Amyloid-beta aggregation implicates multiple pathways in Alzheimer’s disease: Understanding the mechanisms

Abstract

Alzheimer’s disease (AD) is a progressive neurodegenerative condition characterized by tau pathology and accumulations of neurofibrillary tangles (NFTs) along with amyloid-beta (Aβ). It has been associated with neuronal damage, synaptic dysfunction, and cognitive deficits. The current review explained the molecular mechanisms behind the implications of Aβ aggregation in AD via multiple events. Beta (β) and gamma (γ) secretases hydrolyzed amyloid precursor protein (APP) to produce Aβ, which then clumps together to form Aβ fibrils. The fibrils increase oxidative stress, inflammatory cascade, and caspase activation to cause hyperphosphorylation of tau protein into neurofibrillary tangles (NFTs), which ultimately lead to neuronal damage. Acetylcholine (Ach) degradation is accelerated by upstream regulation of the acetylcholinesterase (AChE) enzyme, which leads to a deficiency in neurotransmitters and cognitive impairment. There are presently no efficient or disease-modifying medications for AD. It is necessary to advance AD research to suggest novel compounds for treatment and prevention. Prospectively, it might be reasonable to conduct clinical trials with unclean medicines that have a range of effects, including anti-amyloid and anti-tau, neurotransmitter modulation, anti-neuroinflammatory, neuroprotective, and cognitive enhancement.

Introduction

The deposition of aggregated amyloid-beta (Aβ) peptide is the hallmark of Alzheimer’s disease (AD), a neurodegenerative disorder that progresses over time (Riyaz Basha et al., 2005). Due to changes in the brain and the formation of plaques and tangles, it has been linked to neuronal damage and death (Kehoe et al., 2009). AD is the most prevalent type of dementia, accounting for 60% to 70% of dementia cases among older people (Burns and Iliffe, 2009; Zhang et al., 2018).

AD symptoms include short-term memory loss, as well as a progressive decline in the patient’s capacity for thought, judgment, problem-solving, communication, and self-care (Mount and Downton, 2006; Prasansuklab and Tencomnao, 2013). The daily life of an AD patient is also impacted by symptoms including confusion, impatience, aggression, mood swings, personality and behavior changes, issues with attention and spatial orientation, and loss of long-term memory (Prasansuklab and Tencomnao, 2013).

The fifth-leading cause of death in persons over 65 years is AD (Winston, 2020). Over 26.6 million individuals worldwide suffer from it, and its prevalence is significantly increasing yearly (Prince et al., 2014; Olajide and Sarker, 2020). More than 106 million AD patients are anticipated to exist worldwide by 2050. The disease will affect 1 in 85 people, according to estimates (Brookmeyer et al., 2007), as the population ages and environmental factors take effect (Prince et al., 2014).

AD is a leading cause of disability and life reliance among elderly adults worldwide (Nichols et al., 2019), and has a profound influence on individuals, their families, and societies at large (Winston, 2020). The estimated cost of dementia in 2015 was $818 billion, or 1.1% of the GDP (Youssef et al., 2019). In 2020, it was anticipated that treating AD would cost $305 billion in total, and as the population ages, the amount is expected to rise to more than $1 trillion (Winston, 2020). The expense of dementia globally is predicted to reach $2 trillion by 2030 (Wimo et al., 2017). There are currently no effective or disease-modifying medications for AD (Fu et al., 2019). Many of the clinical trials failed in recent years, however, quite a number of the trials are under evaluation. It is essential to advance AD research to suggest new compounds for treatment and prevention. The objective of the current review is to describe the mechanisms behind the implications of Aβ aggregation in AD using multiple pathways. The literature data published between the years 1993 and 2020 were collected using PubMed and Scopus.

Amyloid-beta

The intracellular cleavage of the amyloid precursor protein (APP) by the proteolytic enzymes beta-(β-) secretase and gamma-(γ-) secretase produces the short peptide known as Aβ, which has 40–42 amino acids (Prasansuklab and Tencomnao, 2013). The APP is localized at neuronal synapses and is abundantly expressed in the brain (Thinakaran and Koo, 2008; O’brien and Wong, 2011). It has been linked to synaptic plasticity, cell–cell or cell-matrix interactions, neuroprotection, and regulation of neuronal cell development (Storey and Cappai, 1999).

