β-Amyloid: the key peptide in the pathogenesis of Alzheimer’s disease

1Alzheimer’s disease (AD), also known as Senile Dementia, is a most common age-related neurodegenerative disorder. More than 11 million people per year are estimated to suffer from this disease by 2050, leading to higher cost as well as more burdens on public health and society. Featured by progressive memory loss and cognitive dysfunction, AD induces the loss of motor functions and personality changes, and eventually leads to death. Histopathologically, AD is mainly characterized by extracellular senile plaques (SPs) and intracellular neurofibrillary tangles (NFTs), which results in the loss of neurons and synapses and finally causes gross atrophy of the brain. NFTs are formed by the regulation of the abnormally hyperphosphorylated and glycosylated microtubule-related tau protein, whereas SPs are associated with the aggregation and deposition of amyloid β peptides (Aβ).

2Aβ accumulation is considered to be the distinct morphological hallmark of early onset of AD and it is also proposed to be an activator to induce the sequential lesion events induced by the aggregation of P-Tau. Therefore, Aβ is predicted to be the most potentially efficient target of the drug therapies. Here, this review will focus on this peptide with the aspects of its biogenesis, regulations, as well as degradation and clearance to elucidate the potential significance of these processes for the clinic treatment of AD.

3The factors involved in the pathogenesis of AD have been intensely investigated, however, the mechanisms governing this disease are not fully understood and remain debated. One prevailing proposal is the amyloid cascade hypothesis positing Aβ as the initiator of subsequent events that leads to AD (Figure 1A): Aβ peptides spontaneously aggregate and deposit into soluble oligomers, fibrils and SPs, which then induces oxidative injury, microglial and astrocytic activity as well as alters kinase/phosphatase activity, eventually leading to the neuronal death. Howerver, whether Aβ acts on tau aggregation is still debated.

4Aβ is a small protein composed of 39–43 amino acids with a variety of biophysical states. There are two major isoforms of Aβ, soluble Aβ40, and insoluble Aβ42, and the latter peptide showing higher percentage concentration in AD patients is more prone to aggregate. In a physiological condition, more than 90% of Aβ is in the form of Aβ40 while less than 5% is generated as the longer form of Aβ42.

5Aβ is derived by proteolysis of an evolutionary conserved large transmembrane amyloid precursor protein (APP) through cleavage of β-secretase followed by γ-secretase. Mutations in the gene encoding APP are the main causes of familial AD (FAD;; ). APP can also be processed by α-secretase via non-amyloidogenic pathways to produce non-toxic fragments, which is thought to antagonize Aβ generation (Figure 1B;; ).

6Most of intracellular Aβ normally distribute in the neuronal cytosol, but it is also colocalized with different organelles dependent on where APP, β- and γ-secretase reside. In particular, it has been reported to be produced in the secretory pathway related organelles including endoplasmic reticulum (ER), medial Golgi saccules as well as trans-Golgi network. It has also been found to be correlated with the endocytic endosomes/lysosomes and autophagic vacuoles. Besides, Aβ also resides in mitochondria.

7The physiological role of Aβ is still unknown, but it indeed exists throughout life in healthy individuals. One possible function is to inhibit the activity of γ-secretase in a negative feedback way.

8Aβ aggregation is considered to be the primary reason for the neurotoxicity in the classic view, and Aβ oligomers are the most neurotoxic form. Three “developed-stimulators” may facilitate the aggregation process. The absolute levels of Aβ42 increased by production via APP processing, the ratio of Aβ42 to Aβ40 elevated due to the decreased level of Aβ40 or the soluble oligomeric Aβ. These potential stimulators further promote the accumulation and deposition of Aβ to develop into SPs, which eventually contributes to AD pathology. Moreover, Aβ-induced apoptosis through interaction with cell surface receptors and proteins is also thought to dedicate to the dysfunction of neuronal system.

