| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vol. 7, No. 6, pp. 383-392, November/December 2000
1 Department of Integrative Physiology, Graduate School of Medical Sciences Kyushu University, Fukuoka 812-8582 Japan; 2 Department of Pathology, Osaka University Graduate School of Medicine, Suita 565 Japan
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
RESULTS
DISCUSSION
MATERIALS AND METHODS
REFERENCES
|
|---|
The c-kit receptor tyrosine kinase encoded by the white-spotting (W) gene is highly expressed in rat hippocampal CA1-CA4 regions. We found an impaired spatial learning and memory in homozygous c-kit (Ws/Ws) mutant rats that have a 12-base deletion in the tyrosine kinase domain of the c-kit gene and a very low kinase activity. Electrophysiological studies in hippocampal slices revealed that the long-term potentiation (LTP) induced by the tetanic stimulation (100 Hz, 1 sec) in the mossy fiber (MF)-CA3 pathway, but not in the Schaffer collaterals/commissural-CA1 pathway, was significantly reduced in c-kit mutants compared with wild-type (+/+) rats. The paired-pulse facilitation (PPF) was measured before the tetanus and after the establishment of the LTP in each slice. The initial PPF in the MF-CA3 pathway positively correlated with the amplitude of the LTP in the wild-type rats but not in the c-kit mutant rats. Furthermore, they failed to show the normal characteristics observed in the MF-CA3 pathway of +/+ rats; that is, the negative correlation between the initial PPF and the changes in PPF measured after the LTP. These findings suggest an involvement of SCF/c-kit signaling in hippocampal synaptic potentiation and spatial learning and memory.
| |
INTRODUCTION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
RESULTS
DISCUSSION
MATERIALS AND METHODS
REFERENCES
|
|---|
The c-kit receptor is a member of a receptor
tyrosine kinase family that is structurally related to the
receptors for a colony-stimulating factor-1 (CSF-1) and
platelet-derived growth factor (PDGF) (Qui et al. 1988
). This receptor
is encoded by the proto-oncogene c-kit located at the white
spotting (W) locus in rodents (Chabot et al. 1988). The
ligand for the c-kit receptor tyrosine kinase has been
designated as a stem cell factor (SCF), mast cell growth factor, or
c-kit ligand, which is mapped to the Steel
(Sl) locus (Huang et al. 1990; Zsebo et al. 1990a). The
c-kit receptors are expressed on the cell surface of
erythrocytes, mast cells, melanocytes, germ cells, and the precursors
of these cells, whereas cells surrounding them such as fibroblasts
produce SCF. It has been shown that SCF plays an important role in the
differentiation and proliferation of these cells working in a paracrine
fashion (Williams et al 1992). Furthermore, SCF is synthesized as
membrane bound forms as well as soluble forms (Zsebo et al. 1990b),
suggesting that works as a cell-adhesion molecule (Adachi et al. 1992).
The gene expression of c-kit receptors and SCF has also been
demonstrated in the central nervous system (CNS) of mouse embryo (Matsui et al. 1990). The c-kit/SCF signaling is shown to be
involved in cell migration, proliferation, differentiation of neural
crest-derived melanocyte precursors (Langtimm-Sedlak et al. 1996) and
outgrowth of neurites in dorsal root ganglia neurons of mouse embryos
(Hirata et al. 1993). Furthermore, in adult mice and rats, mRNA for SCF is strongly expressed in the neurons of the entorhinal cortex and the
granule cells in the dentate gyrus, whereas the transcripts of
c-kit are found in pyramidal cells in the area CA1-CA4, among which the expression in the CA3 region is the strongest (Motro et al.
1991; Hirota et al. 1992; Wong and Licinio 1994). The cellular localization of SCF and c-kit proteins in the adult mouse
hippocampus was further characterized by an immunocytochemical study
(Zhang and Fedoroff 1997). In the MF-CA3 pathway, SCF was found in
cell bodies of the granule cells in the dentate gyrus, whereas heavy immunostaining for c-kit was localized in the dendrites of
pyramidal neurons in the stratum radiatum. These findings raise a
possibility that c-kit/SCF signaling is involved in the
synaptic transmission in the hippocampus and possibly in the
hippocampal-dependent spatial learning and memory. In fact, mice
carrying heterozygous mutation at Sl locus
(Sl/Sld) have a deficit in
hippocampal learning and a selective reduction of baseline synaptic
transmission between the dentate gyrus and CA3 pyramidal cells,
although they exhibit normal long-term potentiation (LTP) in this
hippocampal pathway (Motro et al. 1996). This mutation is characterized
by the lack of the membrane-bound form of SCF, but the soluble SCF is
preserved (Flanagen et al. 1991). It is unknown whether or not the
dissociation between spatial memory and LTP is related to the presence
of the soluble SCF and intact c-kit receptors in
Sl/Sld mice.
