Gerstein,H.C. 2001. Stress hyperglycemia and prognosis of stroke in non-diabetic and diabetic patients: a systematic overview. Stroke, 32:2426–2432. PMID:11588337.

Chatham, J.C., and Seymour, A.M. 2002. Cardiac carbohydrate meta-bolism in Zucker diabetic fatty rats. Cardiovasc. Res. 55: 104–112.doi:10.1016/S0008-6363(02)00399-1. PMID:12062713.

Clark, J.B., Palmer, C.J., and Shaw, W.N. 1983. The diabeticZucker fatty rat. Proc. Soc. Exp. Biol. Med. 173: 68–75.PMID:6344096.

Cross, D.A., Alessi, D.R., Cohen, P., Andjelkovich, M., and Hem-mings, B.A. 1995. Inhibition of glycogen synthase-3 by insulinmediated by protein kinase B. Nature (London), 378: 785–789.doi:10.1038/378785a0. PMID:8524413.

Derave, W., Hansen, B.F., Lund, S., Kristiansen, S., andRichter, E.A. 2000. Muscle glycogen content affectsinsulin-stimulated glucose transport and protein kinase B ac-tivity. Am. J. Physiol. Endocrinol. Metab. 279: E947–E955.PMID:11052948.

Gosselin, H., Beliveau, L., Burelle, Y., Clement, R., Lajoie, C., El-Helou, V., and Calderone, A. 2006. Disparate regulation of sig-nalling proteins after exercise and myocardial infarction. Med.

Sci. Sports Exerc. 38: 455–462. doi:10.1249/01.mss.0000205138.02440.79. PMID:16540832.

Haq, S., Choukroun, G., Kang, Z.B., Ranu, H., Matsui, T., Ro-senzweig, A., et al. 2000. Glycogen synthase kinase-3b is a ne-gative regulator of cardiomyocyte hypertrophy. J. Cell Biol.151: 117–129. PMID:11018058.

Kandel, E.S., and Hay, N. 1999. The regulation and activities ofthe multifunctional serine/threonine kinase Akt/PKB. Exp.Cell Res. 253: 210–229. doi:10.1006/excr.1999.4690.PMID:10579924.

Kawanaka, K., Nolte, L.A., Han, D.H., Hansen, P.A., and Holloszy,J.O. 2000. Mechanisms underlying impaired GLUT-4 transloca-tion in glycogen-supercompensated muscles of exercised rats.Am. J. Physiol. Endocrinol. Metab. 279: E1311–E1318.PMID:11093919.

Kovacic, S., Soltys, C.L., Barr, A.J., Shiojima, I., Walsh, K., andDyck, J.R.B. 2003. Akt activity negatively regulates phosphory-lation of AMP-activated protein kinase in the heart. J. Biol.Chem. 278: 39422–39427. doi:10.1074/jbc.M305371200.PMID:12890675.

Lajoie, C., Calderone, A., Trudeau, F., Lavoie, N., Massicotte, G.,Gagnon, S., and Beliveau, L. 2004. Exercise training attenuatedthe PKB and GSK-3 dephosphorylation in the myocardium ofZDF rats. J. Appl. Physiol. 96: 1606–1612. doi:10.1152/japplphysiol.00853.2003. PMID:14698990.

Laviola, L., Belsanti, G., Davalli, A.M., Napoli, R., Perrini, S., Weir,G.C., Giorgino, R., and Giorgino, F. 2001. Effects of streptozocindiabetes and diabetes treatment by islet transplantation on in vivoinsulin signaling in rat heart. Diabetes, 50: 2709–2720.PMID:11723053.

Lo, S., Russell, J.C., and Taylor, A.W. 1970. Determination of gly-cogen in small tissue samples. J. Appl. Physiol. 28: 234–236.PMID:5413312.

Lynch, D.K., Ellis, C.A., Edwards, P.A., and Hiles, I.D. 1999.Integrin-linked kinase regulates phosphorylation of serine 473of protein kinase B by an indirect mechanism. Oncogene, 18:8024–8032. PMID:10637513.

Nguyen, Q.T., Colombo, F., Clement, R., Gosselin, H., Rouleau,J.L., and Calderone, A. 2003. AT1 receptor antagonist therapypreferentially ameliorated right ventricular function and pheno-type during the early phase of remodeling post-MI. Br. J.Pharmacol. 138: 1485–1494. doi:10.1038/sj.bjp.0705212.PMID:12721104.

