Neuron, Vol. 37, 925–937, March 27, 2003, Copyright 2003 by Cell Press
APP Processing and Synaptic Function
tive disorder that is pathologically characterized by ex-
tracellular deposits of ? amyloid (A?) in senile plaques,
intraneuronal neurofibrillary tangles, depressed brain
function, and neuronal death (reviewed in Price and Si-
tion ofA?, asmall peptidewith ahigh propensity toform
aggregates, is central to the pathogenesis of disease
(Selkoe, 2000). Although the potential neurotoxic prop-
et al., 1989), it is still not known how A? participates in
a pathologic cascade resulting in progressive cognitive
decline. It is also not known if A?, which is detected in
both cerebrospinal fluid and plasma in healthy individu-
als throughout life (Seubert et al., 1992), plays a role in
The proteolytic processing pathways leading to the
formation of A? from the amyloid precursor protein
(APP), a type I membrane protein, have been well char-
acterized in a number of cell lines (Figure 1A) (Selkoe,
2000). APP is delivered to the surface membrane where
it is subject to proteolytic processing by ?-secretase.
APP molecules that fail to be cleaved by ?-secretase
can be internalized into endocytic compartments and
tase to generate A?. A fraction of A? peptides are also
generated in the Golgi apparatus and, to a lesser extent,
the endoplasmic reticulum. A? peptides generated in
the Golgi and in recycling compartments can be se-
creted into the extracellular space (Greenfield et al.,
acids in length (A?40), although the smaller fraction of
longer, 42 amino acid species (A?42) have received
to nucleate and drive production of amyloid fibrils (Jar-
rett et al., 1993).
Recent lines of experimental evidence have sug-
gested that excessive amounts of A? are deleterious to
neuronal function, in addition to, or in lieu of, its pro-
posed neurotoxic effects. First, addition of A? in various
aggregation states to neuronal preparations has been
shown to elicit electrophysiological phenotypes (Cullen
et al., 1997; Freir et al., 2001; Hartley et al., 1999; Kim
ing with the peptides, whose biological properties can
depend on aggregation states and peptide size and
composition (Fezoui et al., 2000; Walsh et al., 1999).
Furthermore, the relevant subcellular sites and (patho)-
physiological concentrations are not known and thus
difficult to mimic by exogenous application. Alternative
various naturally occurring familial AD-linked mutants
ological and behavioral consequences of excessive A?
production and accumulation (Chapman et al., 1999;
Fitzjohn et al., 2001; Hsia et al., 1999; Larson et al., 1999;
Westerman et al., 2002). However, the interpretations
of these studies are complicated by the fact that the
transgenes are expressed at high levels throughout de-
velopment and aging. Furthermore, several domains
Flavio Kamenetz,1,2Taisuke Tomita,4
Helen Hsieh,1,3Guy Seabrook,5
David Borchelt,6Takeshi Iwatsubo,4
Sangram Sisodia,7and Roberto Malinow1,2,3,*
1Cold Spring Harbor Laboratory
Cold Spring Harbor, New York 11724
2Graduate Program in Genetics
3Graduate Program in Neuroscience
State University of New York at Stony Brook
Stonybrook, New York 11794
4Department of Neuropathology and Neuroscience
Graduate School of Pharmaceutical Sciences
University of Tokyo
5Merck Research Laboratories
Merck & Company, Incorporated
770 Sumneytown Pike
West Point, Pennsylvania 19486
6Department of Pathology
Johns Hopkins University School of Medicine
Baltimore, Maryland 21287
7Center for Molecular Neurobiology
Department of Neurobiology, Pharmacology,
The University of Chicago
Chicago, Illinois 60637
A large body of evidence has implicated A? peptides
and other derivatives of the amyloid precursor protein
(APP) as central to the pathogenesis of Alzheimer’s
disease (AD). However, the functional relationship of
physiology is not known. Here, we show that neuronal
activity modulates the formation and secretion of A?
peptides in hippocampal slice neurons that overex-
synaptic transmission onto neurons that overexpress
sion depends on NMDA-R activity and can be reversed
by blockade of neuronal activity. Synaptic depression
from excessive A? could contribute to cognitive de-
cline during early AD. In addition, we propose that
duction may normally participate in a negative feed-
back that could keep neuronal hyperactivity in check.
Disruption of this feedback system could contribute
to disease progression in AD.
Alzheimer’s disease (AD), the most common form of
Figure 1. Neural Activity Controls Formation
of APP Cleavage Derivatives
(A) Biochemical pathways leading to the for-
mation of A? from APP. ? and ? cleavage of
APP result in the production of a large, solu-
ble ectodomain (APPs) and a membrane-
associated carboxy-terminal fragment (CTF).
Cleavage of APP by ?-secretase precludes
production of A?. ?-secretase cleavage of
CTFs produce small peptides (A? and p3)
which can be secreted and a truncated CTF
from mice expressing human APPSwe were
trodoxin [TTX], 1 ?M; 10 mM MgCl2; picro-
toxin [PTX], 100 ?M; flunitrazepam [Flu], 1
?M). Media aliquots were collected and ana-
lyzed for A?40 and A?42. Values expressed
as percentage of secretion seen in sister
slices maintained in control media. All values
different from control, p ? 0.05.
(C) Rat organotypic slices were infected (APP
WT, APP WT ? TTX, APP WT ? Mg) or not
dia (Uninf, APP WT), in culture media with 1
uM TTX (APP WT ? TTX), or in culture media
with 10 mM MgCl (APP WT ? Mg) for 24 hr
(n ? 3 for each). Superfusate media was col-
lected and analyzed for A?40 or A?42 as
(D) Western blots of extracts from organo-
typic APPSWEslices in the presence of ?-secre-
therefore aiding fragment detection) treated
with (n ? 8) or without (n ? 8) picrotoxin (100
?M) and 8 mM CaCl2for 36 hr. Blots were
probed with an antibody specific for ?1
BACE cleavage site (3D6) and subsequently
reprobed using CT-15 for detection of full-
length APP. Bands corresponding to the
BACE APP cleavage product and full-length
APP were quantified by scanning.
(E) Western blots of extracts form organotypic APPSWEslices in the presence of L-685,458 and treated with (n ? 3) or without (n ? 3) TTX for
36 hr as in (D). Note that levels of full-length APP are not significantly changed by either blocking or enhancing neuronal activity (control:
1.0 ? 0.13, TTX: 1.0 ? 0.04, p ? 0.9; control: 1.0 ? 0.11, PTX/Ca2?: 1.2 ? 0.15, p ? 0.3).
(F) IP-Western blot of culture media from organotypic APPSWEslices treated for 36 hr with 1 ?m TTX (n ? 8), picrotoxin (100 mm) and 8 mM
CaCl2(n ? 8), or untreated (n ? 8). APPSWEs?was precipitated from culture media using antibody 54, which specifically recognizes the C
terminus of APPSWEs?, and subsequently blotted with the antibody 22C11, which recognizes the N terminus domain of APP. 1 ?m L-685,458
was present in all samples to parallel (D) and (E) above.
within APP have been described that may have impor-
tant physiological functions (Cao and Sudhof, 2001;
Kamal et al., 2001), and thus, the consequence of over-
expressing APP may have unanticipated effects on
physiological and behavioral measures of learning and
memory. We thus sought to develop a system where
the physiologic effects of acute APP overexpression
could be ascertained and in which the relevant molecu-
lar determinants could be dissected. We employed the
Sindbis expression system (Hayashi et al., 2000; Schle-
singer and Dubensky, 1999; Shi et al., 1999; Zhu et al.,
in organotypic hippocampal slices over a period of be-
tween 12 and 72 hr (Stoppini et al., 1991). In this experi-
mental setting, we could determine the effects of APP
overexpression on synaptic transmission, and the ef-
fects of synaptic transmission on APP processing.
