Marys Medicine

Pii: s0166-2236(98)01361-7

E. Perry et al. – Cholinergic components of consciousness
continued firing of nucleus-basalis cholinergic neur-ones, Buzsaki23 concluded ‘it appears that the ascendingcholinergic system alone is capable of keeping the neo-cortex in its operative mode'. However, the excitatoryglutamatergic input to the nucleus basalis that arisesin the brainstem24, together with the major thalamo-cortical glutamatergic projections, indicate that thecombined actions of ACh and glutamate are essentialin this respect. The consensus view on the role of cor-tical cholinergic projections is that they control selec-tive attention. As Delacour1 has speculated that selectiveattention and consciousness overlap, and Baars25 hashighlighted the importance of selective attention inthe ‘theatre' metaphor of consciousness, the two pro-cesses might share a common neural basis. While atten-tional processes can occur at the non conscious–implicit level, conscious awareness, which representsonly a fraction of cerebral activity at any time, clearlyinvolves a selection process. Current theories of therole of cortical ACh include the possibility that it affectsdiscriminatory processes; increases signal–noise ratios;modulates the efficiency of cortical processing of sen-sory and association information; controls the recep-tion and evaluation of stimuli for their level of signifi- Fig. 1. Cholinergic systems in the human brain. Two major pathways project widely to different
cance; modifies cortical responsiveness in terms of the brain areas: basal-forebrain cholinergic neurons [red, including the nucleus basalis (nb) and relevance and novelty; and confines the contents of the medial septal nucleus (ms)] and pedunculopontine–lateral dorsal tegmental neurons (blue). conscious stream3,26–32. Basal-forebrain cholinergic neur- Other cholinergic neurons include striatal interneurons (orange), cranial-nerve nuclei (green ones project not only to all cortical areas but also to circles), vestibular nuclei (purple); and spinal-cord preganglionic and motoneurons (yellow). A group of cholinergic neurons in the thalamic paracentral nucleus (not shown), thought to select thalamic nuclei, including the reticular nucleus33, project to striatum and visual cortex, has recently been identified in macaque brain101. The which has been implicated in selective attention34.
habenula–interpeduncular pathway is also not shown. Combined retrograde labelling and choline acetyl- transferase immunohistochemical studies have estab-lished that 85–95% of brainstem afferents to most novel patterns and cognitive or motor-skill learning thalamic nuclei, which include specific-relay, nonspe- have, however, been reported to be unimpaired37. This cific and reticular nuclei, originate in the rostral brain- suggests that it is not so much information storage and stem where pedunculopontine cholinergic nuclei and retrieval per se that are primarily compromised in AD, lateral dorsal tegmental nuclei are maximally devel- and that ‘cholinergic correlates' of cognitive impair- oped21. These inputs are excitatory and exert their effects ment might instead be correlates of the degree of un- both directly, via fast nicotinic-receptor mediated and awareness experienced by the patient. Lack of aware- slower muscarinic-receptor mediated depolarization, ness of cognitive impairment is common in AD, more and also indirectly via hyperpolarization of GABAergic so than in vascular dementia38, and is correlated with (inhibitory) reticular neurones35. Co-activation of brain- the degree of cognitive impairment39. Brainstem ped- stem and basal-forebrain cholinergic neurones that unculopontine cholinergic neurones are reportedly project rostrally, which occurs in both wakefulness unaffected in AD (Ref. 40) or modestly reduced (by and REM sleep, provides the thalamus and cortex with 30%)41. The latter finding is not only consistent with a role in integrative modulation of distant neurones the presence of intracellular neurofibrillary tangles but (synchronization) that could represent a component also with the presence of ‘ghost' tangles in this nucleus mechanism of conscious awareness.
(where the cell body is no longer identifiable).
In AD, different pathological manifestations, such Cholinergic neuropathology in mental disorders
as cortical and subcortical b-amyloidosis (which results Abnormalities in the cholinergic system have been in plaque formation), abnormal tau (which results in consistently identified in disorders that affect conscious the development of tangles and dystrophic neurites), awareness, which include AD, PD and DLB (reviewed neuronal and synapse loss, and various transmitter in Ref. 3). The pathology in the nucleus of Meynert, deficits, leave clinical-neuropathological correlations which includes neurofibrillary tangles, Lewy bodies or open to a variety of interpretations. Deficits in cholin- neurone loss, together with deficits in the cholinergic ergic neurotransmission are unlikely to account for the system in the cortex and hippocampus, detected at full spectrum of cognitive and non-cognitive symptoms.
