Carlo Cervellati, Giuseppe Valacchi, Giovanni Zuliani

Beta-site amyloid precursor protein (APP)cleaving enzyme-1 (BACE1): brief overview on the biology and its role in Alzheimer’s disease(AD) pathogenesis:AD is a multifactorial and multifaceted disease, with a complex and still not completely understood pathogenesis(Cervellati et al., 2016; Iturria-Medina et al., 2016). Multiple hypotheses have been proposed to explain the pathobiology of the disease. In this plethora of mechanistic models, the central role amyloid beta (Aβ)forming neuritic plaques remains undiscussed.However, the adjective “central” does not mean that Aβ alone can trigger and fuel all the AD neurodegenerative process. Indeed, it is now amply recognized that a combination of multiple biochemical (especially redox),immune system and vascular abnormalities must also occur to define the direction of disease trajectory (Iturria-Medina et al.,2016). This complexity is probably one of the main reasons for the current unavailability of disease-modifying therapies that may prevent or slow the rate of disease progression.

The “original sin” of many of the unsuccessful therapeutic trials on AD lies in the identification of Aβ production/clearance as sole pharmacological target. The concept that decreasing Aβ is not sufficient to achieve positive outcome in AD patients is, among others, confirmed by the failure of inhibitors of BACE1. This enzyme catalyzes the rate-limiting step of Aβ formation, leadiseng to different lengths of Aβ peptides, including the highly neurotoxic Aβ42(Hampel et al., 2020;Figure 1). Considering that BACE1 deficient animals do not produce Aβ, BACE1 represents a strong candidate for therapeutic intervention in AD(Hampel et al., 2020).

Moreover, BACE1 processes other substrates that appear to be involved in synaptic plasticity.In animal model, halting BACE1 activity leads to reduction of Aβ burden and improvement cognitive performance (Truong et al., 2010).As often observed in other research areas,the promising pre-clinical animal studies using BACE1 inhibitors, did not translate into successful clinical treatments, possibly because Aβ is not the only target of BACE1.The knowledge of biology and physiology of this enzyme is limited. BACE1 showed higher expression/activity in the brain (in the vicinity of Aβ plaques), cerebrospinal fluid as well as in plasma/serum of AD patients compared to cognitively normal controls (Wu et al.,2012; Shen et al., 2018; Hampel et al., 2020).Intriguingly, this change seemingly occurs in the early preclinical stage of the disease and correlate with the levels of Aβ deposition, as assessed by positron emission tomography global standard uptake value ratios (Hampel et al., 2020). The current state of art suggests that BACE1 may function as a stress response protein that is upregulated in AD by different conditions including chronic hypoxia due to brain atherosclerosis and related oxidative stress, chronic inflammation, and endothelial dysfunction (Tamagno et al., 2012; Hampel et al., 2020). However, the upstream of this upregulation remains elusive. NoBACE1gene variants have been associated to higher risk of AD. Thus, the most likely culprit should be sought among the many epigenetic (DNA methylation and acetylation, micro-RNA) and post-translational (oxidation, glycosylation and sumoylation), regulators of this protease.

Our recent two studies (Cervellati et al., 2020;Zuliani et al., 2020) have been conducted with the aim to shed some light into the understanding of the implication of BACE1 in AD and its possible use as diagnostic biomarker for the disease

Figure 1｜Role of beta-site amyloid precursor protein cleaving enzyme-1 (BACE1) in Alzheimer’s disease onset.

Evaluation of serum BACE1 activity in AD and other diseases leading to dementia:We believe that our latest population-based study on 598 subjects adds another piece in the still unsolved puzzle (Zuliani et al., 2020).We extended previous findings from a smaller sample (n= 266) (Cervellati et al., 2020) and showed, for the first time, that increased peripheral BACE1 activity is not a specific feature of AD, but also occurs in vascular dementia (VAD) and mixed dementia (increase with respect to cognitively healthy controls: +30% , 35% and 22%, respectively). Interestingly,serum BACE1 levels were not significantly increased in different types of dementia such as frontotemporal dementia and Lewy body disease.

