Abstract
Alzheimer’s disease is a debilitating neurological disease placing significant burden on health care budgets around the world. It is widely believed that accumulation of amyloid-beta (Aβ) in the brain is a key event that initiates neurodegeneration, thus the clearance of Aβ from brain could be a key therapeutic strategy. Aβ exists in an equilibrium in healthy individuals, and recent research would suggest that dysfunction in the clearance pathways is the driving force behind its accumulation. One mechanism of clearance is proteolytic degradation by enzymes, and increasing the expression of these enzymes in animal models of Alzheimer’s disease has indeed shown promising results. This approach could be challenging to translate into the clinic given the likely need for genetic manipulation. We hypothesize that stimulating the activity of these enzymes (as opposed to increasing expression) through pharmacological agents will enhance degradation or at least prevent amyloid deposition, and is therefore another potentially novel avenue to manipulate Aβ levels for therapeutic purposes. We discuss the recent research supporting this hypothesis as well as possible drawbacks to this approach.
INTRODUCTION
The amyloid cascade hypothesis proposed over 20years ago [1, 2] still continues to stimulate research in the Alzheimer’s disease field. It has generated a wealth of data indicating the accumulation of amyloid-β (Aβ), in particular amyloid-β1 - 42 (Aβ1 - 42), is the key event that initiates neurodegeneration [3]. However, despite much basic research and the results of promising clinical trials on inhibitors of Aβ formation, the Aβ hypothesis has failed to produce clinically useful agents leading to the original amyloid hypothesis being recently re-valuated [4–6].
A key conceptual change that has emerged is the understanding that Aβ levels exist in an equilibrium between synthesis and clearance [7]. Aβ is thought to have physiological roles including its action as a transcription factor or a regulatory peptide [8, 9]. Therefore from a therapeutic perspective, complete elimination of Aβ from the brain would not be ideal. Recent studies have suggested that deficient clearance rather than increased production is the driving force behind accumulation of Aβ in the brain [10]. Therefore levels of Aβ can be manipulated by enhancing clearance to prevent the accumulation of harmful levels of amyloid moieties, which could eventually lead to neurodegeneration. Several mechanisms responsible for clearing Aβ have been identified [11, 12] and these include 1) uptake by phagocytes; 2) flow of brain interstitial fluid (ISF) into cerebrospinal fluid followed by drainage of ISF through basement membrane; 3) transport across the blood-brain barrier through interactions with the multi-ligand lipoprotein receptor LRP-1 as well as P-glycoprotein located on the outer surface of the cerebral endothelium [13]; and 4) degradation by enzymes [3, 14]. Although each one of these approaches are important for clearing Aβ, degradation by enzymes is one approach that can be easily manipulated through therapeutic agents. Increasing the activity of these enzymes can result in accelerated clearance of Aβ. Therefore this latter approach has attracted much attention in the recent past [7, 16].
PROPOSED HYPOTHESIS: AMYLOID-β DEGRADING ENZYMES AS A THERAPEUTIC TARGET
Increasing expression of Aβ degrading enzymes
At present, nearly 20 Aβ degrading enzymes have been identified [7] with the enzyme acyl peptide hydrolase being one of the most recent additions to the list [17]. However, neprilysin (NEP), endothelin converting enzyme (ECE), angiotensin converting enzyme (ACE), and insulin degrading enzyme (IDE) are considered the major players [7]. NEP is thought to be the most physiologically relevant enzyme in clearing Aβ [18], however determining the relative contribution of each enzyme in vivo requires further study. Several groups have attempted various strategies to overexpress NEP as a mechanism for manipulating Aβ levels. Strategies for overexpression include transgenic animal models [19, 20], delivery through viral vectors [21], and peripheral delivery [22], as well as pharmacological upregulation of expression [23]. Overall, the outcomes of these approaches appear to indicate that overexpression of NEP can reduce or delay Aβ plaque pathology (Fig. 1) [3]. However, the effect on the reversal of cognitive deficits is not clear. It is believed that continuous overexpression of NEP during mouse development can reduce the levels of other neuropeptide substrates of NEP, thus impacting cognitive function [3, 20]. Furthermore, some studies have suggested that the timing of overexpression is important and that earlier upregulation would be most beneficial [24].
Epigenetic approaches have also been used to manipulate the activity of Aβ degrading enzymes. These approaches have involved the use of small molecules and endogenous peptides. For example binding of neuropeptide hormone somatostatin to IDE increased its affinity for Aβ [25] and the expression of NEP [26], while the binding of norepinephrine to β2 adrenoceptor stimulated Aβ uptake as well as its breakdown by microglia [27]. In addition, GW742, a peroxisome proliferator receptor δ agonist has been shown to increase the expression of NEP in cell culture models, and reduce Aβ burden in 5xFAD mice [28]. Inhibitors of histone deacetylases such as valprioc acid is known to upregulate NEP expression. Valprioc acid is used in the treatment of epilepsy and its usefulness in the setting of neurodegenerative disease is not fully understood. However treatment with valprioc acid resulted in behavioral improvements in a mouse model of Alzheimer’s disease [29].
On the other hand, the effect of overexpressing ECE in the setting of Alzheimer’s disease has been studied less extensively. In one study, an indirect increase in ECE expression was achieved through transgenic overexpression of protein kinase C epsilon isoenzyme, and this did indeed lead to a reduction in Aβ pathology [15].
The various Aβ degrading enzymes and their cellular localizations are covered in a previous review and hence will not be discussed in detail here [7]. The widely known Aβ degrading enzymes such as NEP, ECE-1, and IDE are located in the plasma membrane, thus stimulating their activity is likely to facilitate the clearance of extracellular Aβ [7]. ECE-2 as well as certain homologs of ECE-1 are located within the cytosol and are therefore most likely to regulate the cleavage of intracellular Aβ [30, 31]. However, which specific pool of Aβ to target in order to achieve best outcome from a therapeutic perspective requires further research.
