Directory UMM :Data Elmu:jurnal:B:Biological Psichatry:Vol49.Issue3.2001:

Nicotine and Its Interaction with b-Amyloid Protein:
A Short Review
M. Reza Zamani and Yvonne S. Allen
Two features of Alzheimer’s disease (AD) are b-amyloid
protein (bAP) deposition and a severe cholinergic deficit.
b-Amyloid protein is a 39 – to 43–amino acid transmembrane fragment of a larger precursor molecule, amyloid
precursor protein. It is a major constituent of senile
plaque, a neuropathologic hallmark of AD, and has been
shown to be neurotoxic in vivo and in vitro. The cholinergic neurotransmission system is seen as the primary
target of AD. However, other systems are also found to
show functional deficit. An association between cholinergic deficit and bAP is suggested by a negative correlation between cigarette smoking and AD. Evidence hitherto
suggests that bAP causes neuronal death possibly via
apoptosis by disrupting calcium homeostasis, which may
involve direct activation or enhancement of ligand-gated
or voltage-dependent calcium channels. Selective second
messengers such as protein kinases are triggered that
signal neuronal death. Nicotine or acetylcholinesterase
inhibitors can partially prevent the neurotoxicity of bAP
in vivo and in vitro. However, the exact mechanism by
which nicotine provides its protective effects is not fully
understood, but clearly there are protective roles for

nicotine. Here, some aspects of bAP neurotoxicity and
nicotinic intervention as a protective agent are discussed.
Biol Psychiatry 2001;49:221–232 © 2001 Society of Biological Psychiatry
Key Words: b-Amyloid protein, neurotoxicity, neuroprotection, nicotine, acetylcholine receptor

Introduction

E

arlier findings that cigarette smoking is negatively
correlated with Parkinson’s disease (Bharucha et al
1986) and positively correlated with the delayed onset of
Alzheimer’s disease (AD) (van Duijn and Hofman 1991)
have encouraged studies investigating the neuroprotective
role of nicotine (e.g., Knott et al 1999; Maggio et al 1998;
Perry et al 1996; Ulrich et al 1997). So far, in vivo studies
From School of Biosciences, Cardiff University, Cardiff (MRZ) and the Department of Human Anatomy and Cell Biology, The University of Liverpool,
Liverpool (YSA), United Kingdom.
Address reprint requests to M. Reza Zamani, Cardiff University, School of
Biosciences, PO Box 911, Cardiff CF10 3US, Wales, United Kingdom.

Received June 1, 2000; revised November 14, 2000; accepted November 28, 2000.

© 2001 Society of Biological Psychiatry

have been inconclusive (Behmand and Harik 1992; Janson
and Møller 1993), but in vitro evidence has demonstrated
more convincingly that nicotine inhibits cell loss induced
by excitotoxic agents (Akaike et al 1994; Marin et al
1994).
Nicotinic receptor binding and choline acetyltransferase
are positively correlated with smoking (Parks et al 1996).
In this respect, it is pertinent that AD has long been
associated with a cholinergic deficit, characterized by
marked degeneration of cholinergic innervation from the
nucleus basalis of Meynert to the cerebral cortex (Candy et
al 1983), reduced levels of choline acetyltransferase and
acetyl cholinesterase, and a much decreased density of
nicotinic and M2 muscarinic receptors in the cerebral
cortex and hippocampus (Davies and Maloney 1976;
Flynn and Mash 1986; London et al 1989; for a review, see

Roßner et al 1998). In this review aspects of the nicotinic
acetylcholine system, b-amyloid protein (bAP) neurotoxicity, and the protective role of nicotine against bAP
toxicity are presented.

Nicotinic Acetylcholine Receptors
Nicotinic acetylcholine receptors (nAChRs) are ligandgated calcium-permeable receptors consisting of at least
eight a-like subunit isoforms (a2–a9) and three b-like
isoforms (b2–b4) (Kim et al 1997). These subunits, either
as homomeric (a7–a9) or heteromeric (requiring b subunits), form receptors with distinct spatial and tissuespecific distributions (McGehee and Role 1995; Zarei et al
1999). In the neocortex the predominant subtype is a4b2,
whereas others contain a4 and a3. The hippocampus, on
the other hand, is rich in a7 subunits that could form the
homomeric receptor (Wonnacott 1997). g-Aminobutyric
acid (GABA)– ergic interneurons in the hippocampus
possess nAChRs the activation of which leads to an
increase in GABA release (Alkondon et al 1997; Lena and
Changeux 1997; McMahon et al 1994; Radcliffe and Dani
1998). Enhanced GABA release by nicotine has been
extensively studied at various drug concentrations (low
mmol/L range equivalent to smokers’ intake and higher

mmol/L range similar to synaptic cleft concentrations), and
in all cases nicotine increases the release of the inhibitory
0006-3223/01/$20.00
PII S0006-3223(00)01108-2

222

M.R. Zamani and Y.S. Allen

BIOL PSYCHIATRY
2001;49:221–232

neurotransmitter (Radcliffe et al 1999). Recently it has
been shown that these cholinergic receptors desensitize in
response to nicotine when exposed with a time course and
concentration equivalent to smokers’ intake, thereby reducing GABA release (Fisher et al 1998).
Relative to nonneuronal tissues, neuronal AChRs and
the a7 receptor have higher Ca21 permeability to the
extent that calcium influx through these channels at
presynaptic terminals may cause transmitter release (Castro and Albuquerque 1995; Decker and Dani 1990; Pugh

and Berg 1994; Radcliffe et al 1999; Seguela et al 1993;
Vernino et al 1992, 1994).
Physiologically, nAChRs are distributed at presynaptic
(terminal), postsynaptic (dendrites), and somatic sites
(Wonnacott 1997). In the hippocampus, presynaptic activation of nAChRs is thought to dominate over the postsynaptic activation, which enhances neurotransmitter release
at excitatory, inhibitory, and modulatory synapses (Fu et al
1999; Radcliffe et al 1999; Vidal 1996; Wonnacott 1997).
For example, glutamatergic synapses contain a7 receptors,
which respond to nicotine resulting in an increase in the
frequency of miniature excitatory postsynaptic currents in
a tetrodotoxin (TTX)-independent manner suggesting a
presynaptic location (Araujo et al 1988; Gray et al 1996;
McGehee and Role 1995; Sacaan et al 1996). In studying
the nAChR system in both the cortex and hippocampus,
nicotine is seen to regulate its own release, implying the
existence of self-reinforcing autoreceptors (Araujo et al
1988; Rowell and Winkler 1984; Wilkie et al 1996; for a
review, see Wonnacott 1997). Furthermore, the rate of
expression and distribution of nAChR subunits changes
during development (Broide et al 1995; Court et al 1997;

Zarei et al 1999; Zhang et al 1998) and in humans during
aging (Marutle et al 1998).
The complexity of the nicotinic acetylcholine system
underlines the importance of therapeutic strategies targeting specific brain regions and/or receptor subtypes. Below,
how the neuronal cholinergic system is affected in AD is
considered briefly.

