|
|
|
|||||||
|
|||||||||
(Received for publication, October 17,
1994; and in revised form, December 14, 1994) From the
The Alzheimer amyloid precursor protein (APP) undergoes complex
processing resulting in the production of a 4-kDa amyloid peptide
(A Alzheimer's disease, is defined by specific pathological
lesions in the brain which include neurofibrillary tangles and
Recent studies strongly
suggest that amyloid deposition is linked to the pathogenesis of
Alzheimer's disease(7, 8, 9) . The
A In
recent years, there has been an intense effort to elucidate the
pathways whereby A It has recently been recognized that A The pathways involved in APP processing under non-amyloidogenic
conditions have begun to be studied. Using human embryonic kidney cell
lines transfected with the genes for the human brain muscarinic
acetylcholine receptors, Nitsch et al. have shown that
stimulation of the m1 and m3 receptor subtypes with carbachol increases
the release of APP derivatives(22) . Buxbaum et al.(23) have also demonstrated that cholinergic agonists
stimulates APP secretion in human glioma and neuroblastoma cells as
well as in PC12 cells transfected with the m1 receptor. The biochemical
pathways underlying muscarinic stimulation of APP secretion are not
well understood, although protein kinase C activation has been
implicated (24, 25, 26) . Recently, it has
been shown that A Human neurons express mainly the APP NT2N neurons are derived
from a human teratocarcinoma cell line, Ntera 2/c1.D1 (NT2), that is
induced by treatment with retinoic acid to commit irreversibly to a
neuronal phenotype(29, 30) . Prior extensive studies
have shown that NT2N neurons express many cytoskeletal markers,
cell-surface antigens, and synaptic proteins typical of central nervous
system neurons(29) . They are permanently postmitotic and
develop functional dendrites and axons. They can be purified to yield
>99% pure postmitotic human neurons. The rationale for using NT2N
cells to study the regulation of APP-S secretion is that 1) different
cells process APP differently, 2) unlike non-neuronal cells, NT2N
neurons express predominantly APP
Figure 1:
Identification of muscarinic receptor
subtypes in NT2N neurons. Membranes from NT2N neurons (replate 3) were
analyzed for muscarinic receptor subtypes with specific anti-muscarinic
antibodies as described under ``Experimental Procedures.''
Results are expressed as the mean ± S.E. of receptor density
(pmol/mg) from three experiments.
Figure 2:
Identification of phospholipase C isoforms
and GTP-binding protein subtypes in NT2N neurons. A,
phospholipase C isoforms. NT2 (lanes 1 and 3) and
NT2N neurons (lanes 2 and 4) lysates were purified by
SDS-PAGE, and immunoblotting was performed with specific monoclonal
antibodies. Lanes 1 and 2,
Figure 3:
Radiochromatogram of carbachol-induced DAG
accumulation in NT2N neurons. NT2N neurons were labeled with
[
Figure 4:
Time
course of carbachol-induced diacylglycerol accumulation in NT2N
neurons. NT2N neurons were labeled with
[
Since carbachol-induced
accumulation of DAG may reflect activation of other pathways in
addition to phosphatidylinositol-specific phospholipase C, we also
measured the muscarinic-induced accumulation of the second messenger
Ins(1,4,5)P Ins(1,4,5)P
Figure 5:
Carbachol stimulation of intracellular
Ca
Figure 6:
Time course of APP-S secretion from NT2N
neurons. NT2N neurons were pulse-labeled 30 min with
[
In order to assess A
Figure 7:
Time course of carbachol on A
We have shown that NT2N neurons express m2 and m3 muscarinic
receptors, and upon muscarinic stimulation of normal human NT2N neurons
there is: 1) activation of phospholipase C with release of the second
messengers Ins(1,4,5)P The signal transduction pathway involved in
muscarinic-induced APP-S secretion has several components. The
muscarinic acetylcholine receptor family consists of five cloned and
expressed receptor genes designated m1 through m5 and is part of the
large family of seven transmembrane receptors(51) . These
receptors work by activation of heterotrimeric GTP-binding proteins:
m1, m3, and m5 are coupled to stimulation of phospholipase C, while m2
and m4 inhibit adenylate cyclase(51) . Our study identified the
m3 receptor as the most prominent subtype in NT2N neurons. The m3
subtype is known to be coupled to the heterotrimeric GTP-binding
protein G Among the
various heterotrimeric GTP-binding protein subunits which were
screened, we demonstrated the presence of the G Two
main isoforms ( Constitutive secretion of APP-S was
clearly detected in the supernatant of NT2N neurons consistent with our
previous results which have shown that NT2N neurons mainly secrete
APP A In summary, we have shown that
the muscarinic agonist carbachol stimulates the
muscarinic/phospholipase C signal transduction pathway in normal human
neurons resulting in a transient increase in the second messengers DAG,
Ins(1,4,5)P
Volume 270,
Number 9,
Issue of March 3, 1995 pp. 4916-4922
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
s Disease Amyloid Precursor Protein
Secretion and Amyloid 
-Protein Production in Human Neuronal NT2N
Cells (*)
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
) which has been implicated in the pathogenesis of
Alzheimer's disease. Recent studies have shown that cells can
secrete carboxyl terminus truncated APP derivatives (APP-S) in response
to physiological stimulus. We have used human central nervous system
neurons (NT2N) derived from a teratocarcinoma cell line (NT2) to study
the signal transduction pathways involved in APP-S secretion and A
production. Muscarinic receptors (m2 and m3) as well as the
heterotrimeric GTP-binding protein G
and the 1 isoform
of phospholipase C were present in NT2N neurons. Stimulation of the
muscarinic receptor with carbachol resulted in phospholipase C
activation as shown by a transient increase in the second messengers
1,2-diacyl-sn-glycerol and inositol 1,4,5-trisphosphate.
