Anti-Neurodegenerative Benefits of Acetylcholinesterase Inhibitors in Alzheimer’s Disease: Nexus of Cholinergic and Nerve Growth Factor Dysfunction

Abstract

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder that is increasingly viewed as a complex multi-dimensional disease without effective treatments. Recent randomized, placebo-controlled studies have shown volume losses of ~0.7% and ~3.5% per year, respectively, in the cholinergic basal forebrain (CBF) and hippocampus in untreated suspected prodromal AD. One year of donepezil treatment reduced these annualized rates of atrophy to about half of untreated rates. Similar positive although variable results, have also been found in volumetric measurements of cortex and whole brain in patients with mild cognitive impairment as well as more advanced AD stages after treatments with all three currently available acetylcholinesterase (AChE) inhibitors (donepezil, rivastigmine, and galantamine). Here we review the anti-neurodegenerative benefits of AChE inhibitors and the expected parallel disease-accelerating impairments caused by anticholinergics, within a framework of the cholinergic hypothesis of AD and AD-associated loss of nerve growth factor (NGF). Consistent with the “loss of trophic factor hypothesis of AD”, we propose that AChE inhibitors enhance acetylcholine-dependent release and uptake of NGF, thereby sustaining cholinergic neuronal viability and thus slowing AD-associated degeneration of the CBF, to ultimately delay dementia progression. We propose that improved cholinergic therapies for AD started early in asymptomatic persons, especially those with risk factors, will delay the onset, progression, or emergence of dementia. The currently available competitive and pseudo-irreversible AChE inhibitors are not CNS-selective and thus induce gastrointestinal toxicity that limits cortical AChE inhibition to ~30% (ranges from 19% to 41%) as measured by in vivo PET studies in patients undergoing therapy. These levels of inhibition are marginal relative to what is required for effective symptomatic treatment of dementia or slowing AD-associated neurodegeneration. In contrast, because of the inherently slow de novo synthesis of AChE in the CNS (about one-tenth the rate of synthesis in peripheral tissues), irreversible AChE inhibitors safely produce significantly higher levels of inhibition in the CNS than in peripheral tissues. For example, methanesulfonyl fluoride, an irreversible inhibitor reduces CNS AChE activity by ~68% in patients undergoing therapy and ~80% in cortical biopsies of non-human primates. The full therapeutic benefits of AChE inhibitors, whether for symptomatic treatment of dementia or disease-slowing, thus would benefit by producing high levels of CNS inhibition. One way to obtain such higher levels of CNS AChE inhibition would be by using irreversible inhibitors.

1. INTRODUCTION

AD is a devastating progressive degenerative brain disorder that affects 6.2 million in the U.S. and many millions more worldwide [1]. Since the first decade of the last century [2], AD has been primarily defined by pathology that includes senile plaques (now known to be accumulations of extracellular aggregated β-amyloid) and neurofibrillary tangles (now known to be intracellular aggregates of hyperphosphorylated tau) in postmortem brain. Despite decades of effort and hundreds of failed clinical trials mainly targeting β-amyloid, no effective prevention, disease modification, or treatment for AD has emerged [3–5]. The problem of finding an effective disease-modifying treatment may be due to increasing evidence that there are different subtypes of AD [6,7] that begin many years before the onset of dementia as shown by varying biomarkers including amyloidosis [8,9], tauopathy [7,10,11], inflammation [12,13], and AD-associated loss of NGF [14,15]. Regardless of the particular biomarkers or subtype, however, all AD is associated with a loss of CBF cholinergic function necessary for cognition [16].

