Download Enhanced intrinsic excitability and EPSP

Articles in PresS. J Neurophysiol (December 7, 2011). doi:10.1152/jn.01009.2011
1
1
2
Enhanced intrinsic excitability and EPSP-spike coupling accompany
3
enriched environment induced facilitation of LTP in hippocampal CA1
4
pyramidal neurons
5
6
7
Ruchi Malik and Sumantra Chattarji
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
National Centre for Biological Sciences, Tata Institute of Fundamental Research,
Bangalore, India
Corresponding Author
Sumantra Chattarji, Ph.D.
National Centre for Biological Sciences (NCBS)
Tata Institute of Fundamental Research
GKVK, Bellary Road,
Bangalore 560065, India.
E-mail: [email protected],
Phone: +91 80 23666120; Fax: +91 80 23636662
Running head: Environmental enrichment and hippocampal plasticity
Keywords: Contextual fear learning; action potential threshold; theta-burst stimulation;
miniature excitatory postsynaptic currents, after-hyperpolarization
Acknowledgements
The authors thank Rishikesh Narayanan for helpful discussions during preparation of the
manuscript and Shobha Anilkumar for technical assistance with Golgi-Cox staining.
Grants: The research was funded by intramural research funds from NCBS (Grant
No.1143 & 4143)
Author contributions: R.M. and S.C. designed the experiments; R.M. performed the
experiments and analyzed the data; R.M. and S.C. wrote the paper.
Copyright © 2011 by the American Physiological Society.
2
39
40
ABTRACT
41
Environmental enrichment (EE) is a well-established paradigm for studying naturally
42
occurring changes in synaptic efficacy in the hippocampus that underlie experience-
43
induced modulation of learning and memory in rodents. Earlier research on the effects of
44
EE on hippocampal plasticity focused on long-term potentiation (LTP). While many of
45
these studies investigated changes in synaptic weight, little is known about potential
46
contributions of neuronal excitability to EE-induced plasticity. Here, using whole-cell
47
recordings in hippocampal slices, we address this gap by analyzing the impact of EE on
48
both synaptic plasticity and intrinsic excitability of hippocampal CA1 pyramidal neurons.
49
Consistent with earlier reports, EE increased contextual fear memory and dendritic spine
50
density on CA1 cells. Further, EE facilitated LTP at Schaffer collateral inputs to CA1
51
pyramidal neurons. Analysis of the underlying causes for enhanced LTP shows EE to
52
increase the frequency, but not amplitude, of miniature excitatory postsynaptic currents.
53
However, presynaptic release probability, assayed using paired-pulse ratios and use-
54
dependent block of NMDA-receptor currents, was not affected. Further, CA1 neurons
55
fired more action potentials in response to somatic depolarization, as well as during the
56
induction of LTP. EE also reduced spiking threshold and after-hyperpolarization
57
amplitude. Strikingly, this EE induced increase in excitability caused the same sized
58
EPSP to fire more action potentials. Together these findings suggest that EE may enhance
59
the capacity for plasticity in CA1 neurons not only by strengthening synapses, but also by
60
enhancing their efficacy to fire spikes - and the two combine to act as an effective
61
substrate for amplifying LTP.
62
3
63
INRODUCTION
64
The rodent hippocampus has served as a powerful model system for understanding the
65
physiological and molecular bases of long-term use-dependent changes in synaptic
66
strength and its relationship to certain forms of learning and memory (Lynch 2004;
67
Malenka and Bear 2004; Martinez and Derrick 1996). The most extensively studied
68
synaptic plasticity mechanism underlying memory formation in the hippocampus is long-
69
term potentiation (LTP), in which brief high-frequency activation of afferents induces a
70
persistent increase in synaptic strength (Bliss and Collingridge 1993; Bliss and Lømo
71
1973). Bottom-up strategies using gene-deletion techniques have greatly advanced
72
analyses of the links between LTP, its underlying biochemical signaling mechanisms, and
73
hippocampus-dependent memory. (Chen and Tonegawa 1997; Hédou and Mansuy 2003;
74
Huang et al. 1995; Tonegawa et al. 2003; Tsien et al. 1996). A complementary top-down
75
approach involving experience-induced modulation of learning and memory has also
76
contributed to our understanding of LTP and its underlying mechanisms (Hargreaves et
77
al. 1990; Huang et al. 2005; Kim and Diamond 2002; Weisz et al. 1984). Environmental
78
enrichment is one such rodent model of behavioral experience that has been utilized to
79
study naturally occurring, as opposed to artificially induced, changes in synaptic efficacy.
80
Commonly used paradigms of enriched environment (EE) include a combination of
81
inanimate and social stimuli that facilitate physical activity, social interactions and
82
exploratory behavior (Rosenzweig and Bennett 1996; van Praag et al. 2000).
83
Accumulating evidence indicates that exposure to EE triggers a range of morphological,
84
biochemical, and physiological changes that facilitate hippocampus-dependent learning
85
in enriched rats (Duffy et al. 2001; Nithianantharajah et al. 2008; Rosenzweig 1966;
4
86
Schrijver et al. 2004; van Praag et al. 2000; Woodcock and Richardson 2000). For
87
instance, EE enhances synaptic connectivity in hippocampal circuits by promoting the
88
growth of dendrites and spines (Faherty et al. 2003; Greenough and Volkmar 1973;
89
Leggio et al. 2005; Moser et al. 1994; Rampon et al. 2000). Exposure to EE also elicits
90
changes in biochemical signaling pathways that play a pivotal role in experimentally
91
induced forms of synaptic strengthening (Duffy et al. 2001; Ickes et al. 2000; Mohammed
92
et al. 2002; Paylor et al. 1992; Williams et al. 2001).
93
While there is broad agreement that EE enhances the molecular and structural
94
substrates of synaptic plasticity in a manner that is expected to facilitate hippocampal
95
LTP, the impact of EE varies between different sub-regions of the hippocampus. For
96
instance, in hippocampal area CA1, electrophysiological recordings from acute slices
97
have shown increased basal excitatory transmission following EE (Foster and Dumas
98
2001 ; Irvine and Abraham 2005), which has been interpreted as a manifestation of
99
enhanced synaptic transmission caused by a natural LTP-like phenomenon over the
100
course of EE. However, there are other studies that did not find any effect of EE on basal
101
synaptic transmission but did demonstrate enhancement of LTP in area CA1 when tested
102
after exposure to EE (Artola et al. 2006; Duffy et al. 2001). In contrast to the CA1 area,
103
in vitro and in vivo extracellular field potential recordings in the dentate gyrus (DG) have
104
shown prior exposure to EE to occlude the induction of LTP at perforant path inputs
105
(Eckert et al. 2010; Foster et al. 1996). Further, this occlusion of LTP was accompanied
106
by an increase in the basal synaptic transmission at perforant path synapses to DG
107
granule cells (Foster et al. 1996; Gagné et al. 1998; Green and Greenough 1986; Irvine et
108
al. 2006). A particularly striking finding comes from in vivo extracellular recordings in
5
109
freely moving rats wherein exposure to EE caused a significant increase in the population
110
spike amplitude of DG granule cells (Irvine et al. 2006). Similar increases in population
111
spike amplitudes have been observed in vitro even when EE failed to enhance the level of
112
LTP induced by tetanic stimulation in the DG (Green and Greenough 1986). Importantly,
113
it has been suggested that such a change may reflect a form of activity-induced
114
hippocampal plasticity called EPSP-Spike (E-S) potentiation (Irvine et al. 2006), which is
115
thought to be the result of a stronger coupling between the EPSP and spike, and is
116
mediated by increased neuronal excitability (Andersen et al. 1980; Daoudal and Debanne
117
2003; Daoudal et al. 2002). However, in earlier studies that relied on extracellular field
118
potential recordings, it was not possible to directly demonstrate EE induced modulation
119
of E-S coupling. This issue is also significant because previous research has focused
120
primarily on experience-induced changes in synaptic strength, thereby overlooking
121
potential changes in intrinsic excitability of neurons (Abraham 2008; Frick and Johnston
122
2005; Frick et al. 2004; Johnston and Narayanan 2008; Kim and Linden 2007; Poirazi
123
and Mel 2001; Sjostrom et al. 2008; Zhang and Linden 2003). While extracellular
124
recordings from freely moving rats are strongly suggestive of a role for intrinsic plasticity
125
in mediating E-S potentiation like effects in the DG, little is known about the impact of
126
EE on intrinsic properties of DG neurons. Even in area CA1 where clear evidence exists
127
for EE enhancing LTP, it is not clear if this is caused by changes in synaptic strength
128
alone or if intrinsic plasticity, such as strengthening of E-S coupling, also plays a role.
129
Therefore, in the current study we address some of these unresolved issues by using
130
whole-cell recordings in CA1 pyramidal cells to test if EE modulates both synaptic
6
131
strength and intrinsic excitability, and if the two can combine to act as an effective
132
substrate for amplifying LTP and its behavioral consequences.
133
7
134
MATERIALS AND METHODS
135
Animals used
136
Male Sprague-Dawley rats were used in this study. Animals were maintained in a
137
temperature-controlled room, with a 14 h light: 10 h dark cycle with ad libitum access to
138
food and water. All experimental protocols used in this study were approved by the
139
Institutional Animal Ethics Committee of the National Centre for Biological Sciences.
140
141
Housing conditions
142
At P25, rats were randomly assigned to either control housing or enriched environment
143
(EE) housing conditions. In control housing condition, 2–3 rats were housed together in
144
standard laboratory cages and were handled only for routine animal maintenance
145
procedures (CON, Fig. 1B). In enriched housing condition, 12–13 rats were group housed
146
in larger cages (25 × 20 × 12 in) (EE, Fig. 1C). Each day for 4 hours (during the light
147
phase of the rats), enriched rats were transferred to a playing arena (34 × 34 × 30 in)
148
which contained novel objects (tunnels, ladders, balls etc; Fig. 1D). The placement of
149
these objects was changed on a daily basis and new objects were added to the arena twice
150
every week. Following 30-35 days of exposure to EE, animals were used for
151
morphological, behavioral or electrophysiological analysis. From each batch of enriched
152
rats, 6–8 rats were used for electrophysiological recordings and the rest were used for
153
morphological and behavioral testing. The EE animals were not exposed to the playing
154
arena on the day of experiment and all recordings were carried out 1 day after the end of
155
EE. The experimenter was aware of the housing condition for electrophysiology
156
experiments. The morphological and behavioral analyses were done blind.
