ORIGINAL RESEARCH ARTICLE
Journal of
Interfering of the Reelin/ApoER2/
PSD95 Signaling Axis Reactivates
Dendritogenesis of Mature
Hippocampal Neurons
Cellular
Physiology
2
€
ESTIBALIZ AMPUERO,1* NUR JURY,1 STEFFEN HARTEL,
MARÍA-PAZ MARZOLO,3
1
AND BRIGITTE VAN ZUNDERT *
1
Center for Biomedical Research, Faculty of Biological Sciences and Faculty of Medicine, Universidad Andres Bello, Santiago, Chile
2
SCIAN-Lab, CIMT, Bomedical Neuroscience Institute (BNI), ICBM, Faculty of Medicine, University of Chile, Santiago, Chile
3
Laboratorio de Trafico Intracelular y Se~nalizacion, Departamento de Biología Celular y Molecular, Facultad de Ciencias Biologicas,
Pontificia Universidad Catolica, Santiago, Chile
Reelin, an extracellular glycoprotein secreted in embryonic and adult brain, participates in neuronal migration and neuronal plasticity.
Extensive evidence shows that reelin via activation of the ApoER2 and VLDLR receptors promotes dendrite and spine formation during
early development. Further evidence suggests that reelin signaling is needed to maintain a stable architecture in mature neurons, but, direct
evidence is lacking. During activity-dependent maturation of the neuronal circuitry, the synaptic protein PSD95 is inserted into the
postsynaptic membrane to induce structural refinement and stability of spines and dendrites. Given that ApoER2 interacts with PSD95, we
tested if reelin signaling interference in adult neurons reactivates the dendritic architecture. Unlike findings in developing cultures, the
presently obtained in vitro and in vivo data show, for the first time, that reelin signaling interference robustly increase dendritogenesis and
reduce spine density in mature hippocampal neurons. In particular, the expression of a mutant ApoER2 form (ApoER2-tailless), which is
unable to interact with PSD95 and hence cannot transduce reelin signaling, resulted in robust dendritogenesis in mature hippocampal
neurons in vitro. These results indicate that reelin/ApoER2/PSD95 signaling is important for neuronal structure maintenance in mature
neurons. Mechanistically, obtained immunofluorescent data indicate that reelin signaling impairment reduced synaptic PSD95 levels,
consequently leading to synaptic re-insertion of NR2B-NMDARs. Our findings underscore the importance of reelin in maintaining adult
network stability and reveal a new mode for reactivating dendritogenesis in neurological disorders where dendritic arbor complexity is
limited, such as in depression, Alzheimer’s disease, and stroke.
J. Cell. Physiol. 232: 1187–1199, 2017. ß 2016 Wiley Periodicals, Inc.
During early postnatal brain development, robust
dendritogenesis is followed by an elimination, or pruning, of
excessive and mis-targeted branches (Cline and Haas, 2008; Jan
and Jan, 2010). With maturation, dendritic branches become
decorated with tiny dendritic protrusions that can mature into
spines harboring excitatory synapses and holding hundreds of
proteins in the postsynaptic density (PSD) (Sheng and
Hoogenraad, 2007). NMDA receptors (NMDARs) are located
in the PSD and, as primary recipients of excitatory inputs, these
glutamatergic receptors are critical players in regulating
neuronal architecture in activity- and developmentaldependent manners (Cline and Haas, 2008).
NMDARs can be composed of two obligatory NR1
subunits plus two NR2A-D and/or NR3A-B subunits
(Cull-Candy and Leszkiewicz, 2004). However,
considerable evidence points to a central role of the NR2A
and NR2B subunits in regulating plasticity. Particularly,
synaptic NR2B-rich NMDARs play a critical role in dendrite
formation during early brain development (Espinosa et al.,
2009; Sepulveda et al., 2010; Bustos et al., 2014). With
activity-dependent maturation of the neuronal circuitry,
NR2A subunit expression increases and induces structural
refinement and dendrite stability (Charych et al., 2006;
Henriquez et al., 2013; Bustos et al., 2014). Comparative and
experimental studies further indicate that NR2A-NMDARs
are held in the PSD by the scaffolding protein PSD95 (Sans
et al., 2000; Losi et al., 2003; van Zundert et al., 2004; Elias
et al., 2008; Bustos et al., 2014). Moreover, knock-down of
PSD95 leads to synaptic re-insertion of NR2B-NMDARs
© 2 0 1 6 W I L E Y P E R I O D I C A L S , I N C .
(Be€ıque et al., 2006; Bustos et al., 2014) and consequent
induction of robust dendritic branching in mature
hippocampal neurons (Bustos et al., 2014).
Contract grant sponsor: FONDECYT;
Contract grant numbers: 3130582, 1140301, 1151029, 1150444.
Contract grant sponsor: UNAB Nucleus;
Contract grant number: DI-603-14N.
Contract grant sponsor: FONDEQUIP;
Contract grant number: EQM 140166.
Contract grant sponsor: CONICYT;
Contract grant number: 201161486.
Contract grant sponsor: BNI;
Contract grant number: ICM P09-015-F.
Contract grant sponsor: VISUAL D;
Contract grant number: ACT1402.
*Correspondence to: Dr. Estibaliz Ampuero and Dr. Brigitte van
Zundert, Center for Biomedical Research, Faculty of Biological
Sciences and Faculty of Medicine, Universidad Andres Bello,
Avenida Rep
ublica 217, Santiago, Chile.
E-mail: estiampu@gmail.com (E.A.); bvanzundert@unab.cl (B.Z.)
Manuscript Received: 10 September 2016
Manuscript Accepted: 12 September 2016
Accepted manuscript online in Wiley Online Library
(wileyonlinelibrary.com): 21 September 2016.
DOI: 10.1002/jcp.25605
1187
1188
A M P U E R O E T A L.
Another important molecule regulating the functional
and behavioral development of brain circuits is reelin, a
multifunctional extracellular matrix glycoprotein, the
expression of which after birth is mediated by GABAergic
interneurons (Herz and Chen, 2006; Lee and D’Arcangelo,
2016). In vitro and in vivo studies indicate that reelin is
important for circuit establishment, thus impacting dendrite
and spine development (Lee and D’Arcangelo, 2016).
Particularly, analysis of juvenile and adult reelin-deficient reeler
mutant mice revealed that neurons exhibit reductions in
dendritic tree complexity and dendritic spine density (Liu et al.,
2001; Niu et al., 2004, 2008). However, since these anatomical
studies were performed on mice in which reelin signaling
components were disrupted from the early embryonic stages,
it cannot be determined if reelin signaling regulates either the
formation, maturation, or maintenance of dendrites or spines
in mature neurons.
Reelin activates a core signaling pathway involving the
receptors ApoER2 (Apolipoprotein E receptor 2; also
termed LRP8 [low-density lipoprotein receptor-related
protein 8]) and VLDLR (very low density lipoprotein
receptor), the adapter protein Dab-1 (Disabled-1), and Src/
Fyn kinases (Lee and D’Arcangelo, 2016). PSD95 is part of the
reelin signaling pathway in the adult brain as this synaptic
scaffolding protein can interact with the cytoplasmic domain
of ApoER2, depending on the expression of the exon 19 that
encodes for a 59 amino acid insert in the cytoplasmic tail
(Beffert et al., 2005). Additionally, reelin modulates NMDARmediated synaptic plasticity and promotes memory formation
(Beffert et al., 2005; Chen et al., 2005; Hoe et al., 2006;
Dumanis et al., 2011).
Given these findings, we hypothesized that reelin signaling
via the ApoER2 (or VLDLR)/Dab1/PSD95 axis regulates
neuronal architecture maintenance of mature neurons and,
consequently, that reelin signaling interference should
reactivate a dendritic outgrowth of mature neurons. In this
study, we provide the first description for a novel role of reelin
in stabilizing the neuronal architecture of hippocampal neurons
in vitro and in vivo and identify underlying molecular
mechanisms.
