Proteostasis failure: at the intersection between aging and Alzheimer’s disease

Lifestyle and demographics of the world’s population are causing serious health problems impacting the brain, increasing the incidence of Alzheimer’s disease (AD) and other types of dementia. Although we have gained important insights into the pathogenic mechanisms of AD, only palliative care is available to patients. AD is characterized by the abnormal deposition of protein aggregates in the brain formed by amyloid β and hyper-phosphorylated, Tau in addition to neuroinflammation. The current failure to treat neurodegenerative diseases underlies the fact that the brain is highly plastic and clinical symptoms are manifested when extensive brain damage has progressed. Thus, new concepts need to be tested to develop preventive strategies that extend healthspan. For half of a century, biomedicine has focused on disease as a means to understand health, developing approaches to target pathogenic features. But health is not simply the absence of disease. It is an active process that enables the organism to cope with intrinsic and extrinsic changes (López-Otín and Kroemer, 2021). Thus, a sensible strategy should focus on extending brain healthspan to prevent disease, by improving synaptic function and connectivity, the initial pathological targets, before neurodegeneration and irreversible brain damage have occurred. The decay in the buffering capacity of the proteostasis network has been pointed out as a primary hallmark of aging (Kaushik and Cuervo, 2015), a phenomenon that may contribute to AD pathogenesis. Strategies to improve proteostasis have been tested in multiple models of neurodegenerative diseases, observing outstanding protective effects. The current challenge is to move forward those interventions into suitable strategies that can be tested in the clinic. Because proteostasis is essential to the normal functions of many organs, it is predicted that the chronic administration of small molecules to target proteostasis may have adverse effects in the long term. Gene therapy is emerging as an alternative because specific brain regions can be targeted locally without affecting the function of other organs. We recently tested a gene therapy strategy to improve proteostasis by artificially enhancing the repair programs to improve proteostasis in the context of AD. Our results indicated outstanding effects in improving cognition and reducing amyloid-β plaques in models of AD (Duran-Aniotz et al., 2023). The environment is not static, our individual history is unique, and the genetic heterogeneity is immense; thus, organisms and cells have evolved physiological plasticity, mediated by robust cellular stress responses to repair and adapt. Indeed, neuronal proteostasis is a pivotal cellular process that determines brain healthspan (Hetz, 2021). The dynamic nature of synaptic composition and abundance is a requisite to drive cognitive and motor processes, involving a tight control of protein biosynthesis, quality control, and degradation pathways. Through the concerted action of specialized stress sensors, the proteostasis network monitors and limits the accumulation of damaged proteins and organelles. These stress pathways signal to the cytosol and nucleus to reprogram gene expression and enable adaptive programs to recover cell function. During aging, the capacity of these homeostatic networks to adjust according to the need declines (López-Otín et al., 2023), which may increase the risk of accumulating abnormal protein aggregates, a hallmark of most neurodegenerative conditions such as AD. Proteostasis is maintained by the dynamic integration between the pathways that mediate the synthesis, quality control, folding, targeting, and degradation of proteins. One of the central nodes of the proteostasis network altered during aging involves the function of the endoplasmic reticulum (ER), the main site for protein production in the cell (Taylor and Hetz, 2020). The ER is also highly altered in AD (Gerakis and Hetz, 2018). To cope with ER stress, cells activate an evolutionarily conserved pathway known as the unfolded protein response (UPR), which aims to re-establish proteostasis (Hetz et al., 2020). The UPR reinforces many processes involved in the function of the secretory pathway to improve protein production and sustain cell function, whereas chronic ER stress results in neurodegeneration and cell death (Hetz et al., 2020). The most conserved UPR signaling branch is initiated by the ER stress sensor IRE1, which catalyzes the unconventional splicing of the mRNA encoding XBP1, excising a 26-nucleotide intron that shifts the mRNA reading frame. This event results in the expression of an active transcription factor, termed XBP1s, which enables transcriptional reprogramming. We recently reported that the activity of the IRE1/XBP1 pathway declines in the brain during normal aging in mammals and strategies to enhance the activity of the UPR extend brain healthspan (Cabral-Miranda et al., 2022). Importantly, we showed that strategies to express XBP1s in neurons either using transgenic mice or gene therapy delayed synaptic dysfunction and cognitive decline during normal aging, in addition to reducing the content of senescence cells in the brain (Cabral-Miranda et