In the present study, we investigated the possible role of
c-kit/SCF signaling in hippocampal learning behavior and
synaptic plasticity using homozygous c-kit receptor mutant
(Ws/Ws) rats having a 12-base deletion in the
tyrosine kinase domain of the c-kit gene (Tsujimura et al.
1991) and a very low kinase activity (Tei et al. 1994). The Morris
water maze task was used to test a hippocampal-dependent spatial
learning and memory. Extracellular field potentials were recorded from
the Schaffer collaterals/commissural (S/C)-CA1 and the mossy fiber
(MF)-CA3 pathways in hippocampal slices. We then examined the
paired-pulse facilitation (PPF), LTP, and the changes in PPF after the
establishment of LTP.
| |
RESULTS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
RESULTS
DISCUSSION
MATERIALS AND METHODS
REFERENCES
|
|---|
Spatial Learning and Memory
A hidden platform version of the Morris water maze task was applied
to Ws/Ws (n = 6) and +/+
(n = 6) rats. Each animal received four trials (one block)
twice a day for two consecutive days. The latency to find the platform
is depicted in Figure 1A. There was an
overall significant difference in escape latency between mutant and
control +/+ rats (F1,40 = 26.68, P <0.01,
two-way ANOVA). In both groups, the escape latency remarkably decreased
at the second block in comparison with the first block. However, mutant rats displayed no further improvement in the following blocks, whereas
control rats showed a further decrease in escape latency over training
blocks. A post-hoc test revealed that mutant rats needed significantly
longer time to reach the platform at the fourth block on day 2 than
control rats (19.0 ± 5.0 vs. 6.5 ± 1.4 sec,
P <0.05, Bonferroni's test). Because the swimming
velocity was not different between two groups throughout four blocks
(Fig. 1B), the difference in escape latency was not due to the motor deficit in mutant rats.
|
The probe test was performed just after the fourth block (the sixteenth trial) on day 2. The platform was removed and the animals searched for the missing platform for 60 sec. There was a significant difference in search time in both groups on day 2 ( +/+, F3,20 = 16.80, P <0.01, and Ws/Ws, F3,20 = 15.12, P <0.01, one-way ANOVA) (Fig. 1C). A post-hoc test revealed that both groups spent significantly longer time in the quadrant that had previously contained the hidden platform (the quadrant 1) than in any other quadrants (P <0.01 in both groups, Bonferroni's test). However, the number crossing the trained goal position in mutant rats was significantly smaller than that in control rats (3.8 ± 0.5 (n = 6) vs. 5.2 ± 0.3 (n = 6), P <0.05). Then the animals were returned to individual cages, kept there for three days and re-examined for their performance in the probe test on day 5. Control rats spent significantly longer time in quadrant 1 than in the test on day 2 (F3,20 = 6.08, P <0.05, one-way ANOVA and P <0.01, Bonferroni's test), but mutant rats did not show significant preference for any quadrants (F3,20 = 0.48, P >0.05) (Fig. 1D). In addition, the average of the cross count of mutant rats (1.2 ± 0.5) was significantly lower than that of control rats (4.7 ± 0.6) (P <0.01). These findings indicate an impairment of spatial learning and memory in c-kit mutant rats. Figure 1E shows examples of individual traces for each rat on day 2 (top) and day 5 (bottom).
PPF
Eighteen rats (four each for the S/C-CA1 experiment, and five each for the MF-CA3 experiment) were used for electrophysiological study.
PPF is a short-lasting increase in a second evoked excitatory
postsynaptic potential (EPSP) when it is elicited shortly after the
first one, and is thought to be mediated exclusively by presynaptic mechanisms. Because PPF may depend on the amplitude of the first EPSP,
we adjusted the stimulus intensity so that the amplitude was almost the
same between +/+ and mutant rats (Table 1).
In addition, the intensity was adjusted to evoke about one-half of the
maximum responses in both the CA1 and CA3 regions. The amplitude and
the slope of population EPSP (pEPSP) in absolute values recorded from
the CA1 region were larger than those recorded from the CA3 region.
However, the average value of the stimulus intensity was not different
between mutant (0.34 ± 0.07 mA, n = 12) and controls (0.32 ± 0.05 mA, n = 10) for the CA1 and CA3
experiments (mutant, 0.56 ± 0.05 mA, n = 15; controls,
0.52 ± 0.07 mA, n = 11). This finding suggests that the
basal synaptic transmission was not impaired in mutant rats in either
the CA1 or CA3 region. The average values at interstimulus intervals of
80 ms were not significantly different from those at 40 ms (data not
shown). Thus, the amplitude and slope of the first pEPSP did not change
during PPF trials (10 times every 10 sec at each interstimulus
interval). Because it was possible that active currents from the first
stimulation were still present at the 40-ms interval, which may affect
the second pEPSP, the following analysis of PPF was performed at an interstimulus interval of 80 ms.