Cholinergic System and Alzheimer’s Disease
Deficit or mutation in the cholinergic neurotransmission
system is seen in a number of neurologic disorders, but in
AD, the cholinergic system is believed to be the primary
target (Bowen et al 1976; Davies and Maloney 1976; Perry
et al 1977). This does not preclude the involvement of
other neurotransmission systems, and in fact, since the
report by Davies and Maloney (1976), other neurotransmitter systems have shown either functional deficit in
different brain regions or differential downregulation of
their receptor subtypes (for a review, see Kanfer et al
1999). Chronologically, cholinergic deficit in AD pre-

cedes the deficit in other neurotransmitter systems. This

has led to the conclusion that the deficit in other neurotransmitter systems may be the consequence of cholinergic
denervation reflecting transynaptic function of this system
(Kanfer et al 1999).
The role of nAChRs in cognitive function and development is well documented (e.g., Granon et al 1995; Levin
1992; Meyer et al 1998). The loss of cognitive performance in AD and Parkinson’s disease patients is positively
correlated with the degree of central cholinergic deafferentation (Flynn and Mash 1986; Perry et al 1987a, 1987b;
Rinne et al 1991; Sugaya et al 1990; for a review, see
Roßner et al 1998). Furthermore, nicotine “replacement”
reduces cognitive deficit and improves short-term, visual,
and semantic memory in these patients and in animal
models of AD (Hodges et al 1991a, 1991b; Jones et al
1992; McGurk et al 1991; Newhouse et al 1990; Perry et
al 1996; Voytko et al 1994; for a review, see Vidal 1996).
To identify changes in nAChR subtypes, Martin-Ruiz et
al (1999) studied AD brain tissues after autopsy and found
a significant reduction in a4 receptor subunits but not a7
or a3 ones as compared with age-matched samples. The
same subtype (a4) showed a significant binding elevation
in a subgroup of control individuals who were habitual
smokers. Similar findings are reported by Sugaya et al

(1990) (see also Schroder and Wevers 1998). Although
these results agree with others (Perry et al 1995) showing
a significant drop in a4 –b2 subunits, a greater loss of a7
subunits is seen in the midtemporal gyrus (but not in the
frontal cortex or hippocampus) of the AD brain (Davies
and Feisullin 1981; Sugaya et al 1990; Warpman and
Nordberg 1995). Also, significant reduction in a3 and a4
subunit binding from AD brains with Swedish mutation is
seen (Marutle et al 1999). Moreover, immunocytochemical studies on postmortem AD brains have revealed
significant reduction in a3 and a4 subunits in the temporal
cortex and, in addition to these subunits, a significant
reduction in a7 subunit in the hippocampus (Guan et al
2000). It should be noted that, in contrast with the above
reports, no changes in message expression for a3 and a4
are seen (Hellstrom-Lindahl et al 1999; Terzano et al
1998). In fact, an increase in the level of a7 messenger
RNA (mRNA) is reported by Hellstrom-Lindahl et al
(1999). Although variation in techniques and subject
selection might be a factor in the apparent lack of
concordance in these reports, they clearly signify changes

in posttranscriptional stages of the nicotinic system in AD.

bAP Is Neurotoxic
Abnormal production of the amyloidogenic protein (39 to
43 amino acid residues), which is cleaved from the
amyloid precursor protein (APP) at b- and g-secretase

Nicotine and bAP Toxicity

sites of action, is closely linked to AD (for details, see
Roßner et al 1998). b-Amyloid protein is a major constituent of senile plaque, a neuropathologic hallmark of AD,
and has been shown to be neurotoxic both in vivo and in
vitro. A shorter, hydrophobic fragment of the protein, bAP
(25–35), though not present in biological systems, is
widely used together with, or instead of, the endogenous
fragment (bAP (1– 42/43)) by a number of investigators
and is found to be at least as toxic as the full-length
fragment (Yankner et al 1989).
In a series of experiments bAP (25–35) (0.01 to 100
mmol/L, 5 days), given to 1-week-old hippocampal neuronal cultures, caused toxicity in a dose-dependent manner

(Zamani et al 1997). The percentage of cells surviving was
reduced from 85 6 7% to 29 6 4% at 0.1 mmol/L to 100
mmol/L bAP concentrations, respectively. The scramble
fragment of the peptide did not induce neuronal loss.
Kihara et al (1997) also saw some 60% cell loss with 20
mmol/L bAP (25–35) after a shorter, 48-hour incubation
period. However, since their results show a linear timedependent toxicity effect, more cell loss would be expected after 48 hours of incubation.
Both bAP (1– 42) and (25–35) peptides aggregate in
solution at different rates, with the former requiring some
24 hours for aggregation and the latter a few hours (Pike
et al 1995). Both fragments, whether fresh (diffusible,
nonfibrilar form) or “aged,” however, show neurotoxic
effects (Pike et al 1993; Terzi et al 1994, 1997).

bAP Modification of Ion Channels
To study the mechanism of bAP neurotoxicity, several
laboratories have looked at immediate to late physiologic
events following exposure of neurons or neuronlike cells
to various forms of bAP fragments. One early physiologic
action of bAP is an increase in intracellular calcium

concentration ([Ca21]). In nerve growth factor–treated
PC12 cells, the increase in [Ca21] is immediate and dose
dependent, which suggests a membrane effect (Joseph and
Han 1992). Similarly, some doubling of [Ca21] is reported
in hippocampal primary cultures 24 hours after exposure
to aged bAP (25–35) (Mattson et al 1993b; Mogensen et
al 1998).
How bAP fragments induce increased [Ca21] has been
the subject of several studies. The early events seem to
involve integration of these peptides with the membrane.
In fact, using artificial membranes Terzi et al (1994) and
Terzi et al (1997) found that both short and long fragments—bAP (25–35) and bAP (1-42), respectively—
have a tendency to insert themselves into liposomes.
Moreover, using the whole-cell patch clamp technique,
Sanderson et al (1997) found that aggregated bAP (25–35)
exerts its disruption of calcium homeostasis by forming