In our opinion, the study gives good hints for both theoretical (AD pathobiology) and applicative (diagnostic) aspects of the disease.BACE1 mechanistic link between AD and VAD:VAD, the second common cause of dementia among elderly, is a heterogeneous group of brain disorders caused by cerebrovascular deterioration, with cortical/subcortical ischemic infarctions and leukoaraiosis. AD and VAD are considered as two different diseases from the nosological point of view, based on current diagnostic criteria (Iturria-Medina et al.,2016). However, growing experimental and epidemiological/clinical evidences suggest that they share many risk factors and important neuropathatological features (De La Torre 2004). Indeed, it has been widely documented that cerebrovascular pathology, including cerebral amyloid angiopathy (with amyloid deposition in the walls of leptomeningeal and cortical arteries) occur early during the pathological process in AD, influencing its clinical progression (Iturria-Medina et al., 2016).Accordingly, chronic brain hypoperfusion,and consequent hypometabolism of cerebral tissue, has been observed in the early stage of AD (Iturria-Medina et al., 2016). The coexistence of cerebrovascular abnormalities and neurodegenerative hallmarks has been also mechanistically explained, by proposing a detrimental interplay between Aβ and neurovascular injury: the deposition of aberrant peptides in cerebral vessel causes the damage,which, in turn leads to Aβ deposition in senile plaques (besides functional impairment and brain atrophy) (Iturria-Medina et al., 2016).

“Orthodoxically” speaking, an elevation of BACE1 can be expected in AD, but not VAD.Indeed, according to the still widely advocated“amyloid cascade hypothesis”, the increase of Aβ via amyloidogenic process is an AD-specific phenomenon. However, there is compellingin vitroand animal evidence that links BACE1 with vascular dysregulation and can account for our findings (Hampel et al., 2020). BACE1 is suggested to be highly sensitive to disturbances in energy metabolism in the brain. These are most often outcomes of acute (micro/macro infarctions) or chronic brain hypoperfusion.Brain hypoperfusion reduces the delivery of oxygen and glucose, inducing local hypoxia,oxidative stress and inflammatory processes (De La Torre, 2004; Iturria-Medina et al., 2016).

It has been observed that hypoxic/ischemic insults increase the expression and activity level of BACE1 (but not α-secretase) in neuronal cells, capillary endothelial cells and affected brain regions (Salminen et al., 2017). The effect might be mediated by Hypoxia-inducible factor-1 binding on a controlled the expression of BACE1 gene through the binding in hypoxia response element sequence present in BACE1 promoter (Salminen et al., 2017). Moreover,up-regulation of the proteins could occur via stimulation of reactive oxygen species(ROS) mediated by hypoxia-induced sudden interruption of the mitochondrial electron transport chain (Tamagno et al., 2012).

Derangement of redox homeostasis, leading to oxidative stress is both cause and effect of vascular dysfunction and metabolic dysregulation (Cervellati et al., 2016). Brain and systemic signatures of oxidative damage have been widely found in association with AD. The rate of oxidative by-products, generated by excess in ROS, is related to cerebral damage,and ample human and experimental evidences support the importance of oxidative stress in the pathogenesis of this disease (Cervellati et al., 2016). There is also a significant correlation between BACE1 activity and oxidative markers(e.g. 4-hydroxynonenal) in AD brains (Tamagno et al., 2012). It has been suggested that ROS may increase the expression and the activity of BACE1 through transcriptional, translational and posttranslational mechanisms. For example,in vitroevidence showed that treatment with a mild ROS (hydrogen peroxide) increases the BACE1 promoter activity, enhancing the production of Aβ peptides. Overexpression of BACE1 can also be induced by microRNAs,which are mostly regulated through redoxdependent pathways.

Compellingly, BACE1 could also act as downstream player for cerebrovascular dysfunction. Indeed, as outlined above, Aβ possess vasoactive proprieties. A direct proof of this vascular effect of BACE1 has been recently provided (Durrant et al., 2020), showing that inhibition of the enzyme halted aberrant angiogenesis in AD animal model.

To conclude, the finding of high BACE1 levels in both LOAD and VAD might have the following explanation: BACE1 dysregulation along with AD pathology accumulation might be the cornerstone of both AD (no/minimal cerebrovascular disease) and VAD (significant cerebrovascular disease); owing these considerations, the real existence of pure VAD from the nosological point of view could be at least questionable.

