- Email: [email protected]
Dye coupling among satellite glial cells in mammalian dorsal root ganglia
Brain Research 1036 (2005) 42 – 49 www.elsevier.com/locate/brainres
Research report
Dye coupling among satellite glial cells in mammalian dorsal root ganglia Tian-Ying Huanga, Pavel S. Cherkasa, David W. Rosenthalb, Menachem Hanania,* a
Laboratory of Experimental Surgery, Hebrew University-Hadassah Medical School, Mount Scopus, Jerusalem 91240, Israel b Department of Medicine, Mayo Clinic and Foundation, Rochester, MN 55905, USA Accepted 4 December 2004
Abstract Dorsal root ganglia (DRG) are key elements in sensory signaling under physiological and pathological conditions. Little is known about electrical coupling among cells in these ganglia. In this study, we injected the fluorescent dye Lucifer yellow (LY) into single cells to examine dye coupling in DRG. We found no dye coupling between neurons or between neurons and their attendant satellite glial cells (SGCs). In mouse DRG, we observed that in 26.2% of the cases SGCs that surround a given neuron were dye coupled. In only 3.2% of the cases SGCs that make envelopes around different neurons were coupled. The data from mouse ganglia were very similar to those from rat and guinea pig DRG. The results obtained by injection of the tracer biocytin were very similar to those observed with LY. The coupling incidence within the envelopes increased 3.1-fold by high extracellular pH (8.0), but coupling between envelopes was not affected. Acidic pH (6.8) reduced the coupling. High extracellular K+ (9.4 mM) increased the coupling 2.4-fold and 4.7-fold within and between envelopes, respectively. Low extracellular Ca2+ (0.5, 1.0 mM) partly reversed the effect of high K+ on coupling. The results showed that SGCs in mammalian sensory ganglia are connected by gap junctions. This coupling is very sensitive to changes in pH, and can therefore be modulated under various physiological and pathological conditions. The dependence of the coupling on extracellular K+ and Ca2+ suggests that the permeability of gap junctions can be altered by physiological and pharmacological stimuli. D 2005 Elsevier B.V. All rights reserved. Theme: Cellular and molecular biology Topic: Neuroglia and myelin Keywords: Dorsal root ganglia; Neurons; Satellite glial cells; Gap junctions; Dye coupling; Lucifer yellow; Biocytin
1. Introduction Satellite glial cells (SGCs) are the most abundant cell type in dorsal root ganglia (DRG) but their physiological properties are little understood. In contrast, much is known about glial cells in the central nervous system (CNS), and it is well established that central glia provide structural, metabolic and trophic support to neurons [12,33]. These cells are active partners of neurons in signal processing and synaptic integration [19,26,35]. Intercellular coupling by gap junctions among glial cells is widespread throughout the CNS [11,24], and is believed
* Corresponding author. Fax: +972 2 5823515. E-mail address: [email protected] (M. Hanani). 0006-8993/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2004.12.021
to be important for the functions of these cells [2,18,25]. Glial cells in the peripheral nervous system are also mutually coupled by gap junctions [15,20,22]. Altered coupling among glial cells in the CNS is known to be associated with various pathological states [37,38]. There is ultrastructural evidence for gap junctions in SGCs in sensory ganglia [27], but little is known about their function. We have shown that SGCs are mutually dye coupled and that sectioning of the axons of sensory neurons resulted in an increase in this coupling and in the number of gap junctions [4,16,30]. We hypothesized that increased coupling contributes to neuropathic pain states produced by peripheral nerve injury, and may account for the spread of pain sensation. It is quite likely that gap junctions among SGCs are essential also for the normal function of sensory ganglia, but so far these cells have received very little
T.-Y. Huang et al. / Brain Research 1036 (2005) 42–49
attention. In the present study, we used intracellular dye injection to compare coupling of SGCs in three mammalian species. Dye coupling was measured using two different tracers, and the effects of extracellular pH, K+ and Ca2+ on dye coupling were examined.
2. Materials and methods 2.1. Animals and preparations Experiments were done on Balb/c mice of either sex, male Sprague–Dawley rats, and male guinea pigs, all 3–5 months old. The experimental protocol was approved by the Institutional Animal Care and Use Committee. The animals were sacrificed by CO2. DRG L4,5 were removed and placed in cold (4 8C) Krebs solution (pH 7.4) containing (mM): 118 NaCl, 4.7 KCl, 14.4 NaHCO3, 1.2 MgSO4, 2.5 CaCl2, 1.2 NaH2PO4 and 11.5 glucose. The ganglia were pinned onto the Sylgard bottom of a chamber superfused with Krebs solution bubbled with 95% O2 and 5% CO2 at 23–24 8C. The pH in the bathing solution was altered by substituting HCO3 with Cl or Cl with HCO3 . In the experiments with altered HCO3 , the tissues were first incubated for 30 min in the low- or high-pH solutions, and the injections were made in the same solutions during the next 4 h. High K+ was achieved by substituting Na+ with K+, and low Ca2+ by substituting Ca2+ with Na+. 2.2. Intracellular labeling Experiments were performed using an upright microscope (Axioskop FS, Zeiss, Jena, Germany) equipped with fluorescent illumination and a digital camera (Pixera 120e, Pixera, Los Gatos, USA) connected to a PC. Neurons and SGCs were singly injected with the fluorescent dye Lucifer yellow (LY, Sigma, St. Louis, MO, USA) 3% in 0.5 M LiCl solution, or 4% biocytin (Sigma) in 1 M KCl from a glass microelectrode connected to a preamplifier (Neuro Data Instrument model IR 283, New York, NY, USA). The tip resistance of the microelectrodes was 80–120 M V. Both LY and biocytin were injected intracellularly by current pulses 100 ms in duration and 0.5 nA in amplitude at 10 Hz. LY was driven by hyperpolarizing pulses for 3–5 min, and biocytin by depolarizing pulses for 15 min. LY injection times were determined by observing the staining of the cell in real time. We found that injections of LY for more than 5 min did not alter the extent of dye coupling. During and after LY injections, the living labeled cells were imaged with the digital camera. After the experiments, the DRG were fixed overnight at 4 8C in 4% paraformaldehyde in phosphate buffer solution (PBS, pH 7.4) and washed with PBS. The tissues injected with LY were mounted in Gel/mount (Biomeda, Foster, CA, USA) and the LY-labeled cells were imaged with a Bio-Rad
43
confocal microscope (Bio-Rad, Hercules, CA, USA). Before processing, the tissues with biocytin-injected cells were incubated for 15 min in PBS containing 1% H2O2. After washing with PBS, the tissues were incubated in PBS containing 0.5% Triton X-100 and 1 Ag/ml avidin– HRP (Sigma) for 2 h. After five washes with PBS, the tissues were reacted with diaminobenzidine (Sigma, 2 mg/ ml) and 0.01% H2O2 in PBS for 40 min. They were then dehydrated, cleared with xylene and mounted in Entellan (Merck, Darmstadt, Germany). 2.3. Statistical analysis All average values were expressed as mean F SEM. Fisher’s Exact Test and Mann–Whitney test were used for comparison. P b 0.05 is considered as statistically significant.
3. Results 3.1. Dye coupling experiments Intracellular dye injection showed that none of the DRG neurons (mice, n = 95; rats, n = 93; guinea pigs, n = 82) that were examined in the three species used was coupled to any other cell (neurons or SGCs), as shown previously [16,46]. Furthermore, when the dye was injected into SGCs, there was no dye passage from them into neurons (n = 126, 4 mice; n = 127, 4 rats; n = 142, 5 guinea pigs). We noticed two types of SGCs coupling: between SGCs that make the envelope around a given neuron (Figs. 1A, B), and between SGCs surrounding different neurons (Fig. 1C). The incidence of the dye-injected SGCs that were coupled to other SGCs within the same envelope was 26.2%, 28.3% and 29.6% in mouse, rat and guinea pig, respectively (Fig. 2A). Coupling between SGCs belonging to different envelopes was rare: 3.2%, 3.1% and 3.5% in mouse, rat and guinea pig, respectively (Fig. 2A). Although the incidence of dye coupling among SGCs was very similar in the three species, the number (mean F SEM) of SGCs coupled to the dye-injected cells in the guinea pig was greater than in the mouse and rat, 5.69 F 0.74 (n = 42, 20 ganglia of 5 guinea pigs) vs. 2.15 F 0.24 (n = 33, 16 ganglia of 4 mice) and 2.08 F 0.24 (n = 36, 16 ganglia of 4 rats), (P b 0.001, Mann–Whitney test), as shown in Fig. 2B. Furthermore, the maximum number of SGCs around a given neuron was greater in guinea pig than in mouse or rat ganglia. In guinea pigs, the maximum was 21 SGCs within an envelope (Fig. 1B), whereas in mice and rats it was eight. Vaney [43] showed that due to the small size of biocytin, this tracer can reveal cell coupling not detected with LY. To examine whether biocytin injection could disclose a different pattern of coupling, we injected it intracellularly into neurons and SGCs in mouse DRG L4,5. We found that
44
T.-Y. Huang et al. / Brain Research 1036 (2005) 42–49
experiments. The experiments with LY and biocytin showed that glial cells surrounding small and large neurons displayed similar coupling patterns. 3.2. The influence of pH on dye coupling in mouse DRG
Fig. 1. Micrographs showing dye coupling between and within glial envelopes in DRG. (A) The dye-injected satellite glial cell (SGC) is coupled only to other SGCs in the same envelope; mouse DRG. (B) An example of dye coupling within an envelope in guinea pig DRG. (C) The dye-injected SGC is coupled to other SGCs forming envelopes around different neurons in mouse DRG. (D) An example of intracellular injection of biocytin showing coupling within a glial envelope in mouse DRG. Arrows indicate four labeled SGCs. Panels (A–C) are confocal images. Asterisks indicate the dye-injected cells. In panel (D), the biocytin-injected cell is not indicated because this method does not allow the observation of the injected cells during the experiment. Scale bars = 20 Am.
