The p66Shc redox adaptor protein is induced by saturated fatty acids and mediates lipotoxicity-induced apoptosis in pancreatic beta cells
The role of the redox adaptor protein p66Shc as a potential mediator of saturated fatty acid (FA)-induced beta cell death was investigated.
The effects of the FA palmitate on p66Shc expression were evaluated in human and murine islets and in rat insulin-secreting INS-1E cells. p66Shc expression was also measured in islets from mice fed a high-fat diet (HFD) and from human donors with different BMIs. Cell apoptosis was quantified by two independent assays. The role of p66Shc was investigated using pancreatic islets from p66Shc−/− mice and in INS-1E cells with knockdown of p66Shc or overexpression of wild-type and phosphorylation-defective p66Shc. Production of reactive oxygen species (ROS) was evaluated by the dihydroethidium oxidation method.
Palmitate induced a selective increase in p66Shc protein expression and phosphorylation on Ser36 and augmented apoptosis in human and mouse islets and in INS-1E cells. Inhibiting the tumour suppressor protein p53 prevented both the palmitate-induced increase in p66Shc expression and beta cell apoptosis. Palmitate-induced apoptosis was abrogated in islets from p66Shc−/− mice and following p66Shc knockdown in INS-1E cells; by contrast, overexpression of p66Shc, but not that of the phosphorylation-defective p66Shc mutant, enhanced palmitate-induced apoptosis. The pro-apoptotic effects of p66Shc were dependent upon its c-Jun N-terminal kinase-mediated phosphorylation on Ser36 and associated with generation of ROS. p66Shc protein expression and function were also elevated in islets from HFD-fed mice and from obese/overweight cadaveric human donors.
p53-dependent augmentation of p66Shc expression and function represents a key signalling response contributing to beta cell apoptosis under conditions of lipotoxicity.
KeywordsApoptosis Beta cell Exendin-4 JNK p53 p66Shc Palmitic acid Pancreatic islet
Extracellular signal-regulated kinase
Glucagon-like peptide 1
c-Jun N-terminal kinase
Mitogen-activated protein kinase
Mitogen-activated protein kinase kinase
Protein kinase C
Reactive oxygen species
Short, interfering RNA
Saturated fatty acid (FA)-induced apoptosis of pancreatic beta cells has long been recognised as a major mechanism linking excess dietary fat and beta cell damage, leading to impaired insulin secretion and hyperglycaemia in type 2 diabetes, particularly when associated with visceral obesity [1, 2, 3].
FA-induced beta cell apoptosis involves a variety of signalling mechanisms, including endoplasmic reticulum (ER) stress induction , mitochondrial dysfunction , activation of specific intracellular kinases such as the members of the mitogen-activated protein kinase (MAPK) family c-Jun N-terminal kinase (JNK) and p38 MAPK [6, 7] and protein kinase C (PKC)δ , and peroxisome-generated reactive oxygen species (ROS) [7, 9]. The tumour suppressor protein p53 is also implicated in FA-induced beta cell apoptosis [7, 10], since both palmitate and oleate were shown to stimulate apoptosis of NIT-1 beta cells through p53 , and p53 inhibition was found to be involved in growth factor-dependent promotion of beta cell survival via Akt/protein kinase B . However, p53 signalling in the context of beta cell lipotoxicity is still poorly defined.
p66Shc, a 66 kDa proto-oncogene Src collagen homologue (Shc) adaptor protein, is the largest of three protein isoforms (p66Shc, p52Shc and p46Shc) encoded by the Shc gene [11, 12]. p66Shc possesses specific functions, such as modulation of p46/52Shc complex activation and downstream signalling via MAPK kinase (MEK)–extracellular signal-regulated kinase (ERK) [13, 14, 15, 16] and control of actin cytoskeleton turnover and glucose transport [16, 17]. Importantly, p66Shc is implicated in both sensing and activation of cellular oxidative stress and consequent induction of apoptosis . p66Shc signalling is strictly dependent upon phosphorylation of Ser36 in the protein CH2 domain, triggered by cell exposure to oxidative stress-inducing agents . Recently, p66Shc knockout mice were found to exhibit protection from hyperglycaemia-induced microvascular disease  and from development of diabetic autonomic neuropathy . Levels of p66Shc mRNA and p66Shc protein were found to be increased in the kidney cortex of diabetic mice  and in circulating leucocytes from diabetic patients , suggesting that p66Shc could ‘sense’ the impaired metabolic milieu in diabetes and promote cellular dysfunction. In support of this concept, the p66Shc protein was found to act as a downstream effector of the tumour suppressor p53 gene in oxidative stress-induced apoptosis [18, 22], and a p53 response element in the gene promoter of p66Shc has also been identified . On the other hand, FAs increase the level of p53 acetylation in human monocytes . Therefore, we investigated the involvement of p53 and p66Shc in FA-dependent apoptosis in pancreatic beta cells.