Carbachol also caused an increase in intracellular Ca
levels measured in single NT2N neurons. Under these conditions,
carbachol caused a time-dependent 2-fold increase in APP-S secretion
into the medium. In contrast, prolonged treatment with carbachol caused
a decrease in A production into the medium. These results suggest
that APP-S secretion and A production in NT2N neurons are
regulated by the muscarinic/phospholipase C signal transduction
pathway. Furthermore, activation of this pathway results in
dissociation of APP-S secretion and A production.
-amyloid (A) (
)deposits affecting selected areas
of the brain(1, 2, 3) . This
neurodegenerative disease is associated with neuron loss and death
which affect many neuronal populations. In particular, the cholinergic
cells which arise in the basal forebrain and terminate in the
hippocampus and cerebral cortex are severely affected(1) . The
selective vulnerability of cholinergic neurons has been the focus of
intensive studies to elucidate its biochemical
mechanism(4, 5) . There is a loss of choline
acetyltransferase, the enzyme that synthesizes acetylcholine from
choline and acetyl-CoA. More recently, it has been shown that the high
affinity choline uptake is abnormally high in Alzheimer's
disease(6) . Thus, these observations suggest that
abnormalities in cholinergic function may be involved in the
pathogenesis of Alzheimer's disease. peptide is derived from amyloid precursor proteins (APP). APP is
an integral membrane, tyrosine-sulfated glycoprotein with one
membrane-spanning domain and an extracytoplasmic NH
terminus. APP exists in three major isoforms in the central
nervous system, which are encoded by the same gene on chromosome 21.
Alternative mRNA splicing generates 695- (APP
), 751-
(APP
), and 770-amino acid APP (APP
). The
brain is the richest source of APP, and in particular APP is restricted almost exclusively to the central nervous system
and the peripheral nervous system(10, 11) . is generated from APP. A peptide is a
39-43 amino acid internal sequence that extends from within the
transmembrane domain into the extracytoplasmic domain of APP. There are
several pathways to process APP. In the constitutive secretory pathway
(or 
-secretase), APP is cleaved, within the A sequence at
residue 687 just outside the transmembrane domain, by the action of a
protease, to a large secreted NH-terminal derivative
(APP-S) and a membrane-associated fragment, neither of which can
produce -amyloid protein(12, 13) . The presence
of a -secretase has also been inferred which cleaves APP precisely
at the amino terminus of A (14) . In a third processing
pathway, APP is processed in the endosomal and lysosomal system, and
yields complex COOH-terminal derivatives, some of which are potentially
amyloidogenic(15, 16) . More recently, several groups
have shown that APP processing in the endosomal/lysosomal system
produces a 4-kDa -amyloid protein that is essentially similar to
the deposited amyloid of Alzheimer's disease (reviewed in Refs.
17, 18). is present in
human cerebrospinal fluid from normal and Alzheimer's disease
patients and that cultures of neuronal and non-neuronal cells
transfected with the APP gene secrete A into the
medium(19, 20) . Primary fetal human neurons secrete
A(20, 21) . Importantly, a unique human neuronal
cell line NT2N has been shown to secrete endogenous A into the
culture medium(11) . Furthermore, intracellular A can be
detected in the NT2N neurons. Collectively, these observations strongly
suggest that normal human neurons can generate the A peptide. production is regulated by the muscarinic
pathway(27) . isoform(11) . In addition to astrocytes, microglia, and
vascular cells, neurons are a likely source of A deposited in
amyloid in Alzheimer's disease: it is therefore important to
study APP expression, processing, and regulation in neurons.
Furthermore, since Alzheimer's disease and related
neurodegenerative diseases only occurs in humans and primates (17, 28) and there is no rodent animal model which
recapitulates the development of this disease, these studies should
ideally be performed in human neurons. Postmitotic mature human
neurons, however, are difficult to isolate and maintain in culture.
Because of these limitations we have used a unique culture model of
human neurons, the NT2N neuronal cells., the major APP isoform
expressed in neurons in brain, 3) in NT2N cells, APP can
be easily detected without transfection of the APP gene, and 4) NT2N
neurons constitutively generate intracellular A peptide and
release it into the culture medium(11) . Thus, NT2N neurons
represent a unique and physiological human model system to study APP
processing and A production which closely recapitulates events in
the normal human brain. In the present study, we have used NT2N neurons
to study muscarinic regulation of APP secretion and A production
and to dissect the signal transduction pathways involved.