8
157
158
Contextual fear conditioning
159
Training was conducted in a Plexiglas rodent conditioning chamber with a metal grid
160
floor (Coulbourn Instruments, Lehigh Valley, PA) that was enclosed within a sound
161
attenuating chamber (Coulbourn Instruments). The chamber was dimly illuminated by a
162
single house light and a ventilation fan provided a background noise of 60–65 dB. The
163
floor and the walls of the chamber were wiped using 70% alcohol between each trial. On
164
the day of training each rat was placed in the conditioning chamber and given one
165
footshock (1 s, 1 mA) 16s later. Rats were removed from the chamber 120 s after the
166
shock. Twenty-four hours later, each rat was returned to the chamber and contextual fear
167
learning was quantified manually during a 2 min period (from the videotape). Freezing
168
was used as the index for contextual fear learning. Freezing involved absence of all body
169
movements, except respiration–related movements (Blanchard and Blanchard 1969;
170
Bouton and Bolles 1980; Fanselow 1980; Phillips and LeDoux 1994).
171
172
Golgi staining and morphological analysis
173
Animals were anesthetized using halothane and decapitated. The brain was removed
174
quickly, and blocks of tissue containing the hippocampus were dissected and processed
175
for the Golgi-Cox technique at room temperature. The brains were processed and coronal
176
sections were obtained as described before (Pawlak et al. 2005).
177
By using the NeuroLucida image analysis system (MicroBrightField, Williston,
178
VT, USA) attached to an Olympus BX61 microscope (100×, 1.3 N.A., Olympus BX61;
179
Olympus, Shinjuku-Ku, Tokyo, Japan), all protrusions, irrespective of their
9
180
morphological characteristics, were counted as spines if they were in direct continuity
181
with the dendritic shaft. The primary branch to the apical dendrite (main shaft) of the area
182
CA1 neurons was selected for spine analysis. Primary branches from both short-shaft and
183
long-shaft neurons were included in the analysis. The branches selected for analysis
184
originated 50–150 μm away from the soma. Starting from the origin of the branch, and
185
continuing away from the cell soma, the number of spines was counted in successive
186
steps of 10 μm each, for a total of 8 steps (i.e. extending a total length of 80 μm).
187
188
Hippocampal slice preparation
189
Rats were anaesthetized using halothane and decapitated. The brain was quickly removed
190
from the skull and transferred to oxygenated, ice-cold ACSF containing (in mM): 115
191
NaCl, 25 Glucose, 25.5 NaHCO3, 1.05 NaH2PO4, 3.3 KCl, 2 CaCl2 and 1 MgSO4. The
192
hippocampi were dissected out of the two hemispheres and transverse sections (400 µm)
193
were obtained using Vibratome 1000 Plus (Vibratome, St. Louis, MO, USA). In
194
experiments where the slices were disinhibited during the recordings, area CA3 was
195
surgically removed to avoid spontaneous activity. Slices were transferred to a holding
196
chamber and were allowed to recover for 1 hour at room temperature.
197
Individual slices were transferred to a submerged recording chamber (28 ± 2°C)
198
and visualized using infrared differential interference contrast optics and Dage camera
199
system attached to an Olympus BX51-WI microscope. Cells were selected for recording
200
based on their pyramidal shape, smoothness of the membrane and low contrast
201
appearance.
202
10
203
Whole-cell recordings
204
Patch pipettes (3–5 MΩ, ~ 2 µm tip diameter) were pulled from thick-walled borosilicate
205
glass on a P-97 Flaming-Brown Micropipette Puller (Sutter Instruments, USA). In
206
experiments where evoked responses were recorded, a bipolar electrode (25 µm dia.
207
Platinum/Iridium, FHC, ME) or a silver coated glass electrode filled with extracellular
208
ACSF (~ 2 µm tip dia.) connected to an Iso-Flex stimulus isolator (A.M.P. Instruments
209
Ltd., Jerusalem, Israel) was used to stimulate the Schaffer collateral inputs. Data were
210
recorded using HEKA EPC 10 Plus amplifier (HEKA Electronik, Germany), filtered at
211
2.9 kHz and digitized at 20 kHz. Stimulus delivery and data acquisition were performed
212
using the Patchmaster software (HEKA Electronik, Germany). Cells were used for
213
recording if initial resting membrane potential (Vm) ≤ -60 mV and series resistance (Rs)
214
was 15–25 MΩ. During the course of the experiments, neuron's input resistance (Rin) and
215
series resistance were continuously monitored by applying hyperpolarizing current or
216
voltage pulse and experiments were rejected if Rs or Rin changed by more than 20% of
217
their respective initial values. All analysis of electrophysiological data was performed
218
using custom-written programs in IGOR PRO software (Wavemetrics, Lake Oswego,
219
OR, USA), unless otherwise stated.
220
For current clamp recordings, the patch pipettes were filled with internal solution
221
containing (in mM): 120 K-gluconate, 20 KCl, 10 HEPES, 2 NaCl, 4 MgATP, 0.3
222
NaGTP, 10 phosphocreatine (pH 7.3, KOH, ~290 mOsm). For the voltage clamp internal
223
solution, potassium was replaced with equimolar cesium. For all recordings, neurons
224
were held at -70 mV.
225
11
226
mEPSC recordings
227
CA1 neurons were voltage clamped at -70 mV and 2-amino-3-(5-methyl-3-oxo-1, 2-
228
oxazol-4-yl) propanoic acid receptor (AMPAR) mediated miniature excitatory
229
postsynaptic currents (mEPSCs) were isolated by tetrodotoxin (TTX; 0.5 μM) and
230
picrotoxin (100 μM). Continuous current traces of 5 minutes duration (recorded at least 5
231
min after achieving whole-cell configuration) were analyzed using the Mini Analysis
232
program (Synaptosoft Inc.).
233
234
Paired-pulse measurements
235
Paired stimuli with inter-stimulus interval (ISI) of 50, 175, 100, 150, 200 and 250 ms
236
were delivered every 20 s (10 sweeps each) to the Schaffer collateral inputs while
237
clamping the cell at -70 mV in the presence of picrotoxin (100 μM). Paired Pulse Ratio
238
(PPR) was defined as the ratio of peak amplitude of the second EPSC (measured for 1
239
ms) to the first.
240
241
MK-801 experiments
242
CA1 neurons were voltage clamped at +40 mV in the presence of -7-nitroquinoxaline-
243
2,3-dione (CNQX; 10 μM) and picrotoxin (100 μM), and N-Methyl-D-aspartic acid
244
receptor (NMDAR) mediated EPSCs were recorded by stimulating the Schaffer collateral
245
inputs every 20 sec. After acquiring a baseline of 5 min, MK-801 (5 μM) was added to
246
the bath, and synaptic stimulation was stopped for 10 minutes to ensure equilibration of
247
the drug concentration. At the end of this incubation, synaptic stimulation was resumed
248
and 100 trials were recorded for every cell. The peak amplitudes (measured for 10 ms) of
12
249
NMDAR-EPSCs were measured and normalized to the amplitude of first trace in MK-
250
801. The decay in NMDAR-EPSC amplitudes in the presence of MK-801 was fit to a
251
single exponential (Manabe and Nicoll 1994; Murthy et al. 1997). Time constants (taus)
252
obtained from the exponential fits were used for statistical comparisons.
253
254
Long-term potentiation
255
Recordings were obtained from CA1 neurons in current clamp mode in the presence of
256
picrotoxin (100 μM) and the cells were held at -70 mV (± 2mV) by injecting
257
hyperpolarizing current. The LTP protocol involved acquisition of 5 min stable baseline
258
Excitatory Postsynaptic Potentials (EPSPs; 0.05 Hz) followed by application of theta
259
burst stimulation (TBS). The TBS consisted of 5 bursts (at 5 Hz) of 4 pulses (at 100 Hz)
260
each (Fig 5A). Post-TBS, EPSPs were recorded for 30 min at the baseline stimulation
261
frequency (0.05 Hz). LTP experiments were excluded from the analysis, if the TBS
262
application was not within 10–12 min after achieving whole-cell configuration. The
263
amplitudes of EPSPs during baseline acquisition were kept within 5–10 mV for all cells.
264
LTP was quantified using averaged and normalized initial slope values, defined as the
265
rise in amplitude for the first 1−2 ms of EPSPs. For statistical comparisons, EPSP slope
266
values at 25–30 min after TBS induction was compared to the 5 min average baseline.
267
To analyze burst–induced depolarization, the traces recorded during TBS were
268
filtered at 100 Hz and the action potentials were subtracted from the waves (Fig. 5B,C).
269
The peak amplitudes for every burst of the resultant waveform were used for comparison.
270
271
Passive and active membrane properties
13
272
Current clamp recordings were obtained in the presence of picrotoxin (100 μM), amino-
273
phosphonopentanoic acid (APV; 30 μM) and cyano-7-nitroquinoxaline-2,3-dione
274
(CNQX; 10 μM). Rs was compensated to accurately measure the action-potential
275
properties. The current-voltage relations were obtained by plotting the steady state
276
voltage responses to 600 ms, 10 pA current steps (-50 pA to +50 pA). Input resistance
277
was calculated from the slope of the linear fit of the voltage–current plot (Staff et al.
278
2000). The membrane time constant was calculated from average of the mono
279
exponential fits of 200 ms current steps (-20, -10, 10 and 20 pA). The sag voltage was
280
calculated by subtracting the steady state voltage from the peak voltage responses to 600
281
ms hyperpolarizing current injections (-600 to -300 pA) (Moyer Jr et al. 1996). The
282
resonance frequency was measured from the cell’s response to a sinusoidal current of
283
constant amplitude with its frequency linearly spanning 0–20 Hz in 20 s (Narayanan and
284
Johnston 2007). The peak after-hyperpolarization potential (AHP) was analyzed using a
285
100 ms, depolarizing pulse that reliably elicited a train of 4-5 action potentials (Moyer Jr
286
et al. 1996). The after-depolarization potential (ADP) was measured from the peak
287
amplitude during the action potential repolarization phase. Action potential (AP)
288
properties were measured from single APs elicited by 5 ms, 1 nA current injection. The
289
AP threshold was measured as the voltage at which the 1st derivative of the voltage
290
response (dV/dt) reached 40 mV/ms. Rheobase current was determined as the minimal
291
depolarizing current amplitude (3 ms) required to elicit an AP. AP amplitude was
292
measured from the resting membrane potential to the AP peak, and the duration was
293
measured at the half-amplitude.
14
294
To study the membrane excitability, neurons were injected with 600 ms
295
depolarizing current pulses ranging from 50 to 500 pA. The number of APs elicited by
296
each current injection was counted for individual traces and plotted as a function of
297
injected current amplitude.
298
299
EPSP-spike coupling
300
EPSP-to-Spike (E-S) relationships were measured for a cell by stimulating the Schaffer
301
collateral inputs at 0.05 Hz, while recording the slope of the resulting EPSP in the
302
presence of picrotoxin (100 μM). The E-S data were plotted by binning the slope values
303
and finding a probability to spike for each slope bin. This E-S curve was fit with a
304
sigmoid function and the EPSP value at the 0.5 spike probability point (EC-50)
305
determined for each cell (Daoudal et al. 2002). The EC-50 values were used for statistical
306
comparison.
