Zipora Pittel

M1 muscarinic agonists can modulate some of the hallmarks in Alzheimer's disease: implications in future therapy.

M1 muscarinic receptors (M1 mAChRs) play a role in an apparent linkage of three major hallmarks of Alzheimer’s disease (AD): β-amyloid (Aβ) peptide; tau hyperphosphorylation and paired helical filaments (PHFs); and loss of cholinergic function conducive to cognitive impairments. We evaluated the M1 muscarinic agonists AF102B (Cevimeline, EVOXAC™: prescribed for Sjøgren’s syndrome), AF150(S), and AF267B on some of these hallmarks of AD. Activation of M1 mAChRs with these agonists leads, inter alia, to enhanced secretion of amyloid precursor protein (α-APP), (via α-secretase activation), to decreased Aβ (via γ-secretase inhibition), and to inhibition of Aβ- and/or oxidative stress-induced cell death. In several animal models mimicking different aspects of AD, these drugs restored cognitive impairments, and in select cases induced a decrease in brain Aβ elevation, with a high safety margin, following po administration. Notably, in mice with small hippocampi, unlike rivastigmine and nicotine, AF150(S) and AF267B restored cognitive impairments also on escape latency in a Morris water maze paradigm, in reversal learning. Studies from other labs showed that AF102B and talsaclidine (another M1 agonist) decreased cerbrospinal fluid (CSF) Aβ in AD patients following chronic treatment, being the first reported drugs with such a profile. The clinical significance of these studies remains to be elucidated, yet based on in vivo (rabbits) and in vitro studies (cell cultures), our M1 agonists can decrease brain Aβ, owing to a novel and dual complementary effect (e.g., inhibition of γ-secretase and activation of α-secretase). Remarkably, although M1 agonists can decrease CSF Aβ in AD patients, an increased AD-type pathology in Parkinson’s disease was recently been associated with chronic antimuscarinic treatment. In another aspect, these agonists decreased tau hyperphosphorylation in vitro and in vivo. Notably, nicotinic agonists or cholinesterase inhibitors increased tau hyperphosphorylation. In summary, the M1 agonists tested are effective on cognition and behavior and show unique disease-modifying properties owing to beneficial effects on major hallmarks of AD. This may place such drugs in the first line of modern AD therapies (e.g., β- or γ-secretase inhibitors, vaccines against Aβ, statins, and inhibitors of tau hyperphosphorylation).

Mitogen-Activated Protein Kinase-Dependent and Protein Kinase C-Dependent Pathways Link the m1 Muscarinic Receptor to β-Amyloid Precursor Protein Secretion

Abstract: Full and functionally selective M1 muscarinic agonists (carbachol and AF102B, respectively) activate secretion of the soluble form of amyloid precursor protein (APPs) in PC12 cells expressing the m1 muscarinic receptor (PC12M1 cells). This activation is further augmented by neurotrophins such as nerve growth factor and basic fibroblast growth factor. Muscarinic stimulation activates two transduction pathways that lead to APPs secretion: protein kinase C (PKC)‐dependent and mitogen‐activated protein kinase (MAPK)‐dependent pathways. These pathways operate in parallel and converge with transduction pathways of neurotrophins, resulting in enhancement of APPs secretion when both muscarinic agonist and neurotrophins stimulate PC12M1 cells. These conclusions are supported by the following findings: (a) Only partial blockade of APPs secretion is observed when PKC, p21ras, or MAPK is fully inhibited by their respective specific inhibitors, GF109203X, S‐trans,trans‐farnesylthiosalicylic acid, and PD98059. (b) K252a, which blocks PKC and phorbol 12‐myristate 13‐acetate‐induced APPs secretion, enhances both muscarinic‐stimulated MAPK activation and APPs secretion. (c) Activation of MAPK in PC12M1 cells by muscarinic agonists is Ras‐dependent but PKC‐independent and is enhanced synergistically by neurotrophins. These results suggest that muscarinic stimulation of APPs secretion is mediated by at least two independent pathways that converge and enhance the signal for APPs secretion at the convergence point.

Large-scale purification and long-term stability of human butyrylcholinesterase: a potential bioscavenger drug

Butyrylcholinesterase from human plasma (HuBChE) is a potential drug candidate for detoxification of certain harmful chemicals that contain carboxylic or phosphoric acid ester bonds. Large quantities of purified HuBChE, displaying a high stability upon long-term storage, are required for the evaluation of its therapeutic capacity and its pharmaceutical properties. Several modifications of a previously reported procedure enabled us to purify the enzyme > 15,000-fold from pools of up to 100 1 of human plasma. The three-step procedure is based on precipitation of plasma proteins by ammonium sulfate (step I) and batch adsorption of HuBChE on procainamide-Sepharose 4B gel (step II). Ammonium sulfate was also employed in the third stage to fractionate the final product from procainamide-containing HuBChE solution. The overall yield (63%) of electrophoretically pure enzyme was significantly higher than that previously reported (34%) for the purification of HuBChE from 12.5 1 of plasma or from 5 kg of Cohn fraction IV-4. Purified HuBChE was stored at 5 degrees C in 10 mM phosphate buffer (pH 7.4) containing 1 mM EDTA and 0.02% NaN3. The specific activity, protein migration on gel electrophoresis, thermostability at 54 degrees C and the mean residence time in the circulation of mice remained essentially constant for at least 46 months. The modifications introduced can provide large quantities of purified enzyme that maintains its activity and bioavailability properties for several years.