Cell Culture
The human teratocarcinoma
NTera2/c1.D1 (NT2) cells were maintained in Opti-MEM (Life
Technologies, Inc.) supplemented with 10% fetal bovine serum and
penicillin/streptomycin. NT2N cells (>99% pure, i.e. replate no. 3 cells) were generated as described
previously(29) . Briefly, cells (2 
10
/75-cm flask) were treated with 10 µM retinoic acid twice a week for 5 weeks. Cells were then replated
(Replate no. 1). Two days later, cells were mechanically dislodged,
replated in culture flasks (replate no. 2), and cultured for an
additional 6-8 days, after which the neuronal cells were again
dislodged with trypsin, and replated on Matrigel (Collaborative
Research)/poly-D-lysine coated coverslips or dishes (replate
no. 3). Cells were generally used within 4 weeks after replate no. 3.Measurement of Diacylglycerol Accumulation in NT2N
Neurons
NT2N neurons were labeled with
[
H]arachidonic acid (31, 32, 33, 34) (1 µCi/100-mm
dish) for 24 h. Each dish was then gently washed five times in a
modified Krebs-HEPES buffer (25 mM HEPES (pH 7.40), 115 mM NaCl, 24 mM NaHCO, 5 mM KCl, 2.5
mM CaCl, 1 mM MgCl, 0.1%
bovine serum albumin, 3 mMD-glucose), preincubated
30 min under an atmosphere of 95% air, 5% CO at 37 °C,
and then incubated for 0-30 min with fresh Krebs-HEPES buffer
± 1 mM carbachol. At the end of the incubation, the
medium was removed, and the cells were quenched with ice-cold methanol.
Prior to extraction, carrier amounts (5 µg) of phosphatidylcholine,
monoolein, diolein, and arachidonic acid were added to each tube to aid
in recovery, followed by 1 ml of chloroform and 1 ml of water. Tubes
were vortexed (1 min), sonicated (30 min, 4 °C), and vortexed (1
min). Tubes were centrifuged in a refrigerated table-top centrifuge (15
min, 4 °C, 800 g). The lower organic phase was
transferred with a silanized Pasteur pipette to a clean silanized 13
100-mm conical borosilicate tube. The remaining aqueous upper
phase was re-extracted twice with chloroform (1 ml) and the extracts
combined with the previous organic phase. The organic phase was washed
twice with water (3 ml), concentrated twice under Nitrogen in a
TurboVap evaporator (Zymark, Hopkinton, MA), and reconstituted in 25
µl of chloroform. With this extraction procedure, recovery of
[H]arachidonic acid was
>95%(33, 35) . Samples were spotted onto the
preadsorbent zone of channeled Silica Gel G TLC 20 20-cm plates
(Analtech, Newark, DE) which had been activated 30 min at 110 °C.
Plates were developed for 30-45 min in petroleum ether
(30-60 °C)/diethyl ether/acetic acid (140:60:2,
v/v/v)(36) . The radioactivity of the chromatogram was
quantitated with a Berthold Linear Analyzer 284 (Wallac Inc.,
Gaithersburg, MD) equipped with a position-sensitive proportional high
resolution counter tube (200 mm long, 1380 V) continuously flushed (0.5
liter/min) with P10 gas (90% argon, 10% methane) and a 4-mm entrance
window. Each TLC lane was scanned simultaneously in its entirety (20
cm) for 60 min. The instrument detected radioactive peaks as small as
50-100 disintegrations/min (area under the curve) with a
resolution of 0.5 mm. Data analysis was performed using version 7.19 of
the Berthold 1D-TLC software. Radioactive peaks corresponding to
diacylglycerol and phospholipids were integrated. Peak identity was
assigned by comparison with iodine-stained cold standards and
radiolabeled commercial [H]arachidonic acid.
Typically, the following R
were obtained:
phospholipids (0), monoacylglycerol (0.19), diacylglycerol (0.45),
arachidonic acid (0.63), triacylglycerol (0.88). Radioactivity in the
diacylglycerol peak was normalized to the total radioactivity
incorporated into the phospholipid
fraction(31, 32, 33, 34) .Inositol Phosphates Measurements
NT2N neurons were
labeled with [H]inositol (10 µCi/dish) for 48
h. [H]Inositol-labeled neurons were washed,
preincubated, and incubated with agonists as described above. At the
end of the incubation, the medium was removed, and 0.5 ml of ice-cold
trichloroacetic acid was added to each dish. Cells were scraped,
transferred to a dry-ice/ethanol bath (15 min), vortexed (1 min),
sonicated (30 min), vortexed (1 min), and centrifuged (2,000 g, 15 min, +4
°C)(37, 38, 39) . The supernatant was
extracted with diethyl ether (5 3 ml), adjusted to pH 7.0 and
lyophilized, and reconstituted in 350 µl of 1 mM EDTA pH
7.0 (supplemented with 10 µg of AMP, 38 µg of ADP, 50 µg of
ATP) prior to strong anion-exchange (SAX) HPLC analysis. The HPLC
system consisted of a Varian 9095 injector, Varian 9010 pump, Varian
9050 detector, Whatman Partisil 10 SAX cartridge, Partisil SAX guard
column, and a Whatman Solvecon precolumn. The solvent program (1
ml/min) was: gradient from 0 to 2 min of 20% solvent B (2.5 M NaHPO
, pH 3.8 with NaOH), 20% B at 18 min,
28% B at 20 min, 28% B at 28 min, 100% B at 30 min and maintained for
10 min, 100% solvent A (HO) at 42 min and maintained for 8
min. One-ml fractions were collected every minute and counted in a
liquid scintillation spectrometer following the addition of 4 ml of
In-Flow BD scintillation mixture (In-US Systems Inc, Fairfield, NJ).