The regulatory mechanisms that control A? biosyn-
thesis have received much interest, as these are attrac-
tive targets for therapeutic intervention. While a number
of proteins have been identified whose expression ap-
et al., 1998; Yu et al., 2000), there is little information
about the neuronal mechanisms that modulate A? pro-
duction. Here, we show that neuronal activity regulates
the production and secretion of A? by controlling APP
we report that A? modulates synaptic strength. Taken
together, these two observations suggest a negative
APP Processing and Synaptic Function
of ?-secretase activity. Thus, A? may play a role in nor-
mal synaptic physiology, as well as in pathological pro-
cesses leading to AD.
or if neuronal activity affects the accessibility of APP to
To test if acutely overexpressed APP is under similar
regulation, we used a Sindbis expression system to de-
liver APP to rat organotypic slice neurons. Media from
slices expressing recombinant APP for 24 hr released
more A? (both A?40 and A?42) into the medium com-
pared to uninfected slices. This secretion could be re-
duced by inhibiting neuronal activity during the incuba-
tion period with TTX or high Mg2?(Figure 1C), indicating
that production and secretion of A? from neurons that
acutely overexpress APP is under similar activity-
dependent regulation as that seen in neurons that ex-
press transgene-derived APPSWE. Lastly, we also found
that the low-level secretion of endogenous A? from rat
organotypic slices could also be reduced by incubation
with TTX for 24 hr (A?40: control: 410.5 ? 36.4 pM [n ?
4], TTX:274.0 ?27.3 pM [n? 3],p ? 0.03;A?42: control:
39.4 ? 2.4 pM [n ? 4], TTX: 25.8 ? 3.3 pM [n ? 3],
p ? 0.03), suggesting that endogenous A? secretion
operates under a similar regulatory mechanism.
Neuronal Activity Controls Formation
and Secretion of A?
To examine A? secretion in neuronal tissue, we mea-
sured (Suzuki et al., 1994) A? in the media collected
from organotypic hippocampal slices prepared from
Swedish APP mutation (APPSWE). This mutation causes
autosomal dominant familial AD in two Swedish pedi-
grees and has been shown to enhance production of
of neuronal activity on A? secretion, slices were main-
tained in the presence of pharmacological agents that
either decrease (tetrodotoxin, high magnesium, or fluni-
trazepam [a GABA-A receptor potentiator]) or increase
(picrotoxin [a GABA-A channel blocker]) neuronal activ-
ity. Agents that decreased or increased activity resulted
in significant reductions or elevations, respectively, in
levels of A? (both A?40 and A?42) detected in the me-
dium (Figure 1B). These results indicate that the secre-
tion of A? from neuronal cells that chronically overex-
press APP can be controlled by neuronal activity.
We next wished to evaluate if neuronal activity regu-
age. These two possibilities can be distinguished by
examining the level of ?-CTF, the membrane-tethered
fragment of APP generated by the action of ?-secretase
(BACE), that is also a substrate for ?-secretase (Figure
1A). We reasoned that if neuronal activity enhances only
?-secretase cleavage, then ?-CTF levels should de-
is enhanced by neuronal activity, then ?-CTF levels
ties are enhanced by increased neuronal activity, then
this issue, we conducted Western blot analysis on ly-
sates of organotypic slices prepared from APPSWE tg
mice. We probed blots with an antibody raised against
the carboxy-terminal 15 amino acids of APP that recog-
cally recognizes the amino terminus of ?-CTF (3D6).
As shown in Figure 1D, increasing neuronal activity by
incubating slices in picrotoxin and elevated Ca2?signifi-
cantly enhanced the levels of ?-CTF. Furthermore, de-
creasing neuronal activity by incubating slices in TTX
significantly decreased levels of ?-CTF (Figure 1E). In
addition, we examined the effects of neuronal activity
on the levels of secreted APPs?, which should parallel
the changes in ?-CTF, if APP processing is altered. As
expected, incubating slices in TTX reduced the amount
of APPs?released into the culture medium. The levels
of APPs?were not increased by incubating slices in PTX
and high Ca2?, suggesting that activity may also affect
degradation of APPs?. Together, these results suggest
that the level of BACE cleavage can be controlled by
neuronal activity. Our experiments, however, do not es-
APP Overexpression Depresses
To address whether APP can control synaptic function,
we overexpressed APP or various APP mutants in wild-
type rat hippocampal slice neurons using the Sindbis
expression system. For these experiments, neurons
were sparselyinfected. Infected neuronswere identified
by coexpressing (cytoplasmic) diffusable GFP (see Ex-
perimental Procedures). One day after infection, the re-
combinant APP distributed homogenously throughout
the dendritic tree including dendritic spines (the sites
of excitatory contacts), with no obvious effects on den-
dritic morphology (Figure 2A). We tested the effects of
APP overexpression on synaptic transmission by com-
paring synaptic responses evoked onto side-by-side
pairs of simultaneously recorded postsynaptic neurons
where only one neuron expresses the exogenous pro-
tein. Excitatory synaptic responses onto neurons ex-
one day after expression, while inhibitory (GABA) cur-
rents were unaffected (Figure 2B) (see Experimental
Procedures for details of response isolation and mea-
surement). No such depression occurred onto cells ex-
pressing only GFP (AMPA: control: 100% ? 7.0%, in-
fected: 96% ? 6.1%, n ? 52, p ? 0.3; NMDA: control:
100% ? 9.5%, infected: 90 ? 10.4, n ? 52, p ? 0.2
[Hayashi et al., 2000]). Neurons expressing APP showed
significant decrease in the frequency of miniature
EPSCs, with no change in their amplitude (Figure 2C)
nor in paired-pulse facilitation (Figure 2D). These results
suggest that the depressive effects of APP overexpres-
sion are due to a decrease in the number of functional
To determineif BACE processing ofrecombinant APP
sion, we expressed a mutant form of APP, APPMV(Citron
et al., 1995), which shows little cleavage by BACE at
the ?1 position of A? (cleavage at this site is required
to generate A?40 or A?42). Expression of this mutant
Figure 2. APP
Excitatory Synaptic Transmission: Require-
ments for APP Processing
(A) Two-photon laser scanning microscopic
image of organotypic hippocampal slice neu-
rons infected with Sindbis virus expressing
APP(myc)-IRES-GFP. Slices were fixed, per-
GFP fluorescence; right, anti-myc fluores-
cence. Scale bars, 10 ?m. Note dendritic
spines in lower panels contain recombinant
(B) Graph of AMPA (n ? 29), NMDA (n ? 25),
and GABA (n ? 44) components of synaptic
transmission measured in pairs of neurons
noninfected (filled bars) and infected (open
bars) with virus expressing wild-type APP. In-
set: sample traces of AMPA (top), NMDA
(middle), and GABA (bottom) responses in
noninfected and infected (arrow) cells. Scale
bars for this and subsequent evoked whole-
cell traces: 20 pA, 20 ms.
(C) Miniature EPSC responses recorded in
slices that did not (control, n ? 12) or did
express APP (n ? 10). Frequency of events
was diminished (control: 0.27 ? 0.05 Hz, in-
fected: 0.17 ? 0.02 Hz), with no change in
their mean amplitude (control: 13.2 ? 0.9 pA,
infected: 11.6 ? 0.7 pA) by APP expression.
bars: 5 pA, 100 ms).
(D) Paired-pulse faciliation evoked onto non-
infected (top) and infected (bottom) neurons.
Bar graph shows averages from n ? 20 cells.
(E) Same graph as (B) for neurons expressing
APPMV(AMPA, n ? 59; NMDA, n ? 45).
(F) Same graph as (B) for transmission re-
and maintained in the ?-secretase inhibitor
L-685,458 (AMPA , n ? 29; NMDA, n ? 29).
