Articles in PresS. J Neurophysiol (February 4, 2015). doi:10.1152/jn.00717.2014

1

Experimental and computational evidence for an essential role of

2

NaV1.6 in spike initiation at stretch-sensitive colorectal afferent

3

endings

4

Bin Feng, Yi Zhu, Jun-Ho La, Zachary P. Wills1, and G.F. Gebhart

5

Center for Pain Research, Department of Anesthesiology and 1Department of

6

Neurobiology, School of Medicine, University of Pittsburgh, Pittsburgh, PA 15213,

7

USA

8

Abbreviated title: NaV1.6 in spike initiation of colorectal afferents

9

Manuscript length: 49 pages, 9 figures, 4 tables

10

Words: 231 for Abstract, 489 for Introduction, 1899 for Discussion

11

Conflict of interest: The authors claim no conflict of interests.

12

Acknowledgements: supported by NIH grant R01 DK093525 (GFG). We thank

13

Michael Burcham for assistance in preparation of figures and Dr. Brian Davis for his

14

generous gift of transgenic mice that express YFP in sensory afferents (driven by

15

SNS-Cre)

16

Correspondence:

17

Bin Feng, Ph.D.

18

Center for Pain Research, University of Pittsburgh

19

W1402 BST, 200 Lothrop St., Pittsburgh, PA 15213, USA

20

E-mail: [email protected]

21

Phone: 412-648-8123, fax: 412-383-5466

22

-1-

Copyright © 2015 by the American Physiological Society.

23

ABSTRACT

24

Stretch-sensitive afferents comprise ~33% of the pelvic nerve innervation of

25

mouse colorectum, which are activated by colorectal distension and encode visceral

26

nociception. Stretch-sensitive colorectal afferent endings respond tonically to stepped

27

or ramped colorectal stretch whereas dissociated colorectal DRG neurons generally

28

fail to spike repetitively upon stepped current stimulation. The present study

29

investigated this difference in the neural encoding characteristics between the soma

30

and afferent ending using pharmacological approaches in an in vitro mouse

31

colon-nerve

32

Immunohistological staining and western blots revealed the presence of NaV1.6 and

33

NaV1.7 channels at sensory neuronal endings in mouse colorectal tissue. Responses

34

of stretch-sensitive colorectal afferent endings were significantly reduced by targeting

35

NaV1.6 using selective antagonists (μ-conotoxin GIIIa and μ-conotoxin PIIIa) or TTX.

36

In contrast, neither selective NaV1.8 (A803467) nor NaV1.7 (ProTX-II) antagonists

37

attenuated afferent responses to stretch. Computational simulation of a colorectal

38

afferent ending that incorporated independent Markov models for NaV1.6 and NaV1.7,

39

respectively, recapitulated the experimental findings, suggesting a necessary role for

40

NaV1.6 in encoding tonic spiking by stretch-sensitive afferents. In addition,

41

computational simulation of a DRG soma showed that by adding a NaV1.6

42

conductance, a single-spiking neuron was converted into a tonic spiking one. These

43

results suggest a mechanism/channel to explain the difference in neural encoding

44

characteristics between afferent somata and sensory endings, likely caused by

preparation

and

complementary

-2-

computational

simulations.

45

differential expression of ion channels (e.g., NaV1.6) at different parts of the neuron.

46

-3-

47

INTRODUCTION

48

Irritable bowel syndrome (IBS) patients typically suffer from persistent pain and

49

organ hypersensitivity, manifesting enhanced responses and reduced thresholds to

50

mechanical distension of the distal colorectum (Naliboff et al. 1997). Understanding

51

the afferent innervation of the colorectum is important as targeting colorectal afferents

52

has proven to be effective in alleviating pain and hypersensitivity in IBS patients (e.g.,

53

intra-rectal instillation of local anesthetics (Verne et al. 2003; Verne et al. 2005) and

54

oral intake of the guanylate cyclase-C agonist linaclotide (Busby et al. 2013; Chey et

55

al. 2012; Rao et al. 2012).

56

The colorectum in the mouse is innervated by lumbar splanchnic and pelvic

57

nerve pathways (Brierley et al. 2004; Feng and Gebhart 2011). Using an in vitro

58

colorectum-nerve preparation and an unbiased electric search strategy, we

59

categorized colorectal afferents into five mechanosensitive classes (serosal, mucosal,

60

muscular, muscular-mucosal, mesenteric) and one mechanically-insensitive class

61

(Feng et al. 2012a). Among these classes, only muscular and muscular-mucosal

62

afferents tonically encode circumferential stretch of the colorectum (i.e., are

63

stretch-sensitive) and subserve the encoding of nociceptive colorectal distension

64

(Feng et al. 2010; Feng et al. 2013).