Peak identity was assigned by comparison with commercial radiolabeled
standards and concomitant UV monitoring of AMP, ADP, and
ATP(38, 39) .Identification of Muscarinic Receptors
Muscarinic
receptor subtypes were identified by immunoprecipitation of
[H]quinuclidinyl benzilate-labeled receptors
expressed by NT2N neurons using specific antibodies and receptor
density calculated as described
previously(40, 41, 42, 43) .Measurement of Cytosolic Free
Ca
NT2N neurons were plated on
poly-L-lysine-coated round glass coverslips and bathed in
Hanks' buffer containing 2.6 mM Ca.
Neurons were loaded with 5 µM fura 2AM for 45 min before
use, and then mounted in a temperature-controlled dish on the stage of
a NIKON Diaphot inverted microscope equipped for epifluorescence.
Cytosolic free Ca in a single neuron was calculated
from the ratio of fura 2 fluorescence at 340 and 380 nm as previously
published(44) .Metabolic Labeling and Immunoprecipitation
NT2N
cells were metabolically labeled (30 min for APP-S secretion in
modified Krebs-HEPES medium, 3 h for A production in
Dulbecco's modified Eagle's medium) with
[
S]methionine (100 µCi/ml), washed, and then
stimulated with carbachol (1 mM) as described above. Following
stimulation with carbachol (15-90 min for APP-S secretion,
1-8 h for A production), the supernatant was removed and
processed for immunoprecipitation with a polyclonal anti-APP-S antibody
(KAREN, anti-NH-terminal APP, a kind gift of Dr. Barry
Greenberg) or with monoclonal anti-A antibody (4G8, Institute for
Basic Research, Staten Island, NY) as described
previously(11) . In brief, the conditioned medium from the
neurons was centrifuged for 30 min at 100,000 g, and
proteins in the supernatant were precipitated with equal amounts of
saturated ammonium sulfate at +4 °C overnight. After a high
speed spin, the pellet was resuspended in 1 RIPA (11, 15) buffer followed by immunoprecipitation with
specific antibodies(11) . Immunoprecipitates were then
separated on SDS/7.5% PAGE or 16.5% Tris/Tricine gels, and
radioactivity in the APP-S 90-kDa protein or 4-kDa A peptide band
was calculated after exposure of the gel to a PhosphorImager plate
(Molecular Dynamics) and analysis with the ImageQuant software. Within
each experiment, results were normalized to the condition with the
highest amount of radioactivity (90 min with carbachol for APP-S, 8 h
control for A) and are expressed as a ratio to that condition.Other Methods
Detection of phospholipase C
isoforms and GTP-binding protein subtypes was performed using standard
Western blotting techniques. In brief, NT2N neurons were homogenized in
Laemmli's buffer. SDS-PAGE (7.5-15% polyacrylamide
depending on the protein of interest) was performed on a Bio-Rad
mini-PROTEAN II gel apparatus using standard techniques and rainbow
molecular weight markers. Proteins were then transferred to
nitrocellulose and probed with the appropriate antibody. Detection of
the antibody-labeled protein was typically performed with 
I-protein A or rabbit anti-mouse IgG followed by I-protein A (for monoclonal antibodies). Detection and
quantitation of the bands of interest were performed on a Molecular
Dynamics PhosphorImager. Phospholipase C monoclonal antibodies directed
to the 1, 
1, 2, and 
isoforms were obtained
commercially (UBI, Lake Placid, NY). Antibodies to the GTP-binding
protein subunits were generated as described
previously(45, 46, 47, 48, 49) .
Selected antibodies used in this study include antibody 1190 which
recognizes s, 0116 against all is, 1521 against i2, 3646
against i1, 3642 against i3, 2921 against z, 9072
against o1 and o2, 0121 against 12, 0120 against
13, 946 against q #0130 against 16, and 5357 against
.Data Analysis
Results are expressed as the mean
± S.E. Statistical analysis was performed using version 5.0 of
SSPS for Windows (SSPS Inc., Chicago, IL) or SigmaStat for Windows
(Jandel, San Rafael, CA). Data were analyzed by one-way or two-way
analysis of variance followed by multiple comparisons between means
using the Least Significant Difference test or the Student-Newman-Keuls
method. A probability of p < 0.05 was considered
statistically significant.
Identification of Muscarinic Receptor Subtypes in NT2N
Neurons
NT2N cells are pure, postmitotic, differentiated human
neurons. They represent a unique model of human neurons in culture with
functional dendrites and axons. As shown in Fig. 1, they also
express muscarinic receptors. The most abundant subtype in NT2N neurons
was m3 (0.113 ± 0.032 pmol/mg protein) which is known to be
coupled to the phospholipase C signal transduction pathway, followed by
the m2 subtype (0.072 ± 0.016 pmol/mg protein).
Phospholipase C Signal Transduction Components in NT2N
Neurons
Two major isoforms of phospholipase C were detected by
immunoblotting in NT2N neurons: 1 and 1 (Fig. 2A). No significant amounts of the 2 and
isoforms were observed. As expected, phospholipase C activity was
detected enzymatically in NT2N cell homogenate using
[H]phosphatidylinositol as a substrate (data not
shown) (50) . Since it has been shown in other cell types that
the muscarinic receptor m3 is coupled to the 1 isoform of
phospholipase C via G, a GTP-binding protein, we used
a panel of monoclonal antibodies to -subunits of heterotrimeric
GTP-binding proteins to identify their presence in NT2N neurons. As
shown in Fig. 2B, - and
-subunits were identified as well as 
and
, but not any of the other -subunits tested for
(i, z, 12, and 16). Thus, these results indicate
that all the components of the muscarinic/phospholipase C signal
transduction pathway are present in NT2N neurons.