Animals were kept in an animal house under controlled temperature, humidity and lighting. Animal experimentation respecting the regulations of Italy and the EU were conducted with the approval of the ethics committee (CESA) of the Gaetano Salvatore Research Institute (IRGS), Biogem, Italy (internal ID 0907), in accordance with the National Institutes of Health (NIH) Principles of Laboratory Animal Care (NIH publication no. 85-23, revised 1985). Wild-type (WT) C57Bl/6 mice were purchased from Charles River Laboratories (Calco, Italy). p66Shc−/− mice were on matched C57Bl/6 genetic backgrounds. Generation of p66Shc−/− mice has been previously described . From weaning at the age of 3 weeks onwards, mice received a standard diet. At the age of 3 weeks, mice were randomised to a high-fat diet (HFD; purchased from Mucedola (Settimio Milanese, Milan, Italy) and consisting of 60% fat from palm oil) or continued on a standard diet for an additional 21 days. Blood samples were collected from the tail vein of fed mice.
Mouse islets were isolated by bile duct perfusion and collagenase digestion and maintained in culture, as previously described . After isolation, islets were studied within 3 days. Human pancreatic islets were isolated from nine lean and 13 overweight/obese cadaveric donors (electronic supplementary material [ESM] Table 1). Pancreases were excised and processed with the approval of the regional ethics committee. Islets were obtained and cultured, as previously reported [26, 27], and studied within 3 days from isolation. Cell viability in islets, measured by Trypan Blue exclusion, was higher than 90% after 3 days in culture.
Rat insulin-secreting INS-1E cells (passage 15-30; a kind gift from C. B. Wollheim, University of Geneva, Switzerland) were grown and treated with or without 0.5 mmol/l palmitic acid (Sigma-Aldrich, St Louis, MO, USA), as previously reported [6, 28]. All chemical inhibitors were provided by Calbiochem (Darmstadt, Germany). Exendin-4 (exenatide) was obtained from Ely Lilly (Indianapolis, IN, USA).
Adenoviral transfection studies
Generation of an adenoviral construct encoding for the p66Shc protein and transfection of INS-1E cells with recombinant adenoviruses were carried out according to previously reported procedures . INS-1E cells were also transfected with an adenoviral p66Shc construct carrying a Ser36 to Ala36 mutation, as indicated. An empty adenovirus was used as control for the infection (‘mock’).
Short, interfering RNA transfection studies
INS-1E cells grown to 70% confluence were transiently transfected with p66Shc short, interfering RNA (siRNA) 1 or p66Shc siRNA 2 (Qiagen, Hilden, Germany) or with p53 siRNA (s128540; Life Technologies, Carlsbad, CA, USA). See ESM Methods for further details.
Gene expression analysis by quantitative RT-PCR
RNA isolation and cDNA synthesis were conducted as previously described [6, 27]. Primers were designed using Primer Express version 3.0 (Life Technologies, Carlsbad, CA, USA) and purchased from Life Technologies. Primer sequences are reported in ESM Table 2. See ESM Methods for further details.
Measurements of apoptosis
Apoptosis was measured by evaluating mono- and oligonucleosomes in the cytoplasmic fraction of cell lysates, by annexin V allophycocyanin (APC) labelling or by TUNEL assay. See ESM Methods for further details.
ROS production measurements
Intracellular ROS were detected through the evaluation of dihydroethidium (DHE) oxidation. See ESM Methods for further details.
All data are presented as means ± SEM. Statistical analysis was performed using the two-tailed unpaired Student’s t test or ANOVA, followed by the Tukey multiple comparison test, as appropriate. Statistical significance was set at p < 0.05.
