Báo cáo khóa học: Characterization of presenilin complexes from mouse and human brain using Blue Native gel electrophoresis reveals high expression in embryonic brain and minimal change in complex mobility with pathogenic presenilin mutations pptx

Characterization of presenilin complexes from mouse and human
brain using Blue Native gel electrophoresis reveals high expression
in embryonic brain and minimal change in complex mobility with
pathogenic presenilin mutations
Janetta G. Culvenor
1,2,3
, Nancy T. Ilaya
1,2,3
, Michael T. Ryan
4
, Louise Canterford
1,3
, David E. Hoke
1,3
,
Nicholas A. Williamson
1,3
, Catriona A. McLean
5
, Colin L. Masters
1,3
and Genevie
`
ve Evin
1,3
1
Department of Pathology and
2
Centre for Neuroscience, The University of Melbourne, Australia;
3
Mental Health Research Institute

of Victoria, Australia;
4
Department of Biochemistry, La Trobe University, Bundoora, Australia;
5
National Neuroscience Facility,
University of Melbourne, Australia
The presenilin proteins are required for intramembrane
cleavage of a subset of type 1 membrane proteins including
the Alzheimer’s disease amyloid precursor protein. Previous
studies indicate presenilin proteins form enzymatically act-
ive high molecular mass complexes consisting of hetero-
dimers of N- and C-terminal fragments in association with
nicastrin, presenilin enhancer-2 and anterior pharynx
defective-1 proteins. Using Blue Native gel electrophoresis
(BN/PAGE) we have studied endogenous presenilin 1
complex mass, stability and association with nicastrin,
presenilin enhancer-2 and anterior pharynx defective-1.
Solubilization of mouse or human brain membranes with
dodecyl-
D
-maltoside produced a 360-kDa species reactive
with antibodies to presenilin 1. Presenilin 1 complex levels
were high in embryonic brain. Complex integrity was sen-
sitive to Triton X-100 and SDS, but stable to reducing
agent. Addition of 5
M
urea caused complex dissolution and
nicastrin to migrate as a subcomplex. Nicastrin and pre-
senilin enhancer-2 were detected in the presenilin 1 complex
following BN/PAGE, electroelution and second-dimension

analysis. Anterior pharynx defective-1 was detected as an
18-kDa form and 9-kDa C-terminal fragment by standard
SDS/PAGE of mouse tissues, and as a predominant 36-kDa
band after presenilin 1 complex second-dimension analysis.
Membranes from brain cortex of Alzheimer’s disease
patients, or from cases with presenilin 1 missense mutations,
indicated no change in presenilin 1 complex mobility.
Higher molecular mass presenilin 1-reactive species were
detected in brain containing presenilin 1 exon 9 deletion
mutation. This abnormality was confirmed using cells
transfected with the same presenilin deletion mutation.
Keywords: Alzheimer’s disease; brain; Blue Native PAGE;
presenilin complex.
The presenilins are multispanning membrane proteins
(presenilin 1, PS1 and presenilin 2, PS2) first identified for
their genetic association with Alzheimer’s disease (AD).
They are essential components of a multiprotein protease
complex implicated in regulated intramembrane proteolysis
of several type 1 membrane proteins including the amyloid
precursor protein (APP) and developmentally important
Notch receptors [1]. The variety of identified substrates
indicates a critical role for PS in cell metabolism involving
controlled cleavage of protein transmembrane domains and
signal transduction. The role of PS in c-secretase activity
is compelling and includes lack of Ab amyloid peptide
generation and Notch signalling in PS double knockout
cells, cofractionation of activity with PS, abolition of
activity with mutation of conserved aspartates in trans-
membrane domains six and seven, binding of c-secretase
aspartyl protease inhibitors to PS, as well as identification of

homologues with protease-associated domains [2–6]. Fol-
lowing a primary cleavage event that causes ectodomain
shedding, PS-dependent Ôc-secretaseÕ proteolysis results in
generation of a membrane spanning stub (such as Ab
peptide) and a cytoplasmic fragment such as the APP
intracellular domain or the Notch intracellular domain
which translocates to the nucleus for regulation of gene
expression [1].
The PS proteins undergo endoproteolysis between trans-
membrane domains six and seven to generate heterodimers
of N-terminal fragments (NTF) and C-terminal fragments
(CTF) [7]. Many binding partners for the PS proteins have
been identified including substrates and the catenins (b, d
and p0071) [8]. The interacting components that are
required for c-secretase activity besides PS include the
Correspondence to J. G. Culvenor, Department of Pathology,
The University of Melbourne, Parkville, Victoria, 3010 Australia.
Fax: +61 38344 4004, Tel.: +61 38344 3990,
E-mail:
Abbreviations: AD, Alzheimer’s disease; APH, anterior pharynx
defective; APP, amyloid precursor protein; BACE, b-site APP-clea-
ving enzyme; BN/PAGE, Blue Native polyacrylamide gel electro-
phoresis; CTF, carboxyl-terminal fragment; DDM, n-dodecyl
b-
D
-maltoside; ECL, enhanced chemiluminescence; FAD, familial
Alzheimer’s disease; FTD, frontotemporal dementia; NCT, nicastrin;
NTF, amino-terminal fragment; PEN, presenilin enhancer;
PS1, presenilin 1; PS2, presenilin 2.
(Received 2 October 2003, revised 9 November 2003,

accepted 20 November 2003)
Eur. J. Biochem. 271, 375–385 (2004) Ó FEBS 2003 doi:10.1046/j.1432-1033.2003.03936.x
glycoprotein, nicastrin (NCT) identified by affinity purifi-
cation of PS1 complex using digitonin-solubilization of
membranes from human embryonic kidney cells [9], and
two further membrane proteins PEN-2 and APH-1, iden-
tified by genetic screening in Caenorhabditis elegans [10,11].
NCT is a heavily glycosylated type 1 membrane protein.
Maturation of NCT glycosylation is impaired in PS1-
deficient cells, and down-regulation of NCT inhibits
c-secretase cleavage of APP and Notch [11–13]. PEN-2 is
a small double membrane-spanning protein with N- and
C-termini facing the lumen [14]. APH-1 contains seven
putative transmembrane domains and has a variety of
isoforms with differing C-terminal domains; in humans,
APH-1a long and short forms are encoded by a gene on
chromosome 1 and APH-1b by chromosome 15 [15,16].
Glycerol velocity gradient centrifugation and gel filtration
of detergent-solubilized membrane extracts have character-
ized PS complexes of sizes ranging between 150 kDa and
two million kDa [5,17,18]. There is limited knowledge about
the assembly and role of the PS complex components other
than PS. We have applied Blue Native gel electrophoresis
(BN/PAGE) [19] to analyse further and characterize endo-
genous native complexes from mouse and human brain
tissue and from the neuroblastoma line SH-SY5Y (SY5Y).
Experimental procedures
Antibodies
Synthetic peptides derived from human sequences were
conjugated to diphtheria toxoid and rabbit antibodies