However, aggregation of Aβ, produced from the cleavage of the amyloidogenic pathway causes neurotoxicity. Most of the body’s cells, including vascular endothelial cells, thyroid epithelial cells, and neuronal and nonneuronal cultured cells, produce Aβ monomers (Schmitt et al., 1995; Fukumoto et al., 1999; Hayes et al., 2002; Kitazume et al., 2010). Although compared to other cell types, neuronal cells appear to produce more Aβ (Fukumoto et al., 1999), demonstrating the possibility that the Aβ-peptide is crucial for maintaining proper CNS physiology. According to the increased long-term potentiation (LTP) mediated by Aβ40, there is a theory that Aβ may play a crucial role in synaptic structural-functional plasticity that underlies learning and memory (Koudinov and Koudinova, 2005).

The amyloid hypothesis

According to the amyloid hypothesis, which explains why synaptic dysfunction and neurodegeneration are brought on by the aggregation of the Aβ-peptide (Van Dyck, 2018). The main contributing factor to AD is errors in the mechanisms directing Aβ formation, accumulation, or elimination. Aβ aggregation stages impair cell-to-cell communication and stimulate the immune system, which then causes inflammation and eventually kills brain cells.

Formation of amyloid-beta

The APP is processed in two distinct pathways as shown in Figure 1. Nonamyloidogenic pathway: The α-secretase enzyme first cleaves APP within the Aβ domain, and then γ-secretase cleaves at the C-terminus. Amyloidogenic pathway: Instead of α-secretase, β-secretase (BACE1) cleaves APP first at the N-terminus of the Aβ domain, and γ-secretase then cleaves it at the C-terminus. The Aβ amylogenic peptides are produced by this chain of events, which then assemble into oligomers to create extracellular neurotoxic plaques in the brain. A similar APP intracellular C-terminal domain (AICD) is released from both pathways (Thinakaran and Koo, 2008). When compared to other fragments, Aβ is chemically “stickier” than those formed by APP proteolytic processes. Small clusters (oligomers) are formed by the fragments initially, followed by chains of clusters (fibrils), and finally “mats” of fibrils (beta-sheets). The final stage is the forming of plaques which contain clusters of beta-sheets and other chemicals (Jung et al., 2010). The amyloid cascade hypothesis (ACH) explains AD pathogenesis from the outcome of two significant facts: (i) Identification of Aβ as a key component of senile plaques (SPs). (ii) Mutations of APP genes and the presenilin 1 and 2 genes (PSEN1 and PSEN2) which are typically detected at the early stage of AD. As a result, it is believed that the emergence of Aβ within SPs is caused by these mutations, which also cause dementia and neuronal cell death (Reitz, 2012).

Figure 1.

Beta-amyloid formation from the proteolytic digestion of the APP. AICD: APP intracellular C-terminal domain (Thinakaran and Koo, 2008).

Formation of neurofibrillary tangles

The aggregation of Aβ causes the formation of neurofibrillary tangles (NFTs) from hyperphosphorylation of tau and its accumulation into tangles is another pathological cause of AD (McGleenon et al., 2009). In normal conditions, tau supports neuronal structures and functions in the brain (Kolarova et al., 2012). However, under pathological circumstances, tau became excessively hyperphosphorylated and aggregated into fibrils known as neurofibrillary tangles. The accumulation of abnormal tau and tangles in neurons leads to neurotoxicity and neuronal degeneration (Gómez-Isla et al., 1997). In addition to the formation of NFTs, Tau phosphorylation impairs tau’s ability to bind microtubules, which impacts neuronal activities such as axonal transport and mitochondrial respiration (Ittner and Götz, 2011). Microtubule depolymerization, self-aggregation, and detachment caused by tau hyperphosphorylation result in neuronal cell death (Suganthy et al., 2016).