autopsy, and more recently, using chemical imaging36, The situation is less complex in PD and DLB, in which has been linked to cognitive impairment and memory cortical neurofibrillary tangles are much rarer or ab- impairment, and might also relate to alterations in con- sent, and b amyloid plaques are not invariant. In these sciousness experienced by patients with these diseases.
disorders, neocortical deficits in ACh-mediated neuro- In AD, explicit memory is affected more than implicit transmission are generally greater than in AD and yet memory: the latter involves learning of which the cognitive impairments not as severe.
patient is unaware. Loss of explicit memory occurs due Patients with DLB and, to a lesser extent, those to early degeneration in the medial-temporal-lobe mem- with PD experience hallucinations that are primarily ory systems, which are primarily, if not exclusively, in- visual and frequently persistent2. In DLB, neocortical volved in declarative memory. Implicit memory for ACh-related activities (especially in the temporal and TINS Vol. 22, No. 6, 1999

E. Perry et al. – Cholinergic components of consciousness
Box 1.The discovery of ACh
Rapid eye movement (REM) sleep or dreaming sleep istriggered by firing and release of ACh from pedunculo-pontine cholinergic neurones. In 1921, Otto Loewi,the German-born pharmacologist and physician, dis-covered ACh. The method he used to do this wasinspired by a dreama.
I awoke, turned on the light, and jotted down a few notes on a tiny slip of thin paper. Then I fell asleepagain. It occurred to me at six o'clock in the morningthat during the night I had written down somethingmost important, but I was unable to decipher the scrawl…The next night, at three o'clock, the idea returned. It wasthe design of an experiment to determine whether or notthe hypothesis of chemical transmission that I had utteredseventeen years ago was correct. I got up immediately,went to the laboratory, and performed a single experi-ment on a frog heart according to the nocturnal design.
Fig. 2. Belladonna (Atropa belladonna or deadly nightshade). This
plant, together with other closely related species such as henbane,
Loewi transferred Ringer solution from a frog heart, mandrake and datura, which contain atropine and scopolamine, has which was stimulated by the vagus nerve, to another been used for centuries to induce hallucinations. Atropa mandragora isolated frog heart in which contractions then became (mandrake) was also used by the ancient Romans to abolish pain and slowed, as if its vagus had been stimulated. These results induce sleep during surgical procedures. proved that the vagal nerve does not influence theheart directly, but via the release of a specific chemi-cals. Loewi shared the Nobel prize for physiology and parietal cortex) are lower in patients with halluci- medicine in 1936 for this discovery, together with Dale nations42. While this identifies a cortical ‘cholinergic cor- and Dudley, who were able to identify the reactive relate' of hallucinogenesis in DLB, abnormalities in REM substance in the fluid as ACh. sleep (below) could also implicate the pathology ofpedunculopontine neurones, with intrusions of REM a Byron, J. et al. (1993) Psychic Experiences of the Famous,
into the waking state. The integrated visual images of people or animals that are encountered by hallucinat-ing patients are similar to those experienced followingingestion of muscarinic-receptor antagonists, such as Pathophysiology of REM-sleep patterns
scopolamine or atropine, in ritualistic, recreational ormedical situations (Fig. 2; see Ref. 3 for a review). In Resemblances between wakefulness and paradoxical, both PD with dementia and AD, hallucinations are at- REM-sleep physiology are likely to provide important tenuated by the cholinesterase inhibitors physostig- clues as to the neurobiological mechanisms of con- mine, tacrine or metrifonate43–45. In PD, antimuscarinic sciousness1,9. One of the most important neurophysio- agents affect cognitive shifting (which is assessed by logical events that triggers REM sleep or dreaming is the card-sorting tests) and memory performance. As in AD, firing of cholinergic neurones in the pedunculopon- memory impairment in PD is apparent in tasks that tine nuclei. It is a striking coincidence that the neuro- test explicit-memory function as opposed to implicit- transmitter that activates dreaming mechanisms was memory function46. Both non-demented and demented originally discovered as the result of a dream (Box 1).
