title: a1 adenosine receptor-mediated girk channels ......172 were rinsed in pbs buffer, and then...

1

Title: A1 adenosine receptor-mediated GIRK channels contributes to the resting 1

conductance of CA1 neurons in the dorsal hippocampus 2

3

4

5

Authors: Chung Sub Kim and Daniel Johnston 6

Author contributions: C.S.K and D.J. designed research; C.S.K performed research; 7

C.S.K. analyzed data; C.S.K. and D.J. wrote the paper. 8

9

Department of Neuroscience and Center for Learning and Memory 10

University of Texas at Austin, Austin, TX 78712, USA. 11

12

Running Head: Coupling of A1ARs to GIRK in dorsal CA1 neurons 13

14

Daniel Johnston, Ph.D. 15

Center for Learning and Memory 16

University of Texas at Austin 17

100 E. 24th Street 18

Austin, TX 78712-0805 19

Phone: 512-232-6564 20

E-mail: [email protected] 21

22

23

Articles in PresS. J Neurophysiol (February 4, 2015). doi:10.1152/jn.00951.2014

Copyright © 2015 by the American Physiological Society.

2

Number of pages: 40 24

Number of figures: 7 25

Number of tables: 9 26

Number of words for Abstract: 185 words 27

Number of words for Introduction: 413 words 28

Number of words for Discussion: 1367 words 29

30

Acknowledgments: This study was supported by grants from National Institutes of 31

Health NS084473 and NS077477 (DJ). We thank Drs. Rick Gray, Brian Kalmbach, 32

and Darrin Brager for reviewing the manuscript and comments throughout this study. 33

We also thank Drs. Randy Chitwood, Kelly Dougherty, and Ann Clemens for giving 34

helpful comments. 35

36

Conflict of Interest: The authors declare no competing financial interests 37

38

39

40

41

42

43

44

3

Abstract 45

The dorsal and ventral hippocampi are functionally and anatomically distinct. 46

Recently, we reported that dorsal CA1 neurons have a more hyperpolarized resting 47

membrane potential (RMP), a lower input resistance (Rin), and fire fewer action 48

potentials for a given current injection than ventral CA1 neurons. Differences in the 49

hyperpolarizing HCN conductance between dorsal and ventral neurons have been 50

reported, but these differences cannot fully account for the different resting 51

properties of these neurons. Here we show that coupling of A1 adenosine receptors 52

(A1ARs) to G-protein coupled inwardly rectifying potassium (GIRK) conductance 53

contributes to the intrinsic membrane properties of dorsal CA1 neurons but not 54

ventral CA1 neurons. The block of GIRKs with either barium or the more specific 55

blocker tertiapin-Q revealed that there is more resting GIRK conductance in dorsal 56

CA1 neurons compared to ventral CA1 neurons. We found that the higher resting 57

GIRK conductance in dorsal CA1 neurons was mediated by tonic A1 adenosine 58

receptor activation. These results demonstrate that the different resting membrane 59

properties between dorsal and ventral CA1 neurons are due in part to higher 60

adenosine A1 receptor-mediated GIRK activity in dorsal CA1 neurons. 61

62

63

64

65

66

Keywords: Dorsal and ventral hippocampus; A1 adenosine receptors; GIRK 67

4

Introduction 68

The rodent hippocampus can be divided into dorsal and ventral regions based 69

on biochemical, anatomical, and behavioral observations (van Groen and Wyss 70

1990; Moser and Moser 1998; Dong et al. 2009; Fanselow and Dong 2010). Until 71

recently, the electrophysiological properties of CA1 pyramidal neurons along the 72

longitudinal hippocampal axis were often thought to be uniform. However, growing 73

evidence demonstrates that intrinsic membrane properties of CA1 neurons from the 74

dorsal and ventral hippocampus are significantly different (Dougherty et al. 2012; 75

Marcelin et al. 2012a). For example, dorsal neurons have a more hyperpolarized 76

resting membrane potential (RMP) and a lower input resistance (Rin). As such, they 77

fire fewer action potentials in response to a given current injection compared to 78

ventral neurons (Dougherty et al. 2012). Although functional Ih is different between 79

dorsal and ventral CA1 neurons (Dougherty et al. 2013), the intrinsic membrane 80

properties of these neurons cannot be fully explained by Ih-related differences. 81

Voltage-gated ion channels and neuronal morphology contribute to the 82

intrinsic membrane properties of neurons (Hoffman et al. 1997; Magee 1998; Vetter 83

et al. 2001; Spruston 2008). Differences in RMP and Rin between dorsal and ventral 84

CA1 neurons may therefore be explained in part by the expression of voltage-gated 85

ion channels. Given the more hyperpolarized membrane potential and lower Rin in 86

dorsal neurons compared to ventral neurons, we hypothesized that dorsal CA1 87

neurons possess a greater resting potassium conductance than ventral CA1 neurons. 88

Furthermore, inwardly rectifying K+ (IRK) channels and G protein-coupled inwardly 89

rectifying K+ (GIRK) channels are highly expressed in the CA1 pyramidal neurons 90

5

(Karschin et al. 1996; Liao et al. 1996; Ponce et al. 1996; Drake et al. 1997). G 91

protein-coupled inwardly rectifying K+ channels (GIRKs) are activated by Gi/o-coupled 92

metabotropic receptors and hyperpolarize the membrane potential and decrease Rin, 93

providing a decrease in neuronal excitability (Andrade et al. 1986; Mihara et al. 1987; 94

North 1989; Luscher et al. 1997). 95

In this study, we investigated the resting conductances in dorsal and ventral 96

CA1 pyramidal neurons. We found that dorsal neurons have more Ba2+-sensitive 97

conductance compared to ventral neurons. Application of tertiapin-Q revealed that 98

this greater Ba2+-sensitive conductance in dorsal CA1 neurons was mediated by 99

GIRK channels. Finally, we found that this higher resting GIRK conductance in dorsal 100

neurons was mediated by tonic activation of A1 adenosine receptors (A1ARs). 101

Together, our results suggest that higher resting GIRK conductance activated by A1 102

adenosine receptors is present in dorsal but not ventral CA1 neurons and contributes 103

to intrinsic membrane properties of these neurons. 104

105

106

107

108

109

110

111

112

113

6

Materials and Methods 114

Animals. Hippocampal slices from the dorsal and ventral poles were prepared from 115

11- to 13-week-old male Sprague Dawley rats housed 2-3 per cage on a 12 hr light 116

schedule (on 7:00 A.M. off 7:00 P.M.) with ad libitum access to water and food. All 117

procedures involving animals were approved by the University of Texas at Austin 118

Institutional Animal Care and Use Committee (IACUC). 119

Drugs. Barium chloride was purchased from Fisher Scientific. Tertiapin-Q was 120

purchased from Alomone Lab. DPCPX, ZD7288, 2’ME-CCPA, D-APV, Gabazine, 121

CGP 55845, and DNQX were purchased from Abcam Inc. 122

Slice preparation. Rats were anesthetized with a lethal dose of a ketamine/xylazine 123

mixture (90/10mg/ml) and transcardially perfused with ice-cold artificial cerebral 124

spinal fluid (aCSF) composed of (in mM): 2.5 KCl, 1.25 NaH2PO4 , 25 NaHCO3, 0.5 125

CaCl2, 7 MgCl2, 7 dextrose, 210 sucrose, 1.3 ascorbic acid, and 3 sodium pyruvate, 126

bubbled with 95% O2 - 5% CO2. The brain was then removed, and hemisected along 127

the longitudinal fissure. For dorsal hippocampal slices, a blocking cut was made at 128

20o - 30o from the posterior coronal plane and collected at the anterior end of the 129

forebrain. Ventral hippocampal slices were made from horizontal section. 300 μm 130

thick hippocampal slices were made in ice-cold aCSF using a vibrating microtome 131

(Microslicer DTK-Zero1, DSK, Kyoto, Japan). Slices were then transferred to a 132

holding chamber for 30 min at 35oC containing (in mM) 125 NaCl, 2.5 KCl, 1.25 133

NaH2PO4 , 25 NaHCO3, 2 CaCl2, 2 MgCl2, 12.5 dextrose, 1.3 ascorbic acid, and 3 134

sodium pyruvate, bubbled with 95% O2 - 5% CO2. Slices were then incubated for at 135

least 45 min at room temperature. 136

7

Whole-cell current-clamp recordings. 137

Whole-cell current-clamp recordings were performed as previously described (Kim et 138

al. 2012). Briefly, hippocampal slices were submerged in a recording chamber 139

continuously perfused with aCSF containing synaptic blockers (D-APV 25 μM, 140

Gabazine 2 μM, DNQX 20 μM) heated to 32-34 oC flowing at a rate of 1 to 2 ml/min. 141

ACSF contained (in mM) 125 NaCl, 2.5 KCl, 1.25 NaH2PO4, 25 NaHCO3, 2 CaCl2, 1 142

MgCl2, and 12.5 dextrose, bubbled with 95% O2 - 5% CO2. CA1 pyramidal neurons 143

were visualized using a microscope (Zeiss Axioskop) fitted with differential 144

interference contrast optics (Stuart et al. 1993). Patch pipettes (4–7 MΩ) were 145

prepared with capillary glass (external diameter 1.65 mm, World Precision 146

Instruments) using Flaming/Brown micropipette puller and filled with an internal 147

solution containing (in mM) 120 K-gluconate, 20 KCl, 10 HEPES, 4 NaCl, 7 K2-148

phosphocreatine, 4 Mg-ATP, 0.3 Na-GTP (pH 7.3 with KOH). Neurobiotin (Vector 149

Labs) was used (0.1-0.2 %) for subsequent histological processing. Somatic whole-150

cell current-clamp recordings were performed using a Dagan BVC-700A amplifier 151

