Download Short-Lasting Classical Conditioning Induces

Short-Lasting Classical Conditioning
Induces Reversible Changes of
Representational Maps of Vibrissae in
Mouse SI Cortex—A 2DG Study
It has been known for several years that receptive field properties of
sensory cortical neurons can be altered by learning experiences. We
attempted to visualize a global change of the cortical body map induced by learning. In order to do this a short-duration classical conditioning involving stimulation of a row of mystacial vibrissae in mice
was followed with 2-deoxyglucose (2DG) mapping of functional activity. Three conditioning sessions that paired stimulation of a row of
whiskers with a tail shock produced an increase of the functional
representation in somatosensory cortex (SI) of a row of the whiskers
stimulated during the training. This plastic change of vibrissal representation in SI was visualized with 2DG autoradiography a day after
completion of training. The expansion of representation was the most
pronounced in cortical layer IV, and to a lesser extent in layer Illb.
The expansion was observed in conditioned but not in pseudoconditioned mice or in animals that received only the conditioned stimulus.
If training was discontinued, the enlargement of vibrissal representation progressively faded. The reversal could be accelerated by a
behavioral extinction procedure. This study gives the pictorial demonstration of rapid, transient, and extinguishable learning-dependent
changes in SI cortical maps.
The primary somatosensory cortex (SI) of adult mammals can
undergo plastic reorganization as a result of eliminating peripheral inputs (Wall and Egger, 1971; Merzenich et al., 1983;
for a recent review, see Kaas 1991) or changing sensory experiences (Delacour et al., 1987; Jenkins et al., 1990; Allard et
al., 1991; Diamond et al., 1993). The evidence is accumulating
that neuronal properties in primary sensory cortices also alter
during a special form of plasticity, for instance, learning
(Brons and Woody, 1980; Diamond and Weinberger, 1986;
Gonzalez-Lima and Scheich, 1986; Bakin and Weinberger,
1990; Recanzone et al., 1992a; Edeline et al., 1993;Recanzone
et al., 1993; Weinberger et al., 1993). The process of learning
may influence the properties of neurons at all levels of the
brain (also see Dudai, 1989). A permanent and stimulus-specific change in neuronal properties can be regarded as a part
of the permanent storage of information induced by behavior.
Such change may also be an adaptation of the sensory pn>
cessing brain structures for better perception of behaviorally
relevant stimuli. The work of several laboratories has demonstrated that in adult primary sensory cortical areas excitability of neurons and receptive field properties can be
changed as a result of sensory experience (Brons and Woody,
1980; Diamond and Weinberger, 1986; Gonzalez-Lima and
Scheich, 1986; Bakin and Weinberger, 1990; Recanzone et al.,
1992a, 1993; Edeline et al., 1993; Scheich et al., 1993). Particularly in somatosensory and auditory cortex, topographical
maps reorganize as a result of sensory training (Gonzalez-Iima
and Agudo, 1990; Recanzone et al., 1992a, 1993). The reorganization of neuronal properties can be strongly correlated
with improved perceptual skills (Recanzone et al., 1992b). An
increase of the cortical representation of a sensory surface
magnifies the power of stimuli activating this surface in the
global pattern of cortical activity. In this way, engagement of
specific sensory inputs presumably results in their more sali-
E. Siucinska and M. Kbssut
Department of Neurophysiology, Nencki Institute, 02-093
Warsaw, Poland
ent brain representation. Merzenich et al.(1990) have claimed
that such changes of cortical maps are normal, dynamic cortical processes by which selective, distributed responses of
cortical neurons are shaped by experience throughout life.
Body maps in the somatosensory cortex can be altered by
training involving tactile stimulation. For example, Jenkins et
al. (1990) and Recanzone et al. (1992a) described topographical reorganizations of hand representations in the SI cortices
of monkeys trained in regulated finger contact or frequency
discrimination behavioral tasks. Hand and Hand (1995) reported that the area of 2DG labeling of cortical functional
representation of a mystacial vibrissa (a vibrissal column) in
a rat was enlarged following a prolonged period of pairing
sensory stimulation with alimentary award. A long-lasting sensory stimulation of whiskers, unrelated to behavioral experience, which can be considered habituation, leads to a decrease in activation (2DG uptake) of cortical representation
of vibrissae (Welker et al., 1992).The vibrissal pathway is well
suited for such cortical response "mapping" experiments because rodent SI cortex contains barrel-shaped neuronal aggregates ("barrels") in layer IV that form a pattern replicating the
whisker distribution on the snout; the special anatomical and
cytoarchitectonic structure allows for an easy localization of
the central representations of particular vibrissae (Woolsey
and Van der Loos, 1970). Considerable plasticity of the vibrissal system in adult rodents has been reported by several investigators working with different experimental approaches
and experimental techniques (Hand 1982; Kossut et al., 1988;
Levin and Dunn-Meynell, 1991; Fox 1992; Kossut, 1992; Diamond et all, 1993; Armstrong-James et al., 1994; Glazewski and
Fox, 1994; Siucinska and Kossut, 1994a,b). We have used this
pathway in mice to determine whether or not changes in
cortical body maps can be generated by a brief training that
does not involve an extensive use of special sensory stimulation, and to asses the stability of any induced plastic
changes. In particular; we wanted to know if a body map
change could be induced in SI by a few minutes of classical
conditioning and whether such changes could be reversed by
an extinction procedure, designed to obliterate the previous
signaling significance of the conditioned stimulus.