BIOL PSYCHIATRY
2001;49:221–232

223

membrane ionophores permeable to calcium. In support of
channel formation, Arispe et al (1994a) suggest that the
conductance is large enough to disrupt calcium homeostasis and possibly cause cell death in vivo.
Reports from other laboratories suggest an additional
role for the bAP mechanism. Thus, bAP is found to
enhance the activity of ligand-gated calcium channels or
voltage-dependent calcium channels (VDCCs), with little
or no effects on calcium homeostasis by itself. In the
NIE-115 cell line, nimodipine (L-type channel selective)
blocked the calcium current due to bAP (Davidson et al
1994). However, at the 100 mmol/L concentration used in
their study, nimodipine could also have blocked N-type
channels. Further support for the involvement of L-type
VDCCs comes from Ueda et al (1997), who succeeded in
preventing bAP (25–35) effect with L-type blockers, but
not with blockers of N and P/Q channel. The protective
effect of L-type channel blockers was not due to a
reduction in free radicals, but to their ability to prevent an
increase in [Ca21]. However, the antioxidant vitamin E
prevented the production of free radicals and rise in
[Ca21] by bAP. It is proposed, therefore, that chronic
treatment with bAP may form free radicals, which then
increase [Ca21] via activation of VDCCs (Ueda et al
1997). This finding was supported in a study where the
process of bAP (25–35) or bAP (1– 40) toxicity consisted
of an early, calcium-insensitive redox phase (0 –12 hours
posttreatment) and a late, calcium-sensitive cell death
phase (3 days onwards) (Abe and Kimura 1996). This was
contrary to glutamate-mediated neuronal loss, where the
“early phase” redox period and “late phase” calciumsensitive lactate dehydrogenase (LDH) production period
were almost contemporaneous (Abe and Kimura 1996).
In rat membrane preparation, N-methyl-D-aspartate receptor (NMDAR) involvement is not clear, with some
reporting an enhancement and others suggesting a reduction of binding of MK-801, glutamate, or glycine by bAP
(Calligaro et al 1993; Cowburn et al 1994, 1997). In
hippocampal neuronal cultures, however, Brorson et al
(1995) found bAP (25–35)– or (1– 40)–mediated increase
in [Ca21] leading to an increase in spontaneous bursts that
were inhibited by TTX, AMPA receptor, NMDAR, or
L-type VDCC blockers. But, no direct enhancement of
NMDAR- or kainate receptor-mediated calcium influx
was observed.
Despite a strong support from a number of groups,
the role of VDCCs and glutamatergic receptors in bAP
toxicity has not been resolved completely. Whitson and
Appel (1995), for example, found no VDCC or
NMDAR involvement in their hippocampal neurons
exposed to bAP (1– 40). Furthermore, in PC12 cells,
bAP (25–35) neurotoxicity and calcium homeostasis
were insensitive to calcium channel blockers but sensi-

224

BIOL PSYCHIATRY
2001;49:221–232

tive to antioxidants such as vitamin E. This is attributed
to the ability of bAP (25–35) to produce free radicals
shortly after its preparation in solution (Hensley et al
1994). It is noteworthy that the elevated free radical
formation has been reported in postmortem tissues from
AD patients (Zhou et al 1995).
An intracellular signal for the release of neurotransmitters
is indicated by an increase in presynaptic Ca21. It would
therefore be expected that bAP should enhance neurotransmitter release. This was the case in both young and aged rats,
for excitatory but not inhibitory neurotransmitters under
K1-induced depolarization in vitro (Arias et al 1995). However, if this is followed by depolarization mediated by high
K1 concentration, then the neurotoxicity is prevented (Pike et
al 1996). This phenomenon is attributed to the depolarization-mediated activation of VDCCs, since pretreatment with
a VDCC activator, bay-K, under a “mild” depolarization (10
mmol/L K1), was also found to be protective against bAP
toxicity (Pike et al 1996).
In addition to calcium channel activity, bAP seems to
require active K1 channels. For instance, tetraethylammonium-sensitive delayed-rectifier K1 channels were involved in
the bAP (1– 40)-mediated neurotoxicity in the septal/neuroblastoma hybrid cell line SN56 (Colom et al 1998). The sister
hybrid cell line, SN48, devoid of tetraethylammonium-sensitive K1 channels, resisted bAP (1– 40) neurotoxicity. Noticeably, SN56 cells do not seem to express high threshold
calcium channels and possess low threshold Ni21-sensitive
calcium channels instead that are not involved in bAP (1– 40)
neurotoxicity (Colom et al 1998). The mechanism for this
increased K1 current by bAP and ensuing neurotoxicity is
unknown, but oxidative stress may be a factor (Colom et al
1998). The protective effect of tetraethylammonium is seen
as its ability to raise [Ca21], which is mimicked by increased
depolarization induced by high extracellular K1 concentration ]K1[. And as seen above, high ]K1[ significantly
protects against bAP assault in hippocampal neurons (Pike et
al 1996).
The above reports suggest that bAP exerts its neurotoxic effect via different routes, which are probably tissue
or cell type specific and occur at various developmental
stages. In support of this notion, Scorziello et al (1996)
show that bAP (25–35) causes significantly more neural
toxicity in immature cerebellar granule cell cultures but
only slightly enhanced glutamate-mediated cell loss in
mature (differentiated) cultures. bAP (25–35) alone does
not affect the basal levels of [Ca21] but enhances glutamate-mediated neurotoxicity. This b-amyloid-enhanced
neurotoxicity is not mediated by NMDAR activity, suggesting a different route for Ca21 influx (Scorziello et al
1996). Furthermore, due to its high inhibitory physiology,
cerebellar tissue shows a distinctly different response to
bAP neurotoxicity. According to Price et al (1998),