(G) Western blot of hippocampal extracts
from slices infected and treated as indicated
above. Blots were probed with anti-APP car-
boxy-terminal antibody, CT-15 (Sisodia et al.,
ure 2E); the effect on transmission by APPMVwas signifi-
cantly different from the effect of wild-type APP (AMPA:
p ? 0.014 [n ? 55, 63]; NMDA: p ? 0.015 [n ? 50,
49], K-S test). Parallel experiments showed that slices
infected with this virus express APPMV, and as expected,
the APP-CTF product that would be generated by pro-
cessing by BACE at ?1 site is not apparent (Figure 2G).
This result indicates that BACE cleavage at the ?1 site
is necessary for the depressive effects of APP overex-
pression on transmission. Furthermore, these results
indicate that ?-secretase-generated ?-CTF is unlikely
to be responsible for the depressive effects, since this
product is still generated from APPMV(Figure 2G).
of APP on the depressive effect on synaptic transmis-
tase inhibitor, L-685,458 (Li et al., 2000). Slices were
infected with a virus producing APP and maintained in
the presence of L-685,458 for 24 hr. Transmission onto
neurons expressing APP was not depressed relative to
nearby control cells (Figure 2F). Parallel experiments
showed that slices expressing APP in the presence
of L-685,458 displayed the expected accumulation of
sis and proteolytic steps prior to ?-secretase cleavage
were not impaired by the addition of L-685,458. Further,
these results indicate that ?-secretase processing of
APP is required for the depressive effects of APP and
that this phenotype is independent of the formation of
the large ectodomain of APP following either ? or ?
cleavage events (Furukawa et al., 1996).
A? Production from APP Mediates Depression
The requirement for BACE and ?-secretase processing
of recombinantAPP stronglysuggests thatA? mediates
the observed phenotype on neuronal transmission. On
the other hand, another carboxy-terminal derivative of
APP, termed ?-CTF, or AICD (see Figure 1A) has been
described that may serve as a transcriptional coactiva-
APP Processing and Synaptic Function
Figure 3. A? Domain of APP Is Necessary
and Sufficient to Depress Synaptic Trans-
(A) Schematic diagram of APP constructs
used in this figure. SP indicates signal
(B) Graph of AMPA (n ? 33) and NMDA (n ?
31) components of synaptic transmission
measured in pairs of neurons noninfected
expressing ?-CTF. Inset: sample traces of
and ?40mV (right) from noninfected (top) and
infected (bottom) neurons.
(C) Same as (B) but slices incubated in the
n ? 33).
(D) Western blot of hippocampal extracts
from slices infected with APPMV in the ab-
sence (left) or presence (right) of ?-secretase
inhibitor L-685,458. Blots were probed with
(E) Same as (B) for neurons expressing
?-CTF(KKKQ) (AMPA, n ? 40; NMDA, n ? 32).
(F) Same as (B) for neurons expressing
APLP2/APP Chimera (AMPA, n ? 35; NMDA,
n ? 33).
tor (Cao and Sudhof, 2001). However, ?-CTF accumu-
lates in cells treated with L-685,458 (Figure 3D), indicat-
and thus likely a precursor of ?-CTF. Taken together
with the finding that abundant amounts of ?-CTF are
produced in APPMV-infected cells (Figure 2G), which
show no synaptic depression (Figure 2E), we feel it un-
on synaptic transmission.
To test more directly whether A? is necessary and
sufficient to produce a depression of synaptic transmis-
sion, we infected hippocampal slice neurons with
Sindbis virus harboring cDNA encoding ?-CTF, a poly-
peptide that includes A?, and the entire transmembrane
and cytoplasmic domains of APP (Figure 3A) [Iwata et
al., 2001]). As expected, expression of ?-CTF was suffi-
cient to depress synaptic transmission (Figure 3B), and
this depression was prevented by incubating slices in
the ?-secretase inhibitor L-685,458 (Figure 3C). Indeed,
expression of a truncated ?-CTF (?-CTFKKKQ [Iwata et
al., 2001]; Figure 3A) that contains A?, the entire trans-
membrane domain, and a KKKQ membrane-anchoring
motif, but lacks the intracellular C terminus, was able
to produce a depression of synaptic transmission (Fig-
ure 3E), although not to the extent produced by expres-
sion of ?-CTF (Figure 3B). The reduced depression by
?-CTFKKKQ may be due to inefficient trafficking of this
polypeptide to sites of action, since cellular targeting
et al., 1999). In any event, these results argue that the
A? domain is sufficient to depress transmission. As a
tic transmission, we generated a chimeric ?-CTF con-
struct in which the A? region was replaced by the corre-
sponding region in APLP2, an APP homolog that has
considerable homology with APP throughout most of
its length except in the A? region, where there is consid-
erable divergence (Wasco et al., 1993) (see Figure 3A).
Expression of this APLP2/APP chimera did not depress
view that the A? domain of APP is crucial for the ob-
served synaptic depression. Taken together, these re-
Figure 4. Synaptic Depression Mediated by
APP Overexpression Is Activity Dependent
(A) Amplitude of AMPA-mediated synaptic
transmission recorded in neurons expressing
APP WT for 24 or 48 hr (n ? 27). Slices were
maintained in normal medium (black), in TTX
24–48 (blue, n ? 37). * indicates significant
difference from value obtained at same time
point in different conditions. Inset: sample
traces of transmission measured at ?60mV
for each of the time points.
(B) Graph of AMPA (n ? 33) and NMDA (n ?
32) components of synaptic transmission
measured in pairs of neurons noninfected
expressing APP WT incubated in the pres-
ence of 100 uM D,L AP5. Inset: sample traces
of transmission measured at ?60mV (left)
and ?40mV (right) from noninfected (top) and
infected (bottom) neurons.
n ? 32).
sults indicate that the A? domain is necessary and suffi-
cient for APP overexpression to produce synaptic
period also blocked APP-induced depression (AMPA:
control: 100% ? 11.3%, infected: 100% ? 12%, n ? 28,
9.7%, n ? 27, p ? 0.3). We next tested whether synaptic
NMDA receptor activation could prevent synaptic de-
pression produced by APP overexpression. Slices were
incubated with 100 ?M D,L AP5, an agent that blocks
NMDA-Rs and has previously been shown not to affect
blockade of NMDA receptors prevented the depression
caused by APP overexpression (Figure 4B). Thus, it ap-
pears that NMDA receptor activation by spontaneous
neuronal activity is required for APP overexpression to
depress synaptic transmission.
Our results above suggest that neural activity affects
A? production by controlling BACE cleavage of APP.
Another prediction from this finding is that blockade of
neural activity should not prevent the synaptic depres-
sion produced by expression of ?-CTF. Indeed, while
the ?-secretase inhibitor did block synaptic depression
by ?-CTF (Figure 3C), TTX did not (Figure 4C). Taken
together, these results provide physiological and bio-
chemical support for the notion that neuronal activity
promotes production of A? by modulating BACE pro-
cessing of APP.
APP-Induced Depression Requires
The results shown above indicate that processing of
APP into A? is dependent on neuronal activity and that
formation of A? results in synaptic depression. Thus, a
direct prediction of these findings is that blockade of
neuronal activity should prevent the depressive actions
of APPoverexpression. To testthis prediction,we main-
tained slices in conditions where spontaneous neuronal
activity was blocked during the expression of full-length
APP and subsequently assayed the effects on physiol-
ogy. AMPA-mediated synaptic transmission onto neu-
rons that express recombinant APP and maintained in
TTX showed no depression of synaptic transmission
compared to neurons expressing recombinant APP and
maintained in normal medium (Figure 4A; 24 hr with or
without TTX, p ? 0.006, K-S test). Similar results were
obtained by measuring NMDA responses (data not
shown). It is notable that in conditions where neuronal
activity is suppressed, APP-overexpressing neurons
demonstrate normal synaptic transmission, near base-
line levels of A?42, but still significant (albeit reduced)
levels of A?40 (Figure 1C). This suggests that A?42,
rather than A?40, may more potently contribute to the
observed synaptic depression.