Materials and Methods
'
Behavioral Training
The experiments were performed on 34 young adult (7-8 weeks old
during conditioning) Swiss mice. The mice were accustomed to a
rieck restraint by being placed in a restraining apparatus for 10 min
a day for 3 weeks prior to behavioral conditioning. The training consisted of stroking the whiskers of row B on one side of the muzzle
(conditioned stimulus, CS) with a hand-held fine camel hair brush in
a posterior to anterior direction. The brush was held at an angle so
that it was parallel to the vibrissae. Great care was taken not to touch
adjacent rows of whiskers. The CS lasted 9 sec and consisted of three
strokes. One stroke took 3 sec, at the last second of the last stroke a
single tail shock (UCS) was applied (0.5 sec, 0.5 mA). The electrical
stimulation was discontinued simultaneously with the end of strok-
Cerebral Cortex May/Jun 1996;6:506-513; 1047-3211/96/$4.00
ing. The stimulator was controlled manually. After a 6 sec interval the
trial was repeated. Pairings were repeated four times/min for 10 min/
d for 3 d. Altogether, these animals (n = 7) received 120 pairings of
CS + UCS. One day after the end of training the cortical representation of row B was mapped with 2DG. A second group of mice (n
= 4) had the 2DG mapping experiment done with a delay of 3 d
after the end of training, and a third one (n = 4) with a 5 d delay.
The control group (n = 4) received only the CS delivered at the
same schedule. In a pseudoconditioned group (n = 4) the stimuli
were applied at random; these mice received the same number of
CS and UCS per session as applied for conditioning. In the extinction
procedure, after 3 d of associative CS + UCS conditioning, the mice
(n = 4) received whisker stroking only, delivered at the same frequency as during the training, but with no UCS.
In a separate control group (n = 4) of mice the heart rate was
monitored during the CS + UCS pairing. Pin electrodes were inserted
in to the skin under local anesthesia and an ECG trace was registered
with a polygraph.
2-Deoxygtucose Mapping of Cortical Representation of Row B
of Vibrissae
Images of the cortical representation of row B of vibrissae were obtained with [uC]2-deoxy-D-glucose functional activity mapping (Sokolpff et al., 1977). 2DG experiments were performed 1, 3, or 5 d
after the end of training in the CS + UCS, 1 d after in the CS only,
pseudoconditioned, and training + extinction groups.
Ten minutes prior to 2DG injection all whiskers except row B
were cut close to the skin on both sides of the snout. The mice were
taped by their trunks to plastic, soft-padded blocks. The whiskers of
row B on both sides of the snout were stimulated. We used the standard stimulation format employed in our laboratory for mapping of
cortical representation of vibrissae. The vibrissae were stroked along
the row, with brushes held in mechanical stimulators, stroking with
2 Hz frequency, for 45 min following i.p. injection of [14C]2-deoxy-r>
glucose (Amersham, s.a. 55 Ci/mmol) in dose of 5 (iCi/mouse. Mice
were sacrificed by an overdose of Nembutal and perfused with 3-3%
formalin. The brains were dissected and cut along the midline and
frozen at -70°C in isopentane. They were sectioned on a cryostat at
— I8°C in the plane tangential to the barrel field into 20 u,m sections.
The sections were rapidly dried on a hot plate and the autoradiograms of sections and [14C] radioactivity standards (American Radiochemical) were obtained on Kodak mammography film. After obtaining the autoradiograms, brain sections were counterstained with
cresyl violet for identification of cortical layers and the barrels in
layer TV.
Quantification of the Autoradiograms
The autoradiograms were analyzed with a computer-controlled image
analysis system (Visionetics). Images were corrected for film background and optical distortions by subtraction of a clearfield image of
the film. A calibration curve was created on the basis of the absolute
gray levels of the [14C] standards on thefilm.The width of the labeled
region and its labeling intensity were measured on serial sections
through the barrel cortex. Usually about 40 sections from each hemisphere were measured. The software allowed us to display on a computer screen the image of a stained section from which the autoradiogram was obtained, and to mark the barrel outlines, which were
superimposed on the autoradiogram so that the relations of the labeled regions to the morphological barrels could be accurately determined. The barrels of row B were measured on all sections (6-9)
on which they were visible. The measurements, were performed for
each barrel at its longest axis, so that for barrel B4 the measurement
wastaken close to its wall with B3. The labeled representation of
row B whiskers in layer IV was measured in exactly the same locations (Fig. 1A,B). The labeled representation was also identified and
measured on all sections through layers n/in and V, where the barrels
are not present. We determined the width of labeling of row B representation using the previously established criterion (Kossut et al.,
1988), which considered as activated the regions with a level of 2DG
labeling ([I4C] concentration) 15% higher than in the surrounding
cortex (Fig. \C). We have found that the 15% difference corresponds
to two standard deviations above the mean value of labeling in the
surrounding cortex (Chmielowska et al., 1986). Values obtained in
histological and autoradiographic measurements were averaged to
obtain the mean value of the barrel dimensions and the labeling di-
mensions for the entire row B. In order to correct for the possible
slight differences of the plane of sectioning, which could produce
differences of dimensions of row B barrels, the results were initially
expressed as a ratio of the width of the labeled area to the width of
row B barrels.