M.R. Zamani and Y.S. Allen

Figure 1. Differential toxic effects of b-amyloid protein (bAP)
(25–35) in organotypic hippocampal slices maintained for 6
months in vitro. (Left) A photomicrograph of a slice showing the
granule cell layer exposed to 100 mmol/L bAP and stained with
propidium iodide. No cell death was observed. (Right) In the
same slice bAP-induced cell death in the pyramidal cell layer as
indicated by brightly stained nuclei.

cerebellar granule neurons show an increased bAP-mediated [Ca21] rise through mainly N-type and not L-type
calcium channels. No nonspecific ionophore formation,
due to bAP integration with the membrane, was observed
in their study. Recently, we also found some cell type
specificity for bAP (25–35) toxicity. In young organotypic
hippocampal slices, bAP was found to be toxic both to
dentate gyrus granule cells and CA1/CA3 pyramidal cells.
However, in 6-month-old slices neuronal loss was mainly
confined to the pyramidal cell layer (Y.S. Allen and M.R.
Zamani, unpublished data) (Figure 1). These observations
point to possible developmentally regulated mechanisms
that render pyramidal neurons more vulnerable to bAP
toxicity than granule cells.
In summary, it is evident that bAP is neurotoxic, but
its mode of action may vary depending on several
factors. These may be summarized as 1) neuronal cell
types (e.g., hippocampal vs. cerebellar), 2) uniformity
of culture (whether or not cultures contain glial “scaffolding”), 3) type of culture (acute brain slice, dissociated culture), 4) developmental stage of the culture
(mature vs. immature, differentiated vs. undifferentiated), 5) age of the animal from which the cultures or
brain slices are obtained (embryonic, postnatal, adult),
6) density of cultured cells (Mattson et al 1993a), 7) the
b-amyloid fragment type (25–35 vs. 1– 40) (Kihara et al
1997, 1998), 8) “age” and concentration of the fragment
(Mattson et al 1993a), and 9) duration of exposure of
cells to the bAP fragments.

Cellular Events following Ca21 Mobilization
Since the accumulation of bAP in the brain is gradual and
progressive, the induction of neuronal death should take a

Nicotine and bAP Toxicity

more “physiologic” course rather than the large and
sudden rises in [Ca21] seen in most in vitro preparations.
Specific signal transduction cascades may become involved during bAP accumulation in the affected brain,
which compromise cell membrane integrity by reconfiguring membrane-associated proteins, modulate the activity
of cation channels, cause conformational changes in proteins, induce protein turnover, and eventually cause cell
death.
Support for the above scenarios is accumulating. For
example, long-term modification of VDCCs by bAP may
be sensitive to functional metabotropic glutamate receptors (mGluRs). According to Copani et al (1995), mGluR
agonists prevented bAP (2–35)-mediated cell loss in both
cerebellar granule cells and mixed cortical cultures. Their
results were consistent with the use of agonists for specific
mGluR subtypes for different cultures, since cerebellar
granule cells express mGluR subtypes in different proportions than cortical cultures (Tanabe et al 1992). The data
are interpreted in terms of VDCC inhibition mediated by
mGluR activation. Consistent with this interpretation,
Copani et al (1995) found that direct inhibition of VDCCs
was as “neuroprotective” against bAP toxicity. No Gprotein analysis was done in this work, but Sorrentino et al
(1996) and Singh et al (1997) support the notion of a
pertussis toxin-sensitive bAP-mediated signal transduction pathway.
Ekinci et al (1999), on the other hand, support the
notion of protein modification by bAP. They found that
bAP imposes neurotoxicity by increasing tau phosphorylation. Tau is the major component of neurofibrillary
tangles in the brains of AD patients. Abnormal phosphorylation of tau can lead to the formation of paired helical
filaments of the neurofibrillary tangles. It has therefore
been suggested that bAP-mediated tau phosphorylation is
achieved through activation of the mitogen-activated protein signal transduction pathway leading to phosphorylation of L-type calcium channels with a consequence of
increased VDCC activity and [Ca21] elevation. This
triggers calcium-dependent kinases such as protein kinase
C and calcium-calmodulin-dependent kinase that phosphorylate tau (Ekinci et al 1999).
Activation of transduction pathways can lead to increased genomic activity and protein turnover. Thus, bAP
is shown to induce cycloheximide-sensitive protein synthesis (Pike et al 1996).
The end point of bAP-induced toxicity is cell death
through necrosis and/or apoptosis. In vitro, this seems to
depend on the excitation state of the neural cultures. Thus,
concomitant with elevated neural excitation, membrane
fragmentation and somatic swelling leading to LDH release ensues (Koh et al 1990; Mattson et al 1992).
However, in the absence of elevated neural excitation,

BIOL PSYCHIATRY
2001;49:221–232

225

bAP seems to cause an apoptosis-path neuronal death
(Forloni et al 1993; Loo et al 1993).
Finally, at low concentrations, early in the course of its
synthesis, bAP may have a different biological function
than later on when its local concentration in the affected
brain is high. Thus, at low (nanomolar or picomolar)
concentrations where no direct cell death can be detected
in vitro, bAP disrupts cholinergic system by 1) reducing
the concentration, release, or synthesis of acetylcholine
and 2) reducing choline acetyltransferase and the activity
of pyruvate dehydroxygenase (involved in the production
of acetylcholine) (Auld et al 1998; Hoshi et al 1997; Kar
et al 1996; Roßner et al 1998). If AD in general and bAP
in specific compromise the cholinergic system, intervention techniques such as “replacement” of cholinergic
nuclei or raising the brain’s cholinergic “output” through
pharmacologic means may help to alleviate some of the
symptoms of the disease.