65

Patch clamp recordings reveal that colorectal dorsal root ganglion (c-DRG)

66

somata have small diameters ( (V − ENa ) . Rate coefficients (ms) that are

NaV1.7 – Markov typed model (Fig. 6A) adopted from Gurkiewicz et

- 13 -

247

al.(Gurkiewicz et al. 2011). NaV1.7 current is determined by the open probability < O >

248

and membrane potential V:

249

adopted from Gurkiewicz et al.

250

INa7 = gNa7 < O > (V − ENa ) . All model parameters were

NaV1.8 – Hodgkin-Huxley typed model adopted from Baker (Baker 2005)

INa8 = gNa8m3h(V − ENa ) = m

h −h h = ∞

m∞ − m

τh

τm



 V + 2.58     11.47  

α m = 3.83 1 + exp  − 

m∞ =

αm α m + βm  V + 105    46.33 

251

αh

1.0 αm + βm 

 V − 21.8     11.998  

β h = 0.61714 1 + exp  − 

τh =

α h + βh

 V + 61.2     19.8  



τm =

α h = 0.013536exp  − h∞ =



β m = 6.894 1 + exp 

1.0 α h + βh

NaV1.9 – Hodgkin-Huxley typed model adopted from Baker (Baker 2005)

INa9 = gNa9mh(V − ENa ) = m

h −h h = ∞

m∞ − m

τh

τm



 V − 11.01     14.871  

α m = 1.548 1 + exp  − 

m∞ =

αm α m + βm 

 V + 63.264     3.7193  

V + 112.4     22.9  



τm =

α h = 0.2574 1 + exp  



β m = 8.685 1 + exp 

1.0 αm + βm 

 V + 0.27853    9.0933   

β h = 0.53984 1 + exp  − 

- 14 -

h∞ =

252 253

αh α h + βh

τh =

1.0 α h + βh

KA – Hodgkin-Huxley typed model adopted from Schild et al.(Schild et al. 1994) with minor adjustments of parameters.

IKA = gKA p3q(V − EK ) p =

254 255

p∞ − p

τp

q =

q∞ − q

τq

  V + 15.79   p∞ = 1.0 1.0 + exp    −10.0   

τ p = 5.0 exp ( −0.022 2 (V + 65) 2 ) + 1.5

  V + 58   q∞ = 1.0 1.0 + exp    7.0   

τ q = 45.0 exp ( −0.00352 (V + 10) 2 ) + 10.5

KD – Hodgkin-Huxley typed model adopted from Schild et al.(Schild et al. 1994) with minor adjustments of parameters.

IKD = gKD x3 y(V − EK ) x =

256 257

x∞ − x

τx

y =

y∞ − y

τy

  V + 14.59   x∞ = 1.0 1.0 + exp    −15.0   

τ x = 5.0 exp ( −0.022 2 (V + 65) 2 ) + 3.5

  V + 48   y∞ = 1.0 1.0 + exp    7.0   

τ y = 1800

KS - Hodgkin-Huxley typed model adopted from Schild et al.(Schild et al. 1994) with minor adjustment of parameters.

IKS = gKSn4 (V − EK ) n =

n∞ − n

τn

- 15 -

n∞ =

αn

τn =

α n + βn

 V + 60   80  

V + 52    V + 52    1.0 − exp  −   12.5    12.5  

α n = 0.1 

258 259

βn = 0.125exp  −

Na+,K+-ATPase - adapted from Bondarenko et al.(Bondarenko et al. 2004) with parameters slightly adjusted.

I NaK = I NaK ⋅ f NaK ⋅

[K + ]o ⋅ + 4 / [ Na + ] ) [K ]o + Km,Ko

1 1 + ( Km,Nai

I NaK = I Na, p + I K, p f NaK =

i

I Na, p = 3I NaK

IK, p = −2INaK

1 VF    VF  1 + 0.1245 ⋅ exp  −0.1  + 0.0365 ⋅ σ ⋅ exp  −  RT    RT 

1  7 

 [ Na+ ]o    −1  67.3  

σ = ⋅ exp 

260

1.0 α n + βn

Km,Nai = 13mM

Km,Ko = 1.5mM

Background current

IB = gB(V − EB ) I B = I Na,B + I K ,B I Na,B = gB 261 262 263 264 265 266

EB − EK (V − ENa ) ENa − EK

I K ,B = gB

ENa − EB (V − EK ) ENa − EK

The summary of the ion channels and Na+,K+-ATPase included in modeling is listed in table 1 below.

Mechanosensitive (ms) ion channels

267

To simulate the gating of a mechanosensitive (ms) ion channel, we used a

268

two-state model that includes an open state (O) and a closed state (C), the rates of

269

transition between which are α and β, two exponential functions of membrane tension

- 16 -

270

τ

at the afferent ending : α

⎯⎯ →C O ←⎯ ⎯ β

α=

1  τ −τ 0  exp  −  A  2S 

β=

1  τ −τ 0  exp   A  2S 

271

Assume that the fraction of ms channels in the open state is denoted by p, and we

272

have:

273

p =

α p∞ − p and , in which, p∞ = α +β Tp

Tp =

1 . α +β

274

p∞, the open probability of the ms channel at steady-state, follows Boltzmann’s

275

equation, consistent with previous theoretical and experimental studies on ms

276

channels (Hao and Delmas 2010; Haselwandter and Phillips 2013; Wiggins and

277

Phillips 2005).

