Low-density lipoprotein receptor-related protein 1 (LRP1) is an endocytic and multi-functional type I cell surface membrane protein, which is known to be phosphorylated by the activated platelet-derived growth factor receptor (PDGFR). The tyrosine kinase inhibitor imatinib, which inhibits PDGFR and c-Abl, and which has previously been reported to counteract β-cell death and diabetes, has been suggested to reduce atherosclerosis by inhibiting PDGFR-induced LRP1 phosphorylation. The aim of the present study was to study LRP1 function in β-cells and to what extent imatinib modulates LRP1 activity. LRP1 and c-Abl gene knockdown was performed by RNAi using rat INS-1 832/13 and human EndoC1-βH1 cells. LRP1 was also antagonized by treatment with the antagonist low-density lipoprotein receptor-related protein associated protein 1 (LRPAP1). We have used PDGF-BB, a PDGFR agonist, and apolipoprotein E (ApoE), an LRP1 agonist, to stimulate the activities of PDGFR and LRP1 respectively. Knockdown or inhibition of LRP1 resulted in increased hydrogen peroxide (H2O2)- or cytokine-induced cell death, and glucose-induced insulin release was lowered in LRP1-silenced cells. These results indicate that LRP1 function is necessary for β-cell function and that LRP1 is adversely affected by challenges to β-cell health. PDGF-BB, or the combination of PDGF-BB+ApoE, induced phosphorylation of extracellular-signal-regulated kinase (ERK), Akt and LRP1. LRP1 silencing blocked this event. Imatinib blocked phosphorylation of LRP1 by PDGFR activation but induced phosphorylation of ERK. LRP1 silencing blocked imatinib-induced phosphorylation of ERK. Sunitinib also blocked LRP1 phosphorylation in response to PDGF-BB and induced phosphorylation of ERK, but this latter event was not affected by LRP1 knockdown. siRNA-mediated knockdown of the imatinib target c-Abl resulted in an increased ERK phosphorylation at basal conditions, with no further increase in response to imatinib. Imatinib-induced cell survival of tunicamycin-treated cells was partially mediated by ERK activation. We have concluded that imatinib promotes LRP1-dependent ERK activation, possibly via inhibition of c-Abl, and that this could contribute to the pro-survival effects of imatinib on β-cells.
Imatinib mesylate could be used in the treatment of diabetes, but the mechanisms by which it promotes its effects are not characterized.
We have observed in the present study that imatinib, besides inhibiting PDGFR-induced LRP1 phosphorylation, also increases basal LRP1-mediated ERK activation, probably via inhibition of c-Abl.
Increased basal LRP1 activity might be beneficial for β-cell function, proliferation and survival, and, in part, the mechanism by which imatinib reduces diabetes.
Type 2 diabetes is characterized both by insulin resistance of the liver, muscle and adipose tissue and by improper β-cell function, which results in hyperglycaemia, hyperlipidaemia and long-term diabetes-associated complications . A protein pertinent to the pathogenesis of Type 2 diabetes and its complications may be the widely expressed low-density lipoprotein receptor-related protein 1 (LRP1). LRP1 belongs to the low-density lipoprotein receptor (LDLR) gene family, which consists of the closely related members LRP1, LDLR, multiple epidermal growth factor-like domains 7 (MEGF7), ApoE receptor-2 (ApoER2), glycoprotein 330 and very low-density lipoprotein receptor (VLDLR) . LRP1 is an endocytic and multi-functional type I cell surface membrane protein consisting of an N-terminal 515-kDa heavy chain, which is extracellular, and an 85-kDa light chain, which has a transmembrane region and a cytoplasmic tail. The extracellular part of LRP1 contains cysteine-rich complement type repeats (ligand-binding repeats), epidermal growth factor (EGF) repeats and β-propeller domains, and these domains are considered to mediate ligand and co-receptor binding. Examples of extracellular ligands known to bind LRP1 are lipoproteins, such as apolipoprotein E (ApoE), proteases, matrix proteins, growth factors, such as platelet-derived growth factor (PDGF), and bacterial and viral proteins. On the inside of the cell, the cytoplasmic domain interacts with numerous intracellular adaptor and effector molecules, such as Dab-1 (Src activation), protein kinase Cα (PKCα), c-Jun N-terminal kinase (JNK)-interacting protein 1/2 (JIP1/2) [extracellular-signal-regulated kinase (ERK) activation], Shc (tyrosine kinase activation) and Fe65 (actin processing), which all are involved in directing cellular traffic or cell signalling events. The many LRP1-binding ligands and intracellular adaptors explain the multitude of diverse biological roles exerted by LRP1. Examples of such roles are hepatic clearance of plasma proteins, protection against atherosclerosis via regulation of PDGF- and TGF-β (transforming growth factor β) signalling, control of adipocyte and blood–brain barrier function, modulation of Aβ peptide metabolism and transport in Alzheimer's disease, focal adhesion assembly and cell migration, inflammation, phagocytosis and translocation of glucose transporters to cell surfaces .