The Role of Nicotine against bAP Toxicity
Recently, several laboratories have addressed the interaction between bAP and the cholinergic system
(Cheung et al 1993; Itoh et al 1996; Kihara et al 1997,
1998; Kim et al 1997; Maurice et al 1996; Monteggia et
al 1994; Singh et al 1998a, 1998b; Zamani et al 1997).
In most cases bAP-induced toxicity can be significantly, or even completely, prevented by activating the
nicotinic system.
In our hands (Zamani et al 1997), toxicity mediated by
bAP (25–35) (100 mmol/L, 5 days) was partially, but
significantly, prevented in hippocampal neurons by nicotine in a dose-dependent manner and with maximum
inhibition at 10 mmol/L (Figure 2). The degree of protection was significantly larger when samples were preexposed to nicotine for 5 days (chronic) rather than 15 min
(acute). At all concentrations, the protective effect of
nicotine differed significantly between the two pretreatment time courses.
The protective effect of nicotine was receptor mediated. Thus, we showed that coadministration of
mecamylamine (0.1–10 mmol/L) and nicotine (10
mmol/L) to hippocampal cultures for 5 days significantly diminished the neuroprotection by nicotine (Figure 2). Mecamylamine (10 mmol/L, 5 days) pretreatment alone did not affect neuronal viability or prevent
bAP (25–35) toxicity (Figure 2). Similar results have
been reported (Kihara et al 1997, 1998). The muscarinic
system does not seem to be involved in neuroprotection,
since the application of muscarine itself did not offer
any protection against bAP toxicity (Kihara et al 1997).
Further investigation by Kihara et al (1998) unveiled
the role of specific nAChR subtypes in neuroprotection.

226

BIOL PSYCHIATRY
2001;49:221–232

Figure 2. (Line graph) Nicotine dose response and b-amyloid
protein (bAP) (25–35) toxicity at two different pretreatment
times. Cultures were pretreated with nicotine (0.01–100 mmol/L)
for either 15 min or 5 days before bAP (25–35) (100 mmol/L, 5
days) treatment. Significantly more neuroprotection was seen in
a 5-day nicotine pretreatment protocol. (Bar graph)
Mecamylamine (Mec, 10 mmol/L) coadministered with nicotine
(5-day pretreatment protocol) significantly prevented neuroprotection by nicotine (*p,0.05). Mecamylamine alone had no
effect on neuronal viability or, when added with bAP, did not
prevent neuronal loss. 1, bAP (100 mmol/L) present. (Reproduced with permission from Zamani et al 1997.)

Thus, simultaneous application of dihydro-b-erythroidine, which targets the a4 –b2 nicotinic receptor, prevented the protective effect of nicotine. Cytisine, a
specific agonist of the a4 –b2 receptor, however,
showed a dose-dependent neuroprotective effect. In
contrast, some studies have shown protective effects
against bAP toxicity with an a7 nAChR partial agonist
[3-(2,4-dimethoxybenzylidene)anabaseine] in human
cell lines (Meyer et al 1998). In the same way as with
Kihara et al (1998), the partial agonist in this study was
added simultaneously with bAP (25–35). The apparent
discrepancy in these reports is most likely the result of
tissue specificity in terms of nAChR subtype distribution and function, but the underlying biological pathways involved in nicotinic protection could be very
similar. This point is further elucidated when one
compares the protective role of nicotine with other
neurotoxins such as glutamate. Here, Minana et al
(1998) proposed an a7-sensitive path for glutamate- or
NMDA-mediated toxicity, as opposed to an a4 –b2mediated path for bAP, as found by others (Abe and
Kimura 1996; Kihara et al 1998) and by us (M.R.
Zamani and Y.S. Allen, unpublished data).
The “chronic” exposure of neurons to nicotine in our

M.R. Zamani and Y.S. Allen

study (Zamani et al 1997) was an attempt to mimic the
condition where nicotine intake through cigarette smoking
may precede the onset of AD by many years. The lack of
protection against bAP toxicity by nicotine after acute
exposure was significant and contrasted with studies
where concomitant application of the amyloid fragments
and nicotine resulted in significant protection (Kihara et al
1997, 1998). Explanations for this discrepancy may be
found in bAP concentrations, its “aging” state, and the
length of exposure of neurons to the amyloid fragment.
Aggregation of bAP is not an issue because the solutions
were made several hours before treatment, during which
aggregation would have occurred (Pike et al 1995). We
therefore anticipated that our experimental regimen would
be more detrimental to the neurons than the protocol used
by others (e.g., Kihara et al 1997, 1998) and that our
results pointed to nicotine-induced long-term cellular
events, possibly involving genomic activation, that rendered neurons resistant to bAP toxicity.
In addition to the in vitro studies, many experiments
have been carried out in intact animals to demonstrate
bAP toxicity and its prevention by nicotine treatment. For
example, rodents infused cerebrally with either fresh or
aged bAP (25–35 or 1– 40) show significantly reduced
acetylcholine and dopamine release, and choline acetyltransferase activity (Itoh et al 1996; Maurice et al 1996).
Cognitive functions were also impaired. However, if mice
are treated with nicotine or the cholinesterase inhibitor
tacrine, the bAP effect is no longer detected (Maurice et al
1996; Nitta et al 1994).
Although convincing evidence by us and others (Kihara
et al 1997, 1998) demonstrates the involvement of
nAChRs in nicotine-mediated protection, data from several laboratories suggest a more direct role for nicotine
that bypasses its putative receptors. For instance, Singh et
al (1998a, 1998b) show that nicotine prevents the activation by bAP (25–35) of phospholipase A2 and phospholipase D, but not phospholipase C. This may occur through
an “atypical” acetylcholine pathway, which is not sensitive
to known nAChR antagonists such as hexamethonium and
D-tubocurarine. This is similar to data published by Marin
et al (1997) involving elevation of phospholipase A2
activity by excitatory amino acids. The nicotine inhibition
of bAP toxicity is not likely to be due to the ability of
nicotine to prevent aggregation of the peptide (Salomon et
al 1996), since in similar preparations nicotine failed to
affect bAP-induced elevation in phospholipase C activity
(Singh et al 1998b).
Although the exact mechanism of nicotine protection
against bAP neurotoxicity is yet to be elucidated, reports
examining nicotine and cholinesterase inhibitors point to a
variety of biological events. These range from stimulation
of neurotransmitter release to production of trophic factors