1 isoform; lanes
3 and 4, 1 isoform. B, GTP-binding protein
subtypes. The arrow indicates the 43-kDa
marker.
Carbachol Activates Phospholipase C in NT2N
Neurons
In order to assess phospholipase C activation, NT2N
neurons were labeled with [H]arachidonic acid for
24 h, washed in a modified Krebs-HEPES buffer, preincubated for 30 min
at 37 °C, and then stimulated for 0-30 min with carbachol (1
mM). Neutral lipids including DAG were extracted and purified
by 1D-TLC, and quantitated with a sensitive linear analyzer. Fig. 3shows a typical radiochromatogram from these experiments.
Under non-stimulatory conditions, DAG levels did not change over time (Fig. 4). The addition of carbachol, however, caused a very
rapid and significant increase in DAG levels which peaked at 2 min
(0.49 ± 0.06% of total phospholipid versus 0.25
± 0.02% for controls, p < 0.05), decreasing slightly
at 5 min (0.41 ± 0.03% of total phospholipid versus 0.25 ± 0.02% for controls, p < 0.05), followed
by a gradual decrease over time.
H]arachidonic acid for 24 h (1 µCi/100 mm
dish), washed in modified Krebs-HEPES buffer (25 mM HEPES pH
7.40, 115 mM NaCl, 24 mM NaHCO, 5 mM KCl, 2.5 mM CaCl, 1 mM MgCl, 0.1% bovine serum albumin, 3 mMD-glucose), preincubated 30 min, and then incubated 0 to
30 min with medium ± 1 mM carbachol at 37 °C under
95% air, 5% CO. Diacylglycerol (DAG) was
extracted, analyzed by TLC, and quantitated with a Berthold linear
analyzer. Equal amounts of radioactivity were loaded onto each
lane.
H]arachidonic as in Fig. 3.
Diacylglycerol accumulation was normalized to percent of label
incorporated into the total phospholipid fraction. Results are shown as
the mean ± S.E. for control (solid squares, dashed
line) and 1 mM carbachol (solid circles, solid
line) from 4 to 16
observations/condition.
which directly reflects phospholipase C
activation. NT2N neurons were labeled with
[H]inositol and then stimulated for 2 and 5 min.
Inositol phosphates were extracted and separated by SAX-HPLC. Under
these conditions, carbachol caused a 2.7-fold increase in
Ins(1,4,5)P levels at 2 min (from 18.9 ± 3.4
counts/min to 50.9 ± 17.0 counts/inm, n = 3). No
significant increase was noted at 5 min. Finally, carbachol had no
effect on the levels of the inactive isomer Ins(1,3,4)P or
on Ins(1,3,4,5)P (data not shown). mobilizes Ca from intracellular stores. Since
carbachol activates phospholipase C with release of the second
messengers DAG and Ins(1,4,5)P, we next examined whether
carbachol could affect intracellular Ca levels in
NT2N neurons. In these experiments, NT2N neurons on coverslips were
loaded with fura 2 and then mounted on a microscope equipped for
epifluorescence. Single neuron cytosolic Ca was
calculated from the ratio of fura 2 fluorescence at 340 and 380 nm.
Transient stimulation with carbachol caused a very rapid and
significant increase in intracellular Ca levels (Fig. 5).
in single NT2N neurons. NT2N neurons were loaded
with fura 2 and Ca fluorescence from a single neuron
was quantitated as described under ``Experimental
Procedures.'' The addition of carbachol is indicated by the arrow. Representative of at least three
experiments.
Carbachol Stimulates APP-S Secretion But Decreases A
Since we had shown that carbachol
stimulates the phospholipase C signal transduction pathway in NT2N
neurons, we then examined whether secretion of APP-S was under
muscarinic control. In these experiments, NT2N neurons were
metabolically labeled with [
Production in NT2N NeuronsS]methionine and
then stimulated with carbachol. The supernatant was collected,
immunoprecipitated with an anti-APP antibody (KAREN which recognizes
APP-S), and analyzed by SDS-PAGE. As shown in Fig. 6(top
panel), a prominent 90 kDa band was detected in NT2N lysate and in
the supernatant. Under non-stimulatory conditions (Fig. 6, bottom panel), there was a time-dependent accumulation of
APP-S from 0.007 ± 0.003 relative units at time 0 to 0.438
± 0.062 at 90 min (p < 0.05). The addition of
carbachol caused an increase in APP-S which was highly significant
(0.175 ± 0.027 at 60 min, and 1.000 at 90 min, p <
0.05 versus control).