In view of the complexity of AD, there is growing interest in developing treatments that target key downstream pathophysiological steps that will slow neurodegeneration, delay the onset of mild cognitive impairment (MCI), or prevent the transition from MCI to clinically diagnosed AD dementia [17,18]. One such treatment that targets a downstream step and is able to slow advancement through clinical stages of AD is long-term use of AChE (EC 3.1.1.7) inhibitors [19–25]. Besides delaying clinical deterioration, AChE inhibitors also slow AD-associated atrophy in the CBF, hippocampus, cortex, and whole brain in prodromal (presymptomatic) AD [26–28] and even in mild/moderate stages of dementia [29–31]. Taken together, these findings strongly suggest that AChE inhibitors warrant further consideration as disease-modifying secondary prevention therapies [32,33]. The purpose of this review is to examine the anti-neurodegenerative benefits of AChE inhibition within a framework of cholinergic and NGF dysfunction during the early prodromal stages of AD [34–37]. The hypothesis being explored is that AChE inhibitors stimulate acetylcholine-dependent release and uptake of NGF and thereby produce anti-neurodegenerative benefits in the CBF in AD.

2. THE CHOLINERGIC SYSTEM IN AD AND OTHER DEMENTING DISEASES

It has been long known that the hippocampus, neocortex, and amygdala accumulate high densities of neurofibrillary tangles and show a loss of the enzymes responsible for both the synthesis and metabolism of acetylcholine in AD [38]. There is specifically a severe loss of CBF neuronal cell body immunostaining, especially in the nucleus basalis of Meynert (nbM) [39], the main source of acetylcholine for limbic and neocortical areas [40,41]. The absence of acetylcholine in the neocortex, hippocampus, and related terminal field networks is the proximate cause of the severe cognitive loss in AD dementia [16,42]. Loss of CBF basal forebrain volume, at least in some cases, precedes and predicts the cortical spread of AD [43,44] and is associated with the development of MCI and AD [45–49]. Magnetic resonance imaging (MRI) studies show that AD-associated neurodegeneration occurs in the CBF, hippocampus and neocortex in prodromal as well as in more advanced stages of the disease [9,26–28,43,44,50,51]. There is also severe damage to the nbM in the related disorders of Lewy body dementia [52], Creutzfeldt-Jakob disease [53,54], and frontotemporal dementia [55], identifying other conditions in which improving acetylcholine levels would likely be beneficial.

3. CHOLINERGIC TONE AFFECTS THE RATE OF CLINICAL PROGRESSION IN AD

AChE inhibitors were originally introduced as symptomatic treatments for AD dementia to amplify the impact of the endogenous acetylcholine remaining in the neocortex and hippocampus to improve cognition, but with no expectation that they could affect the rate of AD progression [33,56,57]. However, multiple retrospective studies have confirmed that patients on long-term AChE inhibitor therapy experience slower advancement through increasingly more severe clinical stages of AD [19–24]. For example, a matched cohort of ~17,000 AD patients followed for an average of 5 years showed that donepezil, rivastigmine, or galantamine use was associated with dose-dependent higher cognitive scores and slower cognitive decline as compared to nonusers. However, only galantamine, an AChE inhibitor that is also an allosteric potentiating ligand for nicotinic acetylcholine receptors (nAChR) [58], reduced the risk of advancing to severe dementia [25]. Also, as expected from their opposing pharmacological mode of action, long-term use of CNS-penetrating anticholinergics are associated with increased AD incidence, accelerated progression from normal cognition to MCI, and a higher risk of advancing from MCI to outright dementia [59–67]. Taken together, these data indicate that increasing normal cholinergic tone (via AChE inhibition) or deceasing cholinergic tone (via anticholinergics) slow or accelerate, respectively, the clinical progression of AD [33,68], demonstrating the key role of the cholinergic system in AD.

4. AChE AND BChE (EC 3.1.1.8) INHIBITORS SLOW CNS GRAY AND WHITE MATTER ATROPHY IN AD

CNS atrophy is a major biomarker of AD [69], with whole brain atrophy progressing at 2.3%/year in AD patients as compared to 0.4%/year in healthy age matched controls [70]. Hippocampal atrophy also occurs in normal aging but is also an early biomarker of AD [70,71], with hippocampal volumes in AD patients being consistently less than those in healthy controls [70,72–74]. The rate of hippocampal atrophy is greater and more variable in AD than in healthy controls, with means ranging from up to 15%/year vs 1.5%/year, respectively [74]. The rate of AD-associated atrophy of the hippocampus and cortex is also higher in APOE ε4 carriers [31,75] and MCI patients who often subsequently convert to AD [76].