generated as follows: Ab 98/1, PS1 NT (1–20) [20]; Ab 00/
1 and Ab 00/2 PS1 loop (301–317) [21]; Ab 00/12, PS2 loop
(307–336); Ab 00/19 NCT CT (691–709]) [22]; Ab 00/22
NCT ectodomain (331–346); Ab 02/45 PEN-2 NT (1–15)
and Ab 02/41 APH-1b CT (244–257). Ab 00/6 was raised to
human BACE CT (485–501) [23]. NCT, PEN-2 and APH-1
antibodies were affinity-purified using immunizing peptide
coupled to SulfoLink Coupling Gel (Pierce). Mouse
mAb C1/6.1 raised to APP (676–695) was a gift from
P. Mathews, Nathan Kline Institute for Psychiatric
Research, NY, USA [24].
Cell lines
Human neuroblastoma SY5Y wild-type cells were cultured
as described previously and stably transfected with cDNA
for PS1, PS1 with exon 9 deletion mutation, or PS1 with
D257A artificial mutation following published protocols
[20]. Mouse embryonic neural stem cell lines NS51 (PS1–/–)
and NS52 (PS1 wt) were derived and cultured from PS1-
deficient mice as described [25].
Brain bank tissue
Samples of frontal cortex from human brain were obtained
from the NHMRC Tissue Resource Centre (Melbourne,
Australia) and were well characterized pathologically and
clinically [23]. Details of pathology or mutation analysis of
tissue used in this study is published as follows: FTD1 and
FTD2 [26], FAD PS1 L271V [27], FAD PS1 exon 9 deletion
[28], FAD PS1 S169L [29] and FAD PS1 L219P [30].
Membrane preparations
Cells were washed in NaCl/P
i

and pellets stored at )70 °C.
Mouse brain tissue was obtained from C57/Bl6 mice,
APPC100 transgenic mice (over-expressing last 99 residues
of human APP) [31], APPsw transgenic mice Tg 2576 (over-
expressing APP with Swedish mutation) [32], or transgenic
mice homozygous for human PS1M146 L [33]. Samples
were homogenized in 250 m
M
sucrose, 20 m
M
Hepes
pH 7.4 with protease inhibitors (Sigma P-2714) at 4 °C
with a glass Dounce homogenizer. Centrifugation at 1000 g
for 10 min, 4 °C, yielded a post-nuclear supernatant that
was further centrifuged at 100 000 g for 1 h, 4 °C, and the
resultant membrane pellets were resuspended in sucrose
homogenization buffer with protease inhibitors and stored
at )70 °C. Protein concentration was determined with
bicinchoninic acid reagent (Pierce).
BN/PAGE
Membrane protein (30–100 lg protein per gel lane) was
resuspended in 0.2–0.5% n-dodecyl b-
D
-maltoside (DDM)
(or alternate treatment as indicated) in buffer composed of
50 m
M
NaCl, 5 m
M
6-aminohexanoic acid and 50 m

M
imidazole pH 7.0 and then clarified after 15 min at room
temperature by microfuge centrifugation for 5 min before
addition of 0.5% Coomassie G250 (Sigma) and 50 m
M
6-aminohexanoic acid. Samples were resolved on 4–16.5%
gradient acrylamide gels with cooling; anode buffer con-
tained 25 m
M
imidazole pH 7.0; cathode buffer contained
50 m
M
tricine, 7.5 m
M
imidazole pH 7.0 with 0.02%
Coomassie G 250 [34]. High molecular mass standards for
native electrophoresis were from Amersham and were
stained with Coomassie R250. Following electrophoretic
transfer to poly(vinylidene difluoride) membrane (Immobi-
lon-P, Millipore), proteins were identified by immunoblot-
ting and detection with enhanced chemiluminescence (ECL;
Amersham). For second-dimension analysis, gel bands
corresponding to the PS complex region derived from
2-cm wide lanes, 16-cm length, 1.5-mm thickness slab gels
were excised at 1 cm below the ferritin marker as indicated,
proteins were electroeluted with BN/PAGE buffers, and
samples were solubilized with 1.5% SDS, 2.3
M
urea, 0.1
M

dithiothreitol, 15 m
M
Tris/HCl pH 6.8 with heating at
50 °C for 15 min before electrophoresis on 10% Tris/tricine
gels.
Immunoblotting
For analysis of PS1, NCT, PEN-2 and APH-1, membrane
proteins were separated with reducing agent on Tris/glycine/
SDS or Tris/tricine/SDS polyacrylamide gels and analysed
by immunoblotting as described previously [20,35].
Results
PS1 is detected in a 360 kDa complex with nicastrin
in embryonic mouse brain and SY5Y cells
Preliminary investigation of optimal conditions for PS1
complex analysis from mouse brain by BN/PAGE indicated
that solubilization of membranes with 0.5% DDM
376 J. G. Culvenor et al. (Eur. J. Biochem. 271) Ó FEBS 2003
generated a complex of mobility  360 kDa, which migra-
ted ahead of the ferritin marker (440 kDa) as detected by
Western analysis (Fig. 1A). The complex was reactive with
both PS1 N-terminal (98/1) and loop antibodies (00/1).
Digitonin was also compatible with maintaining the com-
plex stability and yielded a complex of higher apparent
molecular weight of about 460 kDa (Fig. 1A). Chapso
solubilization was also tested as it has proved useful for
maintaining c-secretase activity [5,36] but did not produce a
distinct band of PS1 complex immunoreactivity under these
electrophoresis conditions (data not shown). Embryonic
mouse brain produced stronger PS1 complex immunoreac-
tivity (Fig. 1A) compared to adult tissue. Results for brain

tissue from adult PS1 [M146L] transgenic mice is shown but
was the same for mouse wild-type brain. NCT was readily
detected in association with PS1 complex in embryonic
tissue as indicated by reactivity of NCT ectodomain
antibody. NCT was barely detectable in the PS complex
from adult mouse brain as shown in the last lane of Fig. 1A.
PS1 and NCT were also strongly detected as an apparent
complex in membranes from the human neuroblastoma cell
line SY5Y using PS1 Ab 98/1 and C-terminal NCT Ab 00/
19 (Fig. 2A).
We investigated further these apparent changes in PS1
and NCT detection with developmental stages. NCT levels
in neural stem cell lines and mouse brain are compared in
Fig. 1B after analysis using 12% SDS/PAGE and immu-
nodetection with NCT CT Ab 00/19 (left panel). Antibody
reactivity was removed by preabsorption with immunizing
peptide (middle panel). Consistent with previous reports
[12,37–39] we found that cells derived from PS1 knockout
mice express high levels of less mature NCT ( 110 kDa).
Embryonic mouse brain contained a more intermediate
glycosylated form of NCT ( 120 kDa) than adult brain
where the predominant NCT species was a 140-kDa form.
NCT ectodomain Ab 00/22 reactivity indicated detection of
only the less mature or fully deglycosylated forms of NCT
(data not shown), this finding may explain the poor
detection of NCT in adult mouse brain complex as high
glycosylation may impair antibody access to the epitope. As
expected, the PS1 N-terminal antibody detected a 29-kDa
immunoreactive band from these tissues corresponding to
PS1 NTF which was not detectable in PS1-deficient cells.