PD patients perform normally in automatic or implicit The hypothesis that REM or dreaming involves the mental tasks (for example, word or picture-fragment cholinergic system was first formulated by Jouvet and identification). In PD, a loss of pedunculopontine neur- Hernandez-Péon (see Ref. 48 for a review) over ten years ones (around 50%) occurs41, which could be responsible before a cholinergic hypothesis was applied to the cog- for sleep abnormalities (see below).
nitive impairment that occurs in dementia. Distinct Another feature of DLB that is distinct from AD is the patterns of firing in mesopontine cholinergic neur- prominent fluctuation in symptoms, which include the ones precede and coincide with REM-sleep onset. A level of consciousness with episodes of reduced aware- REM-sleep induction zone in the dorsolateral meso- ness of surroundings2. Patients, while not unconscious pontine tegmentum receives 40% of its input from or asleep, can, for seconds, minutes or hours, cease to these neurones. Carbachol- or glutamate-elicited exci- respond to external stimuli. These absence episodes are tation of brainstem pedunculopontine tegmental chol- not epileptic in origin nor are they obviously the result inergic cells induces REM sleep, and AChE inhibitors, of cardiovascular deficiencies. Reductions in the number such as tacrine, decrease REM-sleep latency and increase of cholinergic projections to the thalamic reticular nu- REM-sleep duration. Neurotoxic lesions of this region cleus have been identified in DLB that do not occur to (produced using kainic acid) result in reductions in the same extent as in AD or PD (Ref. 47), despite the loss the duration of REM sleep in the cat that parallel the of pedunculopontine neurones in the latter. Whether severity of cholinergic- but not noradrenergic-neurone dysfunction of the cholinergic system accounts for these loss. Muscarinic-receptor antagonists, such as atropine changes in the level of conscious awareness remains to or the blockade of the vesicular ACh transporter by be established, although a recently identified mecha- vesamicol49 also decrease REM-sleep activity by increas- nism of anaesthesia that involves the cholinergic system ing its latency or decreasing its density and duration, supports this concept (see below).
TINS Vol. 22, No. 6, 1999 E. Perry et al. – Cholinergic components of consciousness
TABLE 1. Rapid eye movement (REM) sleep abnormalities in degenerative diseases of the human brain
Potential ACh-related correlates
Parkinson's disease Reduced REM latency and duration REM abnormalities could depend on relative pathology of PPN and NbM neurones, or pathology of noradrenergic and serotonergic neurones that inhibit PPN cholinergic neurones REM behaviour disorder, in some Loss of PPN cholinergic neurones that control instances preceding movement muscle atonia (descending projections to spinal cord) and REM generation or abnormalities in afferent projections (for example, GABAergic) to these neurones Dementia with Lewy bodies REM behaviour disorder Similar to Parkinson's disease Alzheimer's disease Decreased REM duration and Basal forebrain, as opposed to PPN, ACh-related density, increased REM latency neuropathology might be primarily implicated REM behaviour disorder Abbreviations: NbM, nucleus basalis of Meynert; PPN, pedunculopontine.
In degenerative brain diseases, loss of neurones or is interesting in view of the importance of dopamine- histological features, such as neurofibrillary tangles or sensitive GABAergic pathways, which project from the Lewy bodies in brainstem serotonergic, noradrenergic output nuclei of the basal ganglia, in controlling ped- or cholinergic neurones, could interfere with REM sleep unculopontine cholinergic neurones64. A syndrome and in turn contribute to disturbances in consciousness, that was recently identified clinically as ‘RBD Dementia' such as hallucinations or delusions. In AD, reductions is thought to represent a form of DLB where patients in REM (Table 1) have been hypothesized to lead to or have greater attention–concentration and perceptual- contribute to progressive loss of memory and other organization deficits than those seen in AD patients cognitive skills61. Decreased REM sleep correlates with the cognitive decline seen in AD (Ref. 58) and there Pathology of pedunculopontine cholinergic or dorso- are reductions in both REM-sleep duration and also in lateral tegmental neurones has been consistently REM-sleep density, which distinguish AD from depres- described in terms of neurone loss in PD (on average sion. There is, however, one case report of abundant 50%), although tangles or Lewy bodies are present in REM sleep in AD (Ref. 62), which highlights the need these cells in AD and DLB. Locus-coeruleus neurone to relate REM patterns to the relative involvement of loss is more common and usually extensive (up to 70%) different brainstem nuclei in individual cases. REM- in all of these disorders, whereas substantia-nigra neur- sleep-behaviour disorder (RBD), which describes the one loss is extensive in PD, moderate in DLB and rare loss of muscle atonia that can occur during REM-sleep in AD, and raphé-nucleus neurone loss occurs in PD.
associated with movement, often violent, during It will be important in the future to determine the dream mentation, has been reported to precede clini- effects of therapy using drugs that affect the choliner- cal symptoms of AD in one case63. In this case, there was gic system, for example, cholinesterase inhibitors and a loss of locus-coeruleus neurones, which inhibit ped- muscarinic- or nicotinic-receptor agonists, on REM-sleep unculopontine cholinergic neurones, in conjunction patterns in patients with these disorders. If REM-sleep with, surprisingly, elevated numbers of mesopontine abnormalities or RBD relate to cholinergic-neurone pathology, they could be attenuated by therapy.