(Dagan, Minneapolis, MN). Electrical signals were filtered at 5 kHz, sampled at 152

20 kHz, and digitized by an ITC-18 interface (HEKA Instruments Inc., Bellmore, NY) 153

connected to a computer running custom software written in IGOR Pro 154

(Wavemetrics). Experiments were terminated if the series resistance exceeded 30 155

MΩ. Membrane potentials were corrected for liquid junction potentials (- 8mV). For 156

tertiapin experiments, we used log-order increases in concentration (0.03, 0.3, and 2 157

μM) 158

Immunohistochemistry. Animals were anesthetized and perfused through the 159

8

ascending aorta using ice-cold artificial cerebral spinal fluid (aCSF) composed of (in 160

mM): 2.5 KCl, 1.25 NaH2PO4 , 25 NaHCO3, 0.5 CaCl2, 7 MgCl2, 7 dextrose, 210 161

sucrose, 1.3 ascorbic acid, and 3 sodium pyruvate, bubbled with 95% O2 - 5% CO2 162

followed by 4% paraformaldehyde (PFA) in PBS described previously (Kim et al. 163

2012). The brain was removed and fixed overnight in 4% PFA, and then transferred 164

to 30% sucrose in PBS. Dorsal and ventral hippocampal slices were prepared on a 165

cryostat and stored in cryoprotectant (30% sucrose, 30% ethylene glycol, 1% 166

polyvinyl pyrrolidone, 0.05M sodium phosphate buffer) until processing for 167

immunohistochemistry. Dorsal and ventral slices were briefly rinsed in PBS buffer, 168

and incubated in 0.1% TritonX-100 for 30 min. Subsequently, slices were blocked in 169

PBS solution containing 5 % normal goat serum, 0.03% TritonX-100 for 1hr, and then 170

incubated in primary antibody diluted in blocking solution overnight at 4oC. Slices 171

were rinsed in PBS buffer, and then incubated in secondary antibody for 1hr at room 172

temperature. Primary antibodies in this study were used as follows; rabbit anti-Kir3.1 173

(1:200, Alomone Labs), rabbit anti-Kir 3.2 (1:200, Alomone Labs), rabbit anti-A1 174

adenosine receptors (1:200, Alomone Labs), and mouse anti-MAP2 (1:1000, Sigma-175

Aldrich). 176

Data analysis. Input resistance (Rin) was determined by the slope of the linear fit of 177

the voltage-current plot from the steady-state voltage response to the injected 178

current ranging from +10- to -150-pA (750 ms). FI curves were determined by 179

plotting the number of action potentials against injected current (30-300 pA, for 750 180

ms in 30 pA intervals). 181

Statistical analyses. All data are expressed as mean ± SEM. Statistical 182

9

comparisons were performed using ANOVA (one-factor or two-factor) followed by 183

Bonferrori post hoc test or Student’s t test (paired or unpaired) with GraphPad Prism 184

software. *p< 0.05 was considered as statistically significant. 185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

10

Results 201

Contribution of Ba2+-sensitive conductance to intrinsic membrane properties 202

of dorsal CA1 neurons 203

Our initial goal in this study was to explore the resting conductances that 204

contribute to the intrinsic membrane properties of CA1 pyramidal neurons from the 205

dorsal and ventral hippocampus. Although functional differences in Ih and distinct α-206

subunit expression profiles of HCN1 and HCN2 were reported between dorsal and 207

ventral CA1 neurons (Dougherty et al. 2013), Ih-related properties cannot fully 208

account for the differences in intrinsic membrane properties between dorsal and 209

ventral CA1 neurons (Dougherty et al. 2012). We therefore hypothesized that dorsal 210

neurons might have a large resting potassium (K+) conductance contributed by 211

inward rectifier-like K+ channels (Kir). A relatively low concentration of barium (50 μM 212

Ba2+) is known to block Kir channels, but not many other K+ currents (Armstrong and 213

Taylor 1980; Quayle et al. 1988; Sah 1996; Wang et al. 1998; Jin et al. 1999; Chen 214

and Johnston 2005; Day et al. 2005). Because ZD7288 caused a gradual 215

depolarization 10-15 min after the hyperpolarization of Vm (not shown), we first 216

applied Ba2+ and then added of ZD7288 for 4 to 5 min (Dougherty et al. 2013) before 217

switching back to the barium-containing aCSF. Membrane potential (Vm) and input 218

resistance (Rin) at RMP were monitored during baseline, application of 50 μM Ba2+ 219

and addition of 10 μM ZD7288 application in dorsal (Fig.1B) and ventral (Fig.1F) 220

CA1 pyramidal neurons. 50 μM Ba2+ significantly depolarized the membrane 221

potential in dorsal neurons (Fig.1, A and C; table 1), but not in ventral neurons (Fig.1, 222

E and G; table 1). Addition of 10 μM ZD7288 produced a significant hyperpolarization 223

11

from baseline in ventral neurons (Fig.1, E and G; table 1), but not in dorsal neurons 224

(Fig.1, A and C; table 1). Rin at RMP for dorsal neurons was significantly increased 225

(Fig.1, A and D; table 1), but not for ventral neurons (Fig.1 E and H; table 1) after 50 226

μM Ba2+ application. Addition of 10 μM ZD7288 led to a significant increase in Rin in 227

dorsal (Fig.1 A and D; table 1) and ventral (Fig.1 E and H; table 1) neurons. For 228

proper comparison between groups, Rin was measured at the same membrane 229

potential. At a common membrane potential (-73 mV), 50 μM Ba2+ led to a significant 230

increase in Rin for dorsal neurons (Fig.1I; table 2), but not for ventral neurons (Fig.1J; 231

table 2). Vm (Fig.1K; table 2) and Rin (Fig.1L; table 2) were significantly different 232

between dorsal and ventral neurons. This difference was absent when either IRKs 233

alone were blocked or were blocked in combination with Ih (Fig.1, K and L; table 2). 234

Because we observed significantly changes in voltage response (Vm) in response to 235

the injected current from RMP and at -73 mV in the presence of Ba2+ in dorsal 236

neurons (Fig.1 A and I), but not in ventral neurons (Fig.1 E and J), steady-state 237

voltage-current (V-I) curves in control, Ba2+, and the difference between the two were 238

plotted to better illustrate the Ba2+-sensitive conductance in the hyperpolarizing 239

direction (Figs. 1M, N & O). The V-I curves before and after Ba2+ application were 240

significantly different in dorsal neurons (Fig. 1M), but not ventral neurons (Fig. 1N) 241

from RMP and at -73 mV, suggesting a greater Ba2+-sensitive conductance in dorsal 242

neurons. 243

244

Comparison of basal GIRK activity between ventral and dorsal CA1 neurons 245

Given the observation that significant changes in Vm and Rin (at RMP and at a 246

common membrane potential) were observed following blockade of a Ba2+-sensitive 247

12

conductance in dorsal neurons (Fig.1), we next examined which ion conductance is 248

responsible for the resting Ba2+-sensitive conductance. Inwardly rectifying potassium 249

currents (Kir) are blocked by low concentrations of Ba2+ (Standen and Stanfield 1978; 250

Williams et al. 1988; Gerber et al. 1991; Chen and Johnston 2005) , while tertiapin-Q 251

is a specific GIRK channel blocker at nanomolar concentration (Jin et al. 1999). 252

Because GIRK1 (Kir3.1) and GIRK2 (Kir3.2) heterotetrameric channels are 253

predominantly expressed in the brain (Liao et al. 1996), we performed an 254

immunohistochemical staining with antibodies against GIRK1 and GIRK2 in the 255

dorsal and ventral hippocampus. To account for the longer radial axis in the CA1 256

region of ventral hippocampus (Dougherty et al. 2012), the quantification of GIRK1 257

and GIRK2 protein expression was normalized by distance (Fig. 2, A and C). 258

Expression of GIRK1 and GIRK2 was higher in the somatic and distal dendritic 259

regions of dorsal CA1 compared to ventral CA1 (Fig.2, B and D; p

13

that there were significant differences in the changes in Vm (Fig.2, F and G; table 3) 271

and Rin (Fig.2, F and G; table 3) between dorsal and ventral neurons after either 0.3 272

μM tertiapin-Q or 2 μM tertiapin-Q application. Based on these results, all 273

subsequent tertiapin-Q experiments were performed using 0.3 µM. 274

275

The Ba2+-sensitive conductance in dorsal CA1 neurons in mostly mediated by 276

GIRKs 277

Because tertiapin-Q changed RMP and Rin in dorsal but not ventral CA1 278

neurons (Fig.2), we next sought to determine whether the Ba2+-sensitive 279

conductance in dorsal neurons was mediated by GIRKs. Vm and Rin (at RMP) were 280

monitored during successive 0.3 μM tertiapin-Q and 50 μM Ba2+ application in dorsal 281

(Fig.3B) and ventral (Fig.3F) neurons. Application of tertiapin-Q significantly 282

depolarized (Fig.3, A and C; table 4) and increased Rin (Fig.3, A and D; table 4) of 283

dorsal neurons. In contrast, tertiapin-Q had no significant effect on ventral neurons 284

(Fig.3, G and H; table 4). There was no further change in either dorsal or ventral 285

neurons when 50 μM Ba2+ was subsequently added to the bath (dorsal: Fig. 3, C and 286