We tested the generality of this activity-dependent
activity. Indeed,10 ?M NBQX (anAMPA receptor antag-
onist) added to the culture media during the incubation
Depressive Effects of A? Are Reversible
To determine if the depressive effects on transmission
slices overexpressing APP in normal culture media for
24 hr; TTX was then added to fresh media for 24 hr,
and synaptic transmission was subsequently assayed.
Remarkably, after such a protocol, synaptic responses
APP Processing and Synaptic Function
Figure 5. Secreted A? Can Depress Trans-
mission onto Nearby Neurons
(A) Schematic of experimental paradigm. Un-
infected neuron is surrounded by infected
neurons in region a; uninfected neuron is sur-
rounded by uninfected neurons in region b.
Either cell a or cell b was assigned coordi-
nates (0,0) for simplicity; s indicates coordi-
nates of stimulating electrode.
(B) Fluorescent (left) and DIC (right) images
of a small region within the CA1 infected with
a high titer APP virus. Arrow indicates an un-
infected cell surrounded by many infected
(C) Comparison of AMPA transmission be-
tween simultaneously recorded pairs of unin-
fected neurons in the two different regions.
Cumulative distributions were computed us-
ing the ratio of the amplitude response of a
neuron in region a to the amplitude response
were computedfor APPSWEandAPPMV. Ampli-
tuderesponseratios are reportedinalog scale
for display purposes. Inset: sample traces of
AMPA transmission for a pair of uninfected
neurons, one neuron surrounded by APPSWE-
expressing neurons (left), or APPMV-express-
ing neurons (right).
(D) Same as (C) for NMDA transmission.
(E) Field recordings of evoked EPSPs moni-
amounts of A? peptides. After obtaining sta-
solved A? peptides (1 ?m A?40 ? 0.05 ?m
A?42 or 10 ?m A?40 ? 0.5 ?m A?42) were
added to the circulating ACSF, and re-
sponses were monitored for the next 40 min.
in infected cells recovered to control levels (Figure 4A).
Expression of APP for 2 days with no TTX treatment
showed no such recovery (Figure 4A; difference be-
tween 24 hr no drug followed by 24 hr TTX versus 48
for NMDA responses (data not shown). Thus, preventing
neuronal activity for 24 hr can reverse the depression
Our experiments, however, do not indicate if the effects
of chronic APP overexpression can be reversed.
the amount of A? reaching a control neuron is likely to
be orders of magnitude lower than the amount secreted
by the infected neuron (the expression protocol was
purposefully chosen to infect neurons sparsely). To test
if secreted A? can indeed act intercellularly, we com-
pared synaptic function onto twouninfected cells (a and
b); cell a was chosen from a region containing many
infected cells, while cell b was chosen from a region
with no infected cells (Figures 5A and 5B). The two
and a stimulating electrode was placed equidistant to
both cells (see Experimental Procedures). To maximize
A? production, we expressed full-length APPSWE. Excit-
from cells a and b, and a ratio of the amplitude response
(response of cell a/response of cell b) was computed.
Uninfected neurons surrounded by APPSWEhad similar
baseline electrophysiological properties as neurons
from uninfected regions (input resistance: uninfected
region, 215 ? 15 MOhm (n ? 7); infected region, 205 ?
12 MOhm [n?14]; p ? 0.6). In accordance with the pre-
diction that A? can affect neurons in a noncell-auto-
nomous manner, uninfected neurons surrounded by
APPSWE-infected neurons had significantly depressed
A? Depresses Transmission in a Noncell-
We wished to test whether A? produced from overex-
pressing neurons can affect neighboring neurons. The
experiments above show that overexpression of APP
in a neuron can cause synaptic depression onto that
pretation of these results is that A? acts only on the cell
that produces it, that is, in a strictly cell-autonomous
control neurons in these experiments can be affected
by the secreted A?, but they fail to respond because
transmission when comparedto distant control neurons
(?50% reduction in both AMPA- [n ? 20, p ? 0.001]
mission; Figure 5C and 5D, inset), suggesting that unin-
fected neurons in infected regions were responding to
the local high concentrations of A?. In further support
onto a nearby noninfected neuron (AMPA: control:
100% ? 7.8%, infected: 94% ? 8.1%, n ? 36, p ? 0.3;
NMDA: control 100% ? 12.3%, infected 93% ? 13.8%,
n ? 33, p ? 0.2; compare with Figure 2B).
To control for effects other than increased production
of A?, we performed a comparable experiment using
surrounded by APPMV-infected neurons had similar lev-
els of transmission when compared to distant control
neurons (as expected, no reduction in either AMPA [n ?
19, p ? 0.7] or NMDA [n ? 19, p ? 0.6]; Figures 5C and
sion onto uninfected neurons was significantly different
from the effect of APPMV(Figures 5C and 5D). Taken
together, these results indicate that A? from neurons
overexpressing APP can depress synaptic transmission
onto nearby neurons when sufficient levels of A? are
As a further test of the depressive effects of secreted
A?, we bath applied synthetic A? (1-40 and 1-42) onto
hippocampal slices. While it is not known what local
a concentration-dependent depression of transmission
A? depresses synaptic transmission.
We next considered under what conditions of electri-
cal activity this regulation would be operative when APP
and A? are expressed at endogenous levels. As our
previous experiments relied on expression of human
APP, it was important to determine if rodent A?, which
differs from human A? at 3 amino acids, produced a
similar phenotype. This sequence difference has been
suggested to be responsible for the enhanced toxic and
biologically active properties of the human form of the
peptide (Johnstone et al., 1991). Expression of rodent
?-CTF in neurons was sufficient to depress synaptic
transmission (AMPA: control: 100% ? 6.2%, infected:
64.5% ? 8.0%, [n ? 30, p ? 0.001]; NMDA: control
100% ? 13.4%, infected 73.5% ? 12.8%; n ? 23, p ?
0.006).Thus, weconclude thatlike humanA?, rodentA?
peptides are biologically active and capable of exerting
electrophysiological effects, despite their inability to
form amyloid pathology in the rodent brain.
In wild-type animals, acute application of L-685,458
had no effect on basal transmission (45 min after drug,
0.95 ? 0.02 of baseline; 45 min after vehicle, 0.96 ?
0.02; p ? 0.8), no effect on long-term depression (LTD)
(after low-frequency stimulation, vehicle control: 70% ?
2.5% [n ? 25], L-685,458: 73% ? 2.7% (n ? 26); p ?
0.4), nor on single-tetanus LTP (Figure 6B). A pairing
protocol (generally considered to be a stronger stimulus
than a single tetanus) revealed a small (but not signifi-
cant) increase in potentiation in the presence of the
drug (Figure 6A). This effect became more obvious with
repeated tetanic stimulation, which produced signifi-
cantly more potentiation in the presence than in the
absence of L-685,458 (Figure 6C). Furthermore, while
acute application of L-685,458 produced no detectable
cation of the drug increased the frequency of miniature
gest that strong acute electrical activity or low chronic
activity can recruit A?-induced synaptic depression in
required to detect effects of endogenous A? is consis-
tent with the fact that brain tissue from rats or mice
contain very low steady-state levels of A? peptides de-
rived by processing at the ?1 BACE site (De Strooper
et al., 1995).
A? as a Negative Feedback Regulator
of Neuronal Activity
The data presented above indicate that increased neu-
ronal activity promotes the formation of A? and that
increased A? formation depresses synaptic function.
While these relationships were obtained primarily with
overexpressed human APP, they suggest the existence
of a negative feedback process wherein higher levels
of neuronal activity may increase the production of A?
from endogenous APP and lead to synaptic depression
that could curtail excessive activity. We therefore de-
signed a series of experiments to examine the effects
of different levels of APP expression, and different stim-
ulation conditions, on A?-induced synaptic depression.
