Biochimica et Biophysica Acta 1783 (2008) 2344–2351

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

j ourna l homepage: www.e lsev ie r.com/ locate /bbamcr

Calreticulin regulates insulin receptor expression and its downstream PI3 Kinase/Aktsignalling pathway

Shahrzad Jalali, Maliheh Aghasi, Behzad Yeganeh, Nasrin Mesaeli ⁎Institute of Cardiovascular Sciences, St. Boniface General Hospital Research Centre 351 Tache Avenue, Winnipeg, Manitoba, Canada R2H 2A6Department of Biochemistry and Medical Genetics, Faculty of Medicine, University of Manitoba 351 Tache Avenue, Winnipeg, Manitoba, Canada R2H 2A6

⁎ Corresponding author. Institute of CardiovascularHospital Research Centre 351 Tache Avenue, WinnipegTel.: +1 204 235 3671; fax: +1 204 231 1151.

E-mail address: nmesaeli@sbrc.ca (N. Mesaeli).

0167-4889/$ – see front matter © 2008 Elsevier B.V. Aldoi:10.1016/j.bbamcr.2008.08.014

a b s t r a c t

a r t i c l e i n f o

Article history:

Defects in insulin signalling Received 25 April 2008Received in revised form 27 August 2008Accepted 28 August 2008Available online 17 September 2008

Keywords:CalreticulinInsulin receptorAktGlucose uptakePI3 kinasep53

and glucose metabolism are associated with the development of diabetes. Insulinsignalling is initiated by the binding of insulin to its receptor and triggering cascades of events includingactivation of PI3kinase/Akt signalling pathway. Calreticulin (CRT) is a calcium binding chaperone moleculelocated in the endoplasmic reticulum. Targeted deletion of CRT in mice is embryonic lethal as a result ofdevelopmental and metabolic abnormalities. Rescued CRT null mice develop severe hypoglycemia the reasonfor which is not known. In addition, ventricular cardiomyocytes isolated from CRT null (crt−/−) mice haveincreased glycogen deposits. Therefore, the aim of this study was to investigate the changes in the glucoseuptake and insulin signalling pathway (mainly PI3 kinase/Akt) in the absence of CRT. Here we show asignificant increase in the glucose uptake by the crt−/− cells. This increase was accompanied by a significantincrease in both insulin receptor β expression, Insulin receptor substrate-1 phosphorylation, GLUT-1expression and in insulin stimulated Akt phosphorylation and kinase activity in the crt−/− cells. Intriguingly,the increased expression of insulin receptor β in the crt−/− was due to decreased levels of p53 protein. Thecurrent study is the first evidence for the up-regulation of insulin receptor density and activity in the absenceof CRT function.

© 2008 Elsevier B.V. All rights reserved.

1. Introduction

The pathogenesis of type 2 diabetes results from the inability ofcells, particularly skeletal muscle cells, to uptake glucose in an insulindependent manner. Insulin mediated glucose uptake is facilitated viaactivation of insulin receptor thus activating insulin receptorsubstrate-1 (IRS-1)-dependent phosphatidylinositol 3-kinase (PI3Kinase) [34]. Receptor mediated stimulation of PI-3 kinase inducesphosphatidylinositol phosphorylationwhich facilitate translocation ofAkt to the plasma membrane. Akt, also known as protein kinase B, is aserine/threonine kinase involved in regulating cell survival, prolifera-tion, apoptosis and insulin response [3]. Translocation of Akt to the cellmembrane results in its phosphorylation at threonine-308 (T308)residue by PDK-1 [38] and the serine-473 (S473) residue. The identityof the kinase responsible for S473 phosphorylation has long beenunder debate [26,36,40]. However, recently the rictor-mTOR complexwas shown to directly phosphorylate Akt at S473 in vivo [29]. Akt inturn phosphorylates many target proteins including GSK3β [6,8].

Sciences, St. Boniface General, Manitoba, Canada R2H 2A6.

l rights reserved.

Calreticulin (CRT) is an endoplasmic reticulum resident proteinwith several biological functions including being a chaperone ofnascent glycoproteins and a regulator of Ca2+ homeostasis [19]. Genetargeted deletion of CRT results in embryonic lethality [18] whichcould be partially rescued by crossing thesemicewith transgenic miceoverexpressing activated calcineurin B in the heart [9]. However, therescued mice are growth retarded and present with significanthypoglycemia which is of unknown origin [9]. A role for CRT hasalso been demonstrated in the folding of insulin receptor [2] and thestability of glucose transporter-1 (GLUT-1) mRNA [35]. Furthermore,ultrastructural analysis of the ventricular cardiomyocytes of the CRTknockout embryos has shown significant increase in glycogendeposition as compared to wild type controls [15]. This build-up ofglycogen suggests that in the absence of CRT the insulin signalling andglucose metabolism are changed. One could then postulate that loss ofCRT function could alter insulin receptor density and activity leadingto hypoglycemia. In addition to change in glucose uptake, activation ofinsulin receptor can lead to activation of the PI3 kinase/Akt pathwayleading to activation of Mdm-2 [8]. Akt phosphorylates Mdm-2 on twoserine residues (S166 and S186) resulting in its translocation to thenucleus [1,16]. In the nucleus Mdm-2 can inhibit p53 function viainterfering with its transcriptional activities and inducing p53ubiquitination thereby targeting it for proteasome degradation[7,25]. Previously, we have shown increased nuclear localization of

Table 1Primer list used for RT-PCR in the current study

Gene Primer sequence

IR-F 5′-CATTCAGGAAGACCTTCGAGG-3′IR-R 5′-TGTGGTGGCTGTCACATTCC-3′GLUT-1-F 5′-TCATGTTGGCTGTGGGAGGA-3′GLUT-1-R 5′-GGCTCTCCGATGCGGTGGTT-3′GLUT-4-F 5′-TTCCAGCAGATCGGCTCTG-3′GLUT-4-R 5′-GTCCCCCAGGACCTTGCCT-3′GAPDH-F 5′-GTTGCCATCAACGACCCCTTC-3′GAPDH-R 5′-CTCCATGGTGGTGAAGACACC-3′

All primers were designed using the published mouse sequences.