Mechanisms of Alzheimer disease

AD pathogenesis starts from the deposition of Aβ which trigger SPs formation, followed by the death of neurons due to NFTs formation (Armstrong, 2011). Neurotoxic mechanisms in the pathology of AD include aberrant protein aggregation, dysfunction of mitochondrial, decreased neurotransmitter production, inflammation, and oxidative stress (Figure 2). However, the buildup of Aβ and the aggregation of tau are the two most prevalent etiologic models of Alzheimer’s pathogenesis (Bloom, 2014).

Figure 2.

Schematic diagram of AD pathology. Created with BioRender.com.

The neuropathological events in AD patients are the result of the toxicity of amyloid oligomers and fibrils, which are from the aggregated forms of Aβ. The bodies regulate the amyloid level via a variety of methods as Aβ accumulates. The concentration of Aβ-peptide is controlled in healthy brain tissue by its production from APP; the influx across the blood–brain barrier (BBB) via its interaction with the receptor for advanced glycation end products (RAGE; Deane et al., 2003, 2009); and its clearance via the low-density lipoprotein receptor-related protein-1 (LRP1) from the brain and enzymatic breakdown in the brain (Selkoe, 2001; Deane et al., 2003, 2009). Additionally, the levels of Aβ affect how RAGE is expressed. RAGE is upregulated when there is excessive Aβ synthesis, and this leads to neurotoxicity (Prasansuklab and Tencomnao, 2013) as shown in Figure 3. Thus, impairments in these regulatory processes may cause excessive Aβ-peptide to build up and deposit in the brains of AD patients. By binding to Aβ_12-28 residues, apolipoprotein E (ApoE) regulates Aβ’s accumulation and lessens its clearance (Prasansuklab and Tencomnao, 2013; Zhang et al., 2018) from the brain (Sagare et al., 2007). Three isoforms of ApoE such as ApoE4 (E4), ApoE3 (E3), and ApoE2 (E2; Liu et al., 2013), regulate cholesterol levels in various ways to influence γ-secretase activity and Aβ synthesis (Osenkowski et al., 2008). According to Bales et al. (2009) and Castellano et al. (2011), the brain Aβ levels and amyloid plaque loading rely on the ApoE isoforms, demonstrating the modulatory involvement of ApoE in Aβ metabolism, aggregation, and deposition (Liu et al., 2013). The differential lipidation status exhibited by ApoE isoforms affects Aβ clearance. The ApoE particles may seize Aβ and stimulate cellular uptake and degradation of ApoE-Aβ complexes (Kim et al., 2009). Aβ clearance at the blood–brain barrier is inhibited by ApoE in an isoform-dependent manner (E4 > E3 and E2). According to studies, E4 is less effective than E3 and E2 at mediating the clearance of Aβ (Deane et al., 2008; Jiang et al., 2008).

Figure 3.

Diagrammatic representation of the regulatory systems for Aβ in an AD patient’s brain. Aβ, amyloid-beta; BBB, Blood–brain barrier; RAGE, Receptor for advanced glycation end products; AICD, APP intracellular C-terminal domain; APP, Amyloid precursor protein; ApoE, apolipoprotein E; E4, ApoE; E3, ApoE3; E2, ApoE2. Created with BioRender.com.

Effects of metal ions on Aβ and tau aggregation

Strong neurotoxic candidates that alter Aβ and tau aggregation include metal dyshomeostasis (Figure 4). Metal ions’ effects on the aggregation of Aβ and tau have been elucidated. Metals like Zn²⁺, Cu²⁺, Fe³⁺, Mn²⁺, Pb²⁺, Cd²⁺, Hg²⁺, and Al³⁺ stimulate amyloidogenic pathways and Aβ aggregation. [red arrow] (O’brien and Wong, 2011). The neurotoxic Aβ-peptide produced by the cleavage of the APP by β- and γ-secretase is secreted into the extracellular space where it spontaneously changes into amyloid plaques. On the other hand, as seen in Figure 4A, the presence of Mg²⁺, Fe²⁺, and Li²⁺ inhibits the production of Aβ [blue arrow] (O’brien and Wong, 2011).

Figure 4.