al., 2022). XBP1s expression in the hippocampus largely corrected the proteomic changes associated with aging, impacting the expression of a cluster of proteins involved in synaptic function, plasticity, and pathways linked to neurodegenerative diseases (Cabral-Miranda et al., 2022). These observations placed XBP1 in the interphase between aging and the possible risk of developing neurodegenerative diseases. With the idea of testing the effects of artificially enforcing UPR adaptive responses in the AD brain, we overexpressed the active XBP1s form in the nervous system using transgenic mice, in addition to the hippocampus using adeno-associated viral (AAV) vectors (Figure 1; Duran-Aniotz et al., 2023). Overexpression of XBP1s dramatically reduced the content of amyloid plaques in the brain and improved cognitive performance and synaptic plasticity in a model of familial AD (5xFAD transgenic animals expressing mutant APP and presenilin-1). Additionally, XBP1s overexpression in the brain improved memory performance on a model of sporadic AD based on the injection of amyloid β oligomers (Duran-Aniotz et al., 2023). The beneficial effects of XBP1s expression in the context of experimental AD and normal aging involve a substantial correction of gene expression patterns associated with synaptic function, neuronal morphology, and connectivity (Cabral-Miranda et al., 2022; Duran-Aniotz et al., 2023). Thus, we speculate that a major protective mechanism of XBP1s in AD relates to its function as a regulator of neuronal physiology that may parallel its effects in reducing amyloid deposition. However, we cannot discard that targeting XBP1s affects other pathways such as eIF2a phosphorylation, which has important roles in synaptic plasticity (Hetz, 2021). It is predicted that the impaired proteostatic activity observed in aging and AD may impair the efficient production of synaptic proteins and other cargoes of the secretory pathways that are relevant to sustain neuronal function (i.e., ion channels, receptors, proteins involved in connectivity). Thus, we propose that UPR-based gene therapies may target the two pathogenic pillars of AD: abnormal protein aggregation and synaptic dysfunction (both reversible events), representing a unique point of intervention.Figure 1: Gene therapy to reset proteostasis in AD.Created with BioRender.com. AAV: Adeno-associated viral; AD: Alzheimer’s disease.Gene therapy to target different modules of the proteostasis network (chaperones and foldases, UPR, heat shock response, integrated stress response, and autophagy) in the brain has been exploited by many laboratories observing outstanding effects over grain function (reviewed in Hetz, 2021). The artificial enforcement of XBP1s-dependent responses using gene therapy provides a variety of beneficial effects in other disease models, including Parkinson’s disease, spinal cord injury, peripheral nerve degeneration, Huntington’s disease, and eye disease (Hetz et al., 2020). We have undertaken a non-conventional approach to address this fundamental problem where instead of targeting one gene (the classical approach of pharmaceuticals), we developed a technology to engage a full network of processes that as a whole will improve neuronal proteostasis and cell function (holistic view) (Figure 1). We believe that the concept behind the beneficial effects of AAV-XBP1s is “rejuvenation:” the idea of artificially engaging an adaptive capacity that young cells naturally have and that is lost as we age and is exacerbated in AD. Despite these remarkable achievements, challenges persist, such as the inefficient distribution of AAV in the human brain as reported in post-mortem studies of patients participating in clinical trials of AD using AAVs (Castle et al., 2020), emphasizing the need to explore capsid engineering and delivery methods to enhance viral spreading. Treating AD with gene therapy is more complex because large areas of the brain are affected beyond the hippocampus. Gene therapy is an active area of development for therapeutics in central nervous system diseases based on the recent success and US Food and Drug Administration approval of AAV-mediated therapies for spinocerebellar ataxia and Leber disease, in addition to coagulation disorders, and cancer among other conditions (Ling et al., 2023). Since gene therapy has continuously passed regulatory approval by the US Food and Drug Administration and has shown outstanding efficacy in treating various diseases in clinical trials, it promises the possibility of intervening pathogenic pathways or delivering disease-modifier agents like XBP1s into the brains of AD patients. Preclinical studies are required in non-human primates to assess the safety issues related to the chronic expression of adaptive UPR mediators. We are currently developing next-generation AAV-XBP1s vectors that are suitable for clinical trials, that are safer, specific, and with better tissue biodistribution. This work was funded by U.S. Air Force Office of Scientific Research FA9550-21-1-0096, FONDAP program 15150012, ANID/FONDEF ID1ID22I10120, FONDECY/ANID 1220573 and the US Army Medical Research Acquisition Activity (USAMRAA) project number AL2201415; DoD Award HT9425-23-1-0990, AL220141 (to CH). C-Editors: Zhao M, Liu WJ, Qiu Y; T-Editor: Jia Y