neurons were not coupled to each other or to SGCs (n = 55, 22 ganglia of 6 mice). With biocytin as a tracer, coupling incidence among SGCs within envelopes was 31.3% (n = 48, P N 0.5, Fisher’s Exact Test, compared with the LY results), see Fig. 1D, and coupling between glial envelopes was 4.2% (P N 0.6). The number of SGCs coupled to the biocytin-injected cells was 2.20 F 0.38 (n = 15, P N 0.9, Mann–Whitney test) vs. 2.15 F 0.24 for LY. Thus, the results obtained with these two tracers in mouse DRG were very similar. Consequently, we used LY in all the following
As the incidence of SGC dye coupling was very similar in the ganglia from the three species used, we focused on mouse DRG for the remainder of this study. Previous studies showed that the permeability of gap junctions is strongly influenced by changes in intracellular pH [39]. To find out whether dye coupling among SGCs depends on pH, we altered the pH in the bathing solution and examined the dye coupling among SGCs. We found that dye coupling among SGCs depended strongly on extracellular pH. Slightly basic pH (7.8 or 8.0) increased the incidence of dye coupling among SGCs within envelopes, and also the number of SGCs coupled to the dye-injected cells. The incidence of dye coupling among SGCs within envelopes increased 2.4-fold to 63.9% (n = 97, 16 ganglia, 4 mice, P b 0.0001, Fisher’s Exact Test) at pH 7.8, and 3.1fold to 81.3% (n = 123, 20 ganglia, 5 mice, P b 0.0001) at pH 8.0. At pH 6.8, the incidence of coupling within envelopes decreased from 26.2% to 12.5% (n = 104, 16 ganglia, 4 mice, P b 0.05). No significant change in dye coupling between envelopes was observed after changing to acidic or basic pH. The average number of SGCs coupled to the dye-injected cells increased 1.5-fold to 3.32 F 0.22 (n = 62, P b 0.05, Mann–Whitney test) at pH 7.8 and 1.6-fold to 3.34 F 0.20 (n = 100, P b 0.05) at pH 8.0, and decreased by half to 1.69 F 0.38 (n = 13, P b 0.05) at pH 6.8 in comparison with 2.15 F 0.24 (n = 33) at pH 7.4, as shown in Fig. 3. To determine whether the effect of pH on dye coupling is reversible, we incubated the ganglia in Krebs solution at pH 8.0 for 2 h, washed for 1 h in normal Krebs solution (pH 7.4) and then examined dye coupling among SGCs. The coupling incidence was 28.4% (n = 95, 12 ganglia, 3 mice, P N 0.7, Fisher’s Exact Test) within, and 3.2% (P N 0.9) between glial envelopes, vs. 26.2% and
Fig. 2. The incidence of dye coupling among SGCs in DRG of three different mammals. (A) The incidence of dye coupling between (black bars) and within (empty bars) glial envelopes in mouse, rat and guinea pig ganglia. (B) Histograms show the number (mean F SEM) of SGCs coupled to the dye-injected glial cells in mouse, rat and guinea pig DRG. * indicates P b 0.001, as compared with mouse and rat ganglia. Only cases were dye coupling was observed were included in this analysis. Mouse-b denotes the data for the mouse DRG after using biocytin injection. Mann–Whitney test was used for comparison.
T.-Y. Huang et al. / Brain Research 1036 (2005) 42–49
45
Fig. 3. Dye coupling among SGCs in the mouse DRG depends on pH. (A) The incidence of dye coupling within envelopes (empty bars) increased at basic pH, whereas coupling between envelopes (black bars) showed weak dependence on pH. (B) Effect of pH on the number (mean F SEM) of SGCs coupled to the dye-injected glial cells. The data for the groups (pH 8.0 Y 7.4 and pH 6.8 Y 7.4) were obtained from the experiments where we tested the reversibility of the high pH effect. In these experiments, we first incubated the DRG at pH 8.0 or 6.8 for 2 h, washed the tissues at pH 7.4 for 1 h and then examined the dye coupling among SGCs at pH 7.4. *P b 0.05, **P b 0.001, as compared with controls (pH 7.4). Fisher’s Exact Test (A) and Mann–Whitney test (B) were used for comparison.
3.2% in control DRG. The number of SGCs coupled to the LY-injected cells was 2.26 F 0.30 (n = 27, P N 0.8) vs. 2.15 F 0.24 in control DRG. Similarly, after incubation of the DRG at pH 6.8 for 2 h and then washing for 1 h, the coupling incidence recovered from 12.5% to 25% (n = 68, 8 ganglia, 2 mice, P b 0.05) within the envelopes, and the incidence of coupling between glial envelopes recovered to 2.9%. The number of SGCs coupled to the LY-injected cells also recovered from 1.69 F 0.38 to 2.12 F 0.39 (n = 17, P b 0.05), see Fig. 3. These results showed that the effect of extracellular pH on dye coupling among SGCs was reversible. 3.3. The effects of K+ and Ca2+ on dye coupling in mouse DRG Extracellular K+ and Ca2+ have been found to influence coupling among astrocytes [5,8]. We therefore decided to examine the effects of these ions on SGC coupling. At short incubation times (10–55 min) after increasing the concentration of extracellular K+ ([K+]o) from 4.7 to 9.4 mM, coupling incidence among SGCs within an envelope increased from 26.2 to 63.2% (P b 0.0001, n = 68, 16 ganglia of 4 mice); however, the incidence of coupling between envelopes did not change (2.9 vs. 3.2% in control, P N 0.1). In contrast, after longer times (70–240 min) in high [K+]o coupling between envelopes increased 4.7-fold to 14.9% (P b 0.01, n = 94, 16 ganglia, 4 mice). The incidence of coupling within SGC envelopes after long incubation times was similar to that observed at short times (67%, P b 0.0001 compared with control, Fisher’s Exact Test). After long incubation times, the number of SGCs coupled to the dye-injected cells increased from 2.15 F 0.24 (n = 33) to 4.68 F 0.54 (n = 65, P b 0.05, Mann– Whitney test). To find out whether the influence of high K+ on dye coupling among SGCs is reversible, we examined coupling after incubation of the DRG in high K+ Krebs
solution for 3 h and then washing for 3 h. The coupling incidence decreased from 67% (high K+) to 31.3% ( P b 0.01, n = 96, 12 ganglia, 3 mice) within an SGC envelope, and from 14.9% (high K+) to 6.3% (P b 0.05) between SGC envelopes and the number of SGCs coupled to the LY-injected cells decreased from 4.68 F 0.54 (high K+) to 2.67 F 0.35 (P b 0.05, n = 30). There was no significant difference between these data and the data obtained from control ganglia (Fig. 4). The results showed that high K+induced change in dye coupling among SGCs was almost completely reversible. To determine whether the incubation time itself affected dye coupling, we compared dye coupling at short (10–55 min) and long (1–4 h) incubation times in normal solution. The results showed that dye coupling between envelopes and within an envelope were 2.5%, 25% (n = 40) during 10–55 min incubation, and 3.5%, 26.7% (n = 86) in 1–4 h incubation (both P N 0.8, Fisher’s Exact Test). Thus, incubation time did not increase dye coupling. The effect of high [K+]o may be due to depolarization of SGCs themselves and/or neurons, which may lead to Ca2+ influx into these cells (SGCs and neurons).To test this possibility, we repeated the high [K+]o experiments at 70–240 min using a medium with low [Ca2+]o. We found that decreasing [Ca2+]o from 2.5 to 1.0 or 0.5 mM partly reversed the high [K+]o effect. In the presence of high [K+]o and low [Ca2+]o, the incidence of dye coupling between envelopes decreased from 14.9% to 3.7% (1.0 mM [Ca2+]o, n = 108, 16 ganglia of 4 mice, P b 0.01, Fisher’s Exact Test) and to 3.2% (0.5 mM [Ca2+]o, n = 93, 16 ganglia of 4 mice, P b 0.01). The effect of low [Ca2+]o on the incidence of dye coupling within envelopes was less striking, this coupling decreased slightly but significantly from 67% to 52.8% (1.0 mM [Ca2+]o, P b 0.05) and to 51.6% (0.5 mM [Ca2+]o, P b 0.05). The mean number of SGCs coupled to the dye-injected cells slightly decreased from 4.68 F 0.54 (n = 65) to 3.28 F 0.33
46
T.-Y. Huang et al. / Brain Research 1036 (2005) 42–49
Fig. 4. Dye coupling among SGCs in the mouse DRG increased following 70–240 min after increasing [K+]o, and also depended on [Ca2+]o. (A) The effect of K+ and Ca2+ on the incidence of dye coupling between glial envelopes. The full effect of high K+, requires a normal external Ca2+. (B) The effect of K+ and Ca2+ on the incidence of dye coupling within envelopes. This type of coupling depended weakly on Ca2+. (C) The effect of K+ and Ca2+ on the number (mean F SEM) of SGCs coupled to the dye-injected glial cells. The data for the group (K+ 9.4 Y 4.7 mM, Ca2+ 2.5 mM) were obtained from the experiments where we tested the reversibility of the high K+ effect. In these experiments, the DRG were first incubated in high extracellular K+ (9.4 mM, Ca2+ 2.5 mM) for 3 h and were then washed in normal Krebs solution (K+ 4.7 mM, Ca2+ 2.5 mM) for 3 h. Following this, dye coupling among SGCs was examined in normal Krebs solution. *P b 0.05, **P b 0.001, as compared with control (K+ 4.7 mM and Ca2+ 2.5 mM). Fisher’s Exact Test (A, B) and Mann–Whitney test (C) were used for comparison.