Materials and Methods
Neuronal cultures
All protocols involving rodents were carried out in accordance
with NIH guidelines, and were approved by the Ethical and Biosecurity Committees of Universidad Andres Bello. Cultures of
hippocampal neurons were prepared from embryonic day (E)
18 Sprague–Dawley rat fetuses as previously described (Sepulveda
et al., 2010; Henriquez et al., 2013; Bustos et al., 2014). Briefly,
pregnant were deeply anesthetized with CO2 and hippocampi
were excised and placed into ice-cold PBS containing 50 mg/ml
penicillin/streptomycin. The extracts were minced and incubated
for 20 min at 37°C in pre-warmed PBS containing 0.25% trypsin
and then transferred to a tube containing Dulbecco’s modified
Eagle’s medium supplemented with 10% horse serum and
100 U/ml penicillin/streptomycin. Then, cells were resuspended
by mechanical agitation through fire-polished glass Pasteur
pipettes of decreasing diameters. Cells were counted and plated
on freshly preparedpoly-L-lysine-coated 24 well plates (1 mg/ml;
Sigma P2636, St Louis, MO). Plating media was replaced by
growth media Neurobasal (Life Technologies Corp., Carlsbad, CA,
21103-049) supplemented with B27 (Life Technologies Corp.
17504044), 2 mM L-glutamine (Life Technologies Corp. 25030081), 100 U/ml penicillin/streptomycin (Life Technologies Corp.
15070-063). On day 2, hippocampal neurons were treated with
2 mM cytosine arabinoside for 24 h, after that growth media was
replaced with half of new media every 2–3 days.
JOURNAL OF CELLULAR PHYSIOLOGY
Transient transfections
Transfection methodologies differed with cell type and
developmental stage. Hippocampal neurons at 15 DIV were
transfected by MagnetofectionTM using the Neuromag protocol
according to manufacturer instructions (OZ Biosciences,
Marseille, France) with slight modifications, and as previously
described (Henriquez et al., 2013; Bustos et al., 2014). Briefly,
30 min before magnetofection, medium was replaced with prewarmed Neurobasal media. Plasmid DNA, copGFP or ApoER2tailless, was incubated with Neuromag beads, in a ratio of 500 ng of
DNA per 0.75 ml of nanobeads in 100 ml of Neurobasal media. This
mixture was added drop wise to each 24-well plate and incubated
for 15 min at 37°C over a magnetic plate (OZ Biosciences), and old
medium was restored after 45 min.
Hippocampal neurons at 7 DIV were transfected by a CaPO4
transfection as previously described (Sepulveda et al., 2010; Bustos
et al., 2014). Briefly, hippocampal growth media was replaced with
prewarmed MEM (Life Technologies Corp. 21103-049) 20 min
prior to transfection. DNA/calcium complexes were added to the
plates drop wise and incubated for 1 h at 37°C and 5% CO2. Cells
were washed three times with pre-equilibrated MEM/EBSS at 10%
CO2 for 20 min to dissolve the DNA-CaPO4 precipitates. Cells
were left at 37°C and 5% CO2 until 12 DIV to performed
morphological analyses.
Reelin signaling interference
To interfere with the reelin/ApoER2/PSD95 signaling axis, neurons
were transfected (see details in “Transient transfections”) with
ApoER2-tailless, as previously described (Cuitino et al., 2005;
Fuentealba et al., 2007). To inhibit reelin binding to the ApoER2 and
VLDLR receptors, cultures were incubated during 5 days 20 mM
GST-RAP (Hiesberger et al., 1999; Cuitino et al., 2005). GST-fusion
proteins were expressed in Escherichia coli (BL21) and purified
according to the manufacturer’s instructions (16100, Pierce
ThermoFisher, Rockford, IL) with the addition of CompleteTM
protease inhibitor cocktail in the lysis buffer (PBS, 1% Triton-X-100,
10 mM EDTA). All the purified proteins were dialyzed against
50 mM Tris–HCl pH 8.0 (Cuitino et al., 2005). To reduce directly
extracellular reelin levels, hippocampal cultures were incubated
during 5 days with 2 mg/ml of the reelin neutralizing antibody CR50
(MBL Int. Corp., Woburn, MA, MBL-D223-3) (Utsunomiya-Tate et
al., 2000; Groc et al., 2007; Cuchillo-Iban~ez et al., 2013). CR50 (2 ml)
was also applied in vivo through stereotaxic injections into the
dentate gyrus region (see below).
Morphological analysis in vitro
GFP-transfected hippocampal neurons were fixed in 4%
paraformaldehyde (PFA) with 4% sucrose in PBS, mounted with
fluoromont and then visualized with a confocal (Olympus FV 1000,
Tokyo, Japan) or epi-fluorescence (Nikon eclipse Ti, Tokyo, Japan)
microscope.
As previously described (Sepulveda et al., 2010; Henriquez et al.,
2013; Bustos et al., 2014), to measure the number and length of
individual dendrites, every branch segment arising from the soma
was digitally marked from the origin of the branch to its termination
using the plugin NeuronJ of the ImageJ software (NIH). Also, to
measure the complexity of the dendritic arbor the Sholl analysis was
performed. All the analysis was blinded to the experimental
conditions. To visualize spines in culture, cells were fixed for 20 min
with 4% PFA plus 1% glutaraldehyde in PBS and followed by
immunofluorescent staining for GFP. Same confocal parameters
were used to obtain the images of spines in vivo (see below).
Immunofluorescence and cluster analysis
Immunofluorescence assays were performed as previously
described (Bustos et al., 2014; Segovia-Miranda et al., 2015). For
REELIN SIGNALING MAINTAINS ADULT NETWORK STABILITY
immunostaining assays of non-permeabilized cells, hippocampal
cultures were fixed with formaldehyde 4% in PBS for 10 min at
4°C. For immunofluorescence of permeabilized cells, hippocampal
cultures were rinsed twice in ice-cold PBS and fixed for 20 min in a
freshly prepared solution of 4% PFA with 4% sucrose in PBS. Then,
the cells were rinsed three times in cold PBS and permeabilized
for 5 min with 0.2% Triton X-100 in PBS. After that, for both
conditions the cells were rinses in ice-cold PBS and incubated in 1%
BSA plus 3% donkey serum in PBS for 30 min at room temperature,
followed by an overnight incubation at 4°C with primary
antibodies. Primary antibodies used were: Reelin (1:100, Millipore,
Temecula, CA, MAB5364), PSD-95 (1:500; UC Davis/NIH
NeuroMab Facility, 75-028), Synapsin Iab (1:1000, Santa Cruz
Biotechnology, Dallas, TX, sc-20780), NR2B (1:500, Molecular
Probes/Invitrogen, Life Technology, Carlsbad, CA, A-6474),
Bassoon (1:1000, Enzo, Farmingdale, NY, SAP7F407) MAP2 (1:400,
Santa Cruz Biotechnology, sc-20172), GFP (1:1000; Invitrogen, Life
Technology Corp. A-21311), and ApoER2 (1:1000, Sigma–Aldrich,
St. Louis, MO, A-3481) to detect the endogenous receptor. In
order to detect the tailless version of transfected ApoER2 an antiHA antibody (1.25 ng/ml, mouse) was used that detects the HA
epitope present at the N-terminal (extracellular domain) of the
receptor (Fuentealba et al., 2007). Cells were washed three times
with PBS, then incubated with the corresponding Alexaconjugated secondary antibodies (1:500, Life Technologies Corp.)
for 30 min at 37°C. Coverslips were mounted with Fluoromont-G
(Electron Microscopy Sciences, Hatfield, PA) and analyzed by
confocal laser microscopy (Olympus FV 1000). Images were
analyzed using NIH ImageJ software. Dual and triple
immunofluorescent images were captured by multitracking
imaging of each channel independently, to eliminate possible crosstalk between the different fluorochromes. For cluster
quantification, 8-bits images of maximal projections were analyzed
using plugins of the Fiji software. All clusters with a minimal
arbitrary gray level pixel intensity of 50 (out of 255) and a size
larger than 0.02 mm2 were analyzed on primary and secondary
dendrites of GFP-positive hippocampal neurons.
Stereotaxic injection
Adult mice were anesthetized with saline (5 ml saline/gram body
weight) containing 170 mg/kg ketamine plus 17 mg/kg xylazine.
Granular layer of dentate gyrus of adult mice (2–4 month-old
C57B6/SJL) were slowly injected with 0.2 ml of p1005 HSV-GFP
viruses (3 108 transducing units/ml) into the dentate gyrus
following coordinates: 1.5 mm lateral; 2 mm anteroposterior;
2.3 mm ventral from Bregma (Tashiro et al., 2006). We used
HSV-GFP viruses because the expression is robust and rapid (i.e.,
initiated 2–3 h post-injection) (Neve et al., 2005). Concomitant
with HSV injection, 2 ml of CR50 was co-injected bilaterally.