Amyloid precursor protein, b-catenin, and b-secretase
are not tightly associated with the PS1 complex
To investigate possible association of the c-secretase
substrate APP or APP C-terminal fragments with the
complex, BN/PAGE blots of membranes from SY5Y,
APPC100 transgenic mouse brain and APP over-expressing
transgenic mouse brain (Tg2576) were probed with APP
mAb C1/6.1 to the APP C terminus. Fig. 2A shows that
APP was detected as a smear of reactivity that spanned the
PS1 complex mobility. Some increase in APP reactivity
corresponding to PS1 complex mobility was observed for
APP Tg2576 tissue when contrast of the blot was modified,
indicated as Tg2576* on Fig. 2A. Antibody for the PS1
binding partner, b-catenin, produced a smear after BN/
PAGE of molecular mass greater than the 440 marker
consistent with interaction with many alternate binding
proteins.
Fig. 2B compares immunoreactivity for b-site APP-
cleaving enzyme (BACE), PS2 and PS1 in wild-type mouse
brain membrane extracts after BN/PAGE. Immunoreac-
tivity in a broad band of molecular mass greater than
550 kDa did not suggest association of BACE with the PS1
complex. PS2 immunoreactivity indicated that the PS2
complex has a slightly higher complex mass of  420 kDa
compared with  360 kDa for PS1.
Complex stability to detergents and denaturing agents
Samples of adult mouse brain membranes were solubilized
with alternative detergents and subjected to BN/PAGE to
Fig. 1. BN/PAGE of PS1 complex reveals a 360-kDa species reactive
with antibodies to PS1 and NCT. (A) Protein (100 lg) from membranes

of embryonic day 14 (E14)
4
mouse brain (E14MB), or from adult
mouse brain with PS1 [M146L] (AdMB), were treated with DDM or
digitonin before BN/PAGE and Western blotting for PS1 or NCT
detection. Sharp bands of PS1 reactivity indicated that the 360-kDa
PS1 complex contained both PS1 N-terminal reactivity and PS1
C-terminal reactivity in DDM. A broader band of 460 kDa PS1
immunoreactivity was found after digitonin solubilization. Reactivity
for PS1 and NCT was stronger in E14 mouse brain than for adult
mouse brain. NCT was detected in a 360-kDa complex with DDM and
low levels were detected as apparent monomer of  125 kDa; NCT
was detected only weakly in adult brain. (B) Reactivity of NCT
C-terminal antibody after SDS/PAGE for PS1-deficient mouse neural
stem cells, PS1 wild-type neural cells, E16 mouse brain and adult
mouse brain showed three major bands of 110, 120 and 140 kDa which
were absent after preabsorption with immunizing peptide. Right panel
shows PS1 antibody detection of PS1 29-kDa NTF after SDS/PAGE
(20 lg membrane protein per lane).
Ó FEBS 2003 Analysis of mouse and human brain presenilin complex (Eur. J. Biochem. 271) 377
determine PS1 complex stability. Fig. 3A shows that sample
preparation with 1% SDS, or a combination of 1% Triton
X-100/1% Nonidet P-40 abolished the 360-kDa PS1
complex signal that was observed with 0.5% DDM
solubilization. Treatment with 1% Triton X-100 caused
incomplete complex dissociation as a weak 360-kDa signal
was observed together with smearing in the 100–200-kDa
range after long exposure (Fig. 3A; ECL 60 min compared
with ECL 5 min). Thus, we confirm that detergent type is
critical for solubilization of the complex without disruption,

as shown by others [5,17]. The complex was stable in
0.5% DDM even after addition of reducing agent (1%
b-mercaptoethanol), indicating strong interaction, inde-
pendent of maintenance of disulfide bonds. Stability to
2
M
urea further supports this strong interaction. However,
the complex breaks down with increased urea concentration
of 5
M
in presence of reducing agent and 0.2% DDM
(Fig. 3B, left panel). At the low SDS concentration of 0.1%
with 0.2% DDM, PS1 subcomplexes of 90 and 60 kDa
were detected (Fig. 3B).
We also investigated the association of NCT with PS
complex using embryonic mouse brain where we have
shown optimal NCT detection. Paralleling the results
observed for PS1, we found that the low SDS concentration
of 0.1% caused a loss of 360-kDa immunoreactivity for
NCT, using either ectodomain or C-terminal directed
antibodies (Fig. 3B, right panels). A 140-kDa signal was
observed that is consistent with the apparent molecular
mass of NCT. The intermediate complex of 280 kDa has
a size consistent with that expected for a NCT dimer, and
was sensitive to the ionic detergent SDS but stable to
b-mercaptoethanol and 5
M
urea (Fig. 3B). It was hardly
detectable with antibody 00/22, suggesting that the mole-
cular interaction prevented antibody access to its epitope

that is located within the median region of the ectodomain.
Detection with the C-terminal antibody 00/19 was robust,
excluding interaction through the C-terminal domain. This
intermediate NCT complex was similar to the subcomplex
described previously on BN/PAGE for transfected cells
solubilized with digitonin [40,41].
Second dimension denaturing electrophoretic analysis
of semipurified PS complex
Prior to investigation of PEN-2 and APH-1 association with
the PS complex, mouse- and human-derived membrane
preparations were examined for PEN-2 immunoreactivity
andrevealedstrongdetectionofa10-kDabandinPS1+/
+neural stem cells, embryonic brain, and SY5Y cells; lower
levels were detected in PS1–/– cells, and in adult brain from
mouse and human (Fig. 4A). Down-regulation of PEN-2
with PS1-deficiency is in agreement with other reports [42].
APH-1 analysis using APH-1b CT [244–257] antibody in
the same samples detected APH-1b immunoreactivity at
18 kDa and 9 kDa (Fig. 4B). Levels of APH-1b were
similar for PS1-deficient and wild-type neural stem cells
indicating that levels are not coordinated with PS1 expres-
sion. Detection was low in adult human brain as compared
with SY5Y cells. Detection of the APH-1b products was
reduced by peptide absorption (data not shown). APH-1a
expression was also investigated but was detected more
weakly with the reagents currently available.
To analyse further the composition of the PS1 complex,
samples with high expression of PS1 complex were selected
for further study. BN/PAGE gel bands corresponding to the
region of PS1 complex were excised as indicated in Fig. 4C.