(n = 54, 1.0 mM [Ca2+]o, P b 0.05, Mann–Whitney test) and to 3.32 F 0.30 (n = 62, 0.5 mM [Ca2+]o, P b 0.05), as shown in Fig. 4.
4. Discussion Intracellular labeling of SGCs with LY and biocytin revealed two types of dye coupling among these cells: coupling within the glial envelope of individual neurons, and between glial cells belonging to envelopes around different neurons. Dye coupling of SGCs within an envelope was much more prevalent and its incidence changed with extracellular pH and K+. In contrast, dye coupling of SGCs in different envelopes was rare and its incidence was not affected by pH, but increased by high [K+]o. Interestingly, the effect of high [K+]o on dye coupling between envelopes was completely abolished by decreasing [Ca2+]o. Our observation of the inhibitory effects of acidic pH (which blocks gap junctions) on dye coupling and previous ultrastructural studies support the assumption that the dye coupling is mediated by gap junctions [16,30]. The results on SGC coupling incidence for mice, rats and guinea pigs were very similar, but the number of SGCs coupled to the dye-injected cells in guinea pig DRG was greater than those in the mouse and rat. This difference may be due to the larger number of SGCs within an envelope around a given neuron in guinea pigs. This is in accord with the results of Ledda et al. [21], who showed that in large animals DRG neurons are larger than in small animals, and the number of SGCs around a given neuron increases with the neuronal volume. Whether coupling among SGCs contributes to the function of DRG is still unknown. Coupling among glial cells is known to be prominent in the CNS [11,24], in the developing nervous system [3,10], and in cultured brain tissue [8,9]. Glial coupling is essential for spatial K+
buffering in the CNS [2,18,25,34]. Similarly, coupling among SGCs is likely to contribute to normal functions of DRG. Altered coupling among glial cells is associated with various pathological states in the brain [37,38]. Axotomy of the sciatic nerve was found to increase dye coupling among SGCs in mouse DRG [16,30], and it was proposed that augmented glial coupling can contribute to neuropathic pain. The idea of intercellular coupling in DRG has been alluded to in several previous studies. There is evidence for bcross talkQ between DRG neurons [6], but the proposed mechanism is not electrical coupling, but chemical transmission [1]. Miletic and Lu [23] presented evidence for signal transmission between DRG neurons, and proposed that it is mediated either by chemical synapses or by electrotonic coupling. It can be suggested that glial coupling described here might contribute to such inter-neuronal coupling. This will require that neurons and glial cells interact via chemical messengers. However, although previous studies suggested that SGCs and sensory neurons in DRG can interact [14,17], the precise nature of this interaction is still not known. The topic of neuron-glia communication in DRG clearly warrants further investigation. It should be mentioned that the lack of dye coupling between cells does not necessarily indicate the lack of electrical coupling [32]. This certainly holds when using LY, which in some systems appears not to cross gap junctions [43]. To test whether a smaller tracer molecule will reveal a greater extent of coupling, we injected biocytin into SGCs and neurons, but the results were very similar to those obtained with LY. This indicates that the gap junctions in SGCs are equally permeable to LY and biocytin, suggesting that their pore is relatively large. Under control conditions (pH 7.4), only 26.2% of the cells was dye coupled within the envelopes. The reason for this low incidence is not clear. At slightly basic pH (8.0), this coupling incidence increased to 81.3%, suggesting that the majority of the SGCs are connected by gap junctions, which are mostly closed at pH 7.4. One possible explanation
T.-Y. Huang et al. / Brain Research 1036 (2005) 42–49
for this observation is that because of the experimental conditions the intracellular pH is acidic. Previous studies have shown that changing the extracellular pH altered the intracellular pH in the same direction but with smaller amplitude, when using bicarbonate as a buffer [36], and that maximal gap junctional conductance occurred at intracellular pH around 7.6–7.8 [40]. The present results for the effect of extracellular pH on dye coupling among SGCs are consistent with these observations. Furthermore, low pH might be the result of damage, but in the present case we believe that this is unlikely because we worked under standard conditions used for isolated tissues (see Materials and methods). In previous studies, in our laboratory the same experimental conditions were used to study dye coupling in enteric glial cells, and a high (about 90%) incidence of coupling was found [15,22]. Therefore, we propose that under basal conditions many of the gap junctions in SGCs are closed, and are opened by a slightly basic pH or by other, yet unknown, physiological parameters. The incidence of dye coupling between SGCs in different envelopes was very low (3.2%) under control conditions. Pannese et al. [28] found that DRG neurons are sometimes arranged in clusters surrounded by a common connective tissue space. This arrangement is more common in immature animals than adult animals [29]. As we used young adult animals in the present work, a probable explanation is that dye coupling between envelopes is due to the presence of these clusters. This explanation is based on the assumption that SGCs surrounding a cluster are dye coupled. Therefore, when a tracer molecule was injected into individual SGCs within the sheath surrounding a cluster, SGCs around two or more neurons could be stained. This type of coupling was not affected by pH, but was previously found to be blocked by octanol [16]. The reason for the lack of influence of pH on the inter-envelope coupling is not yet clear. We observed that high [K+]o (9.4 mM) increased the incidence of dye coupling both within and between envelopes. The effect of high [K+]o depended on [Ca2+]o. Decreasing [Ca2+]o from 2.5 to 1 or 0.5 mM slightly decreased the high [K+]o-induced rise in coupling within envelopes but completely reversed the increase in dye coupling between the envelopes. The duration of the incubation in high [K+]o had a striking influence on the effects observed. A relatively brief (10–55 min) exposure of the tissue to high [K+]o caused an increase in the coupling among SGCs within the envelope, but did not affect the inter-envelope coupling. In contrast, longer exposures (70–240 min) in high [K+]o caused a marked (4.7-fold) increase in the inter-envelope coupling with little influence on the coupling within the envelope. Furthermore, inter-envelope coupling was much more sensitive to the lowering of [Ca2+]o. We can offer two possible explanations for these results: (1) It can be proposed that depolarization of SGCs by high [K+]o induces Ca2+ influx
47
into these cells through voltage dependent Ca2+ channels. The resulting elevation of intracellular Ca2+ concentration may increase the number of open gap junctions. Such a model has been proposed to explain the increase in coupling among astrocytes under high [K+]o [5]. (2) High [K+]o causes the depolarization of neurons, which increases the Ca2+ conductance of the neuronal plasma membranes, followed by Ca2+ influx and an elevation of intracellular Ca2+ concentration. This in turn may lead to the release of chemical messenger(s) from the neurons, which then alter intercellular coupling. The long delay for the effect on the inter-envelope coupling suggests that long term metabolic processes are involved. The chemical messenger(s) for this high [K+]o effect is not yet known but one candidate is nitric oxide (NO). High [K+]o is known to depolarize neurons and increase its NO production via elevation of cytoplasmic Ca2+ [45]. NO activates soluble guanylate cyclase leading to an increase in cyclic GMP levels, which then promote neurite growth in DRG neurons [42]. In DRG, NO was found to elevate cGMP levels in SGCs via the activation of guanylate cyclase [13,41]. Consequently, it is conceivable that high [K+]o-induced depolarization of NOS containing neurons in DRG leads to NO synthesis, which in turn acts on the neighboring SGCs. The production of cGMP in SGCs may lead to the formation of new gap junctions among these cells, or to the opening of existing gap junctions. Of course, a combination of direct depolarization of SGCs by high [K+]o and mediation by neurons cannot be excluded. The validation of either scheme will require further work. The short term influence of high K+ on dye coupling requires a different explanation from the one proposed above. Enkvist and McCarthy [8] reported that high [K+]o induced an increase in dye coupling among cultured astrocytes and suggested that it might be related to a direct effect of depolarization on gap junction conductance. Depolarization was also found to induce alkalinization in astrocytes [31]. Thus, the increased dye coupling within glial envelopes may be due to the opening of existing gap junctions between SGCs around the same neuron via depolarization and/or alkalinization of SGCs. However, a more likely explanation is that depolarization induced Ca2+ influx into the cells, which initiated events leading to the opening of gap junction channels as proposed by De PinaBenabou et al. [5]. The functional significance of SGC coupling remains to be explored. There is much evidence from studies in the CNS that glial coupling is important for K+ buffering and possibly for other functions [34,37], and it can be proposed that SGCs perform similar roles. We found that SGC coupling is very sensitive to changes in pH, and as pH changes can occur under various circumstances such as inflammation and high metabolic activity, it is likely that SGC coupling is altered under certain physiological and pathological conditions. The findings on the long term effect of high [K+]o indicate that gap junctions among SGCs are
48
T.-Y. Huang et al. / Brain Research 1036 (2005) 42–49
highly plastic, as found previously following axotomy [16]. There is evidence that SGCs in sensory ganglia have functional receptors for bradykinin [7,14] and ATP [14,44], and it is conceivable that the activation of these receptors can also alter SGC coupling. In summary, we showed by intracellular dye injection that SGCs are mutually connected via gap junctions. The dye coupling was quite similar in mice, guinea pigs and rats, suggesting that the findings are typical for mammals in general. We showed that certain experimental conditions can alter SGC coupling and it will be most interesting to explore how coupling changes under a variety of situations, such as aging and disease.