Fig. 1. Changes in insulin receptor β expression level in cells with altered CRTexpression. (A) A representative western blot for insulin receptor β (IR β) subunit inwild type (wt) and calreticulin deficient (crt−/−) MEF cells. Bar graph shows a significantincrease in the ratio of insulin receptor β subunit to actin in the crt−/− cells as comparedto the wt cells. (B) RT-PCR showing the mRNA expression level of IR and GAPDH(internal control) in wt and crt−/− cells. Bar graph represents a significant increase inthe ratio of IR to GAPDH mRNA level in the crt−/− cells as compared to the wt cell. Datapresented are mean±SE of 4–5 different experiments. ⁎pb0.05, significantly differentfrom the wt cells.

2345S. Jalali et al. / Biochimica et Biophysica Acta 1783 (2008) 2344–2351

Mdm-2 and decreased level of p53 protein in crt−/− cells [17].However, no data is available on the changes in activity of insulinreceptor or its downstream PI3 kinase/Akt signalling pathways uponaltered CRT expression (or function). Therefore, the aim of this studywas to investigate the changes in insulin signalling and its down-stream PI3 kinase/Akt pathway in the absence of CRT.

2. Materials and methods

2.1. Materials

PI3 kinase inhibitors, wortmannin, Ly294002 and MG132 werepurchased from Calbiochem (EMD Biosciences Inc., USA). Antibodiesused were actin (Sigma-Aldrich, St. Louise, USA), GSK3β and insulinreceptor (β-subunit) (Chemicon International Inc., USA). Akt kinaseassay and all other antibodies were from Cell Signalling Technology(Pickering, ON, Canada). Serine/threonine and tyrosine phosphataseassay kits were fromMolecular Probes (Invitrogen, Canada). Enhancedgreen fluorescent protein-p53 (EGFP-p53) was purchased fromClontech Laboratories (California, USA).

2.2. Cell culture

Wild type (wt) and calreticulin deficient (crt−/−) mouse embryonicfibroblast (MEF) cells were isolated from 14-day oldmouse embryos asdescribed previously [17]. 106 cells were plated in Dulbecco's modifiedEagle's medium (DMEM) containing 10% FBS. 24 h after plating, thecells were starved by first washing with PBS then incubation in DMEMcontaining 0.5% bovine serum albumin for 14–16 h. Cells were thenstimulated for 10 minwith DMEM alone or DMEM containing 100 nMInsulin. To determine the role of PI3 kinase in the observed responses,cells were incubated with 200 nM wortmannin or 10 μM LY294002 inDMEM1 h prior to stimulationwith insulin. Cells were lysedwith RIPAbuffer (containing 250mMNaCl, 20 mM Tris, 1 mM EDTA,1 mM EGTA,1 mM sodium vanadate, 30 mM sodium fluoride, 10 mM tetrasodiumpyrophosphate, 0.5% sodium deoxycholate, 1% non-iodet p-40, 1%Triton X-100; 0.5 μM PMSF and a protease inhibitor cocktail) followedby centrifugation at 7000 rpm for 10min. The supernatants were usedfor protein assay and western blot analysis.

2.3. Transient transfection

wt and crt−/− cells were transiently transfected with 10 μg of aplasmid encoding CRT-HA (described in [17]) or EGFP-p53 usingLipofectamine 2000. 48 h later, cells were lysed in RIPA buffercontaining protease inhibitors. 30 μg protein of total cell lysate wereused for western blot.

2.4. Western blot

Cell lysates containing 30 μg protein were separated on a 10% SDS-polyacrylamide gel and blotted to nitrocellulose membrane. Themembrane was blocked with 5% milk powder in Tris-buffered saline

(TBS) containing 0.1% Tween-20. The primary antibodieswere diluted inappropriate dilution buffer as recommended by the manufacturer.Cytoskeletal actin was used as a loading control for each blot. HRP-conjugated secondary antibodies (Jackson Laboratories) were usedfollowed by enhanced chemiluminescence solutions. The bands corre-sponding to each specific protein and actin were captured by Fluor-SMax (Bio-Rad) andquantified usingBio-RadQuantityOne software. Datawere presented as the ratio of the specific proteins to the level of actin.

2.5. Akt kinase assay

Akt kinase activity was measured using an in vitro Akt kinase assay(Cell Signalling Technology, Inc.) following the manufacturer'sinstruction. Briefly, cells were lysed using 1 ml lysis buffer supple-mented with 1 mM PMSF. Cell lysates containing 500 μg protein and20 μl immobilized Akt antibody were incubated 3 h at 4 °C, and thencentrifuged at 7000 rpm for 1 min. After washing with lysis buffer andkinase buffer, the pellets were suspended in 40 μl kinase buffercontaining 200 μMATP and 1 μg GSK3 fusion protein. Themixture wasincubated at 30 °C for 30 min and the reaction was terminated byadding 3× SDS sample buffer. Samples were resolved on SDS-PAGE gel

Fig. 2. Insulin receptor is activated in the absence of CRT. (A) Glucose uptake in wt andcrt−/− cells was measured in the presence and absence of insulin (0.1 or 1 μM). Cellswere starved in DMEM overnight and then incubated with a buffer containing 3Hdeoxy-D-glucose with or without insulin (0.1 or 1 μM) for 10 min. Data presented aremean±SE of pmole 3H-glucose uptake/mg protein, from three separate experimentscarried out in duplicates. ⁎pb0.05, significantly different from the DMEM. †pb0.05,significantly different from the corresponding treated wt cells. (B) Glycogen content issignificantly increased in CRT deficient cells as compared to wt cells. Glycogen contentwas determined using a modified phenol-sulfuric acid colorimetric technique asdescribed in “Materials and methods”. Total glycogen in cell lysate was then calculatedusing a glycogen standard curve. The values were then normalized to the proteincontent in each sample and presented as μg glycogen/μg protein. ⁎pb0.05, significantlydifferent from wt cells.