Morphological analysis in vivo
Three days post infection the animals were transcardially perfused
with 4% PFA in PBS. Afterwards, the brain was removed
immediately, post-fixed overnight, and cryopreserved in 30%
sucrose. The brains were cut serially in 40 mm sections on a
cryostat (Leica, CM 152S, Germany). For morphometric analysis,
at least 10 granular cells of the superior granular layer of the
dentate gyrus of each experimental condition were selected, if
they fulfilled the following criteria: (1) GFP positive signal along the
entire dendritic field with high signal-to-noise ratio for GFP, (2)
isolated from neighboring GFP infected granular cells, and (3) lack
of truncated dendrites.
For the dendritic architecture, low-magnification images were
acquired using a confocal laser scanning microscope Leica LSI
Macro-Zoom with 5 air objective (NA ¼ 0.08, LWD Plan
Apochromatic) plus optical zoom at 22, excitation with solid
state laser at 488 nm. Eight-bit TIFF images of 1024 1024 pixels
JOURNAL OF CELLULAR PHYSIOLOGY
were acquired with xy pixel size of 240 and 500 nm between
z-sections. Fifty to sixty z-sections were acquired depending on the
dendrite arbor of the neurons. Neurite Tracer of the Fiji software
was used to measure the dendritic architecture and Sholl analysis
plugins were used to evaluate the complexity of the dendritic
arbor. Specifically, supragranular neurons located in the dentate
gyrus were analyzed.
For spine analysis, high magnification images were acquired with
an UltraView RS spinning disk microscope (Perkin–Elmer) with a
100 oil objective (NA ¼ 1.3, C-Apochromat) with 1.6 optobar,
excitation with a 488 nm diode laser (Omicron), a 12-bit CCD
camera (Hamamatsu ORCA-ER), and Volocity 4.2 software
(Improvision). Sixteen-bit TIFF images of 1344 1024 pixels were
acquired with xy pixel size of 66 and 100 nm between z-sections.
Images were deconvolved, segmented and 3D reconstructions
made. For quantification of spine density, secondary dendrite
shafts (30 mm) were selected from supragranular neurons
located in the dentate gyrus. Segmentation of dendrites, spines,
and quantification of spine density was performed as describes
before (Tortosa et al., 2011; Posada-Duque et al., 2016). For
dendritic spine density, mushroom, stubby and thin spines, but not
filopodia, were quantified.
Statistical analyses
An ANOVA followed by the Bonferroni post hoc was used to
evaluate statistic significance between experimental groups.
Student’s t-test was applied when two populations of responses
were examined. In all figures, error bars represent the SEM;
P 0.05, P 0.01, P 0.001.
Results
Reelin surface expression robustly increases during in
vitro hippocampal neuron development
To elucidate reelin contributions to spine and dendrite
stabilities of mature hippocampal neurons, reelin expression
was first assessed during the development of hippocampal
cultures, between 2 and 20 days in vitro (DIV). To detect
predominantly intracellular levels of reelin, double
immunofluorescent staining was performing with antibodies
against reelin and MAP2 in cultures that were extensively
washed (to reduce surface reelin) and permeabilized with
Triton-X-100. While only a few reelin immunoreactive (IR)
positive neurons were detected during the first days of
development (3% at 2 DIV), a strong increase was observed
in mature cultures (32% at 20 DIV) (Fig. 1A and B). In parallel,
reelin-IR intensity strongly increased during hippocampal
culture maturation (Fig. 1C). Considering that reelin is
secreted by GABAergic neurons in hippocampal cultures to
function as an ECM signal molecule on neighboring neurons
(Gonzalez-Campo et al., 2009), subsequent analyses were
performed to determine reelin surface expression on
hippocampal neurons. For this, double immunofluorescent
staining with reelin and MAP2 was performed without
permeabilization conditions and extensive washing. Surface
reelin expression on hippocampal neuron dendrites increased
threefold between 7 and 20 DIV (Fig. 1D and E); the lack of
MAP2-IR validated that neurons were not permeabilized.
These results, together with previous analyses (Sinagra et al.,
2005; Gonzalez-Campo et al., 2009), indicate that reelin surface
expression robustly increases during hippocampal
development in vitro.
Reelin signaling interference with CR50 or GST-RAP
induces dendritogenesis in mature hippocampal neurons
We hypothesized that reelin signaling via the ApoER2/VLDRL/
PSD95 axis regulates neuronal architecture maintenance of
1189
1190
A M P U E R O E T A L.
Fig. 1. Reelin surface expression increases during development and is reduced by GST-RAP and CR50. (A–C) Intracellular and surface reelin
expression in hippocampal neurons significantly increase during development. Double immunostainings were performed on cultures
permeabilized with Triton-X-100 to detect intracellular and extracellular reelin expression. (A) Representative fluorescence microscopy
images show IR of reelin, the neuronal marker MAP2, and the merge of both. (B and C) Quantifications of (B) the percentage of reelin-IR
positive neurons and (C) the total Reelin-IR intensity, both relative to 2 DIV neurons. (D and E) Surface Reelin expression, particularly on the
dendritic branches (arrows), significantly increases during hippocampal development. Immunostainings were performed as in A, but without
permeabilization, as evidenced by a deficiency of MAP2-IR. (D) Representative confocal images with reelin and MAP2 immunostaining.
(E) Quantifications of total surface reelin-IR intensity (including soma and dendrites), relative to 7 DIV neurons. (F and G) Surface reelin
expression is significantly reduced by CR50 or GST-RAP bath application. Cultures (15 DIV) were incubated for 5 days with CR50 (2 mg/ml),
a reelin-neutralizing antibody, or GST-RAP (20 mM), a competitive inhibitor of the reelin receptors VLDLR and ApoER2. Reelin and MAP2
immunostainings were performed without permeabilization. (F) Representative confocal images with Reelin and MAP2 immunostainings in
untreated (control) or treated (CR50 or GST-RAP) 20 DIV hippocampal cultures. (G) Quantifications of the surface reelin-IR intensity in
CR50- or GST-RAP-treated cultures, relative to untreated neurons. Insets: amplifications of boxed areas. For each developmental stage and
condition, at least 10 neurons, obtained from three independent experiments, were analyzed. Figures show Means S.E.M. P 0.05,
P 0.01, P 0.001 (One-way ANOVA followed by Bonferroni post hoc test).
JOURNAL OF CELLULAR PHYSIOLOGY
REELIN SIGNALING MAINTAINS ADULT NETWORK STABILITY
mature hippocampal neurons, and, consequently, reelin
signaling pathway interference should reactivate dendritic
outgrowth of mature neurons. To test this hypothesis,
two different molecules were used to block the reelin
signaling pathway: CR50, an antibody that interacts with
and neutralizes reelin (Utsunomiya-Tate et al., 2000), and
GST-RAP, in which the RAP protein inhibits reelin binding
to the ApoER2 and VLDLR receptors (Hiesberger et al.,
1999). Applications of either CR50 (2 mg/ml) or GST-RAP
(20 mM) to mature hippocampal neurons for 5 days
significantly decreased reelin surface expressions twofold
(Fig. 1F and G).
Next, 15 DIV hippocampal neurons were transfected with
GFP and cultures were incubated with CR50 or GST-RAP for
5 days to assess the dendritic architecture of fixed neurons at
20 DIV. Mature neurons treated with either CR50 or GST-RAP
exhibited a much more complex dendritic architecture, with
significant increases in the summed quantity of secondary and
total dendritic branches outgrowths relative to control
neurons (Fig. 2A and B). Additionally, CR50, but not GST-RAP,
Fig. 2. Reelin signaling interference with CR50 or GST-RAP increases dendritogenesis in mature hippocampal neurons. Treatment with
CR50 or GST-RAP significantly increases the dendritic architecture complexity of mature hippocampal neurons. Neurons at 15 DIV were
transfected with a plasmid coding for GFP to visualize the morphology. Cultures were untreated (control) or treated with CR50 or GST-RAP
at 15 DIV, and fixed at 20 DIV (as in Fig. 1F). (A) Representative contrast-enhanced images of untreated (control) or treated (CR50 or GSTRAP) 20 DIV hippocampal cultures. Insets: amplification of boxed areas. (B) Quantifications reveal that bath application of CR50 or GST-RAP
significantly increases the summed outgrowth of secondary and total dendritic branches, relative to control neurons. (C) Quantifications show
that CR50 treatment significantly increases the number of the secondary and total dendritic branches, relative to control neurons.