D, ventral: Fig.3, G and H; table 4). Blockade of both GIRK and Ba2+-sensitive 287

conductances led to a significant depolarization (dorsal: Fig.3, A and C; ventral: Fig.3 288

E and G; table 4) and an increased Rin (dorsal: Fig.3, A and D; ventral: Fig.3, E and 289

H; table 4) in dorsal and ventral neurons. At a common membrane potential (-73 mV), 290

a change in Rin for dorsal neurons was significantly increased in the presence of 0.3 291

μM tertiapin-Q (Fig.3I; table 5), but not for ventral neurons (Fig.3J; table 5). When 292

subsequent 50 μM Ba2+ was applied after blockade of GIRK conductance, there was 293

14

no further change in Rin in dorsal (Fig.3I; table 5) and ventral (Fig.3J; table 5) 294

neurons. When both GIRK and Ba2+-sensitive conductances were blocked, Rin for 295

dorsal neurons (Fig.3I; table 5) was significantly increased, but not for ventral 296

neurons (Fig.3J; table 5). Action potentials evoked in response to depolarizing 297

current steps (30-300 pA in 30 pA increments for 750 ms) were determined during 298

successive 0.3 μM tertiapin-Q and 50 μM Ba2+ application at a common membrane 299

potential (-73 mV) in dorsal (Fig.3K) and ventral (Fig.3L) neurons. As expected, 300

tertiapin-Q application significantly increased action potential firing in dorsal neurons 301

(Fig.3K), but had no effect on firing in ventral neurons (Fig.3L) compared to baseline. 302

Addition of 50 μM Ba2+ had no further effects on firing in either dorsal (Fig.3K) or 303

ventral (Fig.3L) neurons. 304

305

Comparison of basal GABAB receptor activity in dorsal and ventral CA1 306

neurons 307

The data presented thus far suggest that there was more basal GIRK activity in 308

dorsal vs ventral neurons (Fig.2 and Fig.3). We next examined whether there was 309

more basal activation of neurotransmitter receptors known to couple with GIRK 310

channels in dorsal vs ventral neurons. In CA1 neurons, the activation of GABAB 311

receptors, A1 adenosine receptors, and 5-HT1A serotonin receptors results in 312

hyperpolarization of resting membrane potential by GIRK channels (Luscher et al. 313

1997). Thus, it is possible that there is more activity in any one of these receptor 314

systems in dorsal compared with ventral neurons. We first examined RMP and Rin in 315

dorsal and ventral neurons during bath application of a potent and selective GABAB 316

15

receptors antagonist, CGP52432 (5 μM) (Lanza et al. 1993). There was no 317

significant change in Vm and Rin (at RMP) up to 30-35 min after wash-in of 318

CGP55845 in either dorsal (Fig.4, B and C; table 6) or ventral neurons (Fig.4, E and 319

F; table 6). Because we did not observe any changes in Vm and Rin after blockade of 320

GABAB receptors in dorsal or ventral neurons, we next applied a GABAB receptor 321

agonist, baclofen, to test for the presence of functional GABAB receptors. Vm and Rin 322

(at RMP) were monitored during 100 μM baclofen application in dorsal (Fig.4G) and 323

ventral (Fig.4J) neurons. Baclofen significantly changed Vm (dorsal: Fig.4H; ventral: 324

Fig.4K; table 6) and Rin (dorsal: Fig.4I; ventral: Fig.4L; table 6) in both dorsal and 325

ventral neurons. The magnitude of changes in Vm (dorsal: -7.6±0.8 mV vs. ventral: -326

8.2±0.6 mV; p>0.05; Fig.4M; n=6) and Rin at RMP (dorsal: -28.3±5.2 % vs. ventral: -327

23.2±3.6 %; p>0.05; Fig.4N; n=6) was no significantly different after activation of 328

GABAB receptors. 329

330

Effect of A1 adenosine receptor-mediated activation on GIRK conductance in 331

dorsal CA1 neurons 332

We next examined whether A1 adenosine receptors (A1ARs) are involved in 333

the activation of the resting GIRK conductance in these neurons. 1,3-Dipropyl-8-334

cyclopentylxanthine (DPCPX) is a selective A1 adenosine receptor antagonist 335

(Alzheimer et al. 1989). Vm and Rin (at RMP) were monitored during 100 nM 336

DPCPX wash-in in dorsal (Fig.5A) and ventral (Fig.5E) neurons. Dorsal neurons had 337

significantly depolarized membrane potential (Fig.5, B and C; table 7) and an 338

increased Rin at RMP (Fig.5, B and D; table 7) after blockade of A1 adenosine 339

16

receptors, whereas no changes were observed in ventral neurons (Fig.5, G and H; 340

table 7). The magnitude of changes in Vm (dorsal 4.7±0.5 mV vs. ventral 1.1±0.5 mV; 341

p

17

and B; n=3). Because there was a significant difference in A1AR protein expression 363

in somatic region of dorsal and ventral CA1, we thus examined the effects by 2’ME-364

CCPA, a selective A1 receptor agonist. Vm and Rin (at RMP) were monitored during 1 365

μM 2’ME-CCPA application in the dorsal (Fig.7C) and ventral (Fig.7G) CA1 neurons. 366

1 μM 2’ME-CCPA significantly hyperpolarized the membrane potential (Fig.7, D and 367

E; table 9) and decreased Rin at RMP (Fig.7, D and F; table 9) in dorsal but not 368

ventral neurons (Fig.7, H-J; table 9). The magnitude of changes in Vm (dorsal -369

3.9±0.3 mV vs. ventral -0.3±0.3 mV; p

18

Discussion 384

In this study, we examined the resting conductances that contribute to the 385

different intrinsic membrane properties in dorsal and ventral CA1 neurons. We found 386

that there was more Ba2+-sensitive conductance in dorsal than in ventral CA1 387

neurons. Furthermore, Ba2+-dependent changes in Vm and Rin were due to block of 388

GIRK channels. We also found that this resting GIRK conductance was mediated by 389

A1 adenosine receptors in dorsal CA1 neurons, but not in ventral CA1 neurons. 390

Until recently, the intrinsic electrophysiological properties of CA1 pyramidal 391

neurons along the longitudinal hippocampal axis were assumed to be uniform. 392

However, we and others have recently reported significant differences in the intrinsic 393

electrophysiological properties between dorsal and ventral neurons (Dougherty et al. 394

2012; Marcelin et al. 2012a). Although differences in the expression of some voltage-395

gated ion channels between dorsal and ventral CA1 neurons have been reported 396

(Marcelin et al. 2012b; Dougherty et al. 2013), the ionic mechanisms underlying the 397

different intrinsic properties remain unclear. Because dorsal CA1 neurons showed a 398

lower somatic Rin compared with ventral CA1 neurons, this could be due to more 399

dendritic surface area for dorsal than ventral neurons (Dougherty et al. 2013), 400

although Marcelin et al., reported that somatic and dendritic capacitances were not 401

different between dorsal and ventral neurons (Marcelin et al. 2012a; Marcelin et al. 402

2012b). A more hyperpolarized resting membrane potential in dorsal neurons 403

compared to ventral neurons, however, could be due to dorsal CA1 neurons 404

possessing a relatively larger resting potassium (K+) conductance, which could also 405

contribute to a decrease in Rin. Indeed, low micromolar application of Ba2+ that 406

19

blocks inwardly rectifying K+ (KIR) channels (Chen and Johnston 2005) had 407

significant effects of Vm and Rin (at RMP and at a common membrane potential -73 408

mV) for dorsal neurons, but not for ventral neurons. In addition, the changes in V-I 409

curves were greater for dorsal neurons than ventral neurons. It should be noted that 410

higher concentrations of extracellular barium (Ba2+, > 200 µM) block A-type 411

potassium channels (Gasparini et al. 2007), whereas low concentrations of barium 412

(Ba2+, < 200 µM) are known to block inwardly rectifying potassium currents (Kir) 413

(Standen and Stanfield 1978; Williams et al. 1988; Gerber et al. 1991; Chen and 414

Johnston 2005). Furthermore, when 10 μM ZD7288 was applied after blockade of 415

Ba2+-sensitive conductance, we observed that the change in Vm for ventral neurons 416

was larger than dorsal neurons, consistent with more Ih in ventral CA1 neurons 417

(Dougherty et al. 2013). Interestingly, the absolute values of Vm and Rin were similar 418

between dorsal and ventral neurons after blockade of Ba2+-sensitive conductance 419

and Ih, suggesting a larger Ba2+-sensitive conductance in dorsal neurons compared 420

to ventral neurons and more Ih in ventral CA1 neurons compared to dorsal CA1 421

neurons. Both Ba2+-sensitive conductance and Ih might account for these differences 422

on intrinsic membrane properties between dorsal and ventral CA1 neurons. 423

Two classes of inwardly rectifying potassium channels, Kir channels, are highly 424

expressed in hippocampal neurons: IRK channels (Kir2.x), and GIRK channels 425

(Kir3.x) (Karschin et al. 1996; Luscher et al. 1997; Chen and Johnston 2005). Both 426