4.1. Anti-neurodegenerative Benefits of Selective AChE Inhibition on Gray Matter

AChE inhibitors slow CNS atrophy in prodromal as well as in more advanced stages of disease. In suspected prodromal AD patients, randomized placebo-controlled trials (RCT) of short-term (1 year) treatment with donepezil, an AChE-selective inhibitor, produced ~60% reduction in the rate of basal forebrain CBF atrophy as compared to placebo-treated controls [27]. Donepezil also produced a similar ~55% reduction in the rate of hippocampal atrophy [28] and a nonsignificant trend toward reducing the rate of neocortex thinning [26] (Table 1). In MCI patients, RCT’s of donepezil (48 weeks) [33] or galantamine (two years), both of which are AChE selective inhibitors, failed to show benefit in the hippocampus [30], although both treatments reduced whole brain atrophy (Table 1). In patients in the galantamine trial, slowing of whole brain atrophy was limited to those who were APOE ε4 positive [30]. However, in patients with mild to moderate AD, short-term donepezil (24 weeks) produced a slowing of hippocampal atrophy [29] (Table 1). These results show that AChE inhibitor therapy has anti-neurodegenerative benefits in the CBF that sometimes extend to its gray matter postsynaptic projection areas in both prodromal and more advanced stages of AD.

Table 1.

Anti-Neurodegenerative Effects of AChE Inhibitors in Prodromal AD by Structure as Compared to MCI and Mild/moderate AD

Dementia Stage	Brain Structure	Duration	Drug1	Annualized % Change - Control	Annualized % Change - Drug	Drug Benefit (Annualized %)	Author/Year
Prodromal	Basal Forebrain	1 yr	DNZ 10 mg/d	−0.74% N=88	−0.30% N=75	+0.44% p=0.008	Cavedo et al. (2017) [27]
Prodromal	Hippocampus	1 yr	DNZ 10 mg/d	−3.47% N=92	−1.89% N=82	+1.58 % p<0.001	Dubois et al. (2015) [28]
Prodromal	Cortex (mean)	1 yr	DNZ 10 mg/d	−1.01% (SEM .25%) N=92	+0.075% (SEM 0.23%) N=82	+1.08% (see note below2)	Cavedo et al. (2016) [26]
MCI	Hippocampus	48 wk	DNZ 10 mg/d	−2.23% N=125	−1.86% N=105	+0.37% p=0.446 ns	Schuff et al. (2011) [31]
	Entorhinal cortex			−3.00% N=112	−0.85% N=100	+2.15% p=0.33 ns
	Whole brain			−0.54% N=90	+0.01% N=74	+0.55% p<0.001
MCI	Hippocampus	2 yr	GAL 16–24 mg/d	−1.70% N=160	−1.94% N=142	−0.24% ns	Prins et al. (2014) [30]
	Whole brain			−0.63% N=130	−0.46% N=112	+0.17% p<0.05
Mild/moderate	Hippocampus	24 wk	DNZ 10 mg/d	−17.7% N=31	−0.87% N=32	+16.8% p=0.01	Krishnan et. al. (2003) [29]

¹DNZ = donepezil, GAL = galantamine.

²Note: Significant drug effects shown in bold. Cavedo et al. [26] shows mean of seven neocortical areas, drug vs. placebo comparisons in all individual areas p<0.05 (same patients studied for hippocampal atrophy by Dubois et al. [28]).

