John H. Growdon1
Summary. The neuropharmacological consequences of finding evidence for deficient acetylcholine neurotransmission in AD have been complex. The initial optimism for a quick cure from choline or lecithin precursor administration, inspired by the success of levodopa in Parkinson's disease, quickly faded when put to the test. Nonetheless, the cholinergic hypothesis of memory dysfunction in AD was valid, and eventually it led to the introduction of AChEI drugs to increase acetylcholine transmission. Drugs of this class are the mainstays of current treatment for AD, even though their effects are generally modest. In the search for improved symptomatic and possibly neuroprotective treatments, acetylcholine may have an unexpected role. The observation that Ml and M3 receptor stimulation with cholinergic drugs drives APP processing into the a-secretase pathway adds a modern coda to the acetylcholine-AD story that is still unfinished.
As my personal note on the history of Alzheimer's disease (AD) research, I would like to highlight a paper entitled "Increase in hippocampal acetylcholine after choline administration," by Madelyn J. Hirsch, John H. Growdon, and Richard J. Wurtman, (Hirsch et al. 1977) for sentimental reasons: it was my first paper on acetylcholine and it set the stage for investigations in AD that would occupy much of my subsequent career.
Neurologic Disease, Neurotransmitters and Neuropharmacology
In 1975, I joined Dr. Richard Wurtman's laboratory at the Massachusetts Institute of Technology as a postdoctoral research fellow. As a neurology resident, I had been intrigued by a clinical report that post-hypoxic intention myoclonus was linked to deficient serotonin neurotransmission (Lhermitte et al. 1972), which prompted me to develop an animal model of myoclonus produced by a serotonin neurotoxin (Stewart et al. 1976). I went to work in the Wurtman laboratory because he was a leading authority on serotonin metabolism, and I wished to learn more about the factors that controlled the synthesis and effects of this neurotransmitter. I knew that Wurtman and his colleague, John Fernstrom, had made a startling discovery: variations in plasma levels of the serotonin precursor amino acid tryptophan caused parallel changes in the amounts of serotonin synthesized in the brain (Fernstrom and Wurtman 1971). I had
1 Department of Neurology, Harvard Medical School and Massachusetts General Hospital, Boston, MA., USA
j ust joined the laboratory when Wurtman, along with a graduate student, Edith Cohen, published a paper showing that brain levels of another transmitter, acetylcholine, were regulated by the availability of choline, the natural dietary precursor for acetylcholine biosynthesis (Cohen and Wurtman 1976). These unexpected effects on brain of naturally occurring dietary constituents that are precursors for neurotransmitters raised the intriguing possibility that dietary substances could be used in the treatment of brain diseases (Growdon et al. 1977a).
Linking neurotransmitter abnormalities to neurologic disease was an exciting new approach to many neurodegenerative diseases (Moskowitz and Wurtman 1975). The most stunning example of neurotransmitter replacement treatment was, of course, Parkinson's disease: administration of the naturally occurring biochemical intermediate in dopamine synthesis, levodopa, produced dramatic suppression of many symptoms, including tremor and bradykinesia. A similar, although less dramatic, benefit occurred in post-hypoxic intention myoclonus following treatment with the serotonin biochemical intermediate L-5-hydroxy tryptophan (Growdon et al. 1976). As a prelude to determining whether cholinergic precursor treatment would benefit diseases linked to deficient acetylcholine neurotransmission, I thought it would be necessary to show that 1) the increase in acetylcholine induced by choline administration was in pre-synaptic terminals and thus available for synaptic release, 2) choline administration to humans would increase plasma levels of choline as it did in rats and 3) choline administration to humans would increase choline levels in brain or cerebrospinal fluid (CSF).