Nicotine and bAP Toxicity

and inhibition of b-sheet formation. A few scenarios are
considered here. Nicotine replenishment of the central
nervous system (CNS) may substitute for the downregulated cholinergic system of AD patients leading to increased neurotransmitter release at cholinergic sites. This
should result in improved cognitive functions. Whether
this nicotinic replacement results in slowing down of the
progression of AD neuropathology is hard to prove, but
epidemiologic studies from smokers, who on average
show a delayed onset of AD, would support this conclusion. However, no evidence suggests that the course of the
disease can be reversed by current therapeutic approaches.
Nicotine may also induce the production of trophic factors.
In support of this view, Pike et al (1996) found a resistance to
toxicity by bAP (1– 42) and (25–35) in hippocampal cultures
that were pretreated with depolarizing concentrations of K1.
This process was dependent on VDCC activity and protein
synthesis. Pre-exposure to high K1 concentrations may
trigger synthesis of certain proteins in neurons that render
them resistant to bAP toxicity (Pike et al 1996). Similarly,
cells more prone to bAP toxicity, when cultured at low
density, were rescued by basic fibroblast growth factor
pretreatment (Mattson et al 1993b).
Our data (Zamani et al 1997) and those of others
suggest that inhibition of b-sheet formation by nicotine (as
reported by Salomon et al 1996) does not play a major role
in neuroprotection, since chronic nicotine pretreatment
significantly prevented bAP toxicity, whereas concomitant exposure with bAP did not. Nicotinic receptor modulation and/or trophic factor production may play a more
significant role in this process.
Finally, recent data suggest that nicotine could protect
against bAP by inducing the release of secreted forms of
bAPP. This has been shown by Kim et al (1997) to occur in
PC12 cells without affecting APP mRNA expression. The
secreted bAPP, produced by a-secretase, is known to have
neurotrophic effects and regulate [Ca21] levels (Mattson et al
1993b). The release of secreted bAPP is induced by nicotine
and requires functional nAChRs. Furthermore, these soluble
fragments can activate fast inactivating K1 channels (Furukawa et al 1996). Nicotine may, therefore, use the same
amyloid protein processing machinery to induce a reduction
in neural excitation, which then counters the toxic effects of
bAP. Whether nicotine can modulate neurotransmitter release in the CNS of smokers in a similar manner or whether
comparable mechanisms are responsible for neuroprotection
against bAP and possibly in the late onset of AD and
Parkinson’s disease is yet uncertain.

Summary
The pathogenesis of AD involves abnormal production or
processing of bAP, especially since Down’s syndrome

BIOL PSYCHIATRY
2001;49:221–232

227

(trisomy 21) invariably results in AD (the gene for APP is
located on chromosome 21) (Kola and Hertzog 1997).
Alzheimer’s disease is also associated with impairment of
the cholinergic system, particularly the nicotinic component, with a loss of acetylcholinesterase and choline
acetyltransferase. Early findings by Bowen et al (1976),
Davies and Maloney (1976), and Perry et al (1977) have
yielded much research. However, in this brief review
attention was focused on the neural nicotinic system, bAP
toxicity, and nicotine “replacement” for protection against
bAP toxicity.
How bAP production and its gradual deposition lead to
neuronal loss is not fully understood, but growing evidence supports the following model:
1. b-Amyloid protein membrane insertion destabilizes
the membrane, affecting its fluidity (Avdulov et al
1997; Muller et al 1995; for a review, see Kanfer et
al 1999).
2. There is an increase in [Ca21] (Mogensen et al
1998) through either production of cation ionophores or activation of ligand-gated calcium channels and/or VDCCs, directly, or through G proteinsensitive signals (Arispe et al 1994b; Sanderson et al
1997).
3. Calcium-sensitive cascades are then triggered, leading to enhanced transmitter release, increased tau
protein phosphorylation resulting in formation of
neurofibrillary tangles, and protein turnover.
4. Neuronal death ensues either by apoptosis or necrosis of primarily the cholinergic system.
In considering therapeutic approaches to AD and the use
of cholinergic drugs, one is reminded of the great complexity of the nicotinic cholinergic system in the CNS,
which is due to the fact that 1) nAChR subunits change
during development and aging and are present in most
areas of the brain that may or may not be affected by the
disease; 2) nAChR subunits can form functional receptors
in various combinations with potentially different pharmacologic properties; 3) the neuronal cholinergic system
comprises two different forms of local circuitry and
projection fibers emanating from specific brain nuclei/
regions; 4) these fibers may form presynaptic, postsynaptic, or somatic innervation; 5) presynaptic nAChRs enhance the release of neurotransmitters including
acetylcholine; and 6) the presence of axo-axonic synapses
with sensitivity to low concentrations of nicotine suggests
that at least under some conditions the nicotinic system
may enhance transmission in a diffuse manner rather than
take part in the activity of a specific neural network for
information processing (Sivilotti and Colquhoun 1995;
Wonnacott 1997). The muscarinic system may also be
influenced by therapeutic drugs such as acetylcholinester-

228

BIOL PSYCHIATRY
2001;49:221–232

ase inhibitors, resulting in adverse effects such as receptor
downregulation or desensitization in areas of the brain
unaffected by AD (however, for an alternative suggestion,
see Abdallah and el-Fakahany 1991).
To minimize the undesirable side effects associated
with nonspecific drugs, recent developments have resulted
in the production of pharmacologically more selective
cholinergic drugs. An example of such a drug specifically
targeting a7 nicotinic receptor subtypes is the partial
agonist 3-(2,4-dimethoxybenzylidene)anabaseine, which
is currently in clinical trails (Meyer et al 1998) and is
shown to have positive therapeutic effects.
An alternative approach for the elevation of cholinergic
output in the brain of AD patients is the use of cholinesterase inhibitors. Noticeably, their effect is indiscriminate,
resulting in enhanced excitatory, inhibitory, modulatory,
or even acetylcholine release at presynaptic sites where
nicotinic receptors are found but may be relatively spared
by AD. However, if the “soup theory of the brain” is to be
favored (Sivilotti and Colquhoun 1995), minimal side
effects would be expected.
Cholinesterase inhibitors (some reversible and some
irreversible) such as tacrine, donepezil, physostigmine,
eptastigmine, metrifonate, rivastigmine, and galantamine
have now been used in several studies in patients with
Alzheimer-type syndrome and have generally resulted in
some improvement, sometimes lasting several years
(Eagger et al 1991; Krall et al 1999; Nordberg 1993;
Nordberg and Svensson 1998; Summers et al 1986).
However, a degree of “habituation” to some of these drugs
is apparent.
In summary, drugs targeting specific nicotinic (or muscarinic) subtypes with preferential central activity will be
of future interest, but the cholinesterase inhibitors may
still prove to be the most promising in alleviating symptoms of AD.