(D) Quantifications show that neither CR50 nor GST-RAP significantly alters average branch lengths. (E) Sholl analyses reveal that, relative to
untreated neurons, CR50 increases branches throughout the dendritic arbor, while GST-RAP enhances branching at larger distances from the
soma (>75 mm). For each condition, at least 10 neurons, obtained from three independent experiments, were analyzed. Figures show
Means S.E.M. P 0.05, P 0.01 (One-way ANOVA followed by Bonferroni post hoc test).
JOURNAL OF CELLULAR PHYSIOLOGY
1191
1192
A M P U E R O E T A L.
significantly increased the quantity of secondary and total
dendritic branches relative to control neurons (Fig. 2C).
Neither CR50 nor GST-RAP significantly increased average
branch lengths (Fig. 2D). Sholl analyses further revealed that,
relative to untreated neurons, CR50 increased branches
throughout the dendritic arbor, while GST-RAP enhanced
branching at distances further from the soma (>75 mm)
(Fig. 2E). These findings indicate that reelin signaling pathway
activation, via ApoER2 and/or VLDRs, maintains the neuronal
architecture of mature neurons.
Reelin interference with CR50 or GST-RAP in developing
hippocampal neurons leads to a decrease in
dendritogenesis
In contrast to the current findings, previous in vitro studies
with dissociated hippocampal cultures derived from reeler mice
embryos indicate that reelin promotes the dendritic growth
and complexity of immature neurons (Niu et al., 2004; Jossin
and Goffinet, 2007; MacLaurin et al., 2007; Matsuki et al., 2008);
however, in all prior studies, immature neurons were analyzed.
To directly investigate if reelin-mediated effects are
determined by the neuronal development stage,
dendritogenesis was tested in developing hippocampal
neurons. Specifically, GFP-transfected neurons were incubated
with CR50 or GST-RAP at 7 DIV and fixed at 12 DIV to assess
dendritic architecture. Both CR50 and GST-RAP strongly
impacted dendritic tree complexity (Fig. 3A) by significantly
decreasing total dendrite outgrowth of the primary, secondary,
and tertiary dendritic branches relative to control neurons
(Fig. 3B). Moreover, CR50 and GST-RAP decreased the
number of tertiary and total branches (Fig. 3C). No effects on
average dendritic length were observed (Fig. 3D).
Complementary, Sholl analysis showed that reelin blockage
Fig. 3. Reelin signaling interference with CR50 or GST-RAP decreases dendritogenesis in early developed hippocampal neurons. Treatment
with CR50 or GST-RAP significantly decreases the dendritic architecture complexity of intermediate developed hippocampal neurons.
Similar to Figure 2, but at now at earlier developmental stages, GFP-transfected neurons were maintained untreated (control) or treated at
7 DIV with CR50 or GST-RAP, and fixed at 12 DIV. (A) Representative contrast-enhanced images of control neurons or neurons treated with
CR50 or GST-RAP. Insets: amplification of boxed areas. (B) Quantifications reveal that CR50 or GST-RAP applications significantly decrease
the summed outgrowth of the primary, secondary, tertiary, and total dendritic branches, relative to control neurons. (C) Quantifications
show that CR50 and GST-RAP significantly decrease the number of the tertiary and total dendritic branches, relative to control neurons.
(D) Quantifications show that CR50 and GST-RAP do not significantly alter averaged branch lengths. (E) Sholl analyses reveal that CR50 and
GST-RAP reduce branches throughout the dendritic arbor, relative to untreated neurons. For each condition, at least 10 neurons, obtained
from three independent experiments, were analyzed. Figures show Means S.E.M. P 0.05, P 0.01 (One-way ANOVA followed by
Bonferroni post hoc test).
JOURNAL OF CELLULAR PHYSIOLOGY
REELIN SIGNALING MAINTAINS ADULT NETWORK STABILITY
reduced branching throughout the dendritic arbor relative to
untreated neurons (Fig. 3E).
Expression of the dominant negative form of ApoER2,
ApoER2-tailless, increases dendritogenesis in mature
hippocampal neurons
To strengthen the results that reelin signaling interference
reactivates dendritogenesis in mature hippocampal neurons,
cultures were transfected at 15 DIV with GFP and the
dominant negative form of the ApoER2, ApoER2-tailless, and
the dendritic architecture was measured at 20 DIV. ApoER2tailless is able to bind reelin but because it lacks the cytoplasmic
domain required to interact with PSD95, this construct
interferes with the reelin/ApoER2/PSD95 signaling axis (Beffert
et al., 2005; Cuitino et al., 2005; Fuentealba et al., 2007).
The expression of ApoER2-tailless, detected with an anti-HA
antibody (Fig. 4A), resulted in a more complex dendritic
architecture, represented by the morphology of GFPtransfected cells (Fig. 4B). Quantifications further revealed that
ApoER2-tailless-transfected neurons had significantly
increased summed outgrowths (Fig. 4C), as well as secondary
and total dendrites quantities (Fig. 4D) relative to control.
No changes were observed in average dendrite lengths after
ApoER2-tailless expression (Fig. 4E). Together these results
show that dendritogenesis is reactivated in mature neurons
through reelin signaling interference with the use of three
different approaches, CR50, GST-RAP, and ApoER2-tailless.
Reelin signaling interference decreases PSD95-IR clusters
and mature spines on adult neurons
The obtained ApoER2-tailless findings indicate a key role
for PSD95 in the reelin-mediated dendritic architecture
maintenance in mature neurons. As the most abundant
scaffolding protein in the excitatory postsynaptic density,
PSD95 drives the maturation of glutamatergic synapses and
spines (El-Husseini et al., 2000; Losi et al., 2003; Ehrlich et al.,
2007; Elias et al., 2008) and induces dendritic arbor
refinement (Charych et al., 2006; Henriquez et al., 2013;
Bustos et al., 2014). Given the above, and that reduced
synaptic PSD95 levels lead to dendritogenesis reactivation in
mature hippocampal neurons (Bustos et al., 2014),
approaches were used to directly determine if reelin signaling
interference leads to reduced synaptic PSD95 levels. For this,
GFP-transfected neurons were incubated at 15 DIV with
Fig. 4. Expression of the dominant negative form of ApoER2, ApoER2-tailless, increases dendritogenesis in mature hippocampal neurons.
Treatment with ApoER2-tailess significantly increases the dendritic architecture complexity of mature hippocampal neurons. Neurons at
15 DIV were transfected with a plasmid coding for GFP alone (control) or together with a plasmid coding for ApoER2-tailess, a construct that
lacks the C-terminal tail and that is required to interact with PSD95, and contains a HA-tag. Cultures were fixed at 20 DIV to detect ApoER2IR and HA-tag-IR by immunostaining (A), and to analyze the neuronal morphology of contrast-enhanced images (B–E). (A) Representative
confocal images show that ApoER2-tailess-transfected neurons display IR for the HA-tag. (B) Representative contrast-enhanced images show
that, relative to control neurons, ApoER2-tailess-transfected neurons display a more complex dendritic architecture. (C) Quantifications
reveal that ApoER2-tailess expression significantly increases the summed outgrowth of the secondary and total dendritic branches, relative to
control neurons. (D) Quantifications show that ApoER2-tailess expression significantly increases the number of secondary and total dendritic
branches, relative to control neurons. (E) Quantifications show that ApoER2-tailess does not significantly alter average branch lengths. For
each condition, at least 10 neurons, obtained from three independent experiments, were analyzed. Figures show Means S.E.M. P < 0.05
(t-test).
JOURNAL OF CELLULAR PHYSIOLOGY
1193
1194
A M P U E R O E T A L.
CR50 or GST-RAP, and at 20 DIV, cultures were fixed and
double immunofluorescent stained with specific antibodies to
detect IR for PSD95 and the mature presynaptic protein
synapsin 1 (Syn1). In accordance with previous studies (Perez
de Arce et al., 2010; Bustos et al., 2014), untreated mature
hippocampal neurons displayed abundant PSD95-IR clusters
on the primary and secondary dendritic branches (Fig. 5A–C);
the close opposition of PSD95-IR clusters to syn1-IR puncta
further indicates that PSD95 clusters in mature neurons were
post-synaptically localized (insets Fig. 5B and C). In contrast,
and as expected, mature neurons treated with either CR50 or
GST-RAP displayed a significant decrease in the quantity of
PSD95-IR clusters on secondary branches (Fig. 5B). A
significant reduction in the number of PSD95 clusters was also
observed on primary branches of neurons treated with GSTRAP, while CR50-treated cultures did not evidence a
significant reduction (Fig. 5C). Treatment with CR50 and
GST-RAP also reduced the number of Syn1 clusters on axon
terminals contacting the postsynaptic sites of GFP-transfected
neurons (Fig. 5B and C). Spine morphology analysis further
demonstrated that CR50 and particularly GST-RAP
decreased the density of mature spines while increasing
filopodia-like structures (Fig. 5D). Given that PSD95 knockdown leads to mature spine loss (Nakagawa et al., 2004;
Bustos et al., 2014), the present findings indicate that reelin
signaling is required to hold PSD95 at the postsynaptic
density, and thereby maintain mature spines.