S]methionine, washed, and then chased with 1
mM carbachol in modified Krebs-HEPES medium as described under
``Experimental Procedures.'' APP-S secretion into the
supernatant was measured after ammonium sulfate precipitation and
immunoprecipitation with the anti-APP KAREN antibody. Proteins were
separated by SDS/7.5% PAGE and analyzed with a PhosphorImager. Top
panel, representative gel of APP-S secretion. A and B, lysate; C and D, supernatant. The time
(0, 60, and 90 min) is indicated on the x axis. Bottom
panel, time course of APP-S secretion into the supernatant. APP-S
secretion was quantitated by PhosphorImager as the amount of
radioactivity in the 90 kDa band and is expressed as the ratio to the
condition within each experiment with the highest counts (90 min,
carbachol). Results are shown as the mean ± S.E. of APP-S
secretion from six separate experiments. Control, hatched
bars; carbachol, solid bars.
production,
slightly different labeling conditions were used. NT2N neurons were
labeled for 3 h with [S]methionine, and then
stimulated with carbachol. As shown in Fig. 7(top
panel), a 4 kDa peptide band was detected. In most experiments,
the related 3-kDa peptide was not detected. There was a significant and
time-dependent accumulation of A peptide over 8 h. Under 1 h,
A levels were too low to quantitate. Stimulation with carbachol
caused a decrease in A levels which represented a 42% decrease by
8 h (0.588 ± 0.079 relative units versus 1.000 for
control, p < 0.05).
production from NT2N neurons. NT2N neurons were pulse-labeled 3 h with
[S]methionine, washed, and then chased with 1
mM carbachol in modified Krebs-HEPES medium as described under
``Experimental Procedures.'' A production into the
supernatant was measured after ammonium sulfate precipitation and
immunoprecipitation with the anti-A 4G8 antibody. Proteins were
separated on 16.5% Tris-Tricine gels and analyzed with a
PhosphorImager. Top panel, representative gels of A
production. Bottom panel, time course of A production
into the supernatant. A secretion was quantitated by
PhosphorImager as the amount of radioactivity in the 4 kDa band and is
expressed as the ratio to the condition within each experiment with the
highest counts (8 h, control). Results are shown as the mean ±
S.E. of A production from three to four separate experiments.
Control, hatched bars; carbachol, solid
bars.
and DAG, 2) increased intracellular
Ca levels, and 3) time-dependent secretion of APP-S
associated with decreased A production. These studies represent
the first demonstration in non-transfected human neurons of muscarinic
regulation of APP-S secretion and A production, and extend
previous studies which have shown muscarinic regulation of A
production(27) . in other systems(52, 53) .
Interestingly, we did not find any significant levels of the m1
muscarinic receptor subtype which has been implicated in APP secretion
in human embryonic kidney cell lines transfected with the genes for the
m1 subtype(22) . Although significant levels of m2 receptors
were also found in NT2N cells, this and previous studies implicate
activation of the m1/m3-coupled phospholipase C signal transduction
pathway in APP-S secretion(22, 23) .- and
-subunits. There are four members (G,
G
G
, and G
)
of the G class of -subunits(54) . There is
overwhelming evidence demonstrating that G, a 42-kDa
protein, directly regulates phospholipase C-(54) . Thus,
purified bovine brain phospholipase C- was shown to be markedly
stimulated by brain G(55) . The identification
of G in NT2N neurons is an important link in the
muscarinic receptor/phospholipase C signal transduction cascade. The
presence of the -subunit also may have functional
implications since it has recently been shown that
stimulates various isoforms of phospholipase C-, although it
stimulates more the 3 isoform than the 1
isoform(56) . was also identified in NT2N
neurons. In most cells, the -subunit, a pertussis
toxin-sensitive G protein, is thought to inhibit Ca channel activity(57) . Recently, however, it has been
proposed that APP itself may function as a membrane receptor coupled to
G(58) . The cytoplasmic APP sequence
His
-Lys
had a specific G activating function and was necessary to form a APP
G complex. Although this observation was demonstrated in
vitro, its physiological significance and role in the pathogenesis
of Alzheimer's disease are unclear at the present time.1, 1) of phospholipase C were present in NT2N
neurons. In other cells, activation of phospholipase C-1 is
typically the result of growth factor occupancy of the growth factor
receptor and autophosphorylation by a receptor-tyrosine
kinase(59) . Activation by epidermal growth factor and
platelet-derived growth factor causes translocation of phospholipase C
to the membrane(60) . Whether these growth factors have any
role in regulating APP secretion remains to be determined. However,
since activation of phospholipase C-1 results in hydrolysis of
polyphosphoinositides and accumulation of the second messengers
Ins(1,4,5)P and DAG, it is also conceivable that these
agonists will regulate APP secretion. Phospholipase C-1 is the
phospholipase C isoform which is known to be regulated by
heterotrimeric GTP-binding proteins(61) . In particular, the
muscarinic receptor, m3, is most likely an activator of phospholipase
C-1(62) . Based on post-mortem studies, brain
phosphoinositide metabolism appears abnormal in Alzheimer's
disease as reflected by decreased levels of phosphoinositides and
aberrant accumulation of phospholipase C- in the temporal cortex
and hippocampus(63) . Indeed, it has been postulated that these
abnormalities may be related to the characteristic cellular pathology
of Alzheimer's disease(63) , although there are
well-established changes in other classes of phospholipids such as
phosphatidylcholine (64) . Since activation of phospholipase C
stimulates APP secretion and decreases A production, this signal
transduction pathway may be a relevant target for the development of
Alzheimer's disease.-S(11) . Muscarinic stimulation of NT2N
neurons results in regulated APP-S secretion which is measured over the
background of constitutive secretion, and which most likely reflects
activation of the putative -secretase. These studies are important
since they directly demonstrate for the first time that activation of
the muscarinic/phospholipase C signal transduction pathway results in
APP-S secretion in normal human neurons with normal levels of
muscarinic receptors and endogenous levels of APP as opposed to
non-neuronal or neuronal cell lines over-expressing muscarinic
receptors and transfected with the APP gene(22, 23) .