4.2. Anti-neurodegenerative Benefits of Dual AChE and BChE Inhibition on White Matter

In contrast to gray matter (Table 1), the anti-neurodegenerative benefits seen in white matter may not be due entirely to AChE inhibition but may also involve BChE inhibition. For example, a 3 to 4-year retrospective analysis of rivastigmine, a mixed inhibitor of both AChE and BChE, showed reduced white matter loss and less ventricle enlargement in women, but not men, although there was no effect noted on hippocampal volumes [77]. A 20-week experiment comparing the anti-neurodegenerative effects of rivastigmine, donepezil, and galantamine in 30 patients with mild-to-moderate AD showed that only rivastigmine, but not the AChE-selective inhibitors donepezil or galantamine, preserved both neocortical white matter and gray matter [78]. The anti-neurodegenerative effect of rivastigmine in white matter is thus attributed to both AChE and BChE inhibition [79], both of which are involved in cholinergic signaling in the myelination of neurons [80]. AChE and BChE also impact muscarinic and nicotinic acetylcholine receptors that are expressed on myelinating oligodendroglia, and acetylcholine-dependent intracellular signaling that occurs in these myelinating glia [80]. BChE inhibition may thus be a useful therapeutic target in the treatment of AD [81].

5. CEREBROSPINAL FLUID (CSF) BIOMARKERS SUGGEST THAT ACHE INHIBITION MAY PRODUCE ANTI-NEURODEGENERATIVE BENEFITS BY STIMULATING CNS REGENERATION

Sustained, although sometimes enigmatic, changes in CSF cholinergic markers include increases and decreases in both AChE and BChE enzyme protein levels and enzymatic activities after long-term AChE inhibitor therapy. For example, patients treated with the readily reversible (competitive) AChE inhibitors donepezil and galantamine show paradoxical increases in CSF AChE activity [82–84]. On the other hand, treatment with the pseudo-irreversible inhibitor rivastigmine (dual AChE and BChE inhibitor) results in decreases in CSF AChE activity [83–85], but no change [84] or sometimes decreased levels of [83,85] BChE activity.

An increase or no change in CSF enzyme activities [82–85] is contrary to the a priori expectation that cholinesterase inhibitors would decrease AChE and/or BChE activity. However, the 20% to 35% increase in CSF AChE activity that followed long-term donepezil treatment was concurrently accompanied by 120% to 160% increases in CSF AChE protein content. The increase in AChE protein was initially interpreted as a compensatory upregulation of enzyme synthesis as a response to AChE inhibition [86]. However, in vivo PET scans of AD patients treated with galantamine show 30% to 36% inhibition of CNS AChE activity that is accompanied by an increase in the activity of CSF choline acetyltransferase (ChAT), the intracellular enzyme responsible for acetylcholine synthesis that is a biomarker of cholinergic neurons [87]. It is interesting to note that the galantamine treatment also resulted in a parallel increase in synaptic nAChR binding sites in most regions of the AD brain [87], a finding that is also reported after rivastigmine treatment [88] and after chronic donepezil or galantamine treatment in the brains of old rats [89]. Thus, changes in cholinergic neuronal and synaptic markers induced by AChE inhibitors suggest that they likely cause generalized synaptic regeneration, especially of nicotinic cholinoceptive neurons, as well as improved CNS cholinergic signaling [87].

6. THE INTERSECTION OF THE CHOLINERGIC AND NGF TROPHIC FACTOR HYPOTHESES IN THE BASAL CHOLINERGIC FOREBRAIN

The AChE inhibitor-induced increases in CNS nAChR binding [87–89] and CSF ChAT activity [87] reviewed above suggest that AChE inhibitor-induced anti-neurodegenerative benefits might be due, at least in part, to CBF nicotinic cholinergic stimulation that has been shown to upregulate NGF-related genes and prolong NGF release after direct administration of nicotine [90]. NGF levels in the hippocampus are also selectively upregulated in response to nicotinic cholinergic stimulation [91]. Therefore the purpose of the following section is to examine the reciprocal relationship between cholinergic modulation of NGF and the essential role of NGF the maintenance of CBF neurons, especially as related to AD-associated loss of NGF, cholinergic dysfunction in AD, and AChE inhibitor-induced anti-neurodegenerative benefits in AD.