Reelin signaling blockage increases NR2B-IR clusters on
dendrites of mature neurons
In addition to changes in synaptic PSD95 levels, several postsynaptic density components are altered when reelin is
deficient (Groc et al., 2007; Ventruti et al., 2011). Considering
previous reports that PSD95 knock-down alone cannot
reactivate dendritic growth in adult neurons unless synaptic
NR2B-NMDARs are inserted into the synapse (Sepulveda
et al., 2010; Bustos et al., 2014), subsequent analyses were
performed to determine if reelin signaling blockage increases
synaptic NR2B-NMDARs clusters. For this, GFP-transfected
neurons were incubated at 15 DIV with CR50 or GST-RAP, and
at 20 DIV, cultures were fixed and double immunofluorescent
stained with specific antibodies to detect IR for the NR2B
subunit of the NMDAR and the mature presynaptic protein
Bassoon. In accordance with previous studies (Li et al., 1998;
Bustos et al., 2014), very few NR2B-IR clusters were present
on the primary and secondary dendritic branches of untreated
mature hippocampal neurons (Fig. 6A–C). In contrast, and as
expected, mature neurons treated with CR50 showed
significant increased NR2B-IR cluster quantities on both
secondary (Fig. 6B) and primary (Fig. 6C) branches. GST-RAP
also significantly increased NR2B-IR cluster quantities on
primary branches (Fig. 6C), but significance was not reached
regarding secondary branches (Fig. 6B). Similar to syn1-IR
clusters, treatment with CR50 also reduced Bassoon-IR cluster
quantities on axon terminals contacting the postsynaptic sites
of GFP-transfected neurons (Fig. 6B and C). Therefore, the
current findings indicate that reelin signaling interference in
mature neurons leads loss of mature spines and synaptic
contacts, and in parallel, to the synaptic insertion of NR2BNMDARs.
Reelin signaling interference induces structural plasticity
in dentate gyrus neurons in vivo
Finally, to determine similarities with in vitro findings, the role
of reelin signaling in controlling spine and dendrite morphology
in adult hippocampal neurons in vivo was assessed. To visualize
the morphology of dentate gyrus neurons in the adult
JOURNAL OF CELLULAR PHYSIOLOGY
hippocampus, viral herpes simplex virus (HSV) particles
expressing GFP (HSV-GFP) were bilaterally injected by
steoreotaxic surgery in 8–12-week-old mice (Fig. 7A). While
the dentate gyrus covers the supragranular, infragranular, and
hilus layers, only neurons located within the supragranular
layer were analyzed. To interfere with the reelin signaling
pathway, a single dose of CR50 (2 ml, 2 mg/ml) was
concomitantly performed with the HSV-GFP injection. Three
days post-injection, animals were perfused (4% PFA), and
coronal slices were prepared from hippocampi to analyze the
morphology and density of dendrites (Fig. 7B–D) and spines
(Fig. 7E and F). Consistent with in vitro findings that reelin
signaling interference induced dendritogenesis in mature
hippocampal neurons, mature granular neurons treated with
CR50 exhibited a much more complex dendritic architecture,
with significantly increases in total dendritic length relative to
control neurons (Fig. 7B and C). Sholl analyses further revealed
that, relative to untreated neurons, CR50-treated granular
neurons increased dendritic branch quantities located >50 mm
from the soma, at the expense of more proximally located
branches (Fig. 7D).
To determine the role of reelin in spine density, dendritic
protrusions on second order branches of selected SG granular
neurons were analyzed. Under control conditions, mature
granular neurons exhibited substantial quantities of thin,
stubby, and mushroom spines, with few filopodia (Fig. 7E). In
contrast, CR50-treated neurons displayed dendrites
decorated with less mushroom-like spines, leading to a
significant decrease in spine density (Fig. 7E and F). Additionally,
CR50-treatment resulted in the development of filopodia-like
structures (Fig. 7E). Altogether, the obtained data show that
CR50 treatment in vivo leads to increased dendrite arbor
complexity and a loss of dendritic spine density in mature
granular neurons, indicating that the reelin signaling pathway is
needed to maintain neurons is a refined mature state to limit
plasticity.
Discussion
We investigated the role of reelin signaling in the neuronal
architecture of mature hippocampal neurons and demonstrate
that reelin signaling interference results in increased
dendritogenesis and reduced spine density in mature
hippocampal neurons in vitro (20 DIV) and in mature dentate
granular hippocampal neurons in vivo (2–4-month-old mice).
Our data underscore the importance of reelin signaling
through ApoER2 and PSD95 in maintaining neuronal circuit
stability once neurons have matured.
Our results showing that reelin signaling interference
increases dendritic growth and branching contrast with
previous studies that consistently show that deficiency in reelin
(Niu et al., 2004; Jossin and Goffinet, 2007), ApoER2 or VLDLR
(Trommsdorff et al., 1999), or Dab-1 (Tabata and Nakajima,
2002; Olson et al., 2006; MacLaurin et al., 2007) impair dendrite
development. However, all prior research on the role of reelin
signaling has been performed in either immature neurons or in
mature neurons with disrupted reelin signaling since embryonic
stages. Considering that dendrites dynamically form and prune
in the developing brain, before becoming largely stable in the
mature neuronal circuitry (Chow et al., 2009; Koleske, 2013),
these aforementioned studies indicate that reelin signaling is
required for dendritic tree formation in the developing brain.
Consistent with this, we showed that the same tools to
interfere in the reelin signaling responsible for dendritogenesis
in mature neurons resulted in reduced dendritic growth and
branching in developing immature hippocampal neurons. Based
on existing literature and the results obtained in the current
study, we propose that the neuronal development stage is
critical for reelin effects on dendritic morphology; thus, reelin
REELIN SIGNALING MAINTAINS ADULT NETWORK STABILITY
Fig. 5. Reelin signaling interference with CR50 or GST-RAP decreases PSD95-IR clusters and mature spines in mature hippocampal neurons.
(A) Representative confocal image of a GFP-transfected mature hippocampal neuron (20 DIV; green) with double immunostainings to detect
IR of PSD95, the presynaptic marker synapsin I (syn1), and the merge of both. An example of a primary (1st) and secondary (2nd) dendritic
branch is indicated. B and C) Treatment of mature hippocampal neurons with CR50 or GST-RAP significantly decreases PSD95-IR clusters in
secondary, but not in primary, dendritic branches. Similar to Figure 2, neurons at 15 DIV were transfected GFP, treated with CR50 or GSTRAP, and fixed at 20 DIV. Representative confocal images of (B) secondary or (C) primary dendritic branches showing IR of PSD95, the
presynaptic marker syn1, and the merge of both. Quantifications are also shown for the total number of PSD95-IR clusters (left graphs)
and syn1-IR clusters (right graphs) per 100 mm of dendritic branches of untreated (control) and treated (CR50 or GST-RAP) neurons. (D)
Treatment of mature hippocampal neurons with CR50, and particularly with GST-RAP, decreases dendritic mushroom-shape spines
(arrows), while increasing filopodia-like structures processes (arrowheads). Developing quaternary processes are also evident in the GST-RAP
treated neurons ( ). Representative confocal images of secondary dendritic branches of untreated (control) and treated (CR50 or GST-RAP)
neurons. For each condition, at least 10 neurons, obtained from three independent experiments, were analyzed. Figures show Means S.E.M.
P 0.05, P 0.01 (One-way ANOVA followed by Bonferroni post hoc test).
JOURNAL OF CELLULAR PHYSIOLOGY
1195
1196
A M P U E R O E T A L.
Fig. 6. Reelin signaling interference with CR50 or GST-RAP increases NR2B-IR clusters and mature spines in mature hippocampal neurons.
(A) Representative confocal image of a GFP-transfected mature hippocampal neuron (20 DIV; green) with double immunostainings to detect
IR of NR2B, the presynaptic marker Bassoon, and the merge of both. An example of a primary (1st) and secondary (2nd) dendritic branch is
indicated. (B and C) Treatment of mature hippocampal neurons with CR50 significantly increases NR2B-IR clusters in secondary and primary
dendritic branches. Application of GST-RAP significantly increases NR2B-IR clusters in primary, but not secondary, dendritic branches.
