Transcript PSD-95 - Fichier

Interaction between SAP97 and PSD-95,
Two Maguk Proteins Involved
in Synaptic Trafficking
of AMPA Receptors
Received for publication,May 31, 2005, and in revised form, October 26,
2005 Published, JBC Papers in Press, December 6, 2005
Chunlin Cai, Hong Li, Claudio Rivera, and Kari Keinänen
From the Department of Biological and Environmental Sciences, Division of
Biochemistry and Division of Physiology and the
Institute of Biotechnology, Viikki Biocenter, University of Helsinki.
Ceccaldi Benoît
De la Crompe Brice
Master 2 Neurosciences 2011 Bordeaux
UE Cellular and Molecular Neurobiology
Plan

INTRODUCTION

EXPERIMENTAL PROCEDURES



Experimental design
GST Pull-down assay
MATERIALS AND METHODS – RESULTS

Part 1: Protein interaction






SAP97 Associates with PSD-95: in Vivo and in Vitro
Mapping the PSD-95 Binding Site in SAP97
Mapping the SAP97 Binding Site in PSD-95
GST Pull-down Analysis of SAP97-PSD-95 Interaction
Part 2: Study of SAP97-PSD-95 Interaction in Cultured Neurons
DISCUSSION



Proteins interaction
Role of PSD-95/SAP97 in GluR-A-containing AMPAR synaptic clustering
Authors’ hypothesis on trafficking GluR-A-containing AMPAR
INTRODUCTION

SAP97 and PSD-95 are two Maguk proteins
(membrane-associated guanylate kinase homologs)
implicated in the synaptic targeting and anchoring of
AMPA receptors:

SAP97 bind directly to the C-terminus of the GluR-A subunit. SAP97
overexpression promote the synaptic delivery of GluR-A-containing
AMPAR

PSD-95 bind indirectly via stargazin and TARPs to GluR-A. PSD-95
overexpression trigger the synaptic trafficking GluR-A-containing
AMPAR in synaptic spines

Experiment using RNAi Knock down of SAP97 and PSD-95 inhibit the
clustering of GluR-A
AMPA receptor

PDZ of SAP97 directly interacts with the C-terminal part of GluR-A

PSD-95 associates with AMPA-R via stargazin and TARPs
MAGUKs (Membrane-Associated Guanylate Kinase homologs)
PSD-95 (PostSynaptic Density-95)
SAP97 (Synapse-Associated Protein 97)
L27
Others: PSD-93/Chapsyn-110 , SAP102

The oligomeric nature of Maguks oriented the authors to
examinate the potentiality for SAP97 and PSD-95 to
form heteromeric complexe

Experimental questions:

Interaction studing of the interaction between these two proteins
(domains interaction)

Impact studing of this interaction into the synaptic transport of GluR-Acontaining AMPAR
EXPERIMENTAL
PROCEDURES
Experimental design
Part
1: Proteins interaction
Authors use coimmunoprecipitation experiments to study
interaction between PSD-95 and SAP97.
Firstable they analyse the formation of a complex in vivo (rat brain lysat)
and in vitro (transfected HEK293 cells)



Then they show domains involved in the interaction.
They validate results by using GST-pull down assay in HEK293 cells
Part
2: study of SAP97-PSD-95 Interaction in Cultured
Neurons
They study in hippocampal neurons (mouse embryos, E17) the effect of an
overexpression of PSD-95 on the synaptic clustering of SAP97.

After they analyse the impact of the complex formation on the GluR-A –
containing AMPAr synaptic clustering.

GST Pull-down assay
•Pull-down assay is an in vitro
experiment use to determine physical
interaction between 2 or more proteins.
•GST fusion proteins is transfected and
expressed in Esherichia coli BL21.
•Then they are purified by interaction of
glutathione-Sepharose with GSTprotein.
•After centrifugation we obtain only the
fusion protein in solution.
•In the time we obtain a other solution
by lysing the cell wich contain the prey
protein.
•After, to allow the formation of the
prey/bait complex, we mix the two
solutions.
•The solution is wash and elute in SDS
buffer then submit to electrophoresis
and western blotting.
Materials and methods - Results
Part 1: Protein interaction
SAP97 Associates with PSD-95 in Vivo
Materials and Methods

Detergent extract of cerebella from adulte Wistar rats were
submitted to co-immunoprecipitation using SAP97 N
antiserum, the corresponding preimmune serum or a PSD-95specific antibody.

The obtained supernatants are subjected to immunoblotting
with SAP97 N and anti-PSD-95.
SAP97 Associates with PSD-95 in Vivo
Results
Non palmitoylated
Palmitoylated
Fig 1A: Coimmunoprecipitation of SAP97
and PSD-95 in Rat brain detergent extract.