REM-sleep disturbances have been reported more However, restoring normal sleep patterns (which has frequently in PD than in AD and include reductions in been reported to occur with tacrine66), or REM sleep REM-sleep latency and also RBD, which are relieved by might not always be beneficial. There is a report of two selegeline or L-dopa (Table 1). A striking observation AD patients who have experienced Aricept-induced was made by Schenck53 that RBD preceded clinical Parkinsonism (by over ten years on average) in 38% of Clinical responses to drug therapy
29 older male PD patients. Hallucinating PD patientsexperience significantly decreased REM-sleep duration The low frequency of identifiable synaptic-membrane (3 min versus 50 min) and percentage REM-sleep– differentiations on choline acetyltransferase immuno- total-sleep time (5% versus 20%) compared with non- stained axon terminals in the rat cortex and hippo- hallucinating PD patients52. Hypnapompic or hypna- campus68, indicates that the dominant mode of corti- gogic hallucinations (which can occur normally on cal ACh-mediated transmission might be ‘diffuse', as waking or falling asleep) are thought to consist of opposed to point synaptic. Moreover, reversal of behav-brief intrusions of REM into the waking state. Hallu- ioural deficits in basal-forebrain cholinergic-neurone- cinations in PD or DLB could, therefore, have a simi- lesioned rats, following implantation of ACh-secreting lar but, owing to brainstem pathology, extended basis.
cells, indicates that impulse-dependent regulated syn- REM-sleep-behaviour disorder has been identified in aptic release of ACh might not be necessary for func- isolated cases of DLB and incidental (or otherwise tional recovery69. These characteristics might provide asymptomatic) Lewy-body disease (Table 1), and the an explanation for functional correlates of systemically latter is associated with loss of locus-coeruleus and administered drugs that affect the cholinergic system substantia-nigra neurones. Substantia-nigra pathology in patients with AD and related disorders.
TINS Vol. 22, No. 6, 1999 E. Perry et al. – Cholinergic components of consciousness
TABLE 2. Response to drug therapy in Alzheimer's diseasea
Decreased agitation Non-dose-dependent improvement in CGIC Improvement in attentional as opposed to Decreased delusions and apathy; mnemonic function (CANTAB) reduced disinhibition No difference between drug and placebo Improvement in ADAS non-cognitive items (for example, delusions, co-operation) Improvement in 8 out of 11 ADAS cognitive items (for example, word recall, comprehension, language production, orientation) Dose-dependent improvement (CIBIC) (ADAS-COG and MMSE) Improvement (ADAS-COG and MMSE) Decreased hallucinations Reduced delusions, hallucinations and behavioural disturbances Improvement in attention but not memory aReports from 1993 onwards; for a recent review, see Ref. 81.
bAll cholinesterase inhibitors except xanomeline, a muscarinic-receptor agonist, and nicotine, a nicotinic-receptor agonist.
cPlacebo controlled.
Abbreviations: ADAS-COG, Alzheimer's Disease Assessment Scale (cognitive subscale); CANTAB, Cambridge Neuropsychological Test AutomatedBattery; CGIC, Clinician Global Impression of Change; MMSE, Mini Mental-State Exam.
Since the introduction of the cholinesterase inhib- of general (volatile) anaesthetics have been compli- itors tacrine (Cognex) and, more recently, donepezil cated by the diversity of chemical agents used. In the (Aricept) and rivastigmine (Exelon), for the treatment 1980s, the fluidizing or disordering effects of anaes- of AD, clinical outcome (Table 2) has generally been thetics on membrane lipids were the main focus of assessed in terms of the recovery of cognitive function attention and related to their ability to disrupt neur- [using, for example, the Alzheimer's Disease Assesment onal excitability. However, it soon became evident Scale (cognitive subscale) or Mini Mental-State Exam that the disordering of membrane lipids was, in effect, (MMSE)]. Cognitive functions that improve include small and did not correlate with the relative potencies word recall, word recognition, orientation, language of different anaesthetic agents. The research focus then production, comprehension, word finding and com- shifted to proteins, in particular voltage-gated ion mand following. Symptomatic improvements are gen- channels. Although Na1, K1 and Ca21 channels are all erally modest and confined to a minority of patients, affected by anaesthetics, the doses required are usually although whether such therapy provides additional supratherapeutic. More recently ligand-gated ion protection against further cognitive decline is still being channels have been intensively studied, including, evaluated. Neuropsychiatric or non-cognitive functions in particular, glutamate NMDA, GABA , glycine and have been assessed to a lesser extent but appear to be nicotinic receptors (see Ref. 82 for a review).