In the measurements of labeling intensity the outline of the labeled area was marked on the screen and the average gray level
reading was taken. The ratio of labeling on the experimental and
control side was calculated for each animal.
Results
Row B Representation 1 Day after CS + UCS Pairing
The effects of training upon row B representation were assessed by comparing the labeling evoked by stimulation of
the "trained" row B of whiskers and the control row B, unstimulated during the behavioral training, on the other side
of the snout. We previously ascertained (Chmielowska et al.,
1986; Kossut et al., 1988) that there were no ipsilateral
changes of 2DG uptake produced by unilateral whisker stimulation in the way applied during the 2DG mapping, and
therefore we could directly compare the labeling of "trained"
and untrained rows in the same animal. We also found previously that in intact untrained mice, the 2DG uptake evoked
by stimulation identical to that used in the present experiment applied to a row of vibrissae on both side of the snout
is very similar, the width of labeling varying by about 10%
(Siucinska and Kossut, 1994b; Jablonska et al., 1995).
Comparison of the extent of 2DG uptake evoked in the
barrel cortex by stimulation of the "trained" row B and control row B of whiskers on the other side of the muzzle (not
stroked during the classical conditioning) revealed that following training, the "trained" rows had a larger representation
in layer IV of all examined animals than the control row (Figs.
2-4). The 2DG labeled representation of row B in layer IV
was wider, on the average, by 45% (490 ± 4 1 u,m in control
rows and 710 ± 105 u.m in trained rows;/> < 0.01, MannWhitney U test). We measured the ratio of the width of 2DG
labeled zone to the width (long axis) of cortical barrel in
order to correct for variations in the plane of sections; however, in most cases the long axes of the corresponding barrels
of row B in both hemispheres were the same and no corrections were made. We measured the width of labeling rather
than the area because in some cases the sections from the
level of layer IV did not contain the entire row B, and parts
of either Bl or B4 barrel •were missing; the •width gave us a
more reliable measure. All quantitative data come from measurements of autoradiograms of tangentially cut sections; one
brain was sectioned paracoronally to visualize the effect in a
different plane (Fig. 5). No statistically significant changes in
labeling were found in cortical layers other than IV when
layers II/m values were averaged together (Fig. 6). If the data
were taken only from the deepest four sections of layer HI, a
statistically significant enlargement of the width of labeling
was found (25%, Mann-Whitney U test,p < 0.05); (Figs. 2, 4,
6). The remaining sections of layers II/III, although they did
not differ in the width of labeling, as determined by the applied criterion, showed an altered pattern of labeling. Around
the central dense core there was a radiating fringe of lower
intensity label (Fig. 4). On the control side the labeling terminated more abruptly. The radiating projections -were also
observed in the labeling of layer rv sections, but they were
usually of higher labeling density.
The labeling intensity of row B representation remained
unchanged after CS + UCS pairing, as compared to labeling
of control row B representation in the opposite hemisphere
(mean difference between intensity of labeling between rows
B in the two hemisphere amounted to 3% ± 17, Mann-Whitney U test,p > 0.05).
Cerebral Cortex May/Jun 1996, V 6 N 3 507
Flg.2
CONTROL
Figure 1. Method of autoradiogram analysis. A, Nissl-stained section at the level of layer IV and pseudocolored autoradiogram of the same section showing row B labeling in a
control mouse. Row B labeling marked with an arrow. B, Enlargement of fragments of A Lines across the labeled zone on the autoradiogram and across the barrels on the Nissl-
508 Visualization of Learning-Induced Plasticity • Siucinska and Kossut
In the groups of mice that received only CS or were
trained in a pseudoconditioning schedule, in which CS and
UCS were randomly applied, no changes in the dimensions or
labeling intensity of cortical representations of row B of vibrissae were found (Figs. 3,6).
Row B Representation after Longer Delays Post-CS
+UCS Pairing and after Extinction Procedure
The changes in row B cortical representation were tested after a waiting period of 3 and 5 d following training. During
the waiting period the animals were left in their cages. Three
days after training the row B representation in layer IV was
smaller than 1 d posttraining, but still significantly larger than
in controls (/> < 0.05); 5 d after training it returned to control
values (Fig. 6).