307
308
EPSP-amplification
309
The EPSP slope data were binned in 0.3 mV/ms bins (range on X-axis: 0.2 mV/ms to 2.3
310
mV/ms) and the average EPSP amplitude was calculated for individual CA1 cells. A
311
linear fit of the amplitude vs. slope plot was obtained for individual cells and the slope of
312
the linear fit for individual cells was used for statistical comparison.
313
314
Statistical analysis
315
All values are expressed as mean ± SEM. Statistical comparisons were done after using
316
Levene's test and single sample Kolmogorov-Smirnov (K-S) test for appropriate
15
317
assumptions of variance and normality of distribution. Comparisons between two groups
318
were done using unpaired Student's t test. Comparison for cumulative distributions was
319
done using two sample K-S test. All statistical analyses were conducted using SPSS 9.0
320
(SPSS Inc., Chicago, IL, USA) or IGOR PRO (Wavemetrics).
321
322
Chemicals
323
Most of the chemical and toxins were obtained from Sigma (St. Louis, MO, USA) unless
324
mentioned otherwise. TTX was obtained from Alomone Labs. MK-801 maleate was
325
obtained from Tocris Biosciences.
326
16
327
RESULTS
328
Enriched environment improves contextual fear learning and enhances spine
329
density on hippocampal CA1 neurons
330
To confirm the efficacy of our experimental protocol for enriched environment (EE) we
331
relied on behavioral and cellular measures that have been established by earlier studies.
332
At the behavioral level, we utilized previous findings on hippocampus-dependent
333
contextual fear learning being enhanced by EE (Duffy et al. 2001; Woodcock and
334
Richardson 2000). At the cellular level, we took note of earlier reports on EE leading to
335
an increase in dendritic spine-density in hippocampal area CA1 neurons (Faherty et al.
336
2003; Greenough and Volkmar 1973; Leggio et al. 2005; Moser et al. 1994; Rampon et
337
al. 2000).
338
Many lines of evidence support the involvement of hippocampus-dependent
339
spatial learning in contextual fear acquisition (Lee and Kesner 2004; Maren et al. 1998).
340
Therefore, we compared contextual fear conditioning between rats subjected to EE and
341
rats that were housed for the same period of time in control cages (Materials & Methods).
342
Memory for contextual fear conditioning, measured in the same context 24 h after
343
training, was significantly enhanced (95 % increase, p<0.01) in enriched rats (% freezing,
344
CON: 31.3 ± 6.2, N=13 rats; EE: 61.2 ± 6.3, N=14 rats; Fig. 2A).
345
Having demonstrated the efficacy of our EE paradigm in improving a
346
hippocampus-dependent form of learning, we next focused on a widely used synaptic
347
correlate of structural plasticity that is also known to be enhanced in the hippocampus of
348
enriched rats – the number of dendritic spines on CA1 pyramidal neurons. We quantified
349
spine-density on primary branches of apical dendrites from Golgi-impregnated CA1
17
350
pyramidal neurons (Fig. 2B). Compared to control neurons, there was a significant
351
increase (31% increase, p<0.01) in the number of spines per 10 µm, measured along an
352
80-µm dendritic segment in EE neurons (CON: 11.1 ± 0.6, n=14 cells, N=4 rats; EE: 13.5
353
± 0.4, n=16 cells, N=4 rats; Fig. 2C). A more detailed segmental analysis, in steps of 10
354
µm, showed significantly higher spine density in all 10-µm segments along the entire 80-
355
µm length of the apical dendrite in EE neurons (Fig. 2D). Thus, consistent with previous
356
reports, the EE paradigm used in the present study also caused structural plasticity in the
357
excitatory neurons in hippocampal area CA1.
358
359
Enriched environment increases the frequency, but not the amplitude, of mEPSCs
360
in CA1 pyramidal neurons
361
Does the higher spine-density after EE have a physiological correlate that is manifested
362
as an increase in excitatory synaptic transmission? To test this we used whole-cell
363
voltage-clamp recordings to compare the frequency and amplitude of spontaneous
364
miniature excitatory postsynaptic currents (mEPSCs) in CA1 pyramidal cells from
365
control and enriched rats (Fig. 3A1, A2). These mEPSCs were completely blocked by
366
CNQX, confirming that they were AMPAR-dependent synaptic currents (data not
367
shown). EE neurons exhibited a significant increase (57%, p<0.01) in mEPSC frequency
368
(CON: 0.28 ± 0.02 Events/s, n=14; EE: 0.44 ± 0.03 Events/s, n=13; Fig. 3B1), whereas
369
mEPSC amplitude was not affected (CON, 22.4 ± 1.5 pA, n=14; EE: 21.4 ± 0.53 pA,
370
n=13; Fig. 3B2). The increase in mEPSC frequency was also reflected in a leftward shift
371
of the cumulative probability plot of inter-event interval for EE neurons relative to
372
controls (Fig. 3B1). Further, the decay-time of mEPSCs was significantly longer (17%
18
373
increase, p<0.01) in EE neurons (CON: 5.88 ± 0.21 ms, n=14; EE: 6.9 ± 0.33 ms, n=13;
374
Fig. 3B3).
375
376
Enriched environment has no effect on measures of presynaptic release probability
377
While the enhanced mEPSC frequency is consistent with an increase in the number of
378
spines after exposure to EE, the increase in mEPSC frequency may also be indicative of
379
enhanced presynaptic release probability (Turrigiano and Nelson 2004). Therefore, we
380
analyzed the potential impact of EE on presynaptic release probability using two different
381
assays (Murthy et al. 1997). First, we measured paired-pulse ratios (PPR) of EPSCs
382
across a range of inter-stimulus intervals at Schaffer collateral inputs using whole-cell
383
voltage-clamp recordings from CA1 neurons (Fig. 3C). We found no difference in PPR
384
of evoked EPSCs in EE cells compared to controls (PPR at 50 ms inter-pulse interval,
385
CON: 2.5 ± 0.2, n=12; EE: 2.5 ± 0.3, n=11; Fig. 3C). This lack of effect on PPR reflects
386
an absence of change in presynaptic release probability after EE. This was probed further
387
using a second assay that involved repeated stimulation of Schaffer collateral inputs to
388
CA1 cells in the presence of the NMDAR open-channel blocker MK-801. This led to a
389
progressive decay of NMDAR-EPSCs (Fig. 3D1-D2), the time constant of which is
390
inversely related to the probability of release (Manabe and Nicoll 1994; Rosenmund et al.
391
1993). The decay kinetics were fit by a single exponential, and the time-constants of
392
decay between the two groups were not found to be significantly different (CON: 59 ±
393
10.6 ms, n=10; EE: 63.6 ± 10.1 ms, n=10; Fig. 3D2, inset). Together these data showed
394
that EE had no effect on the release probability at Schaffer collateral inputs to CA1
395
neurons, suggesting that the increase in mEPSC frequency is likely to be an
19
396
electrophysiological correlate of the higher number of synapses on CA1 neurons in EE
397
rats (Prange and Murphy 1999; Turrigiano and Nelson 2004).
398
399
Enriched environment enhances LTP at Schaffer collateral inputs to CA1
400
pyramidal neurons
401
The morphological and electrophysiological results presented thus far, point to an overall
402
enhancement of excitatory synaptic transmission by EE. In addition, at the behavioral
403
level the same EE improves hippocampus-dependent memory. A large body of evidence
404
has identified a pivotal role for hippocampal synaptic plasticity mechanisms, such as
405
Long-Term Potentiation (LTP), in mediating learning and memory (Bliss and
406
Collingridge 1993; Malenka and Bear 2004). Indeed, several earlier studies have reported
407
EE-induced enhancement of LTP in the CA1 area of the hippocampus (Artola et al. 2006;
408
Duffy et al. 2001). A majority of these earlier studies, however, used extracellular field
409
potential recordings to examine the effects of EE on LTP. We, therefore, analyzed the
410
impact of EE at the level of single CA1 pyramidal neurons in hippocampal slices. To this
411
end, we utilized an LTP induction protocol, theta-burst stimulation (TBS), that resembles
412
the physiologically relevant in vivo firing patterns in the theta frequency range (4-8 Hz)
413
seen during memory acquisition and retrieval in rodents (Bland 1986; Larson and Lynch
414
1986; Larson et al. 1986; Nguyen and Kandel 1997). In hippocampal slices obtained from
415
control rats, TBS applied to Schaffer collateral inputs to CA1 pyramidal neurons led to
416
robust LTP (% increase in EPSP slope relative to pre-TBS baseline, CON: 101.9 ± 19.4,
417
n=11; Fig. 4A1). Strikingly, the same TBS protocol induced significantly greater LTP in
418
CA1 neurons from enriched rats (% increase in EPSP slope relative to pre-TBS baseline,
20
419
EE: 191.7 ± 43.1; n=11; 90% increase compared to LTP in control slices, p<0.05; Fig. 4).
420
Thus, EE facilitates the ability of excitatory glutamatergic synapses in area CA1 to
421
undergo LTP, and this in turn is consistent with the enhanced hippocampal memory.
422
423
Enriched environment enhances action potential firing during LTP induction
424
What may be the underlying mechanisms that lead to enhanced LTP in CA1 neurons after
425
EE? Physiological and morphological changes in the excitatory synapses of CA1 neurons
426
provide an ideal substrate for enhanced LTP, but may not be the only determinants of the
427
magnitude of LTP induced. The level of postsynaptic depolarization reached during the
428
delivery of LTP-inducing stimuli is known to play an important role in its efficacy to
429
elicit potentiation (Urban and Barrionuevo 1996). Therefore, we first carried out a
430
detailed analysis of the levels of membrane depolarization achieved through the 4
431
stimulus pulses, given 10 msec apart (i.e. intra-burst frequency of 100 Hz), which
432
constituted each of the 5 bursts delivered at an inter-burst interval of 200 msec (i.e. 5 Hz
433
frequency; Fig. 5A; Materials and Methods). The peak amplitude of each of the five
434
depolarizing envelopes elicited by the 5 bursts were quantified for each cell and averaged
435
for the control and EE groups (Fig. 5B,C). This analysis showed that the mean amplitude
436
of peak depolarization underlying each individual burst of the TBS was not different
437
between the two groups for any of the 5 bursts (mean burst depolarization, CON: 14.4 ±
438
2.4 mV, n=11; EE: 15.7 ± 2.3 mV, n=11; Fig. 5D). Therefore, we shifted our focus to the
439
number of action potentials fired by EE and control cells within each of the 5 bursts of
440
the TBS induction protocol used to elicit LTP (Pike et al. 1999; Thomas et al. 1998) (Fig.