|
Figure 2 Aa and Ba show the averaged
waveforms recorded from the areas CA1 and CA3, respectively, at a 80-ms
interstimulus interval in control (+/+, top) and mutant
(Ws/Ws, bottom) rats before tetanic stimulation. PPF
in area CA1 was not significantly different between +/+ and mutant rats
at both 40 and 80 ms (Bonferroni's test, P >0.05). PPF
recorded from CA3 in mutant rats was reduced at both 40- and 80-ms
interstimulus intervals compared with that in control rats (Fig. 2Bb)
(40 ms, 141.7 ± 9.0 vs. 179.2 ± 9.7%, P <0.01;
80 ms, 112.1 ± 4.3 vs. 140.8 ± 6.5%, P <0.01;
n = 15 and 11, respectively).
|
LTP
After observing the PPF, a tetanic stimulation (100 Hz, 1 sec) with
the same intensity used in the PPF study was delivered to stimulate the
CA1 or CA3 synapses. Induced synaptic enhancement as measured by the
slope of pEPSP at the CA1 and CA3 regions in control and mutant rats is
shown in Figure 3Aa and Ba. All of the
responses in the post-tetanic period were significantly greater than
those of pre-tetanic stimulation baseline. The pEPSP slope of the CA1
and CA3 regions in mutant rats was significantly smaller in the
post-tetanic periods of 2-47 min (P <0.01, Bonferroni's test), and 11-60 min (P <0.01), respectively, compared
with those in control rats. The data are summarized in Figure 3Ab and
Bb in the forms of post-tetanic potentiation (PTP, mean pEPSP slope during the period 1-2 min after tetanus) and persistent LTP (55-60 min). There is an overall difference between the groups (for CA1, F1,80 = 14.21, P <0.01, two-way ANOVA, and
for CA3,1,96 = 26.57, P <0.01). As for the
S/C-CA1 pathway, Bonferroni's post-hoc test indicated that PTP
(207.7 ± 12.5%) but not LTP (180.6 ± 15.9%) in mutant rats
were significantly smaller than that in control rats (PTP,
287.0 ± 36.4%, P <0.05; LTP, 206.9 ± 9.7%,
P >0.05). In the MF-CA3 pathway, the differences were
significant in LTP (139.5 ± 5.2 vs. 224.1 ± 16.0%,
P <0.01), but not in PTP (217.7 ± 18.7 vs.
255.5 ± 18.1%, P >0.05).
|
PPF after Tetanic Stimulation
Sixty minutes after the induction of LTP, PPF was again examined at
the same stimulus intensity and interstimulus intervals. Examples of
the averaged waveforms of control and mutant rats in both CA1 (Fig.
4Aa) and CA3 (Fig. 4Ba) were recorded from
the same slice preparation as those shown in Figure 2Aa (CA1) and Ba
(CA3) corresponding to the respective traces.
|
Although the mean value of CA1 PPF in control rats seemed to decrease after tetanic stimulation (40 ms, before tetanus, 210. 2 ± 16.2% vs. after tetanus, 172.5 ± 16.8%; 80 ms, 192.8 ± 15.8 vs. 174.2 ± 25.4%), there was no significant difference. In mutant rats, the mean CA1 PPF also showed no significant changes after tetanic stimulation (40 ms, before tetanus, 179.7 ± 14.1% vs. after tetanus, 166.4 ± 11.6%, P >0.05; 80 ms, 187.6 ± 15.4 vs. 170.0 ± 11.5%, P >0.05). Thus, there were no differences in the mean CA1 PPF between controls and mutants (Fig. 4Ab) as observed before tetanus (Fig. 2Ab).
In the MF-CA3 pathway, the averages of PPF after tetanic stimulation in control rats were significantly smaller than those before tetanus (40 ms, before tetanus, 179.2 ± 9.7% vs. after tetanus, 121.8 ± 5.4%, P <0.01; 80 ms, 140.8 ± 6.5% vs. 119.4 ± 2.0%, P <0.01). In mutant rats, the mean CA3 PPF showed no significant changes after tetanic stimulation (40 ms, before tetanus, 141.7 ± 9.0% vs. after tetanus, 136.5 ± 5.3%, P >0.05; 80 ms, 112.1 ± 4.3% vs. 120.1 ± 5.4%, P >0.05). Thus, in CA3, the difference in PPF between control and mutants that was significant before tetanus (Fig. 2Bb), disappeared when examined after LTP (Fig. 4Bb).
Correlation between PPF, PTP, and LTP in CA1
To further characterize the interrelations of synaptic potentiations
in control and mutant rats, the correlation analysis was performed by
use of the initial PPF at an 80-ms interstimulus interval, PTP and LTP
of each slice. In the S/C-CA1 pathway, there was a positive
correlation between PPF and PTP (Fig. 5A,
and a broken line, r = 0.78, P <0.01,
n = 10, y = 1.79x 
57.2) in control rats, whereas CA1 PPF did not correlate with PTP in mutant rats
(Fig. 5A, 
and solid line, r = 0.37,
P >0.05, n = 12, y = 0.29x + 152.4). In contrast to the positive correlation between PPF and PTP,
there was no significant correlation between PPF and LTP (Fig. 5B,
, r = 0.27, P >0.05, n = 10,
y = 0.14x + 179.7) and between PTP and LTP (Fig.