The small tyrosine kinase inhibitors imatinib mesylate (imatinib), which inhibits the platelet-derived growth factor receptor (PDGFR), c-Abl, c-Kit and DDR1 tyrosine kinases, and sunitinib, which inhibits PDGFR, c-Kit, vascular endothelial growth factor (VEGFR) and others, are mainly used in oncology, but could have beneficial effects in other diseases as well. Indeed, imatinib and sunitinib counteract Type 1 diabetes in non-obese diabetic (NOD) mice [3,4]. This has led to the proposal that imatinib could be used in the treatment of Type 1 diabetes patients [5,6], and a clinical study has recently been initiated to test this hypothesis (http://clinicaltrials.gov/ct2/show/study/NCT01781975). Imatinib also improves metabolic control in Type 2 diabetes patients [7,8]. Imatinib and sunitinib have been proposed to counteract Type 2 diabetes or diabetes-induced complications by affecting islet inflammation , peripheral insulin sensitivity , β-cell survival [3,11], glycosaminoglycan synthesis , Rho activation  and serum adiponectin . Some of these pathways might involve inhibition of PDGFR. It is known that PDGFR signalling is augmented in arteriosclerosis and diabetes, possibly via high glucose, low PPARγ (peroxisome-proliferator-activated receptor γ), ApoE and adiponectin, resulting in not only arteriosclerosis, but possibly also in increased insulin resistance [15–18]. PDGFR signalling is modulated by LRP1, which functions as a co-receptor with PDGFR. When PDGF binds to its receptor it induces tyrosine phosphorylation at the NPXY motif of LRP1, either directly or indirectly via Src family kinase activation [19,20], leading to Shc- and Ras-mediated ERK activation. The PDGFR–LRP1 interaction is affected by imatinib, leading to protection against diabetes-associated arteriosclerosis [21,22]. Also, β-cells display insulin resistance and impaired function in Type 2 diabetes , and since β-cells are known to express PDGFRs , it is conceivable that imatinib-mediated protection against diabetes, at least in part, involves inhibition of the PDGFR and its interaction with LRP1 also in islet cells. Therefore, the aim of the present investigation was to determine whether LRP1 plays a general role in β-cell function and survival and, more specifically, to what extent imatinib affects the LRP1–PDGFR interaction. We have observed that LRP1 down-regulation restricts β-cell function and survival and that imatinib increases basal LRP1-induced ERK and Akt signalling via inhibition of c-Abl.
MATERIALS AND METHODS
Imatinib mesylate was from Euroasian Chemicals. Sunitinib was from LC Laboratories. Lipofectamine™ 2000 was obtained from Invitrogen. Diethylenetriamine NONOate (DETA/NO) was from the Cayman Chemicals. Human very low-density lipoprotein (VLDL) (ApoE) was from the GenWay Biotech. Human and rat PDGF-BB was from PeproTech and R&D Systems respectively. Human recombinant interleukin (IL)-1β, murine- and human interferon (IFN)-γ were from PeproTech. Human recombinant low-density lipoprotein receptor-related protein associated protein 1 (LRPAP1) was from Sino Biological. PD98059 was from Calbiochem and tunicamycin was from Sigma.