441
5E). This analysis showed that the number of action potentials fired during the first of the
21
442
5 bursts was significantly higher (80 % increase, p<0.01) in EE cells compared to control
443
cells (CON, 2.1 ± 0.25, n=11; EE, 3.8 ± 0.24, n=11; Fig. 5E). During the remaining 4
444
bursts the enriched cells continued to fire more action potentials, although the difference
445
was no longer statistically significant (Fig. 5E). The total number of action potentials
446
fired through all 5 bursts was significantly higher in CA1 cells from enriched rats (CON,
447
3.36 ± 0.9, n=11; EE, 7.2 ± 1.3, n=11; Fig. 5E, inset). Importantly, the number of action
448
potentials fired during TBS correlated positively with the magnitude of potentiation in
449
EPSP slope for all EE and control cells (r=0.7, p<0.05; Fig. 5F). The number of action
450
potentials fired and the magnitude of LTP in EE cells spanned a greater range compared
451
to their control counterparts (Fig. 5F). Taken together, these results highlight differences
452
in the efficacy of firing action potentials during the induction of LTP, which in turn
453
correlates with the level of synaptic potentiation achieved.
454
455
Enriched environment enhances intrinsic excitability of CA1 pyramidal cells
456
Since our results suggested that increased spiking of CA1 cells might have contributed to
457
the effects of EE on LTP, we examined if and how key parameters related to neuronal
458
excitability were modulated by exposure to EE. We first examined if action potentials,
459
evoked by somatic injection of increasing steps of depolarizing currents, differed between
460
EE and control cells (Fig. 6A). CA1 neurons from enriched rats fired a significantly
461
higher number of action potentials relative to controls for several values of current
462
injected (Number of action potentials for 200 pA current injection, CON: 9.7 ± 0.9, n=16;
463
EE: 12.4 ± 0.73, n=16; p<0.05; Fig. 6B). Further, the instantaneous frequencies for the 1st
464
inter-spike intervals (ISI) were significantly higher in EE neurons (instantaneous
22
465
frequency for 200 pA current injection; CON: 41.6 ± 5.7 Hz, n=16; EE: 66.26 ± 6.43 Hz,
466
n=16, p<0.01; Fig. 6C). Thus, consistent with our observations on EE neurons exhibiting
467
enhanced spiking during TBS-induced LTP, somatic injections of depolarizing currents
468
also led to enhanced action potential firing.
469
Next we examined the basis of this EE induced enhancement in firing by
470
comparing the threshold to fire action potentials in EE versus control cells. The voltage
471
threshold at which somatic depolarization evoked an action potential was significantly
472
reduced in CA1 neurons from EE rats (CON: -51.5 ± 1.1 mV, n=16; EE: -55.5 ± 1.3 mV,
473
n=16; p<0.05; Fig. 6D2). Correlating with a decrease in the voltage threshold, a
474
significant reduction was also observed in the current threshold or the rheobase of EE
475
neurons (25 % decrease; CON: 0.7 ± 0.03 nA, n=16; EE: 0.56 ± 0.03 nA, n=16; p<0.01;
476
Table 1). Next we focused on another facet of neuronal excitability – the after-
477
hyperpolarization potential (AHP) following a depolarizing somatic current injection that
478
reliably elicited a train of 4 to 5 action potentials (Fig. 6E1). Area CA1 pyramidal
479
neurons from EE rats exhibited a significant reduction (42 % decrease, p<0.01) in AHP
480
amplitude (CON: 2.7 ± 0.2 mV, n=16; EE: 1.9 ± 0.2 mV, n=16; Fig. 6E2). Analysis of
481
other active membrane properties (Table 1) did not show any effects of EE on the after-
482
depolarization potential (ADP), action potential amplitude and half-width, or the peak
483
dV/dt of the action potential. There were no differences between EE and control cells in
484
resting membrane potential (Vm), input resistance (Rin) and membrane time constant (τ)
485
(Table 1). The cumulative impact of these changes in neuronal excitability would explain
486
why CA1 neurons from EE rats are prone to firing more action potentials when they are
487
activated.
23
488
489
Enriched environment strengthens EPSP-spike coupling in CA1 pyramidal cells
490
In our earlier analysis of the possible reasons underlying enhanced LTP in EE neurons,
491
two observations were prominent. First, in hippocampal slices from EE rats, CA1
492
neurons fired a higher number of action potentials during the activation of synaptic
493
afferents with TBS (Fig. 5). This in turn was consistent with the increased intrinsic
494
excitability assessed through somatic depolarization (Figs. 6A–E). Second, these
495
measures of enhanced spiking and excitability caused by EE stood in striking contrast to
496
the absence of any effect on the sub-threshold membrane depolarization seen during TBS
497
delivery (Fig. 5D). Further, this lack of effect was also consistent with the finding that EE
498
did not affect the amplitude of mEPSCs. This suggested that the likelihood of firing
499
action potentials was greater in EE neurons during LTP induction despite no apparent
500
increase in the amplitude of postsynaptic depolarization caused by TBS. How does TBS-
501
induced activation of synaptic inputs that leads to comparable levels of postsynaptic
502
depolarization in EE and control cells nonetheless lead to enhanced action potential firing
503
in EE neurons? A possible mechanism is suggested by earlier studies that report
504
enrichment induced increase in population spike amplitudes in DG granule cells (Green
505
and Greenough 1986; Irvine et al. 2006). These findings have also suggested that this
506
effect is reminiscent of tetanus-induced EPSP-Spike (E-S) potentiation, a form of
507
hippocampal plasticity wherein a stronger coupling between the EPSP and spike is
508
manifested as greater population spike amplitude even in the absence of any potentiation
509
of the synaptic response (Andersen et al. 1980; Chavez-Noriega et al. 1989; Chavez-
510
Noriega et al. 1990; Daoudal and Debanne 2003; Daoudal et al. 2002). These earlier
24
511
reports on EE induced E-S potentiation like effects relied on extracellular field potential
512
recordings in the hippocampus. In this study we used whole-cell recordings that provide a
513
sensitive test of this idea at the single cell level by quantifying the strength of the E-S
514
coupling, an index of the probability of firing an action potential for a given synaptic
515
depolarization. The E-S relationships were compared between CA1 cells taken from EE
516
and normal rats under control conditions, not after inducing LTP. To this end, the
517
Schaffer collateral inputs were stimulated, while recording the slope of the resulting
518
EPSP. The E-S data was plotted by binning the slope values and finding a probability to
519
spike for each slope bin. This E-S curve was then fit with a sigmoid function and the
520
EPSP value at the 0.5 spike probability point determined for each cell. This analysis
521
shows a leftward shift in the E-S curve for EE neurons (Fig. 6F). The EPSP slope values,
522
at 0.5 spike probability were significantly lower (22% decrease) for EE neurons (CON,
523
3.26 ± 0.16 mV/ms, n=10; EE, 2.68 ± 0.27 mV/ms, n=8; p<0.05; Fig. 6F). Further, the
524
amplitude and slope of the EPSPs recorded during the E-S coupling experiments were
525
also analyzed to test if EE affected the EPSP amplification (amplitude/slope ratios). We
526
found no significant difference in EPSP amplification between the two groups (CON,
527
6.98 ± 0.3, n=10; EE, 6.29 ± 0.7, n=8; p=0.4; data not shown) (Campanac and Debanne
528
2008). Thus, exposure to EE strengthens the E-S coupling such that the same sized EPSP
529
is likely to cause the firing of more action potentials in CA1 neurons from enriched rats.
530
531
25
532
DISCUSSION
533
In this study we characterized the impact of environmental enrichment (EE) on
534
hippocampal plasticity and its functional consequences using a combination of
535
electrophysiological, morphological and behavioral analyses. Repeated exposure to EE
536
for a month gave rise to naturally occurring plasticity manifested as an increase in both
537
the structural and physiological substrates of excitatory synaptic transmission.
538
Importantly, EE also enhanced intrinsic excitability and the coupling between synaptic
539
drive and action potential firing. This naturally occurring synaptic and intrinsic plasticity,
540
in turn, served as an ideal cellular substrate for supporting further synaptic plasticity that
541
is manifested as enhanced LTP. The combined impact of these cellular and synaptic
542
changes is consistent with the significant improvement in a form of contextual learning
543
that depends on the hippocampus. Thus, the changes in synaptic transmission and
544
neuronal excitability that occurred in vivo in the intact animal during the course of the EE
545
appear to facilitate artificially induced LTP in hippocampal slices ex vivo. Indeed, these
546
two cellular mechanisms act in concert to improve new learning after EE, as shown here
547
and in previous studies.
548
549
Strengthening of the structural and physiological basis of excitatory synaptic
550
transmission
551
Previous studies have identified growth of dendrites and spines as hallmarks of structural
552
plasticity induced by EE (Faherty et al. 2003; Greenough and Volkmar 1973; Leggio et
553
al. 2005; Moser et al. 1994; Rampon et al. 2000). The EE protocol used in the present
554
study also increased spine density on the primary branches of apical dendrites of CA1
26
555
pyramidal neurons. This EE-induced increase in spine-density in the stratum radiatum of
556
area CA1, which receives Schaffer collateral inputs, is consistent with earlier reports of
557
increased excitatory synaptic transmission at the same afferents, assessed using
558
extracellular field recordings of input-output relationships (Irvine and Abraham 2005).
559
We probed the basis of this enhancement in greater detail using whole-cell recordings
560
from CA1 pyramidal cells. We observed no effects on the amplitude of spontaneous
561
mEPSCs, suggesting a lack of any significant impact of EE on the strength of individual
562
functional synapses (Turrigiano and Nelson 2004). However, we found the frequency of
563
mEPSCs to be higher in EE rats. Along with an increase in mEPSC frequency, we report
564
a significant increase in mEPSC decay time course after exposure to EE. This
565
prolongation of the decay of mEPSCs may be caused by the previously reported increase
566
in basal responsiveness of CA1 neurons to exogenous application of AMPA (Foster et al.
567
1996; Gagné et al. 1998). A change in subunit composition in the AMPAR can also
568
contribute to this increase in the decay time course. A more detailed electrophysiological
569
characterization of the kinetics and rectification properties of AMPAR will be required to
570
investigate these possibilities (Gagné et al. 1998; Naka et al. 2005). In light of the
571
increase in mEPSC frequency, we also assessed the impact of EE on paired-pulse ratios
572
and use-dependent block of NMDA-receptor currents at Schaffer collateral inputs to area
573
CA1. We found that EE affected neither of these measures of presynaptic release (Murthy
574
et al. 1997). Taken together, our results suggest that EE enhances basal excitatory
575
transmission by increasing the number of synapses on area CA1 pyramidal neurons. The
576
findings reported here are in agreement with a previous study (Foster and Dumas 2001)
577
that carried out quantal analysis at CA3-CA1 synapses to characterize the relative
27
578
contributions of presynaptic and postsynaptic changes to the increase in synaptic strength
579
caused by EE. Further support for postsynaptic mechanisms comes from a more recent
580
report showing that exposure to EE causes similar postsynaptic changes in the developing
581
hippocampus (He et al. 2010). It is interesting to note that presynaptic changes have also
582
been seen after exposure to EE, but in older animals (Artola et al.). While the nature of
583
pre and postsynaptic changes may vary with the age of experimental animals or features
584
of the enrichment paradigm used, the strengthening of both the structural and
585
physiological basis of excitatory synaptic transmission reflects a robust form of naturally
586
occurring potentiation of synaptic transmission that is developed during exposure to EE.