We first tested the effects of a standard LTP-inducing
protocol in the context of APP overexpression. Follow-
ing a pairing protocol (Figure 6A), neurons from organo-
tiation which returned to baseline levels within 25 min
of LTP induction; potentiation persisted in control neu-
results are consistent with the view that acute enhance-
ment of synaptic activity (e.g., a pairing protocol) can
drive the production of A?, which subsequently de-
presses synaptic function and offsets LTP. This view
is supported by the finding that when using the same
conditions, bath application of L-685,458 (the ?-secre-
tase inhibitor) permitted pairing-induced LTP in neurons
from APPSWEslices (Figure 6A).
Activity-Dependent APP Processing
and A? Secretion
Our studies have identified a novel regulatory mecha-
nism by which individual neurons or neuronal networks
may control A? production and secretion. Generation
of both A?40 and the more fibrillogenic A?42 can be
controlled by neuronal electrical activity. This occurs in
the context of APP overexpression in either transgenic
(chronic) or virally (acutely) driven settings and also un-
der endogenous levels of APP. These results comple-
ment earlier studies indicating that neuronal electrical
ucts in wild-type tissue (Nitsch et al., 1993). Our studies
are consistent with the notion that neuronal electrical
activity modulates APP processing at the ?-secretase
site. It is now fairly well established that neural activity
can regulate the trafficking of proteins at synaptic sites
APP Processing and Synaptic Function
Figure 6. The ?-Secretase Inhibitor L-685,458
A?-Induced Depression in Wild-Type Hippo-
or APPSWEorganotypic mouse slices in the
presence or absence of ?-secretase inhibi-
tion. After a short baseline, LTP was induced
in one pathway by a pairing protocol, while
the other pathway was not paired to monitor
baseline responses. Evoked AMPA-R-medi-
ated synaptic responses from the paired path-
way (top) and control pathways (bottom). Leg-
filledtriangles: Tgslices,DMSO treated;filled
squares: WT slices, L-685,458 treated; open
circles: Tg slices, L-685,458 treated. Inset:
sample traces of AMPA-R mediated synaptic
transmission before (thin) and after (thick)
whole-cell pairing protocol.
(B) Field recordings of evoked EPSPs moni-
tored from acute rat slices exposed (open
symbols, n ? 23) or not exposed (filled sym-
bols, n ? 26) to 1 ?M L-685,458. At arrow,
tetanic stimulation (1 s, 100 Hz) was de-
(C) Same as (A) except ten tetanic stimuli
were delivered, each tetanus (arrow) sepa-
rated by 3 min. Responses following ten te-
tanic stimuli were significantly different, p ?
0.03 (control n ? 17; treated n ? 17). Scale
bars: 200?V, 10 ms.
(D) Miniature EPSC responses recorded in
whole-cell mode from neurons in wild-type
organotypic slices maintained in the absence
(control, n ? 22) or presence of 1 ?M
L-685,458 (n ? 18). Frequency of events was
0.75 ? 0.9 Hz) with no change in their mean
amplitude (control: 11.8 ? 0.5 pA; L-685,458:
11.9 ? 0.9 pA) by drug treatment. Right, sam-
ple traces for each condition.
(e.g., AMPA-Rs [Carroll et al., 2001; Malinow et al.,
2000]), and hence, it is possible that neuronal activity
promotes the endocytosis of surface APP, enhancing
It is of interest, and potential clinical relevance, that
inhibiting excitatory synaptic transmission or blocking
NMDA-Rs prevents the synaptic depressive effects of
APP overexpression. Our results are consistent with the
findings of two clinical studies. First, the benzodiaze-
pines, agents that enhance inhibitory transmission and
thereby decrease excitatory drive, have been found to
protect against Alzheimer’s disease (Fastbom et al.,
1998). This observation is notable in view of our findings
that benzodiazepines reduce secretion of A? peptides
from hippocampal slice neurons (Figure 1B). Second,
aptic depression in our studies, and NMDA-R antago-
nists have been shown to be effective in slowing cogni-
tive decline in mild to severe AD patients (Reisberg et
al., 2000; Winblad and Poritis, 1999).
lieved that A? accumulation in the brain is a critical
component. While A? can be neurotoxic (Yankner et al.,
1990), there is growing evidence that cognitive decline
can occur before, or independent of, neuronal loss, as
amyloid-dependent physiological and behavioral defi-
cits in transgenic mice can occur in the absence of cell
et al., 2002). Here, we show that processing of overex-
pressed APP into A? leads to depression of synaptic
transmission. This result is consistent with previous
studies of transgenic mice expressing APP harboring
mutations known to cause early onset familial Alzhei-
mer’s disease (Fitzjohn et al., 2001; Hsia et al., 1999).
Such mice produce more A? and show depressed syn-
aptic function before amyloid plaque deposition be-
comes evident. Our molecular dissection of the APP
molecule reveals that A? is responsible for this synaptic
depression. Such effects could contribute to the cogni-
tive dysfunction in AD (Selkoe, 2002; Walsh et al., 2002).
A Normal Physiologic Role for A??
While our results are consistent with the notion that
high A? levels may disrupt synaptic function, our data
suggest that A? may also have a normal negative feed-
back function. Increased neuronal activity produces
A? Disrupts Neuronal Transmission: a Mechanism
for the Early Cognitive Defects of AD?
The mechanisms responsible for the cognitive decline
underlying AD are not understood, but it is widely be-
2000). Full-length APP constructs contain the 12 amino acid myc
sequence immediately before the termination codon. APPMVwas
generated using the Quick Change Mutagenesis Kit (Stratagene).
Constructs were cloned into pSinRep5 shuttle vector and infective
sindbis pseudoviruses were generated as described (Malinow et al.,
Organotypic hippocampal slices were prepared from 6-day-old rat
or APPSWEtransgenic mice (Borchelt et al., 1996) and maintained
using standard methods (Stoppini et al., 1991). Four slices were
placed on each membrane. To measure A? secretion from cultured
transgenic mouse slices, slices were maintained with or without
drugs for 4 days before samples were collected for measurements.
Individual values represent pooling media from the subsequent 9
days. A? was measured using a two-site ELISAs that specifically
detect the C terminus of A?, as previously described (Tomita et al.,
1997). For transient APP overexpression studies, A? was measured
from media of infected slices 24 hr after infection. For whole-cell
electrophysiology, slices were allowed to express recombinant pro-
tein for ?24 hr after infection, unless indicated otherwise. Slices
were then transferred to a recording chamber and perfused with
solution (22?C–25?C) containing 119 mM NaCl, 2.5 mM KCl, 4 mM
and 0.002 mM 2-chloroadenosine, at pH 7.4 and gassed with 5%
CO2/95% O2. 100 ?M picrotoxin was included in the bath when
measuring AMPA or NMDA responses; 20 ?M NBQX and 100 ?M
D,L AP5 were added insteadof picrotoxin in experiments measuring
inhibitory currents. 2-chloroadenosine was included to prevent
bursting. Patch recording pipettes (3–6M?) were filled with intracel-
lular solutions containing 115mM cesium methanesulfonate, 20 mM
CsCl, 10 mM HEPES, 2.5 mM MgCl2, 4 mM Na2ATP, 0.4 mM Na3GTP,
10 mM sodium phosphocreatine, and 0.6 mM EGTA, at pH 7.25.