Metal ions’ effects on the aggregation of Aβ and tau. (A) Amyloid Plaques, (B) Tau tangle. Aβ, amyloid-beta; CDK5, Cyclin-dependent kinase; GSK-3β, Glycogen synthase kinase-3beta; NFTs, Neurofibrillary tangles; PP2A, protein phosphatase 2A (O’brien and Wong, 2011).

Tau hyperphosphorylation and aggregation are promoted by metal ions like Zn²⁺, Cu²⁺, Fe³⁺, Mg²⁺, Mn²⁺, Pb²⁺, Cd²⁺, Hg²⁺, and Al³⁺ [red arrow]. Numerous kinases, including glycogen synthase kinase-3 beta (GSK-3β; Rankin et al., 2007), cyclin-dependent kinase 5 (CDK-5), and others, strongly phosphorylate tau (Kimura et al., 2014). If protein phosphatase 2A (PP2A) is not activated, the hyperphosphorylation of tau may persist (Goedert, 1993). Tau that has been hyperphosphorylated forms NFTs. As depicted in Figure 4B, metal ions like Fe²⁺, and Li²⁺, however, lessen tau hyperphosphorylation [blue arrow].

Induction of oxidative stress by Aβ in AD

The polymeric forms of Aβ cause alterations in biochemical components and brain cell activities that lead to neuropathology associated with AD symptoms. According to reports, one of the earliest clinical manifestations of AD is increased oxidative stress. Hydrogen peroxide (H₂O₂) created due to the reduction of metal ions by Aβ-peptides served as a mediator of the oxidative stress as shown in Figure 5 (Huang et al., 1999; Atwood et al., 2003). Aβ-peptides act as powerful oxidation catalysts and can capture transition metal ions like Cu, Fe, and Zn (Miura et al., 2000). In addition, it was shown that Aβ was toxic to neuronal cultures, and Cu²⁺ ions made it more toxic (Cuajungco et al., 2000). Reactive oxygen species can be produced by the Aβ/Cu(Fe) complexes as a toxin mediator (Huang et al., 1999). Furthermore, AD brains have an extracellular and intracellular accumulation of metal ions with high concentrations of Aβ plaques (Lovell et al., 1998; Religa et al., 2006), which produced free radicals. Because of lipid peroxidation and oxidative protein modification, several biomolecules in the AD brain experience conformational and structural changes that impair their ability to function, which, in turn, affects a variety of cellular processes (Qi et al., 2005). By upregulating the expression of the BACE1 gene, increasing oxidative stress enhances APP processing and ultimately increases Aβ generation (Tong et al., 2005; Coma et al., 2008; Quiroz-Baez et al., 2009). This causes oxidative stress and endoplasmic reticulum (ER) stress by increasing ROS and the accompanying rise in abnormal APP and phosphorylated tau. The ER function can be severely damaged by long-term ER stress, which also causes apoptotic signaling (Ogata et al., 2006; Kouroku et al., 2007). Aβ promotes Ca²⁺ release from neurons’ ER Ca²⁺ pools, increasing intracellular free Ca²⁺ (Zhou et al., 2020). Increased expression of the NR2B subunit of NMDAR causes a rise in Ca²⁺ ion concentration in extrasynaptic regions (Jusko et al., 2008), which, then increases the level of intracellular endoplasmic reticulum Ca²⁺ production (Yin et al., 1994). Studies revealed that Ca²⁺ overload could increase ER stress and facilitate mitochondrial Ca²⁺ uptake by suppressing the expression of the anti-apoptotic protein B-cell lymphoma 2 (Bcl2) and increasing the phosphorylation of extracellular regulated protein kinases (Erk) protein, which would ultimately lead to cytotoxicity and cellular apoptosis (Hajnóczky et al., 2003; Zieg et al., 2008; Zhang et al., 2018).

Figure 5.

Diagram showing how Aβ and metal ions combine to cause oxidative stress in AD. Aβ, amyloid-beta; LP, Lipid peroxidation; NMDAR, N-methyl-D-aspartate receptor; VDCC, Voltage-dependent calcium channel; ER, endoplasmic reticulum; APP, Amyloid precursor protein; ATP, Adenosine triphosphate. Created with BioRender.com.