6.1. Trophic Factors in the Normal Maintenance of CNS Neurons and Dysfunction in Disease

As early as 1981 it was proposed that maintenance of certain neuronal cell bodies depends on stimulating the tissues that receive their projections (target tissues) to release neurotrophic hormones (factors) that then undergo retrograde transport to maintain the viability of presynaptic neuronal cell bodies and it is the loss of these factors that result in neurodegeneration [92]. In accordance with this hypothesis, basal forebrain atrophy in rats, monkeys, and humans following cortical damage [93,94], disease [53,54], normal aging, and in various neurodegenerative disorders [95–97] is attributed to a loss of NGF retrograde transport. An example showing that neurotrophic support is essential for the survival of specific neuronal cell bodies in neurodegenerative disease is also found in Huntington’s disease, where the aberrant microtubule-associated protein, huntingtin, appears to be solely responsible for interfering with microtubule-dependent retrograde transport of brain derived neurotrophic factor (BDNF) and the specific pattern of striatal atrophy characteristic of Huntington’s disease [98]. In a parallel example relevant to AD, phospho-tau is associated with impaired microtubule-dependent NGF/TrkA retrograde signaling, reduced NGF trophic support, and reduced axonal trafficking-mediated support of CBF neurons in aged rats [99]. It is tempting to speculate that AD-associated phospho-tau may be responsible for impaired microtubule-dependent NGF/TrkA retrograde transport [34] and the AD-specific pattern of CBF basal forebrain atrophy [100]. Such a possibility is consistent with the early appearance of AD-associated patterns of aberrant tau [101] and the resulting progression of AD neurodegeneration. In any event, NGF metabolic pathways are known to be seriously dysregulated in both AD and Down’s syndrome [96].

6.2. NGF and Basal Forebrain Cholinergic Neuron Survival

As shown in (Fig. 1), acetylcholine (ACh) from cholinergic basal forebrain (CBF) projections stimulates activity-dependent synthesis and release of the NGF precursor protein (proNGF) in the neocortex and hippocampus, where it then undergoes extracellular conversion to mature NGF (mNGF) in the synapse [14,37,90,91,102–104]. Mature mNGF is then taken up into presynaptic terminals after binding to TrkA followed by its uptake into endosomes for microtubule-dependent retrograde transport of mNGF to CBF cell bodies [14,37,102,104–107]. Survival of CBF neuronal cells depends on this retrograde transport of NGF and its subsequent actions on the abundant NGF receptors localized to cholinergic neurons in the basal forebrain in both nonhuman primates [108] and humans [109,110]. NGF receptors in the neocortex are mainly limited to CBF projection areas [108,110]. The specific distribution of NGF and its receptors demonstrate a precise interrelationship between CBF neurons and their NGF-producing neocortical and hippocampal targets, areas that are both severely damaged in AD [110–113].

Fig. (1).

An illustration of a non-exhaustive list of AD-associated changes to CBF/NGF functions that support direct enhancement of NGF as a therapeutic strategy in AD [37]. As also reviewed elsewhere [15,37] these pathophysiological changes include impaired conversion of proNGF to mNGF (I in Fig. 1C) [114], a failure that causes: proNGF accumulation, reduced mNGF levels, [115], an imbalance in TrkA/p75 receptor ratio (II in Fig. 1C), impaired retrograde transport of NGF (III in Fig. 1C), and an increase in inflammation due to the loss of acetylcholine-mediated-stimulation of glial cell anti-inflammatory effects (IV in Fig. 1C) [37,116]. (Created using Biorender.com online software).

6.3. Anti-Neurodegenerative Benefits of AChE Inhibitors May Occur in Response To Acetylcholine-dependent Stimulation of NGF, Likely Mediated at Least in Part, by Nicotinic Cholinergic Receptors

It has long been appreciated that the loss of the CBF nbM is attributed to an AD-associated failure of NGF retrograde transport [117] that occurs during AD progression [15,34], including NGF-related changes that begin during prodromal AD stages [35,36,118]. The loss of NGF support is known to cause CBF neurons to lose their cholinergic phenotype yet still persist, but just in an atrophied state [119,120], which provides an opportunity to re-establish their normal functions if NGF levels can be restored. Therefore, we hypothesize that the anti- neurodegenerative benefits of long-term AChE inhibitor treatments [26–30] likely also result from direct acetylcholine-dependent-stimulation of NGF-induced regeneration as well as AChE inhibitor-induced increases in CNS nAChR upregulation [88–91].