Whole-cell recordings were obtained simultaneously from two post-
synaptic CA1 neurons (typically one infected expressing GFP), and
signals were amplified with Axopatch-1D amplifiers (Axon Instru-
ments). Synaptic responses were evoked by one or two bipolar
electrodes with single voltage pulses (200 ?s, less than 20V). The
stimulating electrodes were placed over Schaffer collateral fibers
?300–500?mfrom therecordedCA1cells.Stimuluslevel wassetto
produce a synaptic response of ?40 pA in the control cell. Synaptic
responses at ?60mV, 0mV, and ?40mV were averaged over 50–100
trials.AMPA receptor-mediatedresponsewasmeasured byaverag-
ing a 5 ms window about the peak response at ?60mV; GABA
receptor-mediated response was measured by averaging a 5 ms
window about the peak response at 0mV. NMDA receptor-mediated
response was measured by averaging a 10 ms window 150 ms
after the stimulus artifact of responses recorded at ?40mV. For
experiments measuring synaptic responses from distant uninfected
cell pairs at positions [0,0] and [x,y], the stimulation electrode was
placed at a site [x?,y?] equidistant for the two cells and a distance
(?300?m) fromthe cellbody layerusing asimple quadraticformula.
trolling stage position. Miniature EPSC events were recorded in the
presence of 1 ?m TTX (no adenosine) and analyzed using the Mini-
Analysis software (Synaptosoft). At least 200 events were obtained
from each cell. Paired pulse facilitation was elicited by using an
interstimulus interval of 50 ms. For whole-cell LTP experiments,
potentiation was induced by pairing 2 Hz stimulation with depolar-
ization of the postsynaptic neuron to 0mV for 120 s; recordings were
maintained for at least 35 min after pairing. The EPSC amplitude
after pairing was normalized to the average amplitude of 20–30
slices were prepared from P14–P21 animals. Slices were preincu-
bated for 1 hr in drug (1 ?M L-685,458) or vehicle (0.1% DMSO) and
placed in a recording chamber containing solution indicated above
(no 2-chloroadenosine). Responses were evoked by alternating
stimuli through two bipolar stimulating electrodes (1V–10V, 200 ?s)
and recorded with glass electrodes placed in CA1 s. radiatum and
amplified with Cyberamp 320 (Axon Instruments). Responses from
four slices were recorded simultaneously; drug and no drug experi-
statistical significance was set at p ? 0.05. Statistical differences of
Figure 7. Negative Feedback Model Indicating Proposed Interac-
tion between Neural Activity and APP Processing
Neural activity regulates ?-secretase actions on APP. Formation of
A? depresses synaptic transmission. Synaptic depression de-
creases neural activity.
tic function; the depressed synaptic function will de-
tic homeostasis have recently been reported (Davis et
al., 1998; Turrigiano et al., 1998), as well as intercellular
depression following strong tetanic stimulation (Scanzi-
ani et al., 1996), although the signaling molecules medi-
ating these processes have not been identified. In sup-
port of this model, we find that in addition to human A?,
rodent A? can also depress synaptic transmission. This
is important because rodent A? is believed not to have
amyloidogenic properties. We find that in wild-type rat
tissue, multiple tetani lead to greater synaptic poten-
tiation in the presence of the ?-secretase inhibitor
L-685,458. This suggests that multiple tetani drive APP
processing, producing a synaptic depression (in addi-
tion to LTP) that can be revealed by ?-secretase inhibi-
tion. This phenomenon is seen when multiple tetani are
delivered suggesting that the negative feedback system
mediated by APP processing may normally only be rap-
idly recruited under very high activity levels. This may
explain the enhanced kainate-induced seizure activity
in APP knockout mice (Steinbach et al., 1998). We also
find that 24 hr application of ?-secretase inhibitor
L-685,458 leads to enhanced synaptic transmission (in-
creased miniature EPSC frequency, Figure 6D). The ab-
sence of an overt phenotype in mice lacking BACE (Cai
et al., 2001; Luo et al., 2001) or APP (Zheng et al., 1996)
suggests that other mechanisms can compensate for
this in these mice.
How could disturbances in this proposed negative
feedback loop contribute to AD? One can envision a
numberof scenarios.Forinstance,if synapseslosesen-
sitivity to A?-induced depression, persistently elevated
neuronal activity may go unchecked. High levels of neu-
ronal activity could lead to excitotoxicity (Zoghbi et al.,
2000), as well as higher levels of secreted A? peptides,
which may in turn form neurotoxic fibrils that eventually
kill neurons (Cotman et al., 1992; Lambert et al., 1998;
synaptic activity), with resulting synaptic depression
and neuronal toxicity.
Construction of Plasmids and Pseudoviruses
APP695 (human) and GFP (Clonetech) were coexpressed by using
an internal ribosomal entry site (IRES) construct (Hayashi et al.,
APP Processing and Synaptic Function
the means were determined using Wilcoxon for paired measure-
ments and t test for nonpaired measurements. To compare the
effects on transmission produced by two different APP constructs,
a cumulative distribution of response ratios (response in infected
cell/response incontrol cell) weregenerated for eachconstruct. The
Kolmogorov-Smirnov (K-S) test was used to determine statistically
significant differences between the two cumulative distributions.
aged amyloid precursor protein transgenic mice. Nat. Neurosci. 2,
Citron, M., Oltersdorf, T., Haass, C., McConlogue, L., Hung, A.Y.,
Seubert, P., Vigo-Pelfrey, C., Lieberburg, I., and Selkoe, D.J. (1992).
Mutation of the beta-amyloid precursor protein in familial Alzhei-
mer’s disease increases beta-protein production. Nature 360,
Citron, M., Teplow, D.B., and Selkoe, D.J. (1995). Generation of
Cotman, C.W., Pike, C.J., and Copani, A. (1992). Beta-amyloid neu-
rotoxicity: a discussion of in vitro findings. Neurobiol. Aging 13,
Cullen, W.K., Suh, Y.H., Anwyl, R., and Rowan, M.J. (1997). Block
of LTP in rat hippocampusin vivo by beta-amyloid precursor protein
fragments. Neuroreport 8, 3213–3217.
Davis, G.W., DiAntonio, A., Petersen, S.A., and Goodman, C.S.
(1998). Postsynaptic PKA controls quantal size and reveals a retro-
grade signal that regulates presynaptic transmitter release in Dro-
sophila. Neuron 20, 305–315.
De Strooper, B., Simons, M., Multhaup, G., Van Leuven, F., Bey-
reuther, K., and Dotti, C.G. (1995). Production of intracellular amy-
loid-containing fragments in hippocampal neurons expressing human
amyloid precursor protein and protection against amyloidogenesis by
subtle amino acid substitutions in the rodent sequence. EMBO J. 14,
Fastbom, J., Forsell, Y., and Winblad, B. (1998). Benzodiazepines
may have protective effects against Alzheimer disease. Alzheimer
Dis. Assoc. Disord. 12, 14–17.
Fezoui, Y., Hartley, D.M., Harper, J.D., Khurana, R., Walsh, D.M.,
Condron, M.M., Selkoe, D.J., Lansbury, P.T., Jr., Fink, A.L., and
Teplow, D.B. (2000). An improved method of preparing the amyloid
beta-protein for fibrillogenesis and neurotoxicity experiments. Amy-
loid 7, 166–178.
Fitzjohn, S.M., Morton, R.A., Kuenzi, F., Rosahl, T.W., Shearman,
M., Lewis, H., Smith, D., Reynolds, D.S., Davies, C.H., Collingridge,
G.L., et al. (2001). Age-related impairment of synaptic transmission
but normal long-term potentiation in transgenic mice that over-
express the human APP695SWE mutant form of amyloid precursor
protein. J. Neurosci. 21, 4691–4698.
Freir, D.B., Holscher, C., and Herron, C.E. (2001). Blockade of long-
term potentiation by beta-amyloid peptides in the CA1 region of the
rat hippocampus in vivo. J. Neurophysiol. 85, 708–713.
Furukawa, K., Barger, S.W., Blalock, E.M., and Mattson, M.P. (1996).
Activation of K? channels and suppression of neuronal activity by
secreted beta-amyloid-precursor protein. Nature 379, 74–78.
Greenfield, J.P., Tsai, J., Gouras, G.K., Hai, B., Thinakaran, G.,
Checler, F., Sisodia, S.S., Greengard, P., and Xu, H. (1999). Endo-
tions of Alzheimer beta-amyloid peptides. Proc. Natl. Acad. Sci.
USA 96, 742–747.
Hartley, D.M., Walsh, D.M., Ye, C.P., Diehl, T., Vasquez, S., Vassilev,
P.M., Teplow, D.B., and Selkoe, D.J. (1999). Protofibrillar intermedi-
ates of amyloid beta-protein induce acute electrophysiological
changes and progressive neurotoxicity in cortical neurons. J. Neu-
rosci. 19, 8876–8884.
