Cell Metabolism
Previews Proteostasis Impairment: At the Intersection between Alzheimer’s Disease and Diabetes Danilo B. Medinas1,2 and Claudio Hetz1,2,3,4,* 1Biomedical
Neuroscience Institute, Faculty of Medicine, University of Chile, Santiago, Chile for Molecular Studies of the Cell, Program of Cellular and Molecular Biology, Institute of Biomedical Sciences, University of Chile, Santiago, Chile 3Neurounion Biomedical Foundation, Santiago, Chile 4Department of Immunology and Infectious Diseases, Harvard School of Public Health, Boston, MA 02115, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cmet.2013.11.009 2Center
Altered insulin signaling and neuroinflammation are emerging features of Alzheimer’s disease. Lourenco et al. (2013) identify a pathogenic mechanism that is shared between Alzheimer’s disease and diabetes and contributes to memory loss through a common molecular event: the control of protein synthesis by PKR and the downstream phosphorylation of eIF2a.
Alzheimer’s disease (AD) is the most common cause of dementia, affecting one in eight individuals over 65 years in the US. The accumulation of b-amyloid (Ab) oligomers and amyloid plaques in the cerebral cortex and hippocampus is thought to underlie synaptic dysfunction and neuronal death, leading to cognitive impairment and memory loss in AD (Cornejo and Hetz, 2013; Holtzman et al., 2011). Thus, it is of paramount importance to define the age-related factors responsible for the burden of Ab peptides. An emerging concept suggests that AD may be viewed as a form of diabetes in the brain (Bomfim et al., 2012). Recently, Bomfim et al. (2012) described the beneficial effects of administering exendin-4, an FDA-approved antidiabetic agent, in a transgenic mouse model of AD. In the study, exendin-4 treatment corrected insulin resistance and improved cognitive deficits in the AD mice (Bomfim et al., 2012). Insulin signaling in the brain orchestrates energy balance through the concerted action on different brain circuits, ranging from control of appetite in the hypothalamus to consolidation of learning and memory in the hippocampus (Fernandez and Torres Alema´n, 2012). In this way, it appears that peripheral organs and the nervous system share common mechanisms of metabolic regulation, and hence, molecular events associated with metabolic disorders may also participate in neurodegenerative cascades. In this issue, Lourenco et al. (2013) shed light on the connection between AD and diabetes by reporting that Ab oligomers
cause impaired insulin signaling and memory deficits through a mechanism involving tumor necrosis factor alpha (TNF-a) and downstream activation of the protein kinase R (PKR) (Lourenco et al., 2013) (Figure 1). Over the last century, we have witnessed a dramatic increase in life expectancy along with many changes in lifestyle. The metabolic stress triggered by lack of physical exercise combined with high calorie intake causes an array of interconnected health problems, such as obesity and diabetes. At the molecular and cellular levels, the increase in energy availability disrupts the homeostasis of several organelles, including the endoplasmic reticulum (ER) (Hotamisligil, 2010). This leads to an abnormal condition termed ER stress, characterized by the accumulation of misfolded proteins in the ER lumen and activation of a stress signaling pathway termed the unfolded protein response (UPR) (Hetz et al., 2013). The UPR decreases protein synthesis through PKR-like ER kinase (PERK)-mediated inhibitory phosphorylation of the eukaryotic translation initiation factor 2a (eIF2a). eIF2a phosphorylation is part of an integrated stress response that decreases protein translation of a stressed cell and also induces downstream transcriptional responses through the expression of activating transcription factor 4 (ATF4) (Hetz et al., 2013). Importantly, genetic or pharmacological inhibition of eIF2a kinases leads to enhanced synaptic plasticity and cognitive function in wild-type mice and