5C, , r = 0.50, P >0.05, n = 10, y = 0.12x + 173.8) in control rats, or in mutant
rats (PPF vs. LTP, Fig. 5B, , r = 0.43,
P >0.05, n = 12, y = 0.37x + 111.3; PTP vs. LTP, Fig. 5C, , r = 0.01,
P >0.05, n = 12,
y = 0.02x + 183.7).
|
Correlation Between PPF, PTP, and LTP in CA3
In the area CA3, PTP positively correlated with the initial PPF in
control rats (Fig. 6A, ,
r = 0.89, P <0.01, n = 11,
y = 2.47x 92.0) as in area CA1 (Fig. 5A).
Furthermore, there was also positive correlation between PPF and LTP
(Fig. 6B, , r = 0.80, P <0.01,
n = 11, y = 1.95x 50.6) and
between PTP and LTP (Fig. 6C, , r = 0.72,
P <0.05, n = 11,
y = 0.63x + 63.0). It is possible that not only
MF but also the associational/commissural pathway, which is another
excitatory input to CA3 pyramidal cells showing the NMDA-dependent LTP,
may also be stimulated in the present study. However, it was clearly
demonstrated that PPF and PTP correlated positively with LTP in CA3
(Fig. 6B,C), whereas there was no correlation between them recorded
from area CA1 in control rats (Fig. 5). In addition, the coaxial
stimulating electrode (tip diameter, 0.3 mm) was placed just on the
granule cell layer. Therefore, taken together, the contamination of the
associational/commissural pathway seemed to be only the minimum in the
present study.
|
In mutant rats, PTP positively correlated with the initial PPF (Fig.
6A, , r = 0.75 P <0.01, n = 15,
y = 3.7x 200.2), as in control rats (Fig.
6A, ). However, the positive correlation between PPF and LTP (Fig.
6B, , r = 0.15, P >0.05, n = 15,
y = 0.21x + 116.5) and between PTP and LTP (Fig.
6C, , r = 0.01, P >0.05,
n = 15, y = 0.03x + 138.8) was
abolished in c-kit mutant rats.
Correlation between PPF before and after LTP
It has been shown that PPF changed specifically in association with
LTP in the S/C-CA1 pathway, when the PPF is measured before and after
LTP (Schulz et al. 1994). Thus, the correlation between the initial PPF
and changes in PPF after the establishment of LTP was investigated. In
Figure 7A, each value of PPF before tetanic stimulation at an 80-ms interstimulus interval (initial PPF) was plotted against changes in PPF after LTP (calculated as PPF at test
stimulation intensity measured after LTP minus initial PPF). In +/+
control rats, the changes in CA1 PPF with LTP showed an inverse
relationship with the initial PPF (Fig. 7A, , r = 0.63, P < 0.05, n = 10,
y = 0.37x + 41.7), which was consistent with the findings in the previous studies (Schulz et al. 1994). In Ws/Ws mutant rats, the inverse relationship between
them was also observed as in control rats (Fig. 7A, and a solid
line, r = 0.66, P <0.05, n = 12,
y = 0.47x + 69.7).
|
In the area CA3, the changes in PPF after LTP in control rats
correlated inversely with initial PPF (Fig. 7B, and broken line,
r = 0.95, P < 0.01, n = 11,
y = 0.96x + 113.5) as in area CA1. This
change in PPF in association with LTP is thought to be consistent with
a predominant involvement of presynaptic components in the mechanisms
of formation of MF-CA3 LTP. In contrast to CA1, there was no
significant correlation between the initial PPF and changes in PPF
after LTP in mutant rats (Fig. 7B, and solid line,
r = 0.40, P > 0.05, n = 15,
y = 0.62x + 77.3).
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
RESULTS
DISCUSSION
MATERIALS AND METHODS
REFERENCES
|
|---|
The aim of this study was to examine the involvement of the
c-kit signaling in learning and memory and synaptic
potentiation by use of c-kit mutant rats, in which the
activity of c-kit receptor tyrosine kinase is very low (Tei et
al. 1994). The previous study using heterozygous
Sl/Sld mutant mice, which are characterized
by the lack of the membrane-bound form of SCF, but not the soluble-type
SCF (Flanagen et al. 1991), showed a deficit in spatial learning and
memory, reduction of baseline synaptic transmission but normal LTP in
the MF-CA3 pathway (Motro et al. 1996). In the present study, the
c-kit (Ws/Ws) mutant rats also demonstrated
an impairment of spatial learning. However, in contrast to
Sl/Sld mutant mice, the synaptic
potentiation such as LTP and PPF and the positive correlation between
PPF and LTP in the MF-CA3 pathway were deteriorated in c-kit
mutant rats. Although there is a difference in species (mice vs. rats),
these findings give further evidence for an importance of
c-kit/SCF signaling in learning and memory and hippocampal
synaptic potentiation. Furthermore, our findings suggest an importance
of the soluble-type SCF in hippocampal synaptic potentiation. The
normal LTP observed in Sl/Sld mice (Motro
et al. 1996) may be due to the presence of soluble-type SCF and intact
c-kit receptors.