Fig. 3. Expression of GLUT1 and GLUT-4 in CRT deficient cells. (A) A representative RT-PCR (upper panel) and western blot (lower panel) of four independent experimentsshowing no change in GLUT-4 mRNA and protein in the absence of CRT. (B) Arepresentative RT-PCR showing GLUT-1 mRNA expression in crt−/− cells as compared towt cells. Bar graph shows mean±SE of ratio of GLUT-1 to GAPDH, from four differentexperiments. (C) A representative western blot showing changes in GLUT-1 proteinexpression in crt−/− cells as compared to wt cells. Bar graph shows mean±SE of ratio ofGLUT-1 to GAPDH, from three different experiments. ⁎pb0.05, significantly differentfrom the wt.

2346 S. Jalali et al. / Biochimica et Biophysica Acta 1783 (2008) 2344–2351

followed by western immunoblotting using an anti-phospho-GSK3α/β antibody (1:1000 dilution).

2.6. Glucose uptake assay

To measure glucose uptake, wt and crt−/− cells were starvedovernight in DMEM. Next day, the cells were washed twice with 1 mlKrebs-Ringer HEPES (KRH) buffer (121 mM NaCl, 4.9 mM KCl, 1.2 mMMgSO4, 0.33 mM CaCl2, and 12 mMHEPES, pH 7.4) and incubated witha cocktail of KRH buffer containing 0.1 mM deoxy-D-glucose and1.0 μCi/ml deoxy-D-glucose, 2-[1,2-3H (N)] in either the presence orabsence of 100 nM insulin at 37 °C for 30min. Subsequently, cells werewashed with 1 ml cold KRH buffer containing 25 mM D-glucose andlysed in 300 μl digitonin release buffer (0.25Mmannitol,17mMMOPS,2.5 mM EDTA and 0.25 M digitonin). Cell lysates were centrifuged at7000 rpm for 5 min and 100 μl of the supernatant was utilized todetermine the radioactivity in each sample using scintillation counter.The amount of 3H-glucose in each sample was divided by proteinconcentration and data is presented as pmole/mg protein.

2.7. Glycogen assay

To measure changes in cellular glycogen level upon loss of CRTfunction a modified phenol-sulfuric acid colorimetric method was

used as described by Lo et al. [14]. Briefly, wt and crt−/− cells weregrown to 80% confluency. Cells were lysed using RIPA buffer anddigested using an alkaline digestion buffer (30% KOH saturated withNa2SO4) followed by 10 min boiling. Total glycogen was thenprecipitated using 95% ethanol. Glycogen was then re-dissolved indistilled water. 5% phenol was then added to each tube followed byaddition of 96% sulforic acid and incubation on ice for 30 min. Theabsorbance of each sample was then read using a microplate reader at500 λ. The concentration of glycogen was then calculated using aglycogen standard curve (concentration range 12–50 μg/ml). Protein

Fig. 4. Insulin mediated phosphorylation of Insulin Receptor Substrate-1 (IRS-1) iselevated in the crt−/− cells. (A) Upper panel shows a representative western blot for p-Insulin Receptor Substrate-1 (p-IRS-1) in wt and crt−/− cells. Cells were serum starvedovernight followed by incubation with DMEM containing 100 nM insulin for 10 min.Cell lysates were then used for western blot with an antibody specific for p-IRS-1 (Tyr941). Bar graph of mean±SE of ratio of p-IRS-1 band intensity to actin, from threedifferent experiments. (B) A representative western blot showing increased IRS-1expression in crt−/− cells as compared towt cells. Bar graph is mean±SE of ratio of IRS-1to actin intensity, from four independent experiments. ⁎pb0.05, significantly differentfrom the wt.

Fig. 5. Akt kinase activity in the crt−/− cells. Upper panel shows a representativewesternblot for p-GSK3α/β demonstrating Akt kinase activity in the wild type (wt) and crt−/−cells. Cells were serum starved for 14–16 h followed by incubation with DMEM alone orDMEMwith 100 nM Insulin for 10 min. Cell lysate was then used for immunoprecipitat-ing Akt and subsequent Akt kinase assay as described in the “Materials and methods”.Bar graphs show the mean±SE of four independent experiments. The intensity of p-GSK3β band from the crt−/− cells is presented as percentage of the wt. ⁎pb0.05,significantly different from the wt.

2347S. Jalali et al. / Biochimica et Biophysica Acta 1783 (2008) 2344–2351

concentration in each sample was also measured and the glycogenconcentration was normalized to μg protein in each sample. Eachexperiment was carried out in duplicates and the assay was repeatedwith 7 different cell preparations.

2.8. Reverse transcriptase assay

To determine the level of insulin receptor transcription, total RNAwas isolated from the wt and crt−/− cells using RNeasy Mini Kit(Qiagen). Isolated RNAs were subject to First-Strand cDNA synthesisusing SuperScript® II Reverse Transcriptase (Invitrogen), followed byPCR amplification for different genes (insulin receptor, GLUT-1, GLUT-4 and GAPDH) using primers listed in Table 1. In preliminaryexperiments the optimum Tm and cycle number for each gene PCRamplificationwere determined. PCR for GAPDH, a housekeeping gene,was carried out using the same cDNA as an internal control fordifferent samples. An aliquot of the PCR reaction was resolved on a 2%agarose gel and visualized using the Bio-Rad Gel-Doc system. Thedensity of the bands corresponding to insulin receptor and GAPDHwere quantified using a Bio-Rad Quantity One Program (Cambridge,MA). Data presented as the ratio of insulin receptor to GAPDH level(mean±SE of three separate mRNA preparations).