transgenic mouse models of AD (Zhu et al., 2011; Ma et al., 2013), possibly due to increased expression of a large cluster of synaptic proteins. Another eIF2a kinase, termed PKR, is activated by inflammatory signals and is involved in learning and memory (Zhu et al., 2011). Thus, stress adaptation and memory consolidation converge at the level of eIF2a phosphorylation, and aberrantly elevated levels of phosphorylated eIF2a are likely involved in AD pathogenesis. Lourenco et al. (2013) now examine in more detail the connection between AD and diabetes. Using TNF-a / and PKR / mice, they demonstrate that TNF-a and PKR lie upstream of eIF2a phosphorylation and establish the TNF-a/PKR axis as a mechanism of synapse loss and cognitive impairment triggered by Ab oligomers. In addition, they show that ER stress mediates the inhibitory phosphorylation of insulin receptor substrate 1 (IRS-1), leading to impaired insulin signaling in the brain. Remarkably, insulin treatment reverted phosphorylation of PKR and eIF2a in hippocampal neurons. Furthermore, the administration of antidiabetic agents (i.e., liraglutide and exendin-4) and chemical chaperones (i.e., 4-phenylbutyrate [4-PBA], attenuators of ER stress) reduced phosphorylation of eIF2a and ER stress levels, increasing synaptic density and cognitive performance in animal models of AD (Figure 1). Finally, liraglutide reduced phospho-eIF2a levels in nonhuman primates injected
Cell Metabolism 18, December 3, 2013 ª2013 Elsevier Inc. 771
Cell Metabolism
Previews
with Ab oligomers, increasing perpetuating a vicious cycle the clinical relevance of the that eventually leads to cell study. demise (Figure 1). Overall, Lourenco et al. These reports provide (2013) convincingly convey pieces of the same puzzle, the idea that Ab oligomers where ER stress is at the exert their neurotoxic activity crossroads of inflammation, by instigating a TNF-a/PKR metabolic deregulation, and signaling pathway, leading to inhibition of protein translaattenuated protein synthesis tion (Figure 1). Further studies due to sustained phosphorytargeting proteostasis dislation of eIF2a and insulin turbances should provide resistance by the phosphoryvaluable insights into the contribution of these events lation of IRS-1. Moreover, the to the pathogenesis of AD. authors provide valuable mechanistic clues into the beneficial effects of antidiaACKNOWLEDGMENTS betic drugs in animal models of AD. Nonetheless, some We acknowledge support from FONDECYT (1100176), Millennium specific points remain open Institute (P09-015-F), Ring Initiative for further investigation. (ACT1109), FONDEF (D11I1007), According to the authors, Alzheimer’s Disease Association microglia likely represent the (to C.H.), and FONDECYT (3130351 to D.B.M.). source of TNF-a. If so, do Ab oligomers directly engage Figure 1. Crosstalk between ER Stress, PKR, Inflammation, and signaling events in microglia REFERENCES Insulin Signaling in Alzheimer’s Disease that then connect with the Accumulation of Ab oligomers (AbOs) leads to production of TNF-a by microBomfim, T.R., Forny-Germano, L., glia as well as phosphorylation and activation of PKR in neurons. Active PKR TNF-a/PKR pathway? What Sathler, L.B., Brito-Moreira, J., Houleads to the inhibition of protein synthesis by phosphorylation of eIF2a. Ab oligis the exact mechanism zel, J.C., Decker, H., Silverman, omers also cause ER stress and eventually insulin resistance due to serine involved the phosphorylation M.A., Kazi, H., Melo, H.M., phosphorylation of insulin receptor substrate 1 (IRS1). ER stress and insulin McClean, P.L., et al. (2012). J. Clin. of IRS-1 by Ab oligomers? resistance feed a vicious cycle, promoting further accumulation of Ab oligoInvest. 122, 1339–1353. mers by JNK3-stimulated processing of amyloid precursor protein (APP) and Regarding the beneficial inhibition of protein synthesis due to PKR- and PERK-mediated phosphorylaCornejo, V.H., and Hetz, C. (2013). action of the antidiabetic tion of eIF2a. The reduction in protein synthesis impairs synaptic plasticity, Semin. Immunopathol. 