He Et Al 2010 SI

Vol 467 | 2 September 2010 | doi:10.
1038/nature09325
LETTERS
Gamma-secretase activating protein is a therapeutic
target for Alzheimers disease
Gen He1, Wenjie Luo1, Peng Li2, Christine Remmers1, William J. Netzer1, Joseph Hendrick2, Karima Bettayeb1,
Marc Flajolet1, Fred Gorelick3,4, Lawrence P. Wennogle2 & Paul Greengard1
Accumulation of neurotoxic amyloid-b is a major hallmark of

Alzheimers disease1. Formation of amyloid-b is catalysed by
c-secretase, a protease with numerous substrates2,3. Little is known
about the molecular mechanisms that confer substrate specificity
on this potentially promiscuous enzyme. Knowledge of the
mechanisms underlying its selectivity is critical for the development of clinically effective c-secretase inhibitors that can reduce
amyloid-b formation without impairing cleavage of other c-secretase
substrates, especially Notch, which is essential for normal biological
functions3,4. Here we report the discovery of a novel c-secretase
activating protein (GSAP) that drastically and selectively increases
amyloid-b production through a mechanism involving its interactions with both c-secretase and its substrate, the amyloid precursor
protein carboxy-terminal fragment (APP-CTF). GSAP does not
interact with Notch, nor does it affect its cleavage. Recombinant
GSAP stimulates amyloid-b production in vitro. Reducing GSAP
concentrations in cell lines decreases amyloid-b concentrations.
Knockdown of GSAP in a mouse model of Alzheimers disease
reduces levels of amyloid-b and plaque development. GSAP represents a type of c-secretase regulator that directs enzyme specificity by
interacting with a specific substrate. We demonstrate that imatinib,
an anticancer drug previously found to inhibit amyloid-b formation
without affecting Notch cleavage5, achieves its amyloid-b-lowering
effect by preventing GSAP interaction with the c-secretase substrate,
APP-CTF. Thus, GSAP can serve as an amyloid-b-lowering therapeutic target without affecting other key functions of c-secretase.
We have reported that imatinib (also known as STI571 or Gleevec)
decreases production of all amyloid-b species by inhibiting c-cleavage
of APP-CTF5. To identify the direct target responsible for imatinibs
selective amyloid-b-lowering activity, we synthesized a photoactivatable azido imatinib derivative, G01 (Supplementary Methods and
Supplementary Fig. 2). When 125I-G01 was incubated with a membrane preparation followed by photolysis, none of the four components of c-secretase were labelled. Rather, 125I-G01 labelled a
,16-kDa protein (Fig. 1a, left panel) which co-immunoprecipitated
with the more slowly migrating 18-kDa presenilin-1-CTF (Fig. 1a,
right panel). This result was confirmed by intact-cell photolabelling
using cell-permeable 3H-G01: the 3H-imatinib derivative did not bind
to any of the four c-secretase components, but did label a band of
,16 kDa that co-immunoprecipitated with presenilin 1 (PS1, also
known as PSEN1; Fig. 1a, middle panel).
To purify the potential target protein, we synthesized a biotinylated
derivative of imatinib, biotin-imatinib (Supplementary Methods and
Supplementary Fig. 3). Solubilized c-secretase components, including
presenilin 1, presenilin enhancer 2 and nicastrin, were specifically
captured by the immobilized biotin-imatinib (Fig. 1b, left panel). A
,16-kDa band was observed by silver staining (Fig. 1b, right panel)
after biotin-imatinib-bound proteins were separated by SDS
polyacrylamide gel electrophoresis, in agreement with the photolabelling results. Peptide fragments, derived from this band after trypsin
digestion, and analysed by tandem mass spectrometry, corresponded
to the C-terminal region of an uncharacterized protein, pigeon homologue protein (PION; human accession number, NP_059135). The
identification was made on the basis of two unique tryptic peptides
(766LWDHPMSSNIISR778 and 779NHVTRLLQNYKK790) covering
approximately 20% of the 16-kDa fragment. Its sequence, especially
the C-terminal region, is highly conserved among multiple species
from chicken to human (Supplementary Fig. 4). Expression pattern
analysis indicates that this gene is expressed in diverse tissues, including brain (Supplementary Fig. 5). In this report, we characterize PION
as a GSAP.
On the basis of its predicted sequence, the full open reading frame
of human GSAP encodes a protein of 854 amino acids (,98 kDa). To
determine whether the 16-kDa fragment was derived from a highmolecular-weight precursor, the metabolism of endogenous GSAP in
cells was monitored by pulsechase analysis. The results showed that
GSAP is synthesized as a holoprotein (,98 kDa) and is rapidly processed into a ,16-kDa C-terminal fragment (GSAP-16K) (Fig. 1c).
In the steady state, the 16-kDa fragment is the predominant form
(Fig. 1c).
Incubation of cells with 3H-G01, followed by photolysis and
immunoprecipitation with anti-GSAP antibody, confirmed that
imatinib directly binds GSAP-16K (Fig. 1d). When GSAP levels were
reduced using short interfering RNA, the amount of c-secretase
(represented by PS1-CTF in Fig. 1e) associated with biotin-imatinib
drastically decreased. This indicates that the affinity of imatinib for
the c-secretase complex depends on GSAP.
The effect of GSAP on amyloid-b generation is shown in Fig. 2.
When siRNA was used to reduce the GSAP concentration (by
72 6 15%) in N2a cells overexpressing APP695, the concentration
of amyloid-b decreased by about 50 6 7% (Fig. 2a); imatinib had little
or no additional effect on amyloid-b concentrations. This result
indicates that GSAP is the molecule through which imatinib lowers
amyloid-b. GSAP knockdown resulted in decreased concentrations of
all major amyloid-b species; Ab38 by 43 6 8%, Ab40 by 53 6 13% and
Ab42 by 48 6 7% (Fig. 2b). GSAP showed no detectable effect on aand b-cleavages (Supplementary Fig. 6). To further investigate
whether GSAP can modulate c-secretase activity, we examined the
effect of purified GSAP on amyloid-b production in an in vitro
c-secretase assay. When recombinant GSAP-16K (amino acids 733
854 of full-length human GSAP), isolated after expression in E. coli,
was added to membrane preparations from HEK cells containing
1
Laboratory of Molecular and Cellular Neuroscience, The Rockefeller University, 1230 York Avenue, New York, New York 10065, USA. 2Intra-Cellular Therapies, Inc., Audubon
Biomedical Science and Technology Park, 3960 Broadway, New York, New York 10032, USA. 3Department of Internal Medicine and Cell Biology, Yale University School of Medicine,
New Haven, Connecticut 06520, USA. 4VA Connecticut Healthcare, 333 Cedar Street, New Haven, Connecticut 06520, USA.
95
2010 Macmillan Publishers Limited. All rights reserved
LETTERS
NATURE | Vol 467 | 2 September 2010
IP
Control
IgG
IP
PS1
Ab
Control
IgG
3H-G01
intact-cell
photolabelling
Mw
16 kDa
GSAP
PEN2
Input
c
1
+
+
1.5
IP
GSAP
Ab
+
GSAP-16K
+
+
80
**
100
50
600
**
400
200
Amyloid-
GSAP-16K (10 g ml1)
L-685,458 (1 M)
d
GSAP shRNA
GSAP-16K
3H-G01
intact-cell
photolabelling
overexpressed b-secretase cleaved C-terminal fragment of APP (APPb-CTF), the concentration of amyloid-b was increased and that of
APP intracellular domain (AICD) was reduced (Fig. 2c).
APP-CTF is cleaved by c-secretase in the middle of its transmembrane domain to generate amyloid-b (c-cleavage) and near its cytosolic membrane boundary to generate AICD (e-cleavage). The effect
of GSAP on AICD production was examined in N2a cells overexpressing APP695. Both GSAP knockdown and imatinib treatment
increased concentrations of AICD (Supplementary Fig. 7a). GSAP
overexpression in HEK293 cells reduced AICD production (Supplementary Fig. 7b). These results indicate that GSAP differentially
regulates c- and e-cleavage of APP-CTF to form amyloid-b and
AICD, respectively.
One distinctive feature of imatinib is its selective inhibition of
amyloid-b production while sparing Notch cleavage5. The effect of
GSAP on Notch cleavage was evaluated using cells expressing
NotchDE (Notch without its extracellular domain), which is the
Notch substrate for c-secretase. As shown in Fig. 2d, the concentration
20
+
**
PS1-CTF
Figure 1 | Identification of GSAP as an imatinib target. a, A PS1-associated