IRK and GIRK channels undergo voltage-dependent intracellular Mg2+ block, which 427

imparts a rapid hyperpolarization-activated gating mechanism to these channels, and 428

they become blocked and unblocked with depolarization and hyperpolarization, 429

20

respectively. Because dorsal CA1 neurons showed a larger response to a relatively 430

low concentration of Ba2+ than ventral CA1 neurons, we used a specific GIRK 431

channels blocker tertiapin-Q (Jin et al. 1999) in an attempt to identify molecular 432

targets responsible for the resting Ba2+-sensitive conductance. Interestingly, we 433

found that dorsal neurons showed significant changes in Vm and Rin (at RMP and at a 434

common membrane potential -73 mV), but not ventral neurons after tertiapin-Q 435

application, indicating that there was more basal GIRK conductance. Furthermore, 436

when action potentials were elicited from a somatic current injection at a common 437

membrane potential (-73 mV), tertiapin-Q application significantly increased action 438

potential firing in dorsal neurons, but not in ventral neurons. In agreement with these 439

results, the protein expression of GIRK1 and GIRK2 was significantly enriched in 440

somatic and dendritic CA1 region of dorsal hippocampus compared to ventral 441

hippocampus. The somatic and dendritic protein expression of GIRK subunits in the 442

hippocampus have also been observed by other groups. (Liao et al. 1996). Most 443

interestingly, successive tertiapin-Q and Ba2+ application revealed that a greater 444

Ba2+-sensitive conductance for dorsal CA1 neurons stems from the GIRK 445

conductance. 446

Hippocampal G protein-coupled inwardly rectifying K+ (GIRK) channels are 447

activated by a number of neurotransmitters systems, including GABAB receptors, 448

serotonergic 5HT1A receptors, and adenosine A1 receptors (Andrade et al. 1986; 449

North et al. 1987; North 1989; Luscher et al. 1997). Given that ventral neurons have 450

a weaker GABAergic fast synaptic inhibition than dorsal CA1 neurons 451

(Papatheodoropoulos et al. 2002), it is possible that there is more GABAB-mediated 452

21

slow tonic inhibition, which could activate more GIRK conductance in dorsal neurons 453

compared to ventral neurons. We observed, however, that there are similar changes 454

in Vm and Rin (at RMP) after activation of GABAB receptors in dorsal and ventral 455

neurons, suggesting that coupling of GABAB receptors to GIRK channels might be 456

similar. Lee et al. reported that dorsal CA1 region had a greater A1 adenosine 457

receptor density, resulting in the suppression of extracellular excitatory postsynaptic 458

potentials (EPSPs) in the presence of N6-cyclohexyladenosine, a selective A1 459

adenosine receptor agonist, compared to ventral CA1 region (Lee et al. 1983). In 460

accordance with these results, somatic whole-cell current-clamp recordings revealed 461

that dorsal CA1 neurons showed significant changes in Vm and Rin in the presence of 462

A1 adenosine receptor agonist or antagonist, but not in ventral neurons, suggesting 463

again that A1 adenosine receptors are more heavily expressed in dorsal CA1 464

neurons. Consistent with these results, the protein expression of A1 adenosine 465

receptor was highly present in the somatic CA1 region of dorsal hippocampus 466

compared with ventral hippocampus. Interestingly, when A1 adenosine receptors 467

were blocked in dorsal neurons, the absolute values of Vm and Rin at RMP were 468

similar to ventral neurons. Furthermore, successive Ba2+ and DPCPX application 469

revealed that the activation of GIRK conductance is mediated by A1 adenosine 470

receptors in dorsal CA1 neurons. These results suggest that A1 adenosine receptor-471

mediated GIRK conductance might contribute to intrinsic membrane properties in 472

dorsal CA1 neurons. 473

Because GIRK and the adenosinergic system play important roles in the 474

pathophysiology of diseases such as epilepsy, depression, and anxiety, our results 475

22

suggest that dorsal CA1 neurons/region (predominantly expressing A1 adenosine 476

receptors-mediated GIRK conductance) might be important for chronic neurological 477

disorders compared to ventral hippocampus. GIRK channels play an important role 478

in regulating neuronal excitability (Signorini et al. 1997). The deletion of Kir3.2 479

(GIRK2) channels led to an increase in susceptibility to a convulsant agent and 480

showed sporadic seizures (Signorini et al. 1997). In the present study, we found that 481

there is a more basal GIRK activity (i.e. physiological responses to specific GIRK 482

channels blocker at RMP) in dorsal CA1 neurons compared to ventral CA1 neurons, 483

which might contribute to different susceptibility to a convulsant agent within the 484

hippocampus. Adenosinergic systems have been implicated in anxiety and 485

depression. In preclinical studies, activation of adenosine A1 receptors led to 486

antidepressant- and anxiolytic-like behaviors, whereas blockade of A1 adenosine 487

receptors resulted in an increase in anxiety (Kaster et al. 2004; Prediger et al. 2006; 488

Maximino et al. 2011). In a clinical study, large amount of caffeine, a non-selective A1 489

adenosine receptor antagonist, are known to cause anxiety and depression in both 490

normal and vulnerable subjects (Broderick and Benjamin 2004). Ligand-stimulated 491

Gi/o protein-coupled receptors such as A1 adenosine receptors induce an inhibition of 492

adenylyl cyclase, which results in a decrease in cAMP concentration (van Calker et 493

al. 1978). Because HCN channels are modulated by cAMP, resulting in an increase 494

in Ih (Wainger et al. 2001), stimulation of A1 adenosine receptors-induced decrease 495

in cAMP levels might be interacting with HCN channels. 496

Summary 497

In summary, we have investigated the resting conductances that contribute to 498

23

intrinsic membrane properties of CA1 pyramidal neurons from the dorsal and ventral 499

hippocampus. We found that dorsal CA1 neurons possess more basal activity of 500

GIRK conductance, which was activated by A1 adenosine receptors. Our results 501

suggest that A1 adenosine receptor-mediated GIRK conductance contributes to the 502

different intrinsic membrane properties in dorsal and ventral CA1 pyramidal neurons. 503

504

505

506

507

508

509

510

511

512

513

514

515

516

517

518

519

520

521

24

Figure legends 522

Fig.1. Dorsal neurons show greater response to Ba2+ compared with ventral CA1 523

neurons. (A and E) Representative voltage responses with step current commands 524

at RMP. (B and F) Time courses of changes in Vm and Rin during successive Ba2+ 525

(50 μM) and ZD7288 (10 μM) wash-in experiments in dorsal (B) and ventral (F) CA1 526

pyramidal neurons. (C and D) Dorsal CA1 pyramidal neurons showed a significantly 527

depolarized Vm (C) and increased Rin (D) in the presence of Ba2+. Subsequent 528

ZD7288 application resulted in a significantly hyperpolarized Vm (C) a further 529

increase Rin (D). (G and H) Ventral CA1 pyramidal neurons showed no significant 530

changes in Vm (G) and Rin (H) in the presence of Ba2+. Addition of ZD7288 led to a 531

significantly hyperpolarized Vm (G) and an increased Rin (H). (I and J) The left 532

displays representative voltage responses with step current commands at -73 mV. 533

Dorsal CA1 neurons showed a significantly increased Rin, but not ventral CA1 534

neurons at a common membrane potential (-73 mV) in the presence of Ba2+. (K and 535

L) Dorsal CA1 neurons showed a more hyperpolarized RMP (K) and a lowered Rin at 536

RMP (L) compared to ventral CA1 neurons. There is no difference in Vm (K) and Rin 537

(L) between dorsal and ventral neurons following successive Ba2+ and ZD7288 538

application. (M and N) Dorsal neurons showed significant changes in V-I curve 539

before and after 50 μM Ba2+ application (M), but not ventral neurons (N). (O) The 540

differences in V-I curves before and after 50 μM Ba2+ application are indicated. (p < 541

0.05 and #p < 0.05 in dorsal vs. ventral groups). 542

Fig.2. Dorsal CA1 neurons show a more basal GIRK activity than ventral CA1 543

neurons. (A and C) Representative dorsal and ventral hippocampal slices 544

25

immunolabeled with antibodies against Kir 3.1 (A; GIRK1) and Kir 3.2 (C; GIRK2). 545

Yellow boxes depict the region of the slice used for quantification of signal intensity 546

(B and D) Quantification of Kir3.1 (B) and Kir3.2 (D) protein expression from 547

perisomatic region to the distal dendritic region of CA1 from dorsal and ventral 548

hippocampus. The protein expression of Kir3.1 (B) and Kir 3.2 (D) was significantly 549

increased in somatic and distal dendritic region of CA1 from dorsal hippocampus 550

compared to ventral hippocampus. Vertical blue bars highlight part of the CA1 radial 551

axis that show significant differences in Kir 3.1 (B) and Kir 3.2 (D) protein expression 552

between dorsal and ventral CA1 region. (E) Representative voltage responses with 553

step current commands at RMP. (F and G) 0.3 μM tertiapin-Q significantly changed 554

Vm and Rin, in dorsal but not ventral neurons. 2 μM tertiapin-Q produced similar 555

effects on Vm and Rin compared to 0.3 μM tertiapin-Q wash-in in dorsal and ventral 556

neurons. Dorsal CA1 neurons showed significant changes in Vm and Rin in the 557

presence of 0.3 μM or 2 μM tertiapin-Q compared to ventral CA1 neurons. *p < 0.05 558

and #p < 0.05 vs. ventral group. 559

560

Fig.3. Ba2+-sensitive conductance-mediated changes in Vm and Rin stem from GIRK 561

conductance in dorsal CA1 neurons. (A and E) Representative voltage responses 562

with step current commands at RMP are shown. (B and F) Time courses of changes 563

in Vm and Rin during successive 0.3 μM tertiapin and 50 μM Ba2+ application in dorsal 564

(B) and ventral (F) CA1 pyramidal neurons. (C and D) Dorsal CA1 neurons showed a 565

significantly depolarized Vm (C) and an increased Rin (D) following bath application of 566

tertiapin-Q (0.3 μM). Subsequent blockade of Ba2+-sensitive conductance showed no 567