Impairment of Synaptic Potentiation in c-kit Mutants
The present study demonstrated (1) that the reduction of the initial PPF was evident in CA3 but not in CA1 regions (Fig. 2Ab, Bb), and (2) that the LTP in the MF-CA3 pathway, but not in the S/C-CA1 pathway, was significantly smaller than that of +/+ control rats (Fig. 3). Therefore, the impairment of c-kit mutant rats seemed to be more severe in the MF-CA3 pathway than in the S/C-CA1 pathway. In the present study, we analyzed not only the averages of PPF, PTP, and LTP but also the correlation among them, which were shown to be useful to detect the differences in synaptic potentiation between control and mutant rats.
The induction and expression mechanisms of LTP in the MF-CA3 pathway
are shown to be primarily of presynaptic origin (Zalutsky and Nicoll
1990; Castillo et al. 1994; Nicoll and Malenka 1995; Son and Carpenter
1996a). The PPF, as well as the PTP, is also thought to be a
presynaptically mediated phenomenon that is associated with an increase
in transmitter release probability and mean quantal content resulting
from a transient increase in presynaptic Ca2+ triggered by
the first stimulation (Kuhnt and Voronin 1994; Schulz et al. 1994).
Thus, it is likely that in the MF-CA3 pathway, PPF, LTP, and PTP share
common mechanisms. In fact, there was positive correlation between PPF
and LTP, between PPF and PTP, and between PTP and LTP in the CA3 region
of control rats (Fig. 6), as observed previously (Son and Carpenter
1996a). In the S/C-CA1 pathway, there was also the positive
correlation between PPF and PTP, but no correlation between PPF and LTP
or between PTP and LTP in control rats (Fig. 5). This may be reasonable
because the mechanism of CA1 LTP is considered to be mainly
postsynaptic in origin, whereas both PPF and PTP are presynaptic.
The positive correlation between PPF and PTP in CA1, which was observed
in control rats, was abolished in c-kit mutant rats (Fig. 5A).
This might be due to the reduction of the size of PTP (Fig. 3Ab) in
mutant rats. In contrast, the positive correlation between PPF and PTP
in CA3 was not abolished in mutant rats (Fig. 6A). In the MF-CA3
pathway, the amplitude of PPF in mutant rats was smaller than that in
controls (Fig. 2Bb), but the size of PTP was not affected (Fig. 3Bb).
Therefore, it is possible that the mechanisms that determine the
amplitude of PTP may be different between CA1 and CA3. On the other
hand, the positive correlation between PPF and LTP and between PTP and
LTP in CA3, which was observed in control rats, was abolished in mutant
rats (Fig. 6B,C). These findings suggest that c-kit mutant
rats have an impairment of the presynaptically mediated synaptic
potentiation, and thus c-kit may play a significant role in
synaptic potentiation in the MF-CA3 pathway. This finding is
consistent with our recent observation that the induction of the
tetanus-induced LTP in the MF-CA3 pathway of mouse slice preparations
is completely blocked by preincubation of slices with
anti-c-kit antibody for 2 hr (Katafuchi et al. 1998).
Negative Correlation between PPFs before and after LTP
There are controversial reports as to whether or not the
tetanus-induced LTP can modulate the amplitude of PPF in the S/C-CA1 pathway. The attenuation (Wang and Kelly 1997), no change (Manabe et
al. 1993), and both increase and decrease (Schulz et al. 1994) in PPF
magnitude after LTP have been reported. In spite of the diversity of
the changes in PPF amplitude, Schulz et al. (1994) have shown that a
larger initial PPF is associated with a decrease in PPF when measured
after LTP, whereas a smaller initial PPF is associated with an increase
in PPF after LTP (negative correlation). This may be explained as
follows: the presynaptic terminals with the low initial probability
(Pre-low) must show the large PPF, whereas those with high
probability (Pre-high) must show the small PPF. When LTP
induces an increase in the probability of neurotransmitter release at
the Pre-low, the originally large PPF would become small. On
the other hand, when LTP increases new release sites having a low
probability at the Pre-high, the small PPF would become large,
thereby producing the negative correlation between the initial PPF and
changes in PPF after LTP. Thus, the LTP-induced negative correlation
may indicate, at least in part, an involvement of presynaptic locus in
the expression mechanism of CA1 LTP, although others have shown that
the attenuation of PPF after LTP may be mediated by postsynaptic
mechanisms including postsynaptic Ca2+/CaM signaling pathways
(Wang and Kelly 1997). In the present study, when the changes in PPF
after LTP were plotted against the initial PPF, there was a negative
correlation between them in the S/C-CA1 pathway (Fig. 7A) as reported
by Schulz et al. (1994). Furthermore, we found that there was also the
negative correlation between them in the MF-CA3 pathway of control
rats (Fig. 7B). This is consistent with the view that both the
induction and expression mechanisms of LTP may be of a presynaptic
origin in the MF-CA3 pathway (Zalutsky and Nicoll 1990; Castillo et
al. 1994; Nicoll and Malenka 1995; Son and Carpenter 1996a).