Rat INS1 832/13 cells , a gift from Professor H. Mulder, Lund University, Sweden, were cultured in RPMI 1640 medium (Sigma) supplemented with 10% FBS (Sigma), penicillin, glutamine+sodium pyruvate and 50 μM of 2-mercaptoethanol. Human EndoC-βH1 cells, a gift from Professor R. Scharfmann and Professor P. Ravassard, INSERM, Paris, France, were cultured in extracellular matrix (ECM)/fibronectin-coated plates in low-glucose Dulbecco's modified Eagle's medium (DMEM) with supplements as described previously . All the cells were incubated at 5% CO2, 37°C in a humidified air incubator.
siRNA-mediated silencing of LRP1 and c-Abl
For knockdown of LRP1, INS-1 832/13 or EndoC-βH1 cells were plated 1 day before transfection to achieve 50% confluency at the time of transfection. On the day of transfection, cells were incubated in serum- and antibiotic-free medium, and Mission siRNA Universal Negative Control #1 (Sigma), SASI_Hs01_00104587 human LRP1 siRNA (Sigma) or SASI_Rn02_00284970 rat LRP1 siRNA (Sigma) were combined with Lipofectamine™ 2000 for 30 min at room temperature. Cells were incubated for 2 h with the siRNA/liposome mixtures (30 nM), after which culture medium containing twice the normal concentration of serum was added (INS-1 832/13 cells). For knockdown of c-Abl, EndoC-βH1 cells were incubated with Universal Negative Control siRNA (Sigma) or SASI Hs01 00174552 human c-Abl siRNA (Sigma) using the same procedure as given above.
Cells were pre-incubated for 30 min in Hepes-balanced Krebs–Ringer bicarbonate buffer (KRBH) containing 0.5% BSA and then incubated for 1 h with 1 mM of glucose and another 1 h with 10 mM of glucose, both incubations at 37°C and in KRBH containing BSA. Insulin concentrations were measured using an Insulin Assay Kit (Meso Scale Discovery).
Evaluation of cell viability
INS-1 832/13 and EndoC-βH1 cells cultured in 96-well plates were incubated with the cell death agents H2O2 (0.1 mM) or IL-1β (20 ng/ml)+IFN-γ (20 ng/ml) for 24 h. In some experiments, EndoC-βH1 cells were cultured in the presence of 280 μg/ml of the LRP1 antagonist LRPAP1, which was added 3 h before cytokines and palmitate. Cell via-bility was measured by staining cells with propidium iodide (Sigma) (20 μg/ml) and bisbenzimide (5 μg/ml) for 10 min at 37°C. The medium was replaced with PBS, and the red and blue fluorescence was detected using the Kodak 4000 MM image station. The ratio of red to blue was taken as a relative measure of cell death (necrosis and late apoptosis) and was quantified using Carestream MI Digital Science ID software, version 220.127.116.11. In some experiments, cell viability was determined by staining the cells with propidium iodide (Sigma) (20 μg/ml) for 10 min at 37°C. After washing, cells were trypsinized and analysed for red fluorescence (FL-3) using flow cytometry (FACSCalibur, BD).
Cells were washed with ice-cold PBS three times, scraped and centrifuged. The cell pellets were collected and lysed in RIPA buffer supplemented with 1 mM of PMSF and Halt™ protease and phosphatase inhibitor cocktail for half an hour. After centrifugation, supernatants were supplemented with LRP1 antibodies (Abcam) and kept on ice for 1 h. LRP1 was then precipitated with Protein A–Sepharose. The samples were boiled for 5 min in SDS-sample buffer and separated by SDS/PAGE.