Induction of neuroinflammation by Aβ in AD

The expression of pro-inflammatory cytokines was increased in response to neuropathological insults induced by Aβ and its interaction with vascular RAGE (Deane et al., 2008). Microglia enhance the clearance of Aβ, but a constant generation of Aβ causes the microglia to become chronically activated, which promotes more amyloid deposition (Hickman et al., 2018). According to Kim and Choi (2015), exposure to Aβ results in microglial activation, which, in turn, causes the generation of reactive oxygen species and neurotoxic pro-inflammatory cytokines. Tau hyperphosphorylation is a result of ROS-activating p38 mitogen-activated protein kinases (p38 MAPK; Giraldo et al., 2014). p38 MAPK has been linked to neuroinflammation and AD due to its ability to activate nuclear factor-B (NF-κB; Kheiri et al., 2018), a master regulator of neuroinflammation gene transcription in the brains of AD patients (Chen et al., 2012; Liao et al., 2016; Olajide and Sarker, 2020). But data indicate that nuclear factor E2-related factor 2 (Nrf2) is negatively regulated by NF-κB (Liu et al., 2008; Kim and Vaziri, 2010; Yu et al., 2011). Substantial evidence connects the activation of the Nrf2 protection mechanism to NF-κB-mediated inflammatory actions (Nair et al., 2008; Sandberg et al., 2014). To uphold the aforementioned finding, Rojo et al. (2010) showed that cyclooxygenase-2 (COX-2), inducible nitric oxide synthases (iNOS), IL-6, and TNF-α levels are elevated when microglia are activated in Nrf2-deficient rats. Ramsey et al. (2007) first noticed this, reporting that the hippocampus of AD patients’ brains had lower amounts of Nrf2 than normal. According to Lee and Kim (2017), through the activation of p38 MAPK, Aβ plaques cause neuronal impairments such as mitochondrial dysfunction, apoptosis, tau phosphorylation, and synaptic dysfunction; the primary cause of neuroinflammation in AD is increased microglial p38 MAPK signaling brought on by Aβ, which results in the production of pro-inflammatory mediators such interleukin-1β (IL-1β), tumor necrosis factor-α (TNF-α), cyclooxygenase-2 (COX-2), and inducible nitric oxide synthase (iNOS); the pathophysiology of the AD brain is worsened by the production of IL-1β from microglia, which increases p38 MAPK activation in neurons and astrocytes; Aβ plaques and IL-1β generated an increase in P38 MAPK activation in astrocytes. By releasing iNOS, COX-2, and TNF-α, this activation accelerates neuroinflammation (Figure 6).

Figure 6.

Diagrammatic representation of how Aβ causes neuroinflammation in AD. Aβ, amyloid-beta; NF-κB, Nuclear factor-κB; p38 MAPK, p38 Mitogen-activated protein kinases; Nrf2, Nuclear factor E2-related factor 2; IL-1β, Interleukin-1β; TNF-α, Tumor necrosis factor-α; COX-2, Cyclooxygenase-2; and iNOS, Inducible nitric oxide synthase (Schnöder et al., 2016; Lee and Kim, 2017). Created with BioRender.com.

Effects of Aβ on acetylcholine in AD

As a neurotransmitter, acetylcholine (Ach) aids in the communication between nerve cells and is essential for memory and learning processes (Kihara and Shimohama, 2004; Francis, 2005). A report revealed that Alzheimer’s patients have reduced amounts of Ach in their brains (Kihara and Shimohama, 2004). Ach is decreased because oxidative stress is induced and inflammatory cytokines are produced by Aβ (Esposito et al., 2006). Free radicals produced due to amyloid peptides have been shown to lower the concentration of Ach by causing cholinergic neurons in the hippocampus to degenerate (Vinod et al., 2009). Additionally, acetylcholinesterase (AChE) activity increases and deactivates acetylcholine in synaptic clefts in the vicinity of amyloid plaques (Mordn et al., 1993; Sberna et al., 1997). According to another study, the amyloid peptide inhibits the production of acetylcholine (ACh) by causing choline to seep through cell membranes (Ehrenstein et al., 1997). Ach deficiency caused cognitive impairment and ultimately AD (Parent et al., 2013; Deture and Dickson, 2019) as shown in Figure 7.