AF150(S) and AF267B: M1 muscarinic agonists as innovative therapies for Alzheimer's disease.

The M1 muscarinic agonists AF102B (Cevimeline, EVOXACTM: prescribed in USA and Japan for Sjogren’s Syndrome), AF150(S) and AF267B—1) are neurotrophic and synergistic with neurotrophins such as nerve growth factor and epidermal growth factor; 2) elevate the non-amyloidogenic amyloid precursor protein (α-APPs) in vitro and decrease β-amyloid (Aβ) levels in vitro and in vivo; and 3) inhibit Aβ- and oxidative-stress-induced cell death and apoptosis in PC12 cells transfected with the M1 muscarinic receptor. These effects can be combined with the beneficial effects of these compounds on some other major hallmarks of Alzheimer’s disease (AD) (e.g. τ hyperphosphorylation and paired helical filaments [PHF]; and loss of cholinergic function conducive to cognitive impairments.) These drugs restored cognitive impairments in several animal models for AD, mimicking different aspects of AD, with a high safety margin (e.g. AF150[S] >1500 and AF267B >4500). Notably, these compounds show a high bioavailability and a remarkable preference for the brain vs. plasma following po administration. In mice with small hippocampi, unlike rivastigmine and nicotine, AF150(S) and AF267B restored cognitive impairments also on escape latency in a Morris water maze paradigm in reversal learning. Furthermore, in aged and cognitively impaired microcebes (a natural animal model that mimics AD pathology and cognitive impairments), prolonged treatment with AF150(S) restored cognitive and behavioral impairments and decreased τ hyperphosphorylation, PHF and astrogliosis. Our M1 agonists, alone or in polypharmacy, may present a unique therapy in AD due to their beneficial effects on major hallmarks of AD.

Reversible and irreversible inhibition of high-affinity choline transport caused by ethylcholine aziridinium ion.

The effect of ethylcholine aziridinium ion (AF64A) on choline transport in hippocampal, striatal, and cerebrocortical synaptosomes was studied. Synaptosomes prepared from these three brain regions were equally sensitive to AF64A. Low concentrations of AF64A produced a reversible inhibition (IC50 values = 1.35–2.25 μM), whereas higher concentrations produced an irreversible inhibition (IC50 values = 25–30 μM), which started as competitive. The irreversible component of the inhibition was independent of extracellular Na+ concentration, a finding suggesting that the choline transporter is alkylated at its outward position. The kinetics of the inhibition were rapid and similar in the three brain regions examined. The high‐affinity choline transport was more sensitive to the toxin than the low‐affinity choline transport. Based on these results, we propose a kinetic model that explains the reversible and the irreversible inhibitions induced by AF64A. The possible relationships between the concentrations that in vitro produce reversible and irreversible inhibition and those that in vivo produce selective and nonselective cholinergic hypofunc‐tion are discussed.

Muscarinic control of amyloid precursor protein secretion in rat cerebral cortex and cerebellum

It was previously shown by us and by others that activation of muscarinic acetylcholine receptors evoke amyloid precursor protein (APP) secretion in various cell lines. Here we examined if such muscarinic control of APP secretion occurs also in normal brain tissues. We found that the secretion of APP from rat cerebrocortical slices (rich in M1 receptors) was significantly increased by K+ depolarization, the non-selective agonist, carbachol (CCh), and the M1-selective agonist, AF102B. CCh also increased APP secretion from cerebellar slices (rich in M2 receptors) while AF102B had no significant effect in this brain region. Despite of its stimulatory effect on APP release in the cerebellum, CCh had no effect on phosphoinositide (PI) metabolism in this brain region. In the cerebral cortex PI metabolism was significantly increased by CCh but only partially increased by AF102B. These results suggest that APP secretion in the brain is mediated via muscarinic receptors. In the cerebral cortex APP secretion seems to be regulated via M1 receptors. Our results also suggest that PI metabolism is not a pronounced step in mediating APP processing.