equally if not more amenable to therapy with cholin- Although there is still no consensus on whether all esterase inhibitors81. Physostygmine, tacrine and met- volatile anaesthetics act via a single, identical mecha- rifonate have been reported to decrease psychosis (hal- nism, nor any consensus on whether there is a specific lucinations and delusion), agitation, apathy, anxiety, receptor involved, evidence for the involvement of the disinhibition, pacing and aberrant motor behaviour, cholinergic system, particularly nicotinic receptors, is and to improve cooperation in AD (Table 2). Such evi- growing. In the Torpedo electric organ and mam- dence, that enhancing the activity of cholinergic neur- malian myotubes, nicotinic receptors have been impli- ones attenuates a broad spectrum of cognitive and non- cated as a sensitive target for many years (Table 3).
cognitive functions, is consistent with a general role Agents such as isoflurane, butanol and chloroform for ACh in selective attention, and suggests that ACh increase channel opening rate and increase rates of is involved centrally in the mechanism of conscious fast and slow desensitization at concentrations similar to those reported for anaesthetic actions on the GABA - receptor channel. More recently, it has been reported The mechanism of action of anaesthetics:
that the subtype of nicotinic receptor found in the involvement of ACh
CNS (a4b2) is more sensitive than the muscle subtype, The identity of the neurochemical systems that are values for halothane or isoflurane being involved in consciousness can be inferred from the 10–35 times higher in muscle than in the CNS, and mechanisms of action of general anaesthetics, which that the a4b2 receptor is more-sensitive to isoflurane induce loss of consciousness and awareness of sensory than the most sensitive GABA receptor or glycine stimuli. The theories behind the mechanism(s) of action receptor previously reported. The extreme sensitivity TINS Vol. 22, No. 6, 1999 E. Perry et al. – Cholinergic components of consciousness
TABLE 3. ACh-related mechanisms of anaesthesia
Effects of volatile anaesthetics
(at clinically relevant concentrations)
Muscle nicotinic receptors (Torpedo Isoflurane, butanol and chloroform increase channel opening, and both fast electric organ or mammalian myotubes) and slow desensitization CNS nicotinic-receptor subtype a4b2 Thiopental inhibits receptor-mediated current More sensitive to halothane or isoflurane than the muscle subtype of nicotinic receptor More potently inhibited by isoflurane than GABA receptors or agonists at the glycine site Muscarinic M receptor Halothane inhibits Ca21-dependent Cl2 currents in M -receptor 90 transfected oocytes High-affinity choline uptake Inhibited by halothane in cortical synaptosomes Nicotine-elicited release of dopamine Inhibited in the striatum by halothane Reduced in the cat medial pontine reticular formation by halothane Reduced in the rat cerebral cortex by isoflurane of the neuronal nicotinic receptor to such compounds of AChE, N-methyl-4-piperidyl acetate has been used suggests that its inhibition is relevant, at least in con- to demonstrate consistent reductions in the levels of junction with effects on other members of this super- AChE in AD, which are more prominent in the par- family of fast neurotransmitter-gated receptors, to the ietal and temporal cortices than in the frontal, occipi- loss of conscious awareness that they produce.
tal and sensorimotor cortices95. Iodinated quinucli- Other evidence that links ACh to anaesthesia (Table dinyl benzilate {[123I]QNB} binding, measured using 3) includes halothane inhibition of the mechanism of SPECT, is reduced in advanced but not moderate AD high-affinity choline uptake into rat cortical synapto- cases96. Using iododexetimide, a muscarinic-receptor somes; decreased nicotine-elicited release of striatal antagonist that might be more specific for the M2 neurotransmitters such as dopamine; and inhibition by receptor subtype, it has been demonstrated, in one halothane of the muscarinic-receptor induced Ca21 de- SPECT study, that a reduction in muscarinic-receptor pendent Cl2 current. Historically, muscarinic-receptor levels in the temporal and parietal cortices is apparent antagonists pre-dated inhalational agents in anaesthe- in mild probable AD (Ref. 97). Administration of the sia. Naturally occurring alkaloids, such as atropine and muscarinic-receptor antagonist, scopolamine, decreased hyoscine (scopolamine), have been used in anaes- [123I]QNB binding in controls but had the opposite effect thesia for over a century and records suggest this appli- in AD patients, which indicates a differential receptor cation could date back to early Roman times. Scopol- sensitivity in the disease98. Reductions in [11C]nicotine- amine induces ‘twilight sleep' in which the patient is binding in temporal cortex of AD patients, which is awake but unaware and subsequently amnesic for the reversed by tacrine, have been reported in PET stud- event. Although the cholinergic hypothesis of geriatric ies99. Other potential imaging markers of the choliner- memory impairment was partly created on the basis of gic system are in development and alterations in cer- results obtained from experimental models of scopol- ebral perfusion that result from treatment with drugs amine-induced memory loss, muscarinic-receptor that affect this system are also being investigated.