The authors thank The Wellcome Trust for financially supporting this
work, and very much appreciate the intellectual involvement of Professor
Jeffery Gray in the initial stages of this project.
Aspects of this work were presented at the symposium “Nicotine
Mechanisms in Alzheimer’s Disease,” March 16 –19, 2000, Fajardo,
Puerto Rico. The conference was sponsored by the Society of Biological
Psychiatry through an unrestricted educational grant provided by Janssen
Pharmaceutica LP.

References
Abdallah EA, el-Fakahany EE (1991): Lack of desensitization of
muscarinic receptor-mediated second messenger signals in rat
brain upon acute and chronic inhibition of acetylcholinesterase. J Biochem Toxicol 6:261–268.
Abe K, Kimura H (1996): Amyloid beta toxicity consists of a

M.R. Zamani and Y.S. Allen

Ca(21)-independent early phase and a Ca(21)-dependent
late phase. J Neurochem 67:2074 –2078.
Akaike A, Tamura Y, Yokota T, et al (1994): Nicotine-induced
protection of cultured cortical neurons against N-methyl-Daspartate receptor-mediated glutamate cytotoxicity. Brain Res
644:181–187.
Alkondon M, Pereira EF, Barbosa CT, Albuquerque EX (1997):
Neuronal nicotinic acetylcholine receptor activation modulates gamma-aminobutyric acid release from CA1 neurons of
rat hippocampal slices. J Pharmacol Exp Ther 283:1396 –
1411.
Araujo DM, Lapchak PA, Collier B, Quirion R (1988): Characterization of N-[3H]methylcarbamylcholine binding sites and
effect of N-methylcarbamylcholine on acetylcholine release
in rat brain. J Neurochem 51:292–299.
Arias C, Arrieta I, Tapia R (1995): beta-Amyloid peptide
fragment 25-35 potentiates the calcium-dependent release of
excitatory amino acids from depolarized hippocampal slices.
J Neurosci Res 41:561–566.
Arispe N, Pollard HB, Rojas E (1994a): The ability of amyloid
beta-protein [A beta P (1-40)] to form Ca21 channels
provides a mechanism for neuronal death in Alzheimer’s
disease. Ann N Y Acad Sci 747:256 –266.
Arispe N, Pollard HB, Rojas E (1994b): beta-Amyloid Ca(21)channel hypothesis for neuronal death in Alzheimer disease.
Mol Cell Biochem 140:119 –125.
Auld DS, Kar S, Quirion R (1998): Beta-amyloid peptides as
direct cholinergic neuromodulators: A missing link? Trends
Neurosci 21:43– 49.
Avdulov NA, Chochina SV, Igbavboa U, et al (1997): Amyloid
beta-peptides increase annular and bulk fluidity and induce
lipid peroxidation in brain synaptic plasma membranes.
J Neurochem 68:2086 –2091.
Behmand RA, Harik SI (1992): Nicotine enhances 1-methyl-4phenyl-1,2,3,6-tetrahydropyridine neurotoxicity. J Neurochem 58:776 –779.
Bharucha NE, Stokes L, Schoenberg BS, et al (1986): A
case-control study of twin pairs dicordant for Parkinson’s
disease: A search for environmental risk factors. Neurology
36:284 –288.
Bowen DM, Smith CB, White P, Davison AN (1976): Neurotransmitter-related enzymes and indices of hypoxia in senile
dementia and other abiotrophies. Brain 99:459 – 496.
Broide RS, O’Connor LT, Smith MA, et al (1995): Developmental expression of alpha 7 neuronal nicotinic receptor messenger RNA in rat sensory cortex and thalamus. Neuroscience
67:83–94.
Brorson JR, Bindokas VP, Iwama T, et al (1995): The Ca21
influx induced by beta-amyloid peptide 25-35 in cultured
hippocampal neurons results from network excitation. J Neurobiol 26:325–338.
Calligaro DO, O’Malley PJ, Monn JA (1993): beta-Amyloid
(25-35) or substance P stimulates [3H]MK-801 binding to rat
cortical membranes in the presence of glutamate and glycine.
J Neurochem 60:2297–2303.
Candy JM, Perry RH, Perry EK, et al (1983): Pathological
changes in the nucleus of Meynert in Alzheimer’s and
Parkinson’s diseases. J Neurol Sci 59:277–289.
Castro NG, Albuquerque EX (1995): alpha-Bungarotoxin-sensi-

Nicotine and bAP Toxicity

tive hippocampal nicotinic receptor channel has a high
calcium permeability. Biophys J 68:516 –524.
Cheung NS, Small DH, Livett BG (1993): An amyloid peptide,
beta A4 25-35, mimics the function of substance P on
modulation of nicotine-evoked secretion and desensitization
in cultured bovine adrenal chromaffin cells. J Neurochem
60:1163–1166.
Colom LV, Diaz ME, Beers DR, et al (1998): Role of potassium
channels in amyloid-induced cell death. J Neurochem 70:
1925–1934.
Copani A, Bruno V, Battaglia G, et al (1995): Activation of
metabotropic glutamate receptors protects cultured neurons
against apoptosis induced by beta-amyloid peptide. Mol
Pharmacol 47:890 – 897.
Court JA, Lloyd S, Johnson M, et al (1997): Nicotinic and
muscarinic cholinergic receptor binding in the human hippocampal formation during development and aging. Brain
Res Dev Brain Res 101:93–105.
Cowburn RF, Messamore E, Li ML, et al (1994): beta-Amyloid
related peptides exert differential effects on [3H]MK-801
binding to rat cortical membranes. Neuroreport 5:405– 408.
Cowburn RF, Wiehager B, Trief E, et al (1997): Effects of
beta-amyloid-(25-35) peptides on radioligand binding to excitatory amino acid receptors and voltage-dependent calcium
channels: Evidence for a selective affinity for the glutamate
and glycine recognition sites of the NMDA receptor. Neurochem Res 22:1437–1442.
Davidson RM, Shajenko L, Donta TS (1994): Amyloid betapeptide (A beta P) potentiates a nimodipine-sensitive L-type
barium conductance in N1E-115 neuroblastoma cells. Brain
Res 643:324 –327.
Davies P, Feisullin S (1981): Postmortem stability of alphabungarotoxin binding sites in mouse and human brain. Brain
Res 216:449 – 454.
Davies P, Maloney AJF (1976): Selective loss of central cholinergic neurons in Alzheimer’s disease. Lancet 2:1403.
Decker ER, Dani JA (1990): Calcium permeability of the
nicotinic acetylcholine receptor: the single-channel calcium
influx is significant. J Neurosci 10:3413–3420.
Eagger SA, Levy R, Sahakian BJ (1991): Tacrine in Alzheimer’s
disease. Lancet 337:989 –992.
Ekinci FJ, Malik KU, Shea TB (1999): Activation of the L
voltage-sensitive calcium channel by mitogen-activated protein (MAP) kinase following exposure of neuronal cells to
beta-amyloid. MAP kinase mediates beta-amyloid-induced
neurodegeneration. J Biol Chem 274:30322–30327.
Fisher JL, Pidoplichko VI, Dani JA (1998): Nicotine modifies
the activity of ventral tegmental area dopaminergic neurons
and hippocampal GABAergic neurons. J Physiol (Paris)
92:209 –213.
Flynn DD, Mash DC (1986): Characterization of L-[3H]nicotine
binding in human cerebral cortex: comparison between Alzheimer’s disease and the normal. J Neurochem 47:1948 –
1954.
Forloni G, Chiesa R, Smiroldo S, et al (1993): Apoptosis
mediated neurotoxicity induced by chronic application of beta
amyloid fragment 25-35. Neuroreport 4:523–526.
Fu Y, Matta SG, McIntosh JM, Sharp BM (1999): Inhibition of
nicotine-induced hippocampal norepinephrine release in rats