The present study demonstrated that the negative correlation between
the initial PPF and the changes in PPF after LTP in the MF-CA3 pathway
was abolished in c-kit mutant rats (Fig. 7B), but was
preserved in the S/C-CA1 pathway in mutant rats (Fig. 7A). This
finding again suggests that c-kit/SCF signaling may play a
significant role in presynaptically mediated potentiation. It has been
demonstrated that in the MF-CA3 pathway c-kit receptor mRNA
is mainly expressed in CA3 pyramidal cells but very low in density in
the granule cells of the dentate gyrus in normal rats (Hirota et al.
1992). In mice there were no expressions of c-kit mRNA in
granule cells (Motro et al. 1991). These findings, taken together,
suggest that the c-kit/SCF signal transduction at the postsynaptic site (CA3 pyramidal cells) produces a retrograde signal(s)
that diffuses to the presynaptic membranes (MF terminals) and affects
transmitter release processes.
Possible Mechanisms of Deficit in Synaptic Potentiation in c-kit Mutants
The mechanisms of the c-kit/SCF signaling in neuronal cells
have not yet been determined. However, it has been shown that SCF
activates phospholipase D (PLD) through a phosphatidylinositol 3'-kinase-dependent pathway in cultured porcine aortic endothelial cells transfected with c-kit receptors, leading to the
production of phosphatidic acid. Phosphatidic acid is then metabolized
to yield diacylgycerol (DAG), which serves as an activator of protein kinase C (PKC), and as a precursor to arachidonic acid (AA) by the
action of DAG lipase and monoacylglycerol lipase. In addition, AA is
released by SCF through an alternative route that involves an
activation of phospholipase A2 (PLA2) (Kozawa et
al. 1997). Thus, one of the candidates that mediate the retrograde
synaptic transmission triggered by SCF may be AA, which is shown to
enhance depolarization-evoked glutamate release from hippocampal MF
nerve endings in a Ca2+- and presynaptic PKC-dependent way
(Zhang et al. 1996). We have shown that the initial value of PPF
negatively correlated with changes in PPF after the establishment of
LTP in the MF-CA3 pathway by use of mouse slice preparations, as
observed in control rats in the present study. Furthermore, bath
application of mouse recombinant SCF for 30 min induced a similar
negative correlation between the PPFs before and after the perfusion
(Katafuchi et al. 1998). Recently we have observed that the SCF-induced
negative correlation between the PPFs is abolished by the simultaneous
application of PLA2 inhibitor, quinacrine (T. Katafuchi, unpubl.).
AA has been shown to have synergistic action of DAG for PKC activation
(Shinomura et al. 1991), and induce PKC-dependent facilitation of
glutamate release from MF nerve endings (Zhang et al. 1996). Activation
of PKC is not sufficient but necessary for induction of LTP in the
MF-CA3 pathway (Son and Carpenter 1996b). Thus, the impairment of
synaptic potentiation in c-kit mutant rats may involve a
down-regulation of PKC activity in the presynaptic terminals. Furthermore, the suppressive action of subtype 2 of the metabotropic glutamate receptor (mGluR2) on synaptic transmission in the MF-CA3 pathway (Kamiya et al. 1996), may be enhanced in c-kit mutant rats, as PKC reduces the inhibitory effect produced by subtype 2 of
mGluR (Swartz et al. 1993).
Several other mechanisms may be considered as the cause of the
impairment of synaptic potentiation in c-kit mutant rats.
First, the cyclic AMP (cAMP)-protein kinase A (PKA) cascade is also
important presynaptically for LTP in the MF-CA3 pathway (Huang et al.