6.4. The NGF/Cholinergic Hypothesis is not Exclusive of Other Pathophysiological Processes

AChE inhibition may exert anti-neurodegenerative effects through cholinergic modulation of AD- associated pathophysiological mechanisms other than NGF. For example, there are other complex interactions that occur between CBF cholinergic neurons, microglia and astrocytes that stimulate anti- inflammatory pathways (Figure 1C) [37,114,116,] that are mediated, at least in part, by α7-nAChRs [37,121]. In addition, AChE inhibition has more general effects, such as increasing regional cerebral blood flow and functional connectivity in cholinergic pathways in AD [122], which may directly or indirectly affect other pathophysiological processes like glucose hypometabolism and dysmyelination that occur in AD [123]. These various alternative mechanisms support the possibility that AChE inhibition affects AD through more than one pathway.

7. LIMITATIONS OF CURRENT ACHE INHIBITOR THERAPY

The main barrier to obtaining the full clinical benefit of AChE inhibitors in AD, whether for symptomatic treatment or anti-neurodegenerative benefit, is the inadequate level of CNS inhibition produced by the currently approved inhibitors [32]. Donepezil, galantamine, and rivastigmine produce a range of only 19% to 41% inhibition of neocortical AChE in AD patients undergoing therapy [124–128] (Table 2), levels that are marginal compared to the need for much greater than ~40% AChE inhibition that is required for optimal efficacy [129,130]. This is because they are not CNS-selective and, thus induce peripheral AChE inhibitor toxicity, especially in the gastrointestinal tract, which prevents the higher dosing essential for full cognitive or anti-neurodegenerative benefits [129,130]. Another inhibitor, metrifonate, a short-acting pseudo-irreversible AChE inhibitor [131,132], was also dose-limited by gastrointestinal toxicity, but its development was discontinued because it also caused respiratory depression and organophosphate-induced delayed neuropathy [133]. In summary, a need for higher levels of CNS AChE inhibition for treating AD [32,129,130] and the finding of a positive relationship between dose and cognitive benefit [25] strongly suggest that the real benefits of AChE inhibitors for symptomatic and anti-neurodegenerative success in AD has only begun to be explored because of the dose-limiting gastrointestinal toxicity caused by the currently available short-acting drugs [32] (Table 2). This has led to a call to identify more robust CNS cholinergic therapies [32,33] as will be described below.

Table 2.

CNS AChE Inhibition in AD patients with current therapies.

Author/Year	Drug1/Dose	Duration/Number of Subjects (N)	Structure	Mean Inhibition +/− SD2
Kuhl et al. (2000) [126]	DNZ 10 mg/d	5 wk (N=6)	Cerebral cortex	27% +/− 11% **
Kaasinen et al. (2002) [125]	DNZ 10 mg/d	3 mo (N=6)	Frontal Temporal Parietal	39% +/− 5.9% ** 29% +/− 11% * 28% +/− 17% *
	RIV 9 mg/d	3–5 mo (N=5)	Frontal Temporal Parietal	37% +/− 7% ** 28% +/− 16% * 28%+/− 18% ns
Bohnen et al. (2005) [124]	DNZ 10 mg/d	12 wk (N=14)	Mean cortical	19% +/− 9% ***
Kadir et al. (2008) [127]	GAL 16–24 mg/d	3 mo (N=10) 12 mo (N=11)	Mean cortical	36% +/− 5% *** 41% +/− 6% ***
Ota et al. (2010) [128]	DNZ 5 mg/d	5 mo (N=16)	Mean cortical	36% +/− 7% ***

¹DNZ = donepezil, RIV = rivastigmine, GAL = galantamine.