In a group where extinction procedure applied for 2 or 5
d after training, no increases of row B representation were
observed. Already 2 d of extinction completely obliterated the
results of training—the width of the trained row B representation was not significantly different from the control values
(Fig. 6). The intensity of 2DG uptake was the same in both
hemispheres of control mice and in all experimental groups;
it was not affected by any of these experimental procedures.
Behavioral Results
During the CS + UCS pairing we observed that the behavior
of mice changed visibly. By the end of the first session stroking of row B of vibrissae evoked a rigid posture, signifying an
expectation of an unpleasant stimulus. This behavior was not
observed in pseudoconditioned mice. In a group of four mice
where the heart rate was monitored during the CS + UCS
pairing, deceleration of the heart rate was observed during
the pairing sessions. Heart rate was measured for 30 sec before the first trial in each session when the mouse was in the
experimental setup and during the CS applications. The heart
rate values were averaged for blocks of 10 trials. The heart
rate declined significantly during the experimental session (/> <
0.02, ANOVA, mixed design followed by Duncan test). The
mean decline during the last block of trials in each session,
as compared to the presession rate, was 7% (p < 0.01). Heart
rate decelerations developed in all four animals, in two during
the first training session and in two during the second day of
training.
Discussion
CS + UCS Pairing
The main finding of this study is that under the influence of
a relatively short lasting (30 min), behaviorally salient stimulus, the responses of neurons of mouse barrel cortex can be
modified so that the cortical representation of the stimulated
mechanoreceptors is enlarged in thalamorecipient layers. This
change is transient, no longer observable 5 d after the end of
training.
In naive mice, 2DG labeling in the barrel cortex obtained
with this type of stimulation overlays the appropriate barrels
in layer IV of SI cortex and extends beyond the barrel bound-
aries, over the septa and parts of adjacent barrels (Chmielowska et al., 1986; Siucinska and Kossut, 1994b). This pattern
of labeling reflects an activation of barrel neurons by inputs
not only from their principal whiskers, but also from adjacent
vibrissae, presumably at least partly due to activity of intracortical excitatory connections (Armstrong-James et al., 1991).
In untrained animals and on the control sides of mice from
all the experimental groups, the ratio of the width of the
labeled representation of row B to the long axis of the corresponding row B barrel was about 1.4. In the trained group,
it was 2.03. The expansion of the row representation over
the adjacent rows of barrels increased on die average by 245
u.m on each side and overlapped about a half of adjacent row
A and C of barrels.
Modifications of responses of cortical neurones during the
process of learning have been investigated using several experimental models (forreview,see Weinberger, 1995). The results of these studies revealed changes in excitability and in
receptive field properties. Specific CS-linked alterations in receptive fields were found, such as increases of cortical representation of skin receptors used during conditioning, increases response to frequency of conditioned stimulus and of
cortical representation of the tone frequency used as CS. The
cause for increasing the cortical representation of sensory input paired with reinforcement may lay with die action of
neuromodulatory systems regulating arousal, attention, and
the stimulus-to-background ratio in cortical activity (Singer,
1987). It was shown that elimination of cholinergic input prevents plasticity in the adult somatosensory cortex (Juliano et
al., 1991) and impairs tactile learning (Jacobs and Juliano,
1995), while in die auditory cortex , microapplications of cholinergic agonists can modify receptive field characteristics of
neurons in a manner similar to those obtained during classical
conditioning (Cox et al., 1992). Coactivation of basal forebrain
and cutaneous receptors leads to a long-term enhancement
of somatosensory evoked potentials (Rasmusson and Dykes,
1988).
Reactivity of cortical neurons is also under very strong
control of inhibitory interneurons, as seen by die fact that die
GABAergic transmission is influential in shaping the properties of cortical receptive fields (Dykes et al., 1984; Sillitoe,
1975). The GABAergic system in adult sensory cortex is
downregulated by elimination of afferent input (Hendry and
Jones, 1988; Warren et al., 1989; Welker et al., 1989a; SkangielKramska et al., 1994). There is also evidence of increased GAD
immunoreactivity after prolonged sensory stimulation of vibrissae; however, this stimulation resulted in lowering of metabolism of the stimulated rows of barrels, presumably due to
habituation (Welker et al., 1989b, 1992). The investigation of
interactions of inhibitory and excitatory systems in our conditioning paradigm revealed no changes in die GABAA receptor binding in die trained row of barrels, while the NMDA
and AMPA receptor binding was transiently elevated in barrels
of the trained row (Skangiel-Kramska et al., 1993).
The stimulus used during the 2DG mapping was different
from that applied during the classical conditioning phase of
the experiment. This experimental design allowed us to ob-
stained section show how the measurements of the barrel long axis and the width of labeling were taken at the same site. C, The 2DG labeling density (in arbitrary units) profile
of the barrel field, from which we defined the cortical area with labeling density 15% higher than over the surrounding cortex Regions of highest optical density pseudocolored red
(lowest) to blue. Scale bars: A 1 mm; B and C, 500 n.m.
Figure 2. Examples of pseudocolored autoradiograms of 2DG labeling cortical representation of "trained" (CS + USC) and control row B in different cortical layers of one mouse.