9The aggregation of Aβ might also promote free radicals such as reactive oxidative species (ROS) to react rapidly with several moieties of proteins and lipids, whose structures or functions are then altered to potential “toxic” oxidized proteins and peroxided lipids. Protein oxidation may cause harm to the membrane integrity or damage the sensitivity to oxidative modification of the enzymes such as glutamine synthetase (GS) and creatine kinase (CK), which are critical to neuronal function. Lipids peroxidation usually causes the toxic product such as 4-hydroxy-2-nonenal (HNE) and 2-propenal (acrolein) that migrates to different parts of the neurons to cause multiple deleterious alterations of cellular function. It includes loss of Ca2+ homoeostasis, inhibition of ion-motive ATPases and glial cell Na+-dependent glutamate and disruption of signaling pathways, all of which are associated with neuronal death. Aβ-induced oxidative stress has also been reported to cause the DNA oxidation, leading to DNA damage.

10Continuous Aβ aggregation or sustained elevation of Aβ would cause a chronic response of the innate immune system by activating microglia through some immunological receptors such as Toll-like Receptors 2 (TLR2), TLR4, TLR6, their co-receptors CD14, CD36, and CD47, which can probably destroy functional neurons by direct phagocytosis. Besides, it also results in inflammatory response, concomitantly releasing a lot of inflammation related mediators including complement factors, eicosanoids, chemokines, and proinflammatory cytokines, which can impair microglial clearance of Aβ and the neuronal debris and increase microglia-mediated neuronal death and loss of neuronal synapses, contributing greatly to AD pathogenesis. Aβ deposition also induces tau pathology by promoting the intraneuronal formation of NFTs which consist of hyperphosphorylated tau proteins. It influences the late-stage of AD pathogenesis. The process is probably mediated by the microglia-driven neuroinflammatory response or by indirectly regulating kinase/phosphatase activity.

11In addition, Aβ precursor APP accumulation at mitochondria membrane can cause mitochondrial dysfunction by blocking the translocation of other mitochondrial inner molecules/proteins and disrupting the electron-transport chain (ETC;; ), which may in turn increase excessive Aβ generation to result in more toxicity. Excessive Aβ can also increase mitochondrial ROS production to induce mitochondrial fragmentation by activating mitochondrial fission proteins Drp1 and Fis1. Aβ localized in mitochondria can bind to two pro-apoptotic factors including Aβ-binding alcohol dehydrogenase (ABAD) and cyclophin D (CypD), consequently increasing neurodegenerative cell death that may be toxic to neurons. Hence, there may be a vicious feedback loop between increased Aβ production and mitochondria dysfunction.

12Because of the key role of Aβ in AD pathogenesis, it has been well accepted that reducing Aβ production or enhancing Aβ clearance may be a putative way to inhibit the cascade of Aβ-induced pathological events.

13Aβ biogenesis is tightly correlated with APP metabolism, including processing and trafficking. There are three isoforms of APP, APP695, APP751, and APP770. APP695 lacking the Kunitz-type protease inhibitor (KPI) domain is predominantly expressed in neurons while the other two isoforms are distributed in most tissues. Some evidence show that APP751 and APP770 up-expression in brains are primarily associated to Aβ deposition. APP processing is mainly regulated by α, β, and γ-secretases (Table 1A). Alpha-secretase plays an essential role in precluding the generation of intact Aβ on account of the cleavage site within the Aβ domain. As a membrane-bound endoprotease, α-secretase usually cleaves APP at plasma membrane. Several members of the a Disintegrin and metalloproteinase (ADAM) family listed in Table 1A have been reported to possess α-secretase activity, which is responsible for APP processing. β- and γ-secretases are devoted to Aβ production via amylodogenic pathway (Figure 1B). Beta-site APP cleaving enzyme 1 (BACE1) and BACE2 are the β-secretases while γ-secretase is complex and composed of presenilins (PSEN1 or PSEN2), Nicastrin, Presenilin enhancer 2 (PEN2), and anterior pharynx defective 1 (APH-1; ). Substantial evidence has shown that manipulation of these secretases could perturb the generation of Aβ. For example, with the α-cleavage abolished in ADAM17-deficient cells affectd Aβ generation. Knock-out of BACE1 in mice completely depleted neuronal Aβ secretion. Mutations of PSEN1 and PSEN2 affected APP cleavage, thereby altering Aβ production. Moreover, regulators related with these secratase, such as the γ-secretase activating protein (GSAP) and CD147, are also likely to be involved in the generation of Aβ.