Nielsen, J.N., Vissing, J., Wojtaszewski, J.F., Haller, R.G., Begum,N., and Richter, E.A. 2002. Decreased insulin action in skeletalmuscle from patients with McArdle’s disease. Am. J. Physiol.Endocrinol. Metab. 282: E1267–E1275. PMID:12006356.

Oku, A., and Asano, T. 2001. Inhibitory effect of hyperglycemia oninsulin-induced Akt/protein kinase B activation in skeletal mus-cle. Am. J. Physiol. 280: E816–E824.

Ouyang, Y.B., Zhang, X.H., He, Q.P., Wang, G.X., Siesjo, B.K.,and Hu, B.R. 2000. Differential phosphorylation at Ser473 andThr308 of Akt-1 in rat brain following hypoglycemic coma.Brain Res. 876: 191–195. doi:10.1016/S0006-8993(00)02618-4.PMID:10973608.

Pelzer, T., Jazbutyte, V., Arias-Loza, P., Segerer, S., Lichtenwald,M., Law, M., Ertl, G., and Neyses, L. 2005. Pioglitazone reversesdown-regulation of cardiac PPARg expression in Zucker diabeticfatty rats. Biochem. Biophys. Res. Commun. 329: 726–732.doi:10.1016/j.bbrc.2005.02.029. PMID:15737646.

Persad, S., Attwell, S., Gray, V., Delcommenne, M., Troussard, A.,Sanghera, J., and Dedhar, S. 2000. Inhibition of integrin-linedkinase (ILK) suppresses activation of protein kinase B/Akt andinduces cell cycle arrest and apoptosis of PTEN-mutant prostate

1212 Can. J. Physiol. Pharmacol. Vol. 84, 2006

# 2006 NRC Canada

cancer cells. Proc. Natl. Acad. Sci. U.S.A. 97: 3207–3212.doi:10.1073/pnas.060579697. PMID:10716737.

Pold, R., Jensen, L.S., Jessen, N., Buhl, E.S., Schmitz, O., Flyvb-jerg, A., et al. 2005. Long-term AICAR administration and exer-cise prevents diabetes in ZDF rats. Diabetes, 54: 928–934.PMID:15793229.

Russell, R.R., III, Li, J., Coven, D.L., Pypaert, M., Zechner, C.,Palmeri, M., et al. 2004. AMP-activated protein kinase mediatesischemic glucose uptake and prevents postischemic cardiac dys-function, apoptosis, and injury. J. Clin. Invest. 114: 495–503.doi:10.1172/JCI200419297. PMID:15314686.

Schubert, K.M., Scheid, M.P., and Duronio, V. 2000. Ceramide in-hibits protein kinase B/Akt by promoting dephosphorylation ofserine 473. J. Biol. Chem. 275: 13330–13335. doi:10.1074/jbc.275.18.13330. PMID:10788440.

Slieker, L.J., Sundell, K.L., Heath, W.F., Osborne, H.E., Bue, J.,Manetta, J., and Sportsman, J.R. 1992. Glucose transporter levelsin tissues of spontaneously diabetic Zucker fa/fa rat (ZDF/drt)and viable yellow mouse (Avy/a). Diabetes, 41: 187–193.PMID:1733808.

Sowers, J.R., Epstein, M., and Frohlich, E.D. 2001. Diabetes, hy-pertension, and cardiovascular disease. Hypertension, 37:1053–1059 PMID:11304502.

Winder, W.W., and Hardie, D.G. 1999. AMP-activated protein ki-nase, a metabolic master switch: possible roles in type 2 dia-betes. Am. J. Physiol. 277: E1–10. PMID:10409121.

Witters, L.A., and Kemp, B.E. 1992. Insulin activation of acetyl-CoA carboxylase accompanied by inhibition of the 5’-AMP-activated protein kinase. J. Biol. Chem. 267: 2864–2867.PMID:1346611.

Wojtaszewski, J.F., Jorgensen, S.B., Hellsten, Y., Hardie, D.G., andRichter, E.A. 2002. Glycogen-dependent effects of 5-aminoimidazole-4-carboxamide (AICA)-riboside on AMP-activated protein kinase and glycogen synthase activities inrat skeletal muscle. Diabetes, 51: 284–292. PMID:11812734.

Woodgett, J.R. 1990. Molecular cloning and expression of glyco-gen synthase kinase-3/factor A. EMBO J. 9: 2431–2438.PMID:2164470.

Lajoie et al. 1213

# 2006 NRC Canada