Age-dependent effects on sensory axonal excitability in normal mice

Neuroscience Letters 611 (2016) 81–87

Contents lists available at ScienceDirect

Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet

Research paper

Age-dependent effects on sensory axonal excitability in normal mice Chimeglkham Banzrai, Hiroyuki Nodera ∗ , Saki Higashi, Ryo Okada, Yusuke Osaki, Atsuko Mori, Ryuji Kaji Department of Neurology, Tokushima University, Tokushima, Japan

h i g h l i g h t s • Age-dependent changes occur on axonal excitability of both motor and sensory neurons. These changes may reflect alterations in passive membrane properties.

• Alterations of an ion conductance (e.g., H conductance) and cable properties (e.g., Barrett–Barrett conductance) explain the interval changes of sensory axonal excitability with maturation and aging.

• Dynamic changes of sensory axonal excitability may explain age-dependent sensory symptoms.

a r t i c l e

i n f o

Article history: Received 26 August 2015 Received in revised form 9 November 2015 Accepted 20 November 2015 Available online 25 November 2015 Keywords: Axonal excitability Age-related change Sensory nerve Ion conductance

a b s t r a c t Serial recordings were performed to measure sensory excitability in peripheral nerves and elucidate age-dependent changes in neuronal ion currents in the peripheral sensory nervous system. The threshold tracking technique was used to measure multiple excitability indices in the tail sensory nerves of five normal male mice at four time points (6, 10, 14, and 19 weeks of age). A separate group of four mice was also measured at 43 weeks and at 60 weeks of age. Maturation was accompanied by an increase in early hyperpolarization and superexcitability at 10 weeks. At 60 weeks, the hyperpolarizing electrotonus shifted downward, while superexcitability became greater and subexcitability (double stimuli) decreased. Computer modeling showed that the most notable age-related interval changes in excitability parameters were Barrett–Barrett, H, and slow K+ conductances. Understanding age-related changes in the excitability of sensory axons may provide a platform for understanding age-dependent sensory symptoms and developing age-specific channel-targeting therapies. © 2015 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Increasing age affects the structure and function of the peripheral nervous system. During maturation, the axonal diameter of human peripheral nerves increases during the first 5 years, the myelin thickens until 14 years of age, and the internodal length increases until the second decade [9,20]. Histopathological changes consistent with aging become evident by the fifth decade, such as the loss of myelinated and unmyelinated fibers [1,5,20]. Animal studies have shown a decrease in axon diameter, disruption of the myelin sheath, a pronounced increase in collagen fibers in the endoneurium and perineurium, the disruption of axoglial junctions, and separation of the myelin loops from the paranodal axolemma,

∗ Corresponding author at: Department of Neurology, 3-18-15 Kuramotocho, Tokushima City 770-8503, Japan. Fax: +81 88 633 7208. E-mail address: [email protected] (H. Nodera). http://dx.doi.org/10.1016/j.neulet.2015.11.032 0304-3940/© 2015 Elsevier Ireland Ltd. All rights reserved.

which widens the nodes of Ranvier of peripheral nerves in old animals [1,5]. Studies using threshold tracking to assess axonal membrane excitability in-vivo have provided insight into the properties of axonal membranes under normal conditions and in many peripheral-nerve disorders [4,7,12,14,16]. The establishment of a threshold tracking model of animal sensory nerves and a mathematical model of human sensory axons provides detailed information about the molecular mechanisms underlying axonal membrane function, and these models have enabled the study of axonal excitability in sensory nerves [7]. Previous studies have uncovered differences between motor and sensory nerve function [19]. A recent study reported that there is an increased hyperpolarization-activated current (Ih) and a smaller nodal slow K+ current (IKs) in human sensory axons than in motor axons. Therefore, we hypothesized that there are age-related changes to neuronal ion currents in the sensory nervous system that may be

82

C. Banzrai et al. / Neuroscience Letters 611 (2016) 81–87

different from those in the motor system. The present study utilized axonal excitability in normal mice to determine the age-related effects on sensory axons, describing the underlying molecular mechanisms with mathematical modeling.

The level of significance was P < 0.05. All data are presented as means ± SD.