Cells were washed in ice-cold PBS, lysed in SDS-sample buffer, boiled for 5 min and separated by SDS/PAGE. Proteins were electrophoretically transferred on to a Hybond-P membrane (GE Healthcare). Membranes were incubated with the following primary antibodies against LRP1 (Abcam), phosphotyrosine (4G10; Millipore), phospho-ERK (P-ERK) (Thr202/Tyr204) and phospho-Akt (P-Akt) (Ser473) (Cell Signaling Technology), total ERK (Santa Cruz) and c-Abl (Oncogene). Bound antibodies were removed from filters by incubating for 40 min at 55°C in 2% (w/v) SDS, 100 mM Tris/HCl, pH 6.8, and 0.5 M 2-mercaptoethanol. The immunodetection was performed as described for the ECL immuno-blotting detection system (GE Healthcare) using the Kodak 4000 MM image station. The intensities of the bands were quantified by densitometric scanning using Carestream MI Digital Science ID software. Filters were finally stained for total protein with Amido Black (0.2% in 5% acetic acid and 10% methanol) for 1 min and then washed repeatedly in water. The filters were then let to dry before scanning and densitometric analysis. Intensities of specific bands were then normalized to Amido Black intensities to correct for differences in protein loading and transfer.
LRP1 levels in INS-1 832/13 and EndoC-βH1 cells are reduced by apoptosis-inducing cell culture conditions
We have analysed LRP1 levels using an antibody that recognizes the LRP1 85-kDa light chain, which contains the transmembrane and intracellular domains. This fragment is shortly after its synthesis cleaved from the extracellular chain by the protease furin but remains attached to the large fragment non-covalently at the cell membrane . The LRP1 antibody detected a 90-kDa band in immunoblot analysis of INS-1 832/13 cells (Figure 1A). To test whether LRP1 is present also in human insulin-producing cells, we also have used EndoC-βH1 cells, a previously generated human β-cell line , and observed the same LRP1 immunoreactive band in these cells (Figure 1B). This band was decreased significantly in INS-1 832/13 cells cultured for 24 h in the presence of 150 μM of sodium palmitate (Figure 1A). There was also a trend of decreased LRP1 in response to the cytokines IL-1β and IFN-γ, but this decrease did not reach statistical significance. The death rates of INS-1 832/13 cells, expressed as a percentage of control, were increased both in response to 150 μM of palmitate and to IL-1β+IFN-γ (Figure 1A). In EndoC-βH1 cells, LRP1 protein was decreased significantly by cytokines but not by 2 mM of palmitate (Figure 1B). EndoC-βH1 cell death rates were increased in response to cytokines but not in response to palmitate. Thus LRP1 protein is present in insulin-producing cells, and its levels are decreased in response to treatments that promote β-cell death.
Effects of cytokines and sodium palmitate on LRP1 levels and relative cell death rates in INS-1 832/13 and EndoC-βH1 cells
LRP1 is required for β-cell glucose-induced insulin release and cell survival
We next down-regulated LRP1 using RNAi and the LRP1 antagonist LRPAP1. LRP1 siRNA treatment in INS-1 832/13 cells resulted in an average 40% decrease in the LRP1 light chain (Figure 2A). INS-1 832/13 cells with down-regulated LRP1 levels released insulin at a low glucose concentration to the same extent as control cells (Figure 2B). However, at a high glucose concentration, LRP1-silenced cells released less insulin than control cells. INS-1 832/13 cells with LRP1 knockdown were also analyzed for cell death rates in response to a 24-h exposure period to H2O2 and the combination of cytokines IL-1β and IFN-γ. The cell death rates were significantly increased in LRP1 knockdown cells when exposed to cytokines (Figure 2C). Treatment of EndoC-βH1 cells with LRPAP1 resulted in a similar decrease in LRP1 (Figure 2D). EndoC-βH1 cells treated with LRPAP1 exhibited modest but significant increases in H2O2- and cytokine-induced cell death rates (Figure 2E).
Down-regulation of LRP1 in INS-1 832/13 and EndoC-βH1 cells resulted in lowered glucose-induced insulin release and increased cell death
LRP1 is tyrosine phosphorylated by PDGFR in INS-1 832/13 cells
Having observed that LRP1 protein levels are decreased during β-cell death, and that LRP1 is necessary for β-cell function, we next investigated LRP1 interaction with PDGFR. PDGFRs (PDGFRα/β) are expressed at high levels in young β-cells and at low levels in old β-cells and have been reported to control age-dependent β-cell proliferation . INS-1 832/13 cells were pre-incubated with the two PDGFR inhibitors–imatinib and sunitinib–and then stimulated with 10% FBS for 10 min or with 20 ng/ml PDGF-BB for 3 min. PDGF-BB is known to activate all PDGFR dimers (PDGFRα/α, PDGFRα/β and PDGFRβ/β). We have observed that immunoprecipitated LRP1 was tyrosine phosphorylated at basal conditions (Figure 3). Both serum and PDGF-BB stimulation resulted in a significant increase in LRP1 tyrosine phosphorylation (Figure 3). This effect was completely abolished by imatinib or sunitinib. These results indicated that PDGFR mediates LRP1 tyrosine phosphorylation in response to serum or PDGF-BB stimulation.
