* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Enhanced intrinsic excitability and EPSP
Donald O. Hebb wikipedia , lookup
Limbic system wikipedia , lookup
Neuromuscular junction wikipedia , lookup
Endocannabinoid system wikipedia , lookup
Mirror neuron wikipedia , lookup
Dendritic spine wikipedia , lookup
Neuroplasticity wikipedia , lookup
Adult neurogenesis wikipedia , lookup
Metastability in the brain wikipedia , lookup
Memory consolidation wikipedia , lookup
Central pattern generator wikipedia , lookup
Action potential wikipedia , lookup
Holonomic brain theory wikipedia , lookup
Neural oscillation wikipedia , lookup
Multielectrode array wikipedia , lookup
Clinical neurochemistry wikipedia , lookup
Neural coding wikipedia , lookup
Biological neuron model wikipedia , lookup
Development of the nervous system wikipedia , lookup
Neurotransmitter wikipedia , lookup
Electrophysiology wikipedia , lookup
Premovement neuronal activity wikipedia , lookup
End-plate potential wikipedia , lookup
Sexually dimorphic nucleus wikipedia , lookup
Molecular neuroscience wikipedia , lookup
Stimulus (physiology) wikipedia , lookup
Neuropsychopharmacology wikipedia , lookup
Apical dendrite wikipedia , lookup
Long-term depression wikipedia , lookup
Neuroanatomy wikipedia , lookup
Spike-and-wave wikipedia , lookup
Optogenetics wikipedia , lookup
Epigenetics in learning and memory wikipedia , lookup
Single-unit recording wikipedia , lookup
Nervous system network models wikipedia , lookup
De novo protein synthesis theory of memory formation wikipedia , lookup
Synaptogenesis wikipedia , lookup
Feature detection (nervous system) wikipedia , lookup
Environmental enrichment wikipedia , lookup
Pre-Bötzinger complex wikipedia , lookup
Long-term potentiation wikipedia , lookup
Synaptic gating wikipedia , lookup
Chemical synapse wikipedia , lookup
Activity-dependent plasticity wikipedia , lookup
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