587
588
Augmentation of LTP
589
Natural strengthening of synaptic transmission could have diverse functional
590
consequences on electrically-induced synaptic plasticity, tested after EE exposure. For
591
instance, it could use up some of the available capacity of CA1 neurons to support further
592
synaptic plasticity, thereby occluding subsequent induction of electrically-induced LTP.
593
Alternatively, the stronger synapses could serve as an effective substrate to facilitate
594
further LTP. We find that despite EE leading to naturally occurring increase in the
595
number of spines and frequency of mEPSCs, it did not impair the ability of CA3-CA1
596
synapses to undergo further LTP after EE. Not only were these synapses able to support
597
LTP, they did so with magnitudes greater than those seen in control animals. This is in
598
clear contrast to previous reports on the absence of an effect or occlusion of electrically
599
induced LTP at the perforant path inputs to DG neurons after exposure to EE (Eckert et
600
al. 2010; Feng et al. 2001; Green and Greenough 1986).
28
601
In this connection, it is also worth noting that earlier studies employed LTP
602
induction protocols involving high-frequency tetanic stimuli that are significantly
603
stronger than the theta-burst stimulation (TBS) used here. Stronger induction protocols
604
are likely to elicit LTP that push synaptic strengths closer to saturating levels, thereby
605
leaving less room to evaluate the full impact of EE-induced increase in LTP. In contrast,
606
the TBS paradigm resembles naturally occurring hippocampal theta rhythm seen in
607
rodents during exploratory behavior (Bland 1986). Earlier studies have also established
608
the efficacy of TBS as an optimal paradigm for triggering LTP that uses fewer electrical
609
pulses and hence is more physiologically relevant compared to high frequency
610
stimulation protocols that involve much higher levels of sustained afferent activity (Chen
611
et al. 2006; Larson et al. 1986).
612
613
Postsynaptic activity during LTP induction and changes in intrinsic excitability of
614
CA1 pyramidal cells
615
To probe how EE may enhance LTP, we first focused on the levels of postsynaptic
616
activity achieved by CA1 neurons during the application of the TBS induction protocol.
617
To this end we compared two measures of postsynaptic activity – the number of action
618
potentials fired and the underlying sub-threshold depolarization – both of which have
619
been shown to be correlated with the magnitude of LTP achieved (Linden 1999; Pike et
620
al. 1999; Thomas et al. 1998; Urban and Barrionuevo 1996). Although exposure to EE
621
had no effect on the mean amplitude of each of the depolarizing envelopes elicited by the
622
five bursts of TBS, the total number of action potentials fired during these bursts was
623
significantly higher in CA1 cells from EE rats. Notably, the enhanced levels of LTP in
29
624
EE cells exhibited a positive correlation with the number of action potentials fired during
625
TBS. These observations led us to focus on the factors that may contribute to the greater
626
efficacy in firing action potentials during the induction of LTP in cells from EE animals.
627
We found three measures of neuronal excitability to be enhanced following EE. First,
628
somatic injections of depolarizing currents led CA1 neurons from EE rats to fire more
629
action potentials. Second, there was a reduction in the threshold for firing action
630
potentials in EE cells. This reduction in action potential threshold was unlikely to be
631
caused by a change in number or properties of voltage-gated sodium channels because we
632
did not observe a change in peak dV/dt values. Finally, EE caused a decrease in AHP
633
amplitude in CA1 neurons, which is in agreement with a similar finding on reduction of
634
AHP in aged rats exposed to EE (Kumar and Foster 2007). These changes are likely to
635
act together to enhance the capacity of EE cells to fire more spikes during TBS and
636
thereby support larger LTP.
637
The increase in spiking rates in EE cells may also have interesting functional
638
consequences for the propagation of action potentials from the soma back along the
639
dendrites. A recent study reported a reduction in currents mediated by the A-type K+
640
channels in oblique dendrites of CA1 neurons from EE rats (Makara et al. 2009). This
641
would increase excitability of dendritic branches and facilitate the conduction of back
642
propagating action potentials (bAP), which are known to play an important role in
643
associative synaptic LTP (Magee and Johnston 1997). Thus, the enhanced LTP seen in
644
the present study could be mediated by the higher number of action potentials generated
645
at the CA1 soma, and the consequent increase in the number of bAPs and their
30
646
conduction along CA1 dendrites (Chen et al. 2006). This possibility awaits further
647
investigation.
648
649
Stronger EPSP-spike coupling
650
Enhanced action potential firing in response to somatic depolarization, along with a
651
reduction in the action potential threshold, explains the increase in the number of action
652
potentials fired during the application of TBS in hippocampal slices from EE rats. This
653
finding, however, does not explain why EE cells fire more action potentials despite
654
having the same levels of postsynaptic depolarization triggered by TBS (Fig. 5). This gap
655
was bridged by the finding that EE also strengthens baseline EPSP-Spike (E-S) coupling
656
(without induction of LTP), i.e. the same sized EPSP evoked by activation of Schaffer
657
collateral inputs is likely to fire more action potentials in CA1 neurons from EE rats (Fig.
658
6). Stronger E-S coupling is also known to contribute to hippocampal E-S potentiation
659
(Daoudal and Debanne 2003; Daoudal et al. 2002). E-S potentiation in the CA1 area was
660
first observed after induction of electrically induced LTP, as enhancement of population
661
spike amplitudes over and above what is expected from the potentiation of the field EPSP
662
alone, or even in the absence of any potentiation of the EPSP (Andersen et al. 1980;
663
Chavez-Noriega et al. 1989; Chavez-Noriega et al. 1990). Using both in vitro and in vivo
664
extracellular field potential recordings, EE has also been shown to induce E-S
665
potentiation like effects at perforant path inputs to the DG (Irvine and Abraham 2006;
666
Green and Greenough 1986). Interestingly, exposure to EE also occludes the induction of
667
LTP at the same inputs to DG. In contrast, we find that EE enhances both LTP and E-S
668
coupling in CA1 pyramidal cells. Indeed, a stronger coupling between the EPSP and
31
669
spike appears to create optimal conditions for enhancing synaptic potentiation, not
670
impeding it, in area CA1.
671
Our results on EE leading to enhanced LTP, along with increases in intrinsic
672
excitability and E-S coupling, are also consistent with earlier reports on molecular
673
signaling mechanisms activated by EE. In particular, EE is known to up-regulate the
674
levels of cAMP response element-binding (CREB) and Protein Kinase C (PKC) (Paylor
675
et al. 1992; Ickes et al. 2000; Williams et al. 2001; Mohammed et al. 2002). Interestingly,
676
not only are PKC and CREB key mediators of synaptic plasticity, they also regulate
677
neuronal excitability (Astman et al. 1998; Lopez de Armentia et al. 2007). Hence, future
678
studies would be required to investigate if EE induced activation of these signaling
679
mechanisms provide a common molecular substrate for enhancing intrinsic excitability
680
and LTP, and the synergy between them.
681
682
Functional implications for hippocampal learning and memory
683
The range of changes in CA1 pyramidal neurons described here is ideally positioned to
684
support enhanced synaptic plasticity and its functional consequences at the behavioral
685
level. Accumulating evidence has established a critical role for LTP in area CA1 in forms
686
of spatial memory that depend on the hippocampus. In the present study, rats were given
687
a relatively brief period of exposure to the spatial context before receiving the
688
unconditioned stimulus (i.e. footshock). In other words, in this task we deliberately
689
restricted the time available to the animals to form a robust representation of the context
690
in which they were subjected to the aversive conditioning. Strikingly, prior exposure to
691
EE enabled the rats to overcome this challenge and exhibit significantly stronger recall
32
692
compared to control animals. This result is in agreement with an earlier report
693
(Woodcock and Richardson 2000) showing that differences between control and EE rats
694
in contextual conditioning were evident with a 16 sec pre-shock period, but not a longer
695
pre-shock period. These findings suggest that EE-induced facilitation of LTP and
696
neuronal excitability act in concert to enhance the capacity of CA1 pyramidal cells in a
697
manner that enables the animals to perform better in a more difficult task requiring better
698
discriminative ability (Fanselow 1986; 2000).
699
Our findings also point to potential differences in the effects of EE on synaptic
700
plasticity in different sub-regions of the hippocampus. As reported earlier, exposure to
701
EE caused E-S potentiation, but occluded LTP at excitatory synaptic inputs to DG
702
granule cells (Green and Greenough 1986; Irvine et al. 2006). In the CA1 area, in
703
contrast, EE enhances both LTP and E-S coupling at the CA3-CA1 synapses. Rapid
704
progress in cell-type-restricted gene ablation techniques has greatly advanced our
705
understanding of the distinct roles played by synaptic plasticity mechanisms in the DG
706
and CA1 area in different facets of spatial learning and memory (Nakazawa et al. 2004).
707
For instance, NMDARs in CA1 pyramidal cells are known to play a pivotal role in the
708
acquisition of spatial reference memory (Nakazawa et al. 2004; Tsien et al. 1996). On the
709
other hand, mutant mice lacking NMDARs in the DG are reported to have impaired
710
ability to distinguish two similar contexts, but perform normally in contextual fear
711
conditioning (Lee and Kesner 2004; McHugh et al. 2007; Nakazawa et al. 2004). These
712
findings on region-specific differences elicited by EE highlight the need to further
713
explore their functional implications for specific mnemonic functions performed by
714
specific microcircuits within the hippocampus.
715
33
716
REFERENCES
717
Abraham WC. Metaplasticity: tuning synapses and networks for plasticity. Nat Rev
718
Neurosci 9: 387, 2008.
719
Andersen P, Sundberg SH, Sveen O, Swann JW, and Wigström H. Possible
720
mechanisms for long-lasting potentiation of synaptic transmission in hippocampal slices
721
from guinea-pigs. J Physiol 302: 463-482, 1980.
722
Artola A, Von Frijtag JC, Fermont PCJ, Gispen WH, Schrama LH, Kamal A, and
723
Spruijt BM. Long-lasting modulation of the induction of LTD and LTP in rat
724
hippocampal CA1 by behavioural stress and environmental enrichment. Eur J Neurosci
725
23: 261-272, 2006.
726
Astman N, Gutnick MJ, and Fleidervish IA. Activation of Protein Kinase C Increases
727
Neuronal Excitability by Regulating Persistent Na+ Current in Mouse Neocortical Slices.