3. Results

3.1. Insulin receptor expression and function in crt−/− cells

A recent study showed that CRT deficient mice that were rescuedfrom embryonic lethality by overexpression of activated calcineurin inthe heart, suffer from severe hypoglycemia [9]. Furthermore, glycogendepositionwas significantly higher in CRT deficient mice as comparedto controls [15]. To determine the cause of this significant decrease inserum glucose level and glycogen deposition, we examined theexpression of insulin receptor in MEF cells isolated from the crt−/− andwtmice. Fig.1A, demonstrates that there is a significant increase in theinsulin receptor β subunit protein level in the crt−/− cells as comparedto the wt cells. Furthermore, this increase in insulin receptor βexpression is due to a significant increase in its mRNA level (Fig.1B). Toinvestigatewhether the increase in insulin receptor β expression upondeletion of CRT alters insulin receptor activity, we measured 3H-glucose uptake in the wt and crt−/− cells. As shown in Fig. 2A, insulininduced a significant increase in glucose uptake in both wt and crt−/−cells. However, the level of glucose uptake was significantly higher inthe crt−/− cells following insulin stimulation as compared to theinsulin treatedwt cells (Fig. 2A). Fig. 2B shows that the higher glucoseuptake in crt−/− cells was accompanied by a significant increase inglycogen deposition in these cells when compared to similarlyculturedwt cells. Intriguingly, the non-treated crt−/− cells also showeda slightly (but not significantly) higher basal level of glucose uptake ascompared to the non-treated wt cells (Fig. 2A, DMEM). Therefore, weexamined the expression of glucose transporters GLUT-1, GLUT-3 andGLUT-4 in wt and crt−/− cells. As shown in Fig. 3A, there was nodifference in the protein and mRNA level of GLUT-4 in wt and crt−/−cells. However, GLUT-1 mRNA (Fig. 3B) and protein (Fig. 3C)expression was significantly higher in crt−/− cells as compared to wtcells. Moreover, several attempts to detect GLUT-3 mRNA by RT-PCRfailed to show any message in both wt and crt−/− MEF cells (data notshown). Similarly, using western blot analysis we did not detect anyGLUT-3 protein in these cells.

Fig. 6. Effect of Insulin on Akt phosphorylation in the wt and crt−/− cells. (A) Upperpanel shows a western blot with antibodies specific for p-Akt (S473), non-phosphorylated Akt, and actin (loading control). Cells were serum starved for 14–16 hfollowed by incubation with DMEM alone, or DMEM containing 100 nM Insulin for10 min. Bar graph shows mean±SE of ratio of p-Akt band intensity to actin bandintensity, from four different experiments. ⁎pb0.05, significantly different from the wt.(B) A representative western blot carried out with antibody specific for the T308 p-Akt,and non-phosphorylated Akt. Cells were serum starved then treated with DMEM orInsulin. (C)Western blot shows that insulin induced phosphorylation of Akt at S473wasinhibited following pre-treatment of cells with wortmannin or LY294002. Cells wereserum starved and treated with 200 nM wortmannin or 10 μM LY294002 for 1 h thentreated with insulin in the presence of inhibitors for an additional 10 min.

Fig. 7. Insulin mediated phosphorylation of GSK3β in vivo. (A) The level of p-GSK3βwassignificantly increased in crt−/− cells as compared to wt cells. Top panel is arepresentative western blot with a p-GSK3β (S9) antibody. Lower panel is a bar graphof mean±SE of ratio of p-GSK3β band intensity to actin band intensity, from threedifferent experiments. ⁎pb0.05, significantly different from the insulin treated wt cells.(B) Western blot analysis with GSK3β specific antibody showed no significant change inthe endogenous level of GSK3β.

2348 S. Jalali et al. / Biochimica et Biophysica Acta 1783 (2008) 2344–2351

Binding of insulin to its receptor induces phosphorylation of itsdownstream signalling molecule insulin receptor substrate 1 (IRS-1)at tyrosine 941 (for review see [11]). Thus, we determined the level oftyrosine 941 phosphorylation of IRS-1 as another measure of insulinreceptor function. As shown in Fig. 4A, there was a significant (40%)increase in IRS-1 phosphorylation in the absence of CRT. The increasedphosphorylation of IRS-1 was accompanied by a significant increase in

IRS-1 protein expression in crt−/− cells (Fig. 4B). These data suggeststhat increased insulin receptor expression in crt−/− cells leads toactivation of its downstream signalling pathway, the PI3/Akt pathway.

3.2. Akt kinase activity and phosphorylation in crt−/− cells

Insulin can stimulate the PI3 kinase/Akt signalling pathway uponbinding to the insulin receptor. To test the effect of insulin on Aktactivation, wt and crt−/− cells were starved for 14–16 h and thenstimulated with DMEM alone, or DMEMwith 100 nM Insulin. As shownin Fig. 5 (upper panel), Akt kinase activity was not detectable in cellstreatedwith DMEM alonewhereas insulin stimulated Akt kinase activityin thewtandcrt−/− cells. The insulin inducedAkt activitywas significantlyhigher in the crt−/− cells as compared to the wt cells (Fig. 5).

Akt activation is mediated via its phosphorylation at a serine residue(S473) in the kinase domain and at a threonine residue (T308) in theregulatory domain [30]. Therefore, to test if the observed changes in theAktkinase activityweredue to increasedproteinphosphorylationordueto increased Akt protein level itself, we performedwestern blot analysis.Fig. 6A and B show that therewere no significant differences in the levelof total Akt between the wt and crt−/− cells. Western blot analysis withphospho-specific antibodies to Akt showed that insulin inducedphosphorylation of Akt at the two key residues S473 and T308 in bothcell types (Fig. 6A and B). However, insulin stimulated a more robustphosphorylation of Akt in crt−/− cells as compared towt cells (Fig. 6A andB). These results indicate that the increased Akt kinase activity observedin the crt−/− cells was due to increased phosphorylation of Akt ratherthan an increased level of Akt protein.