BIOL PSYCHIATRY
2001;49:221–232

229

by alpha-conotoxins MII and AuIB microinjected into the
locus coeruleus. Neurosci Lett 266:113–116.
Furukawa K, Barger SW, Blalock EM, Mattson MP (1996):
Activation of K1 channels and suppression of neuronal
activity by secreted beta-amyloid-precursor protein. Nature
379:74 –78.
Granon S, Poucet B, Thinus-Blanc C, et al (1995): Nicotinic and
muscarinic receptors in the rat prefrontal cortex: differential
roles in working memory, response selection and effortful
processing. Psychopharmacology 119:139 –144.
Gray R, Rajan AS, Radcliffe KA, et al (1996): Hippocampal
synaptic transmission enhanced by low concentrations of
nicotine. Nature 383:713–716.
Guan ZZ, Zhang X, Ravid R, Nordberg A (2000): Decreased
protein levels of nicotinic receptor subunits in the hippocampus and temporal cortex of patients with Alzheimer’s disease.
J Neurochem 74:237–243.
Hellstrom-Lindahl E, Mousavi M, Zhang X, et al (1999):
Regional distribution of nicotinic receptor subunit mRNAs in
human brain: Comparison between Alzheimer and normal
brain. Brain Res Mol Brain Res 66:94 –103.
Hensley K, Carney JM, Mattson MP, et al (1994): A model for
beta-amyloid aggregation and neurotoxicity based on free
radical generation by the peptide: Relevance to Alzheimer
disease. Proc Natl Acad Sci U S A 91:3270 –3274.
Hodges H, Allen Y, Kershaw T, et al (1991a): Effects of
cholinergic-rich neural grafts on radial maze performance of
rats after excitotoxic lesions of the forebrain cholinergic
projection system—I. Amelioration of cognitive deficits by
transplants into cortex and hippocampus but not into basal
forebrain. Neuroscience 45:587– 607.
Hodges H, Allen Y, Sinden J, et al (1991b): Effects of cholinergic-rich neural grafts on radial maze performance of rats
after excitotoxic lesions of the forebrain cholinergic projection system—II. Cholinergic drugs as probes to investigate
lesion-induced deficits and transplant-induced functional recovery. Neuroscience 45:609 – 623.
Hoshi M, Takashima A, Murayama M, et al (1997): Nontoxic
amyloid beta peptide 1-42 suppresses acetylcholine synthesis.
Possible role in cholinergic dysfunction in Alzheimer’s disease. J Biol Chem 272:2038 –2041.
Itoh A, Nitta A, Nadai M, et al (1996): Dysfunction of cholinergic and dopaminergic neuronal systems in beta-amyloid
protein–infused rats. J Neurochem 66:1113–1117.
Janson AM, Møller A (1993): Chronic nicotine treatment counteracts nigral cell loss induced by a partial mesodiencephalic
hemitransection: An analysis of the toal number and mean
volume of neurons and glia in substantia nigra of the male rat.
Neuroscience 57:931–941.
Jones GM, Sahakian BJ, Levy R, et al (1992): Effects of acute
subcutaneous nicotine on attention, information processing
and short-term memory in Alzheimer’s disease. Psychopharmacology 108:485– 494.
Joseph R, Han E (1992): Amyloid beta-protein fragment 25-35
causes activation of cytoplasmic calcium in neurons. Biochem
Biophys Res Commun 184:1441–1447.
Kanfer JN, Sorrentino G, Sitar DS (1999): Amyloid beta peptide
membrane perturbation is the basis for its biological effects.
Neurochem Res 24:1621–1630.
Kar S, Seto D, Gaudreau P, Quirion R (1996): Beta-amyloid-

230

BIOL PSYCHIATRY
2001;49:221–232

related peptides inhibit potassium-evoked acetylcholine release from rat hippocampal slices. J Neurosci 16:1034 –1040.
Kihara T, Shimohama S, Sawada H, et al (1997): Nicotinic
receptor stimulation protects neurons against beta-amyloid
toxicity. Ann Neurol 42:159 –163.
Kihara T, Shimohama S, Urushitani M, et al (1998): Stimulation
of alpha4beta2 nicotinic acetylcholine receptors inhibits betaamyloid toxicity. Brain Res 792:331–334.
Kim SH, Kim YK, Jeong SJ, et al (1997): Enhanced release of
secreted form of Alzheimer’s amyloid precursor protein from
PC12 cells by nicotine. Mol Pharmacol 52:430 – 436.
Knott VJ, Harr A, Mahoney C (1999): Smoking history and
aging-associated cognitive decline: An event-related brain
potential study. Ne