Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Gap Junctions Contribute to the Regulation of Walking-Like Activity in the Adult Mudpuppy (Necturus Maculatus)

  • Igor Lavrov,

    Affiliation Institute of Fundamental Medicine and Biology, Kazan Federal University, Kazan, Russia

    ⨯
  • Lyle Fox,

    Affiliation Departments of Pain Management and Neurosciences, Anesthesiology Institute, Cleveland Clinic, Cleveland, Ohio, United States of America

    ⨯
  • Jun Shen,

    Affiliation Departments of Pain Management and Neurosciences, Anesthesiology Institute, Cleveland Clinic, Cleveland, Ohio, United States of America

    ⨯
  • Yingchun Han,

    Affiliation Department of Pathobiology, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio, United States of America

    ⨯
  • Jianguo Cheng

    chengj@ccf.org

    Affiliation Departments of Pain Management and Neurosciences, Anesthesiology Institute, Cleveland Clinic, Cleveland, Ohio, United States of America

    ⨯

Abstract

Although gap junctions are widely expressed in the developing central nervous system, the role of electrical coupling of neurons and glial cells via gap junctions in the spinal cord in adults is largely unknown. We investigated whether gap junctions are expressed in the mature spinal cord of the mudpuppy and tested the effects of applying gap junction blocker on the walking-like activity induced by NMDA or glutamate in an in vitro mudpuppy preparation. We found that glial and neural cells in the mudpuppy spinal cord expressed different types of connexins that include connexin 32 (Cx32), connexin 36 (Cx36), connexin 37 (Cx37), and connexin 43 (Cx43). Application of a battery of gap junction blockers from three different structural classes (carbenexolone, flufenamic acid, and long chain alcohols) substantially and consistently altered the locomotor-like activity in a dose-dependent manner. In contrast, these blockers did not significantly change the amplitude of the dorsal root reflex, indicating that gap junction blockers did not inhibit neuronal excitability nonselectively in the spinal cord. Taken together, these results suggest that gap junctions play a significant modulatory role in the spinal neural networks responsible for the generation of walking-like activity in the adult mudpuppy.

Citation: Lavrov I, Fox L, Shen J, Han Y, Cheng J (2016) Gap Junctions Contribute to the Regulation of Walking-Like Activity in the Adult Mudpuppy (Necturus Maculatus). PLoS ONE 11(3): e0152650. https://doi.org/10.1371/journal.pone.0152650

Editor: Alexandre Hiroaki Kihara, Universidade Federal do ABC, BRAZIL

Received: April 23, 2015; Accepted: March 17, 2016; Published: March 29, 2016

Copyright: © 2016 Lavrov et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the paper and its Supporting Information files.

Funding: This work was supported by grants to Jianguo Cheng from the National Institutes of Health (NINDS R01 NS052372), the Kentucky Spinal Cord and Head Injury Research Trust (KSCHIRT, Grant# 0-4), and the Cleveland Clinic Anesthesiology Institute. Partial support for analysis of the effects of gap junction inhibitors on locomotor activity was provided to Igor Lavrov by the Russian Science Foundation (RSF grant № 15-15-20036). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Gap junctions are specific structures that link the cytoplasm of adjoining cells and enable direct electrical communications between them. Early in development, intercellular electrical coupling through gap junctions is primarily involved in neurogenesis and axonal targeting [1–3]. At later stages, electrical coupling between neurons contributes to the generation of rhythmic activities in neuronal networks [4–5]. Electrical coupling tends to synchronize spontaneous activities in different brain regions including the neocortex [6, 7], cortex [5, 8], brainstem [9], embryonic retina [10], and the spinal cord [11–12]. Although their roles in development have been extensively studied, much less is known about their function in adult spinal cord [13].

The in vitro forelimb-spinal cord preparation from the adult mudpuppy provides a unique opportunity to address this important question. It generates robust and stable walking-like activity that can last for several days [14–19]. The locomotor-like activity is induced by NMDA or glutamate and is manifested as alternating flexion and extension of the forelimb around the elbow joint and alternating electromyographic (EMG) bursts between the elbow flexor and extensors. The mudpuppy spinal cord may also express gap junctions as suggested by indirect anatomical and electrophysiological evidences [20, 21]. Immunohistochemistry studies suggest that gap junctions expressed in the mudpuppy retina are composed of proteins similar to those found in mammals such as mouse, rat, and human [21, 22]. Pharmacological agents that inhibit electrical coupling in mammals, also inhibit gap junction communication in the mudpuppy and other amphibians [21, 23–26].

In this study we addressed two fundamental questions. Are gap junction proteins expressed in the spinal cord of the adult mudpuppy? If they are expressed, do they contribute to the neural networks for walking? Using immunohistochemistry, we demonstrated that several connexins were expressed in the adult mudpuppy spinal cord. Using four gap junction blocks from three different classes, we demonstrated that gap junctions function to regulate the rhythmicity and amplitude of locomotor-like activity.

Materials and Methods

Animals

A total of 34 mudpuppies were used for the experiments: 6 for gap junction immunohistochemistry and 28 for pharmacological experiments. We only used adult animals with body lengths of 20–30 cm, which indicate the maturity of the animals [14–19]. The experimental protocols were approved by the Animal Care and Use Committee (IACUC) of the Cleveland Clinic.

Immunohistochemistry analysis

Immunohistochemistry was performed according to manufacturer’s instructions and the protocols described [27]. Animals (n = 6) were anesthetized by bath application of tricaine methanesulfonate (MS222) (1.5 g/L) (Sigma, St. Louis, MO) and a dorsal laminectomy was performed on the first six segments of the spinal cord. Segments 1–5 of the spinal cord were removed and fixed with 4% paraformaldehyde in 0.1 M phosphate buffer (pH 7.4) for 2 hours at room temperature. The tissue samples were cryoprotected in 30% sucrose overnight at 4°C. The sections were rehydrated and then blocked with 0.3% Triton X-100 and 3% normal goat serum for 45 to 60 min. Then the slides were incubated overnight at 4°C with the primary antisera directed against mammalian connexins: Cx32 (Alpha Diagnostic, San Antonio TX, USA; Catalog # CX32A11-A), Cx36 (Invitrogen, Carlsbad, CA, USA; Catalog # 36–4600), Cx37 (Alpha Diagnostic; Catalog # CX37A11-A) and Cx43 (Alpha Diagnostic; Catalog # CX43B12-A). Then, the slides were washed with the incubation buffer and the primary antibodies were visualized by incubating for 45 min with Alexa Fluor 488 conjugated goat anti-rabbit antisera (1:1000–1500, Invitrogen, Catalog # A11008). For co-localization of connexin with the astrocyte marker glial fibrillary acidic protein (GFAP), the sections were simultaneously labeled with primary antisera directed against one of the connexins and chicken anti-GFAP (Millipore, Billerica, MA, USA, Catalog # AB5541). The anti-GFAP was visualized with Texas Red conjugated goat anti-chicken (Jackson Immunoresearch, West Grove, PA, USA, Catalog # 103-075-155). The sections were washed with PBS and coverslipped using Vectashield mounting medium (Vector Laboratories, Burlingame, CA, USA) containing DAPI and were viewed under a Leica DM5500 B upright microscope with fluorescent optics and Leica TCS-SP spectral laser scanning confocal microscope. Control sections were handled in a similar manner except that the primary antisera were not added to the incubation buffer. Images were captured using a CCD video camera and QCapture Pro imaging software from QImaging (Surrey, BC, Canada).

In vitro preparation

The spinal cord-forelimb preparation was described elsewhere [14–19, 28–30]. Briefly, animals (n = 28) were anesthetized by bath application of MS222 (1.5 g/l). A dorsal laminectomy was performed and the first five segments of the spinal cord with part of the vertebral columns, the brachial nerves, and the forelimb were isolated from the rest of the body. The preparation was placed in a Petri dish containing 100% oxygenated Ringer’s solution (NaCl 115 mM, KCl 2 mM, CaCl2 2 mM, MgCl2 1.8 mM, HEPES 5 mM and glucose 1 g/l, pH 7.35). Then, the brachial plexus was exposed and the paraspinal muscles were removed. After dissection, the preparation was transferred to a dish with separate recording chambers for the forelimb and spinal cord. The vertebral column was stabilized in the dish by pinning it to the Sylgard (Dow Corning) base of the chamber. The preparation was continuously perfused with cooled (15°C) oxygenated Ringer’s solution throughout the entire experiment at a flow rate of 5 ml/min unless otherwise stated. After a recovery period of 1 hour all preparations showed a withdrawal reflex in response to pinching of the limbs with a pair of blunt forceps. Pairs of Teflon-coated silver wires (75 μm) were inserted into the elbow flexor (Brachialis) and extensor (Extensor ulnae) muscles for EMG recording. The signals were amplified, high-pass (10 Hz) and low-pass (300 Hz) filtered, monitored, and stored in a computer. Commercially available programs were used for recording and data analysis of walking-like activity (Axon Instruments/Molecular Devices, Sunnyvale, CA; Origin, Northampton, MA and SPSS, Chicago, IL).

Neuroactive agent application

Walking-like activity was induced by continuous perfusion (20 ml/min) of D-glutamic acid (0.5 mM, Sigma, St. Louis, MO) or by NMDA (50 μM, Sigma, St. Louis, MO) with 10 μM D-serine (Sigma, St. Louis, MO) as described previously [14–15, 17–19]. Individual gap junction uncoupling agents were added to the D-glutamic acid or NMDA solutions after the walking-like activity stabilized. Dose response curves were generated for four different uncoupling agents, carbenoxolone (CBX, 0.01–2 mM), flufenamic acid (FFA, 0.01–2 mM), 1-Heptanol and 1-Octanol (0.01–3 mM) (Sigma, St. Louis, MO). Stock solutions of CBX, 1-Heptanol, and 1-Octanol were prepared in Ringer’s solution. These blockers are less toxic than other compounds and they are water soluble or form emulsions in 25% ethylene glycol reducing the need for solvents such as DMSO and ethanol that affect the excitability of spinal neurons [19, 31, 32]. FFA was dissolved in 25% polyethylene glycol (Sigma, St. Louis, MO), as it has been shown that polyethylene glycol is not neurotoxic but neuroprotective [33]. All drugs were diluted in 100–200 ml of Ringer’s solution immediately before use. Drugs were washed out after each application with Ringer’s solution for at least 30 min before the next drug application.

Dorsal root reflex testing

The effects of the gap junction inhibitors on the excitability of spinal neurons were additionally monitored using the dorsal root reflex. The C3 dorsal root was stimulated with a 0.5 ms constant cathode current pulse using a stimulator (S88; Grass-Telefactor, Astro-Med Inc, West Warwick, RI, USA) and stimulus isolation unit (PSIU6; Grass-Telefactor, Astro-Med Inc, West Warwick, RI, USA). The stimulus intensity was gradually increased up to 1.5 times the motor threshold and the amplitude of the fast monosynaptic component of response was evaluated from EMG recordings obtained from the bipolar electrodes (A-M System, Carlsborg, WA).

Data analysis

Measurements of the cycle frequency of EMG bursts were means of 20 cycles. EMG records were rectified and integrated for quantification as we and others have previously described [14–17, 29–30]. Integrated areas of the EMG from the flexor muscle (Brachialis) were averages of 10 bursts and normalized to the muscle discharge before drug application for each preparation. All data were reported as mean ± SE. Statistical significance of different groups was evaluated using one-way ANOVA with repeated measures and for non-parametric data we used Kruskal Wallis analysis. The criterion for statistical significance was set as P < 0.05 for all experiments. We used Sigmastat for statistical analysis and Sigmaplot for graph generation (Systat Software Inc).

Results

Immunohistochemistry of gap junctions in the adult mudpuppy spinal cord

Sections of the spinal cord were incubated with antisera directed against four different gap junctions’ proteins, connexin 32 (Cx32), connexin 36 (Cx36), connexin 37 (Cx37), and connexin 43 (Cx43). All four anti-connexin antisera generated punctuate labeling in the mudpuppy spinal cord and the prevalence and pattern of staining were different for each antisera. The Cx43-like staining was more abundant than the staining for the other connexins. The Cx43 antisera produced a punctuate staining on the plasma membranes of the ependymal cells surrounding the central canal (Fig 1A), punctuate staining near nuclei in the ventral horn (Fig 1C), and strings of puncta radiating outward to the sub-pial plexus (Fig 1B). Since Cx43 is predominantly expressed by astrocytes in mammals, we investigated whether Cx43 is expressed by astrocytes in the mudpuppy spinal cord by co-labeling with anti-Cx43 and antisera directed against an astrocyte marker glial fibrillary acidic protein (anti-GFAP). Images from co-labeled spinal cords demonstrated that the punctuate anti-Cx43 (green) fluorescence paralleled the anti-GFAP (red) fluorescence, a staining pattern consistent with astrocytic expression of Cx43 (Fig 1D and 1E). The Cx32 and Cx36 antibodies produced a dense punctuate staining near the nuclei of a few cells in the ventral horn and a sparse punctuate pattern in the white matter (S1 Fig). The Cx37-like immunoreactivity was rare with only a few puncta in the white matter of each section (not presented). It was not practical to determine whether the patterns of Cx32, Cx36, or Cx37 immunoreactivity co-localized with GFAP immunoreactivity because of the scarcity of Cx32, Cx36, or Cx37 staining.

thumbnail
Download:
Fig 1. Examples of Cx43-like immunoreactivity in coronal sections of the mudpuppy spinal cord.

Anti-Cx43 produced punctuate staining (green) near nuclei in the ventral horn and strings of punta radiating outward to the sub-pial plexus (A-C). Confocal images showing the colocalization of Cx43-like (green) and GFAP-like (red) immunofluorescence patterns in coronal sections of the mudpuppy spinal cord (D-E). The wide-spread distribution of the Cx43-like immunoreactivity parallels the GFAP-like staining in the gray and white matters. The confocal images are from a single optical plane. The photos on the right are higher magnification images of the regions enclosed by the white boxes. Nuclei in all images were counter stained with DAPI (blue). The scale bar in A = 200 μM; in B and C = 50 μM; in D = 100 μM; and in E = 250 μM.

https://doi.org/10.1371/journal.pone.0152650.g001

Modulation of walking-like movements and alternating flexor-extensor EMG activity by gap junction inhibitors

Regular rhythmic walking-like limb movement around the elbow joint and alternating flexor-extensor EMG activity were induced within the first few minutes after the application of NMDA with D-serine or D-glutamate. Fig 2 shows a representative recording of EMGs from the elbow flexor (Brachialis) and extensor (Extensor ulnae) muscles induced by NMDA (segment 1, interval before “FFA perfusion On”). The initial activity pattern consisted of alternating bursts of rhythmic activity from the flexor and extensor muscles. The average cycle frequency was 0.75 ± 0.11 Hz for D-glutamate-induced activity and 0.62 ± 0.08 Hz (n = 28; p < 0.05) for NMDA-induced activity. The average amplitude of the EMGs was 11 ± 3.2 mV for the Brahialis and 8.4 ± 2.1 mV for the Extensor ulnae.

thumbnail
Download:
Fig 2. Depiction of the design for electrophysiology experiments.

The figure presents the EMG activity simultaneously recorded from the flexor and extensor muscles during a typical experiment. Walking-like activity was induced with 50 μM NMDA and 10 μM D-serine. The initiation and termination of FFA application (300 μM) is indicated with arrows. Segments used for analysis of the initial changes and tested intervals are shown.

https://doi.org/10.1371/journal.pone.0152650.g002

The effects of the gap junction inhibitors were dose-dependent and also related to the duration of the perfusion. Fig 2 shows the effects of 0.5 mM FFA on the rhythmic activity induced by NMDA, which was continuously infused throughout the whole recording period. FFA produced an increase in cycle frequency during the first 1–2 minutes of perfusion (segment 2, “FFA perfusion On”) and decreased the frequency at later intervals (segment 3, “Tested interval”). In all cases, the walking-like activity was completely restored upon washing out of FFA with Ringer’s solution (segment 5, interval after “FFA perfusion Off”). Taking into account the time-dependent effects of the gap junction inhibitors, we analyzed the EMG activity only after the initial changes were stabilized (segment 3, “Tested interval”).

Perfusion with low concentrations of gap junction inhibitors had a stimulatory effect on the walking-like activity, increasing the cycle frequency and/or EMG amplitude. CBX at a concentration of 50 μM increased the cycle frequency (~ 4%) and amplitude (~ 8%) of the walking-like activity induced by D-glutamate (Fig 3A and 3B). The walking-like activity induced by NMDA was less sensitive to CBX and it was not affected by CBX at concentrations < 500 μM (Fig 3). At much higher concentrations, CBX produced differential effects on the cycle frequency and EMG amplitude. The cycle frequency was reduced by CBX at concentrations > 500 μM (p < 0.05; n = 7) (Fig 3A) while the EMG amplitude was increased by CBX at concentrations >1,300 μM (p < 0.05; n = 7).

thumbnail
Download:
Fig 3. CBX modulates the NMDA- and D-glutamate-induced walking-like activity.

A: Dose-response curves showing the effects of CBX on the cycle frequency. The mean frequency of D-glutamate- (●) and NMDA-induced (○) motor activity patterns were reduced by CBX in a dose-dependent manner. D-glutamate-induced activity was inhibited at lower concentrations than the NMDA-induced activity. B: Dose-response curves showing the effects of CBX on the peak EMG amplitude for D-glutamate-induced activity (●), NMDA-induced activity (○), and the dorsal root reflex (DRR, ▼). The amplitude of D-glutamate-induced activity was inhibited by CBX (concentrations > 100 μM) in a dose-dependent manner, whereas, the amplitude of NMDA-induced patterns were facilitated by CBX (concentrations > 1000 μM). In contrast, the amplitudes of the dorsal root reflexes (DRR) were not significantly affected by CBX. C: Paired recordings from the flexor and extensor muscles showing the effects of CBX on D-glutamate- (right) and NMDA-induced (left) motor activity patterns. Notice that the cycle frequency and EMG amplitude were reduced in the presence of 200 μM CBX (lower pair of records) compared to the control saline (upper pair of records). Values of the means and SEMs as well as their statistical significance are mentioned in the text. Data presented in all graphs are means ± SEMs.

https://doi.org/10.1371/journal.pone.0152650.g003

Application of CBX alone at a concentration that facilitated the EMGs amplitude (1,500 μM) did not induce any rhythmic activity. Instead, it prevented the initiation of walking activities by NMDA (20–100 μM), which only produced tonic activity in the presence of CBX (data not shown). The initiation of walking-like activity by D-glutamate was also completely blocked by CBX at concentrations > 300 μM (Fig 3A and 3B). In contrast, the amplitude of the dorsal root reflex was not significantly affected (p > 0.05; n = 7) (Figs 3B and 4B), indicating that the excitability of motoneurons were not inhibited nonspecifically [34–39]. This is consistent with the observation that CBX increased the EMG amplitude. These data suggest that electrical coupling through gap junctions is required to generate walking-like activity by glutamate or NMDA.

thumbnail
Download:
Fig 4. FFA modulates the NMDA- and D-glutamate-induced walking-like activity.

A: Dose-response curves showing the effects of FFA on the cycle frequency. The mean frequency of D-glutamate- and NMDA-induced motor activity patterns were reduced by FFA concentrations > 100 μM in a dose-dependent manner. D-glutamate-induced activity was inhibited at lower concentrations than the NMDA-induced activity. B: Dose-response curves showing the effects of FFA on the peak EMG amplitude for D-glutamate-induced activity, NMDA-induced activity, and the dorsal root reflex. The amplitude of D-glutamate- and NMDA-induced activity was inhibited by FFA concentrations >100 μM in a dose-dependent manner. In contrast, the amplitudes of the dorsal root reflexes were not significantly affected by FFA at concentrations < 700 μM. C: Paired recordings from the flexor and extensor muscles showing the effects of FFA on D-glutamate- (right) and NMDA-induced (left) motor activity patterns. Notice that the cycle frequency and EMG amplitude were reduced in the presence of 500 μM FFA (lower pair of records) compared to the control saline (upper pair of records). Abbreviations, same as in Fig 3.

https://doi.org/10.1371/journal.pone.0152650.g004

Low concentrations of other gap junction inhibitors also caused a transient increase in the cycle frequency of the walking-like activity induced by D-glutamate, or to a less extent, by NMDA. After 3 min of perfusion with 100 μM FFA, the mean cycle frequency of the NMDA-induced activity increased by 16 ± 6% (p < 0.05; n = 7) (Fig 4A). The amplitude of the EMG bursts was not facilitated by at this concentration (Fig 4B). At concentrations > 300 μM, FFA produced a dose-dependent inhibition of both cycle frequency and EMG burst amplitude. It completely blocked the activity induced by D-glutamate at 400–500 μM and the activity induced by NMDA at 1200–1300 μM (Fig 4A and 4B). The effects of FFA at 400 μM on activities induced by NMDA and glutamate are shown in Fig 4C.

Similarly, Heptanol at 100 μM increased the cycle frequency by 15 ± 6% in NMDA-induced activity and 36 ± 4% in D-glutamate-induced activity (p < 0.05; n = 7) (Fig 5A). Both the cycle frequency and burst amplitude of the activity induced by glutamate were inhibited by Heptanol at concentrations > 600 μM, whereas those induced by NMDA were inhibited at concentrations > 800 μM (p < 0.05; n = 7) (Fig 5). Octanol at 300 μM increased the burst amplitude by 13 ± 7% (p < 0.05; n = 7) but did not affect the cycle frequency of the walking-like activity induced by NMDA (Fig 6A). At concentrations >400 μM, Octanol inhibited the locomotor-like activities induced by either glutamate or NMDA (p < 0.05; n = 7) (Fig 6). These long chain alcohols reduced the dorsal root reflex by ~20%, as exemplified in Fig 5B, but inhibited the locomotor-like activity to a much greater extent.

thumbnail
Download:
Fig 5. Heptanol modulates the NMDA- and D-glutamate-induced walking-like activity.

A: Dose-response curves showing the effects of Heptanol on the cycle frequency. The mean frequency of D-glutamate- and NMDA-induced (○) motor activity patterns was facilitated by Heptanol at low concentrations (100–300 μM) and reduced at concentrations >400 μM in a dose-dependent manner. D-glutamate-induced activity was inhibited at lower concentrations than the NMDA-induced activity. B: Dose-response curves showing the effects of Heptanol on the peak EMG amplitude for D-glutamate-induced activity, NMDA-induced activity, and the dorsal root reflex. The amplitude of D-glutamate- and NMDA-induced activity was facilitated by Heptanol at low concentrations (100–300 μM) and depressed at concentrations >400 100–300 μM. The amplitudes of the dorsal root reflexes were reduced (~20%) by Heptanol at 300 μM. C: Paired recordings from the flexor and extensor muscles showing the effects of Heptanol on D-glutamate- (right) and NMDA-induced (left) motor activity patterns. Notice that the cycle frequency and EMG amplitude were reduced in the presence of 700 μM Heptanol (lower pair of records) compared to the control saline (upper pair of records). Abbreviations, same as in Fig 3.

https://doi.org/10.1371/journal.pone.0152650.g005

thumbnail
Download:
Fig 6. Octanol modulates the NMDA- and D-glutamate-induced walking-like activity.

A: Dose-response curves showing the effects of Octanol on the cycle frequency. The mean frequency of D-glutamate- and NMDA-induced motor activity patterns were reduced by Octanol concentrations >400 μM in a dose-dependent manner. B: Dose-response curves showing the effects of Octanol on the peak EMG amplitude for D-glutamate- and NMDA-induced activity. The amplitude of D-glutamate- and NMDA-induced activity was inhibited by Octanol concentrations >400 μM in a dose-dependent manner. C: Paired recordings from the flexor and extensor muscles showing the effects of Octanol on D-glutamate- (right) and NMDA-induced (left) motor activity patterns. Notice that the cycle frequency and EMG amplitude were reduced in the presence of 700 μM Octanol (lower pair of records) compared to the control saline (upper pair of records). Abbreviations, same as in Fig 3.

https://doi.org/10.1371/journal.pone.0152650.g006

The walking-like activity induced by D-glutamate was more sensitive to gap junction inhibitors than that induced by NMDA (Table 1). The half maximal inhibitory concentration (IC50) of CBX on the cycle frequency of glutamate-induced activity was 150 μM and it was ~ 10 fold higher for the NMDA-induced walking activity (Fig 3A). The IC50 of FFA was 300 μM for glutamate-induced activity and 900 μM for NMDA-induced activity (Fig 4A). The IC50 of Heptanol was 900 and 1100 μM (Fig 5A) and the IC50 of Octanol was 600 and 800 μM respectively for the glutamate- and NMDA-induced walking activities (Fig 6A).

thumbnail
Download:
Table 1. The half maximal inhibitory concentration (IC50) of gap junction blockers on the locomotor-like activities.

https://doi.org/10.1371/journal.pone.0152650.t001

Discussion

Gap junctions consist of six protein subunits (connexins, Cxs) [40] and in vertebrates, they are found in almost all tissues. They are regulated in vivo by a number of factors including pH, voltage, calcium levels, and phosphorylation [41–43]. Their expression in the mammalian spinal cord was originally described as transient with ubiquitous presence in immature animals and less common or even absent in adults [44–47]. Anatomical and physiological studies indicate that gap junctions may serve important physiological functions in the brain and spinal cord in the mammalian [3, 48–49] and vertebrate animals such as the frog [47, 50], the lamprey [51], and the goldfish [52]. However, they may play different roles in the developing and the mature stages of the central nervous system. For example, in the mammalian respiratory system, gap junction blockers cause a dose-dependent inhibition of rhythmic activity in neonatal animals, but an increase of rhythmic activity in adults [53, 54].

Our data indicate that Cx43 is the most abundant gap junction protein in the adult mudpuppy spinal cord, both in the white and gray matters (Fig 1). This is consistent with findings in the rat [55]. Cx43 has been shown to form channels between astrocytes and astrocytes and oligodendrocytes with immunoreactivity primarily observed in astrocytes and rarely in microglia. The ubiquity of Cx43 allows astrocytes to form a functional syncytium that links various regions of the CNS parenchyma in terms of flow of current, ions, and small metabolites [56]. This system also acts as a K+ sink following intense neuronal activity [57] and allows propagation of Ca2+ waves [58–60]. Our data raised the possibility that Cx43 and astrocytes may be involved in the regulation of locomotion.

We found that Cx36 was expressed sparsely in the cell body and the processes of neurons in the spinal cord. A previous study has shown that Cx36 was strongly expressed in immature neurons in rats [61]. It is possible that the expression has been reduced during development and the remaining Cx36 in adulthood may play a role in the modulation of locomotor activity by synchronizing neuronal activities. We observed positive staining for Cx32 in areas surrounding neurons. This is consistent with the finding that Cx32 was expressed predominantly in oligodendrocytes throughout adulthood [61]. In contrast, the staining for Cx37 was sparse in the mudpuppy spinal cord, consistent with the finding that Cx37 is only abundant in embryonic tissues and is expressed in the endothelium of large blood vessels, but not in capillaries, in the brain and other tissues [62]. In summary, the immunohistochemistry data provide evidence that specific gap junction proteins are expressed in the adult mudpuppy spinal cord and may be involved in the generation or modulation of locomotor activities.

Using a battery of gap junction inhibitors, we demonstrated a critical role the gap junction proteins may play in the neural networks for locomotion. It is clear that at high concentrations, all gap junction inhibitors completely blocked the ongoing walking-like activity induced by NMDA or glutamate. In addition, the gap junction inhibitors also prevented NMDA or glutamate from inducing walking-like activities. These effects are not caused by nonspecific inactivation of motoneurons by the gap junction inhibitors, because in the presence of high concentrations of gap junction inhibitors, tonic motor output was consistently induced by NMDA or glutamate, the dorsal root reflex was reliably evoked, and the walking-like activity was completely reversed after washing out the inhibitors.

The main nonspecific cytotoxic effect of CBX was described for concentrations >100μM [36, 63]. CBX, FFA, and the alcohols may also alter the function of ion channels, reduce neuronal excitability [34–39], and inhibit excitatory synaptic transmission [64–66]. Given these considerations, we started with concentrations that were previously shown to block gap junctions with minimal side-effects [67–72] and limited the duration of exposure. The resumption of the locomotor-like activity upon washing out the inhibitors in all cases and the readily evoked dorsal root reflex suggest that the excitability of the spinal neurons was not significantly affected. Octanol and Heptanol induced fast inhibition of the walking-like pattern, compared to other inhibitors. This may be related to its effect on T-type calcium channels that contribute to generation of locomotor activity in lamprey [73, 74]. Collectively, these results suggest that gap junctions play a vital role in the neural networks that are responsible for the generation of locomotion.

We further showed that the effects of the gap junction inhibitors are complex. They are excitatory at low concentrations and inhibitory at high concentrations. At low concentrations the inhibitors increased the cycle frequency and/or EMG amplitude. At high concentrations or after prolonged exposure, the gap junction inhibitors reduced the cycle frequency and/or EMG amplitude or blocked the walking-like activity altogether. Studies of the respiratory rhythm in adult rodent preparations have shown that the inhibition of gap junctions significantly increases the cycle frequency and reduces the amplitude of the phrenic nerve discharges [46]. In contrast, studies on neonatal or postnatal mice have reported a decrease in the cycle frequency without any change in the amplitude of the nerve discharge [36]. The different effects of the gap junction inhibitors in these studies may reflect functional changes of gap junctions in developmental stages from the neonate to the mature nervous systems and/or methodological differences [46].

Although there were differences in the extent, to which various gap junction inhibitors modulate the walking-like activity, the main effects were consistent across inhibitors and reproducible on the walking-like activity induced either by NMDA or by glutamate. The IC50 of each inhibitor is a measure of its efficacy/potency in blocking the ongoing function of the gap junction proteins. We found that the IC50 of each gap junction inhibitor on NMDA-induced activity differed from that on the glutamate-induced activity. This difference likely reflects the susceptibility of the intercellular connections in the neural networks for walking that are activated by NMDA and glutamate respectively. The glutamate-induced locomotor-like activity was more sensitive to gap junction inhibitors, suggesting that electrical coupling plays a more significant physiological role in generating walking-like activity by neural networks that are activated by the naturally existing neurotransmitter glutamate. It is worth noting that the mechanisms of NMDA and glutamate induced locomotor-like activities may differ. In contrast to NMDA, which is a synthetic compound that induces locomotor-like activity through activating NMDA receptors, glutamate is a naturally existing neurotransmitter that may induce rhythmic activity through NMDA, non-NMDA, and metabotropic glutamate receptors [18, 75, 76].

A significant limitation of this study is that we were unable to differentiate the roles played by each individual connexins. None of the available gap junction inhibitors are specific enough to allow us achieve this objective. Some blockers, such as the hemichannel inhibitors lanthanum chloride and mefloquine, were described to be more selective to specific gap junctions. However, recent studies indicate that most of these inhibitors are not specific, as extracellular glutathione accumulation can be equally blocked by lanthanum chloride as well as by carbenoxolone and flufenamic acid [77]. The effects of these blockers become less specific and they may affect most of the other gap junctions in higher concentrations. For example, at low concentrations, mefloquine (1 μM) blocks Cx36 and Cx50 channels expressed in neuroblastoma cells, but it inhibits Cx26, Cx32, and Cx43 channels at higher concentrations (30 μM) [78]. In both vertebrates and invertebrates, gap junctions typically couple similar types of interneurons and motoneurons with restricted distributions in the central nervous system [7, 25, 26, 36]. More than 20 different connexin genes have been identified with similar, but not identical, pharmacological properties [79–81]. A better understanding of the connexin proteins will help elucidate their specific roles in the generation and regulation of locomotor function.

Taken together, our immunohistochemistry data provided evidence for the expression of specific gap junction proteins in the adult mudpuppy spinal cord. Our pharmacological intervention experiments further demonstrated a vital role of gap junctions in the modulation of the walking-like activity. Multiple gap junction inhibitors were able to modulate the cycle frequency and EMG amplitude in a fairly consistent manner. These data support the notion that intercellular coupling through gap junctions plays a critical role in the control of locomotion in adult animals.

Conflict of Interest/Disclosures

None of the authors, Igor Lavrov, Lyle E. Fox, Jun Shen, Yingchun Han, or Jianguo Cheng, or their immediate families have any actual or potential commercial associations that might create a conflict of interest in connection with this manuscript.

Supporting Information

S1 Fig. Examples of Cx32 (A and B) and Cx36-like (C) immunoreactivity in coronal sections of the mudpuppy spinal cord.

The Cx32 and Cx36 antibodies produced a dense punctuate staining near the nuclei of a few cells in the ventral horn (A and C) and a sparse punctuate pattern in the white matter (B). Nuclei in the images were counter stained with DAPI (blue). The scale bar = 50 μM for Cx36. The scale bar = 40 μM for Cx32.

https://doi.org/10.1371/journal.pone.0152650.s001

(TIF)

Author Contributions

Conceived and designed the experiments: JC. Performed the experiments: IL LF JS YH. Analyzed the data: IL LF JC. Contributed reagents/materials/analysis tools: YH JC. Wrote the paper: IL LF JC.

References

  1. 1. Feller MB. Spontaneous correlated activity in developing neural circuits. Neuron 1999 22:653–656. pmid:10230785
  2. 2. Sutor B, Hagerty T. Involvement of gap junctions in the development of the neocortex. Biochim Biophys Acta 2005 1719:59–68. pmid:16225838
  3. 3. Milner LD, Landmesser LT. Cholinergic and GABAergic inputs drive patterned spontaneous motoneuron activity before target contact. J Neurosci 1999 19:3007–3022. pmid:10191318
  4. 4. Bennett MV. Gap junctions as electrical synapses. J Neurocytol 1997 26:349–366. pmid:9278865
  5. 5. Tamas G, Buhl EH, Lorincz A, Somogyi P. Proximally targeted GABAergic synapses and gap junctions synchronize cortical interneurons. Nat Neurosci 2000 3:366–371. pmid:10725926
  6. 6. Galarreta M, Hestrin S. A network of fast-spiking cells in the neocortex connected by electrical synapses. Nature 1999 402:72–75. pmid:10573418
  7. 7. Gibson JR, Beierlein M, Connors BW. Two networks of electrically coupled inhibitory neurons in neocortex. Nature 1999 402:75–79. pmid:10573419
  8. 8. Dupont E, Hanganu IL, Kilb W, Hirsch S, Luhmann HJ. Rapid developmental switch in the mechanisms driving early cortical columnar networks. Nature 2006 439:79–83. pmid:16327778
  9. 9. Solomon IC, Dean JB. Gap junctions in CO(2)-chemoreception and respiratory control. Respir Physiol Neurobiol 2002 131:155–173. pmid:12126918
  10. 10. Wong WT, Sanes JR, Wong RO. Developmentally regulated spontaneous activity in the embryonic chick retina. J Neurosci 1998 18:8839–8852. pmid:9786990
  11. 11. Chang Q, Gonzalez M, Pinter MJ, Balice-Gordon RJ. Gap junctional coupling and patterns of connexin expression among neonatal rat lumbar spinal motor neurons. The Journal of Neuroscience 1999 19:10813–1082 pmid:10594064
  12. 12. Saint-Amant L, Drapeau P. Synchronization of an embryonic network of identified spinal interneurons solely by electrical coupling. Neuron 2001 31:1035–1046. pmid:11580902
  13. 13. Kiehn O, Tresch MC. Gap junctions and motor behavior. Trends Neurosci 2002 25:108–115. pmid:11814564
  14. 14. Wheatley M, Stein RB. An in vitro preparation of the mudpuppy for simultaneous intracellular and electromyographic recording during locomotion. J Neurosci Methods 1992 42:129–137. pmid:1405730
  15. 15. Wheatley M, Edamura M, Stein RB. A comparison of intact and in-vitro locomotion in an adult amphibian. Exp Brain Res 1992 88:609–614. pmid:1587318
  16. 16. Shik ML. Locomotor patterns elicited by electrical stimulation of the brain stem in the mudpuppy. Motor Control 1997 1:354–368.
  17. 17. Cheng J, Stein RB, Jovanovic K, Yoshida K, Bennett D, Han Y. Identification, localization, and modulation of the neural networks for walking in the mudpuppy (Necturus maculatus) spinal cord. J Neurosci 1998 18:4295–4304. pmid:9592106
  18. 18. Lavrov I, Cheng J. Activation of NMDA receptors is required for the initiation and maintenance of walking-like activity in the mudpuppy (Necturus Maculatus). Can J Physiol Pharmacol 2004 82:637–644. pmid:15523521
  19. 19. Lavrov I, Cheng J. Methodological optimization of applying neuroactive agents for the study of locomotor-like activity in the mudpuppies (Necturus maculatus). J Neurosci Methods 2008 174:97–102. pmid:18692523
  20. 20. Roper S. An electrophysiological study of chemical and electrical synapses on neurones in the parasympathetic cardiac ganglion of the mudpuppy, Necturus maculosus: evidence for intrinsic ganglionic innervation. J Physiol 1976 254:427–454. pmid:175154
  21. 21. Delay RJ, Dionne VE. Coupling between sensory neurons in the olfactory epithelium. Chem Senses 2003 28:807–815. pmid:14654449
  22. 22. Ball AK, McReynolds JS. Localization of gap junctions and tracer coupling in retinal Müller cells. J Comp Neurol 1998 393:48–57. pmid:9520100
  23. 23. Bigiani A, Roper SD. Reduction of electrical coupling between Necturus taste receptor cells, a possible role in acid taste. Neurosci Lett 1994 176:212–216. pmid:7830950
  24. 24. Vanoye CG, Vergara LA, Reuss L. Isolated epithelial cells from amphibian urinary bladder express functional gap junctional hemichannels. Am J Physiol 1999 276:279–284.
  25. 25. Li WC, Roberts A, Soffe SR. Locomotor rhythm maintenance: electrical coupling among premotor excitatory interneurons in the brainstem and spinal cord of young Xenopus tadpoles. J Physiol 2009 587:1677–1693. pmid:19221124
  26. 26. Zhang HY, Li WC, Heitler WJ, Sillar KT. Electrical coupling synchronises spinal motoneuron activity during swimming in hatchling Xenopus tadpoles. J Physiol 2009 587:4455–4466. pmid:19635820
  27. 27. Chen MJ, Kress B, Han X, Moll K, Peng W, Ji RR, et al. Astrocytic CX43 hemichannels and gap junctions play a crucial role in development of chronic neuropathic pain following spinal cord injury. Glia. 2012 60:1660–70. pmid:22951907
  28. 28. Cheng J, Jovanovic K, Aoyagi Y, Bennett DJ, Han Y, Stein RB. Differential distribution of interneurons in the neural networks that control walking in the mudpuppy (Necturus maculatus) spinal cord. Exp Brain Res 2002 145:190–198. pmid:12110959
  29. 29. Jovanović K, Petrov T, Greer JJ, Stein RB. Serotonergic modulation of the mudpuppy (Necturus maculatus) locomotor pattern in vitro. Exp Brain Res. 1996 111:57–67. pmid:8891637
  30. 30. Jovanović K, Petrov T, Stein RB. Effects of inhibitory neurotransmitters on the mudpuppy (Necturus maculatus) locomotor pattern in vitro. Exp Brain Res. 1999 129:172–84. pmid:10591891
  31. 31. Winmill RE, Hedrick MS. Gap junction blockade with carbenoxolone differentially affects fictive breathing in larval and adult bullfrogs. Respir Physiol Neurobiol 2003 138:239–251. pmid:14609513
  32. 32. Takahashi M, Alford S. The requirement of presynaptic metabotropic glutamate ceceptors for the maintenance of locomotion. J Neurosci 2002 22:3692–3699 pmid:11978845
  33. 33. Luo J, Borgens R, Shi R. Polyethylene glycol immediately repairs neuronal membranes and inhibits free radical production after acute spinal cord injury. J Neurochem. 2002 83:471–80. pmid:12423257
  34. 34. Lee RJ, Shaw T, Sandquist M, Partridge LD. Mechanism of action of the non-steroidal anti-inflammatory drug flufenamate on [Ca2+]i and Ca(2+)-activated currents in neurons. Cell Calcium 1996 19:431–438. pmid:8793183
  35. 35. Christ GJ, Spektor M, Brink PR, Barr L. Further evidence for the selective disruption of intercellular communication by heptanol. Am J Physiol 1999 276:H1911–1919. pmid:10362670
  36. 36. Rekling JC, Shao XM, Feldman JL. Electrical coupling and excitatory synaptic transmission between rhythmogenic respiratory neurons in the preBotzinger complex. J Neurosci 2000 20:RC113. pmid:11090613
  37. 37. Rouach N, Segal M, Koulakoff A, Giaume C, Avignone E. Carbenoxolone blockade of neuronal network activity in culture is not mediated by an action on gap junctions. J Physiol 2003 553:729–745. pmid:14514879
  38. 38. Vessey JP, Lalonde MR, Mizan HA, Welch NC, Kelly ME, Barnes S. Carbenoxolone inhibition of voltage-gated Ca channels and synaptic transmission in the retina. J Neurophysiol 2004 92:1252–1256. pmid:15028741
  39. 39. Wang D, Grillner S, Wallén P. Effects of flufenamic acid on fictive locomotion, plateau potentials, calcium channels and NMDA receptors in the lamprey spinal cord. Neuropharmacol 2006 51:1038–1046.
  40. 40. Spray DC. Molecular physiology of gap junction channels. Clin Exp Pharmacol Physiol 1996 23:1038–1040. pmid:8977156
  41. 41. Harris AL. Emerging issues of connexin channels: biophysics fills the gap. Q Rev Biophys 2001 34:325–472. pmid:11838236
  42. 42. Yu J, Bippes CA, Hand GM, Muller DJ, Sosinsky GE. Aminosulfonate modulated pH-induced conformational changes in connexin26 hemichannels. J Biol Chem 2007 282:8895–8904. pmid:17227765
  43. 43. Yamamoto K, Jouhou H, Iwasaki M, Kaneko A, Yamada M. Strongly pH-buffered ringer's solution expands the receptive field size of horizontal cells in the carp retina. Zoolog Sci 2008 25:419–427. pmid:18459824
  44. 44. Nelson PG. Interaction between spinal motoneurons of the cat. J Neurophysiol 1966 29:275–287. pmid:5927462
  45. 45. Gogan P, Gueritaud JP, Horcholle-Bossavit G, Tyc-Dumont S. Electronic coupling between motoneurones in the abducens nucleus of the cat. Exp Brain Res 1974 21:139–154. pmid:4373264
  46. 46. Solomon IC, Chon KH, Rodriguez MN. Blockade of brain stem gap junctions increases phrenic burst frequency and reduces phrenic burst synchronization in adult rat. J Neurophysiol 2003 89:135–149. pmid:12522166
  47. 47. Winmill RE, Hedrick MS. Gap junction blockade with carbenoxolone differentially affects fictive breathing in larval and adult bullfrogs. Respir Physiol Neurobiol 2003 138:239–251. pmid:14609513
  48. 48. Rash JE, Dillman RK, Bilhartz BL, Duffy HS, Whalen LR, Yasumura T. Mixed synapses discovered and mapped throughout mammalian spinal cord. Proc Natl Acad Sci 1996 93:4235–4239. pmid:8633047
  49. 49. White TW, Paul DL. Genetic diseases and gene knockouts reveal diverse connexin functions. Annu Rev Physiol 1999 61:283–310. pmid:10099690
  50. 50. Shapovalov AI, Shiriaev BI. Recurrent interactions between individual motoneurones and dorsal root fibres in the frog. Exp Brain Res 1980 38:115–116. pmid:7351224
  51. 51. Christensen BN. Distribution of electrotonic synapses on identified lamprey neurons: a comparison of a model prediction with an electron microscopic analysis. J Neurophysiol 1983 49:705–716. pmid:6834095
  52. 52. Fetcho JR. Spinal network of the Mauthner cell. Brain Behav Evol 1991 37:298–316. pmid:1933252
  53. 53. Bou-Flores C, Berger AJ. Gap junctions and inhibitory synapses modulate inspiratory motoneuron synchronization. J Neurophysiol 2001 85:1543–1551. pmid:11287478
  54. 54. Winmill RE, Hedrick MS. Gap junction blockade with carbenoxolone differentially affects fictive breathing in larval and adult bullfrogs. Respir Physiol Neurobiol 2003 138:239–251. pmid:14609513
  55. 55. Yamamoto T, Ochalski A, Hertzberg EL, Nagy JI. LM and EM immunolocalization of the gap junctional protein connexin 43 in rat brain. Brain Res 1990 508:313–319. pmid:2155040
  56. 56. Mugnaini E. Cell junctions of astrocytes, ependyma, and related cells in the mammalian central nervous system, with emphasis on the hypothesis of a generalized functional syncytium of supporting cells. In: Fedoroff S, Vernadakis A, editors. Astrocytes, vol 1. New York: Academic Press 1986 p 329–371.
    • 57. Holthoff K, Witte OW. Directed spatial potassium redistribution in rat neocortex. Glia 2000 29:288–292. pmid:10642755
    • 58. Cornell-Bell AH, Finkbeiner SM, Cooper MS, Smith SJ. Glutamate induces calcium waves in cultured astrocytes: long-range glial signaling. Science 1990 247:470–473. pmid:1967852
    • 59. Charles AC, Naus CC, Zhu D, Kidder GM, Dirksen ER, Sanderson MJ. Intercellular calcium signaling via gap junctions in glioma cells. J Cell Biol 1992 118:195–201. pmid:1320034
    • 60. Venance L, Stella N, Glowinski J, Giaume C. Mechanism involved in initiation and propagation of receptor-induced intercellular calcium signaling in cultured rat astrocytes. J Neurosci 1997 17(6):1981–92. pmid:9045727
    • 61. Lee IH, Lindqvist E, Kiehn O, Widenfalk J, Olson L. Glial and neuronal connexin expression patterns in the rat spinal cord during development and following injury. 2005 J Comp Neurol 489:1–10 pmid:15977163
    • 62. Reed KE, Westphale EM, Larson DM, Wang HZ, Veenstra RD, Beyer EC. Molecular cloning and functional expression of human connexin37, an endothelial cell gap junction protein, J Clin Invest 1993 91:997–1004. pmid:7680674
    • 63. Juszczak GR, Swiergiel AH. Properties of gap junction blockers and their behavioural, cognitive, and electrophysiological effects: Animal and human studies. Prog Neuropsychopharmacol Biol Psychiatry 2009 33:181–198. pmid:19162118
    • 64. Puil E, el-Beheiry H, Baimbridge KG. Anesthetic effects on glutamate-stimulated increase in intraneuronal calcium. J Pharmacol Exp Ther 1990 255:955–961. pmid:2262913
    • 65. Pocock G, Richards CD. Excitatory and inhibitory synaptic mechanisms in anaesthesia. Br J Anaesth 1993 71:134–147. pmid:7688240
    • 66. Rorig B, Sutor B. Regulation of gap junction coupling in the developing neocortex. Mol Neurobiol 1996 12:225–249. pmid:8884750
    • 67. Kimura H, Oyamada Y, Ohshika H, Mori M, Oyamada M. Reversible inhibition of gap junctional intercellular communication, synchronous contraction, and synchronism of intracellular Ca2+ fluctuation in cultured neonatal rat cardiac myocytes by heptanol. Exp Cell Res 1995 220:348–356. pmid:7556443
    • 68. Tresch MC, Kiehn O. Motor coordination without action potentials in the mammalian spinal cord. Nat Neurosci 2000 3:593–599. pmid:10816316
    • 69. Margineanu DG, Klitgaard H. Can gap-junction blockade preferentially inhibit neuronal hypersynchrony vs. excitability? Neuropharmacol 2001 41:377–383.
    • 70. Ghamari-Langroudi M, Bourque CW. Flufenamic acid blocks depolarizing afterpotentials and phasic firing in rat supraoptic neurones. J Physiol 2002 545:537–542. pmid:12456832
    • 71. Jahromi SS, Wentlandt K, Piran S, Carlen PL. Anticonvulsant actions of gap junctional blockers in an in vitro seizure model. J Neurophysiol 2002 88:1893–1902. pmid:12364515
    • 72. Srinivas M, Spray DC. Closure of gap junction channels by arylaminobenzoates. Mol Pharmacol 2003 63:1389–1397. pmid:12761350
    • 73. Todorovic SM, Perez-Reyes E, Lingle CJ. Anticonvulsants but not general anesthetics have differential blocking effects on different T-type current variants. Mol Pharmacol. 2000 58:98–108 pmid:10860931
    • 74. Tegnér J, Hellgren-Kotaleski J, Lansner A, Grillner S. Low-voltage-activated calcium channels in the lamprey locomotor network: simulation and experiment. J Neurophysiol 1997 77:1795–812. pmid:9114237
    • 75. Alford S, Grillner S. CNQX and DNQX block non-NMDA synaptic transmission but not NMDA-evoked locomotion in lamprey spinal cord. Brain Res 1990 506:297–302 pmid:1967966
    • 76. Takahashi M, Alford S. The requirement of presynaptic metabotropic glutamate ceceptors for the maintenance of locomotion. J Neurosci 2002 22:3692–3699 pmid:11978845
    • 77. Rana S, Dringen R. Gap junction hemichannel-mediated release of glutathione from cultured rat astrocytes. Neurosci Lett. 2007 415:45–8. pmid:17222973
    • 78. Cruikshank SJ, Hopperstad M, Younger M, Connors BW, Spray DC, Srinivas M. Potent block of Cx36 and Cx50 gap junction channels by mefloquine. Proc Natl Acad Sci. 2004 101:12364–9. pmid:15297615
    • 79. Saez JC, Berthoud VM, Branes MC, Martinez AD, Beyer EC. Plasma membrane channels formed by connexins: their regulation and functions. Physiol Rev 2003 83:1359–1400. pmid:14506308
    • 80. McCracken CB, Roberts DC. Neuronal gap junctions: expression, function, and implications for behavior. Int Rev Neurobiol 2006 73:125–151. pmid:16737903
    • 81. Giaume C, Theis M. Pharmacological and genetic approaches to study connexin-mediated channels in glial cells of the central nervous system. Brain Res Rev 2010 632:160–176.