Saturated FAs increase p66Shc protein expression in pancreatic islets and beta cells in vitro
Palmitate-induced apoptosis involves p66Shc
Role of p66Shc in high glucose-induced apoptotic beta cell death
The effects of other apoptotic agents, such as high glucose, were also investigated. INS-1E cells challenged with 25 mmol/l glucose levels for up 24 h showed selective augmentation of p66Shc protein levels and Ser36 phosphorylation by ~2.5-fold, as well as increased apoptosis (p < 0.05; ESM Fig. 3c, d). Overexpression of p66Shc enhanced both basal and glucose-stimulated apoptosis compared with control (p < 0.05; ESM Fig. 6a). On the other hand, siRNA-mediated reduction of p66Shc protein levels resulted in partial inhibition of glucose-induced apoptosis by approximately 50% (p < 0.05; ESM Fig. 6b). These results show that p66Shc contributes to beta cell apoptosis induced by high glucose.
The tumour suppressor p53 protein mediates augmentation of p66Shc expression and function in response to FAs
Ser36 phosphorylation of p66Shc is necessary for palmitate-induced apoptosis
p66Shc signals via generation of ROS in FA-mediated apoptosis
Lipotoxic conditions are associated with increased expression of p66Shc in pancreatic islets in vivo
The mRNA levels of p66Shc were also increased several-fold in pancreatic islets from human donors with BMI >24.9 kg/m2 compared with donors with BMI ≤24.9 kg/m2 (p < 0.05; Fig. 8c), and this was associated with an elevation in p53 mRNA levels (p < 0.05; Fig. 8b). Intriguingly, islets from obese donors showed twofold higher mRNA levels of BAX (encoding BCL2-associated X protein), 2.4-fold higher mRNA levels of CASP3 (encoding caspase 3) and 2.9-fold higher mRNA levels of CYCS (encoding cytochrome c somatic) compared with non-obese donors (p < 0.05; Fig. 8d), suggesting greater activation of the pro-apoptotic response. Thus, conditions with high mobilisation of FAs are associated with augmentation of p66Shc expression in pancreatic islets in vivo.
Dysfunction of beta cells in type 2 diabetes may be further exacerbated in states with elevated plasma FAs [29, 30]. In this study, we show for the first time that p66Shc, an important redox sensor and pro-apoptotic member of the Shc protein family of molecular adaptors, is implicated in the negative effects of FAs on pancreatic beta cells.
Accumulating evidence suggests that palmitic acid, a saturated FA, plays an important role in beta cell death [31, 32]. Prolonged exposure of beta cells to elevated palmitate concentrations has been shown to cause downstream JNK activation and increased p53 expression, events which are in part mediated by enhanced oxidative stress and typically linked to cellular damage [7, 24, 33, this study]. The identification of a p53 response/binding element in the promoter region of p66Shc indicated that p66Shc is a p53 target gene and that p66Shc is indispensable for p53-induced apoptosis [22, 23, 34]. In this study, murine and human islets, as well as INS-1E cells, showed increased levels of p66Shc when chronically exposed to elevated palmitate concentrations. Moreover, pancreatic islets isolated from HFD-fed mice displayed robust increases in p66Shc mRNA expression, establishing a link between in vivo lipotoxicity and p66Shc regulation. Both the use of pifithrin-α, an inhibitor of p53 function, and siRNA-mediated p53 knockdown allowed us to prove that the ability of saturated FAs to promote p66Shc expression is mediated by p53, in line with similar results in other cell systems . We found that p53 expression was increased in response to palmitate in both mouse islets and INS-1E beta cells, as well as in human islets from overweight/obese compared with normal-weight donors.
Beta cell challenge with palmitate resulted in enhanced p66Shc mRNA and protein expression as well as phosphorylation of this protein on Ser36, a well-characterised pro-apoptotic event [18, 35, 36]. By using multiple approaches, i.e. siRNA-mediated p66Shc gene silencing, use of islets from p66Shc−/− mice and forced expression of p66Shc in INS-1E cells, the palmitate-triggered apoptosis was shown to require p66Shc. In mouse embryo fibroblasts  and vascular cells , targeted p66Shc gene deletion conferred protection against apoptosis in response to oxidative stress and HFD, respectively. Furthermore, our results show that p66Shc also has a role in beta cell glucotoxicity, since high glucose increased p66Shc protein expression and its Ser36 phosphorylation, and manipulations of p66Shc content affected high glucose-induced beta cell apoptosis. However, p66Shc knockdown resulted in apparently slightly greater inhibition of apoptosis induced by palmitate than by high glucose (compare Fig. 2 and ESM Fig. 6).