Akt activation can occur via PI3 kinase dependent or PI3 kinaseindependent mechanisms. Thus, we examined the effect of PI3 kinaseinhibition on the insulin mediated increase in Akt phosphorylationobserved in the crt−/− cells. As shown in Fig. 6C, both wortmannin andLY294002 treatment were able to completely abolish Akt phosphoryla-tion at S473 following insulin stimulation in both crt−/− andwt cells.

2349S. Jalali et al. / Biochimica et Biophysica Acta 1783 (2008) 2344–2351

3.3. Insulin stimulated phosphorylation of GSK3β is enhanced in crt−/−cells

Phosphorylation of GSK3β by Akt inhibits its function [12]. Todeterminewhether theelevatedAkt activity can affect the level of GSK3βphosphorylation in vivo, cell lysates from insulin treated andnon-treatedcells were used for immunoblotting. Insulin stimulation increasedGSK3β phosphorylation in both the wt and crt−/− cells (Fig. 7A).

Furthermore, this increase in phosphorylated GSK3β was significantlyhigher in the crt−/− cells as compared to thewt cells (Fig. 7A). The changein the in vivo phosphorylation level of GSK3β corresponds to theincreased Akt kinase activity (Fig. 5). Altered CRT expression has beenshown to affect the expression of different proteins such as calnexin, BiP[13], p53 [17] and bradykinin receptor [22]. Thus, to determine whetherthe observed increase in the in vivo phosphorylated GSK3β is due tochanges in its expression or phosphorylationwe carried outwestern blotanalysis with a non phospho-specific antibody to GSK3β. As shown inFig. 7B, therewasno significant change in the level of totalGSK3βproteinbetween the crt−/− andwt cells.

3.4. Altered CRT and p53 expression results in changes in insulin receptorβ protein levels

To determine the mechanism of increased insulin receptor βexpression (at mRNA and protein levels) in the crt−/− cells weexamined the ability of exogenously expressed CRT to reverse thiseffect. wt and crt−/− cells were transiently transfected with CRT-HAexpression plasmid. Western blot for HA confirms the expression ofexogenous CRT in the transfected cells (Fig. 8A, HA), albeit the level ofHA in the CRT-HA-crt−/− cells was higher as compared to the CRT-HA-wt cells. However, the level of expression of CRT in the CRT-HAtransfected crt−/− cells was still significantly less than the non-transfected wt cells (Fig. 8A, CRT). Fig. 8A (IR β antibody) alsodemonstrates that increasing CRT level decreases insulin receptor βexpression in both the wt and crt−/− cells. We previously showed thatloss of CRT function results in a significant decrease in p53 proteinlevel and its function [17] which can be prevented upon inhibition ofproteasome activity (Uvarov and Mesaeli unpublished observations).Furthermore, p53 has been shown to inhibit insulin receptorexpression by suppressing its promoter activity [39]. Thus weexamined whether decreased levels of p53 in the crt−/− cells couldresult in increased insulin receptor expression. To investigate theeffect of increased p53 protein level, CRT deficient cells weretransiently transfected with EGFP-p53 or treated with proteasomeinhibitors and insulin receptor β expressionwasmeasured by westernblot analysis. As shown in the Fig. 8B, transient transfection of crt−/−cells with EGFP-p53 resulted in a significant decrease in the insulinreceptor β protein level. Furthermore, Fig. 8C shows that treating thecrt−/− cells with MG132 (25 μg/ml for 8 h) results in a significantincrease in the endogenous level of p53 proteinwhich is accompaniedby a significant decrease in the insulin receptor β protein expression.

4. Discussion

Normal glucose metabolism is facilitated by signalling pathwaysinitiated by insulin [34]. As a result, any alteration at the level ofinsulin or defect in insulin signalling can potentially contribute to thedevelopment of diabetes. The present study is the first to demonstratethat CRT function is important in the regulation of glucose uptake andinsulin signalling pathway. In the current study we utilized MEF cells

Fig. 8. Expression of insulin receptor β (IR β) in the crt−/− cells is regulated by increasedCRT or p53 expression. (A) A representative western blot showing the expression of IR βin the cells transiently transfected with CRT. wt and crt−/− cells were transfected withpcDNA-CRT-HA or pcDNA empty vector (controls). Cell lysates were utilized for westernblot analysis with IR β, HA, CRT and actin antibody. Bar graph is mean±SE of ratio of IRβband intensity to actin band intensity (from 4 separate experiments). (B) Arepresentative western blot showing the effect of p53 transfection on the level of IR βprotein expression in the crt−/− cells. wt and crt−/− cells were transiently transfectedwith either EGFP or EGFP-p53 containing plasmid constructs. Cell lysates were used forwestern blot analysis with IRβ, actin and p53. Bar graph is mean ± SE of ratio of IRβband intensity to actin band intensity (from 4 separate experiments). (C) Arepresentative western blot (n=2) shows the effect of proteasome inhibition on IRβand p53 protein expression level. The wt and crt−/− cells were treated with 20 μg/mlMG132 for 8 h at 37 °C. Cells lysate were then used for western blot analysis withantibodies to IR β, p53 and actin.