35, 277–292. agents, crucial questions leading to memory deficits in AD. Administration of antidiabetic agents (ADAs) or the chemical chaperone 4-phenylbutyrate (4-PBA) blocks the vicious have yet to be answered. Fernandez, A.M., and Torres-Alecycle and improves cognition in mouse models of AD. ma´n, I. (2012). Nat. Rev. Neurosci. Most important, by which 13, 225–239. mechanism do the antidiabetic agents reduce the amount of Ab inhibition of insulin signaling, augmented Hetz, C., Chevet, E., and Harding, H.P. (2013). Nat. oligomers? Are these drugs acting on production of Ab oligomers, and attenua- Rev. Drug Discov. 12, 703–719. neurons, or do they have an unanticipated tion of protein synthesis. The outcome of Holtzman, D.M., Morris, J.C., and Goate, A.M. action in glial cells? Are the levels of such detrimental synergism is weakening (2011). Sci. Transl. Med. 3, sr1. TNF-a also reduced upon treatment of synapses and cognitive deficits that Hotamisligil, G.S. (2010). Cell 140, 900–917. with these antidiabetic agents? Finally, eventually translate into dementia. Along although ER stress was shown to induce these lines, a recent report proposed a Lourenco, M.V., Clarke, J.R., Frozza, R.L., Bomfim, T.R., Forny-Germano, L., Batista, A.F., IRS-1 inhibition, this effect was inde- positive feedback loop for sustained Sathler, L.B., Brito-Moreira, J., Amaral, O.B., pendent of eIF2a. The contribution of toxicity of Ab oligomers that involves Silva, C.A., et al. (2013). Cell Metab. 18, this other UPR signaling branches (i.e., the translational block and c-Jun N-terminal issue, 831–843. IRE1/XBP1 pathway) to the disease pro- kinase 3 (JNK3) activation (Yoon et al., Ma, T., Trinh, M.A., Wexler, A.J., Bourbon, C., cess remains to be determined in this 2012), which enhances the generation of Gatti, E., Pierre, P., Cavener, D.R., and Klann, E. context. Ab oligomers. Further accumulation of (2013). Nat. Neurosci. 16, 1299–1305. The fact that insulin signaling can be these species, in turn, blocks protein Yoon, S.O., Park, D.J., Ryu, J.C., Ozer, H.G., Tep, impaired by Ab oligomers and, at translation by inhibiting the mam- C., Shin, Y.J., Lim, T.H., Pastorino, L., Kunwar, Walton, J.C., et al. (2012). Neuron 75, the same time, activated to reduce the malian target of rapamycin (mTOR) A.J., 824–837. burden of such species, suggests the complex through the activation of the occurrence of a feedforward cycle metabolic sensor AMP kinase. Altered Zhu, P.J., Huang, W., Kalikulov, D., Yoo, J.W., Placzek, A.N., Stoica, L., Zhou, H., Bell, J.C., Friedin which inflammation and metabolic protein translation then induces ER lander, M.J., Krnjevic, K., et al. (2011). Cell 147, pathways intersect at the ER, causing stress and further activation of JNK3, 1384–1396.
772 Cell Metabolism 18, December 3, 2013 ª2013 Elsevier Inc.
Previews Proteostasis Impairment: At the Intersection between Alzheimer’s Disease and Diabetes Danilo B. Medinas1,2 and Claudio Hetz1,2,3,4,* 1Biomedical
Neuroscience Institute, Faculty of Medicine, University of Chile, Santiago, Chile for Molecular Studies of the Cell, Program of Cellular and Molecular Biology, Institute of Biomedical Sciences, University of Chile, Santiago, Chile 3Neurounion Biomedical Foundation, Santiago, Chile 4Department of Immunology and Infectious Diseases, Harvard School of Public Health, Boston, MA 02115, USA *Correspondence: [email protected] http://dx.doi.org/10.1016/j.cmet.2013.11.009 2Center
Altered insulin signaling and neuroinflammation are emerging features of Alzheimer’s disease. Lourenco et al. (2013) identify a pathogenic mechanism that is shared between Alzheimer’s disease and diabetes and contributes to memory loss through a common molecular event: the control of protein synthesis by PKR and the downstream phosphorylation of eIF2a.