2. Methods

The commercially available Bostock model of the human motor axon was used in the simulation of axonal excitability (MEMFit, QtracP version 17/10/2014), as previously explained in detail [8,13]. Parameter adjustments were made to improve the fit to the normal human recovery cycle (RC), strength-duration time constant (SDTC), current-threshold relationship (I/V), and threshold electrotonus (TE). To reflect better the characteristic waveform changes (see Results section), the weighting factors were set as follows: TE, 2; RC, 1; SDTC, 0.5; and I/V, 1. The tested parameters were as follows: nodal and internodal resting potentials (ENR and EIR, respectively), nodal Na+ permeability (PNa), percent persistent Na+ channels (PNap), nodal and internodal slow K+ conductance (GKs), nodal and internodal fast K+ conductance (GKf), internodal H conductance (GH), nodal and internodal leak conductance (GLk), Barrett–Barrett conductance (GBB), and total pump currents (IPump). First, the recording of the 14-week-old mice was taken as a reference and best fits were obtained by changing the aforementioned parameters. The recordings for the other ages were then fitted by changing single parameters to reduce the discrepancy. If that did not yield a satisfactory reduction (i.e., arbitrarily less than 65%), another fitting was performed by first changing two parameters, and then up to three parameters.

2.1. Study protocol The experiment was approved by the local animal facility at the Tokushima University. ICR normal male mice (SLC, Hamamatsu, Japan) were tested. Serial electrophysiological testing was performed in five mice at four time points, i.e., juvenile, adolescent, young-adult, and mature (6, 10, 14, 19 weeks of age, respectively). A different group of four male mice was tested at 43 weeks of age (adult) and 60 weeks (aged). 2.2. Axonal excitability study Electrophysiological studies were performed on the tail under 1.5% isoflurane anesthesia. Sensory nerve action potentials were recorded orthodromically and the setup of the electrodes, temperature maintenance, and neuronal excitability testing were performed as previously described in detail [18]. In brief, stimulation was controlled by a PC running the QtracS program (Institute of Neurology, London, UK), and the TRONDNF multiple excitability recording protocol was used for excitability tests. A set of excitability parameters was derived from the recordings, as previously described [17]. One cycle of multiple excitability tests took approximately 20 min.

2.4. Modeling of the excitability data

3. Results

2.3. Data analysis

3.1. Sensory nerve excitability changes with maturation

Axonal excitability data from serial recordings were compared using one-way repeated measures ANOVA with Bonferroni corrections where appropriate, and data from three different time points (19, 43, and 60 weeks old) were compared using one-way ANOVA with Post Hoc tests (SPSS version 22: IBM, New York, NY, USA).

To examine the sequential changes to sensory nerve excitability with maturation, serial recordings were tested at four time points (Table 1, Figs. 1 and 2A and B). The peak latency decreased gradually from 1.87 ± 0.1 ms for the 6 week-old mice to 1.31 ± 0.08 ms for 19 week-old mice, suggesting greater conduction velocities with

Table 1 Changes in sensory axon excitability parameters with maturation in mice (depolarizing threshold electrotonus, TEd; hyperpolarizing threshold electrotonus, TEh; #: P < 0.05; *: P < 0.01).

Amplitude (␮V) Rheobase Peak latency (ms) Threshold electrotonus (TE) TEd (10–20 ms) TEd (40–60 ms) TEd (90–100 ms) TEh (10–20 ms) TEh (20–40 ms) TEh (90–100 ms) S2 accommodation TEh (peak: −70%) (−70%) S3 accommodation Recovery cycle (RC) at 2 ms Refractoriness Superexcitability (%) at 7 ms Superexcitability Late subexcitability (%) RC2-subexcitability (%) Current/threshold relationship (I/V) Resting I/V slope Minimum I/V slope Hyperpolarizing I/V slope Stimulus-response (SR) relationship max response Stimulus for 50% Strength-durationtime constant (SDTC)

6 weeks old (A) 10 weeks old (B) 14 weeksold (C) 19 weeks old (D) ANOVA P values

With Bonferroni correction P values

61.7 ± 25.3 0.42 ± 0.1 1.87 ± 0.1

112.3 ± 45.2 0.35± 0.05 1.49 ± 0.09

128.3 ± 62.7 0.31 ± 0.06 1.39 ± 0.1

135.7 ± 46.5 0.31 ± 0.06 1.31 ± 0.08

0.9 0.06 0.001*

0.02# (A/B);0.032# (A/C);0.032# (A/D)

45.2 ± 2.4 43.3 ± 1.4 41.9 ± 2.2 −61.9 ± 2.9 −72.1 ± 3.8 −59.87 ± 9.9 3.5 ± 2.6 −140.0 ± 21.4 41.8 ± 5.3

42.2 ± 2.7 40.5 ± 1.7 39.1 ± 1.3 −58.0 ± 3.3 −58.8 ± 4.9 −48.09 ± 5.5 3.2 ± 1.4 −119.1 ± 9.2 46.2 ± 7.3

44.1 ± 3.4 42.2 ± 3.3 40.6 ± 1.1 −56.7 ± 5.9 −60.0 ± 7.2 −53.21 ± 5.0 3.7 ± 2.2 −122.9 ± 14.0 39.8 ± 14.5

45.3 ± 1.8 41.1 ± 1.2 40.2 ± 0.4 58.5 ± 2.0 −64.9 ± 2.9 −58.9 ± 4.3 4.8 ± 1.9 −137.0 ± 4.9 46.4 ± 12.1

0.07 0.13 0.1 0.2 0.005* 0.051 0.4 0.6 0.1

0.003* (A/B)

2.9 ± 5.9 −5.62 ± 2.8 −1.3 ± 2.7 1.8 ±1.6 1.6 ± 1.3

2.3 ± 2.9 −1.82 ± 0.8 2.0 ± 1.1 1.5 ±1.3 2.5 ± 0.4

−1.5 ± 3.7 −5.32 ± 2.4 −1.0 ± 2.4 2.5 ±1.0 4.5 ± 0.5

−2.5 ± 4.1 −7.1 ± 2.4 −2.5 ± 0.6 1.8 ±1.1 3.7 ± 1.1

0.013# 0.023# 0.037# 0.57 0.01#

0.017# (B/D)

0.85 ± 0.04 0.58 ± 0.1 0.83 ± 0.2

0.99 ± 0.1 0.81 ± 0.03 1.0 ± 0.06

0.92 ± 0.07 0.67 ± 0.07 1.2 ± 1.0

0.78 ± 0.1 0.65 ± 0.06 1.3 ± 0.8

0.03 0.007* 0.7

0.027# (B/D)

0.59 ± 0.2 0.15 ± 0.03

0.46 ± 0.07 0.15 ± 0.04

0.43 ± 0.07 0.17 ± 0.02

0.42 ± 0.09 0.16 ± 0.03

0.07 0.6

C. Banzrai et al. / Neuroscience Letters 611 (2016) 81–87

83

Fig. 1. Interval changes in sensory axonal excitability with maturation in mice. Serial recordings of sensory excitability at four time points show the following interval changes: (1) left shift of absolute and relative excitability in stimulus-response curves (A and B); (2) inward shift of the current-threshold relationship (I/V) by hyperpolarizing conditioning currents at 10 weeks (open triangles), with further stabilization at 14 weeks (filled circles), and 19 weeks (filled triangles) (C); (3) unchanged strength-duration relationship (D); (4) increased hyperpolarizing electrotonus (TEh) at 10 weeks, and further stabilization (E); and (5) increased superexcitability at 10 weeks was decreased thereafter, and subexcitability (double stimuli) increased up to 14 weeks then stabilized (F).

maturation (Table 1). Excitability studies showed consistent values for the SDTC, but showed maturation-related changes in the TE, I/V, and RC.

not consistent. The refractoriness did not change significantly (Fig. 1F).

3.4. Stimulus-response and strength-duration relationship 3.2. Threshold electrotonus and current-threshold relationship Depolarizing electrotonus responses were mostly constant with maturation, whereas the early hyperpolarizing response increased until 10 weeks (P = 0.003) and stabilized thereafter. However, the peak hyperpolarizing response increased at 10 weeks and gradually decreased thereafter until 19 weeks, but this was not significant (Fig. 1E). Similarly for the I/V, there was an inward shifting of the hyperpolarizing plots until 10 weeks and this then returned to the outward direction (Fig. 1C).

3.3. Recovery cycle Superexcitability, a reflection of the depolarizing afterpotential, increased abruptly at 10 weeks (P < 0.05) and further decreased below the 6-week-old level, while subexcitability (double stimuli) increased gradually up to 14 weeks (P < 0.05) and stabilized thereafter. The overall subexcitability percentage was

Axonal excitability indices for the stimulus-response (SR) and strength-duration relationship were stable or did not change significantly during maturation (Fig. 1A, B, and D). Other excitability indices, except for those mentioned above, were not changed with maturation in the sensory nerve.

3.5. Trends for changes with maturation The trends for changes in the TE and RC with maturation are shown in Fig. 2A and B. The early hyperpolarizing electrotonus (TEh 20–40 ms) increased from 6 weeks to 10 weeks and then stabilized further. The peak response of the hyperpolarizing TE (TEh peak: −70%) increased from 6 weeks to 10 weeks and then there was a decreasing trend. Superexcitability at 7 ms in the RC increased suddenly from 6 weeks to 10 weeks and then decreased gradually. The subexcitability of double stimuli in the RC increased from 6 weeks to 14 weeks and then stabilized.

84

C. Banzrai et al. / Neuroscience Letters 611 (2016) 81–87

60

5

A

B

40 4

3

0 -20

6

-40

10

14

19

*

-60 -80 -100

Threshold change (%)

Threshold elelctrotonus (%)

20

# 2

1

0

6

10

14

19

-1 -120 -2

-140 -160

-3

Time course (weeks)

TEd (10-20ms)

TEd (90-100ms)

TEh (20-40ms)

TEh (peak: -70%)

Time course (weeks)

Superexcitability at 7 ms Late subexcitability

Fig. 2. The trends for changes in TE and RC indices with maturation. The trends for parameters of sensory excitability that were significantly altered with maturation are shown: significant changes in hyperpolarizing electrotonus (TEh, bottom) and unchanged depolarizing electrotonus (TEd, top) (A); changed superexcitability (bottom) and subexcitability (top) with maturation (B). (#: P < 0.05; *: P < 0.01).

3.6. Sensory nerve excitability changes with aging The age-related changes to sensory nerve excitability were compared for three different time points, 19 weeks, 43 weeks, and 60 weeks (Table 2, Figs. 3 and 4A and B). For almost all the excitability parameters except peak latency and superexcitability, there was no difference between 19 weeks and 43 weeks. Changes became evident by 60 weeks.

change was observed at 60 weeks (P = 0.04), suggesting that highthreshold axons required greater stimulation with age. The increase in threshold was associated with an augmentation in rheobase from 0.31 ± 0.06 mA for 19-week-old mice to 0.52 ± 0.08 mA for 60week-old mice (P = 0.009). There was no significant change in SDTC (Fig. 3D) with age (P = 0.8).

3.10. Trends in changes with aging 3.7. Threshold electrotonus and current-threshold relationship The early phase (TEh 20–40 ms) and peak response (TEh peak: −70%) of the hyperpolarizing TE decreased significantly by 60 weeks (P = 0.007 and P = 0.002, respectively). The depolarizing TE was fairly stable during the time course. There was a trend for hyperpolarizing plots to shift outward in the current-threshold relationship (Fig. 3E), but this change was not significant (P = 0.6).

The trends in the parameters that changed with aging are illustrated in Fig. 4A and B. The relatively stable level in the peak response of the TEh decreased abruptly at 60 weeks. The same trend was seen in early hyperpolarizing threshold accommodation (TEh 20–40 ms). In the RC, superexcitability at 7 ms decreased in 43 week-old mice and stabilized thereafter. The late subexcitability percentage gradually decreased with age.

3.8. Recovery cycle

3.11. Modeling of nerve excitability changes with age

Supernormality significantly decreased from 19 weeks to 43 and 60 weeks, suggesting smaller depolarization after potential or less leaking current. However, late subnormality decreased by 60 weeks without significance, whereas double conditioning stimuli for late subexcitability revealed a significant threshold reduction in slow K+ currents by 60 weeks (Fig. 3F), as previously reported in patients with amyotrophic lateral sclerosis who showed greater sensitivity with double rather than single conditioning stimuli [21].

Table 3 shows the results of modeling studies to identify parameters that explain the interval changes in the excitability studies. With 14 week-old mice as the reference (to represent adolescence), the data from the four study time points were fitted by changing a single parameter, or two parameters if changing a single parameter did not yield a satisfactory discrepancy reduction. (1) The interval difference in the 6-week-olds was best explained by an increase in the GBB; (2) for 10-week-olds, single, two, or three (results not shown) parameter changes were unsatisfactory, but increasing the H conductance (GH) and possibly increasing the GBB best explained the interval changes; (3) the interval changes for 43-week-olds were best explained by a reduction in GKs; (4) the interval changes for 60-week-olds were best explained by a reduction in internodal GKs, followed by increased GBB.

3.9. Stimulus-response and strength-duration relationship The SR curves (Fig. 3A and B) shifted to the right side with age and the stimulus required for 50% of maximal amplitude of the sensory nerve action potential (SNAP) increased. The greatest

C. Banzrai et al. / Neuroscience Letters 611 (2016) 81–87

85

Table 2 Changes in sensory axon excitability parameters with aging in mice (depolarizing threshold electrotonus, TEd; hyperpolarizing threshold electrotonus, TEh; #:P < 0.05; *: P < 0.01). 19 weeks old (A)

43 weeks old (B)

60 weeks old (C)

Amplitude (␮V) Peak latency (ms)

135.7 ± 46.5 1.3 ± 0.08

124.2 ± 47.3 1.1 ± 0.07

98.5 ± 45.2 1.3 ± 0.4

Threshold electrotonus (TE) TEd (10–20 ms) TEd (40–60 ms) TEd (90–100 ms) TEd (undershoot) TEh (10–20 ms) TEh (20–40 ms) TEh (90–100 ms) S2 accommodation TEh (peak: −70%) (−70%) S3 accommodation

45.3 ± 1.8 41.1 ± 1.2 40.2 ± 0.4 −2.9 ± 1.8 −58.5 ± 2.0 −64.9 ± 2.9 −58.9 ± 4.3 4.8 ± 1.9 −137.0 ± 4.8 46.4 ± 12.08

47.6 ± 2.5 43.5 ± 1.2 41.5 ± 1.04 −3.6 ± 4.2 −56.4 ± 3.8 −61.6 ± 4.5 −56.6 ± 8.0 6.0 ± 1.3 −132.4 ± 11.2 42.7 ± 12.3

48.8 ± 5.3 44.5 ± 3.5 44.1 ± 5.4 −14.8 ± 11.5 −64.1 ± 10.9 −76.8 ± 7.8 −82.9 ± 13.5 4.5 ± 0.8 −171.2 ± 17.4 53.5 ± 21.6

Recovery cycle (RC) Refractoriness at 2 ms Superexcitability (%) Superexcitability at Late subexcitability (%) RC2-subexcitability (%)

−6.7 ± 2.7 −7.0 ± 2.4 −2.5 ± 0.6 1.8 ± 1.1 3.6 ± 0.8

−8.4 ± 2.0 −8.8 ± 0.9 −5.2 ± 0.4 1.6 ± 0.9 4.0 ± 0.5

−6.7 ± 2.3 − 7.9 ± 2.6 −5.1 ± 1.8 0.6 ± 1.0 1.8 ± 1.4

0.5 0.4 0.004* 0.3 0.03#

Current/threshold relationship (I/V) Resting I/V slope Minimum I/V slope Hyperpolarizing I/V slope

0.78 ± 0.1 0.65 ± 0.06 1.3 ± 0.8

0.89 ± 0.05 0.59 ± 0.1 0.7 ± 0.0

0.7 ± 0.02 0.53 ± 0.06 1.8 ± 1.9

0.08 0.2 0.6

Stimulus-response (SR) relationship Rheobase Stimulus for 50% max response Strength-duration time constant (SDTC)

0.31 ± 0.06 0.42 ± 0.09 0.16 ± 0.03

0.35 ± 0.07 0.49 ± 0.1 0.17 ± 0.02

0.52 ± 0.08 0.63 ± 0.1 0.18 ± 0.07

0.009 * 0.048 # 0.8

7 ms

Table 3 Modeling of nerve excitability data in comparison with the excitability parameters for 14-week-old mice (discrepancy reduction, DR; Barrett–Barrett conductance, GBB; H conductance, GH; internodal slow K+ conductance, GKsI; nodal slow K+ conductance, GKsN; leak conductance, GLk; internodal pump currents, IPumpI; nodal pump currents, IPumpN; percent persistent Na+ channels, PNap). (1) Best fits obtained by changing single parameters

Parameter GBB GKsI GH IPumpI

6-week-old Change +40% −48% −54% +1200%

DR 69.4% 63.4% 57.1% 46.4%

1 2 3 4

GH GLk GKsN IPumpI

10-week-old +116% +7300% +80% −917%

45.8% 37.2% 35.7% 29.3%

1 2 3 4

GKsI GKsN IPumpI GH

43-week-olda −54% -92% +1220% −53%

71.9% 61.5% 58.6% 53.5%

1 2 3 4

GKsI GBB GH IPumpN

60-week-old −83% +87% −73.8% +1137%

90.0% 83.1% 79.8% 47.1%

1 2 3 4

(2) Best fits obtained by changing pairs of parameters

1 2 3 4 5

Parameter 1 GH GH GH PNap(%) GKsN

Change +264% +153% +118% +44% +31%

10-week-old Parameter 2 GBB IPumpN IPumpI GH GH

Change +53% −151% +190% +98% +65%

DR 50.8% 50.7% 48.7% 48.6% 48.5%

a Sensory excitability test results for 19 week-old and 43 week-old mice were similar.

ANOVAP values 0.6 0.028 # 0.3 0.09 0.17 0.051 0.25 0.007 * 0.005 * 0.3 <0.001* 0.6

Post Hoc

P values

0.03 # (A/B)

0.021# (A/C) 0.007* (B/C) 0.009* (A/C) 0.007* (B/C) 0.005* (A/C) 0.003* (B/C)

0.006* (A/B) 0.014* (A/C) 0.03# (B/C)

0.008 * (A/C) 0.03 # (B/C) 0.04 # (A/C)

4. Discussion This study has identified age-related changes in sensory axonal membrane properties associated with the maturation and aging of peripheral nerves in mice. Our results showed dynamic alterations in both the passive membrane properties and conductances of multiple ion channels.

4.1. Variability of axonal excitability Several factors influence axonal excitability, such as the type of fiber (e.g., sensory vs. motor), species (human vs. animal), age/developmental stages, and location (i.e., proximal vs. distal). Previous reports have indicated dynamic changes in axonal excitability with maturation and aging, primarily based on studies of motor excitability [3,22]. Boërio, et al. studied caudal motor nerves in 4–19-week-old mice and showed a sharp decrease in early TEh from weeks 4–13, while superexcitability progressively increased. Yang, et al. studied rat caudal motor nerves in three age groups (i.e., immature, young, and mature), and showed that early TEh decreased and stabilized thereafter, while there were still changes in the TEd. Although the reason for the different trends in depolarizing electrotonus in these studies is unclear, it is possible that different age ranges and recording/stimulating techniques contributed to the changes [15]. In contrast to the results of motor studies, we are not aware of any age-related changes to sensory axons in animals. In humans, changes were restricted to threshold measures and the stimulus-response slope [14]. In our study, hyperpolarizing currents in the TE decreased significantly. This finding suggests that the Ih became lower with aging. The possibility of a hyperpolarizing shift in resting potential is unlikely, because there should have been associated changes in other measures; specifically, an increase in depolarizing accommodation (fanning-out), refractoriness, and the relative refractory period (RRP), and a decrease in the SDTC [11].

86

C. Banzrai et al. / Neuroscience Letters 611 (2016) 81–87

Fig. 3. Interval changes in sensory axonal excitability in aged mice. Sensory axonal excitability tests at three different time points (19 weeks, filled triangles; 43 weeks, filled squares; and 60 weeks, open squares) show the following interval changes: (1) right shift of absolute and relative excitability in stimulus-response curves (P = 0.04, at 60 weeks) (A and B); (2) outward shift of hyperpolarizing conditioning currents that was not significant (C); (3) stable strength-duration relationship (D); (4) significantly decreased early and peak response of hyperpolarizing threshold electrotonus (TEh) at 60 weeks and stable depolarizing response (E); and (5) decreased superexcitability and late subexcitability (double stimuli) at 60 weeks (F).

The modeling study has suggested that the following three conductances explain the interval changes in sensory excitability with maturation and aging: (1) GBB, (2) internodal GKs (GKsI), and (3) GH. Maturation was accompanied by a reduction in the GBB, and an increase in the GKs and GH, whereas with aging there was an increase in the GBB, and a reduction in the GKs and GH conductance. Possible interpretations are discussed in the following sections.

4.2. Alterations in conductance 4.2.1. The Barrett and Barrett conductance According to the Barrett and Barrett axoplasmic membrane model, an internode is ‘leaky’ and the current that passes through the internode (axo-glial junction) is known as the Barrett and Barrett conductance [2]. The GBB conductance in sensory axons was reduced with maturation up to the young-adult stage, reflecting myelination and the formation of axo-glial junctions, which is consistent with the changes previously described for human motor axons [6]. By contrast, the GBB increased in aged animals, reflecting the progressive disruption of myelination and axo-glial junctions. Furthermore, although the GBB could explain the excitability changes over time, it is related to the function of the axo-glial junction, as seen in the excitability changes with maturation [6].

4.3. Slow K+ conductance Maintenance of neuronal excitability is crucial for adequate functioning of the nervous system. Many interacting factors contribute to nerve excitability, including nodal and internodal ion channels, the Na+ -K+ -ATPase pump, the membrane potential, and passive membrane properties. Slow K+ channels, which are distributed diffusely but are predominantly located at the nodes of Ranvier in nerves, become activated to prevent the axonal membrane from being excessively excitable. A reduction in the GKs with aging has not been reported in previous studies. In our study, a GKs reduction best explained the change in the 60-week-old mice. It is also consistent with a reduction in late subexcitability in the recovery cycle. 4.4. H conductance Hyperpolarization-activated cyclic nucleotide gated (HCN) channels conduct Ih, and the HCN1 and HCN2 subunits are thought to be the dominant forms in the peripheral nervous system. They are located in the internodal rather than the nodal region. HCN channels are activated by membrane hyperpolarization near the sub-threshold potential and this allows repetitive firing of the axon. HCN channels play a key role in determining the resting membrane potential [8]. Greater Ih in human motor axons has been related to

C. Banzrai et al. / Neuroscience Letters 611 (2016) 81–87

3

A

50

B 2

25

1

0

19

43

60

-25

# -50

*

-75

TEd (10-20ms) -100

TEd (90-100ms)

Threshold change (%)

Threshold electrotonus (%)

87

0

19

43

60

-1

Superexcitability at 7 ms

-2

Late subexcitability -3

TEh (20-40ms)

-125

TEh (peak: -70%)

-4

-150 -5 -175

*

-200

*

Time course (weeks)

* -6

# Time course (weeks)

Fig. 4. The trends for changes in TE and RC parameters with aging. The trends for parameters of sensory excitability that were significantly altered with aging are shown: fairly stable depolarizing electrotonus (TEd, top) and significantly altered hyperpolarizing electrotonus (TEh, bottom) (A); decreased subexcitability (top) and superexcitability (bottom) (B). (#: P < 0.05; *: P < 0.01).

aging [10]. Alterations in the GH in sensory axons could be related to changes in the internodal length and axonal diameter, resulting in an increase with maturation and a reduction with aging [1,5]. 5. Conclusion In conclusion, the present study showed dynamic agedependent changes of axonal excitability in the tail sensory nerve in mice. Computer modeling showed the most relevant excitability parameters to account for the dynamic changes were the Barrett–Barrett, H, and slow K+ conductances. References [1] A.M. Adinolfi, J. Yamuy, F.R. Morales, M.H. Chase, Segmental demyelination in peripheral nerves of old cats, Neurobiol. Aging 12 (1991) 175–179. [2] E.F. Barrett, J.N. Barrett, Intracellular recording from vertebrate myelinated axons: mechanism of the depolarizing afterpotential, J. Physiol. 323 (1982) 117–144. [3] D. Boërio, L. Greensmith, H. Bostock, Excitability properties of motor axons in the maturing mouse, J. Peripher. Nerv. Syst. 14 (2009) 45–53. [4] H. Bostock, K. Cikurel, D. Burke, Threshold tracking techniques in the study of human peripheral nerve, Muscle Nerve 21 (1998) 137–158. [5] M.H. Chase, J.K. Engelhardt, A.M. Adinolfi, S.S. Chirwa, Age-dependent changes in cat masseter nerve: an electrophysiological and morphological study, Brain Res. 586 (1992) 279–288. [6] M.A. Farrar, S.B. Park, C.S. Lin, M.C. Kiernan, Evolution of peripheral nerve function in humans: novel insights from motor nerve excitability, J. Physiol. 591 (2013) 273–286. [7] A. George, H. Bostock, Multiple measures of axonal excitability in peripheral sensory nerves: an in vivo rat model, Muscle Nerve 36 (2007) 628–636. [8] J. Howells, L. Trevillion, H. Bostock, D. Burke, The voltage dependence of I(h) in human myelinated axons, J. Physiol. 590 (2012) 1625–1640. [9] J.M. Jacobs, S. Love, Qualitative and quantitative morphology of human sural nerve at different ages, Brain 108 (Pt. 4) (1985) 897–924.

[10] S.K. Jankelowitz, P.A. McNulty, D. Burke, Changes in measures of motor axon excitability with age, Clin. Neurophysiol. 118 (2007) 1397–1404. [11] M. Kiernan, H. Bostock, Effects of membrane polarization and ischaemia on the excitability properties of human motor axons, Brain 123 (Pt. 12) (2000) 2542–2551. [12] M.C. Kiernan, D. Burke, K.V. Andersen, H. Bostock, Multiple measures of axonal excitability: a new approach in clinical testing, Muscle Nerve 23 (2000) 399–409. [13] M.C. Kiernan, G.K. Isbister, C.S. Lin, D. Burke, H. Bostock, Acute tetrodotoxin-induced neurotoxicity after ingestion of puffer fish, Ann. Neurol. 57 (2005) 339–348. [14] M.C. Kiernan, C.S. Lin, K.V. Andersen, N.M. Murray, H. Bostock, Clinical evaluation of excitability measures in sensory nerve, Muscle Nerve 24 (2001) 883–892. [15] M. Moldovan, S. Alvarez, C. Krarup, Motor axon excitability during Wallerian degeneration, Brain 132 (2009) 511–523. [16] H. Nodera, R. Kaji, Nerve excitability testing and its clinical application to neuromuscular diseases, Clin. Neurophysiol. 117 (2006) 1902–1916. [17] H. Nodera, S.B. Rutkove, Changes of the peripheral nerve excitability in vivo induced by the persistent Na+ current blocker ranolazine, Neurosci. Lett. 518 (2012) 36–40. [18] Y. Osaki, H. Nodera, C. Banzrai, S. Endo, H. Takayasu, A. Mori, Y. Shimatani, R. Kaji, Effects of anesthetic agent s on in vivo axonal HCN current in normal mice, Clin. Neurophysiol. 126 (10) (2015) 2033–2039. [19] M. Panizza, J. Nilsson, B.J. Roth, J. Rothwell, M. Hallett, The time constants of motor and sensory peripheral nerve fibers measured with the method of latent addition, Electroencephalogr. Clin. Neurophysiol. 93 (1994) 147–154. [20] J.M. Schröder, J. Bohl, K. Brodda, Changes of the ratio between myelin thickness and axon diameter in the human developing sural nerve, Acta Neuropathol. 43 (1978) 169–178. [21] Y. Shibuta, H. Nodera, A. Nodera, T. Okita, K. Asanuma, Y. Izumi, R. Kaji, Utility of recovery cycle with two conditioning pulses for detection of impaired axonal slow potassium current in ALS, Clin. Neurophysiol. 121 (2010) 2117–2120. [22] Q. Yang, R. Kaji, N. Hirota, Y. Kojima, T. Takagi, N. Kohara, J. Kimura, H. Shibasaki, H. Bostock, Effect of maturation on nerve excitability in an experimental model of threshold electrotonus, Muscle Nerve 23 (2000) 498–506.