²*p<0.5, **p<0.01, ***p<0.001 inhibition compared to no inhibitor condition.

In contrast to the competitive and pseudo-irreversible AChE inhibitors described above, very high levels of CNS inhibition can be produced by irreversible AChE inhibition [32]. This type of inhibition dissipates only as new AChE enzyme is synthesized in each affected tissue [134]. The major advantage of irreversible inhibition, as illustrated below (Figure 2), is that de novo replacement of AChE in the CNS occurs at only about one-tenth the rate at which it occurs in peripheral tissues [132,134]. The naturally slow rate of CNS enzyme replacement thus results in the accumulation of very high levels of CNS AChE inhibition after repeated dosing (Figure 2) like that which would be used in long-term treatments for AD [32,134,135]. The stark difference in CNS AChE inhibition expected from a short-acting AChE inhibitor versus an irreversible inhibitor is demonstrated in Figure 2.

Fig. (2).

A side-by-side computational comparison of expected competitive versus irreversible AChE inhibition through 21 days of treatment. This figure shows the expected accumulated AChE inhibition in the brain (upper red line) vs the intestine (lower blue line) after three weeks of daily doses of a competitive inhibitor. Panel A: competitive inhibitor computations based on donepezil. The upward saw-tooth appearance of the lines show 25% inhibition in the CNS versus 20% inhibition in the intestine produced by each daily dose, a ratio of 1.25 selectivity toward the CNS. The downward slope between doses shows spontaneous recovery from donepezil-induced competitive inhibition with a half-time of 76 hours. Panel B: irreversible inhibitor computations based on methanesulfonyl fluoride. The upward saw-tooth appearance of the lines shows an equal increment of 10% inhibition of remaining active AChE in both intestine and CNS with each daily dose. The downward slope between doses is the reduction of AChE inhibition that is produced by new enzyme synthesis that occurs with a half-time of ~12 days in the CNS and a half-time of ~1 day in the intestine [134]. The separation between the accumulated AChE inhibition in CNS and intestines shown in Panel B is due entirely to the slow rate of de novo synthesis of enzyme protein in the CNS versus the peripheral tissues (from Moss et al. [32], with permission, see acknowledgements).

The hypothetical advantage of using an irreversible inhibitor to produce high CNS versus limited peripheral tissue AChE inhibition (Figure 2) has been tested in rats treated three times per week with methanesulfonyl fluoride for 3 weeks, a schedule that simulates actual clinical use of the same compound [135,136]. At 3 weeks, the animals were sacrificed and AChE activity was assayed in the CNS and peripheral tissues [135]. The finding of ~75% CNS AChE inhibition in this rat study (Figure 3) exceeded the expected ~65% shown in Figure 2B. Also in general agreement with the expected results shown in Figure 2B, AChE inhibition in the peripheral tissues remained at or below ~20% and did not induce toxicity. Our results shown in Figure 3 further confirm the high inherent CNS selectivity of irreversible AChE inhibitors is due to the inherent slow de novo protein synthesis of that enzyme in the CNS, which is only about one-tenth the rate of enzyme replacement found in peripheral tissues [134,135].

Fig. (3).

AChE Inhibition accumulated in rat gastrointestinal smooth muscle (ileum), skeletal muscle (pectoral muscle), and heart (cardiac muscle) versus CNS (whole brain) after three weeks of treatment with methanesulfonyl fluoride. ** Brain AChE inhibition (shown in red) is significantly more inhibited than peripheral tissues (shown in blue), while peripheral tissues are not different from each other. The error bars show the standard error of the mean. (from Moss et al. [135] with permission, see acknowledgements).