16-kDa protein is labelled by a photoactivatable imatinib derivative. Left,
photolysis of 125I-G01 with membrane preparations. Middle, photolysis of
3
H-G01 with intact HEK293 cells. Right, PS1-CTF migrated with a slower
mobility than the labelled 16-kDa band and was not labelled by G01. Labelling
specificity was confirmed by competition with unlabelled imatinib. Ab,
antibody; IgG, immunoglobulin-G; Mw, weight-averaged molecular mass.
b, Solubilized endogenous c-secretase components from HEK293 cells were
bound to immobilized biotin-imatinib (left). Among the proteins bound to
biotin-imatinib, a ,16-kDa band was detected by silver staining and was
identified as the C-terminal domain of GSAP (right; arrow and label GSAP).
Biotin-coated beads and an inactive biotin-imatinib derivative
(Supplementary Fig. 3) served as controls. PEN2, presenilin enhancer 2.
c, Endogenous GSAP in N2a cells was synthesized as a full-length 98-kDa
precursor protein (GSAP-FL) and rapidly processed into a 16-kDa C-terminal
fragment. Under steady-state conditions, the predominant cellular form of
GSAP was 16 kDa. d, Endogenous GSAP-16K was specifically labelled by 3HG01 in neuroblastoma cells. e, After GSAP siRNA knockdown in N2a cells,
immobilized biotin-imatinib no longer captured PS1.
**
40
AICD
Biotin-imatinib
bound
+ GSAP siRNA
60
NICD
e
Input
800
150
Steady
2.5 (hr) state
GSAP-16K
Control
IgG
Unlabelled
b
imatinib
200
0
GSAP siRNA
GSAP-FL (98 kDa)
Silver staining
Bound
0.5
Amyloid- (a.u.)
PS1-CTF
Bio
tin
(ac -ima
t
t
Bio ive) inib
ti
(in n-im
ac ati
tiv nib
e)
Pull-down
Nicastrin
**
PS1-CTF blot
Bio
tin
Bio
tin
Bio
tin
(ac -ima
t
Bio tive) inib
ti
(in n-im
ac ati
tiv nib
e)
Bio
tin
Bio
tin
(ac -ima
t
t
Bio ive) inib
tin
(in im
ac ati
tiv nib
e)
membrane
photolabelling
A38 (pg ml1)
125I-G01
APP
Imatinib (10 M)
GSAP siRNA
**
100
75
50
25
0
Amyloid-
16 kDa
GSAP-16K
PS1
Ab
A42 (pg ml1)
PS1
Ab
A40 (pg ml1)
Control
IgG
Unlabelled imatinib
Mw
Amyloid- (a.u.)
IP
200
100
0
+
+
+
+
+
L-685,458 (1 M)
GSAP overexpression
GSAP (anti-HA)
NotchE (anti-Myc)
NICD (anti-Myc)
NICD (anti-Val-1744)
Figure 2 | GSAP regulates amyloid-b production but does not influence

Notch cleavage. a, siRNA-mediated knockdown of GSAP in N2a cells
overexpressing APP695 lowered amyloid-b production. The amyloid-blowering effects of imatinib and of siRNA were not additive (mean 6 s.d.;
**P , 0.01; n $ 3). b, Transfection of N2a cells overexpressing APP695 with
GSAP siRNA reduced the concentrations of Ab38, Ab40 and Ab42
(mean 6 s.d.; **P , 0.01; n $ 3). c, Recombinant GSAP-16K from E. coli
stimulated amyloid-b production in an in vitro c-secretase assay, inhibited
AICD production and had no effect on Notch cleavage. The c-secretase
inhibitor L-685,458 (1 mM) abolished NICD, AICD and amyloid-b
production (mean 6 s.d.; **P , 0.01; n $ 3). d, Neither GSAP knockdown
(left) nor its overexpression (right) affected Notch processing in HEK293
cells overexpressing Notch without its extracellular domain (NotchDE, with
C-terminal Myc tag). NICD was detected using a Myc antibody and a
cleavage-product-specific antibody (Notch1 Val-1744). The c-secretase
inhibitor L-685,458 (1 mM) served as a control. a.u., arbitrary units.
of the c-secretase cleavage product, the Notch intracellular domain

(NICD), was not changed either by reducing GSAP concentrations
using short hairpin RNA (Fig. 2d, left panel) or by overexpressing
GSAP (Fig. 2d, right panel). In addition, GSAP had no effect on
Notch cleavage in an in vitro c-secretase assay (Fig. 2c, left panel).
Thus, GSAP modulates the c-secretase cleavage of APP but not of
Notch.
Additional evidence that endogenous GSAP forms a complex with
c-secretase was provided by examining the distribution of the proteins
in subcellular fractions and in co-immunoprecipitation studies. In
a sucrose gradient, endogenous GSAP co-fractionated with a transGolgi network marker, and with PS1-CTF (Supplementary Fig. 8) and
other c-secretase components (not shown). Using gel filtration to
separate membrane proteins from neuroblastoma cells solubilized
in 1% CHAPSO, we found that endogenous GSAP-16K and c-secretase
co-migrated as a high-molecular-weight complex (Fig. 3a). Furthermore, endogenous GSAP co-immunoprecipitated with c-secretase
components, providing additional evidence that these proteins exist
in a complex (Fig. 3b). Endogenous c-secretase was isolated using
an immobilized biotinylated derivative of the transition-state analogue L-685,4586. Endogenous GSAP-16K co-isolated with the
96
LETTERS
10
11 Fraction no.
GSAP-16K
Nicastrin
PS1-CTF
PS1-NTF
PEN2
PEN2
Input
ntr
ol
IgG
AP
PCT
FA
No
b
tch
E
Ab
Input
Co
ntr
o
GS l IgG
AP
Ab
GSAP-16K
IP
Co
be
ad
s
be
ad
s
tin
b
IP
GS
I
Bi
o
Input
NotchE
APP--CTF
APP--CTF
GSAP-16K
Nicastrin
PS1-CTF
IP
Control GSAP
Ab
Input IgG
Amyloid- level
(percentage of control)
APP-CTF
APP-CTF
bound to GSAP (%)
PEN2
Calnexin
100
80
60
40
20
0
120
100
80
60
40
20
0
+
+
1,500
**
1,000
500
0
8,000
6,000
4,000
**
2,000
0
AD 2 mice without induction
Nicastrin
2,000
A42 (pg per mg

brain weight)
A40 (pg per mg

brain weight)
IP
Control GSAP
Input IgG
Ab
Plaque burden
(per mm2)
158 kDa
+
+
Inactive imatinib
AD 2 mice with induction
300
250
200
150
100
50
0
GSAP RNAi-AD mice without induction

GSAP RNAi-AD mice with induction
**
AD 2 mice
GSAP RNAi-AD mice
Without induction
670 kDa
Imatinib
0
10 20 30 40 50
Compound concentration (M)
APP-CTF
Amyloid-
GSAP-16K (10 g ml1)
L-685,458 (1 M)
Figure 3 | GSAP interacts with c-secretase and APP-CTF but not with
Notch. a, Endogenous GSAP-16K in solubilized membrane preparations
from N2a cells co-migrated with c-secretase components during gel
filtration (void volume: fraction 6). b, Immunoprecipitation of endogenous
GSAP from N2a cells resulted in co-immunoprecipitation of c-secretase
components. c, Endogenous GSAP-16K and c-secretase components are
highly enriched by an immobilized c-secretase transition-state analogue
(GSI beads). d, In HEK293 cells, GSAP-16K and APP-CTF coimmunoprecipitated but NotchDE did not. e, Imatinib treatment reduced
the co-immunoprecipitation of APP-CTF and GSAP in a concentrationdependent manner. An inactive imatinib derivative (IC200001; see
Supplementary Fig. 3) served as a negative control. f, In HEK293 cells, APPCTF without the cytoplasmic domain (APPe-CTF) did not coimmunoprecipitate with GSAP-16K (top); c-cleavage of APPe-CTF was not
stimulated by GSAP-16K in an in vitro assay (bottom).
enzymeinhibitor complex, strongly suggesting that GSAP-16K is a

cofactor for c-secretase (Fig. 3c).
A number of proteases with broad substrate recognition can achieve
specificity through auxiliary factors that couple the core enzyme to
selective substrates7,8. To explore the mechanism by which GSAP
might confer such specificity, we analysed its association with specific
substrates. GSAP-16K co-immunoprecipitated with APP-CTF but
not with NotchDE (Fig. 3d); the interaction was reduced by imatinib
in a concentration-dependent manner (Fig. 3e). Disruption of this
interaction by imatinib probably explains its amyloid-b-lowering
activity. Domain mapping studies demonstrated that the juxtamembrane region of APP-CTF interacts with GSAP (Supplementary
Fig. 9). A truncated form of APP-CTF lacking the cytoplasmic domain
(APPe-CTF)9 did not interact with GSAP and its c-cleavage was no
longer stimulated by GSAP-16K in an in vitro assay (Fig. 3f).
To further determine the structural basis for the selective interaction of GSAP with APP-CTF, we constructed chimaeric proteins by
exchanging the AICD fragment in APP-CTF with the NICD fragment
in NotchDE (Supplementary Fig. 10a). GSAP selectively interacted
with AICD but not with NICD in chimaeric proteins (Supplementary
With induction
Figure 4 | Knockdown of GSAP reduces amyloid-b production and plaque