26

significant changes in Vm (C) and Rin (D) in dorsal CA1 neurons. (G and H) Ventral 568

CA1 neurons showed no significant changes in Vm and Rin in the presence of 569

tertiapin-Q (0.3 μM). Further changes in Vm and Rin were not observed following 570

blockade of Ba2+-sensitive conductance in ventral neurons. Blockade of tertiapin-Q 571

and Ba2+-sensitive conductance produced a significantly depolarized Vm and 572

increased Rin compared to baseline in ventral neurons. (I and J) The left displays 573

representative voltage responses with step current commands at -73 mV. Dorsal 574

CA1 neurons showed a significantly increased Rin (I), but not ventral CA1 neurons 575

(J) at a common membrane potential (-73 mV) following successive tertiapin-Q and 576

Ba2+ application. (K and L) Representative membrane potential responses with 577

depolarizing current steps (300 pA for 750 ms) at a common membrane potential (-578

73 mV). (K) Dorsal neurons showed significant increases in action potential firing 579

after either tertiapin-Q or tertiapin-Q+Ba2+ application, but not ventral neurons (L). *p 580

< 0.05. 581

582

Fig.4. Similar changes in Vm and Rin in the presence of GABAB receptor antagonist 583

and agonist in dorsal and ventral CA1 neurons. (A and D) Time courses of changes 584

in Vm and Rin during CGP55845 (5 μM) wash-in experiments in dorsal (A) and ventral 585

(D) neurons. Dorsal (B and C) and ventral (E and F) neurons showed no changes in 586

Vm and Rin in the presence of CGP55845. (G and J) Time courses of Vm and Rin in 587

the presence of baclofen in dorsal (G) and ventral (J) neurons. Dorsal (H and I) and 588

ventral (K and L) CA1 neurons showed a significantly hyperpolarized Vm and a 589

decreased Rin following activation of GABAB receptors. The changes in Vm and Rin 590

27

were not significantly different in dorsal and ventral CA1 neurons in the presence of 591

CGP55845 or baclofen. *p < 0.05. 592

593

Fig.5. Dorsal CA1 neurons showed greater responses to A1 adenosine receptor 594

antagonist compared to ventral CA1 neurons. (A and E) Time course of changes in 595

Vm and Rin during DPCPX (100 nM) wash-in experiments in dorsal (A) and ventral 596

(E) neurons. (B and F) Representative voltage responses with step current 597

commands at RMP are shown. (C and D) 100 nM DPCPX caused a significant 598

depolarization in dorsal neurons (C) and an increased Rin (D). (G and H) Ventral 599

neurons showed no changes in Vm (G) and Rin (H) after DPCPX (100 nM) wash-in. (I 600

and J) The changes in Vm (I) and Rin (J) after 100 nM DPCPX wash-in were greater 601

for dorsal neurons than ventral neurons. *p < 0.05 and #p < 0.05 vs. ventral group. 602

603

Fig.6. Dorsal CA1 neurons have an A1 adenosine receptor-mediated GIRK 604

conductance. (A) Representative voltage responses with step current commands at 605

RMP. (B) Time courses of changes in Vm and Rin during successive 50 μM Ba2+ and 606

100 nM DPCPX in dorsal CA1 neurons. Dorsal CA1 neurons showed a significantly 607

depolarized Vm (C) and an increased Rin (D) at RMP in the presence of Ba2+. 608

Subsequent bath application of DPCPX had no further effects on Vm (C) and Rin (D) 609

at RMP in dorsal neurons. (E) The left displays representative voltage responses 610

with step current commands at -73 mV. Successive Ba2+ and DPCPX application led 611

to a significantly increased Rin at a common membrane potential (-73 mV). *p < 0.05. 612

613

28

Fig.7. A1 adenosine receptors and their physiological responses predominantly exist 614

in dorsal, but not in ventral CA1 neurons. (A) Representative dorsal and ventral 615

hippocampal slices immunolabeled with antibodies against A1 adenosine receptors. 616

Yellow boxes depict the region of the slice used for quantification of signal intensity. 617

(B) Quantification of A1ARs protein expression from perisomatic region to the distal 618

dendritic region of CA1 from dorsal and ventral hippocampus. The protein 619

expression of A1 adenosine receptors was highly present in somatic CA1 region of 620

dorsal hippocampus, but not in ventral CA1 region. The blue shade indicates a 621

significant difference in A1ARs protein expression between dorsal and ventral CA1 622

region. (C and G) Time courses of Vm and Rin during 2’ME-CCPA (1 μM) wash-in 623

experiments in dorsal (C) and ventral (G) CA1 neurons. (D and H) Representative 624

voltage responses with step current commands at RMP. (E and F) Dorsal CA1 625

neurons showed a significant hyperpolarization (E) and a decreased Rin at RMP (F) 626

in the presence of 2’ME-CCPA (1 μM). (I and J) Ventral CA1 neurons showed no 627

changes in Vm (I) and Rin at RMP (J) after 2’ME-CCPA (1 μM) wash-in. (K and L) The 628

changes in Vm (K) and Rin (L) after 2’ME-CCPA (1 μM) application were greater for 629

dorsal CA1 neurons than ventral CA1 neurons. (M and N) Activation of A1 adenosine 630

receptors resulted in a significantly decreased Rin at a common membrane potential 631

(-73 mV) in dorsal (M), but not in ventral (N) CA1 neurons. *p < 0.05 and #p < 0.05 vs. 632

ventral group. 633

634

635

636

29

Table 1. Subthreshold properties in successive 50 μM Ba2+ and 10 μM ZD7288 637

application. (Related to Fig. 1) 638

Dorsal CA1 Baseline 50 μM Ba2+

wash-in

10 μM ZD7288

50 μM Ba2+ wash-in

Vm (mV) -69.7±0.5 (n=7) -64.4±1.0 (n=7)* -71.0±0.7 (n=7)

Rin (MΩ) 60.8±4.1 (n=7) 85.7±5.6 (n=7)* 166.4±9.0 (n=7)*

Ventral CA1 Baseline 50 μM Ba2+

wash-in

10 μM ZD7288

50 μM Ba2+ wash-in

Vm (mV) -66.5±0.7 (n=7) -64.8±1.0 (n=7) -71.9±1.5 (n=7)*

Rin (MΩ) 73.7±8.3 (n=7) 84.8±8.4 (n=7) 162.1±12.7 (n=7)*

639

Vm: membrane potential. Rin: Steady-state input resistance. * was considered as 640

statistically significant (P

30

Table 2. Change in Rin at a common membrane potential (-73 mV) in dorsal and 655

ventral CA1 neurons after 50 μM Ba2+ wash-in experiments. (Related to Fig. 1) 656

Dorsal CA1 Baseline 50 μM Ba2+ Ventral CA1 Baseline 50 μM Ba2+

Rin (MΩ)

at -73 mV

60.0±3.5

(n=6) 72.1±4.2*

(n=6)

Rin (MΩ)

at -73 mV

69.1±5.1

(n=7)

72.6±7.2

(n=7)

657

Rin: Steady-state input resistance. 658

* was considered as statistically significant (P

31

Table 3. Dose-response tertiapin-Q experiment in dorsal and ventral CA1 neurons. 676

(Related to Fig. 2) 677

0.03 μM tertiapin-Q 0.3 μM tertiapin-Q 2 μM tertiapin-Q

Dorsal CA1 Baseline Tertiapin Baseline Tertiapin Baseline Tertiapin

Vm

(mV)

-69.9±1.0

(n=7)

-67.8±1.1

(n=7)

-69.0±0.5

(n=7) -64.4±0.5*

(n=7)

-69.4±0.9

(n=6) -65.4±0.9*

(n=6)

Rin

(MΩ)

60.2±2.8

(n=7)

70.5±3.5

(n=7)

62.9±1.2

(n=7) 82.6±2.2*

(n=7)

64.5±3.9

(n=6) 85.6±4.5*

(n=6)

0.03 μM tertiapin-Q 0.3 μM tertiapin-Q 2 μM tertiapin-Q

Ventral CA1 Baseline Tertiapin Baseline Tertiapin Baseline Tertiapin

Vm

(mV)

-65.3±1.0

(n=6)

-64.7±1.4

(n=6)

-64.9±0.5

(n=6)

-63.0±0.7

(n=6)

-64.1±0.8

(n=6)

-62.6±0.9

(n=6)

Rin

(MΩ)

82.5±4.8

(n=6)

87.2±5.8

(n=6)

87.3±3.0

(n=6)

95.8±2.7

(n=6)

85.0±4.8

(n=6)

89.1±3.9

(n=6)

678



32

Table 4. Subthreshold properties in subsequent tertiapin-Q and barium wash-in 690

experiments. (Related to Fig. 3) 691

Dorsal CA1 Baseline 0.3 μM Tertiapin-Q

wash-in

0.3 μM Tertiapin-Q

50 μM Ba2+ wash-in

Vm (mV) -68.0±0.6 (n=7) -63.3±0.7 (n=7)* -61.2±1.1 (n=7)*

Rin (MΩ) 62.4±1.5 (n=7) 80.5±3.6 (n=7)* 92.7±4.7 (n=7)*

Ventral CA1 Baseline 0.3 μM Tertiapin-Q

wash-in

0.3 μM Tertiapin-Q

50 μM Ba2+ wash-in

Vm (mV) -64.6±0.5 (n=5) -62.9±0.6 (n=5) -61.4±0.6 (n=5)*

Rin (MΩ) 86.9±3.0 (n=5) 95.0±3.4 (n=5) 102±1.9 (n=5)*

692



33

Table 5. Change in Rin at a common membrane potential (-73 mV) in dorsal and 705

ventral CA1 neurons in subsequent tertiapin-Q and barium wash-in experiments. 706