Acknowledgments This work was supported by the Israel Science Foundation and by the Hebrew University Center for Pain Research. We thank Dr. Naomi Melamed-Book for helping with the confocal imaging.
References [1] R. Amir, M. Devor, Chemically mediated cross-excitation in rat dorsal root ganglia, J. Neurosci. 16 (1996) 4733 – 4741. [2] F. Amzica, M. Massimini, A. Manfridi, Spatial buffering during slow and paroxysmal sleep oscillations in cortical networks of glial cells in vivo, J. Neurosci. 22 (2002) 1042 – 1053. [3] K. Bittman, D.L. Becker, F. Cicirata, J.G. Parnavelas, Connexin expression in homotypic and heterotypic cell coupling in the developing cerebral cortex, J. Comp. Neurol. 443 (2002) 201 – 212. [4] P.S. Cherkas, T.Y. Huang, T. Pannicke, M. Tal, A. Reichenbach, M. Hanani, The effects of axotomy on neurons and satellite glial cells in mouse trigeminal ganglion, Pain 110 (2004) 290 – 298. [5] M.H. De Pina-Benabou, M. Srinivas, D.C. Spray, E. Scemes, Calmodulin kinase pathway mediates the K+-induced increase in Gap junctional communication between mouse spinal cord astrocytes, J. Neurosci. 21 (2001) 6635 – 6643. [6] M. Devor, P.D. Wall, Cross-excitation in dorsal root ganglia of nerve-injured and intact rats, J. Neurophysiol. 64 (1990) 1733 – 1746. [7] S. England, F. Heblich, I.F. James, J. Robbins, R.J. Docherty, Bradykinin evokes a calcium-activated chloride current in nonneuronal satellite cells isolated from neonatal dorsal root ganglion, J. Physiol. (London) 530 (2001) 395 – 403. [8] M.O.K. Enkvist, K.D. McCarthy, Astroglial gap junction communication is increased by treatment with either glutamate or high K+ concentration, J. Neurochem. 62 (1994) 489 – 495. [9] M.M. Froes, A.H. Correia, J. Garcia-Abreu, D.C. Spray, A.C. Campos de Carvalho, M.V. Neto, Gap-junctional coupling between neurons and astrocytes in primary central nervous system cultures, Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 7541 – 7546. [10] B.P. Fulton, Gap junctions in the developing nervous system, Perspect. Dev. Neurobiol. 2 (1995) 327 – 334. [11] C. Giaume, L. Venance, Intercellular calcium signaling and gap junctional communication in astrocytes, Glia 24 (1998) 50 – 64. [12] J.L. Goldberg, B.A. Barres, The relationship between neuronal survival and regeneration, Annu. Rev. Neurosci. 23 (2000) 579 – 612.
[13] R.V. Haberberger, M. Henrich, K.S. Lips, W. Kummer, Nicotinic receptor alpha7-subunits are coupled to the stimulation of nitric oxide synthase in rat dorsal root ganglion neurons, Histochem. Cell Biol. 120 (2003) 173 – 181. [14] M. Hanani, Satellite glial cells in sensory ganglia: from form to function, Brain Res. Rev. (in press). [15] M. Hanani, O. Zamir, P. Baluk, Glial cells in the guinea pig myenteric plexus are dye coupled, Brain Res. 497 (1989) 245 – 249. [16] M. Hanani, T.Y. Huang, P.S. Cherkas, M. Ledda, E. Pannese, Glial cell plasticity in sensory ganglia induced by nerve damage, Neuroscience 114 (2002) 279 – 283. [17] F. Heblich, S. England, R.J. Docherty, Indirect actions of bradykinin on neonatal rat dorsal root ganglion neurones: a role for nonneuronal cells as nociceptors, J. Physiol. 536 (2001) 111 – 121. [18] K. Holthoff, O.W. Witte, Directed spatial potassium redistribution in rat neocortex, Glia 29 (2000) 288 – 292. [19] J. Kang, L. Jiang, S.A. Goldman, M. Nedergaard, Astrocyte mediated potentiation of inhibitory synaptic transmission, Nat. Neurosci. 1 (1998) 683 – 692. [20] T. Konishi, Dye coupling between mouse Schwann cells, Brain Res. 508 (1990) 85 – 91. .[21] M. Ledda, S. DePalo, E. Pannese, Ratios between number of neuroglial cells and number and volume of nerve cells in the spinal ganglia of two species of reptiles and three species of mammals, Tissue Cell 36 (2004) 55 – 62. [22] N. Maudlej, M. Hanani, Modulation of dye coupling among glial cells in the myenteric and submucosal plexuses of the guinea pig, Brain Res. 578 (1992) 94 – 98. [23] V. Miletic, G.W. Lu, Characteristics of action potentials recorded from cat spinal ganglion neurons in vivo, Brain Res. Bull. 31 (1993) 531 – 538. [24] J.I. Nagy, J.E. Rash, Connexins and gap junctions of astrocytes and oligodendrocytes in the CNS, Brain Res. Rev. 32 (2000) 29 – 44. [25] T. Nakase, S. Fushiki, C.C.G. Naus, Astrocytic gap junctions composed of connexin43 reduce apoptotic neuronal damage in cerebral ischemia, Stroke 34 (2003) 1987 – 1993. [26] E.A. Newman, New roles for astrocytes: regulation of synaptic transmission, Trends Neurosci. 26 (2003) 536 – 542. [27] E. Pannese, The satellite cells of sensory ganglia, Adv. Anat., Embryol. Cell Biol. 65 (1981) 1 – 111. [28] E. Pannese, M. Ledda, G. Arcidiacono, L. Rigamonti, Clusters of nerve cell bodies enclosed within a common connective tissue envelope in the spinal ganglia of the lizard and rat, Cell Tissue Res. 264 (1991) 209 – 214. [29] E. Pannese, P. Procacci, M. Ledda, V. Conte, The percentage of nerve cell bodies arranged in clusters decreases with age in the spinal ganglia of adult rabbits, Anat. Embryol. (Berlin) 187 (1993) 331 – 334. [30] E. Pannese, M. Ledda, P.S. Cherkas, T.Y. Huang, M. Hanani, Satellite cell reactions to axon injury of sensory ganglion neurons: increase in number of gap junctions and formation of bridges connecting previously separate perineuronal sheaths, Anat. Embryol. 206 (2003) 337 – 347. [31] C.A. Pappas, B.R. Ransom, Depolarization induced alkalinization (DIA) in rat hippocampal astrocytes, J. Neurophysiol. 72 (1994) 2816 – 2826. [32] B.R. Ransom, H. Kettenmann, Electrical coupling, without dye coupling, between mammalian astrocytes and oligodendrocytes in cell culture, Glia 3 (1990) 258 – 266. [33] B. Ransom, T. Behar, M. Nedergaard, New roles for astrocytes (stars at last), Trends Neurosci. 26 (2003) 520 – 522. [34] A. Reichenbach, Glial K+ permeability and CNS K+ clearance by diffusion and spatial buffering, Ann. N. Y. Acad. Sci. 633 (1991) 272 – 286. [35] R. Robitaille, Modulation of synaptic efficacy and synaptic depression by glial cells at frog neuromuscular junction, Neuron 21 (1998) 847 – 855.
T.-Y. Huang et al. / Brain Research 1036 (2005) 42–49 [36] A. Roos, W.F. Boron, Intercellular pH, Physiol. Rev. 61 (1981) 296 – 434. [37] N. Rouach, E. Avignone, W. Meme, A. Koulakoff, L. Venance, F. Blomstrand, C. Giaume, Gap junctions and connexin expression in the normal and pathological central nervous system, Biol. Cell 94 (2002) 457 – 475. [38] R. Rozental, A.C. Campos de Carvalho, D.C. Spray, Nervous system diseases involving gap junctions, Brain Res. Rev. 32 (2000) 189 – 191. [39] D.C. Spray, M.V. Bennett, Physiology and pharmacology of gap junctions, Annu. Rev. Physiol. 47 (1985) 281 – 303. [40] D.C. Spray, A.L. Harris, M.V.L. Bennett, Gap junctional conductance is a simple and sensitive function of intracellular pH, Science 211 (1981) 712 – 715. [41] T. Thippeswamy, R. Morris, The roles of nitric oxide in dorsal root ganglion neurons, Ann. N. Y. Acad. Sci. 962 (2002) 103 – 110.
49
[42] S. Tsukada, K. Keino-Masu, M. Masu, J. Fukuda, Activation of protein kinase A by nitric oxide in cultured dorsal root ganglion neurites of the rat, examined by a fluorescence probe, ARII, Neurosci. Lett. 318 (2002) 17 – 20. [43] D.I. Vaney, Many types of retinal neurons show tracer coupling when injected with biotin or neurobiotin, Neurosci. Lett. 125 (1991) 187 – 190. [44] M. Weick, P.S. Cherkas, W. H7rtig, T. Pannicke, O. Uckermann, A. Bringmann, M. Tal, A. Reichenbach, M. Hanani, P2 receptors in satellite glial cells in trigeminal ganglia of mice, Neuroscience 120 (2003) 969 – 977. [45] J. Zhang, S.H. Snyder, Nitric oxide in the nervous system, Annu. Rev. Pharmacol. Toxicol. 35 (1995) 213 – 233. [46] E. Zuriel, M. Devor, Dye coupling does not explain functional crosstalk within dorsal root ganglia, J. Peripher. Nerv. Syst. 6 (2001) 227 – 231.