Row B labeling marked with arrows. Scale bar, 500 [x.m.
Figure 3. Examples of pseudocolored autoradiograms and Nissl-stained sections of layer IV of control and experimental hemispheres of mice belonging to A [conditioning (CS +
UCS) group] and B (pseudoconditioning). Row B labeling marked with arrows. Au, auditory cortex; Sll, second somatosensory area. Scale bar, 1 mm. Scale bar, 500 \m.
Cerebral Cortex Mayflun 1996, V 6 N 3 509
Figure 4. Representative example of labeling pattern of row B representation in
layer Illb and IV binary thresholded image, showing in uniform black-only areas
activated 15% higher than background
surrounding cortex Arrowheads point to
row B representation. Scale bar, 500 ujn.
cs-ucs
control side
Figure 5. Labeling of row B representation in a mouse from CS + UCS group, computer-reconstructed autoradiograms of paracoronal sections cut in a plane orthogonal to rows
of barrels according to Chmielowska et aL (1389). A Experimental hemisphere. B, Control hemisphere. Arrowheads point to cross section of row B representation. Scale bar, 500 JJJTI.
510 Visualization of Learning-Induced Plasticity • Siucinska and Kossut
layer II / III
1000 r
I
*
:oo
l d.y
d.iay
j dayi
d.iay
1000 r
E
I control side
s daya
a.,.,
2 myi
.nine.,
4 a ,y,
layer IV
serve changes of responsiveness of neurons in the involved
cortical area to a standard stimulus. We therefore described
changes in sensitivity to a stimulus of characteristics different
than ones used during the training, but applied to the same
receptive surface. This is similar to the phenomenon of reorganization of hand representation after operant conditioning described by Recanzone et al. (1992a), where changes in
properties of cortical neurons were induced by training with
high-frequency flutter stimulus, but the spatial maps of the
cortex were obtained with fine-pressure probes. The experiments of Recanzone et al. (1992b) demonstrated also that discrimination training brings about changes in temporal properties of neurons in the part of SI cortex where the trained
skin area is represented. Moreover, some of this changes in
area 3b can be correlated with perceptual abilities of the
trained animals. The very simple classical conditioning paradigm used in the present study, cannot be compared with
discrimination training used by Merzenich's group or the
roughness discrimination by vibrissae employed by Guic et
al. (1993), which produced large changes of cortical maps.
However, the restricted changes of topographical maps observed in our study also represent remodeling of the cortex
by the process of learning.
CSOnly
Stroking of row B vibrissae during the three experimental
sessions, without application of the noxious stimulus, did not
alter the size of row B representation or the intensity of its
metabolic labeling. No changes 'were also observed after the
application of CS alone by Hand and Hand (1995), after 3 d
of training. However, they found a decrease of labeling intensity after a longer stimulation time (10 d). Similarly, a decrease
of 2DG uptake was recorded by Welker et al. (1992) after 4
d of continuous stimulation of vibrissae. These results were
attributed to habituation. Apparently in our case the total of
30 min of stimulation was insufficient to induce visible
changes in labeling dimension or intensity.
S00 •
200
1 day
dalay
3 daya
dalay
1000
S daya
dalay
2 daya .
•ninaioa
4 daya
IAHCUM
layer V
S00 '
m
200
1 day
d*Uy
3 daya
d*Uy
5 day*
d«*ay
2 daya
axttndio*
4 daya
Hgure 6. Width of row B representation in all experimental groups, in cortical layers
ll/lll, IV, and V. The values represent the mean width of labeling of row B representation
± standard deviation. *p < 0.05, **p < 0.01, Mann-Whitney U test n = 7 in CS +
UCS group, n = 4 in all other groups.
Laminar Localization of the Body Map Change
In view of evidence suggesting the importance of intracortical connections, especially via supragranular layers, in plastic
changes of barrel cortex of adult rats (Fox, 1992; ArmstrongJames et al., 1994; Diamond et al., 1994; Glazewski and Fox,
1994), it is puzzling that this brief training regime influenced
mostly the vibrissal representation in layer IV. We cannot attribute our result to peculiarities of the 2DG method, for example, preferential labeling of layer IV by 2DG, because in
our previous experiments large increases in the cortical representation of whiskers spared from denervation or sensory
deprivation were observed not only in layer IV but also in
supragranular and especially infragranular layer layers (Kossut
et al., 1988; Siucinska and Kossut, 1994a). Similar localization
of changes in cortical representation after classical conditioning using a single vibrissa was obtained by Hand and Hand
(1995), who employed a long lasting (60 d) appetitive training procedure. This effect is therefore not due to the changes
being observed "in statu nascendi." Indeed, the recent papers
describing plastic changes in neuronal responsiveness after
application of a whisker-pairing paradigm in rats, when all
vibrissae except two adjacent ones are clipped close to the
skin, report that altered receptive fields can be observed in
neurons of nongranular layers already after 24 hr (Diamond
et al., 1994) or that long latency responses of the barrel neurons, of intracortical origin, are the first to show receptive
field changes (Armstrong-James et al., 1994). These changes
later decrease and are replaced by short latency (presumably
thalamocortical) alterations in receptive fields (ArmstrongJames et al., 1994). It is therefore likely that in our experi-
Cerebral Cortex Mayflun 1996, V 6 N 3 511
mental animals the plastic changes have already proceeded
far enough to be localized in primarily in layer IV It is also
possible that the classical conditioning paradigm that we used
may trigger different circuits subserving plastic changes than
deafferentation or pairing-induced plasticity of representational maps, circuits where the thalamocortical component is
dominant. For example, the efficiency of thalamocortical synapses was found to be altered in the motor cortex during
associative conditioning (Meftah and Rispal-Padel, 1994). Alternatively, this specific form of CS or UCS stimulation may
result in different competitive or modulatory input consequences in the barrel field. Interestingly, training induced
changes in cortical representation of fingers in monkeys were
observed by Recanzone et al. (1992a) at cortical depth of
700-900 (j,m, which include layer IV.