Similar to Figure 2, neurons at 15 DIV were transfected GFP, treated with CR50 or GST-RAP, and fixed at 20 DIV. Representative confocal
images of (B) secondary or (C) primary dendritic branches showing IR of NR2B, the presynaptic marker Bassoon, and the merge of both.
Quantifications are also shown for the total number of NR2B-IR clusters (left graphs) and Bassoon-IR clusters (right graphs) per 100 mm of
dendritic branches of untreated (control) and treated (CR50 or GST-RAP) neurons. For each condition, at least 10 neurons, obtained from
three independent experiments, were analyzed. Figures show Means S.E.M. P 0.05, P 0.01 (One-way ANOVA followed by Bonferroni
post hoc test).
signaling promotes dendritogenesis in immature neurons but
favors dendrite maintenance in mature neurons.
In contrast to dendritic branches, individual dendritic
spines continue to dynamically form, alter in morphology,
and prune during mature nervous system refinement
(Koleske, 2013). Since various forms of synaptic plasticity
(i.e., learning and memory) are defined by experiencedependent structural plasticity (Nishiyama and Yasuda,
2015), considerable efforts have been made to elucidate the
role of reelin in dendritic spine morphology. In particular,
detailed morphological analyses of reeler mutant mice
demonstrated that the dendritic spine density is strongly
reduced in juvenile (P21) and adult (P60–80) hippocampal
and motor neurons (Liu et al., 2001; Niu et al., 2008).
Similarly, spine density in juvenile (1-month-old) cortical
pyramidal neurons was significantly decreased in ApoER2
deficient mice (Dumanis et al., 2011). However, all these
studies used embryonic models for reelin or ApoER2
JOURNAL OF CELLULAR PHYSIOLOGY
deficiency, thereby being unable to determine if reelin
signaling is required for the formation, maturation, or
maintenance of dendritic spines in mature hippocampal
neurons. We overcame this limitation by specifically and
directly manipulating reelin signaling in mature neurons.
Reelin signaling interference in mature hippocampal neurons
in vitro and in vivo led to a robust decrease in spine density,
strongly indicating that reelin signaling is necessary for
dendritic spine stability in mature neurons. These findings
are consistent with previous studies that increased reelin
either through exogenous applications or overexpression of
the reelin protein gene. Specifically, conditioned mice with
reelin overexpression in the adult brain evidenced the
development of hippocampal spines exhibiting a mushroomtype appearance and marked hypertrophy of the spine head
(Pujadas et al., 2010). Additionally, exogenous reelin
applications in the adult brain increase the dendritic spine
density of CA1 pyramidal neurons (Rogers et al., 2011).
REELIN SIGNALING MAINTAINS ADULT NETWORK STABILITY
Fig. 7. Reelin signaling interference in vivo with CR50 increases dendritogenesis and reduces spine density of dentate granule neurons in
adult mice. (A) Stereotaxic injection of HSV particles expressing GFP (HSV-GFP) was performed concomitantly with a single administration
of CR50 (2 ml, 2 mg/ml) into the dentate gyrus (DG; indicated by ) of adult mice (2–4-month-old). DAPI was used to stain nuclei. The superior
blade of the DG (supra DG) is indicated as well as granular layer (GL), inner molecular layer (IML), medium molecular layer (MML), outer
molecular layer (OML), and the hilus (H). Granular cells in the supra-granular layer GL were selected for morphological analyses. Three days
post-injection, animals were perfused (4% PFA), and coronal slices were prepared from hippocampi to analyze (B–D) the dendritic
architecture, as well as (E and F) spine morphology and density. (B–D) CR50 treatment induces dendritogenesis of dentate granule neurons in
vivo. (B) Representative confocal images of maximal z-projections, as well as the corresponding 3D reconstruction, are shown from isolated
control and CR50-treated granule neurons located within the dentate gyrus layer. (C) Quantifications show that total dendritic length of
CR50-treated granule neurons is significantly increased, relative to control neurons. (D) Sholl analyses reveal that, relative to untreated
neurons, CR50-treated granule neurons increase the number of dendritic branches located at >75 mm from the soma, at the expense of
proximally located branches (<50 mm). (E and F) CR50 treatment leads to reduced spine density of DG neurons. (E) Representative
convoluted confocal images are shown from the secondary dendritic branch from control and CR50-treated granule neurons. CR50 treatment
reduces dendritic mushroom-shape spines (arrows), while increasing filopodia-like structures (arrowheads). (F) Quantifications show that
total dendritic spine density is significantly decreased, relative to control neurons. For each condition, at least 10 neurons, obtained from
three independent experiments, were analyzed. Figures show Means S.E.M. P 0.05 (t-test).
Activation of the reelin signaling pathway that leads to
increased dendritic growth and arborization in developing
neurons is dependent on the mTor pathway and is mediated by
the activation of Dab-1, PI3K, and Akt (Jossin and Goffinet,
2007). Additionally, the amyloid precursor protein (APP; Hoe
et al., 2009) and a cdc42/Rac1 guanine nucleotide exchange
factor (aPIX/Arhgef6; Meseke et al., 2013) have been implicated
in the reelin-mediated dendritic outgrowth of immature
hippocampal neurons (Lee and D’Arcangelo, 2016). Our findings
that reelin signaling interference in mature hippocampal neurons
result in dendritogenesis argue against the involvement of these
downstream reelin molecules. Rather, our data showing that the
NR2B receptor subunit is re-inserted into synapses after reelin
signaling impairment suggests that synaptic NR2B-NMDARs and
underlying signaling pathways are involved in dendritic
reactivation. This idea is substantiated by the findings that
synaptic NR2B-NMDARs induce dendrite formation in
developing neurons (Espinosa et al., 2009; Sepulveda et al., 2010;
Bustos et al., 2014), as well as by reports that a synaptic
re-insertion of NR2B-NMDARs through knockdown of PSD95
JOURNAL OF CELLULAR PHYSIOLOGY
reactivates robust dendritic branching in mature hippocampal
neurons in vitro (Bustos et al., 2014). Furthermore, because the
direct coupling of NR2B-NMDARs to RasGRF1 (Krapivinsky
et al., 2003) or CaMKII (Barria and Malinow, 2005) is required
for dendritogenesis in hippocampal neurons through ERK/CREB
signaling pathway activation (van Zundert et al., 2004; Sepulveda
et al., 2010), it is likely that these signal transduction molecules
are also involved in reactivation of dendritogenesis in mature
neurons with impaired reelin signaling. Further studies are
needed to identify the signaling pathway(s) and cytoskeletal
component(s) (Kulkarni and Firestein, 2012; Koleske, 2013)
underlying dendrite stability and reactivation in mature neurons.
The developmental exchange from predominantly NR2B- to
NR2A-containing NMDARs is not only mediated by neuronal
activity and experience (van Zundert et al., 2004), but also
involves the extracellular environment, including reelin (Sinagra
et al., 2005; Groc et al., 2007; Gonzalez-Campo et al., 2009;
Ventruti et al., 2011), modulating hereby synaptic plasticity and
cognition (Weeber et al., 2002; Beffert et al., 2005; Qiu et al.,
2006). Functional electrophysiological and calcium studies have
1197
1198
A M P U E R O E T A L.
particularly indicated that blocking the expression, release, or
function of reelin in intermediate developing hippocampal
neurons (10–12 DIV) prevents the maturation-dependent
synaptic reduction of NR2B-NMDAR (Sinagra et al., 2005; Groc
et al., 2007; Gonzalez-Campo et al., 2009). Moreover, singleparticle tracking has shown that reducing extracellular reelin
levels reduces NR2B-NMDARs surface mobility and increases
the time spent by NR2B-NMDARs in synapses (Groc et al.,
2007). It is possible that reelin signaling interference in mature
neurons similarly alters NR2B-NMDARs surface mobility and
might increase synaptic dwell time; however, differences would
exist in the underlying mechanisms. In particular, in the current
study we show that treatment of mature hippocampal cultures
with both the reelin-sequestering antibody CR50 and the
lipoprotein receptor antagonist GST-RAP increased synaptic
NR2B clusters, indicating that ApoER2 and/or VLDR2 are
required for this process. In contrast, Groc et al. (2007) showed
that treatment of developing hippocampal neurons with CR50,
but not with GST-RAP, affects synaptic localization of NR2BNMDARs, ruling out the involvement of ApoER2/VLDLR
receptors. In fact, reelin function impairment by an antibody
against b1-class integrins indicates that the non-classical reelin
receptor a3b1 regulates NR2B-NMDAR surface distributions
(Groc et al., 2007). Further studies are needed to elucidate
the mechanisms underlying the differential participations of
the reelin/ApoER2-VLDLR and reelin/a3b1pathways in the
subunit composition of synaptic NMDARs during diverse
developmental stages.