Effects of FCS and PDGF-BB on LRP1 tyrosine phosphorylation
LRP1 mediates PDGF-BB-, ApoE- and imatinib-induced ERK and Akt phosphorylation in INS-1 832/13 cells
PDGFR and LRP1 activation is known to promote Akt and ERK phosphorylations [28,29]. To investigate the effects of LRP1 down-regulation on the phosphorylation of ERK and Akt in INS-1 832/13 cells, LRP1 was silenced by RNAi and ERK/Akt phosphorylation in response to PDGF-BB, and ApoE, a ligand to the LRP1 receptor, was studied by immunoblot analysis. In control cells, we have observed increased ERK and Akt phosphorylations in response to PDGF-BB (Figure 4A). This effect was not present in LRP1-silenced cells. ApoE tended to increase ERK and Akt phosphorylation in control cells but not in LRP1 knockdown cells (Figure 4A). We have also analysed the effect of imatinib and the combination of PDGF-BB+ApoE on the phosphorylation of ERK (Figure 4B). PDGF-BB+ApoE promoted an increase in ERK phosphorylation, which appeared more pronounced than that induced by PDGF-BB alone. Surprisingly, also imatinib increased ERK phosphorylation potently (Figure 4B). The effects of both PDGF-BB+ApoE and imatinib were not present in LRP1 knockdown cells.
Effects of LRP1 silencing on PDGF-BB-, ApoE- and imatinib-induced ERK and Akt phosphorylation
LRP1 mediates PDGF-BB-, ApoE- and imatinib-induced ERK phosphorylation in human EndoC-βH1 cells
We next repeated the LRP1 knockdown experiments using EndoC-βH1 cells. In line with the results obtained with INS-1 832/13 cells (Figure 4), PDGF-BB and ApoE stimulated ERK phosphorylation weakly by themselves and potently when combined (Figure 5). The imatinib effect was pronounced and not further stimulated by PDGF-BB+ApoE. These PDGF-BB-, ApoE- and imatinib-induced effects were completely absent from cells treated with LRP1 siRNA (Figure 5).
Effects of LRP1 silencing in human EndoC-βH1 cells on PDGF-BB-, ApoE- or imatinib-modulated ERK phosphorylation
LRP1 mediates imatinib- but not sunitinib-induced ERK phosphorylation in human EndoC-βH1 cells
The finding that imatinib increased ERK phosphorylation in INS-1 832/13 (Figure 4) and EndoC-βH1 cells (Figure 5) prompted us to compare the effects of imatinib with those of sunitinib. We have observed that both imatinib and sunitinib dose-dependently increased ERK phosphorylation during a 3-h incubation period (Figure 6). Interestingly, the imatinib effect was completely abolished by LRP1 knockdown (Figure 6A), whereas the sunitinib effect remained (Figure 6B). This indicates that sunitinib increases ERK phosphorylation by an LRP1 independent pathway.
Effects of LRP1 silencing in human EndoC-βH1 cells on imatinib- and sunitinib-modulated ERK phosphorylation
c-Abl mediates imatinib-induced ERK phosphorylation
To better understand the above observed difference in imatinib- compared with sunitinib-induced phosphorylation of ERK (Figure 6), we have treated human EndoC-βH1 cells with siRNA specific for c-Abl, a target for imatinib but not for sunitinib. Again, in control cells, imatinib promoted a dose- and time-dependent increase in P-ERK (Figure 7A). In c-Abl knockdown cells, however, no such effect was observed. Instead, basal P-ERK levels appeared increased (Figure 7A). This was further investigated in a separate series of experiments in which siRNA-mediated knockdown of c-Abl was compared with that of LRP1. We have observed that knockdown of c-Abl significantly increased basal P-ERK levels in contrast with knockdown of LRP1 (Figure 7B). The imatinib-induced increase in P-ERK levels was counteracted by LRP1 knockdown, whereas knockdown of c-Abl did not affect the imatinib-induced increase in P-ERK (Figure 7B).