Fig. 4C shows that treatment of membranes with 0.1
M
sodium carbonate, pH 11.3, for removal of nonintegral
membrane proteins, enriched for PS1 immunoreactivity.
This treatment was previously shown to be compatible with
Fig. 2. PS complex does not contain b-catenin or BACE. (A) Membrane protein samples (50 lg) from SY5Y, C100 adult brain, or Tg2576 adult
brain in 0.4% DDM were resolved by BN/PAGE and immunoblotted with the indicated antibody. PS1 detection was enriched in SY5Y as a 360-
kDa band. APP was detected as a broad band above 140 kDa; contrast adjustment for Tg2576* indicated increased APP immunoreactivity with the
PS1 complex region. NCT was strongly detected in SY5Y membrane complex and weakly in adult mouse C100 brain complex. b-catenin was
detected as a broad smear above 440 kDa as indicated. (B) Membrane protein (50 lg) from adult mouse brain were compared for immuno-
reactivity with antibody as indicated. BACE was detected as a broad smear of apparent molecular mass greater than 440 kDa. PS2 was detected as
abandof 420 kDa.
378 J. G. Culvenor et al. (Eur. J. Biochem. 271) Ó FEBS 2003
Fig. 3. Stability of PS1 complex to detergents
and urea analysed by BN/PAGE. (A) Adult
mouse brain samples (50 lg) were treated with
0.5% DDM or as indicated before loading.
Triton X-100 1%/1% NP40, or 1% SDS
eliminated the PS1 complex at 360 kDa. ECL
exposure of 1 h revealed low detection of PS1
complex treated with 1% Triton compared
with 0.5% DDM. Addition of 2
M
urea or 1%
b-mercaptoethanol to 0.5% DDM did not
affect complex detection by PS1 antibody.
(B) Treatment of 50 lgE14mousebrainwith
0.2% DDM and 0.1% SDS or 5
M
urea

indicated disruption of PS1 complex. With
0.1% SDS, PS1 was detected as bands of 90
and60kDa.With0.1%SDS,NCTwas
detected largely as monomer with minimal
reactivity corresponding to possible dimer
or subcomplex. After treatment with 1%
b-mercaptoethanol/5
M
urea, NCT was
detected primarily at 280 kDa by NCT CT
antibody and only weakly as a monomer
with NCT ectodomain antibody.
Ó FEBS 2003 Analysis of mouse and human brain presenilin complex (Eur. J. Biochem. 271) 379
preservation of c-secretase activity [36]. Samples were
processed by second-dimension analysis to investigate
composition and presence of complex components.
Second-dimensional analysis of complex from carbonate-
washed membranes of SY5Y and 5-day-old mouse brain
were examined using 10% Tris/tricine gels with reducing
380 J. G. Culvenor et al. (Eur. J. Biochem. 271) Ó FEBS 2003
conditions and SDS (Fig. 4D and E). PS1 NTF and CTF
and PS2 CTF were detected robustly in these samples. Low
levels of full-length PS1 were also detected with N-terminal
antibody. NCT and PEN-2 were also detected in this semi-
purified complex. APH-1 was not detectable with the
reagents tested which included affinity-purified antibodies to
the C-terminal regions of APH-1a (short form) and APH-
1b, and to the loop four region of APH-1b (data not
shown).
Investigation of noncarbonate washed embryonic mouse

brain PS complex revealed similar detection of PS1 NTF,
PS1 CTF, low level of full-length PS1, mature NCT, and
PEN-2 (Fig. 4F). PS1 CTF was detected with PS1 loop
antibody as well as low amounts of full-length PS1. The
epitope for this antibody is largely buried in full-length PS1
as shown previously by us and others [26,43]. APH-1b was
detected for this complex as immunoreactivity at about
36 kDa with low level detection also at about 18 kDa and
54 kDa suggestive of self-association as dimer and trimer;
mobility corresponding to putative association with mature
NCT (140 kDa) was also detected. Evidence for self-
association of APH-1a with itself and with APH-1b was
also reported previously [16]. Morais et al.demonstrated
that NCT and APH-1 were able to interact in vitro in the
absence of PS [44]. Our results suggest that all four putative
complex components could be detected in the PS complex
for embryonic mouse brain after BN/PAGE, and that
APH-1 association may be diminished by carbonate wash.
Analysis of complex components by second-dimension
analysis for control and AD human brain samples without
carbonate wash revealed detection of PS1 NTF, PS1 CTF,
full-length PS1, NCT and PEN-2 (data not shown) but not
APH-1 which may be below detection levels.
Analysis of PS1 complex from sporadic AD and tissue
with early onset familial AD mutations
Membrane fractions were prepared from frontal cortex of
caseswithsporadicADandcomparedwithage-matched
controls. BN/PAGE analysis indicated no major differences
in PS1 complex mobility with AD (Fig. 5A). Human brain
cortex had lower detectable levels of PS1 complex compared

with preparations from SY5Y. No major differences in
expression of PS1 and NCT levels were found in these
tissues as shown in Fig. 5C using Western analysis after
standard SDS gel electrophoresis.
We next examined cases of early onset familial AD with
pathogenic PS1 missense mutations and found no alteration
in complex mobility by BN/PAGE analysis for tissue
containing PS1 [L271V], PS1 [S169L] or PS1 [L219P]
(Fig. 5B). No change in PS1 complex was found for two
fronto-temporal dementia cases associated with altered PS1
transcript expression [26]. However tissue from a case with
pathogenic PS1 exon 9 deletion showed an additional
oligomeric species of about 600 kDa (demonstrated twice).
PS1 expression and NCT levels for some of these early onset
casesisshowninFig.5D.
SY5Y cells stably transfected with PS1 cDNA constructs
were used to examine effects of over-expression of PS1
mutations on complex formation. Abnormal accumulation
of PS1 at about 600 kDa was detected in SY5Y cells over-
expressing the PS1 delta 9 construct. Increased PS1
expression or expression of the artificial loss of function
D257A mutation did not alter complex mobility. Interest-
ingly for all cell lines over-expression of PS1 did cause
increased complex detection by PS1 antibody (Fig. 5E) but
significant amounts of PS1 were not incorporated in the
complex, as indicated by a smear of reactivity between 50
and 100 kDa.
Discussion
The PS complex is critical for normal biological processes
such as Notch signalling in development in addition to the

generation of Ab peptides in pathological processes. Study
of the characteristics of the PS complex will facilitate
elaboration of the molecular processes involved in the
production of Ab. Analysis of these complexes requires
detergent solubilization in conditions that do not disrupt
protein interactions. This can lead to loss of weakly
interacting proteins or generation of artefact from nonspe-
cific associations. BN/PAGE utilizes Coomassie Blue G250
rather than SDS in the cathode buffer and sample buffer to
assist with protein solubilization and addition of charge for
electrophoretic field migration of proteins according to mass
[19]. It is ideal for analysis and partial isolation of membrane
Fig. 4. PEN-2 and APH-1 expression in mouse and human tissues and
detection in PS complex after 2D-complex analyis. (A) Membranes
from tissues and cells were analysed for PEN-2 (1–15) reactivity. Sig-
nificant expression was detected as a 10-kDa band in PS1 wild-type
neural stem cells, E15 mouse brain and SY5Y cells. Low-level
expression was present in PS1 deficient cells, adult mouse brain and in
control and AD human brain cortex (20 lg protein per lane, 15% Tris-
tricine gel). (B) Tissues were also examined for APH-1b CT reactivity.
APH-1b was detected at  18 kDa and as an apparent CTF of 9 kDa
except in adult human brain. Expression was unchanged with PS1
deficiency. (C) Sodium carbonate wash of membranes before BN/
PAGE analysis increased intensity of PS1 complex reactivity (50 lg
membrane protein per lane in 0.4% DDM). A region of the unstained
BN/polyacrylamide gel was excised as indicated by alignment with
standards. (D) Carbonate washed SY5Y membranes (500 lg) were
run on BN/PAGE, proteins electroeluted from the PS complex region,
treated with reducing sample buffer and analysed on 10% Tris/tricine
gels using mini Biorad apparatus. PS1 NTF and full-length PS1 were