1994; Weisskopf et al. 1994). However, it has been shown that the
genetic ablation of either catalytic (C) 
1 or regulatory
(R) I subunit of PKA has no effect on PPF or spatial learning
despite the elimination of the MF-CA3 LTP (Huang et al. 1995). Second,
the significant reduction of the histamine content of the whole brain
in c-kit mutant rats due to the depletion of mast cells
(Sugimoto et al. 1995) may contribute to the impairment of synaptic
potentiation. However, it is unlikely that the MF-CA3 LTP, which is
known to be NMDA independent, is affected by histamine because it
facilitates the induction of LTP, most likely by acting directly on the
NMDA receptor (Brown et al. 1995). In addition, there was no
significant difference in the histamine content of the hippocampus
between c-kit mutant and control rats, possibly because of the
non-mast cell histamine pools including histaminergic neurons (Sugimoto et al. 1995). Finally, it has been shown that c-kit receptor
is also expressed in microglia and that SCF potentiates microglial expression of mRNA for brain-derived neurotrophic factor (BDNF) (Zhang
and Fedoroff 1998). Thus, it is possible that there is an impairment of
BDNF production in c-kit mutant rats, thereby resulting in a
deficit in the synaptic potentiation and learning and memory. Further
studies will be needed to obtain direct evidence for the mechanisms of
the impairment in c-kit mutant rats.
| |
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACT
INTRODUCTION
RESULTS
DISCUSSION
MATERIALS AND METHODS
REFERENCES
|
|---|
Rats
Adult (8-12 wk old), male Ws/Ws mutant and control +/+ rats weighing 200-300g were used. They were originally bred in the Department of Pathology, Osaka University, and were transferred to the colony in the Kyushu University at least 2 wk before the present work started. They were housed at 22 ± 2°C with artificial light illumination from 0700 to 1900 hours. Food and water were provided ad libitum. There was no obvious abnormality in general behavior in both groups of rats. Different animals were used for behavioral and electrophysiological studies.
The origin and breeding procedure of Ws/Ws rats have been
described previously (Niwa et al. 1991). The c-kit mutant rats
of Ws/Ws genotype have black eyes and white hair
because of the lack of melanocytes except for the retina.
TheWs/Ws rats showed hypoplastic anemia at birth, but
it tended to ameliorate 10 wk after birth. In addition, mast cells are
depleted in the mutant rats as in the W
(W/Wv) and Sl
(Sl/Sld) mutant mice (Niwa et al. 1991).
The c-kit cDNA of Ws/Ws mutant allele was found to
have a deletion of 12 bases, which was shown to be a result of the
deletion of the genomic DNA. Four amino acids encoded by the deleted 12 bases are located at two amino acids downstream from the tyrosine
autophosphorylation site in the c-kit kinase and are also
conserved in mouse and human c-kit kinases. The
Ws/Ws rat is the first characterized mutant of the c-kit gene in an animal species other the mouse (Tsujimura et al. 1991). An in vitro immune complex kinase assay indicated that the
c-kit kinase activity was apparently lower in cultured mast cells of Ws/Ws rats than those of +/+ control rats
(Tei et al. 1994).
Morris Water Maze Studies
To measure the ability of spatial memory and learning of rats, we compared the performance of Ws/Ws mutant (n = 6) and +/+ control rats (n = 6) in the hidden-platform version of the Morris water maze. Rats were brought to the testing room at least 5 d before behavioral testing. The apparatus was a black plastic circular pool (diameter, 150 cm; wall height, 76 cm) containing water at 22-23 °C. The pool was surrounded by a gray curtain wall, on which three rectangular, triangular, and circular drawings that were brightly illuminated were placed and served as the spatial cues. A circular, transparent plastic platform (diameter, 12 cm) was placed in one quadrant of the pool 2 cm below the surface of the water. The rats were released from one of four randomly chosen starting points in the circular pool for 60 sec to search for the hidden escape platform. They were allowed to rest for 30 sec on the platform after they found it. If the rats could not find the platform within 60 sec, the experimenter placed the animal on the platform for 30 sec. Rats were then placed for 60 sec in a waiting cage for the next trial and were dried under a heating lamp. The rats received four trials (one block) twice (1000-1200 hours and 1500-1700 hours) a day for two consecutive days. There were no rats that did not show motivation for swimming (floating behavior) both in mutant and control groups. The rats were tracked by an infrared-sensitive camera connected to a maze analysis unit (Toyo Industrial Co., MAZER MZ-40). A probe test was performed after the end of the fourth block. In this test, the rats were allowed to search for 60 sec with no platform present. The duration of cumulative time that they spent in each of the quadrants and the numbers crossing the trained platform position were measured in the probe test.