Pseudoconditioning
The close temporal spacing of trials during conditioning sessions made it possible that sensitization and not conditioning
has occurred. The controls where pseudoconditioning was
applied served to clarify this situation. However, no significant
changes in cortical representation of row B were found after
the pseudoconditioning, indicating that sensitization was not
the cause of the observed body map change.
Duration of Training
We did not map the cortical representation after conditioning
training shorter than three sessions. Changes in unit response
have been observed by Edeline et al. (1993) after a very short
conditioning training in auditory cortex (five trials, 30 sec of
CS). Very fast changes were also found, with 2DG autoradiography, in gerbils auditory cortex following aversive conditioning (Scheich et al., 1993). The changes produced by nonassociative learning, habituation, in cortical maps of vibrissae
seem to require more time—as previously mentioned by
Hand and Hand (1995). They observed cortical maps changes
after 10 sessions but not after 3 sessions of unpaired passive
whisker stroking. We are presently trying to determine the
minimum duration of associative pairing necessary to produce a change of a cortical vibrissal map.
Duration of Body Map Change
The enlargement of the representation of "trained" row B in
this study was not long lasting. Five days after the training no
changes in the labeling dimensions were observed. Three
days after the training the enlargement was still present in
layer IV (although it was not as pronounced as 1 d after the
training) but disappeared from layer Illb. This spontaneous
disappearance was accelerated by application of extinction
procedure, believed to disrupt the effects of conditioning. In
this case, following 3 d of training, a daily session of CS only
was given. After 2 d of this treatment the enlargement of the
row B representation was no longer seen. Again, the presence
of this effect strengthens the notion that the observed change
is related to the learning process.
The spontaneous disappearance of the observed change is
in variance with the results of Weinberger et al. (1993) who
found that in the auditory cortex of guinea pigs changes in
receptive field properties induced by classical conditioning
persist for as long as 8 weeks. The differences in the animal
species, sensory modality, and finally a much longer duration
of the UCS (2 sec vs 0.5 sec) can account for these differences.
The experiments described above were not designed to
map a memory trace in the brain, but to induce an experience-dependent change in a cortical body map. The training
procedure was to provide an intensive stimulation of the vibrissal pathway coupled with aversive reinforcement. The ob-
512 Visualization of Learning-Induced Plasticity • Siucinska and Kossut
served enlargement of the row B cortical map represents a
plastic change in the cortical representation induced by associative training; therefore, a learning-dependent enhancement of the power of a sensory pathway. This change may,
however, constitute some part of the memory trace, since it
disappears when extinction procedure is employed. In any
event, this model should provide a convenient procedure for
conducting pharmacological and genetic manipulation studies of adult cortical plasticity mechanisms involved in changes
induced by both conditioning and extinction.
Notes
We arc indebted to Dr. M. Merzenich and Dr. S. Juliano for comment
on the manuscript. We thank Dr. W. Danysz for help with setting up
the training and Prof. K. Zielinski and Dr. J. Wilson for advice on
behavioral procedures. This research was supported in part by Me
Donnell Foundation Grant No. K 922335.
Address correspondence to Malgorzata Kossut, Department of
Neurophysiology, Nencki Institute, 3 Pasteur Street, 02-093 Warsaw,
Poland.
References
Allard T, Clark SA, Jenkins WM, Merzenich MM (1991) Reorganization
of somatosensory area 3b representations in adult owl monkeys
after digital syndactyly.J Neurophysiol 66:1048-1058.
Armstrong-James M, Callahan CA, Friedman MA (1991) Thalamo-cortical processing of the vibrissal information in a rat. I. Intracortical
origins of the surround but not centre receptive fields of layer IV
neurones in rat SI barrel field.J Comp Neurol 303:193-210.
Armstrong-James M, Diamond ME, Ebner FF (1994) An innocuous
bias in whisker use in adult rats modifies receptive fields of barrel
cortex neurons. J Neurosci 14:6978-6991Bakin JS, Weinberger NM (1990) Classical conditioning induces CSspecific receptive field plasticity in the auditory cortex of the
guinea pig. Brain Res 536:271-286.