728
J Physiol 80: 1547-1551, 1998.
729
Blanchard RJ, and Blanchard DC. Passive and active reactions to fear-eliciting stimuli.
730
J Comp Physiol Psychol 68: 129-135, 1969.
731
Bland BH. The physiology and pharmacology of hippocampal formation theta rhythms.
732
Prog Neurobiol 26: 1-54, 1986.
733
Bliss TVP, and Collingridge GL. A synaptic model of memory: long-term potentiation
734
in the hippocampus. Nature 361: 31-39, 1993.
735
Bliss TVP, and Lømo T. Long-lasting potentiation of synaptic transmission in the
736
dentate area of the anaesthetized rabbit following stimulation of the perforant path. J
737
Physiol 232: 331-356, 1973.
34
738
Bouton M, and Bolles R. Conditioned fear assessed by freezing and by the suppression
739
of three different baselines. Learning & Behavior 8: 429-434, 1980.
740
Campanac E, and Debanne D. Spike timing-dependent plasticity: a learning rule for
741
dendritic integration in rat CA1 pyramidal neurons. The Journal of Physiology 586: 779-
742
793, 2008.
743
Chavez-Noriega LE, Bliss TVP, and Halliwell JV. The EPSP-spike (E-S) component
744
of long-term potentiation in the rat hippocampal slice is modulated by GABAergic but
745
not cholinergic mechanisms. Neurosci Lett 104: 58-64, 1989.
746
Chavez-Noriega LE, Halliwell JV, and Bliss TVP. A decrease in firing threshold
747
observed after induction of the EPSP-spike (E-S) component of long-term potentiation in
748
rat hippocampal slices. Exp Brain Res 79: 633-641, 1990.
749
Chen C, and Tonegawa S. Molecular genetic analysis of synaptic plasticity, activity-
750
dependent neural development, learning, and memory in the mammalian brain. Annu Rev
751
Neurosci 20: 157-184, 1997.
752
Chen X, Yuan L-L, Zhao C, Birnbaum SG, Frick A, Jung WE, Schwarz TL, Sweatt
753
JD, and Johnston D. Deletion of Kv4.2 Gene Eliminates Dendritic A-Type K+ Current
754
and Enhances Induction of Long-Term Potentiation in Hippocampal CA1 Pyramidal
755
Neurons. J Neurosci 26: 12143-12151, 2006.
756
Daoudal Gl, and Debanne D. Long-Term Plasticity of Intrinsic Excitability: Learning
757
Rules and Mechanisms. Learn Mem 10: 456-465, 2003.
758
Daoudal Gl, Hanada Y, and Debanne D. Bidirectional plasticity of excitatory
759
postsynaptic potential (EPSP)-spike coupling in CA1 hippocampal pyramidal neurons.
760
Proc Natl Acad Sci USA 99: 14512-14517, 2002.
35
761
Duffy SN, Craddock KJ, Abel T, and Nguyen PV. Environmental Enrichment
762
Modifies the PKA-Dependence of Hippocampal LTP and Improves Hippocampus-
763
Dependent Memory Learn Mem 8: 26-34, 2001.
764
Eckert MJ, Bilkey DK, and Abraham WC. Altered Plasticity in Hippocampal CA1,
765
But Not Dentate Gyrus, Following Long-Term Environmental Enrichment. J
766
Neurophysiol 103: 3320-3329, 2010.
767
Faherty CJ, Kerley D, and Smeyne RJ. A Golgi-Cox morphological analysis of
768
neuronal changes induced by environmental enrichment. Dev Brain Res 141: 55-61,
769
2003.
770
Fanselow M. Conditional and unconditional components of post-shock freezing.
771
Integrative Physiological and Behavioral Science 15: 177-182, 1980.
772
Fanselow MS. Associative vs topographical accounts of the immediate shock-freezing
773
deficit in rats: Implications for the response selection rules governing species-specific
774
defensive reactions. Learning and Motivation 17: 16-39, 1986.
775
Fanselow MS. Contextual fear, gestalt memories, and the hippocampus. Beh Brain Re
776
110: 73-81, 2000.
777
Feng R, Rampon C, Tang Y-P, Shrom D, Jin J, Kyin M, Sopher B, Martin GM,
778
Kim S-H, Langdon RB, Sisodia SS, and Tsien JZ. Deficient Neurogenesis in
779
Forebrain-Specific Presenilin-1 Knockout Mice Is Associated with Reduced Clearance of
780
Hippocampal Memory Traces. Neuron 32: 911-926, 2001.
781
Foster TC, and Dumas TC. Mechanism for Increased Hippocampal Synaptic Strength
782
Following Differential Experience J Neurophysiol 85 1377-1383 2001
36
783
Foster TC, Gagne J, and Massicotte G. Mechanism of altered synaptic strength due to
784
experience: relation to long-term potentiation. Brain Res 736: 243-250, 1996.
785
Frick A, and Johnston D. Plasticity of dendritic excitability. J Neurobiol 64: 100-115,
786
2005.
787
Frick A, Magee J, and Johnston D. LTP is accompanied by an enhanced local
788
excitability of pyramidal neuron dendrites. Nat Neurosci 7: 126-135, 2004.
789
Gagné J, Gélinas S, Martinoli M-G, Foster TC, Ohayon M, Thompson RF, Baudry
790
M, and Massicotte G. AMPA receptor properties in adult rat hippocampus following
791
environmental enrichment. Brain Res 799: 16-25, 1998.
792
Green EJ, and Greenough WT. Altered synaptic transmission in dentate gyrus of rats
793
reared in complex environments: evidence from hippocampal slices maintained in vitro J
794
Neurophysiol 55 739-750, 1986.
795
Greenough WT, and Volkmar FR. Pattern of dendritic branching in occipital cortex of
796
rats reared in complex environments. Exp Neurol 40: 491-504, 1973.
797
Hargreaves EL, Cain DP, and Vanderwolf CH. Learning and behavioral-long-term
798
potentiation: importance of controlling for motor activity. J Neurosci 10: 1472-1478,
799
1990.
800
He S, Ma J, Liu N, and Yu X. Early Enriched Environment Promotes Neonatal
801
GABAergic Neurotransmission and Accelerates Synapse Maturation. J Neurosci 30:
802
7910-7916, 2010.
803
Hédou G, and Mansuy IM. Inducible molecular switches for the study of long-term
804
potentiation. Philos Trans R Soc Lond B Biol Sci 358: 797-804, 2003.
37
805
Huang C-C, Yang C-H, and Hsu K-S. Do stress and long-term potentiation share the
806
same molecular mechanisms? Molecular Neurobiology 32: 223-235, 2005.
807
Huang Y-Y, Kandel ER, Varshavsky L, Brandont EP, Qi M, Idzerda RL, Stanley
808
McKnight G, and Bourtchouladz R. A genetic test of the effects of mutations in PKA
809
on mossy fiber ltp and its relation to spatial and contextual learning. Cell 83: 1211-1222,
810
1995.
811
Ickes BR, Pham TM, Sanders LA, Albeck DS, Mohammed AH, and Granholm A-C.
812
Long-Term Environmental Enrichment Leads to Regional Increases in Neurotrophin
813
Levels in Rat Brain. Exp Neurol 164: 45-52, 2000.
814
Irvine GI, and Abraham WC. Enriched environment exposure alters the input-output
815
dynamics of synaptic transmission in area CA1 of freely moving rats. Neurosci Lett 391:
816
32-37, 2005.
817
Irvine GI, Logan B, Eckert M, and Abraham WC. Enriched environment exposure
818
regulates excitability, synaptic transmission, and LTP in the dentate gyrus of freely
819
moving rats. Hippocampus 16: 149-160, 2006.
820
Johnston D, and Narayanan R. Active dendrites: colorful wings of the mysterious
821
butterflies. Trends Neurosci 31: 309-316, 2008.
822
Kim JJ, and Diamond DM. The stressed hippocampus, synaptic plasticity and lost
823
memories. Nat Rev Neurosci 3: 453-462, 2002.
824
Kim SJ, and Linden DJ. Ubiquitous plasticity and memory storage. Neuron 56: 582-
825
592, 2007.
826
Kumar A, and Foster T. Environmental enrichment decreases the afterhyperpolarization
827
in senescent rats. Brain Res 1130: 103-107, 2007.
38
828
Larson J, and Lynch G. Induction of Synaptic Potentiation in Hippocampus by
829
Patterned Stimulation Involves Two Events. Science 232: 985-988, 1986.
830
Larson J, Wong D, and Lynch G. Patterned stimulation at the theta frequency is
831
optimal for the induction of hippocampal long-term potentiation. Brain Res 368: 347-
832
350, 1986.
833
Lee I, and Kesner RP. Differential contributions of dorsal hippocampal subregions to
834
memory acquisition and retrieval in contextual fear-conditioning. Hippocampus 14: 301-
835
310, 2004.
836
Leggio MG, Mandolesi L, Federico F, Spirito F, Ricci B, Gelfo F, and Petrosini L.
837
Environmental enrichment promotes improved spatial abilities and enhanced dendritic
838
growth in the rat. Beh Brain Re 163: 78-90, 2005.
839
Linden DJ. The Return of the Spike: Postsynaptic Action Potentials and the Induction of
840
LTP and LTD. Neuron 22: 661-666, 1999.
841
Lopez de Armentia M, Jancic D, Olivares R, Alarcon JM, Kandel ER, and Barco A.
842
cAMP Response Element-Binding Protein-Mediated Gene Expression Increases the
843
Intrinsic Excitability of CA1 Pyramidal Neurons. J Neurosci 27: 13909-13918, 2007.
844
Lynch MA. Long-Term Potentiation and Memory. Physiol Rev 84: 87-136, 2004.
845
Magee JC, and Johnston D. A Synaptically Controlled, Associative Signal for Hebbian
846
Plasticity in Hippocampal Neurons. Science 275: 209-213, 1997.
847
Makara JK, Losonczy A, Wen Q, and Magee JC. Experience-dependent
848
compartmentalized dendritic plasticity in rat hippocampal CA1 pyramidal neurons. Nat
849
Neurosci 12: 1485-1487, 2009.
39
850
Malenka RC, and Bear MF. LTP and LTD: An Embarrassment of Riches. Neuron 44:
851
5-21, 2004.
852
Manabe T, and Nicoll RA. Long-term potentiation: evidence against an increase in
853
transmitter release probability in the CA1 region of the hippocampus. Science 265: 1888-
854
1892, 1994.
855
Martinez JL, and Derrick BE. LONG-TERM POTENTIATION AND LEARNING.
856
Ann Rev Psychol 47: 173-203, 1996.
857
McHugh TJ, Jones MW, Quinn JJ, Balthasar N, Coppari R, Elmquist JK, Lowell
858
BB, Fanselow MS, Wilson MA, and Tonegawa S. Dentate Gyrus NMDA Receptors
859
Mediate Rapid Pattern Separation in the Hippocampal Network. Science 317: 94-99,
860
2007.