14Like other type I transmembrane proteins, APP is synthesized and translocated into ER followed by matured in the Golgi apparatus where APP is mainly concentrated in neurons at the steady state. And then APP would traffic through the constitutive secretory pathway. Once reaching the cell surface, it is either cleaved by α-secretase to produce sAPPα or rapidly re-internalized by recognition of a “YENPTY” motif and subsequently recycled back to the cell surface by the recycling compartments or delivered to the lysosome for degradation through the endosomal–lysosomal systems. Generally, promoting APP delivery or inhibiting APP internalization from the cell surface favors the non-amyloidogenic processing, thereafter antagonizing the generation of Aβ. Elevating retention of APP in acidic compartments such as endosomes greatly adds the chances for amyloidogenic processing and consequent Aβ production. Mutations within the “YENPTY” internalization motif have been addressed to block APP internalization and consequently decrease Aβ generation. In contrast, mutation within extracellular KPI domain existing in APP751 and APP770 that assists APP sorting to plasma membrane causes APP retention in the ER, thereby elevating the Aβ production. Synaptic transmission indicated to accelerate APP endocytosis has also been shown to result in increasing the level of secreted Aβ. Some general modulators that could regulate APP trafficking, such as dynamin I, the RAB GTPase family including RAB1B, RAB6, RAB8, and RAB11, and the SNX family including SNX17 and SNX33, have also been found to be associated with Aβ generation. In addition, factors that function in the trafficking of the three secretases may also change the production of Aβ by affecting APP processing. The G-protein-coupled receptor protein GPR3, which is responsible for the cell surface localization of matured γ-secretase, stimulates Aβ production when it is overexpression.

15Proteolytic degradation is thought to take a large part of responsibility in preventing Aβ aggregation or deposition into plaques. The enzymes or proteases in proteolytic degradation play important roles by cleaving Aβ into shorter soluble fragments without neurotoxic effect. The proteases including cathepsin B (CatB), cathepsin D (CatD), Gelatinase A, serine protease factor Xia (FXIa), matrix metalloprotein-9 (MMP-9), neprilysin (NEP), presequence protease (Prep) and the α2M complex are involved in Aβ clearance, while enzymes such as angiotensin-converting enzyme (ACE), endothelin-converting enzyme (ECE), insulin-degrading enzyme (IDE), and uPT and tPA have been found to be involved in the degradation of Aβ (Table 1B;;;;;; ).

16Besides the proteolysis for Aβ degradation, receptor-mediated endocytosis of Aβ that delivers Aβ to lysosome for degradation also contributes to the clearance of toxic Aβ peptide and Aβ deposits. Low-density lipoprotein receptor-related protein 1(LRP1) is considered to be the vital modulator in this process by probably direct binding to Aβ for uptake or through Aβ receptor such as heparin sulphate proteoglycan (HSPG; ) and GPI-anchored cellular prion protein (PrPc; ) to facilitate Aβ trafficking. In addition, Aβ aggregates may also undergo maropinocytosis or phagocytosis for clearance, of which the critical step about actin polymerization is regulated by LRP1. Apolipoprotein E (ApoE), as a major ligand for LRP1 and an important partner of Aβ, plays dual roles in Aβ clearance. Moreover, induction of another degrading pathway of autophagy serves to accelerate the clearance of both soluble Aβ and Aβ aggregates.

17Various types of animal models related to Aβ have been created to dissect the mechanisms for the development and progression of AD, the majority are overexpression transgenic lines (see the summary in Table 1C;;;;; ).