detected with PS1 [1–20] antibody, PS1 loop antibody detected PS1
CTF, NCT was detected at  140 kDa, PEN-2 at 10 kDa and PS2
CTF detected with PS2 CTF antibody. Parallel poly(vinylidene diflu-
oride) strips were also probed with APH 1a and 1b antibodies but
APH-1 was not detected. (E) Carbonate washed membranes (500 lg)
from mouse brain post-natal day 5, were similarly probed after PS
complex isolation from BN/PAGE and second-dimension Tris/tricine
gel analysis using a large gel (16 · 18 cm) format. PS1 and PEN-2
were easily detected in these samples. NCT was clearly detected as
the mature form with very low detection of an immature form. Again
APH-1 was not detected with the available reagents. (F) E16
membranes were analysed in the absence of carbonate treatment.
Second-dimension analysis after BN/PAGE of the PS complex
detected full-length PS1, NTF, CTF and PEN-2. NCT was detected as
the mature form. APH-1b was detected primarily as a major band of
36 kDa (possible dimer) and as minor bands of 18 kDa and possible
trimer at 54 kDa (after long exposure); immunoreactivity with
mobility corresponding to NCT was also indicated.
Ó FEBS 2003 Analysis of mouse and human brain presenilin complex (Eur. J. Biochem. 271) 381
Fig. 5. BN/PAGE analysis of PS1 complex in membranes from human brain control and AD brains. (A) After BN/PAGE, PS1 complex from control
(CT) or AD cortex was detected in 50 lg membrane protein at 360 kDa. (B) Analysis of samples from AD cases with PS1 missense mutations and
fronto-temporal dementia (FTD) indicated no change in PS1 complex mobility by BN/PAGE. In familial AD (FAD) with exon 9 deletion, a higher
band of  600 kDa was detected as well as the major band for PS1 at 360 kDa (shown twice for confirmation). (C) Detection of PS1 and NCT after
12% Tris-glycine SDS/PAGE of 20 lg membrane samples of human brain analysed in (A) indicated no major difference in PS1 or NCT expression
in AD. (D) Analysis of samples used in (B) for PS1 and NCT expression with 12% Tris-glycine SDS/PAGE, indicated additional full-length PS1
detection with exon 9 deletion and no alteration in NCT expression. (E) Detection of PS1 in SY5Y stably transfected with PS1, PS1 exon 9 deletion
(delta E9), and PS1-D257A following BN/PAGE, indicated increased levels of PS1 complex with PS1 over-expression. Increased immunoreactivity
was detected at  600kDawithPS1deltaE9(30lg protein per lane from 18 000 g · 20-min post-nuclear pellets).
382 J. G. Culvenor et al. (Eur. J. Biochem. 271) Ó FEBS 2003
protein complexes under nondenaturing conditions. This

technique complements immunoprecipitation and velocity
gradient analyses that have been applied more extensively to
examine PS interactions with binding partners and complex
size. The molecular mass of PS1 complex determined here
with BN/PAGE corresponds to the size of  400 kDa
recently determined by velocity gradient centrifugation of
digitonin-solubilized membranes from HEK293 cells [45]. It
is also consistent with the size of Chapso-solubilized PS
complex obtained after over-expression of Drosophila PSin
Drosophila cells or N2a neuroblastoma cells (between 232
and 443 kDa) [46]. The detection of some full-length PS1
after second-dimension analysis of complex components
suggests assembly with full-length PS before endoproteo-
lysis to generate heterodimers of NTF and CTF compo-
nents consistent with previous reports [47].
Sensitivity of the PS complex to SDS and Triton X-100
and stability to reducing agent is consistent with previous
studies [17,48]. Conditions causing partial complex disrup-
tion can contribute to knowledge of complex assembly.
With 0.1% SDS, PS1 high molecular mass complex was
disrupted, and some PS1 with mobility of about 90 kDa
was detected, which may correspond to a subcomplex or
PS1 dimer. A recent report by He
´
bert et al. showed evidence
for PS1 dimerization using 0.1% SDS treatment and
alternative native electrophoresis conditions as well as by
yeast two-hybrid analysis [49]. We also document that NCT
migrates in BN/PAGE as a subcomplex or possible dimer
stable to reducing agent and high urea concentration. This

finding supports previous evidence, that NCT and APH-1
may associate independently of PS1 complexes [40,41,44]
and that NCT may form a subcomplex in the absence of
PS1 [13]. Low or no detection of APH-1 in the partially
purified endogenous PS complex by second-dimension
analysis, and minimal effect on APH-1 expression level
with PS1 deficiency supports the proposal that APH-1 may
be important for early assembly and stabilization of the
complex [50–52].
High levels of PS complex were detected in mouse
embryonic brain consistent with high expression of PS1
in embryogenesis and supporting an important role for
PS in development, presumably due to its involvement in
the Notch signalling system [53]. Indeed studies with PS1
knockout mouse embryos demonstrated severe neural
and skeletal defects and PS1/PS2 double knockouts
display a severe phenotype resembling Notch gene
knockout [3].
Alternative approaches have provided evidence that PS
and cofactors interact directly: affinity isolation of the
complexes [9], coimmunoprecipitation studies [12,37–39],
and complex isolation with c-secretase inhibitor [4,22,54].
Other recent studies have also utilized BN/PAGE for
analysis of PS complexes. DDM at 0.5% has been
recently reported independently to preserve PS complex
integrity in BN/PAGE with reports of complex size from
440 to 600 kDa [12,13,42,55,56]. Digitonin (1%) and BN/
PAGE produced PS1 complex mobility of  250–270 kDa
in cells over-expressing the four main constituents of the
c-secretase complex [40,41]. Estimates of complex mobility

may vary depending on sample preparation, detergent
type and concentration, as well as buffer and gel
composition.
Detection of APH-1b as an 18-kDa and CTF 9-kDa
form indicates that native APH-1 undergoes endoproteo-
lysis. Lack of detection of a free 9-kDa form after second-
dimension complex analysis indicates that this fragment is
unlikely to be associated with the mature complex. This is in
agreement with the report by Kimberley et al.thatAPH-1a
also generated a small CTF which did not associate with
other PS complex components after glycerol velocity
gradient fractionation for tagged APH-1 in transfected cells
[40].
This is the first report of native PS1 complex analysis
from human brain of late onset sporadic AD cases and
from cases with early onset PS1 mutations. We found that
PS1 complex from cortex of sporadic late-onset AD cases
or from pathogenic PS1 point mutations did not have
altered apparent mobility in BN/PAGE. This was consis-
tent with study of PS1 complex from transfected SY5Y
cells reported here for the artificial PS1 Asp257Ala
mutation, and for Asp257 or Asp385 mutations in
transfected mouse embryonic fibroblasts [56]. Additional
high molecular weight PS1 species of  600 kDa were
detected for brain carrying the severe PS1 exon 9 deletion
and for SY5Y cells over-expressing this mutation, indica-
ting impairment of normal complex formation in the
presence of the mutation.
In vitro co-expression studies in transfected cell systems
indicates PS1 or PS2 and NCT, APH-1, and PEN-2 are