Comparable experiments in humans and nonhuman primates are consistent with the CNS AChE inhibition shown in Figure 3. In AD patients, methanesulfonyl fluoride produced ~70% CNS AChE inhibition as estimated by measuring erythrocyte AChE activity [136], an estimate that was later confirmed in cortical biopsies of nonhuman primates that showed ~80% AChE inhibition [32]. As expected from the high selectivity toward CNS AChE inhibition and minimal inhibition in peripheral tissues, neither humans nor nonhuman primates showed troublesome gastrointestinal toxicity or any other noticeable adverse effects [32,135,136]. Taken together, these experiments show that when freed from the barrier of dose-limiting gastrointestinal toxicity, high-level AChE inhibition (>50%) can be obtained and be well tolerated. The high CNS selectivity and low peripheral toxicity shown for methanesulfonyl fluoride would also be expected if using any safe irreversible AChE inhibitor [32]. Thus, irreversible inhibition is one strategy for producing the robust cholinergic enhancement needed to test the limits of AChE inhibitor-induced anti-neurodegenerative benefits [32,33,137].

8. DISCUSSION AND FUTURE DIRECTIONS

Successful disease-modifying interventions that slow AD-associated CNS atrophy can significantly mitigate the economic impact and suffering in AD [17,18,138]. For example, a hypothetical intervention initiated in 2025 that could delay dementia onset by only 5 years could reduce the estimated 13.5 million with AD in 2050 by about half to 7.8 million or even fewer [139]. Similarly, a hypothetical treatment that does not affect AD onset, but slows the progression to more advanced stages of dementia, would reduce the numbers of patients with severe dementia by ~80%, from 6.5 million to 1.2 million by 2050 [139].

Even in the face of the limited CNS inhibition produced by the currently approved AChE inhibitors (Table 2), they do have modest anti-neurodegenerative benefits in the basal forebrain and hippocampus (Table 1). It thus seems likely that if they were started up to two decades before onset of cognitive decline when the first signs of aberrant tau [10], β-amyloid or hippocampal atrophy [9] begin, they could significantly delay the onset of dementia or slow cognitive decline to more advanced stages of dementia. Such prophylactic use of the currently approved AChE inhibitors thus might be worthwhile for treating cognitively normal elderly, especially those who are at risk of AD like those with familial history, APOE ε4 carriers [18], or with abnormal CSF biomarkers linked to accelerated AD progression associated with β-amyloid or tauopathy [50].

Potential prophylactic use of AChE inhibitors in persons without an existing need for symptomatic treatment may, however, introduce ethical and other questions. Short-term tests of AChE inhibitors in cognitively normal healthy older subjects have been shown to produce no significant cognitive disturbance and only minor, if any, other effects [135,140,141]. The potential for AChE inhibitor use in cognitively normal persons thus requires further risk/benefit consideration [142].

Other cholinergic drugs, in addition to AChE inhibitors, are potential preventative agents against AD. For example, both nAChR and muscarinic acetylcholine (mAChR) receptor agonists, much like AChE inhibitors, affect a wide variety of AD-associated pathophysiological processes [80,116,143,144]. Both muscarinic and nicotinic receptor agonists also provide neuroprotection for the cholinergic system and thus are proposed to be rational therapeutic targets for AD [90,91,145–147].

SUMMARY AND CONCLUSION

AChE inhibitors produce neuroprotective/regenerative benefits as shown by slowing AD-associate atrophy of the basal forebrain and hippocampus (Table 1) and stimulate signs of CNS regeneration as shown by CSF biomarkers [87]. It is hypothesized that AChE-induced anti-neurodegenerative benefits are likely to be due to acetylcholine-dependent stimulation of NGF-induced protection of CBF cholinergic neurons [33,36,37], a possibility that deserves further study. Regardless of the mechanism of action, the consistent benefits of AChE inhibitor therapy reviewed here offer unparalleled hope as compared to the prior failed clinical trials of other agents [4,5]. Moreover, the prolonged presymptomatic period that exists in AD [9,10] offers ample opportunity for interventions with benefits compounded over several years, or decades. AChE inhibitors, especially those that yield high level CNS AChE inhibition [32,137], thus offer a new frontier for delaying the onset or slowing the relentless progression of AD.