Research report
Dye coupling among satellite glial cells in mammalian dorsal root ganglia Tian-Ying Huanga, Pavel S. Cherkasa, David W. Rosenthalb, Menachem Hanania,* a
Laboratory of Experimental Surgery, Hebrew University-Hadassah Medical School, Mount Scopus, Jerusalem 91240, Israel b Department of Medicine, Mayo Clinic and Foundation, Rochester, MN 55905, USA Accepted 4 December 2004
Abstract Dorsal root ganglia (DRG) are key elements in sensory signaling under physiological and pathological conditions. Little is known about electrical coupling among cells in these ganglia. In this study, we injected the fluorescent dye Lucifer yellow (LY) into single cells to examine dye coupling in DRG. We found no dye coupling between neurons or between neurons and their attendant satellite glial cells (SGCs). In mouse DRG, we observed that in 26.2% of the cases SGCs that surround a given neuron were dye coupled. In only 3.2% of the cases SGCs that make envelopes around different neurons were coupled. The data from mouse ganglia were very similar to those from rat and guinea pig DRG. The results obtained by injection of the tracer biocytin were very similar to those observed with LY. The coupling incidence within the envelopes increased 3.1-fold by high extracellular pH (8.0), but coupling between envelopes was not affected. Acidic pH (6.8) reduced the coupling. High extracellular K+ (9.4 mM) increased the coupling 2.4-fold and 4.7-fold within and between envelopes, respectively. Low extracellular Ca2+ (0.5, 1.0 mM) partly reversed the effect of high K+ on coupling. The results showed that SGCs in mammalian sensory ganglia are connected by gap junctions. This coupling is very sensitive to changes in pH, and can therefore be modulated under various physiological and pathological conditions. The dependence of the coupling on extracellular K+ and Ca2+ suggests that the permeability of gap junctions can be altered by physiological and pharmacological stimuli. D 2005 Elsevier B.V. All rights reserved. Theme: Cellular and molecular biology Topic: Neuroglia and myelin Keywords: Dorsal root ganglia; Neurons; Satellite glial cells; Gap junctions; Dye coupling; Lucifer yellow; Biocytin
1. Introduction Satellite glial cells (SGCs) are the most abundant cell type in dorsal root ganglia (DRG) but their physiological properties are little understood. In contrast, much is known about glial cells in the central nervous system (CNS), and it is well established that central glia provide structural, metabolic and trophic support to neurons [12,33]. These cells are active partners of neurons in signal processing and synaptic integration [19,26,35]. Intercellular coupling by gap junctions among glial cells is widespread throughout the CNS [11,24], and is believed
* Corresponding author. Fax: +972 2 5823515. E-mail address: [email protected] (M. Hanani). 0006-8993/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.brainres.2004.12.021
to be important for the functions of these cells [2,18,25]. Glial cells in the peripheral nervous system are also mutually coupled by gap junctions [15,20,22]. Altered coupling among glial cells in the CNS is known to be associated with various pathological states [37,38]. There is ultrastructural evidence for gap junctions in SGCs in sensory ganglia [27], but little is known about their function. We have shown that SGCs are mutually dye coupled and that sectioning of the axons of sensory neurons resulted in an increase in this coupling and in the number of gap junctions [4,16,30]. We hypothesized that increased coupling contributes to neuropathic pain states produced by peripheral nerve injury, and may account for the spread of pain sensation. It is quite likely that gap junctions among SGCs are essential also for the normal function of sensory ganglia, but so far these cells have received very little
T.-Y. Huang et al. / Brain Research 1036 (2005) 42–49
attention. In the present study, we used intracellular dye injection to compare coupling of SGCs in three mammalian species. Dye coupling was measured using two different tracers, and the effects of extracellular pH, K+ and Ca2+ on dye coupling were examined.
2. Materials and methods 2.1. Animals and preparations Experiments were done on Balb/c mice of either sex, male Sprague–Dawley rats, and male guinea pigs, all 3–5 months old. The experimental protocol was approved by the Institutional Animal Care and Use Committee. The animals were sacrificed by CO2. DRG L4,5 were removed and placed in cold (4 8C) Krebs solution (pH 7.4) containing (mM): 118 NaCl, 4.7 KCl, 14.4 NaHCO3, 1.2 MgSO4, 2.5 CaCl2, 1.2 NaH2PO4 and 11.5 glucose. The ganglia were pinned onto the Sylgard bottom of a chamber superfused with Krebs solution bubbled with 95% O2 and 5% CO2 at 23–24 8C. The pH in the bathing solution was altered by substituting HCO3 with Cl or Cl with HCO3 . In the experiments with altered HCO3 , the tissues were first incubated for 30 min in the low- or high-pH solutions, and the injections were made in the same solutions during the next 4 h. High K+ was achieved by substituting Na+ with K+, and low Ca2+ by substituting Ca2+ with Na+. 2.2. Intracellular labeling Experiments were performed using an upright microscope (Axioskop FS, Zeiss, Jena, Germany) equipped with fluorescent illumination and a digital camera (Pixera 120e, Pixera, Los Gatos, USA) connected to a PC. Neurons and SGCs were singly injected with the fluorescent dye Lucifer yellow (LY, Sigma, St. Louis, MO, USA) 3% in 0.5 M LiCl solution, or 4% biocytin (Sigma) in 1 M KCl from a glass microelectrode connected to a preamplifier (Neuro Data Instrument model IR 283, New York, NY, USA). The tip resistance of the microelectrodes was 80–120 M V. Both LY and biocytin were injected intracellularly by current pulses 100 ms in duration and 0.5 nA in amplitude at 10 Hz. LY was driven by hyperpolarizing pulses for 3–5 min, and biocytin by depolarizing pulses for 15 min. LY injection times were determined by observing the staining of the cell in real time. We found that injections of LY for more than 5 min did not alter the extent of dye coupling. During and after LY injections, the living labeled cells were imaged with the digital camera. After the experiments, the DRG were fixed overnight at 4 8C in 4% paraformaldehyde in phosphate buffer solution (PBS, pH 7.4) and washed with PBS. The tissues injected with LY were mounted in Gel/mount (Biomeda, Foster, CA, USA) and the LY-labeled cells were imaged with a Bio-Rad
43
confocal microscope (Bio-Rad, Hercules, CA, USA). Before processing, the tissues with biocytin-injected cells were incubated for 15 min in PBS containing 1% H2O2. After washing with PBS, the tissues were incubated in PBS containing 0.5% Triton X-100 and 1 Ag/ml avidin– HRP (Sigma) for 2 h. After five washes with PBS, the tissues were reacted with diaminobenzidine (Sigma, 2 mg/ ml) and 0.01% H2O2 in PBS for 40 min. They were then dehydrated, cleared with xylene and mounted in Entellan (Merck, Darmstadt, Germany). 2.3. Statistical analysis All average values were expressed as mean F SEM. Fisher’s Exact Test and Mann–Whitney test were used for comparison. P b 0.05 is considered as statistically significant.
3. Results 3.1. Dye coupling experiments Intracellular dye injection showed that none of the DRG neurons (mice, n = 95; rats, n = 93; guinea pigs, n = 82) that were examined in the three species used was coupled to any other cell (neurons or SGCs), as shown previously [16,46]. Furthermore, when the dye was injected into SGCs, there was no dye passage from them into neurons (n = 126, 4 mice; n = 127, 4 rats; n = 142, 5 guinea pigs). We noticed two types of SGCs coupling: between SGCs that make the envelope around a given neuron (Figs. 1A, B), and between SGCs surrounding different neurons (Fig. 1C). The incidence of the dye-injected SGCs that were coupled to other SGCs within the same envelope was 26.2%, 28.3% and 29.6% in mouse, rat and guinea pig, respectively (Fig. 2A). Coupling between SGCs belonging to different envelopes was rare: 3.2%, 3.1% and 3.5% in mouse, rat and guinea pig, respectively (Fig. 2A). Although the incidence of dye coupling among SGCs was very similar in the three species, the number (mean F SEM) of SGCs coupled to the dye-injected cells in the guinea pig was greater than in the mouse and rat, 5.69 F 0.74 (n = 42, 20 ganglia of 5 guinea pigs) vs. 2.15 F 0.24 (n = 33, 16 ganglia of 4 mice) and 2.08 F 0.24 (n = 36, 16 ganglia of 4 rats), (P b 0.001, Mann–Whitney test), as shown in Fig. 2B. Furthermore, the maximum number of SGCs around a given neuron was greater in guinea pig than in mouse or rat ganglia. In guinea pigs, the maximum was 21 SGCs within an envelope (Fig. 1B), whereas in mice and rats it was eight. Vaney [43] showed that due to the small size of biocytin, this tracer can reveal cell coupling not detected with LY. To examine whether biocytin injection could disclose a different pattern of coupling, we injected it intracellularly into neurons and SGCs in mouse DRG L4,5. We found that
44
T.-Y. Huang et al. / Brain Research 1036 (2005) 42–49
experiments. The experiments with LY and biocytin showed that glial cells surrounding small and large neurons displayed similar coupling patterns. 3.2. The influence of pH on dye coupling in mouse DRG
Fig. 1. Micrographs showing dye coupling between and within glial envelopes in DRG. (A) The dye-injected satellite glial cell (SGC) is coupled only to other SGCs in the same envelope; mouse DRG. (B) An example of dye coupling within an envelope in guinea pig DRG. (C) The dye-injected SGC is coupled to other SGCs forming envelopes around different neurons in mouse DRG. (D) An example of intracellular injection of biocytin showing coupling within a glial envelope in mouse DRG. Arrows indicate four labeled SGCs. Panels (A–C) are confocal images. Asterisks indicate the dye-injected cells. In panel (D), the biocytin-injected cell is not indicated because this method does not allow the observation of the injected cells during the experiment. Scale bars = 20 Am.
neurons were not coupled to each other or to SGCs (n = 55, 22 ganglia of 6 mice). With biocytin as a tracer, coupling incidence among SGCs within envelopes was 31.3% (n = 48, P N 0.5, Fisher’s Exact Test, compared with the LY results), see Fig. 1D, and coupling between glial envelopes was 4.2% (P N 0.6). The number of SGCs coupled to the biocytin-injected cells was 2.20 F 0.38 (n = 15, P N 0.9, Mann–Whitney test) vs. 2.15 F 0.24 for LY. Thus, the results obtained with these two tracers in mouse DRG were very similar. Consequently, we used LY in all the following
As the incidence of SGC dye coupling was very similar in the ganglia from the three species used, we focused on mouse DRG for the remainder of this study. Previous studies showed that the permeability of gap junctions is strongly influenced by changes in intracellular pH [39]. To find out whether dye coupling among SGCs depends on pH, we altered the pH in the bathing solution and examined the dye coupling among SGCs. We found that dye coupling among SGCs depended strongly on extracellular pH. Slightly basic pH (7.8 or 8.0) increased the incidence of dye coupling among SGCs within envelopes, and also the number of SGCs coupled to the dye-injected cells. The incidence of dye coupling among SGCs within envelopes increased 2.4-fold to 63.9% (n = 97, 16 ganglia, 4 mice, P b 0.0001, Fisher’s Exact Test) at pH 7.8, and 3.1fold to 81.3% (n = 123, 20 ganglia, 5 mice, P b 0.0001) at pH 8.0. At pH 6.8, the incidence of coupling within envelopes decreased from 26.2% to 12.5% (n = 104, 16 ganglia, 4 mice, P b 0.05). No significant change in dye coupling between envelopes was observed after changing to acidic or basic pH. The average number of SGCs coupled to the dye-injected cells increased 1.5-fold to 3.32 F 0.22 (n = 62, P b 0.05, Mann–Whitney test) at pH 7.8 and 1.6-fold to 3.34 F 0.20 (n = 100, P b 0.05) at pH 8.0, and decreased by half to 1.69 F 0.38 (n = 13, P b 0.05) at pH 6.8 in comparison with 2.15 F 0.24 (n = 33) at pH 7.4, as shown in Fig. 3. To determine whether the effect of pH on dye coupling is reversible, we incubated the ganglia in Krebs solution at pH 8.0 for 2 h, washed for 1 h in normal Krebs solution (pH 7.4) and then examined dye coupling among SGCs. The coupling incidence was 28.4% (n = 95, 12 ganglia, 3 mice, P N 0.7, Fisher’s Exact Test) within, and 3.2% (P N 0.9) between glial envelopes, vs. 26.2% and
Fig. 2. The incidence of dye coupling among SGCs in DRG of three different mammals. (A) The incidence of dye coupling between (black bars) and within (empty bars) glial envelopes in mouse, rat and guinea pig ganglia. (B) Histograms show the number (mean F SEM) of SGCs coupled to the dye-injected glial cells in mouse, rat and guinea pig DRG. * indicates P b 0.001, as compared with mouse and rat ganglia. Only cases were dye coupling was observed were included in this analysis. Mouse-b denotes the data for the mouse DRG after using biocytin injection. Mann–Whitney test was used for comparison.
T.-Y. Huang et al. / Brain Research 1036 (2005) 42–49
45
Fig. 3. Dye coupling among SGCs in the mouse DRG depends on pH. (A) The incidence of dye coupling within envelopes (empty bars) increased at basic pH, whereas coupling between envelopes (black bars) showed weak dependence on pH. (B) Effect of pH on the number (mean F SEM) of SGCs coupled to the dye-injected glial cells. The data for the groups (pH 8.0 Y 7.4 and pH 6.8 Y 7.4) were obtained from the experiments where we tested the reversibility of the high pH effect. In these experiments, we first incubated the DRG at pH 8.0 or 6.8 for 2 h, washed the tissues at pH 7.4 for 1 h and then examined the dye coupling among SGCs at pH 7.4. *P b 0.05, **P b 0.001, as compared with controls (pH 7.4). Fisher’s Exact Test (A) and Mann–Whitney test (B) were used for comparison.
3.2% in control DRG. The number of SGCs coupled to the LY-injected cells was 2.26 F 0.30 (n = 27, P N 0.8) vs. 2.15 F 0.24 in control DRG. Similarly, after incubation of the DRG at pH 6.8 for 2 h and then washing for 1 h, the coupling incidence recovered from 12.5% to 25% (n = 68, 8 ganglia, 2 mice, P b 0.05) within the envelopes, and the incidence of coupling between glial envelopes recovered to 2.9%. The number of SGCs coupled to the LY-injected cells also recovered from 1.69 F 0.38 to 2.12 F 0.39 (n = 17, P b 0.05), see Fig. 3. These results showed that the effect of extracellular pH on dye coupling among SGCs was reversible. 3.3. The effects of K+ and Ca2+ on dye coupling in mouse DRG Extracellular K+ and Ca2+ have been found to influence coupling among astrocytes [5,8]. We therefore decided to examine the effects of these ions on SGC coupling. At short incubation times (10–55 min) after increasing the concentration of extracellular K+ ([K+]o) from 4.7 to 9.4 mM, coupling incidence among SGCs within an envelope increased from 26.2 to 63.2% (P b 0.0001, n = 68, 16 ganglia of 4 mice); however, the incidence of coupling between envelopes did not change (2.9 vs. 3.2% in control, P N 0.1). In contrast, after longer times (70–240 min) in high [K+]o coupling between envelopes increased 4.7-fold to 14.9% (P b 0.01, n = 94, 16 ganglia, 4 mice). The incidence of coupling within SGC envelopes after long incubation times was similar to that observed at short times (67%, P b 0.0001 compared with control, Fisher’s Exact Test). After long incubation times, the number of SGCs coupled to the dye-injected cells increased from 2.15 F 0.24 (n = 33) to 4.68 F 0.54 (n = 65, P b 0.05, Mann– Whitney test). To find out whether the influence of high K+ on dye coupling among SGCs is reversible, we examined coupling after incubation of the DRG in high K+ Krebs
solution for 3 h and then washing for 3 h. The coupling incidence decreased from 67% (high K+) to 31.3% ( P b 0.01, n = 96, 12 ganglia, 3 mice) within an SGC envelope, and from 14.9% (high K+) to 6.3% (P b 0.05) between SGC envelopes and the number of SGCs coupled to the LY-injected cells decreased from 4.68 F 0.54 (high K+) to 2.67 F 0.35 (P b 0.05, n = 30). There was no significant difference between these data and the data obtained from control ganglia (Fig. 4). The results showed that high K+induced change in dye coupling among SGCs was almost completely reversible. To determine whether the incubation time itself affected dye coupling, we compared dye coupling at short (10–55 min) and long (1–4 h) incubation times in normal solution. The results showed that dye coupling between envelopes and within an envelope were 2.5%, 25% (n = 40) during 10–55 min incubation, and 3.5%, 26.7% (n = 86) in 1–4 h incubation (both P N 0.8, Fisher’s Exact Test). Thus, incubation time did not increase dye coupling. The effect of high [K+]o may be due to depolarization of SGCs themselves and/or neurons, which may lead to Ca2+ influx into these cells (SGCs and neurons).To test this possibility, we repeated the high [K+]o experiments at 70–240 min using a medium with low [Ca2+]o. We found that decreasing [Ca2+]o from 2.5 to 1.0 or 0.5 mM partly reversed the high [K+]o effect. In the presence of high [K+]o and low [Ca2+]o, the incidence of dye coupling between envelopes decreased from 14.9% to 3.7% (1.0 mM [Ca2+]o, n = 108, 16 ganglia of 4 mice, P b 0.01, Fisher’s Exact Test) and to 3.2% (0.5 mM [Ca2+]o, n = 93, 16 ganglia of 4 mice, P b 0.01). The effect of low [Ca2+]o on the incidence of dye coupling within envelopes was less striking, this coupling decreased slightly but significantly from 67% to 52.8% (1.0 mM [Ca2+]o, P b 0.05) and to 51.6% (0.5 mM [Ca2+]o, P b 0.05). The mean number of SGCs coupled to the dye-injected cells slightly decreased from 4.68 F 0.54 (n = 65) to 3.28 F 0.33
46
T.-Y. Huang et al. / Brain Research 1036 (2005) 42–49
Fig. 4. Dye coupling among SGCs in the mouse DRG increased following 70–240 min after increasing [K+]o, and also depended on [Ca2+]o. (A) The effect of K+ and Ca2+ on the incidence of dye coupling between glial envelopes. The full effect of high K+, requires a normal external Ca2+. (B) The effect of K+ and Ca2+ on the incidence of dye coupling within envelopes. This type of coupling depended weakly on Ca2+. (C) The effect of K+ and Ca2+ on the number (mean F SEM) of SGCs coupled to the dye-injected glial cells. The data for the group (K+ 9.4 Y 4.7 mM, Ca2+ 2.5 mM) were obtained from the experiments where we tested the reversibility of the high K+ effect. In these experiments, the DRG were first incubated in high extracellular K+ (9.4 mM, Ca2+ 2.5 mM) for 3 h and were then washed in normal Krebs solution (K+ 4.7 mM, Ca2+ 2.5 mM) for 3 h. Following this, dye coupling among SGCs was examined in normal Krebs solution. *P b 0.05, **P b 0.001, as compared with control (K+ 4.7 mM and Ca2+ 2.5 mM). Fisher’s Exact Test (A, B) and Mann–Whitney test (C) were used for comparison.
(n = 54, 1.0 mM [Ca2+]o, P b 0.05, Mann–Whitney test) and to 3.32 F 0.30 (n = 62, 0.5 mM [Ca2+]o, P b 0.05), as shown in Fig. 4.
4. Discussion Intracellular labeling of SGCs with LY and biocytin revealed two types of dye coupling among these cells: coupling within the glial envelope of individual neurons, and between glial cells belonging to envelopes around different neurons. Dye coupling of SGCs within an envelope was much more prevalent and its incidence changed with extracellular pH and K+. In contrast, dye coupling of SGCs in different envelopes was rare and its incidence was not affected by pH, but increased by high [K+]o. Interestingly, the effect of high [K+]o on dye coupling between envelopes was completely abolished by decreasing [Ca2+]o. Our observation of the inhibitory effects of acidic pH (which blocks gap junctions) on dye coupling and previous ultrastructural studies support the assumption that the dye coupling is mediated by gap junctions [16,30]. The results on SGC coupling incidence for mice, rats and guinea pigs were very similar, but the number of SGCs coupled to the dye-injected cells in guinea pig DRG was greater than those in the mouse and rat. This difference may be due to the larger number of SGCs within an envelope around a given neuron in guinea pigs. This is in accord with the results of Ledda et al. [21], who showed that in large animals DRG neurons are larger than in small animals, and the number of SGCs around a given neuron increases with the neuronal volume. Whether coupling among SGCs contributes to the function of DRG is still unknown. Coupling among glial cells is known to be prominent in the CNS [11,24], in the developing nervous system [3,10], and in cultured brain tissue [8,9]. Glial coupling is essential for spatial K+
buffering in the CNS [2,18,25,34]. Similarly, coupling among SGCs is likely to contribute to normal functions of DRG. Altered coupling among glial cells is associated with various pathological states in the brain [37,38]. Axotomy of the sciatic nerve was found to increase dye coupling among SGCs in mouse DRG [16,30], and it was proposed that augmented glial coupling can contribute to neuropathic pain. The idea of intercellular coupling in DRG has been alluded to in several previous studies. There is evidence for bcross talkQ between DRG neurons [6], but the proposed mechanism is not electrical coupling, but chemical transmission [1]. Miletic and Lu [23] presented evidence for signal transmission between DRG neurons, and proposed that it is mediated either by chemical synapses or by electrotonic coupling. It can be suggested that glial coupling described here might contribute to such inter-neuronal coupling. This will require that neurons and glial cells interact via chemical messengers. However, although previous studies suggested that SGCs and sensory neurons in DRG can interact [14,17], the precise nature of this interaction is still not known. The topic of neuron-glia communication in DRG clearly warrants further investigation. It should be mentioned that the lack of dye coupling between cells does not necessarily indicate the lack of electrical coupling [32]. This certainly holds when using LY, which in some systems appears not to cross gap junctions [43]. To test whether a smaller tracer molecule will reveal a greater extent of coupling, we injected biocytin into SGCs and neurons, but the results were very similar to those obtained with LY. This indicates that the gap junctions in SGCs are equally permeable to LY and biocytin, suggesting that their pore is relatively large. Under control conditions (pH 7.4), only 26.2% of the cells was dye coupled within the envelopes. The reason for this low incidence is not clear. At slightly basic pH (8.0), this coupling incidence increased to 81.3%, suggesting that the majority of the SGCs are connected by gap junctions, which are mostly closed at pH 7.4. One possible explanation
T.-Y. Huang et al. / Brain Research 1036 (2005) 42–49
for this observation is that because of the experimental conditions the intracellular pH is acidic. Previous studies have shown that changing the extracellular pH altered the intracellular pH in the same direction but with smaller amplitude, when using bicarbonate as a buffer [36], and that maximal gap junctional conductance occurred at intracellular pH around 7.6–7.8 [40]. The present results for the effect of extracellular pH on dye coupling among SGCs are consistent with these observations. Furthermore, low pH might be the result of damage, but in the present case we believe that this is unlikely because we worked under standard conditions used for isolated tissues (see Materials and methods). In previous studies, in our laboratory the same experimental conditions were used to study dye coupling in enteric glial cells, and a high (about 90%) incidence of coupling was found [15,22]. Therefore, we propose that under basal conditions many of the gap junctions in SGCs are closed, and are opened by a slightly basic pH or by other, yet unknown, physiological parameters. The incidence of dye coupling between SGCs in different envelopes was very low (3.2%) under control conditions. Pannese et al. [28] found that DRG neurons are sometimes arranged in clusters surrounded by a common connective tissue space. This arrangement is more common in immature animals than adult animals [29]. As we used young adult animals in the present work, a probable explanation is that dye coupling between envelopes is due to the presence of these clusters. This explanation is based on the assumption that SGCs surrounding a cluster are dye coupled. Therefore, when a tracer molecule was injected into individual SGCs within the sheath surrounding a cluster, SGCs around two or more neurons could be stained. This type of coupling was not affected by pH, but was previously found to be blocked by octanol [16]. The reason for the lack of influence of pH on the inter-envelope coupling is not yet clear. We observed that high [K+]o (9.4 mM) increased the incidence of dye coupling both within and between envelopes. The effect of high [K+]o depended on [Ca2+]o. Decreasing [Ca2+]o from 2.5 to 1 or 0.5 mM slightly decreased the high [K+]o-induced rise in coupling within envelopes but completely reversed the increase in dye coupling between the envelopes. The duration of the incubation in high [K+]o had a striking influence on the effects observed. A relatively brief (10–55 min) exposure of the tissue to high [K+]o caused an increase in the coupling among SGCs within the envelope, but did not affect the inter-envelope coupling. In contrast, longer exposures (70–240 min) in high [K+]o caused a marked (4.7-fold) increase in the inter-envelope coupling with little influence on the coupling within the envelope. Furthermore, inter-envelope coupling was much more sensitive to the lowering of [Ca2+]o. We can offer two possible explanations for these results: (1) It can be proposed that depolarization of SGCs by high [K+]o induces Ca2+ influx
47
into these cells through voltage dependent Ca2+ channels. The resulting elevation of intracellular Ca2+ concentration may increase the number of open gap junctions. Such a model has been proposed to explain the increase in coupling among astrocytes under high [K+]o [5]. (2) High [K+]o causes the depolarization of neurons, which increases the Ca2+ conductance of the neuronal plasma membranes, followed by Ca2+ influx and an elevation of intracellular Ca2+ concentration. This in turn may lead to the release of chemical messenger(s) from the neurons, which then alter intercellular coupling. The long delay for the effect on the inter-envelope coupling suggests that long term metabolic processes are involved. The chemical messenger(s) for this high [K+]o effect is not yet known but one candidate is nitric oxide (NO). High [K+]o is known to depolarize neurons and increase its NO production via elevation of cytoplasmic Ca2+ [45]. NO activates soluble guanylate cyclase leading to an increase in cyclic GMP levels, which then promote neurite growth in DRG neurons [42]. In DRG, NO was found to elevate cGMP levels in SGCs via the activation of guanylate cyclase [13,41]. Consequently, it is conceivable that high [K+]o-induced depolarization of NOS containing neurons in DRG leads to NO synthesis, which in turn acts on the neighboring SGCs. The production of cGMP in SGCs may lead to the formation of new gap junctions among these cells, or to the opening of existing gap junctions. Of course, a combination of direct depolarization of SGCs by high [K+]o and mediation by neurons cannot be excluded. The validation of either scheme will require further work. The short term influence of high K+ on dye coupling requires a different explanation from the one proposed above. Enkvist and McCarthy [8] reported that high [K+]o induced an increase in dye coupling among cultured astrocytes and suggested that it might be related to a direct effect of depolarization on gap junction conductance. Depolarization was also found to induce alkalinization in astrocytes [31]. Thus, the increased dye coupling within glial envelopes may be due to the opening of existing gap junctions between SGCs around the same neuron via depolarization and/or alkalinization of SGCs. However, a more likely explanation is that depolarization induced Ca2+ influx into the cells, which initiated events leading to the opening of gap junction channels as proposed by De PinaBenabou et al. [5]. The functional significance of SGC coupling remains to be explored. There is much evidence from studies in the CNS that glial coupling is important for K+ buffering and possibly for other functions [34,37], and it can be proposed that SGCs perform similar roles. We found that SGC coupling is very sensitive to changes in pH, and as pH changes can occur under various circumstances such as inflammation and high metabolic activity, it is likely that SGC coupling is altered under certain physiological and pathological conditions. The findings on the long term effect of high [K+]o indicate that gap junctions among SGCs are
48
T.-Y. Huang et al. / Brain Research 1036 (2005) 42–49
highly plastic, as found previously following axotomy [16]. There is evidence that SGCs in sensory ganglia have functional receptors for bradykinin [7,14] and ATP [14,44], and it is conceivable that the activation of these receptors can also alter SGC coupling. In summary, we showed by intracellular dye injection that SGCs are mutually connected via gap junctions. The dye coupling was quite similar in mice, guinea pigs and rats, suggesting that the findings are typical for mammals in general. We showed that certain experimental conditions can alter SGC coupling and it will be most interesting to explore how coupling changes under a variety of situations, such as aging and disease.
Acknowledgments This work was supported by the Israel Science Foundation and by the Hebrew University Center for Pain Research. We thank Dr. Naomi Melamed-Book for helping with the confocal imaging.
References [1] R. Amir, M. Devor, Chemically mediated cross-excitation in rat dorsal root ganglia, J. Neurosci. 16 (1996) 4733 – 4741. [2] F. Amzica, M. Massimini, A. Manfridi, Spatial buffering during slow and paroxysmal sleep oscillations in cortical networks of glial cells in vivo, J. Neurosci. 22 (2002) 1042 – 1053. [3] K. Bittman, D.L. Becker, F. Cicirata, J.G. Parnavelas, Connexin expression in homotypic and heterotypic cell coupling in the developing cerebral cortex, J. Comp. Neurol. 443 (2002) 201 – 212. [4] P.S. Cherkas, T.Y. Huang, T. Pannicke, M. Tal, A. Reichenbach, M. Hanani, The effects of axotomy on neurons and satellite glial cells in mouse trigeminal ganglion, Pain 110 (2004) 290 – 298. [5] M.H. De Pina-Benabou, M. Srinivas, D.C. Spray, E. Scemes, Calmodulin kinase pathway mediates the K+-induced increase in Gap junctional communication between mouse spinal cord astrocytes, J. Neurosci. 21 (2001) 6635 – 6643. [6] M. Devor, P.D. Wall, Cross-excitation in dorsal root ganglia of nerve-injured and intact rats, J. Neurophysiol. 64 (1990) 1733 – 1746. [7] S. England, F. Heblich, I.F. James, J. Robbins, R.J. Docherty, Bradykinin evokes a calcium-activated chloride current in nonneuronal satellite cells isolated from neonatal dorsal root ganglion, J. Physiol. (London) 530 (2001) 395 – 403. [8] M.O.K. Enkvist, K.D. McCarthy, Astroglial gap junction communication is increased by treatment with either glutamate or high K+ concentration, J. Neurochem. 62 (1994) 489 – 495. [9] M.M. Froes, A.H. Correia, J. Garcia-Abreu, D.C. Spray, A.C. Campos de Carvalho, M.V. Neto, Gap-junctional coupling between neurons and astrocytes in primary central nervous system cultures, Proc. Natl. Acad. Sci. U. S. A. 96 (1999) 7541 – 7546. [10] B.P. Fulton, Gap junctions in the developing nervous system, Perspect. Dev. Neurobiol. 2 (1995) 327 – 334. [11] C. Giaume, L. Venance, Intercellular calcium signaling and gap junctional communication in astrocytes, Glia 24 (1998) 50 – 64. [12] J.L. Goldberg, B.A. Barres, The relationship between neuronal survival and regeneration, Annu. Rev. Neurosci. 23 (2000) 579 – 612.
[13] R.V. Haberberger, M. Henrich, K.S. Lips, W. Kummer, Nicotinic receptor alpha7-subunits are coupled to the stimulation of nitric oxide synthase in rat dorsal root ganglion neurons, Histochem. Cell Biol. 120 (2003) 173 – 181. [14] M. Hanani, Satellite glial cells in sensory ganglia: from form to function, Brain Res. Rev. (in press). [15] M. Hanani, O. Zamir, P. Baluk, Glial cells in the guinea pig myenteric plexus are dye coupled, Brain Res. 497 (1989) 245 – 249. [16] M. Hanani, T.Y. Huang, P.S. Cherkas, M. Ledda, E. Pannese, Glial cell plasticity in sensory ganglia induced by nerve damage, Neuroscience 114 (2002) 279 – 283. [17] F. Heblich, S. England, R.J. Docherty, Indirect actions of bradykinin on neonatal rat dorsal root ganglion neurones: a role for nonneuronal cells as nociceptors, J. Physiol. 536 (2001) 111 – 121. [18] K. Holthoff, O.W. Witte, Directed spatial potassium redistribution in rat neocortex, Glia 29 (2000) 288 – 292. [19] J. Kang, L. Jiang, S.A. Goldman, M. Nedergaard, Astrocyte mediated potentiation of inhibitory synaptic transmission, Nat. Neurosci. 1 (1998) 683 – 692. [20] T. Konishi, Dye coupling between mouse Schwann cells, Brain Res. 508 (1990) 85 – 91. .[21] M. Ledda, S. DePalo, E. Pannese, Ratios between number of neuroglial cells and number and volume of nerve cells in the spinal ganglia of two species of reptiles and three species of mammals, Tissue Cell 36 (2004) 55 – 62. [22] N. Maudlej, M. Hanani, Modulation of dye coupling among glial cells in the myenteric and submucosal plexuses of the guinea pig, Brain Res. 578 (1992) 94 – 98. [23] V. Miletic, G.W. Lu, Characteristics of action potentials recorded from cat spinal ganglion neurons in vivo, Brain Res. Bull. 31 (1993) 531 – 538. [24] J.I. Nagy, J.E. Rash, Connexins and gap junctions of astrocytes and oligodendrocytes in the CNS, Brain Res. Rev. 32 (2000) 29 – 44. [25] T. Nakase, S. Fushiki, C.C.G. Naus, Astrocytic gap junctions composed of connexin43 reduce apoptotic neuronal damage in cerebral ischemia, Stroke 34 (2003) 1987 – 1993. [26] E.A. Newman, New roles for astrocytes: regulation of synaptic transmission, Trends Neurosci. 26 (2003) 536 – 542. [27] E. Pannese, The satellite cells of sensory ganglia, Adv. Anat., Embryol. Cell Biol. 65 (1981) 1 – 111. [28] E. Pannese, M. Ledda, G. Arcidiacono, L. Rigamonti, Clusters of nerve cell bodies enclosed within a common connective tissue envelope in the spinal ganglia of the lizard and rat, Cell Tissue Res. 264 (1991) 209 – 214. [29] E. Pannese, P. Procacci, M. Ledda, V. Conte, The percentage of nerve cell bodies arranged in clusters decreases with age in the spinal ganglia of adult rabbits, Anat. Embryol. (Berlin) 187 (1993) 331 – 334. [30] E. Pannese, M. Ledda, P.S. Cherkas, T.Y. Huang, M. Hanani, Satellite cell reactions to axon injury of sensory ganglion neurons: increase in number of gap junctions and formation of bridges connecting previously separate perineuronal sheaths, Anat. Embryol. 206 (2003) 337 – 347. [31] C.A. Pappas, B.R. Ransom, Depolarization induced alkalinization (DIA) in rat hippocampal astrocytes, J. Neurophysiol. 72 (1994) 2816 – 2826. [32] B.R. Ransom, H. Kettenmann, Electrical coupling, without dye coupling, between mammalian astrocytes and oligodendrocytes in cell culture, Glia 3 (1990) 258 – 266. [33] B. Ransom, T. Behar, M. Nedergaard, New roles for astrocytes (stars at last), Trends Neurosci. 26 (2003) 520 – 522. [34] A. Reichenbach, Glial K+ permeability and CNS K+ clearance by diffusion and spatial buffering, Ann. N. Y. Acad. Sci. 633 (1991) 272 – 286. [35] R. Robitaille, Modulation of synaptic efficacy and synaptic depression by glial cells at frog neuromuscular junction, Neuron 21 (1998) 847 – 855.
T.-Y. Huang et al. / Brain Research 1036 (2005) 42–49 [36] A. Roos, W.F. Boron, Intercellular pH, Physiol. Rev. 61 (1981) 296 – 434. [37] N. Rouach, E. Avignone, W. Meme, A. Koulakoff, L. Venance, F. Blomstrand, C. Giaume, Gap junctions and connexin expression in the normal and pathological central nervous system, Biol. Cell 94 (2002) 457 – 475. [38] R. Rozental, A.C. Campos de Carvalho, D.C. Spray, Nervous system diseases involving gap junctions, Brain Res. Rev. 32 (2000) 189 – 191. [39] D.C. Spray, M.V. Bennett, Physiology and pharmacology of gap junctions, Annu. Rev. Physiol. 47 (1985) 281 – 303. [40] D.C. Spray, A.L. Harris, M.V.L. Bennett, Gap junctional conductance is a simple and sensitive function of intracellular pH, Science 211 (1981) 712 – 715. [41] T. Thippeswamy, R. Morris, The roles of nitric oxide in dorsal root ganglion neurons, Ann. N. Y. Acad. Sci. 962 (2002) 103 – 110.
49
[42] S. Tsukada, K. Keino-Masu, M. Masu, J. Fukuda, Activation of protein kinase A by nitric oxide in cultured dorsal root ganglion neurites of the rat, examined by a fluorescence probe, ARII, Neurosci. Lett. 318 (2002) 17 – 20. [43] D.I. Vaney, Many types of retinal neurons show tracer coupling when injected with biotin or neurobiotin, Neurosci. Lett. 125 (1991) 187 – 190. [44] M. Weick, P.S. Cherkas, W. H7rtig, T. Pannicke, O. Uckermann, A. Bringmann, M. Tal, A. Reichenbach, M. Hanani, P2 receptors in satellite glial cells in trigeminal ganglia of mice, Neuroscience 120 (2003) 969 – 977. [45] J. Zhang, S.H. Snyder, Nitric oxide in the nervous system, Annu. Rev. Pharmacol. Toxicol. 35 (1995) 213 – 233. [46] E. Zuriel, M. Devor, Dye coupling does not explain functional crosstalk within dorsal root ganglia, J. Peripher. Nerv. Syst. 6 (2001) 227 – 231.