2350 S. Jalali et al. / Biochimica et Biophysica Acta 1783 (2008) 2344–2351

derived from 13.5 day old embryos of wt and CRT knockout mice.Although these cells are not the typical insulin sensitive cell type (e.g.skeletal muscle or adipocytes) but here we show that they do expressinsulin receptor β, IRS-1, GLUT-1 and GLUT-4 (Figs. 2–4). Our resultsdemonstrate that targeted deletion of CRT significantly induces insulinmediated glucose uptake (Fig. 2A). This observation could explain thereported severe hypoglycemia observed in the rescued CRT null mice[9] and increased glycogen deposition in the myocardium of CRT nullembryos [15]. Indeed, our results in Fig. 2B show a significant increasein glycogen content of crt−/− cells. Furthermore, we illustrated that thisincreased responsiveness to insulin was due to increased insulinreceptor β expression (Fig. 1), increased IRS-1 expression (Fig. 4B) andphosphorylation (Fig. 4A), and elevated Akt phosphorylation andfunction (Fig. 5–6). These changes were not due to the alteredexpression level of Akt in the crt−/− cells (Fig. 6A). We alsodemonstrated that increasedAkt phosphorylation following treatmentwith insulinwas PI3 kinase dependent as this increase was completelyblocked by two different PI3 kinase inhibitors; wortmannin orLY294002 (Fig. 6C). The increased activity of Akt in the crt−/− cellsleads to a higher level of phosphorylation of its endogenous substrateGSK3β upon stimulation with insulin as demonstrated in Fig. 7. This isin accordance with the data showing the increased accumulation ofglycogen in crt−/− cells (Fig. 2B) and the ventricular cardiomyocytes ofthe CRT knockout embryos [15]. In fact, insulin mediated phosphor-ylation of GSK3β has been shown to activate glycogen synthase andincreased glycogen accumulation [33].

Insulin activates the PI3 kinase/Akt signalling pathways [5] bybinding to the insulin receptor and inducing phosphorylation of IRS-1[4]. Our data demonstrate that loss of CRT function results in a significantincrease in the expression of insulin receptor β (Fig. 1). Many oligomericmembrane receptors including insulin receptor undergo subunit foldingand assembly in the lumen of endoplasmic reticulum before they aretransported through the golgi apparatus to the cell membrane [2,28].The process ofmaturation and folding recruits a number of endoplasmicreticulum chaperones specifically the lectin-like chaperones (calnexinand CRT) [10,19]. Previously, Bass et al., demonstrated a requirement forcalnexin/CRT function as lectin-like chaperones and Grp78 (non-lectinchaperone) in the proper folding of the insulin receptor [2]. Morerecently Saitoh et al., showed that the monomeric form of the insulinreceptor precursor binds to calnexin alone and the function of HSP90 isnecessary for insulin receptor dimerization and function [27]. Calnexinand CRT are known to work synchronously as lectin-like chaperones toensure the proper folding of N-glucosylated proteins in the lumen ofendoplasmic reticulum [10]. Thus loss of CRT function might affectfolding of these proteins. However, in crt−/− cells the endoplasmicreticulum levels of both calnexin and Grp78 are significantly increased[13,37]. Furthermore, our data showed that in the crt−/− cells theincrease in insulin receptor β subunit expression (Fig. 1A) wasaccompanied by a similar increase in the mRNA of insulin receptor β(Fig. 1B), and these cells demonstrated higher insulin receptor function(Figs. 2 and 4). All together, these data exclude the possibility of directinvolvement of CRT chaperone function in insulin receptor folding,maturation and function.

Our data also demonstrated that the increase in the insulinreceptor βwas due to decreased level of CRT protein (Fig. 8A). Indeed,transient transfection of the wt and crt−/− cells with a plasmidcontaining CRT-HA resulted in a significant decrease in the insulinreceptor protein level. However, the level of insulin receptor proteinwas still slightly higher in crt−/− cells (Fig. 8A, IRβ and bar graph)which could be due to the low level of ectopic CRTexpression achievedin these cells. Here we also showed that the increase in insulinreceptor β protein in crt−/− cells is due to increased levels of insulinreceptormRNA (Fig.1B). A recent paper demonstrated a role for CRT inthe down regulation of GLUT1 mRNA and protein level upon exposureto high glucose [35]. Whether CRT could destabilize the insulinreceptor mRNA is not known and awaits further studies. However, in

the current study we show that the role of CRT on regulation of insulinreceptor mRNA is indirect and could be mainly mediated viatranscriptional regulation. p53 has been shown to suppress insulinreceptor promoter via C/EBP and Sp1 transcription factors [39].Intriguingly, our previous study showed a significant decrease in p53protein level and function in the crt−/− cells [17]. Therefore, wepostulated that the increase in insulin receptor mRNA in the crt−/−cells could be mediated by reduced repression of the insulin receptorpromoter by p53. Indeed, introduction of p53 into crt−/− cells resultedin a reduction in insulin receptor β protein expression (Fig. 8B), whichimplies the p53-mediated regulation of insulin receptor β expressionby CRT. Furthermore, as Fig. 8C shows, the inhibition of proteasomeactivity by MG132 leads to increased p53 protein level which inconcomitant with a significant reduction of insulin receptor β proteinin the crt−/− cells, again emphasizing the role of p53 in the regulationof insulin receptor β in crt−/− cells. Several reports have demonstratedthat in addition to insulin receptor β, p53 negatively regulates thepromoter activity of insulin receptor substrate-3 [32], GLUT-1, GLUT-4[31] and insulin like growth factor [24]. Indeed in crt−/− cells bothGLUT-1 and IRS-1 expression are significantly increased (Figs. 3B,C,and 4), however, we observed no change in GLUT-4 expression (Fig.3A). As a regulator of Ca2+ homeostasis CRT [19] could regulate signaltransduction leading to gene regulation. Gene targeted deletion of CRTresults in a significant increase in the basal level of cytosolic Ca2+

whereas no changes in the free endoplasmic reticulum Ca2+ contentwas observed in these cells [23]. To date, no data is available on theregulation of insulin receptor β or IRS-1 expression by alteredintracellular Ca2+ concentration. However, two early reports illu-strated that increasing intracellular Ca2+ concentration results in asignificant increase in GLUT-1 mRNA expression [20,21].