Input lane verifies the presence of protein
in cells.

The two Co-IP constitute there reciprocal
controls.

In rat brain, PSD-95 and SAP97
coimmunoprecipitate.
=> In rat brain, PSD-95 and SAP97 interacting
to form a complex.
SAP97 Associates with PSD-95 in Vitro
Materials and Methods

In first, HEK293 cells were transfected (CalciumPhosphate coprecipitation) with plasmids containing
fusion tagged-proteins: Myc-SAP97 or GFP-PSD-95.

Myc tag was added on N-Terminal part of proteins

Green fluorescent protein was added on C-terminal

Detergent extract of HEK293 cells were submitted to coimmunoprecipitation using Myc or GFP-specific antibody.

The obtained supernatants were subjected to
immunoblotting with anti-Myc and anti-GFP.
SAP97 Associates with PSD-95 in Vitro
Results(1/2)

In cotransfected HEK293 cells, MycSAP97 and GFP-PSD-95
coimmunoprecipitate.

The Co-IP does not allow to show a
direct proteins interaction.

But, HEK293 cells are non-neurals
so the authors conclude that:
=> interaction between this two tagged
proteins is direct.
=> Authors verified the presence of
endogenous SAP97 or PSD-95 (not
published data). HEK293 cells
express endogenous SAP97.
Fig 1B and1C: Coimmunoprecipitation of SAP97 and PSD-95 in HEK293 detergent extract.
SAP97 Associates with PSD-95 in Vitro
Results(2/2)

Coimmunoprecipitation of GFPPSD-95 and endogenous SAP97
=> GFP-PSD-95 interacting directly
to form a complex with
endogenous SAP97.
=> Because HEK293 cells express
SAP97, we cannot exclude the
possibility that they can express
other MAGUK proteins.
Fig 1A: Coimmunoprecipitation of endogenous SAP97
and GFP-PSD-95 in HEK293 cell detergent extract.
=> To validate the direct interaction,
we purpose to use other
methods like Yeast double hybrid
method or
Mapping the PSD-95 Binding Site in SAP97
Materials and Methods



In first, HEK293 cells were transfected (Calcium-Phosphate
coprecipitation) with plasmids containing fusion tagged-proteins: HisPSD-95 (full length, C-term) and differents Myc tagged domains of
SAP97 (N-term).
His tag was added on C-terminal
The same precedent protocol of Co-IP was used.
Mapping the PSD-95 Binding Site in SAP97
Results

Coimmunoprecipitation of HisPSD-95 with:




Myc-SAP97
Myc-ΔSH3-GK
Myc-NTD
NTD is unique common domain which Co-IP with
PSD-95.
=> The binding site of PSD-95 in SAP97 is the
NTD.
Fig 2A and 2B: Mapping the PSD-95 binding domain in SAP97. HEK293 cells were transfected for expression of His-tagged
full-length PSD-95 together with the indicated Myc-tagged SAP97 domains. The arrows indicate the position of the immunoglobulin
heavy chain band. Molecular size markers are shown on the left. Un., untransfected cells; Neg., no cotransfection with SAP97.
Mapping the SAP97 Binding Site in PSD-95
Materials and Methods
• In first, HEK293 cells were transfected (Calcium-Phosphate
coprecipitation) with plasmids containing fusion tagged-proteins: GFPSAP97 (full length, N-terminal tag) and differents Myc tagged domains of
PSD-95 (N-term).
• The same precedent protocol of Co-IP was used.
Mapping the SAP97 Binding Site in PSD-95
Results

Coimmunoprecipitation of GFPSAP97 with:





Myc-PSD-95
Myc-ΔN-PSD-95
Myc-SH3-PSD-95
Myc-SH3-GK-PSD-95 (fig B)
SH3 is unique common domain which Co-IP with
SAP97.
=> The binding site of SAP97 in PSD-95 is the SH3
domain.
=> Why in first experiment, the SH3-GK-PSD-95
domain did not bind to SAP97?
Fig 3A and 3B: Mapping the SAP97 binding domain in PSD-95. HEK293 cells were transfected
for the expression of GFP-tagged full-length SAP97 together with the indicated Myc-tagged PSD95 domains. The arrows point to the immunoglobulin heavy chain band. Molecular size markers are
shown on the left. Un., untransfected cells.
GST Pull-down analysis of SAP97-PSD-95
interaction - Materials and Methods


GST fusion protein were expressed in E.coli BL21 and
purified with gluthatione-Sepharose bead.
In first experiment:



In second experiment:




HEK293 cells were transfected by C-terminal tagged GFP-PSD-95.
All SAP97 fragments were fusioned with GST then expressed and purified in
E.coli.
HEK293 cells were transfected by C-terminal tagged GFP-PSD-95-SH3.
SAP97ΔNTD and SAP97NTD were fusioned with GST then expressed and
purified in E.coli.
Then GST-Protein-containing solution were mix with cells
lysat.
Finally, the purified solution was submitted to western blot.
GST Pull-down analysis of SAP97-PSD-95
interaction - Results

Fig A: Interaction between GFP-PSD95 and:

GST-SAP97

GST-NTD-SAP97

GST-PDZ1-3-SAP97
=> The domain of GST-SAP97 that binds
to GFP-PSD-95 is NTD of SAP97.
=> GST-PDZ1-3 may constitue a second
binding site.
Fig 4: GST pull-down analysis of SAP97-PSD-95 interaction
A, binding of GFP-PSD-95 to SAP97 domains (transfection of HEK293
cells with GFP-PSD-95 and GST-SAP97 fusion proteins).
B, binding of GFP-PSD-95 SH3 to SAP97 domains (transfection of
HEK293 cells with GFP-PSD-95 and GST-SAP97 fusion
proteins).
Fig B: Interaction between GFP-PSD95-SH3 and GST-SAP97-NTD.
=> The SH3 domain of GFP-PSD-95 bind
directly to the GST-SAP97 NTD.

Part 2: Study of SAP97-PSD-95
Interaction in Cultured Neurons
SAP97-PSD-95 Interaction in Cultured
Neurons Materials and Methods

Hippocampal neurons were fixed.

The fixed hippocampal neurons were permeabilized.

An overnight incubation with the primary Antibodies/antiserum
was realized (SAP97N antiserum, anti-PSD-95 monoclonal
antibody, and/or GluR-ACTD antiserum).

After washing, the secondary antibodies conjugated to
Cyanine 3 (red) or Alexa fluor 488 (green) were added.

Immunostaining was visualized via fluorescence microscope.
SAP97-PSD-95 Interaction in Cultured
Neurons - Results (1/4)
Distribution of SAP97 and PSD-95 in
cultured hippocampal neurons

Both endogenous proteins
are distributed in whole cell.

We can see that PSD-95
(red) is also localized in
synaptic spines unlike
SAP97 (green).

In colocalization study,
SAP97 can be present in
spines but always with
PSD-95 (white arrows in
merge photo).
A, immunofluorescence localization of endogenous SAP97 and PSD-95. Mouse hippocampal neurons (E17; kept for 17 days in vitro)
were fixed and stained with rabbit anti-SAP97 N and mouse PSD-95 antibodies followed by Alexa Fluor-488-conjugated anti-rabbit
and Cy3-conjugated anti-mouse IgGs, respectively. Individual immunofluorescence stainings for SAP97 (green) and PSD-95 (red) are
shown in the upper panel, whereas the lower panel represents a merged picture of the individual immunostainings showing
overlapping staining in yellow (Merge; Zoom for an enlarged view). The arrows indicate spines containing both SAP97 and PSD-95.
SAP97-PSD-95 Interaction in Cultured
Neurons - Results (2/4)
Distribution of SAP97 and PSD-95 in
cultured hippocampal neurons

The overexpression of
PSD-95 trigger the spine
clustering of SAP97
 To verifying this
observations, the authors
conducted the following
experiment.
Induction of SAP97 clustering by overexpression of PSD-95. Cultured neurons transfected for expression of GFP-tagged PSD-95
were fixed and stained with anti-SAP97 N followed by Cy3-conjugated anti-mouse IgG. A merged image of the Cy3 (red) and GFP
(green) fluorescence is shown. Individual fluorescence images of the boxed dendritic area are shown on the right.
SAP97-PSD-95 Interaction in Cultured
Neurons - Results (4/4)