Electrophysiology
Under anesthesia with ether, rats were killed by a blow to the neck
and decapitated. The brain was quickly removed. The hippocampal slices
(400-µm thick) were made perpendicular to the septotemporal axis
for the study of the S/C-CA1 pathway or transverse to the axis for the
study of the MF-CA3 pathway. They were incubated in Krebs-Ringer
solution bubbled with 95% O2 and 5% CO2 at
32-34°C and at pH 7.4. After 2 hr pre-incubation, slices were
transferred to a recording chamber, which was perfused with
Krebs-Ringer solution at a constant rate of ~2.5 mL/min. The
Krebs-Ringer solution contained (in millimolars) the following: 124 NaCl, 4 KCl, 1.3 MgSO4, 1.23 NaH2PO4, 26 NaHCO3, 2.4 CaCl2, and 10 glucose. Extracellular
recording of population pEPSP was made from the stratum radiatum in the area CA1 or from the stratum lucidum in the area CA3 by a glass microelectrode filled with Krebs-Ringer solution (tip diameter, 15-25
µm; DC resistance, 3-5 
). Orthodromic stimuli were delivered through a coaxial bipolar electrode (diameter, 0.3 mm) that was placed
in the stratum radiatum in the CA3 region to stimulate the S/C-CA1
pathway or in the granule cell layer of the dentate gyrus to stimulate
the MF-CA3 pathway. The test-stimulus intensity of 50 µs square
pulses was adjusted to give a pEPSP amplitude of 0.2-0.7 mV at 0.03 Hz
(finally, 0.15-0.65 mA). After having confirmed that the amplitude of
pEPSP was stable for at least 10 min, PPF was measured with
interstimulus intervals of 40 and 80 ms. Paired-pulse stimuli were
applied 10 times at 0.1 Hz for each interval. Stable pEPSPs, the
amplitude of which was confirmed to be the same as that before PPF,
were again obtained for 20 min without changing stimulus intensity.
Then tetanic stimulation was delivered and data were collected for
another 60 min. To induce LTP, a train of 100 Hz for 1 sec was
delivered at the same stimulus intensity used for the test stimulus.
PPF was re-measured 60 min after the end of the tetanic stimulation
with the same interstimulus intervals.
Responses were acquired, digitized, and stored by a Macintosh computer
interfaced with MacLab (AD Instruments) at 20 kHz for 64 ms or at 10 kHz for 128 ms, beginning 4 ms prior to the stimulation. For the
analysis of PPF, the 10 traces at each interstimulus interval were
averaged and the slope of the middle two-thirds of the rising phase of
pEPSP (1.2-1.5 ms for S/C-CA1 and 0.8-1.0 ms for MF-CA3 pathways)
was measured in mV/ms. To avoid bias, the same time points on the
initial slope of pEPSP were analyzed throughout the experiment. PPF was
calculated as a percent ratio of the slope of the second pEPSP to the
first. For LTP analysis, the initial slope of the pEPSP at each time
point was expressed as a percentage of control values (averages before
tetanic stimulation). After a tetanic stimulation, PTP was followed by
a short-term potentiation (STP), which decayed within ~20 min, then
LTP, a more persistent and significant enhancement in the synaptic
transmission, lasted for hours (Bliss and Collingridge 1993). To
analyze data of individual sets on synaptic potentiation, the mean
percentages of the rising slope of the pEPSPs of each animal recorded
from 1 to 2 min and from 55 to 60 min after tetanus were used as PTP
and LTP, respectively. The averages of the pEPSP slopes recorded from 10 to 5 min before tetanus were served as the pre-tetanic stimulation baseline.
Statistics
The data were expressed as mean ± S.E.M. The statistical analysis for the differences in the parameters of a Morris water maze task such as escape latency and search time between c-kit mutant and +/+ rats was performed by two-way ANOVA. When a significant overall F score was obtained, the difference between groups at each block was analyzed by Bonferroni's modified t-test as a post-hoc test. The differences in PPF and synaptic potentiation after tetanus were also examined by Bonferroni's modified t-test. A linear regression test was used for examining the correlation between two parameters among the initial PPF, PPF after the tetanic stimulation, PTP, and LTP. P <0.05 was considered to be statistically significant.
| |
ACKNOWLEDGMENTS |
|---|
This work was supported by a Grant-in Aid for Scientific Research [(B) 10557008 and 12470011 to T.K.] from the Ministry of Education, Science, and Culture, and the Special Coordination Funds for the Science and Technology from Science and Technology Agency (STA) of Japan.
The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 USC section 1734 solely to indicate this fact.
| |
FOOTNOTES |
|---|
Received May 24, 2000; accepted in revised form September 29, 2000.
3 Present address: Cell Biology Laboratory, National Institute of Bioscience and Human Technology, 1-1 Higashi, Tsukuba, Ibaraki 305, Japan.
4 Corresponding author.
E-MAIL kataf{at}physiol.med.kyushu-u.ac.jp; FAX 81-92-642-6093.
Article and publication date are at www.learnmem.org/cgi/doi/10.1101/lm.33900.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
RESULTS
DISCUSSION
MATERIALS AND METHODS
REFERENCES
|
|---|
oncogenic activation of v-kit involves deletion of extracellular domain and C-terminus.
EMBO J.
7:
1003-1011[Medline].
-fluoromethylhistidine.
J. Neurochem.
65:
791-797[Medline].
CiteULike 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
) at an 80-ms interstimulus interval obtained from the S/C-CA1
pathway in +/+ (top) andWs/Ws mutant
(bottom) rats. Calibration bars; horizontal, 10 ms; vertical,
1 mV. (Ab) Mean values of PPF in +/+ (open bar,
n = 10) and mutant (solid bar, n = 12) rats at
each interstimulus interval. (Ba) Averaged waveforms of 10 traces evoked by paired pulses (