In conclusion, our data demonstrates that loss of CRT function in thecrt−/− cells reduces p53 protein expression thus inducing the expressionof insulin receptor β. Elevated insulin receptor β augment the cell'sresponsiveness to insulin which is accompanied by elevated phosphor-ylation of both IRS-1 and Akt. Increased Akt activity and enhancedglucose uptake in the crt−/− cells can likely represents a mechanism ofthe observed hypoglycemia in the rescued CRT deficient mice. Overall,our study provides the first evidence for an indirect relationshipbetween CRT expression and insulin receptor expression and function.

Acknowledgments

This work was supported by a grant to N.M. from the Heart andStroke Foundation of Manitoba and a Canadian Institute of HealthResearch (CIHR MOP# 42424).

References

[1] M. Ashcroft, R.L. Ludwig, D.B. Woods, T.D. Copeland, H.O. Weber, E.J. MacRae, K.H.Vousden, Phosphorylation of HDM2 by Akt, Oncogene 21 (2002) 1955–1962.

[2] J. Bass, G. Chiu, Y. Argon, D.F. Steiner, Folding of insulin receptor monomers isfacilitated by the molecular chaperones calnexin and calreticulin and impaired byrapid dimerization, J. Cell Biol. 141 (1998) 637–646.

[3] D.P. Brazil, J. Park, B.A. Hemmings, PKB binding proteins. Getting in on the Akt, Cell111 (2002) 293–303.

[4] R. Conejo, M. Lorenzo, Insulin signaling leading to proliferation, survival, andmembrane ruffling in C2C12 myoblasts, J. Cell. Physiol. 187 (2001) 96–108.

[5] D.A. Cross, P.W. Watt, M. Shaw, J. van der Kaay, C.P. Downes, J.C. Holder, P. Cohen,Insulin activates protein kinase B, inhibits glycogen synthase kinase-3 andactivates glycogen synthase by rapamycin-insensitive pathways in skeletal muscleand adipose tissue, FEBS Lett. 406 (1997) 211–215.

[6] M. Delcommenne, C. Tan, V. Gray, L. Rue, J. Woodgett, S. Dedhar, Phosphoinositide-3-OH kinase-dependent regulation of glycogen synthase kinase 3 and proteinkinase B/AKT by the integrin-linked kinase, Proc. Natl. Acad. Sci. U. S. A. 95 (1998)11211–11216.

[7] J. Feng, R. Tamaskovic, Z. Yang, D.P. Brazil, A. Merlo, D. Hess, B.A. Hemmings,Stabilization of Mdm2 via decreased ubiquitination is mediated by protein kinaseB/Akt-dependent phosphorylation, J. Biol. Chem. 279 (2004) 35510–35517.

[8] T.M. Gottlieb, J.F. Leal, R. Seger, Y. Taya, M. Oren, Cross-talk between Akt, p53 andMdm2: possible implications for the regulation of apoptosis, Oncogene 21 (2002)1299–1303.

2351S. Jalali et al. / Biochimica et Biophysica Acta 1783 (2008) 2344–2351

[9] L. Guo, K. Nakamura, J. Lynch, M. Opas, E.N. Olson, L.B. Agellon, M. Michalak,Cardiac-specific expression of calcineurin reverses embryonic lethality incalreticulin-deficient mouse, J. Biol. Chem. 277 (2002) 50776–50779.

[10] D.N. Hebert, J.X. Zhang,W. Chen, B. Foellmer, A. Helenius, The number and locationof glycans on influenza hemagglutinin determine folding and association withcalnexin and calreticulin, J. Cell Biol. 139 (1997) 613–623.

[11] A.M. Johnston, L. Pirola, E. Van Obberghen, Molecular mechanisms of insulinreceptor substrate protein-mediated modulation of insulin signalling, FEBS Lett.546 (2003) 32–36.

[12] E.S. Kandel, N. Hay, The regulation and activities of the multifunctional serine/threonine kinase Akt/PKB, Exp. Cell Res. 253 (1999) 210–229.

[13] R. Knee, I. Ahsan, N. Mesaeli, R.J. Kaufman, M. Michalak, Compromised calnexinfunction in calreticulin-deficient cells, Biochem. Biophys. Res. Commun. 304(2003) 661–666.

[14] S. Lo, J.C. Russell, A.W. Taylor, Determination of glycogen in small tissue samples, J.Appl. Physiol. 28 (1970) 234–236.

[15] M.D. Lozyk, S. Papp, X. Zhang, K. Nakamura, M. Michalak, M. Opas, Ultrastructuralanalysis of development of myocardium in calreticulin-deficient mice, BMC Dev.Biol. 6 (2006) 54.

[16] L.D. Mayo, D.B. Donner, A phosphatidylinositol 3-kinase/Akt pathway promotestranslocation of Mdm2 from the cytoplasm to the nucleus, Proc. Natl. Acad. Sci. U.S. A. 98 (2001) 11598–11603.

[17] N. Mesaeli, C. Phillipson, Impaired p53 expression, function, and nuclearlocalization in calreticulin-deficient cells, Mol. Biol. Cell 15 (2004) 1862–1870.

[18] N. Mesaeli, K. Nakamura, E. Zvaritch, P. Dickie, E. Dziak, K.H. Krause, M. Opas, D.H.MacLennan, M. Michalak, Calreticulin is essential for cardiac development, J. CellBiol. 144 (1999) 857–868.

[19] M. Michalak, E.F. Corbett, N. Mesaeli, K. Nakamura, M. Opas, Calreticulin: oneprotein, one gene, many functions, Biochem. J. 344 (Pt2) (1999) 281–292.

[20] Y. Mitani, A. Behrooz, G.R. Dubyak, F. Ismail-Beigi, Stimulation of GLUT-1 glucosetransporter expression in response to exposure to calcium ionophore A-23187,Am. J. Physiol. 269 (1995) C1228–C1234.