Dorsal CA1 Baseline 0.3 μM Tertiapin-Q

wash-in

0.3 μM Tertiapin-Q

50 μM Ba2+ wash-in

Rin (MΩ) at -73 mV 52.7±1.3 (n=6) 60.5±1.6 (n=6)* 62.8±1.7 (n=6)*

Ventral CA1 Baseline 0.3 μM Tertiapin-Q

wash-in

0.3 μM Tertiapin-Q

50 μM Ba2+ wash-in

Rin (MΩ) at -73 mV 72.5±1.6 (n=5) 74.1±3.4 (n=5) 78.1±4.1 (n=5)

708

Rin: Steady-state input resistance. * was considered as statistically significant 709

(P

34

Table 6. Subthreshold membrane properties in 5 μM CGP55845 or 100 μM baclofen 721

wash-in experiments. (Related to Fig. 4) 722

Dorsal CA1 Baseline 5 μM

CGP55845

Ventral CA1 Baseline 5 μM

CGP55845

Vm (mV) -70.0±0.8

(n=4)

-70.2±0.9

(n=4)

Vm (mV) -64.7±0.4

(n=4)

-64.2±0.5

(n=4)

Rin (MΩ)

62.7±2.7

(n=4)

64.5±3.3

(n=4)

Rin (MΩ)

86.4±4.0

(n=4)

89.6±4.0

(n=4)

Dorsal CA1 Baseline 100 μM

baclofen

Ventral CA1 Baseline 100 μM

baclofen

Vm (mV) -69.4±1.6

(n=6)

-76.6±1.0*

(n=6)

Vm (mV) -66.1±0.7

(n=6)

-74.4±1.0*

(n=6)

Rin (MΩ)

66.6±5.7

(n=6)

46.5±2.1*

(n=6)

Rin (MΩ)

83.2±5.7

(n=6)

63.1±3.4*

(n=6)

723



35

Table 7. Subthreshold membrane properties in 100 nM DPCPX wash-in experiments. 733


Dorsal CA1 Baseline DPCPX Ventral CA1 Baseline DPCPX

RMP (mV) -69.0±0.7

(n=7) -64.2±0.5*

(n=7)

RMP (mV) -64.7±0.7

(n=6)

-63.6±0.8

(n=6)

Rin (MΩ)

66.4±4.9

(n=7) 89.3±5.0*

(n=7)

Rin (MΩ)

84.8±7.0

(n=6)

87.2±7.2

(n=6)

735



36

Table 8. Subthreshold membrane properties in successive Ba2+ and DPCPX wash-in 749

experiments. (Related to Fig. 6) 750

Dorsal CA1 Baseline 50 μM Ba2+

wash-in

100 nM DPCPX

50 μM Ba2+ wash-in

Vm (mV) -68.5±0.6 (n=7) -62.8±0.7 (n=7)* -62.4±0.6 (n=7)*

Rin (MΩ) 69.8±2.5 (n=7) 99.7±4.9 (n=7) 103.7±5.0 (n=7)*

Rin (MΩ) at -73 mV 58.8±2.4 (n=7) 65.8±2.2 (n=7) 69.1±2.6 (n=7)*

751



37

Table 9. Subthreshold membrane properties in 2’ME-CCPA wash-in experiments. 766


Dorsal CA1 Baseline 2’ME-CCPA Ventral CA1 Baseline 2’ME-CCPA

Vm (mV) -68.5±0.8

(n=6) -72.3±0.9*

(n=6)

Vm (mV) -65.0±0.9

(n=6)

-65.3±0.94

(n=6)

Rin (MΩ)

60.8±3.8

(n=6) 50.2±3.2*

(n=6)

Rin (MΩ)

75.2±5.4

(n=6)

75.4±6.0

(n=6)

Rin (MΩ)

at -73 mV

56.0±3.3

(n=6) 50.5±3.1*

(n=6)

Rin (MΩ)

at -73 mV

60.2±3.6

(n=5)

60.6.±3.1

(n=5)

768



38

References 781 782

Alzheimer C, Sutor B, and ten Bruggencate G. Transient and selective blockade of adenosine A1-783 receptors by 8-cyclopentyl-1,3-dipropylxanthine (DPCPX) causes sustained epileptiform activity in 784 hippocampal CA3 neurons of guinea pigs. Neurosci Lett 99: 107-112, 1989. 785 Andrade R, Malenka RC, and Nicoll RA. A G protein couples serotonin and GABAB receptors to the 786 same channels in hippocampus. Science 234: 1261-1265, 1986. 787 Armstrong CM, and Taylor SR. Interaction of barium ions with potassium channels in squid giant 788 axons. Biophys J 30: 473-488, 1980. 789 Broderick P, and Benjamin AB. Caffeine and psychiatric symptoms: a review. The Journal of the 790 Oklahoma State Medical Association 97: 538-542, 2004. 791 Chen X, and Johnston D. Constitutively active G-protein-gated inwardly rectifying K+ channels in 792 dendrites of hippocampal CA1 pyramidal neurons. J Neurosci 25: 3787-3792, 2005. 793 Day M, Carr DB, Ulrich S, Ilijic E, Tkatch T, and Surmeier DJ. Dendritic excitability of mouse frontal 794 cortex pyramidal neurons is shaped by the interaction among HCN, Kir2, and Kleak channels. J 795 Neurosci 25: 8776-8787, 2005. 796 Dong HW, Swanson LW, Chen L, Fanselow MS, and Toga AW. Genomic-anatomic evidence for 797 distinct functional domains in hippocampal field CA1. Proc Natl Acad Sci U S A 106: 11794-11799, 798 2009. 799 Dougherty KA, Islam T, and Johnston D. Intrinsic excitability of CA1 pyramidal neurones from the 800 rat dorsal and ventral hippocampus. J Physiol 590: 5707-5722, 2012. 801 Dougherty KA, Nicholson DA, Diaz L, Buss EW, Neuman KM, Chetkovich DM, and Johnston D. 802 Differential expression of HCN subunits alters voltage-dependent gating of h-channels in CA1 803 pyramidal neurons from dorsal and ventral hippocampus. J Neurophysiol 109: 1940-1953, 2013. 804 Drake CT, Bausch SB, Milner TA, and Chavkin C. GIRK1 immunoreactivity is present 805 predominantly in dendrites, dendritic spines, and somata in the CA1 region of the hippocampus. Proc 806 Natl Acad Sci U S A 94: 1007-1012, 1997. 807 Fanselow MS, and Dong HW. Are the dorsal and ventral hippocampus functionally distinct 808 structures? Neuron 65: 7-19, 2010. 809 Gasparini S, Losonczy A, Chen X, Johnston D, and Magee JC. Associative pairing enhances 810 action potential back-propagation in radial oblique branches of CA1 pyramidal neurons. J Physiol 580: 811 787-800, 2007. 812 Gerber U, Stevens DR, McCarley RW, and Greene RW. Muscarinic agonists activate an inwardly 813 rectifying potassium conductance in medial pontine reticular formation neurons of the rat in vitro. J 814 Neurosci 11: 3861-3867, 1991. 815 Hoffman DA, Magee JC, Colbert CM, and Johnston D. K+ channel regulation of signal propagation 816 in dendrites of hippocampal pyramidal neurons. Nature 387: 869-875, 1997. 817 Jin W, Klem AM, Lewis JH, and Lu Z. Mechanisms of inward-rectifier K+ channel inhibition by 818 tertiapin-Q. Biochemistry 38: 14294-14301, 1999. 819 Karschin C, Dissmann E, Stuhmer W, and Karschin A. IRK(1-3) and GIRK(1-4) inwardly rectifying 820 K+ channel mRNAs are differentially expressed in the adult rat brain. J Neurosci 16: 3559-3570, 1996. 821 Kaster MP, Rosa AO, Rosso MM, Goulart EC, Santos AR, and Rodrigues AL. Adenosine 822 administration produces an antidepressant-like effect in mice: evidence for the involvement of A1 and 823 A2A receptors. Neurosci Lett 355: 21-24, 2004. 824 Kim CS, Chang PY, and Johnston D. Enhancement of dorsal hippocampal activity by knockdown of 825 HCN1 channels leads to anxiolytic- and antidepressant-like behaviors. Neuron 75: 503-516, 2012. 826 Lanza M, Fassio A, Gemignani A, Bonanno G, and Raiteri M. CGP 52432: a novel potent and 827 selective GABAB autoreceptor antagonist in rat cerebral cortex. Eur J Pharmacol 237: 191-195, 1993. 828 Lee KS, Schubert P, Reddington M, and Kreutzberg GW. Adenosine receptor density and the 829 depression of evoked neuronal activity in the rat hippocampus in vitro. Neurosci Lett 37: 81-85, 1983. 830 Liao YJ, Jan YN, and Jan LY. Heteromultimerization of G-protein-gated inwardly rectifying K+ 831 channel proteins GIRK1 and GIRK2 and their altered expression in weaver brain. J Neurosci 16: 832 7137-7150, 1996. 833 Luscher C, Jan LY, Stoffel M, Malenka RC, and Nicoll RA. G protein-coupled inwardly rectifying K+ 834 channels (GIRKs) mediate postsynaptic but not presynaptic transmitter actions in hippocampal 835 neurons. Neuron 19: 687-695, 1997. 836