18Despite the existing innate disadvantages. e.g., the transgenic flies that express both human APP and β-secretase BACE1 displayed Aβ accumulation, the animal models are useful to screen genes involved in APP processing, making a great contribution to the development of this field. The secreted-Aβ model in Drosophila is a direct approach to investigate the toxicity caused by Aβ. The Caenorhabditis elegans Aβ-expressing models developed in different tissues are also helpful for examining genes involved in Aβ-induced toxicity. Phenotypes were also analyzed in zebrafish through high-throughout screen by treatment with Alzheimer’s γ-secretase inhibitors to determine efficient compounds for blocking Aβ generation.

19Aβ infusion models are that different species of Aβ peptides is directly injected in the rodent brains. They could mimic the most aspects of AD and deliver experimental results for analysis in a relatively short time. However, these approaches usually induce much higher levels of Aβ in the brains than that exists in the patients, and the results usually vary due to differences in methodology and the concentration of Aβ and the duration treatment. Although most of the models do not show Tau pathology and other shedding fragments from APP processing may also influence neuron systems, transgenic rodent models with overexpression of wild type or mutated human APP can recapitulate some features of AD pathology and provide great convenience to discover more regulators involved in the onset of AD. Nevertheless, no model system is impeccable, further understanding of the molecular mechanisms for Aβ-initiated AD pathology would still be desirable.

20According to the conventional approaches targeting Aβ, therapeutic strategies focus on reducing Aβ production via inhibition of β- and γ-secretases to prevent Aβ aggregation and facilitate Aβ clearance. However, the results are not so inspiring, as all the strategies have failed in clinical trials. Recently, immunotherapies by two monoclonal antibodies against Aβ have been tried. One is Bapineuzumab that could recognize both soluble and insoluble forms of Aβ; the other one is Solanezumab that targets Aβ central domain and recognizes only soluble Aβ. Yet both of them failed to improve the clinical outcomes in patients in phage III trials, which suggests that targeting Aβ alone might not be enough to impede AD progression and multiple steps of Aβ modulations should be taken into consideration according to the different clinical phenotypes in AD patients. e.g., the activity of Foxp3+ regulatory T cells (Tregs) has been reported to be related with Aβ plaque clearance, suggesting novel immunosuppression curing way. Moreover, other approaches besides immunotherapy also need to be explored in order to understand multiple regulations of Aβ for the development of therapies for treating AD.

21The vital role of Aβ as an initiator in the pathology of AD has been well accepted. Aβ production mainly depends on APP processing, whereas Aβ removal is largely associated with proteases and lysosomal enzymes. Subcellularlly, Aβ production together with Aβ precursor protein APP seems closely related with mitochondria, the major source of energy for the brain. Mitochondrial changes including increasing ROS production and reducing ATP generation are in an age-dependent manner. ROS-related oxidative stress induces more Aβ production, while Aβ and APP localized to mitochondrial membranes cause mitochondrial damage by elevating ROS production, blocking the transport of nuclear-encoded mitochondrial protein and disrupting ETC activities. However, the mechanisms of Aβ and APP transport into mitochondrial membranes are still unknown. Future work focus on this part might provide well understanding between mitochondria and APP as well as Aβ production, which might be helpful for exploring new compounds.

22On the other hand, microglial cells play very important roles in the removal of accumulated Aβ not only by phagocytosis but also by releasing proteases such as IDE for degradation, and it is also associated with the innate immune system induced by the aggregated Aβ. Therefore, further researches are needed to find how to keep the clearance function of microglial cells without being impaired by the proinflammatory cytokines.

23Although tremendous progress has been made in the development therapeutic strategies targeting Aβ, more work are still needed to find efficient drugs for curing AD. Network regulations of Aβ should be taken into consideration for the therapy approaches, and it would be instrumental to create good animal models and find more specific biomarkers for the Aβ-mediated pathogenesis of AD.

24The 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.