Our first paper examined hippocampal choline and acetylcholine levels after choline administration to answer the first requirement (Hirsch et al. 1977). The rationale behind this study was that acetylcholine in the hippocampus was largely confined to the axon terminals of the septo-hippocampal tract, and that acetylcholine was released from the hippocampus by septal electrical stimulation. Our study showed that a single interperitoneal injection of choline markedly elevated both choline and acetylcholine levels within the dorsal hippocampus, just as it did in the caudate nucleus and in the whole rat brain (Table 1). In a subsequent study, we showed that lecithin, the naturally occurring dietary source of choline, was more efficient at raising blood levels of choline in normal human volunteers than was a choline salt (Wurtman et al. 1977). Finally, in a separate clinical study, we confirmed that oral choline administration increased serum levels of choline and also produced substantial and significant increases in CSF choline levels (Growdon et al. 1977b)
The next step was to test precursor choline treatment in a neurologic disease linked to deficient acetylcholine tone. Tardive dyskinesia was one such condition. It had been known that intravenous administration of physostigmine decreased chor-eic movements whereas anti-cholinergic drugs such as scopolamine tended to worsen dyskinesia. We conducted a clinical trial in which we administered 8-20 grams per day of choline chloride to 20 patients with tardive dyskinesia, collected blood samples for choline measurements and counted the number of choreic movements over time. During the second week of choline ingestion, choreiform movements decreased substantially in five patients and moderately in four, whereas they were unchanged in 10 and worse in one (Growdon et al. 1977c). We found similar effects with oral doses of lecithin. Thus, administration of choline and lecithin to increase acetylcholine in brain was secure from a scientific standpoint. From a clinical perspective, both compounds
Table 1. Effect of choline chloride administration on choline and acetylcholine concentrations in rat hippocampus and caudate nuclei (modified from Hirsch et al. 1977)
Control Hippocampus Caudate 20 minutes after ChCl Hippocampus Caudate 40 minutes after ChCl Hippocampus Caudate
Groups of 10 rats received choline chloride (ChCl) (60 mg/kg, ip) or its diluent (water) and were killed 20 or 40 minutes after injection. Data are given as means +/- SEM.
had been tested in human subjects, both increased plasma and CSF levels of choline, and both had shown promise in treating a human disease associated with deficient cholinergic neurotransmission.
The Cholinergic Hypothesis of Memory Dysfunction in AD
Multiple and independent reports of a cholinergic deficit in AD (Bowen et al. 1976; Davies and Maloney 1976; Perry et al. 1977) appeared in 1976 and 1977 and signaled a new era in AD research. In addition to confirming the selective decrease in choline acetyltransferase (CAT) activity in AD brain, investigators reported preservation of intact muscarinic receptor sites. These observations raised the possibility that treatments designed to increase the synthesis or release of acetylcholine, or to block its subsequent hydrolysis, might be beneficial in treating AD symptoms. Prior pharmacological studies had already demonstrated the importance of intact cholinergic neurotransmission in memory functions. It had been widely known, for example, that anti-cholinergic drugs could impair memory and even produce amnesia. Drachman and Levitt extended this knowledge and showed that low doses of scopolamine produced a pattern of cognitive deficits that was qualitatively similar to those observed in demented patients (Drachman and Leavitt 1974). It was then shown that cholinergic agonists could enhance memory and reverse the adverse effects of scopolamine (Davis et al. 1978; Sitaram et al. 1978).
Pathologic studies also supported the cholinergic hypothesis of memory dysfunction and provided an explanation for the decreased CAT levels in AD brain. Decreased CAT activity in brains of AD patients correlated with estimates of dementia severity and with neurofibrillary tangle counts (Wilcock et al. 1982). Whitehouse et al. (1981) reported that the cholinergic neurons in the ventral forebrain, including the nucleus basalis of Meynert, whose axons project widely to neocortex and the hippocampus, were severely atrophic. Although atrophy of ventral forebrain nuclei is not unique to
AD, finding atrophic neurons was internally consistent with biochemical indices of decreased cholinergic transmission in the terminal projections of their axons. These three lines of evidence therefore formed the basis of the cholinergic hypothesis of memory dysfunction in AD. From a therapeutic standpoint, the parallels with Parkinson's disease were striking. In Parkinson's disease, neuronal loss in the substantia nigra pars compacta accounts for decreased production of dopamine; it is believed that deficient dopaminergic neurotransmission in the nigral projections to the striatum results in the characteristic extrapyramindal motor signs. Administration of levodopa to increase dopamine biosynthesis improves these motor signs. In AD, atrophy of the nucleus basalis of Meynert might be considered as a lesion comparable to the nigral damage in Parkinson's disease. As a result of the basalis lesion, there is deficient acetylcholine synthesis and reduced cholinergic neurotransmission in cholinergic axon terminal projections, especially to the hippocampus, which results in memory loss and perhaps other cognitive impairments. Could neurotransmitter replacement strategies similar to those used successfully in Parkinson's disease work to palliate cognitive impairments in AD?
Acetylcholine-related Treatments: The Reality Precursors
In contrast to the beneficial effects in tardive dyskinesia, neither oral choline nor oral lecithin administration improved any aspect of cognition or behavior in AD patients. A study of lecithin administration vs. placebo conducted by Sullivan et al. (1982) is illustrative. We treated 18 AD patients with either lecithin or placebo according to a double-blind cross-over design. Lecithin administration increased plasma choline levels two- to four-fold (p < 0.0001) whereas choline levels returned to baseline during periods of washout or placebo administration. Despite the increase in plasma choline levels, no patient improved on any memory test during lecithin or placebo administration.