39

Magee JC. Dendritic hyperpolarization-activated currents modify the integrative properties of 837 hippocampal CA1 pyramidal neurons. J Neurosci 18: 7613-7624, 1998. 838 Marcelin B, Liu Z, Chen Y, Lewis AS, Becker A, McClelland S, Chetkovich DM, Migliore M, 839 Baram TZ, Esclapez M, and Bernard C. Dorsoventral differences in intrinsic properties in developing 840 CA1 pyramidal cells. J Neurosci 32: 3736-3747, 2012a. 841 Marcelin B, Lugo JN, Brewster AL, Liu Z, Lewis AS, McClelland S, Chetkovich DM, Baram TZ, 842 Anderson AE, Becker A, Esclapez M, and Bernard C. Differential dorso-ventral distributions of 843 Kv4.2 and HCN proteins confer distinct integrative properties to hippocampal CA1 pyramidal cell distal 844 dendrites. J Biol Chem 287: 17656-17661, 2012b. 845 Maximino C, Lima MG, Olivera KR, Picanco-Diniz DL, and Herculano AM. Adenosine A1, but not 846 A2, receptor blockade increases anxiety and arousal in Zebrafish. Basic & clinical pharmacology & 847 toxicology 109: 203-207, 2011. 848 Mihara S, North RA, and Surprenant A. Somatostatin increases an inwardly rectifying potassium 849 conductance in guinea-pig submucous plexus neurones. J Physiol 390: 335-355, 1987. 850 Moser MB, and Moser EI. Functional differentiation in the hippocampus. Hippocampus 8: 608-619, 851 1998. 852 North RA. Twelfth Gaddum memorial lecture. Drug receptors and the inhibition of nerve cells. Br J 853 Pharmacol 98: 13-28, 1989. 854 North RA, Williams JT, Surprenant A, and Christie MJ. Mu and delta receptors belong to a family 855 of receptors that are coupled to potassium channels. Proc Natl Acad Sci U S A 84: 5487-5491, 1987. 856 Papatheodoropoulos C, Asprodini E, Nikita I, Koutsona C, and Kostopoulos G. Weaker synaptic 857 inhibition in CA1 region of ventral compared to dorsal rat hippocampal slices. Brain Res 948: 117-121, 858 2002. 859 Ponce A, Bueno E, Kentros C, Vega-Saenz de Miera E, Chow A, Hillman D, Chen S, Zhu L, Wu 860 MB, Wu X, Rudy B, and Thornhill WB. G-protein-gated inward rectifier K+ channel proteins (GIRK1) 861 are present in the soma and dendrites as well as in nerve terminals of specific neurons in the brain. J 862 Neurosci 16: 1990-2001, 1996. 863 Prediger RD, da Silva GE, Batista LC, Bittencourt AL, and Takahashi RN. Activation of adenosine 864 A1 receptors reduces anxiety-like behavior during acute ethanol withdrawal (hangover) in mice. 865 Neuropsychopharmacology 31: 2210-2220, 2006. 866 Quayle JM, Standen NB, and Stanfield PR. The voltage-dependent block of ATP-sensitive 867 potassium channels of frog skeletal muscle by caesium and barium ions. J Physiol 405: 677-697, 868 1988. 869 Sah P. Ca(2+)-activated K+ currents in neurones: types, physiological roles and modulation. Trends 870 Neurosci 19: 150-154, 1996. 871 Signorini S, Liao YJ, Duncan SA, Jan LY, and Stoffel M. Normal cerebellar development but 872 susceptibility to seizures in mice lacking G protein-coupled, inwardly rectifying K+ channel GIRK2. 873 Proc Natl Acad Sci U S A 94: 923-927, 1997. 874 Spruston N. Pyramidal neurons: dendritic structure and synaptic integration. Nature reviews 875 Neuroscience 9: 206-221, 2008. 876 Standen NB, and Stanfield PR. A potential- and time-dependent blockade of inward rectification in 877 frog skeletal muscle fibres by barium and strontium ions. J Physiol 280: 169-191, 1978. 878 Stuart GJ, Dodt HU, and Sakmann B. Patch-clamp recordings from the soma and dendrites of 879 neurons in brain slices using infrared video microscopy. Pflugers Arch 423: 511-518, 1993. 880 van Calker D, Muller M, and Hamprecht B. Adenosine inhibits the accumulation of cyclic AMP in 881 cultured brain cells. Nature 276: 839-841, 1978. 882 van Groen T, and Wyss JM. Extrinsic projections from area CA1 of the rat hippocampus: olfactory, 883 cortical, subcortical, and bilateral hippocampal formation projections. J Comp Neurol 302: 515-528, 884 1990. 885 Vetter P, Roth A, and Hausser M. Propagation of action potentials in dendrites depends on dendritic 886 morphology. J Neurophysiol 85: 926-937, 2001. 887 Wainger BJ, DeGennaro M, Santoro B, Siegelbaum SA, and Tibbs GR. Molecular mechanism of 888 cAMP modulation of HCN pacemaker channels. Nature 411: 805-810, 2001. 889 Wang HS, Pan Z, Shi W, Brown BS, Wymore RS, Cohen IS, Dixon JE, and McKinnon D. KCNQ2 890 and KCNQ3 potassium channel subunits: molecular correlates of the M-channel. Science 282: 1890-891 1893, 1998. 892

40

Williams JT, North RA, and Tokimasa T. Inward rectification of resting and opiate-activated 893 potassium currents in rat locus coeruleus neurons. J Neurosci 8: 4299-4306, 1988. 894 895 896

ADorsal CA1 neurons

Ventral CA1 neurons

050

100150200250

Baseline Ba2+

Rni

)Ω

M(

50 μM Ba2+ 50 μM Ba2+

10 μM ZD728850 μM Ba2+

-80

-75

-70

-65

-60

Vm

)V

m(

0

50

100

150

200

250R

ni)

ΩM(

VmRin

-90

-80

-70

-60

Baseline Ba2+ ZD7288Ba2+

Vm

)V

m(


050

100150200250

Rni

)Ω

M(

-80

-75

-70

-65

-60

Vm

)V

m(

0

50

100

150

200

250

Rni

)Ω

M(

Time (min)12 16 20 24 32 36 40 44 48

50 μM Ba2+ 50 μM Ba2+

10 μM ZD728850 μM Ba2+Vm

RinB

Baseline50 μM Ba2+

10 μM ZD728850 μM Ba2+

200 ms

Vm 01

-70mV

-64mV

-72mV

Baseline 50 μM Ba2+

10 μM ZD728850 μM Ba2+

200 ms

Vm 01

-66mV

-64mV

-71mV

C D

E

F G H

***

* *

**

-90

-80

-70

-60


Vm

)V

m(

**

I

J

ZD7288Ba2+

-80

-75

-70

-65

-60

Vm

)V

m(

enilesaB aB+2

8827DZ aB

+2

Dorsal CA1 Ventral CA1

0

50

100

150

200

Rni

)Ω

M(

#

#

enilesaB aB+2

8827DZ aB

+2

enilesaBaB

+2 8827DZ aB+2

enilesaBaB

+2 8827DZ aB+2

-73 mV

-73 mV

200 ms

Vm 01

200 ms

Vm 01

-150 pA

10 pAIinj

Baseline

50 μM Ba2+

Baseline

50 μM Ba2+

Baseline Ba2+0

2040

6080

100

Rni

Vm 37- ta )

ΩM(

Baseline Ba2+0

2040

6080

100

Rni

Vm 37- ta )

ΩM (

*K

L

-150 pA

10 pAIinj

-150 pA

10 pAIinj

Time (min)11 15 19 23 34 38 42 46 50

Vm(m

V)

M

BaselineBa2+

-150 -100 -50

-15

-10

-5

5

Vm(m

V)

Injected Current (pA)

At -73mV

-150 -100 -50

-15

-10

-5

5

Vm(m

V)


At RMP

Ventral CA1 neuronsN

-150 -100 -50

-15

-5

5

BaselineBa2+

-150 -100 -50

-15

-10

-5

5

Vm(m

V)

Injected Current (pA) Injected Current (pA)

At RMP At -73mV

Dorsal CA1 neurons

* *

-150 -100 -50

-5-4-3-2-1

1

DorsalVentral

At RMP

∆Vm(m

V)

(Bas

elin

e- B

a2+ )


*

O-150 -100 -50

-5-4-3-2-1

1

At -73mVInjected Current (pA)

* ∆Vm(m

V)

(Bas

elin

e- B

a2+ )

DorsalVentral

Dorsal CA1

2

4

6

8

10

0

Cha

nge

in V

m (m

V)

(Bas

elin

e - T

ertia

pin-

Q)

**

#

0.03 0.3 2

Ventral CA1

Tertiapin-Q (μM) 0

.03 0.3 2

0

20

40

60

80

100

Cha

nge

in R

in(%

)(B

asel

ine-

Terti

apin

-Q)

**

Dorsal CA1

0.03 0.3 2

Ventral CA1

Tertiapin-Q (μM) 0

.03 0.3 2

F G

0.03 μM Tertiapin-Q

-70 mV

-150 pA

10 pA

-68.9mV

-69 mV

-64.6 mV -69.4

mV

-64.8 mV

10 m

V

200 ms

0.3 μM Tertiapin-Q 2 μM Tertiapin-Q

E

Before After Before After Before After

-66 mV

-65.3 mV

-65 mV

-63.7 mV

-66 mV

-64.5 mV

10 m

V

200 ms-150 pA

10 pA

0.03 μM Tertiapin-Q 0.3 μM Tertiapin-Q 2 μM Tertiapin-Q

Before After Before After Before After

Dorsal CA1 neurons

Ventral CA1 neurons

##

AKir3.1 (GIRK1)