DAG, the product of phospholipase C hydrolysis of polyphosphoinositides
is an endogenous activator of protein kinase C, a Ca-
and phospholipid-dependent protein kinase which has been implicated in
regulated APP secretion. Phorbol esters have been used as pharmacologic
probes to activate protein kinase C and APP secretion in cells
overexpressing the APP gene such as PC12 cells, Chinese hamster ovary
cells, 293 cells, human umbilical vein endothelial cells, and COS cells (23, 24, 27, 65) . In one study with
Swiss 3T3 fibroblasts overexpressing protein kinase C, the
specific isoform of protein kinase C which mediates phorbol
ester-induced APP secretion was identified as protein kinase C,
although the contribution of other isoforms of protein kinase C in APP
secretion cannot be excluded, specially as neurons are known to express
several isoforms(25, 66) . Ins(1,4,5)P, a
second messenger generated from phospholipase C hydrolysis of
phosphatidylinositol 4,5-bisphosphate, mobilizes calcium from
intracellular stores resulting in increased cytoplasmic Ca levels(67) . Recently, it has been suggested that
increases in intracellular Ca levels in Chinese
hamster ovary cells transfected with cDNA encoding the m1 or m3
muscarinic receptor and APP results in APP secretion
independently of protein kinase C(26) . Our study demonstrates
that the muscarinic-induced transient increase in DAG,
Ins(1,4,5)P, and Ca levels in NT2N
neurons correlates with APP-S secretion. Of note is the fact that
muscarinic-induced increase in these second messengers is transient
(2-5 min) whereas the increase in APP-S secretion is observed
over 90 min, suggesting that intermediate steps, such as protein
phosphorylation and/or
synthesis(24, 26, 27, 65) , are
distally involved in the regulated secretion of APP-S.
production from NT2N neurons was decreased following cholinergic
stimulation. Thus, activation of the muscarinic/phospholipase C pathway
results in opposite effects on APP-S secretion and A production.
Similar results were observed in a variety of cells transfected with
the gene for APP or
APP(24, 26, 27, 65) .
However, one recent study showed that in the human neuroblastoma cell
line SY5Y transfected with cDNA encoding APP, phorbol
esters caused an increase in A production(68) . Our
studies, performed in human neurons expressing endogenous levels of
APP, strongly suggest that the pathways of APP-S secretion and A
production can be dissociated following stimulation of the
muscarinic/phospholipase C signal transduction pathway. Interestingly,
muscarinic-induced decrease in A production was only observed
after prolonged stimulation (8 h) with carbachol which suggest that APP
levels have to be substantially depleted (by increased secretion of
APP-S) before A production is decreased. Alternatively, this may
reflect differences in the release kinetics of APP-S and A and the
difficulty in quantitating low levels of A. Previously, we have
shown that A can be recovered from cell lysates in NT2N neurons (11) . We did not examine the effect of carbachol treatment on
intracellular A because the recovered levels were low and did not
allow for sufficient quantification., Ca, and a sustained increase
in APP-S secretion with a subsequent decrease in A production.
),
-amyloid peptide; APP, Alzheimer amyloid precursor protein; APP-S,
secreted Alzheimer amyloid precursor protein; SAX, strong-anion
exchange; Ins(1,4,5)P, myo-inositol
1,4,5-trisphosphate; DAG, 1,2-diacyl-sn-glycerol; PAGE,
polyacrylamide gel electrophoresis; HPLC, high performance liquid
chromatography; TLC, thin layer chromatography; Tricine, N-tris(hydroxymethyl)methylglycine.