Hayashi, Y., Shi, S.H., Esteban, J.A., Piccini, A., Poncer, J.C., and
Malinow, R. (2000). Driving AMPA receptors into synapses by LTP
and CaMKII: requirement for GluR1 and PDZ domain interaction.
Science 287, 2262–2267.
Howland, D.S., Trusko, S.P., Savage, M.J., Reaume, A.G., Lang,
D.M., Hirsch, J.D., Maeda, N., Siman, R., Greenberg, B.D., Scott,
R.W., et al. (1998). Modulation of secreted beta-amyloid precursor
protein and amyloid beta-peptide in brain by cholesterol. J. Biol.
Chem. 273, 16576–15682.
Hsia, A.Y., Masliah, E., McConlogue, L., Yu, G.Q., Tatsuno, G., Hu,
Plaque-independent disruption of neural circuits in Alzheimer’s dis-
ease mouse models. Proc. Natl. Acad. Sci. USA 96, 3228–3233.
Organotypic slices from APPSWEtransgenic mice or wild-type rats
sis (Sisodia et al., 1993). APP fragments were identified by the cyto-
plasmic terminus antibody CT15, which recognizes full-length APP
as well as CTFs, or the ?-cleavage site-specific antibody 3D6 (Bac-
skai et al., 2001). Secreted APPs?measurements were carried out
by immunoprecipitating APPs?from ?500 ? of culture media using
antibody 54, which specifically recognizes the C terminus of APPSWEs?
(Howland et al., 1998), and subsequently blotting using antibody
22C11, which recognizes the N terminus of APP. Quantifications of
immunoblots were done by enhanced chemiluminescence (ECL;
NEN) and densitometric scanning of the films under linear exposure
Immunohistochemistry was performed on organotypic slices after
1–2 days of infection with the wild-type APP IRES construct that
solution overnight at 4?C and blocked in PBS containing 10% horse
serum, 0.2% sodium azide, and 0.1% triton overnight at 4?C. Slices
were incubated in anti-c-Myc (Calbiochem) in blocking buffer over-
night at 4?C. Following a 2 hr room-temperature incubation with a
biotin-conjugated secondary antibody, avidin-Texas red was added
for 2 hr at room temperature. Immunohistochemistry was imaged
using a custom built two-photon laser scanning microscope.
We thank Alan Nadin (Merck) for the synthesis of L-685,458 and
Charles Glabe (University of California at Irvine) for A?40 and A?42
Received: April 5, 2002
Revised: January 16, 2003
Bacskai, B.J., Kajdasz, S.T., Christie, R.H., Carter, C., Games, D.,
beta deposits in brains of living mice permits direct observation of
clearance of plaques with immunotherapy. Nat. Med. 7, 369–372.
Borchelt, D.R., Thinakaran, G., Eckman, C.B., Lee, M.K., Davenport,
F., Ratovitsky, T., Prada, C.M., Kim, G., Seekins, S., Yager, D., et
al. (1996). Familial Alzheimer’s disease-linked presenilin 1 variants
elevate Abeta1-42/1-40 ratio in vitro and in vivo. Neuron 17, 1005–
Borchelt, D.R., Ratovitski, T., van Lare, J., Lee, M.K., Gonzales, V.,
Jenkins, N.A., Copeland, N.G., Price, D.L., and Sisodia, S.S. (1997).
Accelerated amyloid deposition in the brains of transgenic mice
coexpressing mutant presenilin 1 and amyloid precursor proteins.
Neuron 19, 939–945.
Cai, H., Wang, Y., McCarthy, D., Wen, H., Borchelt, D.R., Price,
D.L., and Wong, P.C. (2001). BACE1 is the major beta-secretase for
generation ofAbeta peptides byneurons. Nat. Neurosci.4, 233–234.
Cao, X., and Sudhof, T.C. (2001). A transcriptively active complex
of app with fe65 and histone acetyltransferase tip60. Science 293,
Carroll, R.C., Beattie, E.C., von Zastrow, M., and Malenka, R.C.
(2001). Role of AMPA receptor endocytosis in synaptic plasticity.
Nat. Rev. Neurosci. 2, 315–324.
Chapman, P.F., White, G.L., Jones, M.W., Cooper-Blacketer, D.,
Marshall, V.J., Irizarry, M., Younkin, L., Good, M.A., Bliss, T.V., Hy-
man, B.T., et al. (1999). Impaired synaptic plasticity and learning in
Hsiao, K., Chapman, P., Nilsen, S., Eckman, C., Harigaya, Y., Youn-
kin, S., Yang, F., and Cole, G. (1996). Correlative memory deficits,
Abeta elevation, and amyloid plaques in transgenic mice. Science
Iwata,H., Tomita,T.,Maruyama, K.,andIwatsubo,T. (2001).Subcel-
lular compartment and molecular subdomain of beta-amyloid pre-
cursor protein relevant to the Abeta 42-promoting effects of Alzhei-
mer mutant presenilin 2. J. Biol.Chem. 276, 21678–21685.
Jarrett, J.T., Berger, E.P., and Lansbury, P.T., Jr. (1993). The C-ter-
minus of the beta protein is critical in amyloidogenesis. Ann. N Y
Acad. Sci. 695, 144–148.
Johnstone, E.M., Chaney, M.O., Norris, F.H., Pascual, R., and Little,
species polymerase chain reaction analysis. Brain Res. Mol. Brain
Res. 10, 299–305.
Kamal, A., Almenar-Queralt, A., LeBlanc, J.F., Roberts, E.A., and
Goldstein, L.S. (2001). Kinesin-mediated axonal transport of a mem-
brane compartment containing beta-secretase and presenilin-1 re-
quires APP. Nature 414, 643–648.
Kim, J.H., Anwyl, R., Suh, Y.H., Djamgoz, M.B., and Rowan, M.J.
amyloid precursor protein on synaptic plasticity in rat hippocampus
in vivo. J. Neurosci. 21, 1327–1333.
Lambert, M.P., Barlow, A.K., Chromy, B.A., Edwards, C., Freed, R.,
Liosatos, M., Morgan, T.E., Rozovsky, I., Trommer, B., Viola, K.L.,
et al. (1998). Diffusible, nonfibrillar ligands derived from Abeta1-42
are potent central nervous system neurotoxins. Proc. Natl. Acad.
Sci. USA 95, 6448–6453.
Larson, J., Lynch, G., Games, D., and Seubert, P. (1999). Alterations
in synaptic transmission and long-term potentiation in hippocampal
slices from young and aged PDAPP mice. Brain Res. 840, 23–35.
Li, Y.M., Xu, M., Lai, M.T., Huang, Q., Castro, J.L., DiMuzio-Mower,
J., Harrison, T., Lellis, C., Nadin, A., Neduvelil, J.G., et al. (2000).
Photoactivated gamma-secretase inhibitors directed to the active
site covalently label presenilin 1. Nature 405, 689–694.
Luo, Y., Bolon, B., Kahn, S., Bennett, B.D., Babu-Khan, S., Denis,
P., Fan, W., Kha, H., Zhang, J., Gong, Y., et al. (2001). Mice deficient
in BACE1, the Alzheimer’s beta-secretase, have normal phenotype
and abolished beta-amyloid generation. Nat. Neurosci. 4, 231–232.
Malinow, R., Hayashi, Y., Maletic-Savatic, M., Zaman, S., Poncer,
J.-C., Shi, S.-H., and Esteban, J. (1999). Introduction of Green Fluo-
rescent Protein into Hippocampal Neurons through Viral Infection
(Cold Spring Harbor, New York: Cold Spring Harbor Press).
Malinow, R., Mainen, Z.F., and Hayashi, Y. (2000). LTP mechanisms:
from silence to four-lane traffic. Curr. Opin. Neurobiol. 10, 352–357.