861
Mohammed AH, Zhu SW, Darmopil S, Hjerling-Leffler J, Ernfors P, Winblad B,
862
Diamond MC, Eriksson PS, and Bogdanovic N. Environmental enrichment and the
863
brain. Prog Brain Res 138: 109-133, 2002.
864
Moser MB, Trommald M, and Andersen P. An increase in dendritic spine density on
865
hippocampal CA1 pyramidal cells following spatial learning in adult rats suggests the
866
formation of new synapses. Proc Natl Acad Sci USA 91: 12673-12675, 1994.
867
Moyer Jr JR, Thompson LT, and Disterhoft JF. Trace Eyeblink Conditioning
868
Increases CA1 Excitability in a Transient and Learning-Specific Manner. J Neurosci 16:
869
5536-5546, 1996.
870
Murthy VN, Sejnowski TJ, and Stevens CF. Heterogeneous Release Properties of
871
Visualized Individual Hippocampal Synapses. Neuron 18: 599-612, 1997.
40
872
Naka F, Narita N, Okado N, and Narita M. Modification of AMPA receptor properties
873
following environmental enrichment. Brain and Development 27: 275-278, 2005.
874
Nakazawa K, McHugh TJ, Wilson MA, and Tonegawa S. NMDA receptors, place
875
cells and hippocampal spatial memory. Nat Rev Neurosci 5: 361-372, 2004.
876
Narayanan R, and Johnston D. Long-Term Potentiation in Rat Hippocampal Neurons
877
Is Accompanied by Spatially Widespread Changes in Intrinsic Oscillatory Dynamics and
878
Excitability. Neuron 56: 1061-1075, 2007.
879
Nguyen PV, and Kandel ER. Brief theta-burst stimulation induces a transcription-
880
dependent late phase of LTP requiring cAMP in area CA1 of the mouse hippocampus.
881
Learn Mem 4: 230-243, 1997.
882
Nithianantharajah J, Barkus C, Murphy M, and Hannan AJ. Gene-environment
883
interactions modulating cognitive function and molecular correlates of synaptic plasticity
884
in Huntington's disease transgenic mice. Neurobiology of Disease 29: 490-504, 2008.
885
Paylor R, Morrison SK, Rudy JW, Waltrip LT, and Wehner JM. Brief exposure to
886
an enriched environment improves performance on the Morris water task and increases
887
hippocampal cytosolic protein kinase C activity in young rats. Beh Brain Res 52: 49-59,
888
1992.
889
Phillips RG, and LeDoux JE. Lesions of the dorsal hippocampal formation interfere
890
with background but not foreground contextual fear conditioning. Learn Mem 1: 34-44,
891
1994.
892
Pike FG, Meredith RM, Olding AWA, and Paulsen O. Postsynaptic bursting is
893
essential for ‘Hebbian’ induction of associative long-term potentiation at excitatory
894
synapses in rat hippocampus. J Physiol 518: 571-576, 1999.
41
895
Poirazi P, and Mel BW. Impact of active dendrites and structural plasticity on the
896
memory capacity of neural tissue. Neuron 29: 779-796, 2001.
897
Prange O, and Murphy TH. Correlation of Miniature Synaptic Activity and Evoked
898
Release Probability in Cultures of Cortical Neurons. J Neurosci 19: 6427-6438, 1999.
899
Rampon C, Tang Y-P, Goodhouse J, Shimizu E, Kyin M, and Tsien JZ. Enrichment
900
induces structural changes and recovery from nonspatial memory deficits in CA1
901
NMDAR1-knockout mice. Nat Neurosci 3: 238-244, 2000.
902
Rosenzweig MR. Environmental complexity, cerebral change, and behavior. Am Psychol
903
21: 321-332, 1966.
904
Rosenzweig MR, and Bennett EL. Psychobiology of plasticity: effects of training and
905
experience on brain and behavior. Beh Brain Res 78: 57-65, 1996.
906
Schrijver NCA, Pallier PN, Brown VJ, and Würbel H. Double dissociation of social
907
and environmental stimulation on spatial learning and reversal learning in rats. Beh Brain
908
Res 152: 307-314, 2004.
909
Sjostrom PJ, Rancz EA, Roth A, and Hausser M. Dendritic excitability and synaptic
910
plasticity. Physiol Rev 88: 769-840, 2008.
911
Staff NP, Jung H-Y, Thiagarajan T, Yao M, and Spruston N. Resting and Active
912
Properties of Pyramidal Neurons in Subiculum and CA1 of Rat Hippocampus. J Physiol
913
84: 2398-2408, 2000.
914
Thomas MJ, Watabe AM, Moody TD, Makhinson M, and O’Dell TJ. Postsynaptic
915
Complex Spike Bursting Enables the Induction of LTP by Theta Frequency Synaptic
916
Stimulation. J Neurosci 18: 7118-7126, 1998.
42
917
Tonegawa S, Nakazawa K, and Wilson MA. Genetic neuroscience of mammalian
918
learning and memory. Philos Trans R Soc Lond B Biol Sci 358: 787-795, 2003.
919
Tsien JZ, Huerta PT, and Tonegawa S. The essential role of hippocampal CA1 NMDA
920
receptor-dependent synaptic plasticity in spatial memory. Cell 87: 1327-1338, 1996.
921
Turrigiano GG, and Nelson SB. Homeostatic plasticity in the developing nervous
922
system. Nat Rev Neurosci 5: 97-107, 2004.
923
Urban NN, and Barrionuevo G. Induction of Hebbian and Non-Hebbian Mossy Fiber
924
Long-Term Potentiation by Distinct Patterns of High-Frequency Stimulation. J Neurosci
925
16: 4293-4299, 1996.
926
van Praag H, Kempermann G, and Gage FH. Neural consequences of enviromental
927
enrichment. Nat Rev Neurosci 1: 191-198, 2000.
928
Weisz DJ, Clark GA, and Thompson RF. Increased responsivity of dentate granule
929
cells during nictitating membrane response conditioning in rabbit. Beh Brain Res 12:
930
145-154, 1984.
931
Williams BM, Luo Y, Ward C, Redd K, Gibson R, Kuczaj SA, and McCoy JG.
932
Environmental enrichment: Effects on spatial memory and hippocampal CREB
933
immunoreactivity. Physiology & Behavior 73: 649-658, 2001.
934
Woodcock EA, and Richardson R. Effects of Environmental Enrichment on Rate of
935
Contextual Processing and Discriminative Ability in Adult Rats. Neurobiol Learn Mem
936
73: 1-10, 2000.
937
Zhang W, and Linden DJ. The other side of the engram: experience-driven changes in
938
neuronal intrinsic excitability. Nat Rev Neurosci 4: 885-900, 2003.
939
940
43
941
FIGURE LEGENDS
942
Fig. 1: Enriched environment protocol
943
A: Schematic showing the duration of the EE protocol. B, C: Photograph of the control
944
and enriched housing cage. D: Photograph of the playing arena for the enriched rats.
945
946
Fig. 2: EE increases contextual fear learning and hippocampal spine density
947
A: Mean (±SEM) value of % freezing during context re-exposure was significantly
948
higher in enriched rats (CON, N = 14; EE, N= 14). *, p<0.05; **, p<0.01, Student’s t-
949
test. B: Representative segments from primary branches of Golgi stained CA1 pyramidal
950
neurons from control (left) and enriched (right) rats (Scale bar: 4µm). C: Mean (±SEM)
951
value for spine-density (calculated as the average number of spines per 10 μm) of CA1
952
pyramidal neurons from enriched rats was higher than the control rats. **, p<0.01,
953
Student’s t-test; (CON, n = 14, N=4; EE, n= 16, N=4). D: Segmental analysis of the mean
954
(±SEM) number of spines in each successive 10-μm segment along primary branches as a
955
function of the distance of that segment from the origin of the branch. Spine density in all
956
segments was significantly higher for neurons from enriched rats.
957
958
Fig. 3: EE increases the frequency and decay-time course of mEPSC events in CA1
959
neurons with no change in release probability
960
A1, A2: Representative sample traces showing mEPSC recordings from neurons in
961
control (grey trace) and enriched rats (black trace). B: Cumulative distribution plots for
962
inter-event interval (B1), amplitude (B2) and decay time (B3), containing 50 randomly
963
selected events from each neuron (CON, grey, n= 14 neurons; EE, black, n=13 neurons).
44
964
The cumulative distributions for inter-event interval and decay time were significantly
965
different between neurons from control and enriched rats. *, p<0.001, K-S test. The insets
966
in B1, B2 and B3 show mean (±SEM) for mEPSC frequency, amplitude and decay time
967
respectively. *, p<0.05; **, p<0.01, Student’s t-test. The traces in B3 illustrate increased
968
decay time in neurons from enriched rats (traces are peak-scaled average of 50 randomly
969
selected events). C: Paired-pulse ratio was not altered in neurons from enriched rats
970
across all inter-stimulus intervals. The data plotted is mean (±SEM) of paired-pulse ratios
971
(CON, n=12 neurons; EE, n=11 neurons). The traces are representative examples of
972
paired-pulse facilitation (at 50 ms inter-stimulus interval, average of 5 traces) in neurons
973
from control and enriched rats. D1: Representative traces illustrating the decay in
974
amplitude of NMDAR EPSCs in presence of MK-801. D2: Normalized mean ((±SEM)
975
amplitudes of NMDAR EPSCs (in presence of MK-801) evoked in neurons from control
976
and enriched rats. The inset shows the mean (±SEM) of tau values for the single
977
exponential fit of decay time course of NMDAR EPSCs (CON, n=10; EE, n=10).
978
979
Fig. 4: EE increases TBS-LTP in CA1 neurons.
980
A1, B1: Time course plots of TBS (solid upward arrow) induced LTP in representative
981
CA1 neurons from control (grey) and enriched (black) rats. Traces show average EPSPs
982
from pre-TBS and post-TBS (9 sweeps each; dotted trace, before TBS; solid trace, after
983
TBS). A2, B2: Time course plots depicting the stability of Rs, Rin and Vm during LTP
984
experiments. C: Summary graph showing the mean (±SEM) EPSP slope (%) in CA1
985
neurons from control (grey circles, n=13) and enriched (black circles, n=12) rats. D: The
986
cumulative probability plot summarizes the data from C. Each point represents the
45
987
magnitude of change relative to baseline as a cumulative fraction of total number of
988
experiments for any given experiment, 25-30 min after TBS.
989
990
Fig 5: EE increases the number of action potentials fired during TBS induction.
991
A: Schematic showing the theta burst stimulation protocol. B1, C1: Representative traces
992
showing the voltage responses to TBS in CA1 neurons from control and enriched rats.