278

279

p∞ =

1  τ −τ 0  1 + exp  − S  

To allow calculation of

τ

from bulk colorectal deformation (circumferential stretch

280

force F, stretch ratio λ, and their derivatives F and λ ), the passive mechanical

281

properties of the colorectal wall tissue was simulated by a lumped parametric model

282

consisting of 2 springs and 1 dashpot (Fig. 1C), which leads to the following

283

equations:

284

 x2 = (λ − 1) ΔL  τ Δw = k2 ( x2 − x1 )  τ Δw = k x + c x 1 1 1 1 

285

Assuming: Tτ =

c1 k1 + k2

m1 =

k1k2 ΔL (k1 + k2 )Δw

- 17 -

m2 =

k2c1 ΔL (k1 + k2 )Δw

m1 (λ − 1) + m2λ − τ Tτ

286

Then: τ =

287

The ms channel conductance is divided into conductance for Na+ and K+ ions:

288

gM = gM p = gNa,M + gK ,M

289

Assuming the conductance for Na+ and K+ are proportional to their respective driving

290

forces, i.e.,

(ENa −V ) and (V − EK ) , then we have:

I Na, M = g M p

ENa − V (V − ENa ) ENa − EK

I K ,M = g M p

V − EK (V − EK ) ENa − EK

I M = I Na,M + I K,M 291

292

Parameters for the ms channel in the study are listed below: A

10 ms

τ0

4.45 mN/m

S

2.07 mN/m

1000 ms

m1

13.6 mN/m

Tτ m2

16900 (mN∙ms)/m

The ms current evoked by a stepped colorectal stretch is plotted in Fig. 1C, which

293

recapitulated features of a typical slowly adapting

294

experimentally in DRG neurons (Rugiero et al. 2010).

295

Passive properties and initial conditions

ms current observed

296

The passive electrical properties Cm, Rm and Ri in both models were set to 1

297

μF/cm2, 10,000 Ω∙cm2, and 123 Ω∙cm, respectively. The initial ion concentrations for

298

the soma model were set to be consistent with current-clamp recording conditions

299

(i.e., 140mM [Na+]o, 4.5mM [Na+]i, 5mM [K+]o, and 130mM [K+]i), so was the resting

300

membrane potential Vm of -64.3mV(Shinoda et al. 2010). The initial ion

301

concentrations for the afferent ending model(145mM [Na+]o, 4mM [Na+]i, 6.3mM [K+]o,

302

and 155mM [K+]i); were determined when the model reached equilibrium condition at

- 18 -

303

Vm of -65mV, a potential well within the range of resting membrane potentials

304

recorded from distal axons of small-diameter DRG neurons (-68 to -53mV) (Vasylyev

305

and Waxman 2012). Simulations were run at 30oC to approximate the experimental

306

conditions of in vitro single-fiber recordings. Rate constants of voltage-dependent

307

channels and Na+,K+-ATPase were multiplied by a temperature factor (i.e.,

308

Q10(

309

al. 1994). The numerical error tolerance in NEURON was set at 10-5.

310

Data Recording and Analysis

t −24) 10

). The Q10 values listed in Table 2 were adapted from Schild et al.(Schild et

311

Action potentials (APs) were recorded extracellularly using a low-noise AC

312

differential amplifier. Activity was monitored on-line, filtered (0.3 to 10 kHz), amplified

313

(x10,000), digitized at 20kHz using a 1401 interface (CED, Cambridge, England) and

314

stored on a PC. APs were discriminated off-line using Spike 2 software (CED). To

315

avoid erroneous discrimination, no more than two clearly discriminable units in any

316

record were studied. The stretch response threshold was defined as the force that

317

evoked the first AP during ramped stretch. SRFs are presented as bins of evoked

318

APs (0-57, 57-113, 113-170mN) by the ramped colorectal stretch. SRFs were

319

normalized to the respective maximum binned spike number in control (baseline)

320

tests. To facilitate comparisons, responses of stretch-sensitive afferents are also

321

presented as total numbers of APs during ramped stretch. Data are presented

322

throughout as mean ± SEM. One-way and two-way analyses of variance (ANOVA) or

323

repeated-measures were performed as appropriate using SigmaPlot v11.0 (Systat

324

software, Inc., San Jose, CA). Bonferroni post-hoc multiple comparisons were

- 19 -

325

performed when F values for main effects were significant. Differences were

326

considered significant when p