Distinct Muscarinic Receptor Subtypes Differentially Modulate Acetylcholine Release from Corticocerebral Synaptosomes

The effect of McN‐A‐343 and oxotremorine on acetylcholine (ACh) release and choline (Ch) transport was studied in corticocerebral synaptosomes of the guinea pig. The synaptosomes were preloaded with [3H]Ch after treatment with the irreversible cholinesterase inhibitor, diisopropyl fluorophosphate, and then tested for their ability to release isotope‐labeled ACh and Ch in the presence and absence of these agents. The kinetics of release were determined at the resting state (basal release) and in the presence of 50 mM K+. Under either condition, McN‐A‐343 enhanced the release of isotope‐labeled ACh, whereas oxotremorine inhibited the K+‐evoked release but had no effect on the basal release. The enhancing effect of McN‐A‐343 on basal ACh release was fully blocked by the selective M1 muscarinic antagonist, pirenzepine (100 nM). In contrast to its enhancing effect on ACh release, McN‐A‐343 potently inhibited Ch efflux as well as Ch influx. These effects were not blocked by atropine, a nonselective muscarinic antagonist. Oxotremorine had no effect on Ch transport. Binding studies showed that McN‐A‐343 was 3.6‐fold more potent in displacing radiolabeled quinuclidinyl benzilate from cerebral cortex muscarinic receptors (mostly M1 subtype) than from cerebellar receptors (mostly M2 subtype), whereas oxotremorine was 2.6‐fold more potent in the cerebellum. The displacements of radio‐labeled pirenzepine and m‐dioxolane confirmed the M1 subtype preference of McN‐A‐343 and the M2 subtype preference of oxotremorine. These observations suggest that (a) activation of one subpopulation of muscarinic receptors by McN‐A‐343 enhances ACh release, whereas activation of another subpopulation by oxotremorine decreases this process (these subpopulations may correspond to the M1 and M2 subtypes, respectively), (b) the effect of the stimulatory receptors is pronounced at the resting state and that of the inhibitory receptors during depolarization, and (c) activation of the stimulatory receptors may be inversely related to the activity of the Ch transporter.

Selective Signaling via Unique Ml Muscarinic Agonistsa

Rigid analogs of acetylcholine (ACh) were designed for selective actions at muscarinic receptor (mAChR) subtypes and distinct second messenger systems. AF102B, AF150, and AF151 are such rigid analogs of ACh. AF102B, AF150 and AF151 are centrally active M1 agonists. AF102B has a unique agonistic profile showing, inter alia: only part of the M1 electrophysiology of ACh and unusual binding parameters to mAChRs. AF150 and AF151 are more efficacious agonists than AP102B for M1 AChRS in rat cortex and in CHO cells stably transfected with the ml AChR subtype. Notably, the selectivity of the new ml agonists is reflected also by activation of select second messenger systems via distinct G‐proteins. These compounds reflect a new pharmacological concept, tentatively defined as ligand‐selective signaling. Thus, agonist/m1AChR complexes may activate different combinations of signaling pathways, depending on the ligand used. Rigid agonists may activate a limited repertoire of signaling systems. In various animal models for Alzheimers disease (AD) the agonists AF102B, AF150 and AF151, exhibited positive effects on mnemomic processes and a wide safety margin. Such agonists, and especially AF102B, can be considered as a rational treatment strategy for AD.

Pharmacological basis for functional selectivity of partial muscarinic receptor agonists.

Muscarinic receptor agonists activate phosphoinositide hydrolysis and adenylate cyclase in Chinese hamster ovary cells transfected with cDNAs encoding the human muscarinic ml and m3 receptors. Whereas carbachol activates similarly both receptor subtypes, 4-[3-chlorophenyl-carbamoyloxy]-2-butynyltrimethyl ammonium chloride (McN-A-343) preferentially activates the m1 subtype over m3, in regard to both phosphoinositide hydrolysis and adenylate cyclase activity. On the other hand, oxotremorine activates phosphoinositide hydrolysis to a similar extent in both cell lines, but it activates preferentially adenylate cyclase in m1 versus m3 receptor expressing cells. Relative to carbachol, both McN-A-343 and oxotremorine activate preferentially phosphoinositide hydrolysis over adenylate cyclase in both cell lines. Prolonged incubation of cells with either carbachol, McN-A-343, or oxotremorine down-regulated the m1 receptors. This was accompanied by a parallel decrease in adenylate cyclase activity, whereas phosphoinositide hydrolysis remained relatively high. Inactivation of the receptors by alkylation with acetylethylcholine mustard, or by blocking with atropine, reduced carbachol-stimulated adenylate cyclase activity more effectively than carbachol-induced phosphoinositide hydrolysis in both m1 and m3 receptor expressing cells. These findings imply that the receptor reserve in these cell lines is greater for phosphoinositide hydrolysis response than for adenylate cyclase response. Yet, the receptor reserve for each of these responses is similar in both m1 and m3 receptor expressing cells. Since the binding affinities of McN-A-343 and of oxotremorine to m1 and m3 receptors are very similar, and since both cell lines contain similar amounts of spare receptors, we propose that the preferential activation of muscarinic m1 over m3 receptor by partial agonists is related to differences in the abilities of the two receptor subtypes to undergo conformational changes following agonist binding. This hypothesis is supported by results showing that the muscarinic m1 but not m3 receptor exhibits two affinity states in a competition binding assay.

AF64A Induced Cholinotoxicity: Behavioral and Biochemical Correlates

The postulated involvement of the cholinergic system in Alzheimer’s disease (AD) has highlighted the research and therapeutic approach of this affliction during the last decade (1). There is little doubt at the present stage that a clear hypofunction of the cholinergic system is in evidence in certain brain areas in AD patients. Other neurotransmitter systems seem to be relatively unaffected (1).