993
B2, C2: highlighted traces in B1 and C1 (after filtering out 100 Hz band). B3, C3: Traces
994
in B2 and C2 after subtraction of spikes (the peak amplitude is measured as Burst
995
depolarization). D: The mean (±SEM) of the depolarization amplitude during individual
996
bursts of TBS was not different between the two groups. E: The mean (±SEM) number of
997
spikes fired during the bursts was significantly higher in neurons from enriched rats. The
998
inset shows mean (±SEM) of total number of spikes for all bursts. ** p<0.01, *, p<0.05,
999
Student’s t-test. F: A significant correlation (r=0.7, p<0.05) was observed for the total
1000
number of action potentials fired during TBS with the % potentiation in EPSP slope for
1001
all neurons from control (grey) and enriched rats (black). The range of number of action
1002
potentials and % potentiation was much larger for the neurons from enriched rats (CON
1003
range, grey dashed lines; EE range, black dashed lines)
1004
1005
Fig. 6: Enrichment increases the excitability of CA1 neurons to both somatic
1006
depolarization and synaptic stimulation.
1007
A: Representative traces of action potentials fired in response to depolarizing current
1008
injections (200 and 300 pA) to the soma of CA1 neurons from control (grey) and
1009
enriched rats (black). B: The mean (±SEM) number of action potentials fired is plotted
46
1010
against the increasing current injections. The curve is shifted to the left for neurons from
1011
enriched rats indicating increased excitability. C: Instantaneous frequency for first ISI
1012
(Hz) was significantly higher in neurons from enriched rats. *, p<0.05; **, p<0.01,
1013
Student’s t-test (CON, n=16; EE, n=16). D1: Representative action potentials fired in
1014
response to depolarizing current step in neurons from control and enriched rats. The
1015
arrowheads indicate the action-potential threshold. D2: The mean (±SEM) action-
1016
potential threshold for CA1 neurons from enriched rats is more hyperpolarized. E1:
1017
Representative traces showing AHP in CA1 neurons from control (grey) and enriched
1018
rats (black). E2: Mean (±SEM) value of AHP amplitude was significantly reduced in
1019
CA1 neurons from enriched rats. *, p<0.05; **, p<0.01, Student’s t-test (CON, n=16; EE,
1020
n=16). F: Mean (±SEM) firing probability is plotted as a function of EPSP slope. The
1021
sigmoid fits of the data (dashed lines) illustrate a leftward shift (solid leftward arrow) in
1022
firing probability for neurons in enriched rats. The traces are EPSPs from representative
1023
CA1 neurons in control (grey) and enriched (black) rats. The inset is the mean (±SEM)
1024
values of EC-50 of the sigmoid fits of E-S coupling. *, p<0.05, Student’s t-test (CON,
1025
n=10; EE, n=8).
1026
47
1027
TABLE 1
1028
Summary of passive and active membrane properties of CA1 neurons from control
1029
and enriched rats
CON
EE
Parameters
Mean ± SEM
n
Mean ± SEM
n
p value
Resting membrane potential (mV)
-64.8 ± 0.8
16
-66.4 ± 0.5
16
0.1
Input resistance (MΩ)
100.6 ± 3.8
16
105.3 ± 4.4
16
0.5
Membrane time constant (ms)
27.9 ± 1
16
31.2 ± 1.7
16
0.1
AHP (mV)
2.7 ± 0.2
16
1.9 ± 0.2
16
0.009 ‡
ADP (mV)
16.2 ± 0.4
16
15.8 ± 0.5
16
0.1
AP amplitude (mV)
120.1 ± 1.1
16
118.8 ± 1.2
16
0.45
AP half-width (ms)
1.6 ± 0.02
16
1.6 ± 0.04
16
0.6
AP threshold (mV)
-51.5 ± 1.1
16
-55.5 ± 1.3
16
0.02
Max. dV/dt (KVs-1)
0.42 ± 0.001
16
0.42 ± 0.001
16
0.9
Current threshold (nA)
0.7 ± 0.03
16
0.56 ± 0.03
16
0.008‡
Sag Voltage (mV)
13.3 ± 0.3
16
12.7 ± 0.5
16
0.6
Resonance frequency (Hz)
1.7 ± 0.2
11
2 ± 0.2
14
0.14
1030
1031
The CA1 neurons in EE group have decreased AHP amplitude and have a reduced action
1032
potential (AP) threshold.
1033
(*, p<0.05, ‡, p<0.01; Student’s t-test; AHP – After-hyperpolarization potential; ADP –
1034
After-depolarization potential; AP – Action potential)
1035
*
A
Contextual fear conditioning
P25 rats
30-35 days
enriched/control
housing
C
B
Control
housing
Spine-density analysis
Electrophysiological recordings
D
Enriched
housing
Enriched rats - playing arena
(4 hrs/day)
A
0
CON
EE
15
5
0
EE
10
Number of spines
(in 10 m segments)
**
CON
EE
40
CON
% Freeezing
60
20
20
20
**
Number of spines
(per 10 m
m)
80
D
C
B
*
15
**
*
*
**
40
50
60
** **
*
10
5
0
10
20
30
70
80
Distance from origin of the branch
(m)
A1
A2
CON
0
0
0.1
0.2
0.0
0.0
0.6
0.4
0.2
5
8000
0
-100
150
x10
pA
-100
-12
200
-200
-150
100
-200
50
25 ms
-250
500
1000
1500
2000
2500
3000
40
60
80
Amplitude (pA)
Trial #
1
10
20
250
0
-50
-150
20
D1
0
500
1000
1500
2000
2500
90
100
0
3000
CON
2.0
300
1.5
2.00
2.04
2.08
s
2.12
200 ms
250
200
pA
1.0
Trial #
1
10
20
150
100
0.5
90
100
50
0
0.0
50
100
150
200
Inter-pulse interval (msec)
250
1.0
2 ms
-0.2
-0.2
20
25
-0.4
-0.4
-0.6
-0.6
0.8
8
-0.8
-1.0
0
0.6
30
s
5
10
10
15
15
20
20
0
5 1510 20
0.4
0.2
**
6
4
2
0
0.0
10000
-50
15
2.00
EE
2.04
2.08
s
2.12
100
0
5
10
15
20
Decay time (ms)
D2
80
1.0
Tau (trials)
6000
EE
50 pA
-12
10
Normalised EPSC amplitude
4000
CON
x10
Paired-pulse ratio
15
50 pA
2000
C
2.5
20
0
Inter-event interval (ms)
3.0
25
0.8
0.0
0
3.5
B3
10
EE
0.2
EE
0.4
0.3
15 s
0.8
60
40
20
EE
0.4
0.6
14
1.0
Cumulative probability
**
0.8
-50
12
CON
B2
CON
10
Cumulative probability
8
s
EE
1.0
6
CON
4
Events/s
Cumulative probability
2
CON
B1
0
-20
10 pA-30
-40
Amplitude (pA)
30
Decay time (ms)
20
50
-10
pA
10
40
EE
10
10
0
0.6
0.4
0
20
40
60
Trial number
80
100
20
-3
10
100
5
0
0
5
10
15
20
1000
25
2000
30
Time after TBS (mins)
200
3000
%
100
-70
5
0
0
500
0
5
10
15
1000 1500 2000 2500 3000
20
25
30
25
30
Rs
100
Rin
0
20
100
0
Vm 0
5
10
15
0
5
10
20
25
30
-60
-5
15
20
25
-80
Vm
-70
0
20
-60
30
-5
Time after TBS (mins)
0
5
10
15
20
Time after TBS (mins)
C
D
Cumulative probabilty
400
Averaged normalized
EPSP slope (%)
5 ms
10
100
0
200
Rin
0
-80 -5
15
200
%
100
10 mV
20
Time after TBS (mins)
B2
Rs
200
-5
mV
%
%
5 ms
0
0
200
mV
10 mV
15
300
-3
200
EE
400
x10
300
Normalized EPSP slope (%)
400
-5
A2
B1
CON
x10
Normalized EPSP slope (%)
A1
300
200
100
EE
CON
1.0
0.8
0.6
0.4
0.2
0.0
0
-5
0
5
10
15
20
Time after TBS (mins)
25
30
0
100
200
300
400
EPSP slope (%)
500
600
B1
40
4
5
EE
60
40
100 ms
mV
20
0
-20
3000
E
1000
25
20
15
10
5
0
1
2
3
4
Theta burst number
5
2000
3000
30
20
10
10
4
**
3
2
8
1000
2000
**
6
4
2
0
1
0
1
2
3
1.0
50 ms
0
0
4000
0.8
50
20
0
0
4000
60
0.6
40
4
Theta burst number
3000
F
4000
Total # of action potentials
2000
s
10 mV
30
Total # of
action potentials
1000
C3
0.4
40
0
0
0.2
50
10
0
Burst depolarization (mV)
60
0.0
50 ms
20
10
C2
1.0
-3
-3
-3
30
20
D
0.8
40
x10
30
0.6
50
x10
40
B3.
10 mV
0.4
-3
60
0.2
50
# of action potentials / burst
0.0
EE
60
B2
x10
-20
x10
0
CON
20
3
C1
20 mV
CON
60
2
1
Theta burst stimulation
mV
A
5
0
1000
2000
3000
12
10
8
6
4
2
0
0
100
200
300
400
500
% potentiation in EPSP slope
A
CON
0.10
0.08
0.08
0.06
0.06
0.04
0.04
0.02
0.02
0.00
0.12
0.00
0.12
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.3
0.4
0.06
0.5
0.04
0.00
0.1
0.2
0.3
0.4
0.5
0.6
0.7
C
s
0.1
0.8
0.2
0.3
0.4
0.5
*
First ISI (Hz)
*
*
* *
100
80
**
40
*
20
0
0
300
400
100
500
200
D2
D1
E1
*
5 mV
mV
5
0
-10
-5
0.50
s
0.60
0.70
0.80
0
0.104
25
F
20
15
mV
1.0
10 ms
10
5
0
0.8
20
30
40
50
ms
60
70
*
80
3
0.6
0.4
2
1
EE
0.103
Firing probability
0.102
s
-20
50 ms
CON
0
10
EC-50
2
15
-30
EE
4
20
-40
5 mV
6
-50
CON
1 ms
AP threshold (mV)
20 mV
8
0.2
0
0.0
1
2
400
500
E2
2
0
300
Current injected (pA)
Current injected (pA)
3
4
5
EPSP slope (mV/ms)
6
**
3
2
1
0
EE
200
CON
100
0.101
0.8
*
**
60
5
0.7
* *
**
AHP amplitude (mV)
# of action potentials
20
*
0.6
s
120
0.100
0.8
0.02
0.00
10
0.7
0.06
100 ms
0.02
15
0.6
s
0.08
V
0.08
B
0.2
0.10
0.10
0.04
600 ms
0.1
0.8
s
40 mV
V
Increasing
steps of
somatic
depolarization
EE
0.12
0.10
V
V
0.12