Effects of imatinib, c-Abl- and LRP-1 silencing in human EndoC-βH1 cells on ERK phosphorylation
The ERK inhibitor PD98059 counteracts imatinib-mediated protection against ER-stress-induced EndoC-βH1 cell death
We finally analysed whether the imatinib-induced and LRP1-mediated ERK activation contributed to an improved β-cell survival when exposed to endoplasmic reticulum (ER) stress. For this purpose, EndoC-βH1 cells were pre-treated with imatinib and then exposed to the glycosylation inhibitor tunicamycin for 21 h. Tunicamycin treatment resulted in an increased cell death, which was efficiently counteracted by imatinib (Figure 8). The imatinib-mediated protection against tunicamycin was partially lost when cells were incubated with both imatinib and the ERK inhibitor PD98059 (Figure 8).
ERK inhibition counteracts imatinib-mediated protection against tunicamycin-induced cell death
The drug imatinib targets LRP1 indirectly, via inhibition of the PDGFR. LRP1 is a multi-functional receptor whose role in β-cells has previously not been investigated. We have observed in the present study that LRP1 is necessary for β-cell survival in response to H2O2 and cytokines in EndoC-βH1 cells and cytokines in INS-1 832/13 cells. We have also demonstrated a correlation between cell death rates and the lowering of LRP1 protein levels. Indeed, INS-1 832/13 cells, which exhibited a higher cell death rate in response to palmitate than to cytokines, displayed a palmitate-induced decrease in LRP1, whereas EndoC-βH1 cells, which exhibited a higher cell death rate in response to cytokines, displayed a cytokine-induced decrease in LRP1. This indicates that LRP1 is not only required for β-cells to survive, but also that cell death inducing agents, at least in part, promote their effects by negatively regulating LRP1 gene expression. Indeed, in other cells than β-cells, the expression of LRP1 is known to be regulated, for example in response to hypoxia, thereby affecting retinal neovascularization .
LRP1 down-regulation also negatively affected glucose-induced insulin release. Because LRP1 is a multi-functional receptor, many different mechanisms of actions pertinent to the secretion of insulin can be envisaged. For example, LRP1 plays an important role in endocytosis , and since β-cells actively engage in exo/endocytosis, this process may be supported by LRP1 so that a deficiency of LRP1 results in a lowered glucose-induced insulin release. Another process that might be modulated by LRP1 in the insulin-producing cell is cell adhesion and interaction with the ECM. LRP1 is known to modulate this process via interaction with calreticulin and the ECM protein thrombospondin-1 , and thrombospondin-1 has previously been observed to be necessary for the maintenance of proper β-cell function in vivo .
Because LRP1 might protect β-cells against apoptosis via increased ERK and Akt signalling, in the present study we chose to focus, among the many alternative pathways by which LRP1 may act in β-cells, on the LRP1–PDGFR interaction. In the present study, we have reported that LRP1 interacts with PDGFR, and that this interaction promotes PDGF-BB- and ApoE-induced ERK phosphorylation, both in rat INS-1 832/13 and human EndoC-βH1 cells. This is in line with previous studies reporting that LRP1 is necessary for PDGFR-induced ERK/Akt activation [28,29]. An increase in LRP1-induced signalling via the ERK and Akt pathways fits well with the anti-apoptotic role of LRP1 indicated by the present results.