Brons JF, Woody CHD (1980) Long-term changes in excitability of
cortical neurons after pavlovian conditioning and extinction. J
Neurophysiol 44:605-615.
Chmielowska J, Kossut M, Chmielowski M (1986) Single vibrissal
column in a mouse labeled with 2-deoxyglucose. Exp Brain Res
63:607-619.
Chmielowska J, Carvell GE, Simons DJ (1989) Spatial organization of
thalamocortical and corticothalamic projection system in rat SmI
barrel cortex.J Comp Neurol 285:325-338.
Cox CHL, Metherate R, Weinberger NM, Ashe JH (1992) Synaptic
potentials and effects of amino acid antagonists in the auditory
cortex. Brain Res Bull 28:401-410.
Delacour J, Houcine O, Talbi B (1987) "Learned" changes in the responses of the rat barrel field neurons. Neuroscience 23:63-71.
Diamond DM, Weinberger NM (1986) Classical conditioning rapidly
induces specific changes in frequency receptive fields of single
neurons in secondary and ventral ectosylvian auditory cortical
fields. Brain Res 372:357-360.
Diamond EM, Armstrong-James M, Ebner FF (1993) Experience-dependent plasticity in adult rat barrel cortex. Proc Natl Acad Sci
USA 90:2082-2086.
Diamond ME, Huang W, Ebner FF (1994) Laminar comparison of
somatosensory cortical plasticity. Science 265:1885-1888.
Dudai Y (1989) The neurobiology of memory: concepts, findings,
trends. Oxford: Oxford UP
Dykes RW, Landry R, Metherate R, Hicks TP (1984) Functional role
of GABA in cat primary somatosensory cortex: shaping receptive
fields of cortical neurons. J Neurophysiol 52:1066-1093.
Edeline JM, Pham P, Weinberger NM (1993) Rapid development of
learning-induced receptive field plasticity in the auditory cortex.
Behav Neurosci 107:539-557.
Fox K (1992) A critical period for experience-dependent synaptic
plasticity in rat barrel cortex.J Neurosci 12:1826-1838.
Glazewski S, Fox K (1994) The time-course of plastic changes in the
barrel cortex of adolescent rats. Soc Neurosci Abstr 20:694.11
Gonzalez-Lima F, Agudo J (1990) Functional reorganization of neural
auditory maps by differential learning. Neuroreport 1:161-164.
Gonzalez-Lima F, Scheich H (1986) Neural substrates for tone-con-
ditioned bradycardia demonstration with 2-deoxyglucose n. Auditory cortex plasticity. Behav Brain Res 20:281-293.
Guic E, Rodriguez E, Caviedes P, Merzenich MM (1993) Use-dependent reorganization of barrel field in adult rats. Soc Neurosci Abstr
19:696
Hand C, Hand PJ (1995) Discrete changes in brain metabolism resulting from changes in stimulus delivery context. J Neurophysiol,
in press.
Hand PJ (1982) Plasticity of the rat cortical barrel system. In: Changing concepts of the nervous system (Strick P, Morrison AD, eds),
pp 49-70. New York: Academic.
Hendry SHC, Jones EG (1988) Activity-dependent regulation of
GABA expression in the visual cortex of adult monkeys. Neuron
1:701-712.
Jablonska B, Gierdalski M, Siucinska E, Skangiel-Kramska J, Kossut M
(1995) Partial blocking of NMDA receptors restricts plastic
changes in adult mouse barrel cortex. Behav Brain Res 66:207216.
Jacobs SE, Juliano SL (1995) The impact of basal forebrain lesions
on the ability of rats to perform a sensory discrimination task
involving barrel cortex. J Neurosci 15:1099-1109.
Jenkins WM, Merzenich MM, Ochs MT, AUard T, Guic-Robles E (1990)
Functional reorganization of primary somatosensory cortex in
adult owl monkeys after behaviorally controlled tactile stimulation. J Neurophysiol 63:82-105.
Juliano SL, Ma W, Eslin D (1991) Cholinergic depletion prevents expansion of topographic maps in somatosensory cortex. Proc Natl
Acad Sci USA 88:780-784.
Kaas JH (1991) Plasticity of sensory and motor maps in adult mammals. Annu Rev Neurosci 14:137-167.
Kossut M (1992) Plasticity of the barrel cortex neurons. Prog Neurobiol 39:389-422.
Kossut M, Hand PJ, Greenberg J, Hand C (1988) Single vibrissal cortical column in SI cortex of rat and its alterations in neonatal and
adult vibrissae deafferented animals: a quantitative 2 DG study. J
Neurophysiol 60:829-852.
Levin BE, Dunn-Meynell A (1991) Adult rat barrel cortex plasticity
occurs at 1 week but not at 1 d after vibrissectomy as demonstrated by the 2-deoxyglucose method. Exp Neurol 113:237-248.