Alzheimer’s disease (AD) is the most common cause of dementia, affecting one in eight individuals over 65 years in the US. The accumulation of b-amyloid (Ab) oligomers and amyloid plaques in the cerebral cortex and hippocampus is thought to underlie synaptic dysfunction and neuronal death, leading to cognitive impairment and memory loss in AD (Cornejo and Hetz, 2013; Holtzman et al., 2011). Thus, it is of paramount importance to define the age-related factors responsible for the burden of Ab peptides. An emerging concept suggests that AD may be viewed as a form of diabetes in the brain (Bomfim et al., 2012). Recently, Bomfim et al. (2012) described the beneficial effects of administering exendin-4, an FDA-approved antidiabetic agent, in a transgenic mouse model of AD. In the study, exendin-4 treatment corrected insulin resistance and improved cognitive deficits in the AD mice (Bomfim et al., 2012). Insulin signaling in the brain orchestrates energy balance through the concerted action on different brain circuits, ranging from control of appetite in the hypothalamus to consolidation of learning and memory in the hippocampus (Fernandez and Torres Alema´n, 2012). In this way, it appears that peripheral organs and the nervous system share common mechanisms of metabolic regulation, and hence, molecular events associated with metabolic disorders may also participate in neurodegenerative cascades. In this issue, Lourenco et al. (2013) shed light on the connection between AD and diabetes by reporting that Ab oligomers
cause impaired insulin signaling and memory deficits through a mechanism involving tumor necrosis factor alpha (TNF-a) and downstream activation of the protein kinase R (PKR) (Lourenco et al., 2013) (Figure 1). Over the last century, we have witnessed a dramatic increase in life expectancy along with many changes in lifestyle. The metabolic stress triggered by lack of physical exercise combined with high calorie intake causes an array of interconnected health problems, such as obesity and diabetes. At the molecular and cellular levels, the increase in energy availability disrupts the homeostasis of several organelles, including the endoplasmic reticulum (ER) (Hotamisligil, 2010). This leads to an abnormal condition termed ER stress, characterized by the accumulation of misfolded proteins in the ER lumen and activation of a stress signaling pathway termed the unfolded protein response (UPR) (Hetz et al., 2013). The UPR decreases protein synthesis through PKR-like ER kinase (PERK)-mediated inhibitory phosphorylation of the eukaryotic translation initiation factor 2a (eIF2a). eIF2a phosphorylation is part of an integrated stress response that decreases protein translation of a stressed cell and also induces downstream transcriptional responses through the expression of activating transcription factor 4 (ATF4) (Hetz et al., 2013). Importantly, genetic or pharmacological inhibition of eIF2a kinases leads to enhanced synaptic plasticity and cognitive function in wild-type mice and
transgenic mouse models of AD (Zhu et al., 2011; Ma et al., 2013), possibly due to increased expression of a large cluster of synaptic proteins. Another eIF2a kinase, termed PKR, is activated by inflammatory signals and is involved in learning and memory (Zhu et al., 2011). Thus, stress adaptation and memory consolidation converge at the level of eIF2a phosphorylation, and aberrantly elevated levels of phosphorylated eIF2a are likely involved in AD pathogenesis. Lourenco et al. (2013) now examine in more detail the connection between AD and diabetes. Using TNF-a / and PKR / mice, they demonstrate that TNF-a and PKR lie upstream of eIF2a phosphorylation and establish the TNF-a/PKR axis as a mechanism of synapse loss and cognitive impairment triggered by Ab oligomers. In addition, they show that ER stress mediates the inhibitory phosphorylation of insulin receptor substrate 1 (IRS-1), leading to impaired insulin signaling in the brain. Remarkably, insulin treatment reverted phosphorylation of PKR and eIF2a in hippocampal neurons. Furthermore, the administration of antidiabetic agents (i.e., liraglutide and exendin-4) and chemical chaperones (i.e., 4-phenylbutyrate [4-PBA], attenuators of ER stress) reduced phosphorylation of eIF2a and ER stress levels, increasing synaptic density and cognitive performance in animal models of AD (Figure 1). Finally, liraglutide reduced phospho-eIF2a levels in nonhuman primates injected
Cell Metabolism 18, December 3, 2013 ª2013 Elsevier Inc. 771
Cell Metabolism
Previews