all required for generation of active c-secretase activity
[40,50,52,57]. The current study of endogenous tissue levels
and detection within native endogenous complex for these
membrane proteins from mouse and human brain contri-
butes to further understanding of the nature of the native
mature presenilin/c-secretase complex complementing
in vitro cell-based studies.
Acknowledgements
We thank P. M. Mathews and R.M.D. Holsinger for antibodies, F. B.
Reinhard for SY5Y-PS1 mutant cell lines, and Q X. Li for transgenic
mouse tissue. The work was supported by the Australian NHMRC
(Grants 114132 and 208978) and ANZ Charitable Trusts. D. H. was
supported by an NIH post doctoral fellowship.
References
1. Fortini, M.E. (2002) Gamma-secretase-mediated proteolysis in
cell-surface-receptor signalling. Nat.Rev.Mol.CellBiol.3,
673–684.
2. Wolfe, M.S., Xia, W., Ostaszewski, B.L., Diehl, T.S., Kimberly,
W.T. & Selkoe, D.J. (1999) Two transmembrane aspartates in
presenilin-1 required for presenilin endoproteolysis and gamma-
secretase activity. Nature 398, 513–517.
3. Herreman, A., Serneels, L., Annaert, W., Collen, D., Schoonjans,
L. & De Strooper, B. (2000) Total inactivation of gamma-secretase
activity in presenilin-deficient embryonic stem cells. Nat. Cell Biol.
2, 461–462.
4. Esler, W.P., Kimberly, W.T., Ostaszewski, B.L.YeW., Diehl, T.S.,
Selkoe, D.J. & Wolfe, M.S. (2002) Activity-dependent isolation of
the presenilin-gamma-secretase complex reveals nicastrin and a
gamma substrate. Proc.NatlAcad.Sci.USA99, 2720–2725.
5. Li, Y.M., Lai, M.T., Xu, M., Huang, Q., DiMuzio-Mower, J.,

Sardana, M.K., Shi, X.P., Yin, K.C., Shafer, J.A. & Gardell, S.J.
(2000) Presenilin 1 is linked with gamma-secretase activity in the
Ó FEBS 2003 Analysis of mouse and human brain presenilin complex (Eur. J. Biochem. 271) 383
detergent solubilized state. Proc. Natl Acad. Sci. USA 97,
6138–6143.
6. Ponting,C.P.,Hutton,M.,Nyborg,A.,Baker,M.,Jansen,K.&
Golde, T.E. (2002) Identification of a novel family of presenilin
homologues. Hum. Mol. Genet. 11, 1037–1044.
7. Thinakaran, G., Borchelt, D.R., Lee, M.K., Slunt, H.H., Spitzer,
L., Kim, G., Ratovitsky, T., Davenport, F., Nordstedt, C., Seeger,
M.,Hardy,J.,Levey,A.I.,Gandy,S.E.,Jenkins,N.A.,Copeland,
N.G., Price, D.L. & Sisodia, S.S. (1996) Endoproteolysis of pre-
senilin 1 and accumulation of processed derivatives in vivo. Neuron
17, 181–190.
8. VanGassen,G.,Annaert,W.&VanBroeckhoven,C.(2000)
Binding partners of Alzheimer’s disease proteins: are they physio-
logically relevant? Neurobiol. Dis. 7, 135–151.
9. Yu,G.,Nishimura,M.,Arawaka,S.,Levitan,D.,Zhang,L.,
Tandon, A., Song, Y.Q., Rogaeva, E., Chen, F., Kawarai, T.,
Supala,A.&Levesque,L.,YuH.,Yang,D.S.,Holmes,E.,Mil-
man, P., Liang, Y., Zhang, D.M., Xu, D.H., Sato, C., Rogaev, E.,
Smith, M., Janus, C., Zhang, Y., Aebersold, R., Farrer, L.S.,
Sorbi, S., Bruni, A., Fraser, P. & St George-Hyslop, P. (2000)
Nicastrin modulates presenilin-mediated notch/glp-1 signal
transduction and betaAPP processing. Nature 407, 48–54.
10. Goutte, C., Tsunozaki, M., Hale, V.A. & Priess, J.R. (2002) APH-
1 is a multipass membrane protein essential for the Notch sig-
naling pathway in Caenorhabditis elegans embryos. Proc. Natl
Acad. Sci. USA 99, 775–779.
11. Francis, R., McGrath, G., Zhang, J., Ruddy, D.A., Sym, M.,

Apfeld, J., Nicoll, M., Maxwell, M., Hai, B., Ellis, M.C., Parks,
A.L., Xu, W., Li, J., Gurney, M., Myers, R.L., Himes, C.S.,
Hiebsch, R., Ruble, C., Nye, J.S. & Curtis, D. (2002) aph-1 and
pen-2 are required for Notch pathway signaling, gamma-secretase
cleavage of betaAPP, and presenilin protein accumulation. Dev.
Cell 3, 85–97.
12. Edbauer, D., Winkler, E., Haass, C. & Steiner, H. (2002) Pre-
senilin and nicastrin regulate each other and determine amyloid
beta-peptide production via complex formation. Proc. Natl Acad.
Sci. USA 99, 8666–8671.
13. Li, T., Ma, G., Cai, H., Price, D.L. & Wong, P.C. (2003) Nicastrin
is required for assembly of presenilin/gamma-secretase complexes
to mediate Notch signaling and for processing and trafficking of
beta-amyloid precursor protein in mammals. J. Neurosci. 23,
3272–3277.
14. Crystal, A.S., Morais, V.A., Pierson, T.C., Pijak, D.S., Carlin, D.,
Lee, V.M. & Doms, R.W. (2003) Membrane topology of gamma-
secretase component PEN-2. J. Biol. Chem. 278, 20117–20123.
15. Lee, S.F., Shah, S. & Li, H., YuC. & , Han, W.G. (2002) Mam-
malian APH-1 interacts with presenilin and nicastrin and is
required for intramembrane proteolysis of amyloid-beta precursor
protein and Notch. J. Biol. Chem. 277, 45013–45019.
16. Gu,Y.,Chen,F.,Sanjo,N.,Kawarai,T.,Hasegawa,H.,Duthie,
M., Li, W., Ruan, X., Luthra, A., Mount, H.T., Tandon, A.,
Fraser, P.E. & St George-Hyslop, P. (2003) APH-1 interacts with
mature and immature forms of presenilins and nicastrin and may
play a role in maturation of presenilin.nicastrin complexes. J. Biol.
Chem. 278, 7374–7380.
17. Capell,A.,Grunberg,J.,Pesold,B.,Diehlmann,A.,Citron,M.,
Nixon, R., Beyreuther, K., Selkoe, D.J. & Haass, C. (1998) The

proteolytic fragments of the Alzheimer’s disease-associated pre-
senilin-1 form heterodimers and occur as a 100–150-kDa
molecularmasscomplex.J. Biol. Chem. 273, 3205–3211.
18. Seeger, M., Nordstedt, C., Petanceska, S., Kovacs, D.M., Gouras,
G.K.,Hahne,S.,Fraser,P.,Levesque,L.,Czernik,A.J.,George-
Hyslop, P.S., Sisodia, S.S., Thinakaran, G., Tanzi, R.E., Green-
gard, P. & Gandy, S. (1997) Evidence for phosphorylation and
oligomeric assembly of presenilin 1. Proc. Natl Acad. Sci. USA 94,
5090–5094.
19. Scha
¨
gger
1
, H. & von Jagow, G. (1991) Blue native electrophoresis
for isolation of membrane protein complexes in enzymatically
active form. Anal. Biochem. 199, 223–231.
20. Culvenor, J.G., Evin, G., Cooney, M.A., Wardan, H., Sharples,
R.A.,Maher,F.,Reed,G.,Diehlmann,A.,Weidemann,A.,
Beyreuther, K. & Masters, C.L. (2000) Presenilin 2 expression in
neuronal cells: induction during differentiation of embryonic car-
cinoma cells. Exp. Cell Res. 255, 192–206.
21. Evin, G., Sharples, R.A., Weidemann, A., Reinhard, F.B., Car-
bone, V., Culvenor, J.G., Holsinger, R.M., Sernee, M.F., Beyre-
uther, K. & Masters, C.L. (2001) Aspartyl protease inhibitor
pepstatin binds to the presenilins of Alzheimer’s disease. Bio-
chemistry 40, 8359–8368.
22. Beher, D., Fricker, M., Nadin, A., Clarke, E.E., Wrigley, J.D., Li,
Y.M., Culvenor, J.G., Masters, C.L., Harrison, T. & Shearman,
M.S. (2003) In vitro characterization of the presenilin-dependent
gamma-secretase complex using a novel affinity ligand. Biochem-

istry 42, 8133–8142.
23. Holsinger,R.M.D.,McLean,C.A.,Beyreuther,K.,Masters,C.L.
& Evin, G. (2002) Increased expression of the amyloid precursor
b-secretase in Alzheimer’s disease. Ann. Neurol. 51, 783–786.
24. Mathews, P.M., Guerra, C.B., Jiang, Y., Grbovic, O.M., Kao,
B.H., Schmidt, S.D., Dinakar, R., Mercken, M., Hille-Rehfeld,
A.,Rohrer,J.,Mehta,P.,Cataldo,A.M.&Nixon,R.A.(2002)
Alzheimer’s disease-related overexpression of the cation-depend-
ent mannose 6-phosphate receptor increases Ab secretion: role for
altered lysosomal hydrolase distribution in b-amyloidogenesis.
J. Biol. Chem. 277, 5299–5307.
25. Culvenor, J.G., Rietze, R.L., Bartlett, P.F., Masters, C.L. & Li,
Q.X. (2002) Oligodendrocytes from neural stem cells express
alpha-synuclein: increased numbers from presenilin 1 deficient
mice. Neuroreport 13, 1305–1308.
26. Evin,G.,Smith,M.J.,Tziotis,A.,McLean,C.,Canterford,L.,
Sharples, R.A., Cappai, R., Weidemann, A., Beyreuther, K.,
Cotton, R.G., Masters, C.L. & Culvenor, J.G. (2002) Alternative
transcripts of presenilin-1 associated with frontotemporal
dementia. Neuroreport 13, 917–921.
27. Kwok, J.B., Halliday, G.M., Brooks, W.S., Dolios, G., Laudon,
H., Murayama, O., Hallupp, M., Badenhop, R.F., Vickers, J.,
Wang, R., Naslund, J., Takashima, A., Gandy, S.E. & Schofield,
P.R. (2002) Presenilin-1 mutation (Leu271VAL) results in altered
exon 8 splicing and Alzheimer’s disease with non-cored plaques
andnoneuriticdystrophy.J. Biol. Chem. 278, 6748–6754.
28. Smith, M.J., Kwok, J.B., McLean, C.A., Kril, J.J., Broe, G.A.,
Nicholson, G.A., Cappai, R., Hallupp, M., Cotton, R.G.,
Masters,C.L.,Schofield,P.R.&Brooks,W.S.(2001)Variable
phenotype of Alzheimer’s disease with spastic paraparesis. Ann.

Neurol. 49, 125–129.
29. Taddei, K., Kwok, J.B., Kril, J.J., Halliday, G.M., Creasey, H.,
Hallupp, M., Fisher, C., Brooks, W.S., Chung, C., Andrews, C.,
Masters, C.L., Schofield, P.R. & Martins, R.N. (1998) Two novel
presenilin-1 mutations (Ser169Leu and Pro436Gln) associated
with very early onset Alzheimer’s disease. Neuroreport 9, 3335–
3339.
30. Smith, M.J., Gardner, R.J., Knight, M.A., Forrest, S.M., Beyre-
uther, K., Storey, E., McLean, C.A., Cotton, R.G., Cappal, R. &
Masters, C.L. (1999) Early-onset Alzheimer’s disease caused by a
novel mutation at codon 219 of the presenilin-1 gene. Neuroreport
10, 503–507.
31. Li, Q.X., Maynard, C., Cappai, R., McLean, C.A., Cherny, R.A.,
Lynch, T., Culvenor, J.G., Trevaskis, J., Tanner, J.E., Bailey,
K.A.,Czech,C.,Bush,A.I.,Beyreuther,K.&Masters,C.L.
(1999) Intracellular accumulation of detergent-soluble amyloido-
genic Ab fragment of Alzheimer’s disease precursor protein
in the hippocampus of aged transgenic mice. J. Neurochem. 72,
2479–2487.
384 J. G. Culvenor et al. (Eur. J. Biochem. 271) Ó FEBS 2003
32. Hsiao, K., Chapman, P., Nilsen, S., Eckman, C., Harigaya, Y.,
Younkin, S., Yang, F.S. & Cole, G. (1996) Correlative memory
deficits, Ab elevation, and amyloid plaques in transgenic mice.
Science 274, 99–102.
33. Duff,K.,Eckman,C.&Zehr,C.,Yu,X.,Prada,C.M.,Pereztur,
J., Hutton, M., Buee, L., Harigaya, Y., Yager, D., Morgan, D.,
Gordon,M.N.,Holcomb,L.,Refolo,L.,Zenk,B.,Hardy,J.&
Younkin, S. (1996) Increased amyloid-b42 (43) in brains of mice
expressing mutant presenilin 1. Nature 383, 710–713.
34. Scha

¨
gger, H. (2001) Blue-native gels to isolate protein complexes
from mitochondria. Methods Cell Biol. 65, 231–244.
35. Culvenor, J.G., McLean, C.A., Cutt, S., Campbell, B.C., Maher,
F., Ja
¨
ka
¨
la
¨
, P., Hartmann, T., Beyreuther, K., Masters, C.L. & Li,
Q.X. (1999) Non-Ab component of Alzheimer’s disease amyloid
(NAC) revisited. NAC and a-synuclein are not associated with Ab
amyloid. Am. J. Pathol. 155, 1173–1181.
36. Esler, W.P., Kimberly, W.T., Ostaszewski, B.L.YeW., Diehl, T.S.,
Selkoe, D.J. & Wolfe, M.S. (2002) Activity-dependent isolation of
the presenilin-c-secretase complex reveals nicastrin and a c sub-
strate. Proc. Natl Acad. Sci. USA 99, 2720–2725.
37. Leem, J.Y., Vijayan, S., Han, P., Cai, D., Machura, M., Lopes,
K.O., Veselits, M.L., Xu, H. & Thinakaran, G. (2002) Presenilin 1
is required for maturation and cell surface accumulation of
nicastrin. J. Biol. Chem. 277, 19236–19240.
38. Tomita,T.,Katayama,R.,Takikawa,R.&Iwatsubo,T.(2002)
Complex N-glycosylated form of nicastrin is stabilized and selec-
tively bound to presenilin fragments. FEBS Lett. 520, 117–121.
39. Yang,D.S.,Tandon,A.,Chen,F.,Yu,G&Yu,H.,Arawaka,S.,
Hasegawa, H., Duthie, M., Schmidt, S.D., Ramabhadran, T.V.,
Nixon, R.A., Mathews, P.M., Gandy, S.E., Mount, H.T., St
George-Hyslop, P. & Fraser, P.E. (2002) Mature glycosylation
and trafficking of nicastrin modulate its binding to presenilins.

J. Biol. Chem. 277, 28135–28142.
40. Kimberly, W.T., LaVoie, M.J., Ostaszewski, B.L., Ye, W., Wolfe,
M.S. & Selkoe, D.J. (2003) Gamma-secretase is a membrane
protein complex comprised of presenilin, nicastrin, Aph-1, and
Pen-2. Proc. Natl Acad. Sci. USA 100, 6382–6387.
41. LaVoie, M.J., Fraering, P.C., Ostaszewski, B.L., Ye, W., Kimberly,
W.T., Wolfe, M.S. & Selkoe, D.J. (2003) Assembly of the gamma-
secretase complex involves early formation of an intermediate sub-
complex of Aph-1 and Nicastrin. J. Biol. Chem. 278, 37213–37222.
42. Steiner, H., Winkler, E., Edbauer, D., Prokop, S., Basset, G.,
Yamasaki, A., Kostka, M. & Haass, C. (2002) PEN-2 is an
integral component of the gamma-secretase complex required for
coordinated expression of presenilin and nicastrin. J. Biol. Chem.
277, 39062–39065.
43. Honda, T., Nihonmatsu, N., Yasutake, K., Ohtake, A., Sato, K.,
Tanaka, S., Murayama, O., Murayama, M. & Takashima, A.
(2000) Familial Alzheimer’s disease-associated mutations block
translocation of full-length presenilin 1 to the nuclear envelope.
Neurosci. Res. 37, 101–111.
44. Morais, V.A., Crystal, A.S., Pijak, D.S., Carlin, D., Costa, J., Lee,
V.M. & Doms, R.W. (2003) The transmembrane domain region of
nicastrin mediates direct interactions with APH-1 and the gamma-
secretase complex. J. Biol. Chem. 278, 43284–43289.
2
45. Gu,Y.,Chen,F.,Sanjo,N.,Kawarai,T.,Hasegawa,H.,Duthie,
M., Li, W., Ruan, X., Luthra, A., Mount, H.T., Tandon, A.,
Fraser, P.E. & St George-Hyslop, P. (2003) APH-1 interacts with
mature and immature forms of presenilins and nicastrin and may
play a role in maturation of presenilin-nicastrin complexes. J. Biol.
Chem. 278, 7374–7380.

46. Takasugi, N., Takahashi, Y., Morohashi, Y., Tomita, T. &
Iwatsubo, T. (2002) The mechanism of gamma-secretase activities
through high molecular weight complex formation of presenilins is
conserved in Drosophila melanogaster and mammals. J. Biol.
Chem. 277, 50198–50205.
47. Saura, C.A., Tomita, T., Davenport, F., Harris, C.L., Iwatsubo,
T. & Thinakaran, G. (1999) Evidence that intramolecular asso-
ciations between presenilin domains are obligatory for endo-
proteolytic processing. J. Biol. Chem. 274, 13818–13823.
48. Thinakaran, G., Regard, J.B., Bouton, C.M., Harris, C.L., Price,
D.L., Borchelt, D.R. & Sisodia, S.S. (1998) Stable association of
presenilin derivatives and absence of presenilin interactions with
APP. Neurobiol. Dis. 4, 438–453.
49. Hebert, S.S., Godin, C., Tomiyama, T., Mori, H. & Levesque, G.
(2003) Dimerization of presenilin-1 in vivo: suggestion of novel
regulatory mechanisms leading to higher order complexes.
Biochem. Biophys. Res. Commun. 301, 119–126.
50. Hu, Y. & Fortini, M.E. (2003) Different cofactor activities in
gamma-secretase assembly: evidence for a nicastrin-Aph-1 sub-
complex. J. Cell Biol. 161, 685–690.
51. Luo,W.J.,Wang,H.,Li,H.,Kim,B.S.,Shah,S.,Lee,H.J.,
Thinakaran, G. & Kim, T.W., Yu, G. & Xu, H. (2003) PEN-2 and
APH-1 coordinately regulate proteolytic processing of presenilin
1. J. Biol. Chem. 278, 7850–7854.
52. Takasugi, N., Tomita, T., Hayashi, I., Tsuruoka, M., Niimura,
M., Takahashi, Y., Thinakaran, G. & Iwatsubo, T. (2003) The
role of presenilin cofactors in the gamma-secretase complex.
Nature 422, 438–441.
53. Lee, M.K., Slunt, H.H., Martin, L.J., Thinakaran, G., Kim, G.,
Gandy,S.E.,Seeger,M.,Koo,E.,Price,D.L.&Sisodia,S.S.

(1996) Expression of presenilin 1 and 2 (PS1 and PS2) in human
and murine tissues. J. Neurosci. 16, 7513–7525.
54. Kimberly, W.T., LaVoie, M.J., Ostaszewski, B.L., Ye, W., Wolfe,
M.S. & Selkoe, D.J. (2002) Complex N-linked glycosylated
nicastrin associates with active gamma-secretase and undergoes
tight cellular regulation. J. Biol. Chem. 277, 35113–35117.
55. Farmery, M.R., Tjernberg, L.O., Pursglove, S.E., Bergman, A.,
Winblad, B. & Naslund, J. (2003) Partial purification and char-
acterization of gamma-secretase from post-mortem human brain.
J. Biol. Chem. 278, 24277–24284.
56. Nyabi, O., Bentahir, M., Horre, K., Herreman, A., Gottardi-
Littell, N., Van Broeckhoven, C., Merchiers, P., Spittaels, K.,
Annaert, W. & De Strooper, B. (2003) Presenilins mutated at
Asp257 or Asp385 restore Pen-2 expression and nicastrin glyco-
sylation but remain catalytically inactive in the absence of wild
type presenilin. J. Biol. Chem. 278, 43430–43436.
3
57. Edbauer,D.,Winkler,E.,Regula,J.T.,Pesold,B.,Steiner,H.&
Haass, C. (2003) Reconstitution of gamma-secretase activity. Nat.
Cell Biol. 5, 486–488.
Ó FEBS 2003 Analysis of mouse and human brain presenilin complex (Eur. J. Biochem. 271) 385