In conclusion, the presented results not only underscore the
importance of reelin in maintaining adult network stability but
also reveal important potential therapeutic implications.
Specifically, reeling signaling interference can be used to
reactivate dendritogenesis in neurological and cognitive
disorders in which dendritic tree complexity is limited, such as
in depression, Alzheimer’s disease, schizophrenia, and stroke,
among others (Kulkarni and Firestein, 2012). Of particular
therapeutic potential is the finding that a single injection of the
CR50 antibody, known to sequester extracellular reelin, is
sufficient to significantly reactivate structural dendritogenesis
in the mouse brain. On the other hand, caution should be
exercised with Alzheimer’s disease patients as a recent study
evidences that reelin protects the brain against pathological
levels of toxic amyloid b species (Lane-Donovan et al., 2015).
Acknowledgments
We thank Pamela Farfan and Luis Melo for their technical
support. This work was supported by several Chilean grants
that are acknowledged in the title page. This study was
supported by the funds FONDECYT (3130582) to EA,
FONDECYT (1140301) to BvZ, UNAB Nucleus (DI-603-14N)
to BvZ, FONDEQUIP (EQM 140166) to BvZ, CONICYT
(201161486) to NJ, BNI (ICM P09-015-F) to SH, VISUAL D
(ACT1402) to SH, FONDECYT (1151029) to SH, and
FONDECYT (1150444) to MPM.
Literature Cited
Barria A, Malinow R. 2005. NMDA receptor subunit composition controls synaptic plasticity
by regulating binding to CaMKII. Neuron 48:289–301.
Beffert U, Weeber EJ, Durudas A, Qiu S, Masiulis I, Sweatt JD, Wei Li, Adelmann G,
Frotscher M, Hammer R, Herz J. 2005. Modulation of synaptic plasticity and memory by
Reelin involves differential splicing of the lipoprotein receptor Apoer2. Neuron
47:567–579.
Be€ıque J-C, Lin D-T, Kang M-G, Aizawa H, Takamiya K, Huganir RL. 2006. Synapse-specipfic
regulation of AMPA receptor function by PSD-95. Proc Natl Acad Sci 103:19535–19540.
Bustos FJ, Varela-Nallar L, Campos M, Henriquez B, Phillips M, Opazo C, Aguayo L,
Montecino M, Inostroza E, Constantine-Paton M, van Zundert B. 2014. PSD95 suppresses
dendritic arbor development in mature hippocampal neurons by occluding the clustering
of NR2B-NMDA receptors. PLoS ONE 9:e94037.
Charych EI, Akum BF, Goldberg JS, J€
ornsten RJ, Rongo C, Zheng JQ, Firestein B. 2006.
Activity-independent regulation of dendrite patterning by postsynaptic density protein
PSD-95. J Neurosci 26:10164–10176.
JOURNAL OF CELLULAR PHYSIOLOGY
Chen Y, Beffert U, Ertunc M, Tang T, Kavalali ET, Bezprozvanny I, Herz J. 2005. Reelin
modulates NMDA receptor activity in cortical neuron. J Neurosci 25:8209–8216.
Chow DK, Groszer M, Pribadi M, Machniki M, Carmichael ST, Liu X, Trachtenberg J. 2009.
Laminar and compartmental regulation of dendritic growth in mature cortex. Nat
Neurosci 12:116–118.
Cline H, Haas K. 2008. The regulation of dendritic arbor development and plasticity by
glutamatergic synaptic input: A review of the synaptotrophic hypothesis. J Physiol
586:1509–1517.
Cuchillo-Iba~nez I, Balmaceda V, Botella-L
opez A, Rabano A, Avila J, Saez-Valero J. 2013.
Beta-amyloid impairs reelin signaling. PLoS ONE 8:e72297.
Cuitino L, Matute R, Bu G, Inestrosa NC, Marzolo MP. 2005. ApoER2 is endocytosed by a
clathrin-mediated process involving the adaptor protein Dab2 independent of its rafts’
association. Traffic 6:820–838.
Cull-Candy SG, Leszkiewicz DN. 2004. Role of distinct NMDA receptor subtypes at central
synapses. Sci STKE 255:re16.
Dumanis S, Cha H-J, Song JM, Trotter JH, Spitzer M, Lee J-Y, Weeber E, Turner RS, Pak D,
Rebeck GW, Hoe HS. 2011. ApoE receptor 2 regulates synapse and dendritic spine
formation. PLoS ONE 6:e17203.
Ehrlich I, Klein M, Rumpel S, Malinow R. 2007. PSD-95 is required for activity-driven synapse
stabilization. Proc Natl Acad Sci 104:4176–4181.
Elias G, Elias L, Apostolides P, Kriegstein A, Nicoll R. 2008. Differential trafficking of AMPA
and NMDA receptors by SAP102 and PSD-95 underlies. Proc Natl Acad Sci
105:20953–20958.
El-Husseini AE, Schnell E, Chetkovich DM, Nicoll RA, Bredt DS. 2000. PSD-95 involvement
in maturation of excitatory synapses. Science 290:1364–1368.
Espinosa J, Wheeler D, Tsien RW, Luo L. 2009. Uncoupling dendrite growth and patterning:
Single cell knockout analysis of NMDA receptor 2B. Neuron 62:205–217.
Fuentealba RA, Barría MI, Lee J, Cam J, Araya C, Escudero CA, Inestrosa N, Bronfman F, Bu
G, Marzolo MP. 2007. ApoER2 expression increases Ab production while decreasing
Amyloid Precursor Protein (APP) endocytosis. Possible role in the partitioning of APP into
lipid rafts and in the regulation of g-secretase activity. Mol Neurodegener 2:14.
Gonzalez-Campo C, Sinagra M, Verrier D, Manzoni OJ, Chavis P. 2009. Reelin secreted by
GABAergic neurons regulates glutamate receptor homeostasis. PLoS ONE 4:e5505.
Groc L, Choquet D, Stephenson FA, Verrier D, Manzoni OJ, Chavis P. 2007. NMDA
receptor surface trafficking and synaptic subunit composition are developmentally
regulated by the extracellular matrix protein Reelin. J Neurosci 27:10165–10175.
Henriquez B, Bustos FJ, Aguilar R, Becerra A, Simon F, Montecino M, van Zundert B. 2013.
Ezh1 and Ezh2 differentially regulate PSD-95 gene transcription in developing hippocampal
neurons. Mol Cell Neurosci 57:130–143.
Herz J, Chen Y. 2006. Reelin, lipoprotein receptors and synaptic plasticity. Nat Rev Neurosci
7:850–859.
Hiesberger T, Trommsdorff M, Howell BW, Goffinet A, Mumby MC, Cooper JA, Herz J.
1999. Direct binding of Reelin to VLDL receptor and ApoE receptor 2 induces tyrosine
phosphorylation of disabled-1 and modulates tau phosphorylation. Neuron 24:481–489.
Hoe H, Pocivavsek A, Chakraborty G, Fu Z, Vicini S, Ehlers MD. 2006. Apolipoprotein E
receptor 2 interactions with the N-Methyl-D-aspartate receptor. J Biol Chem
281:3425–3431.
Hoe HS, Lee KJ, Carney RS, Lee J, Markova A, Lee JY, Howell B, Hyman B, Pak D, Bu G,
Rebeck GW. 2009. Interaction of reelin with amyloid precursor protein promotes neurite
outgrowth. J Neurosci 29:7459–7473.
Jan Y-N, Jan LY. 2010. Branching out: Mechanisms of dendritic arborization. Nat Rev
Neurosci 11:316–328.
Jossin Y, Goffinet AM. 2007. Reelin signals through phosphatidylinositol 3-kinase and Akt to
control cortical development and through mTor to regulate dendritic growth. Mol Cell
Biol 27:7113–7124.
Koleske AJ. 2013. Molecular mechanisms of dendrite stability. Nat Rev Neurosci
14:536–550.