[21] Y. Mitani, G.R. Dubyak, F. Ismail-Beigi, Induction of GLUT-1 mRNA in response toinhibition of oxidative phosphorylation: role of increased [Ca2+]i, Am. J. Physiol.270 (1996) C235–C242.

[22] K. Nakamura, E. Bossy-Wetzel, K. Burns, M.P. Fadel, M. Lozyk, I.S. Goping, M. Opas,R.C. Bleackley, D.R. Green, M.Michalak, Changes in endoplasmic reticulum luminalenvironment affect cell sensitivity to apoptosis, J. Cell Biol. 150 (2000) 731–740.

[23] K. Nakamura, A. Zuppini, S. Arnaudeau, J. Lynch, I. Ahsan, R. Krause, S. Papp, H. DeSmedt, J.B. Parys, W. Muller-Esterl, D.P. Lew, K.H. Krause, N. Demaurex, M. Opas, M.Michalak, Functional specialization of calreticulin domains, J. Cell Biol. 154 (2001)961–972.

[24] C. Ohlsson, N. Kley, H.Werner, D. LeRoith, p53 regulates insulin-like growth factor-I (IGF-I) receptor expression and IGF-I-induced tyrosine phosphorylation in anosteosarcoma cell line: interaction between p53 and Sp1, Endocrinology 139(1998) 1101–1107.

[25] M. Oren, A. Damalas, T. Gottlieb, D. Michael, J. Taplick, J.F. Leal, R. Maya, M. Moas, R.

Seger, Y. Taya, A. Ben-Ze'Ev, Regulation of p53: intricate loops and delicatebalances, Ann. N. Y. Acad. Sci. 973 (2002) 374–383.

[26] S. Persad, S. Attwell, V. Gray, N. Mawji, J.T. Deng, D. Leung, J. Yan, J. Sanghera, M.P.Walsh, S. Dedhar, Regulation of protein kinase B/Akt-serine 473 phosphorylationby integrin-linked kinase: critical roles for kinase activity and amino acidsarginine 211 and serine 343, J. Biol. Chem. 276 (2001) 27462–27469.

[27] T. Saitoh, T. Yanagita, S. Shiraishi, H. Yokoo, H. Kobayashi, S. Minami, T. Onitsuka, A.Wada, Down-regulation of cell surface insulin receptor and insulin receptorsubstrate-1 phosphorylation by inhibitor of 90-kDa heat-shock protein family:endoplasmic reticulum retention of monomeric insulin receptor precursor withcalnexin in adrenal chromaffin cells, Mol. Pharmacol. 62 (2002) 847–855.

[28] M. Sakagami, P. Morrison, W.J. Welch, Benzoquinoid ansamycins (herbimycin Aand geldanamycin) interfere with the maturation of growth factor receptortyrosine kinases, Cell Stress Chaperones 4 (1999) 19–28.

[29] D.D. Sarbassov, D.A. Guertin, S.M. Ali, D.M. Sabatini, Phosphorylation andregulation of Akt/PKB by the rictor-mTOR complex, Science 307 (2005) 1098–1101.

[30] M.P. Scheid, J.R. Woodgett, Unravelling the activation mechanisms of proteinkinase B/Akt, FEBS Lett. 546 (2003) 108–112.

[31] F. Schwartzenberg-Bar-Yoseph, M. Armoni, E. Karnieli, The tumor suppressor p53down-regulates glucose transporters GLUT1 and GLUT4 gene expression, CancerRes. 64 (2004) 2627–2633.

[32] S. Sciacchitano, A. Orecchio, L. Lavra, S. Misiti, A. Giacchini, M. Zani, D. Danese, A.Gurtner, S. Soddu, U. Di Mario, M. Andreoli, Cloning of the mouse insulin receptorsubstrate-3 (mIRS-3) promoter, and its regulation by p53, Mol. Endocrinol. 16(2002) 1577–1589.

[33] L. Stryer, Biochemistry3rd Edition, W.H. Freeman and Company, New York, 1988.[34] S.A. Summers, V.P. Yin, E.L. Whiteman, L.A. Garza, H. Cho, R.L. Tuttle, M.J.

Birnbaum, Signaling pathways mediating insulin-stimulated glucose transport,Ann. N. Y. Acad. Sci. 892 (1999) 169–186.

[35] H. Totary-Jain, T. Naveh-Many, Y. Riahi, N. Kaiser, J. Eckel, S. Sasson, Calreticulindestabilizes glucose transporter-1 mRNA in vascular endothelial and smoothmuscle cells under high-glucose conditions, Circ. Res. 97 (2005) 1001–1008.

[36] A.A. Troussard, N.M. Mawji, C. Ong, A. Mui, R. St -Arnaud, S. Dedhar, Conditionalknock-out of integrin-linked kinase demonstrates an essential role in proteinkinase B/Akt activation, J. Biol. Chem. 278 (2003) 22374–22378.

[37] A.V. Uvarov, N. Mesaeli, Enhanced ubiquitin-proteasome activity in calreticulindeficient cells: a compensatory mechanism for cell survival, Biochim. Biophys.Acta 1783 (2008) 1237–1247.

[38] B. Vanhaesebroeck, D.R. Alessi, The PI3K-PDK1 connection: more than just a roadto PKB, Biochem. J. 346 (Pt3) (2000) 561–576.

[39] N.J. Webster, J.L. Resnik, D.B. Reichart, B. Strauss, M. Haas, B.L. Seely, Repression ofthe insulin receptor promoter by the tumor suppressor gene product p53: apossible mechanism for receptor overexpression in breast cancer, Cancer Res. 56(1996) 2781–2788.

[40] S. Yano, H. Tokumitsu, T.R. Soderling, Calcium promotes cell survival throughCaM-K kinase activation of the protein-kinase-B pathway, Nature 396 (1998)584–587.