More surprising, however, is that also imatinib promoted an LRP1-dependent ERK phosphorylation. As imatinib blocked PDGFR-induced LRP1 phosphorylation, it could be argued that LRP1-induced ERK activation should be attenuated, and not increased, by imatinib. The explanation to this peculiar finding is not clear, but our data instead suggest that imatinib via inhibition of c-Abl increases basal LRP1/PDGFR activity. Indeed, in other cell types than insulin-producing cells, c-Abl has been demonstrated to negatively regulate receptor tyrosine kinase signalling. Both TrkA, the receptor for nerve growth factor, and Met, the receptor for hepatocyte growth factor, associate with c-Abl, and this event leads to attenuated receptor signalling [34–37]. This is also in line with our recent finding that imatinib augments phosphoinositide 3-kinase (PI3K) signalling and ERK activation via inhibition of c-Abl in insulin-producing MIN6 cells . Since imatinib enhanced ERK and Akt phosphorylation in the present study, it is likely that c-Abl dampens PDGFR/LRP1 tyrosine kinase signalling at basal conditions. Thus, it is possible that imatinib, via inhibition of c-Abl, increases basal LRP1/PDGFR activity so that stimulation with PDGF-BB or ApoE does not further enhance ERK phosphorylation. It is not clear how LRP1 mediates increased ERK phosphorylation during conditions of PDGFR inhibition, but it has been reported that LRP1 promotes Src activation, and that this via focal adhesion kinase leads to ERK phosphorylation .
We have observed previously that imatinib protects β-cells against ER stress induced by the glycosylation inhibitor tunicamycin , and it is likely that the treatments used in the present study for the induction of β-cell death promoted ER stress. Interestingly, LRP1 has been demonstrated to counteract ER stress in Schwann cells, and this occurred via decreased CCAAT/enhancer-binding protein-homologous protein 10 (CHOP) protein expression and increased PI3K signalling . The results of the present study indicate an anti-apoptotic role for enhanced ERK activity in response to imatinib. Indeed, we have observed that the ERK inhibitor PD98059 counteracted imatinib-induced protection against tunicamycin in EndoC-βH1 cells. Thus, it may be that imatinib and LRP1 in β-cells suppress apoptosis via reduction in ER stress, and that this effect is, in part, mediated by ERK activation. As the ERK inhibitor did not abolish the entire imatinib effect, it is also possible that LRP1 increases β-cell viability by additional mechanisms, for example via Akt activation, which was demonstrated in the present study and observed previously, to protect against apoptosis in neurons .
In vascular cells and other non-β-cells, exaggerated PDGFR signalling is considered pathological as it promotes, for example, cancer, fibrosis, atherosclerosis and vascular neoformation. This may not be the case in β-cells because PDGFR expression in these cells decreases with age . Therefore, low PDGFR expression in adult β-cells probably rules out excessive PDGFR signalling, and it may instead be that imatinib- or LRP1-enhanced PDGFR signalling improves the function or survival of β-cells. Such a chain of events might, at least in part, explain the previously reported beneficial effects of imatinib on β-cell exposed to diabetic conditions in vivo [3,11,43].
We have observed that LRP1 is necessary for normal β-cell function and survival, and that β-cells regulate expression of the LRP1 gene in response to pro-apoptotic factors. LRP1 interacts with the PDGFR and signals to increased ERK and Akt phosphorylation, which may protect against cell death and impaired function. The drug imatinib blocks PDGFR-induced LRP1 phosphorylation but also promotes inhibition of c-Abl, which elevates basal ERK phosphorylation and overrides the inhibitory effects on PDGFR phosphorylation activity. Thus, both LRP1 and imatinib enhance ERK/Akt signalling, which could be beneficial in aging β-cells of pre-diabetic or diabetic individuals.
Rikard Fred, Santosh Boddeti and Marcus Lundberg performed the experiments. Rikard Fred and Nils Welsh designed the experiments, analysed the results and wrote the manuscript.
EndoC-βH1 cells were kindly provided by Raphael Scharfmann and Philippe Ravassard, Paris, France. N.R.W. is co-author of a patent .
This work was supported in part by the Swedish Research Council [grant number 2010-11564-15-3], the Swedish Diabetes Association, the family Ernfors Fund, the Barndiabetesfonden and the Novo-Nordisk Foundation.
Krebs–Ringer bicarbonate buffer
low-density lipoprotein receptor
low-density lipoprotein receptor-related protein 1
low-density lipoprotein receptor-related protein associated protein 1
platelet-derived growth factor
platelet-derived growth factor receptor