Meftah EM, Rispal-Padel L (1994) Synaptic plasticity in the thalamocortical pathway as one of the neurobiological correlates of forelimb flexion conditioning: electrophysiological investigation in the
cat.J Neurophysiol 72:2631-2647.
Merzenich MM, Kaas JH, Wall JT, Sur M, Nelson RJ, FeUeman D (1983)
Topographic reorganization of somatosensory areas 3b and 1 in
adult monkeys following restricted deafferentation. Neuroscience
8:33-55.
Merzenich MM, Recanzone GH, Jenkins WM, Grajski KA (1990)
Adaptive mechanisms in cortical networks underlying cortical
contributions to learning and nondeclarative memory. Cold Spring
Harbor Symp Quant Biol 55:873-887.
Rasmusson DD, Dykes RW (1988) Long-term enhancement of
evoked potentials in cat somatosensory cortex produced by coactivation of the basal forebrain and cutaneous receptors. Exp
Brain Res 70:276-286.
Recanzone GH, Merzenich MM, Jenkins WM, Grajski K (1992a) Topographic reorganization of the hand representation in cortical
area 3b of owl monkeys trained in a frequency-discrimination
task.J Neurophysiol 67:1031-1056.
Recanzone GH, Merzenich MM, Schreiner CH (1992b) Changes in
the distributed temporal response properties of SI cortical neurons reflect improvements in performance on a temporally based
tactile discrimination task.J Neurophysiol 67:1071-1091.
Recanzone GH, Schreiner CE, Merzenich MM (1993) Plasticity in the
frequency representation of primary auditory cortex following
discrimination training in adult owl monkeys. J Neurosci 13:87103.
Scheich H, Simons C, Ohl F, Tillein J, Thomas H (1993) Functional
organization and learning-related plasticity in auditory cortex of
the Mongolian gerbil. Prog Brain Res 97:135-143.
SillitoeAM (1975) The contribution of inhibitory mechanisms to the
receptive field properties of neurones in the striate cortex of the
c a t j Physiol (Lond) 250:305-329.
Singer W (1987) Activity-dependent self-organization of synaptic
connections as a substrate of learning. In: The neural and molec-
ular bases of learning (Changeux LP, Komishi M, eds), pp 301335. New York: Wiley.
Siucinska E, Kossut M (1994a) Plasticity of mystacial fur representation in SI cortex of adult vibrissectomized rats—a 2DG study.
Neuroreport 5:1605-1608.
Siucinska E, Kossut M (1994b) Short term changes of cortical body
maps following partial vibrissectomy in adult mice. Acta Neurobiol Exp 54:345-352.
Skangiel-Kramska J, Jablonska B, Kossut M (1993) Increase of NMDA
and AMPA receptor sites in cortical representation of whiskers
after peripheral tactile stimulation. Soc Neurosci Abstr 19:69.
Skangiel-Kramska J, Glazewski S, Jablonska B, Siucinska E, Kossut M
(1994) Reduction of GABA-A receptor binding of [3H]muscimol
in the barrel field of mice after peripheral denervation: transient
and long-lasting effects. Exp Brain Res 100:39-46.
Sokoloff L, Reivich M, Kennedy C, Des Rosiers MH, Patlak CS, Pettigrew, KD, Sakurada O, Shimohara M (1977) The (1<Q-deoxyglucose method for measurement of local cerebral glucose utilization: theory, procedure and normal values in conscious and anesthetized albino rat.J Neurochem 28:897-916.
Wall PD, Egger MD (1971) Formation of new connections in adult
rat brains after partial deafferentation. Nature 232:542-545.
Warren R, Tremblay N, Dykes RW (1989) Quantitative study of glutamic acid decarboxylase-immunoreactive neurons and cytochrome oxidase activity in normal and partially deafferented rat
hindlimb somatosensory cortex. J Comp Neurol 288:583-592.
Weinberger NM (1995) Dynamic regulation of receptive fields and
maps in the adult sensory cortex. Annu Rev Neurosci 129-158.
Weinberger NM, Javid R, Lepan B (1993) Long-term retention of
learning-induced receptive-field plasticity in the auditory cortex.
Proc Natl Acad Sci USA 90:2394-2398.
Welker E, Soriano E, Van der Loos H (1989a) Plasticity in barrel cortex of adult mouse: effects of peripheral deprivation on GADimmunoreactivity. Exp Brain Res 74:441-452.
Welker E, Soriano E, Dorf J, Van der Loos H (1989b) Plasticity in the
barrel cortex of adult mouse: transient increase of GAD-immunoreactivity following sensory stimulation. Exp Brain Res 78:659664.
Welker E, Rao SB, Dorfl J, Melzer P, Van der Loos H (1992) Plasticity
in the barrel cortex of the adult mouse: effects of chronic stimulation upon deoxyglucose uptake in the behaving animal. J Neurosci 12:153-170.
Woolsey TA, Van der Loos H (1970) The structural organization of
layer IV in the somatosensory region (SI) of mouse cerebral cortex. Brain Res 17:205-242.
Cerebral Cortex May/Gun 1996, V 6 N 3 513