Stimulating the activity of amyloid-β degrading enzymes
Although the above-mentioned approaches have produced promising data in animal models, they pose challenges for clinical translation. This is particularly the case for strategies that involve genetic manipulation, which would likely require significant ethical hurdles to be overcome prior to embarking on clinical trials.
Therefore we propose a novel hypothesis being that a pharmacological agent which stimulates the activity of Aβ degrading enzymes, in particular that of NEP, will have the same physiological effect as increasing enzyme expression (Fig. 1). Theoretically, stimulation of enzyme activity is expected to tip the equilibrium in favor of clearance, thus preventing the accumulation of harmful Aβ moieties or even reversing plaques already formed. Therefore, stimulating the activity of these enzymes is a potential novel approach to manipulate the levels of Aβ for a therapeutic outcome. This approach appears to have been largely overlooked by researchers over the years, most likely reflecting the lack of appropriate molecules to stimulate enzyme activity. Aβ degrading enzymes have a basal level of activity and most belong to the family of zinc dependent metalloproteases. The feasibility of increasing the activity of any member of this family of proteins was first demonstrated by the discovery of two small molecules that enhanced the activity of angiotensin converting enzyme-2 (ACE-2) [32]. Subsequently small molecule activators of IDE were also reported [33]. The effect of these molecules on the activity of ECE or NEP is yet to be determined [33], and the in vivo efficacy of these molecules in the setting of Alzheimer’s disease is as yet unknown.
Further proof of the feasibility of this approach has come from our recent discovery of a peptide (which we called K49-P1-20), which is based on a toxin found in a snake venom that can directly enhance the activity of both NEP and ECE [34]. Our in vitro data indicate that K49-P1-20 mediated stimulation of ECE and NEP activity can accelerate the degradation of synthetic Aβ1 - 40 [34]. In addition, it stimulated ECE mediated breakdown of soluble Aβ1 - 42 in cerebrospinal fluid of Alzheimer’s disease patients [34]. To the best of our knowledge K49-P1-20 is the first known stimulator of ECE and NEP activity. Our in vitro results already suggest that K49-P1-20 can be made more selective for NEP [34]. Given that NEP is regarded as the most physiologically relevant enzyme in the clearance of Aβ, a drug lead more selective for NEP could be of significant clinical importance. Taken together, the discovery of K49-P1-20 enables the testing of our hypothesis that stimulating the activity of Aβ degrading enzymes (in particular NEP) will reverse Aβ plaque pathology and/or prevent the accumulation of harmful Aβ in the setting of Alzheimer’s disease.
Any pharmacological agent that increases the activity of Aβ degrading enzymes would need the ability to cross the blood-brain barrier. Given that K49-P1-20 is a 20 amino acid peptide it is unlikely to cross the blood-brain barrier in its native form. However, K49-P1-20 could be an excellent research tool to better understand the mechanism(s) behind the stimulation of enzyme activity and in itself could be a template for the development of peptidomimetics. Specifically, this new knowledge can in turn be used in medicinal chemistry approaches aimed at developing synthetic small molecule analogs of K49-P1-20 with pharmacodynamic/pharmacokinetic properties amenable for use in humans. Furthermore, these studies can be the basis for developing molecules that stimulate the activity of other Aβ degrading enzymes, thus having implications for the field of protease biology in general.
Potential pitfalls
Increasing enzyme activity is likely to have unwanted side effects given that these proteases often play a role in biochemical/metabolic pathways other than Aβ degradation. For example, in addition to its role an Aβ degrading enzyme, ECE plays a key role in the cardiovascular system, catalyzing the conversion of big endothelin (BigET) to the potent vasoconstrictor hormone endothelin-1 [35]. Therefore it is logical to assume that increasing the activity of ECE, or its expression levels would lead to enhanced endothelin-1 production and hence increased blood vessel contraction and hypertension. However, it must be noted that this effect will only occur if the increase in ECE expression or activity is accompanied by a corresponding increase in the substrate BigET. Furthermore, the effects of endothelin-1 can be effectively countered through clinically approved endothelin receptor antagonists such as bosentan.
NEP inhibition has been used as a form of therapy in heart failure patients [36, 37]. Therefore, our proposed approach of stimulating NEP activity as a potential approach to treat Alzheimer’s disease may have negative effects in the setting of heart failure. This underscores the importance of modifying this potential treatment approach to specifically target the delivery of appropriate pharmacological agents to the target tissues (i.e., brain). Given that Aβ accumulation is thought to be the event that initiates neurodegeneration, therapeutic interventions aimed at enhancing the activity of Aβ degrading enzymes are likely to be most effective in the early stages of the disease process. This therefore makes it most critical to identify the individuals most susceptible to and/or the development of biomarkers for the early onset of AD.
CONCLUDING REMARKS
Previous studies have shown that increasing the expression of Aβ degrading enzymes can reverse or prevent Aβ pathology in mouse models of Alzheimer’s disease. Given the possible challenges of translating this approach into the clinic, we propose an alternative approach of stimulating the activity of Aβ degrading enzymes through appropriate pharmacological agents. The feasibility of this approach has been demonstrated in vitro using the NEP/ECE stimulator K49-P1-20, as well as small molecules activators of IDE. Promising in vitro data obtained through these pharmacological leads we believe warrant in vivo testing of this hypothesis in well characterized animal models of Alzheimer’s disease.
DISCLOSURE STATEMENT
Authors’ disclosures available online (http://j-alz.com/manuscript-disclosures/16-0492r2).