development in a mouse model of Alzheimers disease. a, GSAP RNAi-AD
mice were generated by crossing AD 32 mice with doxycycline-inducible GSAP
RNAi mice. Six months after inducing GSAP shRNA expression, Ab40 and
Ab42 concentrations in the crossed-mouse brains had decreased by 42 6 13%
and 40 6 7%, respectively (mean 6 s.e.m.; **P , 0.01; n 5 4). Doxycycline
treatment alone did not change amyloid-b concentrations in AD mice. b, Six
months after inducing GSAP shRNA expression, amyloid plaque development
had reduced in the crossed-mouse brains by 38 6 9% (mean 6 s.e.m.;
**P , 0.01; n 5 4). Doxycycline treatment alone did not change plaques in AD
mice. Amyloid plaques were revealed by 6E10 immunostaining.
Fig. 10b). GSAP knockdown selectively increased AICD production

but had no influence on NICD production from the chimaeric
proteins (Supplementary Fig. 10c). These results further demonstrated that the selective effect of GSAP on APP-CTF cleavage by
c-secretase involves GSAP binding to the cytoplasmic domain of
the substrate.
To determine whether our findings are relevant to Alzheimers
disease pathology, we examined the effects of GSAP on amyloid-b
levels and plaque development in vivo. A conditional GSAP RNA interference mouse line was generated by integration of a tetracyclineinducible GSAP shRNA vector into the mouse genomic locus.
GSAP RNAi mice were then crossed with a mouse model of
Alzheimers disease (APPswe and PS1DE9 mutations; doubletransgenic Alzheimers disease (AD 32) mice)10. To examine the
long-term effect of GSAP knockdown on amyloid-b concentrations
andplaquedevelopment,thecrossedGSAPRNAi-AD32micewerecontinuously induced for six months. After induction, GSAP messenger
RNA concentrations in these hybrid mouse brains were reduced by
85 6 12% and similar decreases were achieved in other tissues; after
six months of induction, Ab40 and Ab42 concentrations in the crossed
mouse brains were lowered by 42 6 13% and 40 6 7%, respectively
(Fig. 4a). Amyloid plaque load in crossed mouse brains with GSAP
knockdown was reduced by 38 6 9% relative to the same line of mouse
brains without induction (Fig. 4b). Doxycycline did not have an effect
97
LETTERS
on either amyloid-b or plaque concentrations in AD 32 mice. The

amyloid-b-lowering effects of GSAP knockdown are similar to those
caused by the c-secretase inhibitor dibenzazepine11 (DBZ), administered
at 10 mmol kg21 for five days (Supplementary Fig. 11a). In contrast,
GSAP knockdown did not cause the intestinal mucosal cell metaplasia
seen with DBZ (Supplementary Fig. 11b): the latter effect is mediated by
impaired Notch processing4,11. Furthermore, GSAP knockdown did not
cause any pathological changes in spleen (data not shown), whereas
severe marginal-zone lymphoid depletion was caused by DBZ administration12. These results indicate that GSAP knockdown reduces
amyloid-b concentrations and plaque formation without affecting
Notch-dependent pathways.
Gamma-secretase processes diverse substrates with low homologies
at their cleavage sites13. The various roles of c-secretase during development and in tissue homeostasis require that its activity be tightly regulated. TMP2114 (also known as TMED10), orphan G-protein-coupled
receptor 315 and different Aph1 isoforms16 have been reported to modulate amyloid-b production through c-secretase but to spare Notch
cleavage. However, the underlying molecular mechanisms by which
they impart their specificities were not elucidated in those studies.
Nevertheless, the studies demonstrated that it is possible to selectively
regulate substrate specificity of this vitally important and potentially
promiscuous enzyme. GSAP seems to confer substrate specificity on
c-secretase by forming a ternary complex with c-secretase and the
substrate APP-CTF. The present results support the concept that
appropriate cofactors impart substrate specificity on the c-secretase
core enzyme complex, as they do on a number of other proteases7,8.
The literature on the relationship between c- and e-cleavage of
APP-CTF is controversial. For instance, there is some evidence supporting sequential cleavage of APP-CTF9,17. There is also extensive
evidence reported in the literature that these two types of cleavage can
occur independently1820. Our data support the second hypothesis.
We propose that removal of GSAP from the GSAP/c-secretase/APPCTF ternary complex alters the structural relationship between
c-secretase and APP-CTF, facilitating e-cleavage at the expense of
c-cleavage (Supplementary Fig. 1). To elucidate the detailed mechanism by which GSAP modulates the cleavage of APP-CTF, it will be
important to compare the stoichiometry of the various c-secretase
cleavage products in the presence and absence of GSAP.
Anti-amyloid therapy remains a rational approach to the treatment
of Alzheimers disease. One promising anti-amyloid compound failed
in limited clinical trials, owing to lack of accumulation in the brain21.
Similarly, imatinib is actively excluded from the brain by a highly
potent P-glycoprotein pump, a component of the bloodbrain
barrier22. The development of compounds that accumulate in the
brain and target GSAP represents a valid approach for development
of potential therapies against Alzheimers disease.
METHODS SUMMARY
See Methods for details of in vitro and intact-cell photolabelling, affinity purification using immobilized biotin-imatinib, GSAP knockdown and overexpression, co-immunoprecipitation, gel filtration chromatography, affinity capture of
endogenous c-secretase using an immobilized transition-state analogue, in vitro
c-secretase assay, GSAP RNAi mouse line generation, induction, immunohistochemistry and amyloid-b measurements.
Full Methods and any associated references are available in the online version of
the paper at www.nature.com/nature.
Received 13 September 2009; accepted 5 July 2010.
1.
Selkoe, D. J. Alzheimers disease: genes, proteins, and therapy. Physiol. Rev. 81,
741766 (2001).
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
Steiner, H., Fluhrer, R. & Haass, C. Intramembrane proteolysis by gammasecretase. J. Biol. Chem. 283, 2962729631 (2008).
Lathia, J. D., Mattson, M. P. & Cheng, A. Notch: from neural development to
neurological disorders. J. Neurochem. 107, 14711481 (2008).
Wong, G. T. et al. Chronic treatment with the gamma-secretase inhibitor LY411,575 inhibits beta-amyloid peptide production and alters lymphopoiesis and
intestinal cell differentiation. J. Biol. Chem. 279, 1287612882 (2004).
Netzer, W. J. et al. Gleevec inhibits beta-amyloid production but not Notch
cleavage. Proc. Natl Acad. Sci. USA 100, 1244412449 (2003).
Placanica, L. et al. Pen2 and presenilin-1 modulate the dynamic equilibrium of
presenilin-1 and presenilin-2 gamma-secretase complexes. J. Biol. Chem. 284,
29672977 (2009).
Dougan, D. A., Mogk, A., Zeth, K., Turgay, K. & Bukau, B. AAA1 proteins and
substrate recognition, it all depends on their partner in crime. FEBS Lett. 529, 610
(2002).
Visintin, R., Prinz, S. & Amon, A. CDC20 and CDH1: a family of substrate-specific
activators of APC-dependent proteolysis. Science 278, 460463 (1997).
Lefranc-Jullien, S., Sunyach, C. & Checler, F. APPe, the e-secretase-derived
N-terminal product of the b-amyloid precursor protein, behaves as a type I
protein and undergoes a-, b-, and c-secretase cleavages. J. Neurochem. 97,
807817 (2006).
Jankowsky, J. L. et al. Co-expression of multiple transgenes in mouse CNS: a
comparison of strategies. Biomol. Eng. 17, 157165 (2001).
van Es, J. H. et al. Notch/gamma-secretase inhibition turns proliferative cells in
intestinal crypts and adenomas into goblet cells. Nature 435, 959963 (2005).
Milano, J. et al. Modulation of notch processing by gamma-secretase inhibitors
causes intestinal goblet cell metaplasia and induction of genes known to specify
gut secretory lineage differentiation. Toxicol. Sci. 82, 341358 (2004).
Beel, A. J. & Sanders, C. R. Substrate specificity of gamma-secretase and other
intramembrane proteases. Cell. Mol. Life Sci. 65, 13111334 (2008).
Chen, F. et al. TMP21 is a presenilin complex component that modulates gammasecretase but not epsilon-secretase activity. Nature 440, 12081212 (2006).
Thathiah, A. et al. The orphan G protein-coupled receptor 3 modulates amyloidbeta peptide generation in neurons. Science 323, 946951 (2009).
Serneels, L. et al. c-secretase heterogeneity in the Aph1 subunit: relevance for
Alzheimers disease. Science 324, 639642 (2009).
Takami, M. et al. c-secretase: successive tripeptide and tetrapeptide release from
the transmembrane domain of beta-carboxyl terminal fragment. J. Neurosci. 29,
1304213052 (2009).
Kume, H. & Kametani, F. Abeta 1140/42 production without gamma-secretase
epsilon-site cleavage. Biochem. Biophys. Res. Commun. 349, 13561360 (2006).
Wiley, J. C., Hudson, M., Kanning, K. C., Schecterson, L. C. & Bothwell, M. Familial
Alzheimers disease mutations inhibit gamma-secretase-mediated liberation of
beta-amyloid precursor protein carboxy-terminal fragment. J. Neurochem. 94,
11891201 (2005).
Bentahir, M. et al. Presenilin clinical mutations can affect gamma-secretase
activity by different mechanisms. J. Neurochem. 96, 732742 (2006).
Green, R. C. et al. Effect of tarenflurbil on cognitive decline and activities of daily
living in patients with mild Alzheimer disease: a randomized controlled trial. J. Am.
Med. Assoc. 302, 25572564 (2009).
Dai, H., Marbach, P., Lemaire, M., Hayes, M. & Elmquist, W. F. Distribution of STI571 to the brain is limited by P-glycoprotein-mediated efflux. J. Pharmacol. Exp.
Ther. 304, 10851092 (2003).
Supplementary Information is linked to the online version of the paper at