Ser36 phosphorylation of p66Shc is critical for inducing the apoptotic cascade in cells exposed to several toxic stimuli . Accordingly, in this study, overexpression of a phosphorylation-defective p66Shc mutant protein in INS-1E cells did not affect basal and reduced palmitate-induced apoptosis, respectively. Depending on the cellular context and stimulus, Ser36 phosphorylation of p66Shc was found to be promoted by either the MAP kinases ERK-1/2 or the stress-activated kinases JNK and p38 MAPK [14, 15, 35, 38, 39, 40, 41, 42]. In specific cells, it was shown to be mediated by PKCβ activation [43, 44, 45]. We found that p66Shc phosphorylation was prevented by pretreatment of cells with the specific JNK inhibitor, as in other cell types [36, 39, 42, 46]. Interestingly, the GLP-1 analogue exendin-4, which reportedly prevents FA-induced apoptosis by inhibiting JNK phosphorylation , markedly inhibited p66Shc phosphorylation on Ser36, linking GLP-1 receptor-dependent anti-apoptotic signalling to inhibition of JNK-mediated phosphorylation of p66Shc. Within mitochondria, the p66Shc protein binds cytochrome c and acts as an oxidoreductase, shuttling electrons from cytochrome c to molecular oxygen . This redox activity of p66Shc explains the increase in ROS levels caused by p66Shc overexpression, as well as their decrease in p66Shc knockout cells . Similarly, in beta cells, ROS levels were increased in Ad/p66Shc cells and further augmented in response to palmitate, in close association with the levels of p66Shc phosphorylation on Ser36; moreover, the antioxidant NAC reduced apoptosis both in control and Ad/p66Shc cells, whereas it was without effect in cells with p66Shc knockdown. Altogether, these results suggest that the p66Shc-mediated ROS generation contributed to palmitate-induced apoptosis also in beta cells. Additional mechanisms could, however, be involved, including inhibition of beta cell survival signals. Indeed, overexpression of p66Shc was found to reduce both basal and insulin-stimulated Akt phosphorylation in INS-1E cells (Natalicchio et al, data not shown).
We show that increased p66Shc expression levels in the pancreatic islets can also be observed in vivo, in response to HFD in mice and in association with overweight and non-diabetic obesity in humans. In the islets from obese donors, elevated p66Shc mRNA levels were associated with enhanced expression of p53 and pro-apoptotic genes. Increased beta cell apoptosis has been reported in obese individuals, particularly those with type 2 diabetes . Of note, increased p66Shc levels have been reported in the aorta, kidneys and stem cells of experimental models of diabetes in vivo, as well as in circulating leucocytes from diabetic patients [19, 20, 21, 45, 50].
In summary, this study identifies p66Shc as a novel signalling intermediate in FA-mediated apoptotic beta cell damage. Targeting p66Shc in beta cells in vivo may potentially represent a novel strategy to prevent the deleterious effects of lipotoxicity on glucose control.
We thank M. Gigante (University of Bari Aldo Moro, Bari, Italy) for assistance with the annexin V assays.
This work was supported by Ministero dell’Università e della Ricerca, Italy, PRIN 2007 #200775N24E_004 (FG), PRIN 2010-2011 #2010JS3PMZ_010 (AN) and PRIN 2010-2011 #2010JS3PMZ_004 (PM).
Duality of interest
FG has received grant support from Eli Lilly, AstraZeneca, Sanofi and Lifescan, and lecture fees from Eli Lilly, AstraZeneca, Sanofi, Lifescan, Novo Nordisk, Boehringer Ingelheim, Takeda and Janssen. AA is a consultant for and has received lecture fees and grant support from Eli Lilly, Novo Nordisk, Servier, Sanofi, AstraZeneca, Bristol-Myers Squibb, Boehringer Ingelheim, Mediolanum, Takeda and Janssen. AN, FT, RL, GB, NM, EC, AL, AC, MB, PM, GPF, MG, SP, and LL have nothing to declare.
All authors contributed to the conception and design of the study or to the analysis and interpretation of data, and to drafting the article or revising it critically for intellectual content. All authors have given their final approval of the version to be published. FG is the guarantor of this work.