Acetylcholine Esterase Inhibitors (AChEIs)
An alternative way to increase acetylcholine neurotransmission is to block its hydrolysis and thereby prolong the intrasynaptic effects of released ACh. Physostigmine was the only AChEI in the formulary at the time, and its use was severely restricted because of the requirement for intravenous administration, as no oral preparation with a long duration of action was available. Early studies with AChEIs appeared more promising than precursor administration, and drugs of this class were finally developed and approved nearly 20 years after the discovery of the cholinergic deficit in AD (Lleo et al. 2006). There are now four AChEIs available; these drugs are the current standard of care even though they typically produce only modest clinical benefits (Greenberg et al. 2000). Since their introduction into clinical practice 10 years ago, it has become clear that this line of treatment is sub-optimal and that AChEIs have not become the levodopa of AD. Although the cholinergic deficit is still true, it is an incomplete account of the total AD pathology; correcting acetylcholine transmission is a bandage at best and has little or no effect on the overall progressive deterioration of this illness. To stop AD or slow its progression, the focus has shifted to neuroprotective strategies designed to counteract the underlying causes of AD.
Amyloid deposition is an early event in brains of AD patients and defines much of the histopathology of AD. Amyloid plaques are composed of small peptide fragments called Ap, which are derived by protolytic cleavage of a large transmembrane amyloid precursor protein (APP). Much of the research in AD during the past decade has centered on the molecular aspects of APP processing andits consequences. As described elsewhere in this volume, at least three enzymes are involved in APP metabolism: a-secretase, which cleaves APP within the Ap sequence and generates non-amyloidogenic moieties, and another set of proteases (P- and y-secretases) that generate Api-42 and Api-40 fragments, which are believed to be neurotoxic. Roger Nitsch joined the Wurtman laboratory in 1990 and soon established the initial and most direct link between acetylcholine transmission and amyloid metabolism. We discovered that HEK 293 cells separately transfected with the muscarinic Ml and M3 receptor subtypes increased a-secretase cleavage of APP within minutes of stimulating the receptors with the cholinergic agonist carbacol (Nitsch et al. 1992). In subsequent experiments, we showed that stimulating the M1 receptor subtype also produced a concurrent decrease in Ap secretion (Hung et al. 1993). APP processing in wild type HEK cells, as well as those expressing the M2 and M4 receptor subtypes, was not affected by muscarinic stimulation, indicating that APPs secretion was specifically linked to the M1 and M3 receptor subtypes. We reasoned that if such a sequence were to occur in human beings, administration of an M1 agonist would be expected to decrease levels of Ap in the central nervous system and possibly slow or even reverse the course of AD dementia. To test this hypothesis at the biochemical level, we administered the selective M1 and M3 agonist AF102B to 19 patients with the clinical diagnosis of AD and measured the CSF levels of soluble Ap before and during drug administration. To determine the specificity of the AF102B effect, we administered two other drugs, hydroxychloroquin or physostigmine, to separate sets of AD patients. We found that treatment with AF102B lowered total Ap levels in CSF by 22% in 14 of the patients, whereas levels increased slightly in three and were unchanged in two (Nitsch et al. 2000). The overall decrease in the group as a whole was statistically significant. CSF Ap levels did not change significantly in the nine patients treated with physostigmine or in the 10 patients treated with hydroxychloroquin. These data provided evidence that specific activation of M1 receptors reduced total Ap levels in CSF of AD patients. If this effect were to occur in the brain, M1 agonists might have long-term therapeutic benefits by lowering the Ap load. This hypothesis is nowbeing actively tested in transgenic mice. Caccamo et al. (2006) administered the selective M1 muscarinic agonist AF267B to a triple transgenic mouse that over-expresses both amyloid and tau. They found that AF267B reduced both Ap and tau pathologies in hippocampus and cortex and that these changes were associated with improved performance in a spatial task. Further, they showed that the mechanism underlyingthe effect on Ap pathology was caused by the selective activation of ADAM17, thereby shifting APP processing toward the non-amyloidogenic pathway. In contrast, the M1 antagonist dicyclomine exacerbated the Ap and tau pathologies.
These pre-clinical and clinical experiments raise the possibility that drugs developed to stimulate specifically Ml and M3 muscarinic receptors might prove effective in lowering the Ap burden in human brain and thereby slowing or even reversing the cognitive decline associated with AD.
Amyloid and Genetics
Was this article helpful?
The comprehensive new ebook All About Alzheimers puts everything into perspective. Youll gain insight and awareness into the disease. Learn how to maintain the patients emotional health. Discover tactics you can use to deal with constant life changes. Find out how counselors can help, and when they should intervene. Learn safety precautions that can protect you, your family and your loved one. All About Alzheimers will truly empower you.