Serum BACE1 as candidate biomarker of the diagnosis of AD (and VAD):At present, the definitive diagnosis of AD can be made on the basis of brain biopsies or autopsies. In living patients, the diagnosis of probable AD (but also VAD) relies on clinical criteria, supported by a combination of biomarkers assessed through cerebrospinal fluid (CSF) and brain imaging.However, both types of biomarkers are costly.Moreover, although CSF could represent an ideal diagnostic tool (given its physical vicinity with the brain it reflects pathophysiological changes in AD) the withdrawal procedure is invasive and not well-tolerated procedure, in particular in elderly patients.

Blood-based biomarkers are the ideal solution for overcoming this shortcoming, since they are widely accessible, minimally invasive, and more time- and cost-effective than the others.Besides, this concept is applicable also in a more general clinical context. Indeed, according to the Food and Drug Administration (FDA),an ideal biomarker should be measured in standard biological sources such as serum and urine (Khoury and Ghossoub, 2019). In AD field, a blood-based biomarker can be object of definitive analytical and diagnostic validation whether it fulfill some basic criteria:(1) it discriminates with good accuracy patients with AD from both mild cognitively impaired,cognitively normal individuals, as well as other forms of dementia-related disease; (2) it is involved in the pathogenesis of the disease and its brain levels is altered in association with disease; (3) its level reflects that of diseasespecific neuropathological hallmark .

The evidence accumulated so far clearly suggests that serum/plasma BACE 1 activity fulfills most of these criteria. We have shown that serum BACE1 performs well in discriminating AD(area under curve, AUC = 77%) and VAD (AUC =83%) from cognitively healthy controls (Zuliani et al., 2020). On the contrary, its activity does not significantly change in patients with other forms of dementia. Besides, Shen et al. (2018)reported that in individuals with mild cognitive impairment, (which is widely considered as a prodromal phase of AD) higher levels of BACE1 plasma activity is associated with in an increased probability to develop AD. This study along with another, found a significant correlation with Aβ (as assessed in CSF or by positron emission tomography) and t-Tau(CSF), as well as the cognitive status (Hampel et al., 2020). Last but not least, the analytical method is fast and inexpensive, allowing high throughput assessment.

In conclusion, there are robust data pointing serum/plasma BACE1 as promising diagnostic tool to use in clinical setting. However, owing the marked complexity of AD pathophysiology,the fact that a single blood-based biomarker might have satisfactory diagnostic/prognostic performance is doubtful. Further studies are required to evaluate whether BACE1, alone or in combination with other markers, may have reliable applications in multiple contexts such as trial enrollment and monitoring, proof of mechanism, response and toxicity dose estimation.

The authors would like to thank Dr. Valentina Rosta and Prof. Alessandro Trentini for their excellent and essential contribution to the studies which inspired this work.

Carlo Cervellati*, Giuseppe Valacchi,Giovanni Zuliani

Department of Translational Medicine and for

Romagna, University of Ferrara, Ferrara, Italy(Cervellati C, Zuliani G)

Department of Biomedical and Specialist Surgical Sciences, University of Ferrara, Ferrara, Italy;Plants for Human Health Institute, Animal Science Department, NC State University, Kannapolis, NC,USA (Valacchi G)

*Correspondence to:Carlo Cervellati, PhD,crvcrl@unife.it.https://orcid.org/0000-0003-4777-6300(Carlo Cervellati)

Date of submission:November 27, 2020

Date of decision:January 15, 2021

Date of acceptance:February 5, 2021

Date of web publication:April 23, 2021

https://doi.org/10.4103/1673-5374.313041

How to cite this article:Cervellati C, Valacchi G,Zuliani G (2021) BACE1 role in Alzheimer’s disease and other dementias: from the theory to thepractice. Neural Regen Res 16(12):2407-2408.

Copyright license agreement:The Copyright License Agreement has been signed by all authors before publication.

Plagiarism check:Checked twice by iThenticate.

Peer review:Externally peer reviewed.

Open access statement:This is an open access journal, and articles are distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike 4.0 License, which allows others to remix, tweak, and build upon the work non-commercially, as long as appropriate credit is given and the new creations are licensed under the identical terms.

Open peer reviewers:Baruh Polis, Bar-Ilan University, Israel; Rodolfo Pinto-Almazán, Hospital Regional de Alta Especialidad de Ixtapaluca,Mexico.