block induces a more-global disruption, which might be as relevant to understanding the pathophysiologyof dementia. Tacrine, the first prescription drug for AD, Although the subject of transmitters or other neural has been used since the 1960s as a ventilatory stimu- correlates of consciousness might be considered to be lant and to promote the recovery of consciousness fol- academic, in relation to major diseases of the brain, dis- lowing anaesthesia, an effect that is similar to that pro- turbance of conscious awareness is a major predictor duced by physostigmine, though of greater duration of personal and social dysfunction. It is over 30 years and with fewer side effects.
since ACh release in the cerebral cortex was originallycorrelated with consciousness and shown to increase dur- ing waking and dreaming compared with non-dream- Monitoring cholinergic-neurone activities in vivo pro- ing sleep. Since then, different cholinergic pathways in vides new opportunities for the examination of clini- the brain have been characterized, and their involve- cal correlates of pathological or drug-induced changes ment in brain diseases that affect cognition and con- in the cholinergic system, including alterations in con- sciousness have been reported. As drugs emerge for the scious awareness. Chemical markers of the cholinergic treatment of AD and ligands for imaging the cholinergic system are progressively being developed for PET or system in vivo proliferate, new opportunities arise that SPECT imaging of the human brain in vivo. The vesicu- allow the examination of the role of cholinergic systems lar ACh transporter has been monitored using iodo- in the human brain. Beyond objective measures of cog- benzovesamicol and it has been shown that reduc- nition, memory and behaviour, it will be valuable to tions in its binding capacity correlate with cognitive explore subjective experiences that involve conscious impairment in AD patients. In PD patients, this reduc- awareness, including such components as hallucino- tion of binding capacity is more pronounced in de- genesis, levels of consciousness, and REM sleep or dream- mented than non-demented subjects36. An inhibitor ing. The physiological, pharmacological and pathological TINS Vol. 22, No. 6, 1999 E. Perry et al. – Cholinergic components of consciousness
data reviewed in this article are consistent with the con- 43 Cummings, J.T. et al. (1993) Biol. Psychiatry 33, 536–541
cept that the action of ACh in the cortex and thalamus 44 Hutchinson, M. and Fazzini, E. (1996) J. Neurol. Neurosurg.
Psychiatry 61, 324–325 is essential for the maintenance of the normal experi- 45 Morris, J.C. et al. (1998) Neurology 50, 122–130
ence of conscious awareness. In the words of Alexander 46 Appolonio, I. et al. (1994) Arch. Neurol. 51, 359–367
Karczmar12, ‘no behaviour is a one-transmitter affair… 47 Perry, E.K. et al. (1998) J. Neural. Transm. 105, 915–933
yet, frequently the cholinergic system constitutes the 48 Hobson, J.A. et al. (1992) Curr. Opin. Neurobiol. 2, 759–763
49 Salin-Pascual, R.J. and Jimenez-Anguiano, A. (1995)
significant correlate'. The way in which ACh might Psychopharmacol. 12, 485–487 contribute to generating the integrated, coherent ex- 50 Kostic, V.S. et al. (1989) J. Neurol. 236, 421–423
perience of conscious awareness remains to be established.
51 Askenasy, J.J. (1993) Acta Neurol. Scand. 87, 167–170
52 Cornella, C.L. et al. (1993) Ann. Neurol. 34, 710–714
During preparation of this article, a novel hypothesis, 53 Schenck, C.H. et al. (1996) Biol. Psychiatry 40, 422–425
that ACh enhances the activity of specific circuits in- 54 Tan, A. et al. (1996) Movement Disord. 11, 214–216
volved in conscious awareness by promoting the inter- 55 Uchiyama, M. et al. (1995) Neurology 45, 709–712
action between microtubule-associated protein 2 56 Turner, R.S. et al. (1997) Neurology 49, 523–527
57 Schenck, C.H. and Mahowald, M.W.
(1997) Biol. (MAP2) and microtubules, was published on the basis Psychiatry 42, 527–528 of data from parallel-distribution studies of MAP2 and 58 Prinz, P.N. et al. (1982) Neurobiol. Aging 3, 361–370
postsynaptic muscarinic receptors in the cortex100.