Morgan, D., Diamond, D.M., Gottschall, P.E., Ugen, K.E., Dickey, C.,
Hardy, J., Duff, K., Jantzen, P., DiCarlo, G., Wilcock, D., et al. (2000).
of Alzheimer’s disease. Nature 408, 982–985.
Nitsch, R.M., Farber, S.A., Growdon, J.H., and Wurtman, R.J. (1993).
Release of amyloid beta-protein precursor derivatives by electrical
depolarization of rat hippocampal slices. Proc. Natl. Acad. Sci. USA.
Perez, R.G., Soriano, S., Hayes, J.D., Ostaszewski, B., Xia, W.,
Selkoe, D.J.,Chen, X.,Stokin, G.B., andKoo, E.H.(1999). Mutagene-
tosis, turnover, and the generation of secreted fragments, including
Abeta42. J. Biol. Chem. 274, 18851–18856.
Price, D.L., and Sisodia, S.S. (1998). Mutant genes in familial Alzhei-
mer’s disease and transgenic models. Annu. Rev. Neurosci. 21,
Reisberg, B., Windscheif, U., Ferris, S.H., Hingorani, V.N., Stoeffler,
A., and Moebius, H. (2000). Memantine in Moderately Severe to
Severe Alzheimer’s Disease: Results of a Placebo Controlled
6-Month Trial. In World Alzheimer’s Congress (Washington D.C.).
Sabo, S.L., Ikin, A.F., Buxbaum, J.D., and Greengard, P. (2001).
The Alzheimer amyloid precursor protein (APP) and FE65, an APP-
Sastre, M., Turner, R.S., and Levy, E. (1998). X11 interaction with
beta-amyloid precursor protein modulates its cellular stabilization
and reduces amyloid beta-protein secretion. J. Biol. Chem. 273,
lular interactions in heterosynaptic long-term depression. Nature
Schlesinger, S., and Dubensky, T.W. (1999). Alphavirus vectors for
gene expression and vaccines. Curr. Opin. Biotechnol. 10, 434–439.
Selkoe, D.J. (2000). Toward a comprehensive theory for Alzheimer’s
disease. Hypothesis: Alzheimer’s disease is caused by the cerebral
accumulation and cytotoxicity of amyloid beta-protein. Ann. N Y
Acad. Sci. 924, 17–25.
Selkoe DJ. (2002). Alzheimer’s diseaseis a synaptic failure. Science.
Seubert, P., Vigo-Pelfrey, C., Esch, F., Lee, M., Dovey, H., Davis, D.,
Sinha, S., Schlossmacher, M., Whaley, J., Swindlehurst, C., et al.
(1992). Isolation and quantification of soluble Alzheimer’s beta-pep-
tide from biological fluids. Nature 359, 325–327.
Shi, S.H., Hayashi, Y., Petralia, R.S., Zaman, S.H., Wenthold, R.J.,
Svoboda,K., andMalinow, R.(1999).Rapid spinedelivery andredis-
tribution of AMPA receptors after synaptic NMDA receptor activa-
tion. Science 284, 1811–1816.
Sisodia, S.S., Koo, E.H., Hoffman, P.N., Perry, G., and Price, D.L.
(1993). Identification and transport of full-length amyloid precursor
proteins in rat peripheral nervous system. J. Neurosci. 13, 3136–
Steinbach, J.P., Muller, U., Leist, M., Li, Z.W., Nicotera, P., and
Aguzzi, A. (1998). Hypersensitivity to seizures in beta-amyloid pre-
cursor protein deficient mice. Cell Death Differ. 5, 858–866.
Stephan, A., Laroche, S., and Davis, S. (2001). Generation of aggre-
gated beta-amyloid in the rat hippocampus impairs synaptic trans-
mission and plasticity and causes memory deficits. J. Neurosci. 21,
Stoppini, L., Buchs, P.A., and Muller, D. (1991). A simple method
for organotypic cultures of nervous tissue. J. Neurosci. Methods
Suzuki,N., Cheung,T.T.,Cai,X.D., Odaka,A.,Otvos,L., Jr.,Eckman,
long amyloid beta protein secreted by familial amyloid beta protein
precursor (beta APP717) mutants. Science 264, 1336–1340.
Tomita, T., Maruyama, K., Saido, T.C., Kume, H., Shinozaki, K., To-
kuhiro, S., Capell, A., Walter, J.,Grunberg, J., Haass, C., et al. (1997).
(Volga German families) increases the secretion of amyloid beta
protein ending at the 42nd (or 43rd) residue. Proc. Natl. Acad. Sci.
USA 94, 2025–2030.
Turrigiano, G.G., Leslie, K.R., Desai, N.S., Rutherford, L.C., and Nel-
son, S.B. (1998). Activity-dependent scaling of quantal amplitude in
neocortical neurons. Nature 391, 892–896.
Lomakin, A., Benedek, G.B., Selkoe, D.J., and Teplow, D.B. (1999).
Amyloid beta-protein fibrillogenesis. Structure and biological activ-
ity of protofibrillar intermediates. J. Biol. Chem. 274, 25945–25952.
Walsh, D.M., Klyubin, I., Fadeeva, J.V., Cullen, W.K., Anwyl, R.,
Wolfe, M.S., Rowan, M.J., Selkoe, D.J. (2002). Naturally secreted
term potentiation in vivo. Nature 416, 535–539.
Wasco, W., Gurubhagavatula, S., Paradis, M.D., Romano, D.M., Si-
sodia, S.S., Hyman, B.T., Neve, R.L., and Tanzi, R.E. (1993). Isolation
and characterization of APLP2 encoding a homologue of the Alzhei-
mer’s associated amyloid beta protein precursor. Nat. Genet. 5,
Westerman, M.A., Cooper-Blacketer, D., Mariash, A., Kotilinek, L.,
Kawarabayashi, T., Younkin, L.H., Carlson, G.A., Younkin, S.G., and
Ashe, K.H. (2002). The relationship between Abeta and memory in
the Tg2576 mouse model of Alzheimer’s disease. J. Neurosci. 22,
Winblad, B., and Poritis, N. (1999). Memantine in severe dementia:
APP Processing and Synaptic Function
results of the 9M-Best Study (Benefit and efficacy in severely de-
mented patients during treatment with memantine). Int. J. Geriatr.
Psychiatry 14, 135–146.
Yankner, B.A., Dawes, L.R., Fisher, S., Villa-Komaroff, L., Oster-
Granite, M.L., and Neve, R.L. (1989). Neurotoxicity of a fragment of
the amyloidprecursor associated with Alzheimer’sdisease. Science
Yankner, B.A., Duffy, L.K., and Kirschner, D.A. (1990). Neurotrophic
and neurotoxic effects of amyloid beta protein: reversal by tachyki-
nin neuropeptides. Science 250, 279–282.
Yu, G., Nishimura, M., Arawaka, S., Levitan, D., Zhang, L., Tandon,
A., Song, Y.Q., Rogaeva, E., Chen, F., Kawarai, T., et al. (2000).
Nicastrin modulates presenilin-mediated notch/glp-1 signal trans-
duction and betaAPP processing. Nature 407, 48–54.
Zheng, H., Jiang, M., Trumbauer, M.E., Hopkins, R., Sirinathsinghji,
D.J., Stevens, K.A., Conner, M.W., Slunt, H.H., Sisodia, S.S., Chen,
H.Y., et al. (1996). Mice deficient for the amyloid precursor protein
gene. Ann. N Y Acad. Sci. 777, 421–426.
Zhu, J.J.,Esteban, J.A.,Hayashi, Y.,and Malinow,R. (2000).Postna-
tal synaptic potentiation: delivery of GluR4-containing AMPA recep-
tors by spontaneous activity. Nat. Neurosci. 3, 1098–1106.
Zoghbi, H.Y., Gage, F.H., and Choi, D.W. (2000). Neurobiology of
disease. Curr. Opin. Neurobiol. 10, 655–660.