The overexpression of GFP-PSD-95 (green)
induces the spine clustering of GluR-A (red). (cf fig
A)
The cotransfection with PSD-95-SH3 inhibits the
GluR-A clustering induces by PSD-95. (cf fig B)
The cotransfection with SAP97-NTD inhibits the
GluR-A clustering induces by PSD-95. (cf fig C)
The overexpression of SAP97-NTD alone does not
induce the clustering of GluR-A.
The overexpression of PSD-95-SH3 alone does not
induce the clustering of GluR-A.
PSD-95-induced clustering of GluR-A.
Cultured hippocampal neurons transfected with GFP-PSD-95 plasmid alone (A) or together
with Myc-tagged SAP97 NTD (B) or PSD-95 SH3 (C) constructs were stained with GluR-A
CTD antiserum followed by Cy3-con-jugated anti-rabbit IgG and then visualized for GluR-A
immunoreactivity (red) and for GFP fluorescence (green). The panels on the right show
enlarged images of GluR-A clusters along dendrites of untransfected cells (Un; blue-lined
boxed area in the left panel) and of GFP-positive PSD-95-overexpressing cells (white-lined
boxed area in the left panel).
Intensities of GluR-A immunoreactive clusters in transfected neurons. Clustering of GluR-A was analyzed by immunofluorescence and quantified as
described under “Experimental Procedures.” Cluster intensities in GFP-positive cells are indicated as percentage of control using cluster intensities measured
from neighboring untransfected cells as a 100% reference including a total of 127 clusters in 20 neurons; 100 +/- 8.4%). Statistical significance of the
results was analyzed by using unpaired t test to calculate the indicated two-tailed p values. NS, not significant (p<0.05); NA, not applicable.
SAP97-PSD-95 Interaction in Cultured
Neurons - Results (3/4)
PSD-95-induced clustering of SAP97 to
synaptic spines is inhibited by PSD-95SH3

Overexpression of PSD-95
trigger the spine clustering
of SAP97.

The PSD-95SH3 inhibits the
clustering of SAP97
Number of SAP97 immunoreactive spines in
transfected neurons
Cultured hippocampal neurons were transfected for expression of GFP-tagged PSD-95 alone or together with Myc-tagged PSD-95
SH3 as indicated and then stained with anti-SAP97 N followed by Cy3-conjugated anti-mouse IgG. Merged images of the Cy3 (red)
and GFP (green) fluorescence are shown on the left. The right panels show an enlarged view of the SAP97 immunofluorescence of
the boxed areas. Dendrites of untransfected and transfected cells are indicated by the arrows and arrowheads, respectively.
DISCUSSION
Proteins interaction

In rat brain and in transfected cells, the SAP97 and PSD-95 can be associate directly
as heteromeric complex.

Known mechanisms involving interaction of PSD-95 domains with other Maguks are
PSD-95/Chapsyn110 via NTD and PSD-95/SAP102 via SH3-GK domains.

Authors found a new interaction mechanism between Dlg proteins: an association of
SH3 (PSD-95) and NTD (SAP97).

SAP97 does exist in two states because of intramolecular interactions:

In the closed complex, SH3 and GK interacting. This association prevents the GKAP binding
on GK domain.

In the second state, NTD interacting with SH3 liberate GK domain.

Similary configuration were found in PSD-95.

The authors hypothetis: The binding of NTD(SAP97) with SH3(PSD-95) promotes the
complex anchoring via PSD-95-GK/GKAP interaction. GKAP interacts with Shank
proteins which is a central actor for anchoring of Glutamate receptor in PSD.
Implication of PSD-95/SAP97 in GluRA-containing AMPAR synaptic clustering

In this study authors found:

GFP-PSD-95 overexpression trigger synaptic clustering of
SAP97.

The inhibition induced by cotransfection of PSD-95-SH3 suggest
the major role of PSD95/SAP97 in synaptic clustering of SAP97.

Overexpression of GFP-PSD-95 induces the synaptic clustering
of GluR-A.

The inhibition triggered by co-transfection with SAP97-NTD and
PSD-95-SH3 suggest the importance of SAP97/PSD-95
interaction in GluR-A clustering.
Authors’ hypothesis on trafficking GluR-Acontaining AMPAR

Precedent study has demonstrated that:

Association between GluR-A-containing AMPAR and PSD-95 is mediate by SAP97.

An endocytosis mechanism of synaptic GluR-A-containing AMPAR is based on the
trimeric complex formation of a GluR-A/SAP97/MyosineVI.
Myosine VI is a motor protein implicated in clathrin mediated endocytosis of GluR-A.



Myosine VI can bind to NTD SAP97 like PSD-95.
 competition between PSD-95 and MyoVI.
The PSD-95 binding on the NTD-SAP97 may prevent the binding of MyoVI.
 This interaction can cause the stabilization of GluR-A-containing AMPAR clustering.
 This clustering may be mediate by the interaction of GK PSD-95 domain with GKAP.

Authors « tentavely » suggest that SAP97-PSD-95 interaction serves an
essential role in synaptic accumulation of GluR-A-containing AMPAR,
possibly through stabilization of receptor cluster.
Thank you
Yeast two hybrid system
GKAP
GK
PSD-95
SH3
PSD-95
NTD
SAP97
PDZ
SAP97
C-terminal
GluR-A-containing AMPAR