59 Bliwise, D.L. et al. (1989) Biol. Psychiatry 25, 320–328
60 Bahro, M. et al. (1993) Biol. Psychiatry 34, 482–486
Interactions between ACh and other neurotransmit- 61 Christos, G.A. et al. (1993) Med. Hypotheses 41, 435–439
ters, in particular glutamate and GABA, which control 62 Bliwise, D.L. et al. (1990) Neurology 40, 1281–1284
basal-forebrain and pedunculopontine cholinergic 63 Schenck, C.H. et al. (1996) Neurology 46, 388–393
neurones, are likely to provide further insights into 64 Rye, D.B. et al. (1997) Sleep 20, 757–785
65 Boeve, B.F. et al. (1998) Neurology 51, 363–370
cholinergic correlates of consciousness. Acetylcholine 66 Gillman, P.K. et al. (1997) J. Am. Geriatr. Soc. 45, 1286
will no doubt need to compete with other candidate 67 Ross, J.S. and Shua-Haim, J.R. (1998) J. Am. Geriatr. Soc. 46,
neurochemical correlates of consciousness, as it does 68 Descarries, L., Gisiger, V. and Steriade, M. (1997) Prog.
with other neurotransmitters in determining the Neurobiol. 53, 603–635 physiological response of receptive neurones.
69 Winkler, Y. et al. (1995) Nature 375, 484–487
70 Gorman, D.G. et al. (1993) Neuropsychiatry Neuropsychol.
Behav. Neurol. 6, 229–239 1 Delacour, J. et al. (1995) Neuropsychologia 33, 1061–1074
71 Thal, L.J. et al. (1996) Neurology 47, 1389–1395
2 McKeith, I.G. et al. (1996) Neurology 47, 1113–1124
72 Raskind, M.A. et al. (1997) Arch. Neurol. 54, 836–840
3 Perry, E.K. and Perry, R.H. (1995) Brain Cogn. 28, 240–258
73 Sahakian, B.J. et al. (1993) Psychopharmacology 110, 395–401
4 Sommerhoff, G. and MacDorman, K. (1994) Integ. Physiol.
74 Knopman, D. et al. (1996) Neurology 47, 166–177
Behav. Sci. 29, 177–189 75 Kaufer, D.L. et al. (1995) J. Geriatr. Psychiatr. 9, 1–6
5 Block, N. et al. (1996) Trends Neurosci. 19, 456–459
76 Rogers, S.L. and Friedhoff, L.T. (1996) Dementia 7, 293–303
6 Henke, K. et al. (1994) Int. J. Neurosci. 75, 181–187
77 Rogers, S.L. et al. (1998) Neurology 50, 136–145
7 Markowitsch, H.L. et al. (1995) Neuropsychologia 33, 1181–1192
78 Becker, R.E. et al. (1996) Alzheimer Dis. Assoc. Disord. 10,
8 Moruzzi, G. and Magoun, H.W. (1969) Electroencephalog. Clin.
Neurophysiol. 1, 455–473 79 Bodick, N.C. et al. (1997) Arch. Neurol. 54, 465–473
9 Paré, D. and Llinás, R. (1995) Neuropsychologia 33, 1155–1168
80 Snaedal, J. et al. (1996) Dementia 7, 47–52
10 Bogen, J.E. et al. (1995) Conscious. Cogn. 4, 52–62
81 Cummings, J.T. et al. (1997) Alzheimer Dis. Assoc. Disord. 11,
11 Jones, E.G. et al. (1998) Neuroscience 85, 331–345
12 Karczmar, A.G. (1993) Neuropsychopharmacol. 5, 181–199
82 Franks, N.P. and Lieb, W.R. (1997) Nature 389, 334–335
13 Steriade, M. et al. (1996) J. Neurosci. 16, 392–417
83 Liu, Y. et al. (1994) Mol. Pharmacol. 45, 1235–1241
14 Fisahn, A. et al. (1998) Nature 394, 186–189
84 Lin, L. et al. (1995) Biochem. Pharmacol. 49, 1085–1089
15 Brizer, D.A. and Manning, D.W. (1982) Am. J. Psychiatry 139,
85 Raines, D.E. et al. (1995) Anaesthesiology 82, 276–287
86 Scheller, M. et al. (1997) Anaesthesiology 86, 118–127
16 Tune, L.E. et al. (1981) Lancet 2, 651–653
87 Andoh, T. et al. (1997) Anaesthesiology 87, 1199–1209