www.nature.com/nature.
Acknowledgements We thank E. Woo and B. Chait for their help with protein
identification. We thank Y. M. Li for providing us with the biotinylated
transition-state analogue. We thank B. Turner and S. Ku for their technical support.
This work was supported by NIH grant AG09464 to P.G., DOD grant
W81XWH-09-1-0402 to P.G., the Fisher Center for Alzheimers Research
Foundation and the F. M. Kirby Foundation.
Author Contributions G.H., W.L., P.L., C.R., J.H. and K.B. performed experiments;
W.J.N. was involved in experimental design; M.F. performed sequence analysis;
G.H., W.L., L.P.W. and P.G. designed the study; G.H., W.L., F.G., L.P.W. and P.G.
wrote the paper; all authors discussed the results and commented on the
manuscript.
Author Information Reprints and permissions information is available at
www.nature.com/reprints. The authors declare competing financial interests:
details accompany the full-text HTML version of the paper at www.nature.com/
nature. Readers are welcome to comment on the online version of this article at
www.nature.com/nature. Correspondence and requests for materials should be
addressed to P.G. (greengard@rockefeller.edu).
98
doi:10.1038/nature09325
METHODS
In vitro and intact-cell photolabelling. For in vitro labelling, resuspended membranes isolated from HEK293 cells were incubated with 20 nM 125I-G01 for 3 h at
4 uC before photolysis at 254 nM for 2 min. For intact-cell labelling, HEK293 cells
were incubated with 0.1 mM 3H-G01 in Opti-MEM for 2 h at 37 uC before being
transferred to ice for an additional hour. To examine labelling specificity, either
membrane preparations or cells were treated with 50 mM unlabelled imatinib
together with photoactivatable G01 in parallel assays. Photolysis was conducted
on ice for 2 min at 254 nm. After photolysis, membranes or cells were disrupted in
lysis buffer (50 mM HEPES, 150 mM NaCl, 1% CHAPSO with protease inhibitors)
and immunoprecipitated with PS1-loop antibody. The immunopurified material
was eluted with SDS sample buffer and proteins were separated using a 1020%
Tris-tricine SDSPAGE gel and transferred to PVDF for autoradiography.
Affinity purification using immobilized biotin-imatinib. Membrane preparations of HEK293 cells were solubilized in lysis buffer (50 mM HEPES, 150 mM
NaCl, 5 mM MgCl2, 5 mM CaCl2 and 1% CHAPSO containing protease inhibitors; Roche) and incubated with MyOne streptavidin T1 beads (Invitrogen) containing bound biotin-imatinib for 3 h at 4 uC. Subsequently, the beads were
washed three times with lysis buffer. Bound proteins were eluted with tricine
SDSPAGE sample buffer and separated on 1020% Tris-tricine gels. Silver staining was used to identify protein bands in SDSPAGE gels. The ,16-kDa band was
excised, trypsinized and sequenced by tandem MS/MS mass spectrometry.
GSAP antibody production and metabolic labelling. Rabbit polyclonal antiserum against GSAP was generated against the peptide CFEGHDNVDAEFV
EEAALKHT (corresponding to amino acids 829848 of human GSAP with an
amino-terminal cysteine attached for conjugation). Pulsechase labelling experiments using neuroblastoma 2a (N2a) cells were conducted as described previously23. Cells were pulsed for 15 min, the chase periods were initiated by replacing
the medium with full culture medium and cells were incubated at 37 uC. For
continuous labelling, cells were labelled with 35S Protein Labelling Mix (Perkin
Elmer) for 4 h without chase. Cell monolayers were lysed in RIPA buffer followed
by immunoprecipitation with GSAP antibody. The beads were incubated with
Tris-tricine sample buffer to elute bound proteins, which were then separated by
1020% Tris-tricine gel and transferred to PVDF membrane for autoradiography.
Cellular knockdown and overexpression. For cellular GSAP knockdown
experiments, siRNA of GSAP was purchased from Dharmacon. The sequences
of the siRNA used were as follows: sense sequence, 59-AUGCAGAGCUGGACG
ACAUUU-39; antisense sequence, 59-PAUGUCGUCCAGCUCUGCAUUU-39.
An N2a cell line stably overexpressing APP695 was transfected with siRNA using
DharmaFect 2 reagent at a concentration of 50 nM. Non-targeting control
siRNA (Dharmacon) was transfected in parallel as control. shRNA of GSAP
was purchased from Open Biosystems and transfected into cells using ArrestIn transfection reagent (Open Biosystems). The sequence of human GSAP
shRNA in pGIPZ shRNAmir-GFP vector was as follows: TGCTGTTGACAGTG
AGCGCGGAAATAGAGTGGTGATTAAATAGTGAAGCCACAGATGTATTT
AATCACCACTCTATTTCCATGCCTACTGCCTCGGA. The knockdown efficiencies were examined using a real-time PCR assay with an Applied Biosystems 7900 HT System.
For GSAP overexpression in cells, mammalian expression vector pReceiverM07 with the full-length GSAP coding a C-terminal HA tag was purchased from
Genecopoeia. Plasmid was transfected into a stable HEK293 cell line overexpressing APP695, containing the Swedish mutation, using Lipofectamine 2000
(Invitrogen). pcDNA4-APP-b-CTF expression vector was a kind gift from Dr.
Y. M. Li (Memorial Sloan Kettering Cancer Center). APPe-CTF construct was
derived from the pcDNA4-APP-b-CTF as reported previously9.
The concentrations of amyloid-b species were quantified using a highly sensitive ECL assay from Meso Scale Drug Discoveries. Immunoprecipitation of
amyloid-b was performed as described prevously5.
For Notch cleavage analysis, cells transfected with NotchDE5 were co-transfected
with GSAP-shRNA or GSAP plasmids. After two days of transfection, Notch
expression and cleavage were detected with anti-Myc antibody. The cleaved
Notch intracellular domain (NICD) was detected with a cleavage-specific antibody
(Notch1 Val-1744, Cell Signaling). Cells treated with L-685,458 served as controls.
Co-immunoprecipitation. For co-immunoprecipitation, cells were lysed in
50 mM HEPES, 150 mM NaCl, 5 mM MgCl2, 5 mM CaCl2 and 1% CHAPSO, with
protease inhibitors. Immunoprecipitation was performed using the corresponding