Figure 7.

Aβ and acetylcholine interactions in an AD schematic diagram. APP, Amyloid precursor protein; Aβ, Amyloid-beta; Ach, acetylcholine; AChE, acetylcholinesterase (DeTure and Dickson, 2019). Created with BioRender.com.

Current state of AD treatment

According to Yiannopoulou and Papageorgiou (2013), the formation of amyloid oligomers, which mediates the amyloid cascade, is primarily responsible for neurotoxicity. The main pathophysiologic pillars are oxidation, inflammation, excessive glutamate, and tau hyperphosphorylation. Anti-amyloid disease-modifying treatments (DMTs) have therefore concentrated on three main mechanisms of action (MOAs), including reducing the formation of Aβ₄₂, reducing the burden of Aβ-plaque, and promoting Aβ clearance (Yiannopoulou and Papageorgiou, 2020). Hence, inhibiting the formation of A𝛽-peptide accumulation and tau hyperphosphorylation may be part of the treatment for AD (Mendiola-Precoma et al., 2016). Physical exercise, a healthy diet, and mental stimulation are further AD prevention strategies (Nelson and Tabet, 2015).

Compounds used in clinical trials for the treatment of AD

Acetylcholinesterase inhibitors (AChEIs), such as rivastigmine, donepezil, and galantamine, are clinically effective in increasing the availability of acetylcholine at synapses and thereby inhibiting cognitive decline in AD (Andrieu et al., 2015; Hampel et al., 2018; Cummings et al., 2019). Nevertheless, diarrhea, nausea, and vomiting are some of the typical negative effects of AChEIs on the digestive system (Yiannopoulou and Papageorgiou, 2020). Memantine, which was approved in 2003, selectively binds to open calcium channels that are controlled by NMDA receptors, inhibiting NMDA-mediated ion flux and reducing pathologically excessive glutamate levels (Yiannopoulou and Papageorgiou, 2013; Matsunaga et al., 2015; Cummings et al., 2019). Memantine also reduces the activity of glycogen synthase kinase 3𝛽 (GSK-3𝛽), which, in turn, reduces tau phosphorylation (Prentice et al., 2015; Folch et al., 2016).

Despite extensive and expensive trials, the Food and Drug Administration (FDA) has not approved any DMTs or new medications for AD since 2003 (Anderson et al., 2017; Hukins et al., 2019). The β-secretase (BACE) inhibitors, lanabecestat (Burki, 2018), verubecestat (Egan et al., 2019), and atabecestat (Henley et al., 2019), as well as the anti-amyloid agents such as semagacestat (Doody et al., 2013), bapineuzumab (Vandenberghe et al., 2016), and solanezumab (Neurology, 2016), failed in recent phase 3 clinical trials. The acknowledged explanations for the numerous failures include inadequate understanding of the pathophysiology, inappropriate drug doses, late therapies in disease progression, and wrong therapeutic targets (Gauthier et al., 2016).

Conclusion

The current review explained the molecular mechanisms of Aβ mediating AD via multiple events, including Aβ production and accumulation, tau hyperphosphorylation, metal dyshomeostasis, oxidative stress, neuroinflammation, and inhibition of acetylcholine production. There are presently no efficient or disease-modifying medications for AD. Some of the clinical trials targeting the above events failed in recent years, however, quite a number of the trials are under evaluation. It is necessary to advance AD research to suggest novel compounds for treatment and prevention. Prospectively, it might be reasonable to conduct clinical trials with unclean medicines that have a range of effects, including anti-amyloid, anti-tau, neurotransmitter modulation, anti-neuroinflammatory, neuroprotective, and cognitive enhancement.

Author contributions

MI wrote the manuscript. SM, SO, and AI edited, approved, and concluded the manuscript. All authors contributed to the article and approved the submitted version.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.