Krapivinsky G, Krapivinsky L, Manasian Y, Ivanov A, Tyzio R, Pellegrino C, Ben-Ari Y,
Clapham DE, Medina I. 2003. The NMDA receptor is coupled to the ERK pathway by a
direct interaction between NR2B and RasGRF1. Neuron 40:775–784.
Kulkarni V, Firestein BL. 2012. The dendritic tree and brain disorders. Mol Cell Neurosci
50:10–20.
Lane-Donovan C, Philips GT, Wasser CR, Durakoglugil MS, Masiulis I, Upadhaya A,
Pohlkamp T, Coskun C, Kotti T, Steller L, Hammer R, Frotscher M, Bock H, Herz J. 2015.
Reelin protects against amyloid b toxicity in vivo. Sci Signal 8:ra67.
Lee GH, D’Arcangelo G. 2016. New insights into reelin-mediated signaling pathways. Front
Cell Neurosci 10:1–8.
Li J, Wang Y, Wolfe B, Krueger K, Corsi L, Stocca G, Vicini S. 1998. Developmental changes
in localization of NMDA receptor subunits in primary cultures of cortical neurons. Eur J
Neurosci 10:1704–1715.
Liu WS, Pesold C, Rodriguez M a, Carboni G, Auta J, Lacor P, Condie B, Guidotti A, Costa E.
2001. Down-regulation of dendritic spine and glutamic acid decarboxylase 67 expressions
in the reelin haploinsufficient heterozygous reeler mouse. Proc Natl Acad Sci
98:3477–3482.
Losi G, Prybylowski K, Fu Z, Luo J, Wenthold RJ, Vicini S. 2003. PSD-95 regulates NMDA
receptors in developing cerebellar granule neurons of the rat. J Physiol 548:21–29.
MacLaurin S, Krucker T, Fish K. 2007. Hippocampal dendritic arbor growth in vitro:
Regulation by reelin-disabled-1 signaling. Brain Res 4:1–9.
Matsuki T, Pramatarova A, Howell BW. 2008. Reduction of Crk and CrkL expression blocks
reelin-induced dendritogenesis. J Cell Sci 121:1869–1875.
Meseke M, Rosenberger G, F€
orster E. 2013. Reelin and the Cdc42/Rac1 guanine nucleotide
exchange factor aPIX/Arhgef6 promote dendritic Golgi translocation in hippocampal
neurons. Eur J Neurosci 37:1404–1412.
Nakagawa T, Futai K, Lashuel HA, Lo I, Okamoto K, Walz T, Hayashi Y, Sheng M. 2004.
Quaternary structure, protein dynamics, and synaptic function of SAP97 controlled by
L27 domain interactions. Neuron 44:453–467.
Neve RL, Neve KA, Nestler EJ, Carlezon WA. 2005. Use of herpes virus amplicon vectors to
study brain disorders. Biotechniques 39:381–389.
Nishiyama J, Yasuda R. 2015. Biochemical computation for spine structural plasticity.
Neuron 87:63–75.
Niu S, Renfro A, Quattrocchi CC, Sheldon M, Arcangelo GD. 2004. Reelin promotes
hippocampal dendrite development through the VLDLR/ApoER2-Dab1 pathway program
in neuroscience. Neuron 41:71–84.
Niu S, Yabut O, Arcangelo GD. 2008. The Reelin signaling pathway promotes dendritic spine
development in hippocampal neurons. Cancer Res 28:10339–10348.
Olson EC, Kim S, Walsh C. 2006. Impaired neuronal positioning and dendritogenesis
in the neocortex after cell-autonomous Dab1 suppression. J Neurosci
26:1767–1775.
REELIN SIGNALING MAINTAINS ADULT NETWORK STABILITY
Perez de Arce K, Varela-Nallar L, Farias O, Cifuentes A, Bull P, Couch Koleske BA, Inestrosa
N, Alvarez A. 2010. Synaptic clustering of PSD-95 is regulated by c-Abl through tyrosine
phosphorylation. J Neurosci 30:3728–3738.
Posada-Duque RA, Ramirez O, H€artel S, Inestrosa NC, Bodaleo F, Gonzalez-Billault C,
Kirkwood A, Cardona-G
omez GP. 2016. CDK5 downregulation enhances synaptic
plasticity. Cell Mol Life Sci 68:151–166.
Pujadas L, Gruart A, Bosch C, Delgado L, Teixeira CM, Rossi D, Lecea L, Martinez DelgadoGaracía, Soriano E. 2010. Reelin regulates postnatal neurogenesis and enhances spine
hypertrophy and long-term potentiation. J Neurosci 30:4636–4649.
Qiu S, Zhao LF, Korwek KM, Weeber EJ. 2006. Differential reelin-induced enhancement of
NMDA and AMPA receptor activity in the adult hippocampus. J Neurosci
26:12943–12955.
Rogers JT, Rusiana I, Trotter J, Zhao L, Donaldson E, Pak DTS, Babus L, Peters M, Banko J,
Chavis P, Rebeck GW, Hoe H-S, Weeber E. 2011. Reelin supplementation enhances
cognitive ability, synaptic plasticity, and dendritic spine density. Learn Mem 18:558–564.
Sans N, Petralia RS, Wang YX, Blahos J, Hell JW, Wenthold RJ. 2000. A developmental
change in NMDA receptor-associated proteins at hippocampal synapses. J Neurosci
20:1260–1271.
Segovia-Miranda F, Serrano F, Dyrda A, Ampuero E, Retamal C, Bravo-Zehnder M, Parodi J,
Zamorano P, Valenzuela Massardo L, van Zundert B, Inestrosa and Gonzalez A. 2015.
Pathogenicity of lupus anti-ribosomal P antibodies. Role of cross-reacting neuronal surface
P antigen in glutamatergic transmission and plasticity in a mouse model. Arthritis
Rheumatol 67:1598–1610.
Sepulveda FJ, Bustos FJ, Inostroza E, Zu~niga FA, Neve RL, Montecino M, van Zundert B. 2010.
Differential roles of NMDA receptor subtypes NR2A and NR2B in dendritic branch
development and requirement of RasGRF1. J Neurophysiol 10:1758–1770.
Sheng M, Hoogenraad CC. 2007. The postsynaptic architecture of excitatory synapses. A
more quantitative view. Annu Rev Biochem 76:823–847.
JOURNAL OF CELLULAR PHYSIOLOGY
Sinagra M, Frankova D, Korwek KM, Blahos J, Weeber EJ, Manzoni OJ, Chavis P. 2005.
Apolipoprotein E receptor 2 control somatic NMDA receptor composition during
hippocampal maturation in vitro. J Neurosci 25:6127–6136.
Tabata H, Nakajima K. 2002. Neurons tend to stop migration and differentiate along the
cortical internal plexiform zones in the Reelin signal-deficient mice. J Neurosci Res
69:723–730.
Tashiro A, Zhao C, Gage FH. 2006. Retrovirus-mediated single-cell gene knockout
technique in adult newborn neurons in vivo. Nat Protoc 1:3049–3055.
Tortosa E, Montenegro-Venegas C, Benoist M, H€artel S, Gonzalez-Billault C, Esteban JA,
Avila J. 2011. Microtubule-associated protein 1B (MAP1B) is required for dendritic spine
development and synaptic maturation. J Biol Chem 286:40638–40648.
Trommsdorff M, Gotthardt M, Hiesberger T, Shelton J, Stockinger W, Nimpf J,
Hammer R, Richardson JA, Herz J. 1999. Reeler/disabled-like disruption of neuronal
migration in knockout mice lacking the VLDL receptor and ApoE receptor 2. Cell
97:689–701.
Utsunomiya-Tate N, Kubo K, Tate S, Kainosho M, Katayama E, Nakajimaa K, Mikoshiba
K. 2000. Reelin molecules assemble together to form a large protein complex, which
is inhibited by the function-blocking CR-50 antibody. Proc Natl Acad Sci
97:9729–9734.
van Zundert B, Yoshii A, Constantine-Paton M. 2004. Receptor compartmentalization and
trafficking at glutamate synapses. A developmental proposal. Trends Neurosci
27:428–437.
Ventruti A, Kazdoba TM, Niu S, D’Arcangelo G. 2011. Reelin deficiency causes specific
defects in the molecular composition of the synapses in the adult brain. Neuroscience
189:32–42.
Weeber EJ, Beffert U, Jones C, Christian JM, Forster E, Sweatt JD, Herz J. 2002. Reelin and
ApoE receptors cooperate to enhance hippocampal synaptic plasticity and learning. J Biol
Chem 277:39944–39952.
1199