We are very grateful to Dr. Todd Golde for helpful
comments and discussion.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  M. L. Kim, B. Zhang, I. P. Mills, M. E. Milla, K. R. Brunden, and V. M.-Y. Lee Effects of TNF{alpha}-Converting Enzyme Inhibition on Amyloid {beta} Production and APP Processing In Vitro and In Vivo J. Neurosci., November 12, 2008; 28(46): 12052 - 12061. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Grupe, R. Abraham, Y. Li, C. Rowland, P. Hollingworth, A. Morgan, L. Jehu, R. Segurado, D. Stone, E. Schadt, et al. Evidence for novel susceptibility genes for late-onset Alzheimer's disease from a genome-wide association study of putative functional variants Hum. Mol. Genet., April 15, 2007; 16(8): 865 - 873. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.-W. Klafki, M. Staufenbiel, J. Kornhuber, and J. Wiltfang Therapeutic approaches to Alzheimer's disease Brain, November 1, 2006; 129(11): 2840 - 2855. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Camden, A. M. Schrader, R. E. Camden, F. A. Gonzalez, L. Erb, C. I. Seye, and G. A. Weisman P2Y2 Nucleotide Receptors Enhance {alpha}-Secretase-dependent Amyloid Precursor Protein Processing J. Biol. Chem., May 13, 2005; 280(19): 18696 - 18702. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Cummings Searching for Methods to Detect, Prevent, and Treat Alzheimer's Disease Am J Psychiatry, April 1, 2005; 162(4): 645 - 647. [Full Text] [PDF]
|
|
|
|
|
|
M. Hashimoto, H. Kazui, K. Matsumoto, Y. Nakano, M. Yasuda, and E. Mori Does Donepezil Treatment Slow the Progression of Hippocampal Atrophy in Patients With Alzheimer's Disease? Am J Psychiatry, April 1, 2005; 162(4): 676 - 682. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Chen, J. Sulcove, I. Frank, S. Jaffer, H. Ozdener, and D. L. Kolson Development of a Human Neuronal Cell Model for Human Immunodeficiency Virus (HIV)-Infected Macrophage-Induced Neurotoxicity: Apoptosis Induced by HIV Type 1 Primary Isolates and Evidence for Involvement of the Bcl-2/Bcl-xL-Sensitive Intrinsic Apoptosis Pathway J. Virol., August 12, 2002; 76(18): 9407 - 9419. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. P. Esler and M. S. Wolfe A Portrait of Alzheimer Secretases--New Features and Familiar Faces Science, August 24, 2001; 293(5534): 1449 - 1454. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Borroni, F. Colciaghi, L. Pastorino, C. Pettenati, E. Cottini, L. Rozzini, R. Monastero, G. L. Lenzi, F. Cattabeni, M. Di Luca, et al. Amyloid Precursor Protein in Platelets of Patients With Alzheimer Disease: Effect of Acetylcholinesterase Inhibitor Treatment Arch Neurol, March 1, 2001; 58(3): 442 - 446. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. R. Neelands, A. P. J. King, and R. L. Macdonald Functional Expression of L-, N-, P/Q-, and R-Type Calcium Channels in the Human NT2-N Cell Line J Neurophysiol, December 1, 2000; 84(6): 2933 - 2944. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. M. Skovronsky, D. B. Moore, M. E. Milla, R. W. Doms, and V. M.-Y. Lee Protein Kinase C-dependent alpha -Secretase Competes with beta -Secretase for Cleavage of Amyloid-beta Precursor Protein in the Trans-Golgi Network J. Biol. Chem., January 28, 2000; 275(4): 2568 - 2575. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. S. Kelner, M. Lee, M. E. Clark, D. Maciejewski, D. McGrath, S. Rabizadeh, T. Lyons, D. Bredesen, P. Jenner, and R. A. Maki The Copper Transport Protein Atox1 Promotes Neuronal Survival J. Biol. Chem., January 7, 2000; 275(1): 580 - 584. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. S. Howland, S. P. Trusko, M. J. Savage, A. G. Reaume, D. M. Lang, J. D. Hirsch, N. Maeda, R. Siman, B. D. Greenberg, R. W. Scott, et al. Modulation of Secreted beta -Amyloid Precursor Protein and Amyloid beta -Peptide in Brain by Cholesterol J. Biol. Chem., June 26, 1998; 273(26): 16576 - 16582. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Jolly-Tornetta, Z.-y. Gao, V. M.-Y. Lee, and B. A. Wolf Regulation of Amyloid Precursor Protein Secretion by Glutamate Receptors in Human Ntera 2 Neurons (NT2N) J. Biol. Chem., May 29, 1998; 273(22): 14015 - 14021. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. C. LeBlanc, M. Koutroumanis, and C. G. Goodyer Protein Kinase C Activation Increases Release of Secreted Amyloid Precursor Protein without Decreasing Abeta Production in Human Primary Neuron Cultures. J. Neurosci., April 15, 1998; 18(8): 2907 - 2913. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y.-L. Guo, K. Baysal, B. Kang, L.-J. Yang, and J. R. Williamson Correlation between Sustained c-Jun N-terminal Protein Kinase Activation and Apoptosis Induced by Tumor Necrosis Factor-alpha in Rat Mesangial Cells J. Biol. Chem., February 13, 1998; 273(7): 4027 - 4034. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Mills, D. Laurent Charest, F. Lam, K. Beyreuther, N. Ida, S. L. Pelech, and P. B. Reiner Regulation of Amyloid Precursor Protein Catabolism Involves the Mitogen-Activated Protein Kinase Signal Transduction Pathway J. Neurosci., December 15, 1997; 17(24): 9415 - 9422. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. H. Ulus and R. J. Wurtman Metabotropic Glutamate Receptor Agonists Increase Release of Soluble Amyloid Precursor Protein Derivatives from Rat Brain Cortical and Hippocampal Slices J. Pharmacol. Exp. Ther., April 1, 1997; 281(1): 149 - 154. [Abstract] [Full Text]
|
|
|
|
|
|
R. S. Turner, N. Suzuki, A. S. C. Chyung, S. G. Younkin, and V. M.-Y. Lee Amyloids beta[IMAGE] and beta[IMAGE] Are Generated Intracellularly in Cultured Human Neurons and Their Secretion Increases with Maturation J. Biol. Chem., April 12, 1996; 271(15): 8966 - 8970. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. M. Nitsch, M. Deng, J. H. Growdon, and R. J. Wurtman Serotonin 5-HT2a and 5-HT2c Receptors Stimulate Amyloid Precursor Protein Ectodomain Secretion J. Biol. Chem., February 23, 1996; 271(8): 4188 - 4194. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |