In vivo synaptic plasticity in the dentate gyrus of mice deficient in the neural cell adhesion molecule NCAM or its polysialic acid.

Abstract

The neural cell adhesion molecule NCAM and its associated polysialic acid (PSA) play important roles in synaptic plasticity in the CA1 and/or CA3 regions of the hippocampus in vitro. Here, we address the question of whether NCAM and PSA are involved in regulation of synaptic transmission and plasticity also in vivo at synapses formed by entorhinal cortex axons in the dentate gyrus of mice anaesthetized with urethane. We show that basal synaptic transmission, measured as the slope of field excitatory postsynaptic potentials, was reduced strongly in mice lacking ST8SiaII/STX, the enzyme involved in polysialylation of NCAM in stem cell-derived immature granule cells, but not in mice deficient either in the NCAM glycoprotein or the enzyme ST8SiaIV/PST involved in polysialylation of NCAM in mature neurons. Strikingly, only mice deficient in NCAM, but not in PST or STX, were impaired in long-term potentiation (LTP) induced by theta-burst stimulation, suggesting that LTP in the dentate gyrus depends on the NCAM glycoprotein alone rather than on its associated PSA. As also patterns of synaptic activity during and immediately after induction of LTP were impaired in NCAM-deficient mice, it is likely that induction of LTP requires NCAM. These data are the first to describe that NCAM is necessary for induction of synaptic plasticity in identified synapses in vivo and suggest that polysialylation of NCAM expressed by immature granule cells in the dentate gyrus supports development of basal excitatory synaptic transmission in this region.

Full text

Page 1

In vivo synaptic plasticity in the dentate gyrus of mice
deficient in the neural cell adhesion molecule NCAM or its
polysialic acid
Luminita Stoenica,1,2Oleg Senkov,1,2Rita Gerardy-Schahn,3Birgit Weinhold,3Melitta Schachner1and Alexander
Dityatev1,2
1Zentrum fu ¨r Molekulare Neurobiologie, Universita ¨tsklinikum Hamburg-Eppendorf, Martinistr. 52, D-20246 Hamburg, Germany
2Institut fu ¨r Neurophysiologie und Pathophysiologie, Universita ¨tsklinikum Hamburg-Eppendorf, Martinistr. 52, D-20246 Hamburg,
Germany
3Abteilung Zellula ¨re Chemie, Medizinische Hochschule Hannover, Carl-Neuberg-Str. 1, D-30625 Hannover, Germany
Keywords: carbohydrate, cell adhesion, granule cells, long-term potentiation, perforant path, population spike
Abstract
The neural cell adhesion molecule NCAM and its associated polysialic acid (PSA) play important roles in synaptic plasticity in the CA1
and⁄or CA3 regions of the hippocampus in vitro. Here, we address the question of whether NCAM and PSA are involved in regulation
of synaptic transmission and plasticity also in vivo at synapses formed by entorhinal cortex axons in the dentate gyrus of mice
anaesthetized with urethane. We show that basal synaptic transmission, measured as the slope of field excitatory postsynaptic
potentials, was reduced strongly in mice lacking ST8SiaII⁄STX, the enzyme involved in polysialylation of NCAM in stem cell-derived
immature granule cells, but not in mice deficient either in the NCAM glycoprotein or the enzyme ST8SiaIV⁄PST involved in
polysialylation of NCAM in mature neurons. Strikingly, only mice deficient in NCAM, but not in PST or STX, were impaired in long-
term potentiation (LTP) induced by theta-burst stimulation, suggesting that LTP in the dentate gyrus depends on the NCAM
glycoprotein alone rather than on its associated PSA. As also patterns of synaptic activity during and immediately after induction of
LTP were impaired in NCAM-deficient mice, it is likely that induction of LTP requires NCAM. These data are the first to describe that
NCAM is necessary for induction of synaptic plasticity in identified synapses in vivo and suggest that polysialylation of NCAM
expressed by immature granule cells in the dentate gyrus supports development of basal excitatory synaptic transmission in this
region.
Introduction
NCAM is a widely expressed glycoprotein in the central and
peripheral nervous systems that functions in cell–cell interactions,
such as cell migration, neurite extension and fasciculation, synapto-
genesis and synaptic plasticity and learning and memory (Rutishauser
& Landmesser, 1996; Walsh & Doherty, 1997; Kleene & Schachner,
2004). Initial studies in vitro showed that application of polyclonal
antibodies against NCAM before induction of long-term potentiation
(LTP) reduced levels of LTP in the CA1 region of the adult
hippocampus (Luthi et al., 1994; Ronn et al., 1995). Similarly, LTP
in the hippocampal CA1 region was reduced in constitutively NCAM
deficient mice (Muller et al., 1996, 2000) and in conditionally NCAM-
deficient mice, in which the NCAM gene was ablated in excitatory
neurons postnatally after the cessation of major developmental events
(Bukalo et al., 2004). These observations support the idea that the
deficit in CA1 LTP in constitutively NCAM-deficient mice is not a
consequence of developmental abnormalities but that NCAM glyco-
protein is functional in mature synapses. In the CA3 region,
constitutively NCAM-deficient mice, but not conditionally NCAM-
deficient mice, are abnormal in lamination of mossy fibre projections
and in LTP of mossy fibre synapses, suggesting that NCAM
glycoprotein is required for the normal development of mossy fibre–
CA3 synapses (Cremer et al., 1994, 1998).
As NCAM is the only glycoprotein that carries the unusual
polysialic acid (PSA) in mammals and because expression of PSA is
developmentally regulated, much attention has been given to under-
standing the functions of the NCAM glycoprotein backbone versus
those of PSA. Enzymatic removal of PSA by the highly PSA-specific
glycosidase endosialidase-N inhibited LTP and long-term depression
(LTD) in the CA1 region, LTP-associated formation of perforated
synapses, and spatial learning (Becker et al., 1996; Muller et al., 1996;
Dityatev et al., 2004). Our experiments with mice deficient in either
one of the two polysialyltransferases, ST8SiaII⁄STX or ST8Sia-
IV⁄PST required for polysialylation of NCAM in less or more
differentiated neural cells, respectively (Hildebrandt et al., 1998; Ong
et al., 1998), provided genetic evidence for the importance of PSA and
ST8SiaIV⁄PST in synaptic plasticity in CA1 (Eckhardt et al., 2000;
Angata et al., 2004). Interestingly, polysialylation of NCAM was not
required for induction or maintenance of LTP in mossy fibre synapses
in the CA3 region, despite abnormal lamination of mossy fibre
projections in STX-deficient mice (Angata et al., 2004). As impair-
ment of LTP in conditionally NCAM-deficient mice could be rescued
by elevation of extracellular Ca2+concentration (Bukalo et al., 2004)
Correspondence: Dr A. Dityatev,2Institut fu ¨r Neurophysiologie und Pathophysiologie,
as above.
E-mail: a.dityatev@uke.uni-hamburg.de
Received 31 December 2005, revised 19 February 2006, accepted 22 February 2006
European Journal of Neuroscience, Vol. 23, pp. 2255–2264, 2006
doi:10.1111/j.1460-9568.2006.04771.x
ª The Authors (2006). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd

Page 2

and because peptides blocking the interaction of glycans with the
fourth immunoglobulin-like domain of NCAM (Horstkorte et al.,
1993) reduced LTP when applied before induction of LTP but not
afterwards (Staubli et al., 1998), a role for NCAM in induction of LTP
was suggested. That impairment of LTP in NCAM-deficient mice
depends on the experimental conditions, e.g. extracellular Ca2+
concentration, prompted us to investigate synaptic plasticity in these
mice under the more physiological conditions, i.e. in vivo. It is
interesting in this context that in vivo tetanic stimulation of the
perforant path increases levels of the soluble, most likely proteoly-
tically released, form of NCAM in the extracellular space (Fazeli
et al., 1994). Tetanic stimulation also increases the percentage of
axospinous synapses expressing NCAM180, the largest splice variant
of NCAM with the longest intracellular domain of all NCAM
isoforms, in the dentate gyrus of adult mice (Schuster et al., 1998).
The functional importance of NCAM and PSA in this region of the
hippocampus has, however, not been investigated. We therefore,
decided to perform in vivo recordings of LTP in the dentate gyrus of
adult NCAM-deficient mice. An additional reason for recordings in
the dentate gyrus was that this structure is particularly convenient for
identification of monosynaptic responses and has therefore been
studied most extensively with regard to induction and maintenance of
LTP in vivo. As CA1 LTP was found to be impaired in NCAM-
deficient mice and because the mechanisms of LTP induction in the
dentate gyrus and in the CA1 region share common features, e.g. are
N-methyl-D-aspartate (NMDA) receptor-dependent, we expected to
find synaptic abnormalities in the dentate gyrus of NCAM-deficient
mice. To analyse the role of PSA during early and later stages of
neural development in vivo we used mice deficient in either the
polysialyltransferases STX or PST.
Materials and methods
Animals
Adult age-matched knockout and wild-type mice were used in this
study. Males and females were used for recordings and statistical
evaluation as no gender-related differences were seen in the electro-
physiological parameters measured. Constitutively NCAM-deficient
mice (NCAM– ⁄ –mice; mean age 12 ± 3 weeks ± SD; mean weight
27 ± 4 g ± SD; Cremer et al., 1994) were back-crossed for more than
eight generations onto the C57BL⁄6J background. Wild-type litter-
mates (NCAM+ ⁄ +mice; 14 ± 3 weeks, 24 ± 3 g) derived from
heterozygous breeding pairs served as controls. Mice deficient in
PST(PST– ⁄ –;17 ± 2 weeks;
17 ± 3 weeks;32 ± 4 g)mice
SV129Ola embryonic stem cells carrying a mutation either in the
PST or STX gene into C57BL⁄6J mice (Eckhardt et al., 2000; Angata
et al., 2004). These mice were back-crossed for six generations onto
the C57BL⁄6J background. Heterozygous mice were then intercrossed
to obtain homozygous mice and strains were maintained as homozy-
gous lines for several generations (Weinhold et al., 2005). Both
strains, STX and PST, have the same genetic background, as they bear
the same percentage of C57BL⁄6J and SV129Ola genome, and
therefore only one group of wild-type mice (16 ± 2 weeks; 27 ± 4 g)
was used for comparison with STX– ⁄ –and PST– ⁄ –mice.
All mice were housed individually for at least 1 week before testing
in pathogen-free, climate-controlled conditions under a 12 h light⁄
dark inverted cycle, with free access to standard mouse diet and water.
All surgical procedures described in this study were approved by the
Committee on Animal Health and Care of the governmental body in
Hamburg.
32 ± 4 g)
were
andSTX
by
(STX– ⁄ –;
introducingobtained
Anaesthesia and tracheotomy
Mice were injected intraperitoneally with urethane (Sigma) dissolved
in distilled water with an initial dose corresponding to 1.5 g⁄kg body
weight, except for mice that were more than 30 g, who were injected
with a dose of 0.045 g (corresponding to 30 g animal weight). Fifteen
minutes following initial anaesthesia, all mice were tested with a hind
limb pinch, and additional doses of urethane were administrated every
10–15 min until the reflex was no longer observed. The dosage of
supplementary injections was one-sixth of the initial dose. During the
time of recording the mice were checked regularly for signs of
recovery from anaesthesia and, if needed, an additional dose of
urethane was injected intramuscularly.
A tracheotomy was performed to avoid breathing problems that are
otherwise common during long-term recordings in mice. Tracheotomy
was performed by a transverse cut between two tracheal rings (using a
sharp needle) in the middle part of the trachea. A 1-cm-long plastic
tube with an outer diameter of approximately 1 mm was inserted into
the incised trachea and kept there for the duration of the experiment.
Placement of electrodes
Immediately following the tracheotomy, mice were visually inspected
for 10 min for any sign of breathing problems or discomfort and then
fixed in a stereotactic frame (Narishige, Tokyo, Japan) for the insertion
of electrodes. The temperature of the mice was then monitored
constantly through a rectal probe and kept at 35.5–36 ?C using a
heating pad (rectal probe and pad from HSE-Harvard, March-
Hugstetten, Germany).
The mouse’s head was positioned by means of a mouse nose clamp
adaptor supplemented by ear bars. For all mouse strains, except for the
NCAM– ⁄ –mice, the coordinates for electrode insertion were: AP, 1.5–
2.0 mm; ML, +1–1.5 mm for the recording electrode and 0.5 mm
anterior to lambda; ML, +3.0–4.00 mm, ipsilaterally for the stimulating
electrode. For NCAM– ⁄ –mice, the coordinates were slightly different
because of their smaller brain size (AP, 1.2–1.8 mm; ML, +1–1.5 mm
for the recording electrode and 0.5 mm anterior to lambda, ML, +3.0–
4.00 mm, ipsilaterally for the stimulating electrode). The stimulating
electrodewasinsertedatanangleof70 ?.Anadditionalholewasdrilled
onthesurfaceofthecontralateralfrontalcortexforthegroundelectrode,
which was made of chlorided silver wire. A stimulating electrode was
made of two 125-lm Teflon-insulated stainless steel wires, whereas a
recording electrode was made of 75-lm Teflon-insulated tungsten wire.
Electrophysiological recordings
Theelectrodeswereadvanceduntilwaveformswithalargepositivefield
excitatorypostsynapticpotential(fEPSP)andasuperimposednegative-
going population spike (PS) were evoked. The Z-coordinate was
adjusted to maximize the field responses to a biphasic current pulse of
100 ls and 100–500 lA delivered by a stimulator (A-M Systems,
Everett, WA, USA). The stimulation strength for recordings before and
after induction of LTP was selected to provide a PS of 2–3 mV. The
responseswerecollectedevery30 sandtheaveragesoffiveconsecutive
sweeps were used for measurements. Long-term potentiation was
induced by six trains (intertrain interval of 20 s) of six bursts (interburst
interval of 200 ms) of six pulses applied at 400 Hz (Davis et al., 1997).
The stimulation intensity used for induction of LTP was twofold higher
than that used for baseline recordings. Only experiments without
rundownoffEPSPs(> 90%baseline)wereincludedintheanalysis.The
recorded signals were amplified ·1000 and filtered (0.1-5 kHz band-
pass) using an AC amplifier (A-M Systems) and digitized using a Pico
2256L. Stoenica et al.
ª The Authors (2006). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 23, 2255–2264

Page 3

ADCboard(Tangetal.,2001).Thesignalswereacquiredandmeasured
using the LTP Program (Anderson & Collingridge, 2001). For
measurements of the slope of fEPSPs, only the initial part (?0.5 ms)
of the response was used, as shown in Fig. 4.
Verification of electrode positions
At the end of electrophysiological recordings, anaesthetized mice were
killed by decapitation. Brains were removed, frozen and sectioned
with a Leica CM3050 cryomicrotome (Nussloch, Germany). Sections
were stained with cresyl violet and checked for the proper locations of
the electrodes. In all tested cases, the positions of the recording and
stimulating electrodes were in the hilus (near the granule cell layer)
and angular bundle, respectively (data not shown).
Results
Short-term plasticity
Analysis of short-term synaptic plasticity in mice deficient in either
NCAM, PST or STX was performed using paired-pulse stimulation of
theperforantpathattheangularbundle.Subthreshold(i.e.notelicitinga
PS) double-pulse stimulation with interpulse intervals of 10, 25, 50 and
100 ms produced a small paired-pulse facilitation of fEPSPs in wild-
type and mutant mice (Fig. 1A, C and E), as previously reported for
C57BL⁄6J mice (Bampton et al., 1999). Analysis of paired-pulse
facilitationoffEPSPsusingtwo-wayanalysisofvariance(anova)with
Genotype as between-subjects factor and Interpulse interval as within-
subject factor revealed a significant interaction (P < 0.05) between
thesetwofactorsforPST– ⁄ –,butnotforNCAM– ⁄ –andSTX– ⁄ –mice.In
other words, the two curves shown in Fig. 1C are significantly different
from each other but those in Fig. 1A and E are not. However,
no significant difference in paired-pulse facilitation was detected for
PST– ⁄ –mice for any particular interpulse interval using t-test with
Bonferroni correction for multiple comparisons, although mutants
showed a tendency for larger paired-pulse facilitation at 10 and 25 ms
(Fig. 1C).
Stronger paired-pulse stimulation, eliciting a supramaximal PS in
response to the first stimulus, resulted in a block or strong inhibition of
the PS in response to the second stimulus when the interpulse interval
was 10 or 25 ms (Fig. 1B, D and F). This inhibition is believed to be
mediated by feedback inhibition in the dentate gyrus (Bampton et al.,
1999). Paired-pulse stimulation with intervals longer than 50 ms
produced an increase of the PS amplitude, as also reported previously
(Bampton et al., 1999). This increase most likely reflects enhanced
synchrony in firing of neurons in response to the second pulse.
Analysis of paired-pulse modulation of PSs using two-way anova
revealed a weak but significant interaction (P < 0.05) between
Genotype and Interpulse interval for NCAM– ⁄ –and STX– ⁄ –, but
not for PST– ⁄ –mice. No significant difference in paired-pulse
modulation was detected for any particular interpulse interval using
the t-test with Bonferroni correction for multiple comparisons,
although there was a tendency for the mutant mice to have smaller
depressions at 10-ms and 25-ms intervals and smaller facilitation at
intervals >25 ms (Fig. 1B and F). In summary, there were rather
subtle abnormalities in paired-pulse modulation of either fEPSPs or
population spikes in all genotypes studied.
Basal synaptic transmission and LTP
To record responses before and after induction of LTP, we adjusted the
stimulation intensity to elicit PSs with amplitudes of 2–3 mV. There
was no significant difference in the selected stimulation strength and
PS amplitude between mice of all genotypes analysed (Fig. 2A and B).
In NCAM– ⁄ –and PST– ⁄ –mice, the slopes of fEPSPs recorded at the
selected stimulation strength also matched those in wild-type mice
(Fig. 2C); however, in STX– ⁄ –mice there was a strong reduction in
the slope of fEPSPs. Analysis of stimulus-response curves in STX– ⁄ –
mice showed that relationships between the intensity of stimulation
and the amplitude of PSs were similar in STX– ⁄ –and STX+ ⁄ +mice
(Fig. 3A), whereas the slope of fEPSPs was lower in STX– ⁄ –mice as
compared with STX+ ⁄ +mice for all stimulation strengths tested
(Fig. 3B). Analysis of relationships between the slope of fEPSPs and
the amplitude of PSs, i.e. so-called E-S curves, revealed a significant
shift of the curve to the left (Fig. 3C) in STX– ⁄ –mice, indicating a
reduction in the threshold of action potential generation.
Theta-burst stimulation elicited a robust LTP in urethane-anaes-
thetized wild-type mice, as reported previously (Errington et al., 1997;
Bampton et al., 1999; Bliss et al., 2000). Three hours after induction
of LTP, the magnitude of potentiation in the amplitude of the PS was
on average between 200 and 250%, whereas the magnitude of
potentiation in the initial slope of the fEPSP was approximately 120%
(Fig. 4). In PST– ⁄ –and STX– ⁄ –mice there was no significant
difference in LTP of either the fEPSP or PS, as compared with each
other and to wild-type mice. In NCAM– ⁄ –mice, however, there was
no detectable LTP in the initial slope of fEPSPs and potentiation of the
PS amplitude was impaired substantially (Fig. 4A and B).
The short-term potentiation (STP), i.e. increase of fEPSP and PS
immediately after induction of LTP was also impaired greatly in
NCAM– ⁄ –mice (Fig. 4A and B). For all tested genotypes, except for
NCAM– ⁄ –mice, there was a strong correlation between STP and LTP
levels (Fig. 5A and B). Impaired STP in NCAM– ⁄ –mice suggests that
inductionofLTPisimpairedinthesemutants.Toverifythishypothesis,
we measured PSs and fEPSPs during theta-burst stimulation (Fig. 5C)
and, indeed, observed a strong difference between mouse genotypes: In
NCAM+ ⁄ +mice, the slope of fEPSPs gradually increased after each
theta-burst stimulation, reaching the level of STP recorded during the
firstminuteafterinductionofLTP.Thesamepatternwasobservedforall
other genotypes tested, except for NCAM– ⁄ –mice, which showed no
facilitation(Fig.5D).TheamplitudesofPSselicitedduringinductionof
LTP were also significantly smaller in NCAM– ⁄ –as compared with
NCAM+ ⁄ +mice (Fig. 5E). Thus, there were significant differences
between NCAM– ⁄ –and NCAM+ ⁄ +mice in the pattern of synaptic
activity during and immediately after induction of LTP, suggesting that
theinductionphase isimpairedinNCAM– ⁄ –mice. Thisisinagreement
withpreviousinvitrodataintheCA1regionthatindicatedthatNCAMis
involved in early stages of CA1 LTP (Staubli et al., 1998; Bukalo et al.,
2004).
Discussion
Impaired dentate gyrus LTP in NCAM-deficient mice
Our results provide first electrophysiological evidence that NCAM
regulates synaptic plasticity in identified synapses in vivo and, thus,
close a gap between previous data on the role of NCAM in LTP
in vitro (Luthi et al., 1994; Muller et al., 1996; Dityatev et al., 2000;
Bukalo et al., 2004) and behavioural studies, demonstrating the
importance of NCAM in hippocampus-dependent learning and
memory in the Morris water maze and contextual fear conditioning
(Cremer et al., 1994; Stork et al., 2000; Bukalo et al., 2004). Also for
the first time, we characterized synaptic plasticity in the dentate gyrus
of NCAM-, PST- and STX-deficient mice (Table 1) and suggest that it
is NCAM glycoprotein-, but not PSA-dependent. It is remarkable that
NCAM, PSA and synaptic plasticity in vivo
2257
ª The Authors (2006). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 23, 2255–2264

Page 4

2258L. Stoenica et al.
ª The Authors (2006). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 23, 2255–2264

Page 5

these features of LTP induced in synaptic connections with granule
cells as postsynaptic elements (with a postsynaptic mechanism of
induction) resemble those of mossy fibre LTP, i.e. in a situation where
the granule cells are presynaptic and induction of LTP is presynaptic.
In both cases, induction of LTP requires NCAM but not PSA (Cremer
et al., 1998; Eckhardt et al., 2000; Angata et al., 2004). Thus, in
respect to PSA-dependence, LTP in the dentate gyrus is different from
LTP and LTD in the CA1 region, which are both PSA dependent.
Interestingly, we did not detect any abnormalities in dentate gyrus LTP
in STX– ⁄ –mice. However, the number of immature granule cells in
which polysialylation of NCAM is driven by STX is quite small, and
therefore even strong abnormalities in LTP in this subpopulation of
granule cells might not be well detected using field recordings, which
provide a measure of activity in large neuronal cell populations.
Because the highly polysialylated subpopulation of newly generated
granule cells shows particularly high levels of LTP (Schmidt-Hieber
et al., 2004), it will be of interest to compare LTP in these neurons and
mature granule cells in STX– ⁄ –and NCAM– ⁄ –mice using single-cell
recordings.
Although a wealth of information has been accumulated on the
contributions of cell adhesion and extracellular matrix molecules on
synaptic plasticity in vitro (Benson et al., 2000; Dityatev & Schachner,
2003), experiments in vivo are very rare. For instance, normal dentate
gyrus LTP was found in vivo in anaesthetized mice deficient in the
neural cell adhesion molecule L1 (Bliss et al., 2000), whereas mice
deficient in the cell adhesion molecule Thy-1 show impaired dentate
gyrus LTP in both anaesthetized and freely moving mice (Nosten-
Bertrand et al., 1996; Errington et al., 1997). As Thy-1– ⁄ –mice show
an increase in inhibitory postsynaptic currents in the dentate gyrus
(Hollrigel et al., 1998) and because levels of LTP are normalized in
mutants towards wild-type levels in vitro and in anaesthetized mice
in vivo by a c-aminobutyric acid (GABA)A receptor antagonist
(Nosten-Bertrand et al., 1996; Errington et al., 1997), it is conceivable
that abnormally high GABAergic activity in the dentate gyrus causes
the impairment of LTP in Thy-1-deficient mice. There are no
data available on GABAergic transmission in the dentate gyrus of
NCAM– ⁄ –mice, but transgenic mice overexpressing the secreted
extracellular domain of NCAM exhibit a striking reduction in
inhibitory synapses in the cingulate and frontal association cortices
and in the amygdala (Pillai-Nair et al., 2005), thus suggesting a link
between inhibitory interneurons and NCAM. Our present data show
that paired-pulse inhibition of population spikes, a measure of feed-
back inhibition, was not higher in NCAM– ⁄ –mice than in NCAM+ ⁄ +
mice. In fact, there was a tendency in both NCAM– ⁄ –and STX– ⁄ –
mice to exhibit lower modulation of population spikes, as compared
with wild-type mice. It is therefore more likely that the number of
GABAergic interneurons or their synapses⁄synaptic activity is
decreased, rather than increased, in the dentate gyrus of NCAM– ⁄ –
or STX– ⁄ –mice. Still, lower amplitudes of population spikes elicited
by theta-burst stimulation in NCAM– ⁄ –mice, as found in the present
study, could reflect an increased recruitment of GABAergic interneu-
rons or facilitation of inhibitory currents in response to theta-burst
stimulation.
Another hypothesis to explain the observed deficit in LTP of
NCAM– ⁄ –mice would be that LTP is already saturated at dentate
gyrus synapses because of prior potentiation. However, in this case
one would expect to find elevated levels of basal synaptic transmission
Fig. 1. Small abnormalities in short-term plasticity in mice deficient in NCAM (A and B), STX (C and D) or PST (E and F) vs. wild-type mice. Paired-pulse
facilitation of fEPSPs was elicited by subthreshold double-pulse stimulation of the angular bundle with interpulse intervals of 10, 25, 50, and 100 ms. The data
shown in A, C and E are mean + SEM of the ratio between slopes of fEPSPs elicited by the second and first pulses. Analysis of paired-pulse facilitation using two-
way anova detects a significant interaction between genotype and interpulse interval for PST-deficient mice. Paired-pulse modulation of the population spike (PS;
negative-going fast component) was elicited by supramaximal stimulation of the angular bundle, resulting in a strong suppression of spikes at short intervals (10 and
25 ms) and facilitation of PS amplitude at longer intervals (50, 100, 200, 500 and 1000 ms). The data shown in B, D and F are the mean + SEM of the ratio between
amplitudes of PSs elicited by the second and first pulses. Two-way anova detected a significant interaction between genotype and interpulse interval for NCAM-
and STX-deficient mice. N indicates the number of mice analysed.
Fig. 2. Impaired basal synaptic transmission in STX– ⁄ –mice. Data are
mean + SEM of three parameters characterizing basal synaptic transmission:
intensityof stimulation usedtoevoke LTP (A); amplitudeof PSs(B); andslope
of fEPSPs (C) collected at the stimulation strength used to evoke LTP.
Nindicatesthenumberofmiceanalysed.**P < 0.005,unpairedtwo-sidedt-test.
NCAM, PSA and synaptic plasticity in vivo
2259
ª The Authors (2006). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 23, 2255–2264

Page 6

in NCAM– ⁄ –mice, as, for example, was observed in mice deficient in
the extracellular glycoprotein tenascin-R (Saghatelyan et al., 2001).
Our data, however, show normal basal synaptic transmission in
NCAM– ⁄ –mice and, thus, argue against this idea. Another possibility
is that NMDA receptor activity is reduced in these animals. The
NMDA receptor is critical for induction of LTP in the dentate gyrus of
mice (Kleschevnikov & Routtenberg, 2001). The functional cross-talk
between adhesion molecules and neurotransmitter receptors is not
new. Exciting examples linking cell adhesion molecule function and
NMDA receptors are provided by EphB receptors, integrins and alpha-
neurexin (Takasu et al., 2002; Lin et al., 2003; Kattenstroth et al.,
2004). In this respect, it is noteworthy that tetanic stimulation of the
perforant path results in an increased proportion of axospinous
synapses expressing NCAM180, the largest isoform of NCAM with
the longest cytoplasmic domain of all splice variants, that interacts
with spectrin enriched in postsynaptic densities (Schuster et al., 1998;
Sytnyk et al., 2002). Also NCAM180 and the NR2A subunit of
NMDA receptors colocalize before, and coredistribute to, the periph-
ery of postsynaptic densities after induction of dentate gyrus LTP (Fux
et al., 2003). It is not known how these activity dependent changes in
NCAM expression and cellular localization contribute to synaptic
plasticity in the dentate gyrus but it is conceivable that an association
between NCAM180 and NMDA receptors could be critical for
NMDA receptor activity during the induction of LTP. Additional
indirect support for this notion comes from our in vitro experiments in
which CA1 LTP is restored by an induced increase in Ca2+influx via
NMDA receptors, such as elevation of extracellular Ca2+concentra-
tion (Bukalo et al., 2004) or lowering of extracellular Mg2+
concentration (Kochlamazashvili & Dityatev, unpublished observa-
tions), which is known to facilitate activation of NMDA receptors.
Abnormalities in paired-pulse modulation in
NCAM- and PSA-deficient mice
To assess short-term synaptic plasticity in NCAM- and PSA-
deficient mice, we analysed facilitation of transmitter release elicited
by paired-pulse subthreshold stimulation of the perforant path. Only
PST– ⁄ –mice showed slightly but significantly abnormal paired-
pulse facilitation of fEPSPs (Table 1). This abnormality did not
result in altered modulation of population spikes in PST– ⁄ –mice in
response to paired-pulse supramaximal stimulation; this is unsur-
prising as this parameter is known to reflect excitability of neurons
and synchronization of spiking activity in the population of granule
cells rather than modulation of transmitter release. Excitability and
synchronization depend on intrinsic properties and network activity,
including the activity of GABAergic interneurons. In contrast to
PST– ⁄ –mice, we found a slightly abnormal paired-pulse modulation
of population spikes, although not of synaptic responses, in
NCAM– ⁄ –and STX– ⁄ –mutants. In summary, these data suggest
that polysialylation of NCAM in immature neurons possibly shapes
the organization of the neural network in the dentate gyrus and,
thus, affects the firing pattern of granule cells, whereas polysial-
ylation of NCAM in synapses of mature neurons modulates
transmitter release in synapses formed on these cells by yet
unknown mechanisms.
Fig. 3. Changes in stimulus-response and E-S curves in STX– ⁄ –vs. wild-type
mice. Data demonstrate relationships between the intensity of stimulation and
the amplitude of PSs (A), the intensity of stimulation and the slope of fEPSPs
(B) and the slope of fEPSPs vs. the amplitude of PSs (E-S curves in
C). Means ± SEM are shown; N indicates the number of mice analysed.
Fig. 4. Impaired long-term plasticity in NCAM– ⁄ –mice. Data show mean + SEM of population spike (PS) amplitudes (A, C and E) and the slope of fEPSPs (B, D
and F) expressed as a percentage regarding the baseline (time from )20–0 min). Theta-burst stimulation (applied at time ‘0’) elicited a robust LTP in wild-type,
STX– ⁄ –and PST– ⁄ –mice; however, LTP in NCAM- ⁄ – mice was impaired (A and B). Examples of responses collected before and 3 h after induction of LTP are
shown above the LTP profiles. Horizontal bars in A, C and E indicate the time intervals shown with a higher resolution in B, D and F. Horizontal boxes in B, D and F
indicate the time intervals used for slope measurements. N indicates the number of mice analysed (being the same in B and A, in D and C, and in E and F).
2260L. Stoenica et al.
ª The Authors (2006). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 23, 2255–2264

Page 7

NCAM, PSA and synaptic plasticity in vivo
2261
ª The Authors (2006). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 23, 2255–2264

Page 8

Impaired synaptic transmission in the dentate gyrus
of STX-deficient mice
Another novel finding of the present study is that basal synaptic
transmission in the perforant path–dentate gyrus connection is
impaired in STX– ⁄ –mice, i.e. in mice with deficient polysialylation
of immature neurons that are generated continuously in this hippo-
campal region in adult mice. As presynaptic stimulation elicited
normal population spike activity in the dentate gyrus of STX– ⁄ –mice,
while fEPSPs were reduced, the mutants display a shift in the E-S
curve that is indicative of an increase in excitability of granule cells.
This phenomenon might result from a secondary homeostatic⁄adap-
tive event to maintain spiking activity of granule cells at normal levels
Fig. 5. Abnormal patterns of activity in NCAM– ⁄ –mice during and immediately after induction of LTP. (A and B) Analysis of the relationship between levels of
short-term potentiation (first minute after TBS) and long-term potentiation (3 h after TBS) of fEPSPs (A) and population spikes (PSs; B). N is the number of mice
analysed, r, the coefficient of correlation, P, the significance of correlation. ‘Others’ include NCAM+ ⁄ +, PST– ⁄ –, STX– ⁄ –and their wild-type controls. Solid and
dashed lines are linear regression lines drawn through data from NCAM– ⁄ –mice and other mice (‘others’), respectively. (C) Examples of fEPSPs elicited by the
sixth train of six bursts for NCAM+ ⁄ +and NCAM– ⁄ –mice. Negative-going components are PSs. Arrows point to reduced PSs in the NCAM– ⁄ –mouse. (D and E)
Data are mean + SEM of slope of the first fEPSPs (D) and of PS amplitude (E) measured for each of 6 · 6 bursts applied for induction of LTP.
2262L. Stoenica et al.
ª The Authors (2006). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 23, 2255–2264

Page 9

despite reduced synaptic input. As paired-pulse facilitation of fEPSPs,
which is related to the probability of release, is normal in STX– ⁄ –
mice, it is unlikely that probability of release is reduced in this mutant.
More plausible reasons for the reduction in the basal synaptic
transmission in STX– ⁄ –mice would be a reduction in the number of
synapses formed by perforant pathway axons in the dentate gyrus or
smaller postsynaptic responses elicited by single vesicle release. The
hypothesis that PSA synthesized by the developmentally early acting
polysialyltransferase STX might be important for proper synaptogen-
esis in the dentate gyrus is supported by recent in vitro data showing
that polysialylation of immature hippocampal neurons promotes
synapse formation in a choice situation when NCAM-deficient and
PSA-NCAM-expressing neurons are cocultured (Dityatev et al., 2000,
2004). In the dentate gyrus, a choice situation is created by the
progenitor cell-derived immature neurons, which are highly polysial-
ylated by STX (Hildebrandt et al., 1998; Angata et al., 2004) and
situated in a cellular territory with a lower expression of PSA by
mature granule cells. This scenario favours preferable synapse
formation on PSA-enriched immature neurons in wild-type mice. In
STX– ⁄ –mice preferential PSA-mediated synapse formation might no
longer take place and immature granule cell neurons would thus
receive less synaptic input. As our recordings revealed impaired basal
synaptic transmission in adult STX– ⁄ –mice, we have to postulate that
the deficit in formation of synapses on immature granule cells might
have long-lasting consequences and result in reduced synaptic
coverage of mature granule cells. As basal synaptic transmission is
normal in the dentate gyrus of NCAM– ⁄ –mice, the deficit in synaptic
transmission in STX mutants highlights the interesting possibility that
abnormal polysialylation of NCAM leads to a more severe abnormal-
ity of synaptic transmission than the absence of the NCAM
glycoprotein itself. This interpretation is consistent with data on
mutant mice deficient in both PST and STX, which exhibit numerous
abnormalities in axonal projections that are not observed in single
NCAM– ⁄ –mice or mice deficient in the three genes PST, STX and
NCAM (Weinhold et al., 2005). As PSA is a strongly anti-adhesive
structure, a PSA-negative (although NCAM glycoprotein-positive)
environment in which immature granule cells are situated in STX-
deficient mice might be too adhesive for optimal axonal outgrowth and
remodelling, processes that would appear necessary for synapse
formation on newly generated granule cells.
In summary, our major findings support the view that NCAM is
important for induction and maintenance of synaptic plasticity,
whereas PSA generated specifically on progenitor cell-derived imma-
ture granule neurons by STX is involved in setting normal levels of
synaptic transmission in the dentate gyrus of live mice.
Acknowledgements
We thank M. Errington for very valuable methodological advice at the initial
stages of this project, K. Angata and M. Fukuda for STX-deficient mice, H.
Cremer for NCAM-deficient mice, A. Dahlmann for genotyping, G. Engler and
C. Moll for help with establishing of tracheotomy in mice, A. Engel for his
support and A. Kleschevnikov for valuable comments on the manuscript. This
workwassupportedbyErasmusandDAADfellowshipstoL.S.andtheDeutsche
Forschungsgemeinschaft (Di 702 ⁄ 4–1,2 to A.D., Ge 801 ⁄ 3–3 to R.G.S.).
Abbreviations
E-S curve, curve depicting relationship between the slope of fEPSPs and the
amplitude of population spike; fEPSP, field excitatory postsynaptic potential;
GABA, c-aminobutyric acid; LTD, long-term depression; LTP, long-term
potentiation; NCAM, neural cell adhesion molecule; NMDA, N-methyl-D-
aspartate; PS, population spike; PSA, polysialic acid; PST, polysialyltransferase
ST8SiaIV; STP, short-term potentiation; STX, polysialyltransferase ST8SiaII.
References
Anderson, W.W. & Collingridge, G.L. (2001) The LTP Program: a data
acquisition program for on-line analysis of long-term potentiation and other
synaptic events. J. Neurosci. Meth., 108, 71–83.
Angata, K., Long, J.M., Bukalo, O., Lee, W., Dityatev, A., Wynshaw-Boris, A.,
Schachner, M., Fukuda, M. & Marth, J.D. (2004) Sialyltransferase ST8Sia-II
assembles a subset of polysialic acid that directs hippocampal axonal
targeting and promotes fear behavior. J. Biol. Chem., 279, 32603–32613.
Bampton, E.T., Gray, R.A. & Large, C.H. (1999) Electrophysiological
characterisation of the dentate gyrus in five inbred strains of mouse. Brain
Res., 841, 123–134.
Becker, C.G., Artola, A., Gerardy-Schahn, R., Becker, T., Welzl, H. &
Schachner, M. (1996) The polysialic acid modification of the neural cell
adhesion molecule is involved in spatial learning and hippocampal long-term
potentiation. J. Neurosci. Res., 45, 143–152.
Benson, D.L., Schnapp, L.M., Shapiro, L. & Huntley, G.W. (2000) Making
memories stick: cell-adhesion molecules in synaptic plasticity. Trends Cell
Biol., 10, 473–482.
Bliss, T., Errington, M., Fransen, E., Godfraind, J.M., Kauer, J.A., Kooy, R.F.,
Maness, P.F. & Furley, A.J. (2000) Long-term potentiation in mice lacking
the neural cell adhesion molecule L1. Curr. Biol., 10, 1607–1610.
Bukalo, O., Fentrop, N., Lee, A.Y., Salmen, B., Law, J.W., Wotjak, C.T.,
Schweizer, M., Dityatev, A. & Schachner, M. (2004) Conditional ablation of
the neural cell adhesion molecule reduces precision of spatial learning, long-
term potentiation, and depression in the CA1 subfield of mouse
hippocampus. J. Neurosci., 24, 1565–1577.
Cremer, H., Chazal, G., Carleton, A., Goridis, C., Vincent, J.D. & Lledo, P.M.
(1998) Long-term but not short-term plasticity at mossy fiber synapses is
impaired in neural cell adhesion molecule-deficient mice. Proc. Natl. Acad.
Sci. USA, 95, 13242–13247.
Cremer, H., Lange, R., Christoph, A., Plomann, M., Vopper, G., Roes, J.,
Brown, R., Baldwin, S., Kraemer, P. & Scheff, S.& (1994) Inactivation of the
N-CAM gene in mice results in size reduction of the olfactory bulb and
deficits in spatial learning. Nature, 367, 455–459.
Davis, S., Bliss, T.V., Dutrieux, G., Laroche, S. & Errington, M.L. (1997)
Induction and duration of long-term potentiation in the hippocampus of the
freely moving mouse. J. Neurosci. Meth., 75, 75–80.
Dityatev, A., Dityateva, G. & Schachner, M. (2000) Synaptic strength as a
function of post- versus presynaptic expression of the neural cell adhesion
molecule NCAM. Neuron, 26, 207–217.
Table 1. Summary of electrophysiological findings in the dentate gyrus of
mice deficient in NCAM or the polysialyltransferases ST8SiaII ⁄ STX and
ST8SiaIV ⁄ PST
Parameter
Findings relative to those in WT mice
NCAM– ⁄ –
STX– ⁄ –
PST– ⁄ –
Basal level of transmission
fEPSP
PS
Paired-pulse modulation
fEPSP
PS
Short-term potentiation
fEPSP
PS
Long-term potentiation
fEPSP
PS
¼
¼
fl
¼
¼
¼
„
¼
¼
„
¼
„
fl
fl
¼
¼
¼
¼
fl
fl
¼
¼
¼
¼
The following conclusions can be drawn. (i) Basal level of synaptic trans-
mission and paired-pulse modulation of excitability depend on polysialylation
of immature neural cells; (ii) paired-pulse facilitation of transmitter release
requires polysialylation of mature cells; and (iii) NCAM, but not PSA, is
necessary for short- and long-term potentiation induced by theta-burst stimu-
lation. Abbreviations: fEPSP, field excitatory postsynaptic potential; PS, pop-
ulation spike; ¼ , not different from wild-type mice; „ , significantly different
from wild-type mice; fl, significantly reduced compared with wild-type mice.
NCAM, PSA and synaptic plasticity in vivo
2263
ª The Authors (2006). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 23, 2255–2264

Page 10

Dityatev, A., Dityateva, G., Sytnyk, V., Delling, M., Toni, N., Nikonenko, I.,
Muller, D. & Schachner, M. (2004) Polysialylated neural cell adhesion
molecule promotes remodeling and formation of hippocampal synapses. J.
Neurosci., 24, 9372–9382.
Dityatev, A. & Schachner, M. (2003) Extracellular matrix molecules and
synaptic plasticity. Nat. Rev. Neurosci., 4, 456–468.
Eckhardt, M., Bukalo, O., Chazal, G., Wang, L., Goridis, C., Schachner, M.,
Gerardy-Schahn, R., Cremer, H. & Dityatev, A. (2000) Mice deficient in the
polysialyltransferase ST8SiaIV ⁄ PST-1 allow discrimination of the roles of
neural cell adhesion molecule protein and polysialic acid in neural
development and synaptic plasticity. J. Neurosci., 20, 5234–5244.
Errington, M.L., Bliss, T.V., Morris, R.J., Laroche, S. & Davis, S. (1997) Long-
term potentiation in awake mutant mice. Nature, 387, 666–667.
Fazeli, M.S., Breen, K., Errington, M.L. & Bliss, T.V. (1994) Increase in
extracellular NCAM and amyloid precursor protein following induction of
long-term potentiation in the dentate gyrus of anaesthetized rats. Neurosci.
Lett., 169, 77–80.
Fux, C.M., Krug, M., Dityatev, A., Schuster, T. & Schachner, M. (2003)
NCAM180 and glutamate receptor subtypes in potentiated spine synapses:
an immunogold electron microscopic study. Mol. Cell Neurosci., 24, 939–
950.
Hildebrandt, H., Becker, C., Murau, M., Gerardy-Schahn, R. & Rahmann, H.
(1998) Heterogeneous expression of the polysialyltransferases ST8Sia II and
ST8Sia IV during postnatal rat brain development. J. Neurochem., 71, 2339–
2348.
Hollrigel, G.S., Morris, R.J. & Soltesz, I. (1998) Enhanced bursts of IPSCs in
dentate granule cells in mice with regionally inhibited long-term potentiation.
Proc. Biol. Sci., 265, 63–69.
Horstkorte, R., Schachner, M., Magyar, J.P., Vorherr, T. & Schmitz, B. (1993)
The fourth immunoglobulin-like domain of NCAM contains a carbohydrate
recognition domain for oligomannosidic glycans implicated in association
with L1 and neurite outgrowth. J. Cell Biol., 121, 1409–1421.
Kattenstroth, G., Tantalaki, E., Sudhof, T.C., Gottmann, K. & Missler, M.
(2004) Postsynaptic N-methyl-D-aspartate receptor function requires alpha-
neurexins. Proc. Natl. Acad. Sci. USA, 101, 2607–2612.
Kleene, R. & Schachner, M. (2004) Glycans and neural cell interactions. Nat.
Rev. Neurosci., 5, 195–208.
Kleschevnikov, A.M. & Routtenberg, A. (2001) PKC activation rescues LTP
from NMDA receptor blockade. Hippocampus, 11, 168–175.
Lin, B., Arai, A.C., Lynch, G. & Gall, C.M. (2003) Integrins regulate NMDA
receptor-mediated synaptic currents. J. Neurophysiol., 89, 2874–2878.
Luthi, A., Laurent, J.P., Figurov, A., Muller, D. & Schachner, M. (1994)
Hippocampal long-term potentiation and neural cell adhesion molecules L1
and NCAM. Nature, 372, 777–779.
Muller, D., Djebbara-Hannas, Z., Jourdain, P., Vutskits, L., Durbec, P., Rougon,
G. & Kiss, J.Z. (2000) Brain-derived neurotrophic factor restores long-term
potentiation in polysialic acid-neural cell adhesion molecule-deficient
hippocampus. Proc. Natl. Acad. Sci. USA, 97, 4315–4320.
Muller, D., Wang, C., Skibo, G., Toni, N., Cremer, H., Calaora, V., Rougon, G.
& Kiss, J.Z. (1996) PSA-NCAM is required for activity-induced synaptic
plasticity. Neuron, 17, 413–422.
Nosten-Bertrand, M., Errington, M.L., Murphy, K.P., Tokugawa, Y., Barboni,
E., Kozlova, E., Michalovich, D., Morris, R.G., Silver, J., Stewart, C.L.,
Bliss, T.V. & Morris, R.J. (1996) Normal spatial learning despite regional
inhibition of LTP in mice lacking Thy-1. Nature, 379, 826–829.
Ong, E., Nakayama, J., Angata, K., Reyes, L., Katsuyama, T., Arai, Y. &
Fukuda, M. (1998) Developmental regulation of polysialic acid synthesis in
mouse directed by two polysialyltransferases, PST and STX. Glycobiology,
8, 415–424.
Pillai-Nair, N., Panicker, A.K., Rodriguiz, R.M., Gilmore, K.L., Demyanenko,
G.P., Huang, J.Z., Wetsel, W.C. & Maness, P.F. (2005) Neural cell adhesion
molecule-secreting transgenic mice display abnormalities in GABAergic
interneurons and alterations in behavior. J. Neurosci., 25, 4659–4671.
Ronn, L.C., Bock, E., Linnemann, D. & Jahnsen, H. (1995) NCAM-antibodies
modulate induction of long-term potentiation in rat hippocampal CA1. Brain
Res., 677, 145–151.
Rutishauser, U. & Landmesser, L. (1996) Polysialic acid in the vertebrate
nervous system: a promoter of plasticity in cell–cell interactions. Trends
Neurosci., 19, 422–427.
Saghatelyan, A.K., Dityatev, A., Schmidt, S., Schuster, T., Bartsch, U. &
Schachner, M. (2001) Reduced perisomatic inhibition, increased excitatory
transmission, and impaired long-term potentiation in mice deficient for the
extracellular matrix glycoprotein tenascin-R. Mol. Cell Neurosci., 17, 226–
240.
Schmidt-Hieber, C., Jonas, P. & Bischofberger, J. (2004) Enhanced synaptic
plasticity in newly generated granule cells of the adult hippocampus. Nature,
429, 184–187.
Schuster, T., Krug, M., Hassan, H. & Schachner, M. (1998) Increase in
proportion of hippocampal spine synapses expressing neural cell adhesion
molecule NCAM180 following long-term potentiation. J. Neurobiol., 37,
359–372.
Staubli, U., Chun, D. & Lynch, G. (1998) Time-dependent reversal of long-
term potentiation by an integrin antagonist. J. Neurosci., 18, 3460–3469.
Stork, O., Welzl, H., Wolfer, D., Schuster, T., Mantei, N., Stork, S., Hoyer, D.,
Lipp, H., Obata, K. & Schachner, M. (2000) Recovery of emotional
behaviour in neural cell adhesion molecule (NCAM) null mutant mice
through transgenic expression of NCAM180. Eur. J. Neurosci., 12, 3291–
3306.
Sytnyk, V., Leshchyns’ka, I., Delling, M., Dityateva, G., Dityatev, A. &
Schachner, M. (2002) Neural cell adhesion molecule promotes accumulation
of TGN organelles at sites of neuron-to-neuron contacts. J. Cell Biol., 159,
649–661.
Takasu, M.A., Dalva, M.B., Zigmond, R.E. & Greenberg, M.E. (2002)
Modulation of NMDA receptor-dependent calcium influx and gene
expression through EphB receptors. Science, 295, 491–495.
Tang, J., Wotjak, C.T., Wagner, S., Williams, G., Schachner, M. & Dityatev, A.
(2001) Potentiated amygdaloid auditory-evoked potentials and freezing
behavior after fear conditioning in mice. Brain Res., 919, 232–241.
Walsh, F.S. & Doherty, P. (1997) Neural cell adhesion molecules of the
immunoglobulin superfamily: role in axon growth and guidance. Annu. Rev.
Cell Dev. Biol., 13, 425–456.
Weinhold, B., Seidenfaden, R., Rockle, I., Muhlenhoff, M., Schertzinger, F.,
Conzelmann, S., Marth, J.D., Gerardy-Schahn, R. & Hildebrandt, H. (2005)
Genetic ablation of polysialic acid causes severe neurodevelopmental defects
rescued by deletion of the neural cell adhesion molecule. J. Biol. Chem., 280,
42971–42977.
2264L. Stoenica et al.
ª The Authors (2006). Journal Compilation ª Federation of European Neuroscience Societies and Blackwell Publishing Ltd
European Journal of Neuroscience, 23, 2255–2264