antibody and protein G plus/protein A beads for 2 h on ice. Immunoprecipitated
proteins were resolved by SDSPAGE and analysed by immunoblots. Presenilin 1
loop antibody (EMD Biosciences) was used to detect PS1-CTF; PEN2 antibody was
purchased from EMD Biosciences. Nicastrin antibody was from BD Biosciences.
HA monoclonal antibody and Myc tag polyclonal antibody were from Genscript.
APP-CTF was detected using the 369 antibody24. 6E10 and 4G8 antibodies from
Covance were used to detect amyloid-b.
Gel filtration chromatography. Solubilized membrane preparations from N2a
cells (0.2 ml, ,1 mg of solubilized protein, in 50 mM HEPES, 150 mM NaCl, 1%
CHAPSO, 5 mM MgCl2, 5 mM CaCl2) were applied to a Superdex 200 10/300 GL
column (GE Healthcare) of an AKTA fast performance liquid chromatography
system. Fractionation was performed in the lysate buffer at a flow rate of
0.5 ml min21 and 1-ml fractions were collected. Endogenous GSAP was detected
after immunoprecipitated with GSAP antibody. Each fraction was analysed by
immunoblot using c-secretase antibodies.
Affinity capture of endogenous c-secretase using an immobilized transitionstate analogue. Compound 4, a biotinylated c-secretase transition-state analogue6, was a kind gift from Dr. Y. M. Li. HEK293 cell lysates in 50 mM HEPES,
150 mM NaCl, 1% CHAPSO, 5 mM MgCl2 and 5 mM CaCl2 were incubated with
compound 4 immobilized on MyOne streptavidin magnetic beads for 2 h at 4 uC.
The beads were then washed three times with lysate buffer. The captured proteins
were eluted with SDS sample buffer, separated by SDSPAGE and processed for
immunoblot analysis.
In vitro c-secretase assay. The in vitro c-secretase assay was based on that
described previously25 except for the use of APP-CTF or NotchDE overexpressed
in HEK293 cells rather than recombinant proteins from E. coli. Recombinant
GSAP-16K (amino acids 733854 of the human GSAP) was expressed in BL21
DE3 E. coli and purified. After 2 h of 35S labelling, membrane preparations from
HEK293 cells overexpressing APP-CTF were resuspended in 200 ml of assay
buffer with 2 mg recombinant GSAP-16K or the same amount of BSA as control.
A parallel system with 1 mM L-685,458 (c-secretase inhibitor) was also used as a
control. The membrane suspension was pre-incubated at 4 uC for 1 h and then
incubated for 2 h at 37 uC to allow in vitro generation of amyloid-b. Amyloid-b
was immunoprecipitated from the lysate using 4G8 antibody, separated on 1020%
Tris-tricine gel and transferred to PVDF membrane for autoradiography.
GSAP RNAi mouse line generation. Inducible RNAi mice were generated by
incorporating GSAP shRNA 59-TCCCGGAACTCCATGATTGACAAATTTC
AAGAGAATTTGTCAATCATGGAGTTCCTTTTTA-39 into the mouse genome
(B6/129S6 background) under the control of a H1-Tet promoter, as described
previously26 (TaconicArtemis). Heterozygous RNAi mice were then crossed with
an Alzheimers disease mouse model with APPswe and PS1D9 mutations (AD
32 mice) to generate GSAP-RNAi AD mice for amyloid-b analyses.
Induction of GSAP RNAi-AD mouse, measurement of amyloid-b concentration and histochemical analysis. shRNA was induced in two-month-old GSAP
RNAi-AD mice with doxycycline for one month by adding 2 mg ml21 doxycycline (Sigma) into drinking water containing 10% sucrose. Control mice were
fed drinking water containing 10% sucrose. GSAP knockdown efficiency in mice
was assayed using real-time PCR. Amyloid-b concentrations from mouse hippocampus were measured by ELISA (Wako Chemicals). Intestinal dissection and
histochemical staining (H&E and PAS staining) were conducted as described
elsewhere11.
For immunohistochemistry studies, mouse brains were processed and labelled
with the anti-amyloid-b antibody 6E10 (Novus Biologicals) to visualize extracellular amyloid plaques using an MOM immunodetection kit (Vector Laboratories).
23. Wang, H. et al. Presenilins and gamma-secretase inhibitors affect intracellular
trafficking and cell surface localization of the gamma-secretase complex
components. J. Biol. Chem. 279, 4056040566 (2004).
24. Xu, H. et al. Estrogen reduces neuronal generation of Alzheimer beta-amyloid
peptides. Nature Med. 4, 447451 (1998).
25. Li, Y. M. et al. Presenilin 1 is linked with gamma-secretase activity in the detergent
solubilized state. Proc. Natl Acad. Sci. USA 97, 61386143 (2000).
26. Seibler, J. et al. Reversible gene knockdown in mice using a tight, inducible shRNA
expression system. Nucleic Acids Res. 35, e54 (2007).
doi: 10.1038/nature09325
SUPPLEMENTARY INFORMATION
Supplementary Information
Methods
Materials: 2,5-dioxopyrrolidin-1-yl 4-azido-2-hydroxybenzoate (NHS-ASA) was
purchased from ProChem. Inc (Rockford, IL). 6-Methyl-N1-(4-(pyridin-3-yl)pyrimidin2-yl)benzene-1,3-diamine and N-(4-methyl-3-(4-(pyridin-3-yl)pyrimidin-2ylamino)phenyl)-4-(piperazin-1-ylmethyl)benzamide (N-desmethyl imatinib) were
purchased from ChemPacific Inc (Baltimore, MD). 2,5-dioxopyrrolidin-1-yl 5((3aS,4S,6aR)-2-oxo-hexahydro-1H-thieno[3,4-d]imidazol-4-yl)pentanoate (Biotin-OSu),
N-(chloro(dimethylamino)methylene)-N-methylmethanaminium hexafluorophosphate
(TCFH), trifluoroacetic acid (TFA), 1H-benzo[d][1,2,3]triazol-1-ol (HOBt) and N,Ndiisopropylethyl amine (DIPEA) were purchased from Sigma-Aldrich (St. Louis, MO).
Tert-butyl 2-(piperazin-1-yl)ethylcarbamate was purchased from Astatech Inc (Bristol,
PA).
Synthesis of an imatinib derived photo-affinity label, G01: DIEPA (63 l, 0.36 mmol)
was added to a solution of NHS-ASA (50 mg, 0.18 mmol), HOBt (25 mg, 0.18 mmol),
and 6-Methyl-N1-(4-(pyridin-3-yl)pyrimidin-2-yl)benzene-1,3-diamine (50 mg, 0.18
mmol) in DMF (2 ml). The reaction mixture was stirred at room temperature overnight
under argon atmosphere. The generated crude product was purified by a semi-preparative
HPLC to give 54 mg of the titled compound with a yield of 68%. The product, G01, 4azido-2-hydroxy-N-(4-methyl-3-(4-(pyridin-3-yl)pyrimidin-2-ylamino)phenyl)benzamide,
was confirmed by mass spectral analysis using an ESI-MS in the positive mode [M+H]+,
demonstrating a m/z of 439.1.
1
www.nature.com/nature
doi: 10.1038/nature09325
Radioiodination of G01 by 125I was performed without carrier using a modification of a

Chloramine-T procedure and the iodinated product was purified by HPLC. Specifically,
in a UV protected "V" vial, total volume 0.9 ml, ~10 mCi of 125I stock isotope (volume =
25 l) was added to 200 l of 0.2M phosphate buffer, pH 7.2. G01 was dissolved to 1
mg/ml in ethanol and 25 l of this solution was combined with chloramine-T at 1 mg/mL
in water (50 l) and then added to the V-vial. The reaction proceeded for 1 min and was
terminated by the addition of 50 L of 1 mg/ml meta-bisulfite. The reaction mixture was
chromatographed on a 25 cm Waters RP-C18 column, using 0.1% TFA in water as the
"A" solvent and 0.1% TFA in acetonitrile as the "B" solvent. A gradient was run at 1
ml/min from 0% B to 50% B for 45 minutes and held at 50% B for 15 minutes. The
product demonstrated a retention time of 54.5 min as followed by radiochemical
detection, and had a specific activity of 2000 Curies per millimole. The I125 labeling
experiment was performed by PerkinElmer Life and Analytical Sciences, Inc.
3
H-G01 was prepared by ViTrax Radiochemicals via catalytic tritium exchange of G01.
The labeled product was purified by HPLC. The composition of the purified product was
verified by co-injection of the tritium labeled product with its cold precursor and both
compounds co-chromatographed on an analytical HPLC.
Cellular A production assays and incubation with G01.
Neuroblastoma 2a cells stably overexpressing human APP695 were treated with 10 M
G01 for 3 hr. Cells treated with DMSO, or DMSO plus imatinib, were used as controls.
After 3 hr, conditioned medium was collected and A immunoprecipitation was
conducted using 4G8 antibody. The immunoprecipitated A was separated on 10-20%
Tris-tricine gel, transferred to PVDF membrane and detected by 6E10 antibody.
2
doi: 10.1038/nature09325
Synthesis and kinase profiling of biotin-imatinib (active and inactive form): Inactive
biotin-imatinib, (IC200001) was synthesized by reacting N-desmethyl imatinib with
Biotin-OSu. Active biotin-imatinib, (IC339239) was synthesized from the key
intermediates, tert-butyl 2-(piperazin-1-yl)ethylcarbamate and 6-methyl-N1-(4-(pyridin-3yl)pyrimidin-2-yl)benzene-1,3-diamine, via 4 steps, as shown in supplementary figure 1.
The kinase profiling was performed by Millipore Inc. using the standard assays
for Abl kinase and PDGF receptor (ATP = 45 M). Compound IC200001 showed no
significant inhibitory activity toward either kinase, while compound IC339239 had an
IC50 of 146 nM against Abl kinase (imatinib had an IC50 of 79 nM) and an IC50 of 6.6
M against PDGF receptor (imatinib had an IC50 of 4.8 M). Thus, we refer to
IC200001 as inactive biotin-imatinib and IC339239 as active biotin-imatinib.
Construction of APP-CTF truncated forms
APP-CTF-T1 is the truncated form of APP--CTF spanning from its N-terminus to
HHGV64; APP-CTF-T2 is the truncated form of APP--CTF spanning from its Nterminus to VMLKK55. Both truncated forms were generated by introducing a stop
codon in related positions of the CT100 (APP-CTF) in pcDNA4. Mutagenesis was
performed using QuickChange Site-Directed Mutagenesis kit (Stratagene) according to
the manufacturer's instructions. The primers used for APP-CTF-T1 are: Forward: 5'
CATTCATCATGGTGTGTAGGAGGTTGACGCCGC 3'. Reverse: 5'
GCGGCGTCAACCTCCTACACACCATGATGAATG 3'. The primers used for APPCTF-T2 are: Forward: 5'
CTTGGTGATGCTGAAGAAGTAACAGTACACATCCATTC 3' Reverse: 5'
3
doi: 10.1038/nature09325
GAATGGATGTGTACTGTTACTTCTTCAGCATCACCAAG 3'. The presence of the

stop codon and integrity of the cDNA were verified by sequencing.
Construction, expression, and analysis of APP/NICD and NotchE/AICD
APP/NICD and NotchE coding sequences were synthesized by Genscript Inc. and are
illustrated in supplementary Fig. 10. Both sequences were then incorporated into a
pcDNA3.1 vector coding for a C-terminal Myc tag. APP-CTF/NICD, APP-CTF, and
NotchE/AICD were overexpressed separately in HEK293 cells. Immunoprecipitation
was conducted using gSAP antibody. APP/NICD and NotchE/AICD were detected by
c-myc antibody. APP-CTF was detected by 369 antibody.
APP-CTF/NICD, APP-CTF, and NotchE/AICD were transfected into N2a cells
with/without gSAP siRNA knockdown. NICD was detected by both myc antibody and
cleavage specific Val1844 antibody (Cell Signaling); AICD was detected by myc
antibody. AICD production from APP-CTF was detected by 369 antibody.
Mouse administration of a -secretase inhibitor dibenzazepine (DBZ)
gSAP RNAi mice (6 months old) were administered dibenzazepine (DBZ) (10 mol/kg)
once daily for 5 days by intraperitoneal injection. DBZ was suspended finely in 0.5%
(w/v) hydroxypropylmethylcellulose and 0.1% (w/v) Tween 80. Mice were treated with
DBZ or with vehicle, with 4 mice in each group. After sacrificing, mouse brain was
removed for A ELISA assays (Invitrogen); Mouse intestine was processed for H&E and
PAS staining.
doi: 10.1038/nature09325
Supplementary Figure 1
Supplementary Figure Legends

Supplementary Figure 1. gSAP action on APP processing. Ternary complex of gSAP,
APP and -secretase (top) is associated with elevated -cleavage (A production) and
reduced -cleavage (AICD production). In the absence of gSAP (bottom), the binary
complex of APP and -secretase is associated with decreased -cleavage and increased cleavage.
Supplementary Figure 2. a: structures of imatinib, G01, and 125I-G01. b: G01
significantly reduces levels of A in N2a cells.
Supplementary Figure 3. a: Procedures for the synthesis of biotinylated imatinib
derivative, IC339239: reagents and conditions: (a) 4-(bromomethyl) benzoic acid, K2CO3,
DMF, room temperature, 2 h. (b) 6-methyl-N1-(4-(pyridin-3-yl)pyrimidin-2-yl)benzene1,3-diamine, TCFH, DIPEA, DMF, room temperature, overnight. (c) TFA, CH2Cl2, room
temperature, 30 min. (d)Biotin-OSu, HOBt, DIPEA, room temperature, overnight, and
then HPLC purification. Compound IC200001 was synthesized by reacting N-desmethyl
imatinib with Biotin-OSu.
b: Kinase profiling results shows that IC339239 has activities comparable to those of
imatinib, while IC200001 showed no activity. Therefore, IC339239 is designated as
active biotin-imatinib and IC200001 as inactive biotin-imatinib.
Supplementary Figure 4. Alignments of gSAP sequences among species. Red: identical
doi: 10.1038/nature09325

Supplementary Figure 4. Alignments9of gSAP sequences among species. Red: identical
residues; Blue/Green: conservative substitutions. The C-terminal region of gSAP is

highly conserved. The gSAP-16K region is underlined.
doi: 10.1038/nature09325
a
N
NH
(a)
N
H
OH
H
N
H2N
N
N
H
N
H
N
N
H
N
H
(c)
(b)
H
N
N
N
H
N
(d)
O
NH H
HN
HN
N
O
S
H
N
N
N
IC339239

reduced -cleavage (AICD production). In the absence
of gSAP (bottom), the binary
O
HN
NH
complex of APP
-cleavage
N and -secretase is associated with
N and increased H decreased
H
H
N
N
O
NH H
HN
H
cleavage.
HN
H
N
H
N
H
N
O
Supplementary
Figure 2. a: structures of imatinib, G01, and 125I-G01. b: G01
IC339239: active
IC200001: inactive
N
N

10
doi: 10.1038/nature09325
Homo
Canis
Bos
Mus
Rattus
Gallus
---------------------------------------MALRLVADFDLGKDVLPWLRA
--------------------------------------------------MTQNLSWPGH
---------------------------------------MALRLIADFDLEKDVLPWLRV
---------------------------------------MALRLVTHFDVLEDVLPSLLT
---------------------------------------MALRLVTHFDVLADVLPSLLV
MAVAAPQQPARCGGQRPPECGRVGPRLRALPSGGRRSQAGRESPRAAHGAASPLLPSGPG
Homo
Canis
Bos
Mus
Rattus
Gallus
QRAVSEASGAGSG----------------------------------GADVLENDYES-SSNGSCIRFPVAG----------------------------------GTAVL-------QLAASAAAGARGG----------------------------------GPGVLENNYEC-QAATTDEGDRAGV----------------------------------LETTYG----S-QAATADEGDEGA------------------------------------ETTLG----S-RLEATGGRGNGGGASGRPQLRGLSPPAPLPCGGCAGPELRGLTLSLCGGSALDTSEKSSA
Homo
Canis
Bos
Mus
Rattus
Gallus
LHVLNVERNGNIIYTYKDDKGNVVFGLYDCQTRQNELLYTFEKDLQVFSCSVNSERTLLA
---------------------------WQGAVSSIQGLGTADHELPTWRAYEQLPYADLA
LRVLNVERNRNIIYTYKDNKGNVFFGLYDYQTKQNEHLYTFEKDLQVVSCSVNKEKTLLA
LRVLNIERNGNIIYTYKDNKGNAVFGLYDCQTRQNEHLYTFEKDMQAVSCSVNSERTVLA
LRVLNIERNGDIIYTYKDNKGNAVFGIFDCQTRENEHLYTFEKDMQAVSCSVNSERTVLA
LYIVNVERNGKIIYTWKGNQRSTHIGLYDLQTKENEHLYTFEKDLRIISCSVNSERTLLA
Homo
Canis
Bos
Mus
Rattus
Gallus
ASLVQSTK-EGKRNELQPGSKCLTLLVEIHPVNNVKVLKAVDSYIWVQFLYPHIESHPLP
SVDLCIQL-LIPFAFIPTGSKCLTLLVEIHPVNNVKVLKAVDSSIWVQFLYPQVESHPPP
TSLVQAAK-EGRSNELQPGSKCLTLLVEIHPINNVKVLKAVDSYIWVQFLYPHVESCPQP
ASFIQYTT-EGVKNDLQPGSKCLTLLVEIHPVNNVKVLKAVDSCVWVQFLYPQAESHLLP
ASFIQYT--EGVRSELQPGSKCLTLLVEIHPVNNVTVLKAVDSCVWVQFLYPQAESHLLA
VSFRQYTEEERVTHLLQSVSKYLALLIEIHPINNVKVLKAVDSCVRVQFLYPVEDRNSST
Homo
Canis
Bos
Mus
Rattus
Gallus
ENHLLLISEEKYIEQFRIHVAQEDGNRVVIKNSGHLPRDRIAEDFVWAQWDMSEQRLYYI
ENHLLLISEEKYIEKFHIHVIQEDGNKVVLRDSGHLPRERVAEDFVWAQWDMSEQRLYYI
KNHLLLLSEEKYIEQFHIQVVQEDGNRVVIKNSGHLPRERIAEDFVWAQWDMSEQRLYYI
QNHLLLISEEKYIERFHIQITREDGDRVVIRNSSHLPRDRLAEDFVWAQWDLSEQRLYYI
QNHLLLISEEKYIERFHIQITREDGNRVVIRNSSHLPRERIAEDFVWAQWDVSEQRIHYI
ESHLLLVSEDKYIEQFDIHVAEEE-HRVVIQNSGQLPRARVADDLIWAQWDMTEQRLFYI
Homo
Canis
Bos
Mus
Rattus
Gallus
DLKKSRSILKCIQFYADESYNLMFEVPLDISLSNSGFKLVNFGCDYHQYRDKFSKHLTLC
VLKKSRSILKCIQFSANEKFNLMFEAPLDITLSASGFELVNFGCDDLQDQGNLSKHLTLC
DLKKSRSVLKCIQFYAEEHFNLMFEAPLDISLSDSGFKLVNFGYSDLQDKEELSEHLTLC
ELKESRSILKCIQFRADESFNLMFEMPLDITLTGLRFKLVNFGYDYRQDREKLCNQPSLC
ELQESRSILKCVQFWADESFTIMFEMPLDISLSGLRFKLVNFGYDYRQDQAKLCHQPSLC
VPKESRSILRCVQFYPDENFNSTLESQLDISVNDKRVKLVNFGYNDCEDRDVPPKSLNLQ
Homo
Canis
Bos
Mus
Rattus
Gallus
VFTNHTGSLCVCYSPKCASWGQITYSVFYIHKGHSKTFTTSLENVGSHMTKG-------VFTNHTGSLCVCYSPKFDSWEKITYSVFYFHKGHSKTFTAALGSVDSLVTKG-------VFTNHTGSLCVCYCPNFDSWEQITYSVFYFHKGHSKTFTTTLGSVDSHVTKG-------IFTNHTGSLCMCYSPKSDSREEITYSVFYLHKGYRKIFTAAPGSADSQVTNGADSQVTDG
IFTNHTGSLCVCYSPKSDSWKEITYSVFYLHKGYRKTFTVAPGSTDSQVANG-------VFTNKAG--------------------------FSKTFTASLERPETPQLKE--------
Homo
Canis
Bos
Mus
Rattus
ITFLNLDYYVAVYLPGHFFHLLNVQHPDLICHNLFLTGNNEMIDMLPHCPLQSLSGSLVL
LTFLNLDYYVAVYLPGHFFHLLNIQHPDLICHSLFLTGNNEVVDMLPHSPLQSLSGSLVL
ITFLNLDYYVAVYLPGHFFHLLNIQHPDLICHSLFLTENSEVIDMLPHSPLQSLSGSLVL
IAFLNLGYFVAVYSPGHFLHLLNIQHPDLVCHSLFLTGNNKIAAVLPPSPLQSLPGSLVL
VTFLNLGYFVAVYSPCRFLHLLNIRHPDLICHSLFLTGNNKTAAVLPPSPLQSLPGSLIL
11
doi: 10.1038/nature09325
Gallus
VAFLNLDYYVAAYLPGQFLHLLNIQHPDLLCYSLFLTGEDARIDMLPNCSIQSPLVSTVL
Homo
Canis
Bos
Mus
Rattus
Gallus
DCCSGKLYRALLSQSSLLQLLQNTCLDCEKMAALHCALYCGQGAQFLEAQIIQWISENVS
DWCSGKLYRALLNQSYLLQFLWNTQLDCEKMAVLHCVLSCGRDPRFLEAKIIQWISENIS
DSRSGKLYRVLLNQSYLVEFLRSARLDCERMALLHCALSHGRDPRRLEAKIIQWISENIS
DCYSGKVYRVTLDQSYLLRFLWNAHLDCERMAALHCILSCSQDPGFPEEQIIQWISEHVS
DCSSGKVYRATLDQSYLMGFLWNAQLDCEKMAALHCALSCDSDPGFPE-QIVQWVSERVS
DCCIGRLYAMSISDSALLKYLQNSKRDSERLAALHCALLCVRRTTDLEMKIIWWISENLS
Homo
Canis
Bos
Mus
Rattus
Gallus
ACHSFDLIQEFIIASSYWSVYSETSNMDKLLPHSSVLTWNTEIPGITLVTEDIALPLMKV
TCHSFDLIQEFIIASSYWSIYPETSNIDKLLPYSSVLTWNTEIPGITLVTEEITLPFMKV
ACHSFDLIQEFIIASSYWSIYPETSNMDKLLPYSSLLTWDTEIPGITLVTEEIPLPLMKV
ACHSFDLIQEFLIASSYWSVYAELDDMGMLLQYSSVLTWNTEIPGIKFTTEELPLPLMKV
ACHSFDLIQEFLIASSYWSVYPGLDDVDLLLPYSSVLTWDTEIPGMKLVTEELPLPLMKV
TCHSFDPIQEFIIASLYCRMCPETNNLDKLLPYTSLLDWTGVIPGVACATDIISLPVLEM
Homo
Canis
Bos
Mus
Rattus
Gallus
LSFKGYWEKLNSNLEYVKYAKPHFHYNNSVVRREWHNLISEEKTGKRRSAAYVRNILDNA
HSFKGYWEKLNSNLEYVKCSKPCLLYNNSMVKREWHSLISEEKTGRRRSMVYVRNIFDNA
HSFKGYWEKLNSNLEYVKYSKPHLHYNNSVVRREWHNLISEEKTGKRRSTVYVRNILDNA
YGLKGYWAKLNSNLEYIKYTKPHLHYHNSVVRREWHNLISEERTGKRRSTMYVRNILENA
YSLKGYWAKLNSNLEYIKYTKPHLHYHNSVVRREWHNLISEERTGKRRSTMYVRNILDNA
QNSKGFWEKLDSNLESVKYAEPHLHYHNNVLRREWRNLSEE-------------------
Homo
Canis
Bos
Mus
Rattus
Gallus
Homo
Canis
Bos
Mus
Rattus
Gallus
VKVISNLEARNLGPRLTPLLQEEDSHQRLLMGLMVSELKDHFLRHLQGVEKKKIEQMVLD
Supplementary
Figure Legends
MKVISNLEARNLEPRLTPLFQEEDYHQRLLIGLMVSELREHLLRHLQGIGKKKIEQMVLD
Supplementary
Figure 1. gSAP action on APP processing. Ternary complex of gSAP,
IKVISNVEAKNLEPRLTPLFQEEDTHQQLLIGLMVSELREHLLRHLQGVEKRKIEQMVLD
MKVIASMETRTLEPRLIPFLQEEDRHQRLLMGLMVSELRDHLLRHLQGVEKKKIEQMVLD
APP VKVISNMEMKTFEPRLIPLLQEEDRHQRLLMGLMVSELRDHLLRHLQGVEKKKIEQMVLD
and -secretase (top) is associated with elevated -cleavage (A production) and
---------------------------------MVAQLKDHLMRHLQYVGKKKIDQIVLD
YISKLLDLICHIVETNWRKHNLHSWVLHFNSRGSAAEFAVFHIMTRILEATNSLFLPLPP
complex
of APP and -secretase is associated with decreased -cleavage and increased YISKLLDLICQILETSWRTHHLHPWVLHL--RASAAEFTVFHIMTRILEATMSLFLPLPP
YVSKLLDLICQILEASWRKHNLHPWALHFNRQASAAEFAVFHIMTRILEATNTLFLPLPP
cleavage.
YISKLLDLIWCLLETSWRKHSMHPLVLHLNSHCSAADFEVFHLMTRILDAASSLCLPLPP
YISKLLDLVWCLLETSWRKHSVHPWVLHLNEHGSPADFEVFHLMTRILDAASSLCFPLPP
Supplementary
Figure 2. a: structures of imatinib, G01, and 125I-G01. b: G01
YVANLLNLVHRIMKEVWKIHQLHSCIFCFDERGSEAEFRVFHIMSRILEAANGMCMPLPP
Homo
Canis
Bos
Mus
Rattus
Gallus
significantly
reduces levels of A in N2a cells.
GFHTLHTILGVQCLPLHNLLHCIDSGVLLLTETAVIRLMKDLDNTEKNEKLKFSIIVRLP
GFHTLHTILGVHCLPLHNLLHYIDSGVLLLTETAVIRLMKDLDNSENNEKLKFSIIVRLP
Supplementary
Figure 3. a: Procedures for the synthesis of biotinylated imatinib
GFHTLHMILGVRCLPLHNLLHYIDHGVLLLTEAAVTRLMKDLDNTEKNEKLKFSIIMRLP
GFHSLHTILGVHCLPLYSLLHYIDNGVLLLTETAVTRLMKDLDNSEKNEQLKFSIIVRLP
GFHSLHTILGVHCLPLYNLLHYIDNGVLLLTETVVTRLMKDLDNSEKNEKLKFSIIVRLP
GFHSLHLGLGVRCLPLHTLLHYIDNGVLHLTETCVRKLLKDLDDNEKNEKLKFSIVTRLP
DMF, room temperature, 2 h. (b) 6-methyl-N1-(4-(pyridin-3-yl)pyrimidin-2-yl)benzene-
Homo
Canis
Bos
Mus
Rattus
Gallus
PL----IGQKICRLWDHPMSSNIISRNHVTRLLQNY-KKQPRNSMINKSSFSVEFLPLNY
1,3-diamine, TCFH, DIPEA, DMF, room temperature, overnight. (c) TFA, CH2Cl2, room
PH----IGQKICRLWDHPMSSNIISRNHVKQLLLNY-KKQPQSSMIDKSPGSVEFLPLNY
PL----TGQKICRLWDHPVSSNIISRNHVKRLLQNY-NKQPWSSVMDKSSFSVEFLPLNY
temperature,
30 min. (d)Biotin-OSu, HOBt, DIPEA, room temperature, overnight, and
PL----IGQKVCRLWDHPMSSNIISRNHVARLLKNY-RKEPRNSMIDKSSFPVEFLPLNY
PL----IGQKVCRLWDHPMSSNIISRNHVAQLLKNY-KKEPQSSMIDKSSFPVEFLPLNY
then HPLC
purification. Compound IC200001 was synthesized by reacting N-desmethyl
EVTLDALGLKARQFWDHPVNANFRARKYVKLLLEKLGNRQCSRPVPERHPVCVEFLPLNY
Homo
Canis
Bos
Mus
Rattus
Gallus

FIEILTDIESSNQALYPFEGHDNVDAEFVEEAALKHTAMLLGL--FIEILTDIESSNQALYAFEGHDNVDAKFVEEAALKHTTMLLGL--b: Kinase
profiling results shows that IC339239 has activities comparable to those of
FIHILTDIESSNPALYAFEGHDNVDAKFVEEAALKHTAMLLGL--FIEILMGLESSNQALYGFEGHDNVDAEFVEEAALKHTTMLLGL--imatinib,
while IC200001 showed no activity. Therefore, IC339239 is designated as
FIEILMHLESSNQALHGFEGHDNVDAEFVEEAALKHTTSLLGL--LTNVLAEIES--QGVHLYEKQDHINVRFVEEAALKHTMMLLGLRYS
active
biotin-imatinib and IC200001 as inactive biotin-imatinib.
Supplementary Figure 4. Alignments of gSAP
12 sequences among species. Red: identical
Supplementary Figure 5. gSAP mRNA expression levels in different tissues. Tissues

from 3 month old wild type BL/6 mice were harvested and gSAP levels were quantitated
doi: 10.1038/nature09325
Supplementary Figure 5.

Supplementary Figure 5. gSAP mRNA expression levels in different tissues. Tissues
from 3 month old wild type BL/6 mice were harvested and gSAP levels were quantitated
using real time PCR. Both actin and GAPDH served as internal controls (n=6). Tissue
5 prior to analysis.
extracts were adjusted to the same protein levels
Supplementary Figure 6. gSAP knockdown does not influence the generation of -, or
-APP-CTF from full length APP. N2a cells overexpressing APP695 were pre-treated
with the -secretase inhibitor, L685,458 [1 M] and the cleavages of APP were
monitored by pulse-chase labeling (35S-methionine) followed by immunoprecipitation.
Proteins were separated by SDS-PAGE and transferred to PVDF membrane for
autoradiography.
Supplementary Figure 7. Regulation of AICD production by gSAP. a. Either gSAP
knockdown or imatinib treatment increases AICD levels in N2a cells overexpressing
APP695 (**p < 0.01; n=3). b. Transfection of gSAP into HEK293 cells overexpressing
APP-CTF led to reduced AICD production and increased A production. Cleavage of
APP-CTF was monitored by pulse-chase labeling
13 at indicated time point. AICD and A
levels after the 3 hr chase were quantitated (**p < 0.01; n=3).
Supplementary Figure 8. After separation of organelles from N2a cells on a continuous
www.nature.com/nature sucrose gradient, endogenous gSAP co-localizes with -secretase in a Golgi-enriched
fraction (No. 6), which also contains endosomes.
10
doi: 10.1038/nature09325
extracts were adjusted to the same protein levels prior to analysis.
autoradiography.
APP-CTF was monitored by pulse-chase labeling at indicated time point. AICD and A
sucrose gradient, endogenous gSAP co-localizes with -secretase in a Golgi-enriched
Supplementary Figure 9. Truncation of APP-CTF and immunoprecipitation using gSAP
antibody through gSAP demonstrates that gSAP interacts with the juxtamenbrane region
of APP-CTF. APP-CTF-T1 is the truncated form of APP--CTF spanning from its Nterminus to HHGV64; APP-CTF-T2 is the truncated form of APP--CTF spanning from
its N-terminus to VMLKK55. Truncated forms were overexpressed in HEK293 cells and
11
doi: 10.1038/nature09325
autoradiography.
15


12
doi: 10.1038/nature09325
Supplementary
Figure 8
autoradiography.
immunoprecipitated with gSAP antibody. 6E10 antibody was used for immuno-detection.
13
doi: 10.1038/nature09325
Supplementary
Figure 9
autoradiography.
immunoprecipitated with gSAP antibody. 6E10 antibody was used for immuno-detection.
17
14
doi: 10.1038/nature09325
Supplementary Figure 10. Domain exchange studies demonstrate that gSAP regulates cleavage of APP-CTF but not NotchE, through selective interaction with AICD. a.
Design of APP-CTF/NICD and NotchE/AICD constructs. b. APP-CTF/NICD, APPCTF, and NotchE/AICD were overexpressed separately in HEK293 cells. Both APPCTF and NotchE/AICD interact with gSAP, while APP-CTF/NICD does not. c. Effects
of gSAP knockdown on the cleavage of APP-CTF/NICD, APP-CTF, and NotchE/AICD
by -secretase. Individual constructs were overexpressed in N2a cells. Upon gSAP
knockdown, AICD production from APP-CTF (center panel) as well as from
NotchE/AICD (right panel) increased, but NICD production from APP-CTF/NICD was
not influenced (left panel).
Supplementary Figure 11. Comparison of the effects of either reducing gSAP or of a secretase inhibitor on A levels and histopathology. a. Mice were given a -secretase
inhibitor DBZ (10 mol/kg) for 5 days. This resulted in reduced A40 and A42 levels of
449% and 475% respectively. b. No histopathological changes were observed in
mouse small intestine after gSAP knockdown (H&E and PAS staining). However,
increased amounts of violet-stained goblet cells18
were observed after DBZ administration,
www.nature.com/nature a finding typical of Notch inhibition.
15
doi: 10.1038/nature09325
b
Supplementary Figure 10. Domain exchange studies demonstrate that gSAP regulates cleavage of APP-CTF but not NotchE, through selective interaction with AICD. a.
Design of APP-CTF/NICD and NotchE/AICD constructs. b. APP-CTF/NICD, APPCTF, and NotchE/AICD were overexpressed separately in HEK293 cells. Both APPCTF and NotchE/AICD interact with gSAP, while APP-CTF/NICD does not. c. Effects
of gSAP knockdown on the cleavage of APP-CTF/NICD, APP-CTF, and NotchE/AICD
by -secretase. Individual constructs were overexpressed in N2a cells. Upon gSAP
knockdown, AICD production from APP-CTF (center panel) as well as from
NotchE/AICD (right panel) increased, but NICD production from APP-CTF/NICD was
not influenced (left panel).
Supplementary Figure 11. Comparison of the effects of either reducing gSAP or of a secretase inhibitor on A levels and histopathology. a. Mice were given a -secretase
inhibitor DBZ (10 mol/kg) for 5 days. This resulted in reduced A40 and A42 levels of
449% and 475% respectively. b. No histopathological changes were observed in
mouse small intestine after gSAP knockdown (H&E and PAS staining). However,
increased amounts of violet-stained goblet cells were observed after DBZ administration,
a finding typical of Notch inhibition.
19
16

He Et Al 2010 SI

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

He Et Al 2010 SI

Uploaded by

Copyright:

Available Formats

Vol 467 | 2 September 2010 | doi:10.

Accumulation of neurotoxic amyloid-b is a major hallmark of

NATURE | Vol 467 | 2 September 2010

Figure 1 | Identification of GSAP as an imatinib target. a, A PS1-associated

GSAP-FL (98 kDa)

A38 (pg ml1)

A42 (pg ml1)

A40 (pg ml1)

Figure 2 | GSAP regulates amyloid-b production but does not influence

of the c-secretase cleavage product, the Notch intracellular domain

NATURE | Vol 467 | 2 September 2010

A42 (pg per mg

A40 (pg per mg

AD 2 mice with induction

GSAP RNAi-AD mice without induction

GSAP RNAi-AD mice

enzymeinhibitor complex, strongly suggesting that GSAP-16K is a

Figure 4 | Knockdown of GSAP reduces amyloid-b production and plaque

Fig. 10b). GSAP knockdown selectively increased AICD production

2010 Macmillan Publishers Limited. All rights reserved

NATURE | Vol 467 | 2 September 2010

on either amyloid-b or plaque concentrations in AD 32 mice. The

Supplementary Information is linked to the online version of the paper at

protease inhibitors. Immunoprecipitation was performed using the corresponding

2010 Macmillan Publishers Limited. All rights reserved

Radioiodination of G01 by 125I was performed without carrier using a modification of a

GAATGGATGTGTACTGTTACTTCTTCAGCATCACCAAG 3'. The presence of the

Supplementary Figure Legends

Supplementary Figure 4. Alignments of gSAP sequences among species. Red: identical

Supplementary Figure Legends

residues; Blue/Green: conservative substitutions. The C-terminal region of gSAP is

Supplementary Figure Legends

significantly reduces levels of A in N2a cells.

Supplementary Figure 3. a: Procedures for the synthesis of biotinylated imatinib

residues; Blue/Green: conservative substitutions. The C-terminal region of gSAP is

imatinib with Biotin-OSu.

Supplementary Figure 5. gSAP mRNA expression levels in different tissues. Tissues

Supplementary Figure 2. a: structures of imatinib, G01, and 125I-G01. b: G01

fraction (No. 6), which also contains endosomes.

APP-CTF led to reduced AICD production and increased A production. Cleavage of

sucrose gradient, endogenous gSAP co-localizes with -secretase in a Golgi-enriched

You might also like