Kir3.2 (GIRK2)

500 μm

500 μm

Kir3.1 Kir3.1

Kir3.2 Kir3.2

Dorsal Ventral

Dorsal Ventral

SOSPSRSLM

SPSO

SR

SLM

B

C D

* *

Normalized Relative Distancein CA1 region

Rel

ativ

e In

tens

ity(A

rbitr

ary

Uni

t)

Dorsal CA1Ventral CA1

SP SR SLM

0 0.5 1.00

50

100

150

Rel

ativ

e In

tens

ity(A

rbitr

ary

Uni

t)

0 0.5 1.00

50

100

150

SP SR SLM


*


SP SL

SP

*

L

Iinj

Iinj

#

-70 mV Vm 01

200 ms

-150 pA

10 pA

-65 mV -63 mV

Baseline 0.3 μM Tertiapin-Q0.3 μM Tertiapin-Q

50 μM Ba2+

*

50

100

150

Rni

)Ω

M (

*

-80

-70

-60

-50

Vm

)V

m(

eniilesaB

Q-nipaitreT

Q-nipaitreT

aB

+2

-66 mV -64 mV-62 mV

Vm 01

200 ms

-150 pA

10 pA

*

-80

-70

-60

-50

A

Rni

)Ω

M(

60

80

100

14 18 2210Time (min)

-75

-70

-65

-60

Vm

)V

m(

120

140

26 30 38 42 46 51 55

VmRin

0.3 μM Tertiapin-Q

60

80

100

9 13 17 21Time (min)

-75

-70

-65

-60

120

140

25 29 3640 44 4852

VmRin

B C D

E

F G H0.3 μM Tertiapin-Q

50 μM Ba2+

50 μM Ba2+

Baseline 0.3 μM Tertiapin-Q0.3 μM Tertiapin-Q

50 μM Ba2+

Ventral CA1 neurons

Dorsal CA1 neurons

-73 mV

Vm 01

200 ms

-150 pA

10 pA

Baseline

0.3 μM Tertiapin-Q

0.3 μM Tertiapin-Q50 μM Ba2+

-150 pA

10 pA

IVentral CA1 neuronsDorsal CA1 neurons

0

20

40

60

80

100

eniilesaB

Q-nipaitreT

Q-nipaitreT

aB

+2

Rni

Vm 37- ta )

ΩM(

**

0

20

40

60

80

100

Rni

Vm 37- ta )

ΩM(

JBaseline

0.3 μM Tertiapin-Q

0.3 μM Tertiapin-Q50 μM Ba2+

-73 mV

Vm 01

200 ms

**

*

0

50

100

150

0

Iinj

Iinj

Rni

)Ω

M(

Vm

)V

m(

Rni

)Ω

M(

Vm

)V

m(

eniilesaB

Q-nipaitreT

Q-nipaitreT

aB

+2

eniilesaB

Q-nipaitreT

Q-nipaitreT

aB

+2 eniilesaB

Q-nipaitreT

Q-nipaitreT

aB

+2

eniilesaB

Q-nipaitreT

Q-nipaitreT

aB

+2

Iinj Iinj

0

5

10

15

20

# of

AP

s

0 60 120 180 240 300Injected Current (pA)

0 60 120 180 240 3000

5

10

15

20


# of

AP

s

BaselineTertiapin-QTertiapin-Q + Ba2+

-73 mV -73 mV

Dorsal CA1 neurons Ventral CA1 neuronsBaseliine 0.3 μM Tertiapin-Q 50 μM Ba2+

20 m

V

200 ms

20 m

V

200 ms

Baseliine 0.3 μM Tertiapin-Q 50 μM Ba2+

K L

*

300 pA 300 pA

0.3 μM Tertiapin-Q 0.3 μM Tertiapin-Q

BaselineTertiapin-QTertiapin-Q + Ba2+

DADorsal CA1 neurons Ventral CA1 neurons

Time (min)

Vm

Rin

0

50

120

-80

-70

-60

0 8 16

Vm (m

V) R

in

24

VmRin

-85

-75

-65

-55

Baseline Baclofen

*

-85

-75

-65

-55 *

0

50

100

150*

0

50

100

150 *

32

5 μM CGP55845

100

5 μM CGP55845

Vm (m

V)

-75

-70

-65

-60

Baseline CGP55845

0

50

100

150

Rin

Baseline CGP55845

0

50

100

150

Baseline CGP55845

-75

-70

-65

-60

Baseline CGP55845

Time (min)

0

50

120

-80

-70

-60

0 8 16 24 32

100

B C E F

Baseline Baclofen Baseline Baclofen Baseline Baclofen

0

30

60

90

120

-80

-70

-60

Time (min)0 2 4 6 8

0

30

60

90

120

-80

-70

-60

Time (min)0 2 4 6 8

100 μM Baclofen 100 μM Baclofen

JG

H I K L

M N

Cha

nge

in R

in (%

) 0-20

-40

-60

-80

-100

Dorsal VentralDorsal Ventral

-10

-8

-6

-4

-2

0

Cha

nge

in V

m (m

V)

Baclofen Baclofen

Dorsal CA1 neurons Ventral CA1 neurons

Rin

Vm (m

V)

Vm (m

V)

Rin

Vm

RinVmRin

Vm (m

V) R

in

Rin

Vm (m

V)

Vm (m

V)

Rin V

m (m

V)

Rin

EADorsal CA1 neurons Ventral CA1 neurons

-67 mV -67 mV-70 mV

-150 pA

10 pA

5 m

V

200 ms

-65 mV

Baseline 100 nM DPCPX Baseline

-150 pA

10 pA

100 nM DPCPX

5 m

V

200 ms

*

-80

-70

-60

-50

Vm(m

V)

Baseline DPCPX-80

-70

-60

-50

Baseline DPCPX0

50

100

150

Rin

Baseline DPCPX

*

0

50

100

150

Baseline DPCPX

Time (min)

VmRin

406080100120140

-80

-75

-70

-65

-60

0 5 10 15 20

Vm (m

V) Rin

100 nM DPCPX

Time (min)

VmRin

406080100120140

-80

-75

-70

-65

-60

0 5 10 15 20

100 nM DPCPX

Cha

nge

in R

in (%

)

Dorsal Ventral0

20

40

60

80

100

B

C D

F

G H

0

2

4

6

8

10

Dorsal Ventral

Cha

nge

in V

m (m

V)

# #

I JV

m (m

V) Rin

Iinj Iinj

Vm(m

V)

Rin

Baseline Ba2+ DPCPXBa2+

-80

-70

-60

-50

Vm (m

V)

0

50

100

150

Rin��

��

Baseline Ba2+ DPCPXBa2+

VmRin

B

C D **

**

m

0

50

100

-80

-70

-60

-50

Vm (m

V)

150-40

10 14 18

Time (min)22

50 μM Ba2+ 50 μM Ba2+100 nM DPCPX

Rin ��

��

-70 mV

10 m

V

200 ms

-150 pA

10 pA

-64 mV -64 mV

Baseline 50 μM Ba2+100 nM DPCPX

50 �M Ba2+A

Dorsal CA1 neurons

-73 mV

-150 pA

10 pA

Baseline

50 μM Ba2+

100 nM DPCPX50 �M Ba2+

E

10 m

V

200 ms

Basel

iine Ba2+

DPCP

X

Ba2

+

**

0

20

40

60

80

100

Rin��

��

��

��

26 37 41 45 49 53

Iinj

Iinj

GCDorsal CA1 neurons Ventral CA1 neurons

-66 mV -67 mV-69 mV

-150 pA

10 pA

-73 mV

Baseline 1 μM 2’ME-CCPA Baseline

-150 pA

10 pA

5 m

V

200 ms

-80

-70

-60

-50

Vm

(mV

)

Baseline 2’ME-CCPA-80

-70

-60

-50

0

50

100

150

Rin��

��

*

0

50

100

150

Cha

nge

in R

in (%

)

Dorsal Ventral0

20

40

60

80

100

D

E F

H

I J

K LDorsal Ventral

#-5

-4

-3

-2

-1

0

Cha

nge

in V

m (m

V)

Baseline 2’ME-CCPABaseline 2’ME-CCPA Baseline 2’ME-CCPA

5 m

V

200 ms

1 μM 2’ME-CCPA

Time (min)

VmRin

020406080100

-80

-75

-70

-65

-60

10 12 14 16 18

Vm (m

V) Rin ��

��

1 μM 2’ME-CCPA

20 22

VmRin

0

20406080100

-80

-75

-70

-65

-60

1 μM 2’ME-CCPA

Time (min)11 13 15 17 19 21 23 25

*

#

020406080

100

Rin��

��

�

��

��

�

Baseline 2’ME-CCPA

*

020406080

100

Rin��

��

�

��

��

�

Baseline 2’ME-CCPA

M NDorsal

CA1 neuronsVentral

CA1 neurons

Vm (m

V) Rin ��

��

Iinj Iinj

Vm

(mV

)

Rin��

��

500 μm

A1 ARDorsal

SOSPSRSLM

A1 ARVentral

500 μm


Rel

ativ

e In

tens

ity(A

rbitr

ary

Uni

t)


0 0.5 1.00

50

100

150*

SP SR SLM

200SOSP

SRSLM

SO

SSL

A B

Article FileFigure 1Figure 2Figure 3Figure 4Figure 5Figure 6Figure 7

title: a1 adenosine receptor-mediated girk channels ......172 were rinsed in pbs buffer, and then...

Documents