17 Petit, D. et al. (1993) Neurobiol. Aging 14, 141–145
88 Violet, J.M. et al. (1997) Anaesthesiology 86, 866–874
18 Montplaisis, J. et al. (1996) Eur. Neurol. 36, 197–200
89 Flood, P. et al. (1997) Anaesthesiology 86, 859–865
19 Dierks, T. et al. (1994) Pharmacopsychiatry 27, 72–74
90 Minami, K. et al. (1998) Eur. J. Pharmacol. 339, 237–244
20 Ikeda, K. et al. (1995) Brain Res. 688, 171–183
91 Griffiths, R. et al. (1994) Anaesthesiology 81, 953–958
21 Bentivoglio, M. and Steriade, M. (1990) in The Diencephalon and
92 Keifer, J.C. et al. (1996) Anaesthesiology 84, 945–954
Sleep (Mancia, M. and Marini, M., eds), pp. 7–29, Raven Press 93 Salord, F. et al. (1997) Anaesthesiology 86, 632–641
22 Mesulam, M.M. et al. (1995) The Neurosciences 7, 297–307
94 Shichino, T. et al. (1998) Br. J. Anaesth. 80, 365–370
23 Buzsaki, G. et al. (1988) J. Neurosci. 26, 735–744
95 Iyo, M. et al. (1997) Lancet 349, 1805–1809
24 Rasmusson, D.D. et al. (1994) Neuroscience 60, 665–677
96 Wyper, D.J. et al. (1993) Eur. J. Nuclear Med. 20, 379–386
25 Baars, B.J. et al. (1998) Trends Neurosci. 21, 58–62
97 Claus, J.J. et al. (1997) Eur. J. Nuclear Med. 24, 602–608
26 Sahakian, B. et al. (1988) Handbook Psychopharmacol. 20, 393–424
98 Sunderland, T. et al. (1995) Psychopharmacology 121, 231–241
27 Warburton, D.M. and Rusted, J.M. (1993) Neuropsychobiology
99 Nordberg, A. et al. (1997) Dementia 8, 78–84
100 Woolf, N.J. et al. (1997) Conscious. Cogn. 6, 574–596
28 Bloklund, A. et al. (1995) Brain Res. Rev. 21, 285–300
101 Rico, B. and Cavada, C. (1998) Eur. J. Neurosci. 10, 2346–2352
The authors thank 29 Voytko, M.L. et al. (1996) Behav. Brain Res. 75, 13–25
30 Everitt, B.J. and Robbins, T.W. (1997) Annu. Rev. Psychol. 48,
Clive Ballard, Ian McKeith, Andrew 31 Wenk, G.L. et al. (1997) Neurobiol. Learn. Mem. 67, 85–95
Fairbairn, Daniel 32 Sarter, M. and Bruno, J.P. (1997) Brain Res. Rev. 23, 28–46
Collerton and 33 Heckers, S. et al. (1992) J. Comp. Neurol. 325, 68–82
34 Mitrofanis, J. and Guillery, R.W. (1993) Trends Neurosci. 13,
Heather Ashton for Homeostatic plasticity in neuronal networks: the more 35 McCormick, D.A. et al. (1990) in Brain Cholinergic Systems
(Steriade, M. and Biesold, D., eds), pp. 236–264, Oxford
things change, the more they stay the same, by Gina G.
Lloyd for providing Turrigiano, Vol. 21, pp. 221–227.
36 Kuhl, D.E. et al. (1998) Ann. Neurol. 40, 399–410
a constant stream of In Fig. 3B, some mathematical symbols were omitted. It 37 Hirono, N. et al. (1997) Dementia 8, 210–218
relevant papers, and 38 Wagner, M.J. et al. (1997) Alzheimer Dis. Assoc. Disord. 11, 125–131
should show negative values on the x-axis, and read Dawn Houghton 39 Lopez, O.L. et al. (1994) Eur. Neurol. 34, 277–282
‘Reduced activity 4 2.68' and ‘Enhanced activity 3 1.58'.
40 Woolf, N.J. et al. (1989) Neurosci. Lett. 96, 277–282
and Lorraine Hood We apologize to the author and readers for this error.
41 Jellinger, K. et al. (1988) J. Neurol. Neurosurg. Psychiatry 51,
for manuscript 42 Perry, E.K. et al. (1993) J. Neural. Transm. 6, 167–177
TINS Vol. 22, No. 6, 1999

Source: http://walkerlab.berkeley.edu/reprints/PerryWalkerGrace_TiNS_1999.pdf