with Ab oligomers, increasing perpetuating a vicious cycle the clinical relevance of the that eventually leads to cell study. demise (Figure 1). Overall, Lourenco et al. These reports provide (2013) convincingly convey pieces of the same puzzle, the idea that Ab oligomers where ER stress is at the exert their neurotoxic activity crossroads of inflammation, by instigating a TNF-a/PKR metabolic deregulation, and signaling pathway, leading to inhibition of protein translaattenuated protein synthesis tion (Figure 1). Further studies due to sustained phosphorytargeting proteostasis dislation of eIF2a and insulin turbances should provide resistance by the phosphoryvaluable insights into the contribution of these events lation of IRS-1. Moreover, the to the pathogenesis of AD. authors provide valuable mechanistic clues into the beneficial effects of antidiaACKNOWLEDGMENTS betic drugs in animal models of AD. Nonetheless, some We acknowledge support from FONDECYT (1100176), Millennium specific points remain open Institute (P09-015-F), Ring Initiative for further investigation. (ACT1109), FONDEF (D11I1007), According to the authors, Alzheimer’s Disease Association microglia likely represent the (to C.H.), and FONDECYT (3130351 to D.B.M.). source of TNF-a. If so, do Ab oligomers directly engage Figure 1. Crosstalk between ER Stress, PKR, Inflammation, and signaling events in microglia REFERENCES Insulin Signaling in Alzheimer’s Disease that then connect with the Accumulation of Ab oligomers (AbOs) leads to production of TNF-a by microBomfim, T.R., Forny-Germano, L., glia as well as phosphorylation and activation of PKR in neurons. Active PKR TNF-a/PKR pathway? What Sathler, L.B., Brito-Moreira, J., Houleads to the inhibition of protein synthesis by phosphorylation of eIF2a. Ab oligis the exact mechanism zel, J.C., Decker, H., Silverman, omers also cause ER stress and eventually insulin resistance due to serine involved the phosphorylation M.A., Kazi, H., Melo, H.M., phosphorylation of insulin receptor substrate 1 (IRS1). ER stress and insulin McClean, P.L., et al. (2012). J. Clin. of IRS-1 by Ab oligomers? resistance feed a vicious cycle, promoting further accumulation of Ab oligoInvest. 122, 1339–1353. mers by JNK3-stimulated processing of amyloid precursor protein (APP) and Regarding the beneficial inhibition of protein synthesis due to PKR- and PERK-mediated phosphorylaCornejo, V.H., and Hetz, C. (2013). action of the antidiabetic tion of eIF2a. The reduction in protein synthesis impairs synaptic plasticity, Semin. Immunopathol. 35, 277–292. agents, crucial questions leading to memory deficits in AD. Administration of antidiabetic agents (ADAs) or the chemical chaperone 4-phenylbutyrate (4-PBA) blocks the vicious have yet to be answered. Fernandez, A.M., and Torres-Alecycle and improves cognition in mouse models of AD. ma´n, I. (2012). Nat. Rev. Neurosci. Most important, by which 13, 225–239. mechanism do the antidiabetic agents reduce the amount of Ab inhibition of insulin signaling, augmented Hetz, C., Chevet, E., and Harding, H.P. (2013). Nat. oligomers? Are these drugs acting on production of Ab oligomers, and attenua- Rev. Drug Discov. 12, 703–719. neurons, or do they have an unanticipated tion of protein synthesis. The outcome of Holtzman, D.M., Morris, J.C., and Goate, A.M. action in glial cells? Are the levels of such detrimental synergism is weakening (2011). Sci. Transl. Med. 3, sr1. TNF-a also reduced upon treatment of synapses and cognitive deficits that Hotamisligil, G.S. (2010). Cell 140, 900–917. with these antidiabetic agents? Finally, eventually translate into dementia. Along although ER stress was shown to induce these lines, a recent report proposed a Lourenco, M.V., Clarke, J.R., Frozza, R.L., Bomfim, T.R., Forny-Germano, L., Batista, A.F., IRS-1 inhibition, this effect was inde- positive feedback loop for sustained Sathler, L.B., Brito-Moreira, J., Amaral, O.B., pendent of eIF2a. The contribution of toxicity of Ab oligomers that involves Silva, C.A., et al. (2013). Cell Metab. 18, this other UPR signaling branches (i.e., the translational block and c-Jun N-terminal issue, 831–843. IRE1/XBP1 pathway) to the disease pro- kinase 3 (JNK3) activation (Yoon et al., Ma, T., Trinh, M.A., Wexler, A.J., Bourbon, C., cess remains to be determined in this 2012), which enhances the generation of Gatti, E., Pierre, P., Cavener, D.R., and Klann, E. context. Ab oligomers. Further accumulation of (2013). Nat. Neurosci. 16, 1299–1305. The fact that insulin signaling can be these species, in turn, blocks protein Yoon, S.O., Park, D.J., Ryu, J.C., Ozer, H.G., Tep, impaired by Ab oligomers and, at translation by inhibiting the mam- C., Shin, Y.J., Lim, T.H., Pastorino, L., Kunwar, Walton, J.C., et al. (2012). Neuron 75, the same time, activated to reduce the malian target of rapamycin (mTOR) A.J., 824–837. burden of such species, suggests the complex through the activation of the occurrence of a feedforward cycle metabolic sensor AMP kinase. Altered Zhu, P.J., Huang, W., Kalikulov, D., Yoo, J.W., Placzek, A.N., Stoica, L., Zhou, H., Bell, J.C., Friedin which inflammation and metabolic protein translation then induces ER lander, M.J., Krnjevic, K., et al. (2011). Cell 147, pathways intersect at the ER, causing stress and further activation of JNK3, 1384–1396.
772 Cell Metabolism 18, December 3, 2013 ª2013 Elsevier Inc.
