SEARCH

SEARCH BY CITATION

Keywords:

  • Helisoma trivolvis;
  • growth cone;
  • filopodia;
  • intracellular Ca;
  • neuronal excitability

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

In addition to acting as a classical neurotransmitter in synaptic transmission, acetylcholine (ACh) has been shown to play a role in axonal growth and growth cone guidance. What is not well understood is how ACh acts on growth cones to affect growth cone filopodia, structures known to be important for neuronal pathfinding. We addressed this question using an identified neuron (B5) from the buccal ganglion of the pond snail Helisoma trivolvis in cell culture. ACh treatment caused pronounced filopodial elongation within minutes, an effect that required calcium influx and resulted in the elevation of the intracellular calcium concentration ([Ca]i). Whole-cell patch clamp recordings showed that ACh caused a reduction in input resistance, a depolarization of the membrane potential, and an increase in firing frequency in B5 neurons. These effects were mediated via the activation of nicotinic acetylcholine receptors (nAChRs), as the nAChR agonist dimethylphenylpiperazinium (DMPP) mimicked the effects of ACh on filopodial elongation, [Ca]i elevation, and changes in electrical activity. Moreover, the nAChR antagonist tubucurarine blocked all DMPP-induced effects. Lastly, ACh acted locally at the growth cone, because growth cones that were physically isolated from their parent neuron responded to ACh by filopodial elongation with a similar time course as growth cones that remained connected to their parent neuron. Our data revealed a critical role for ACh as a modulator of growth cone filopodial dynamics. ACh signaling was mediated via nAChRs and resulted in Ca influx, which, in turn, caused filopodial elongation. © 2013 Wiley Periodicals, Inc. Develop Neurobiol 73: 487–501, 2013


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

During early development, neurons extend neurites to connect to appropriate target cells. Growth cones at the tip of growing neurites are important for pathfinding and its filopodia serve as sensors to probe the environment for guidance cues (Rehder et al., 1996; Gomez and Zheng, 2006; Farrar and Spencer, 2008). ACh, besides serving as a classical neurotransmitter in synaptic transmission, has been shown to play an unconventional role in axonal growth (Lauder and Schambra, 1999; Phillis, 2005). ACh inhibits neurite outgrowth in several neuronal cell types (Owen and Bird, 1995; Small et al., 1995; Rudiger and Bolz, 2008), and induces positive turning responses of growth cones in Xenopus spinal neurons in vitro (Zheng et al., 1994). In molluscan nervous systems, ACh is suggested to have a role in neurite extension (McCobb et al., 1988). However, neither the functions of ACh at the level of the growth cone nor the underlying mechanisms by which ACh affects growth cone filopodial dynamics have yet been fully understood.

Electrical activity has been shown to affect the neurite outgrowth rate of developing neurons (Neely and Nicholls, 1995). Evoked action potentials cease neurite outgrowth and growth cone advance in both vertebrate and invertebrate neurons (Cohan and Kater, 1986; Fields et al., 1990), and the depolarization-induced suppression of neurite elongation requires an increase in cytoplasmic Ca (Cohan, 1992). A more recent study further identified the critical role of electrical activity in growth cone turning induced by various guidance cues (Ming et al., 2001). Whereas ACh is known to be involved in synaptic transmission and synapse formation in gastropods (Haydon, 1988; Elliott and Vehovszky, 2000), a comprehensive study of how cholinergic modulation of neuronal electrical activity may be linked to growth cone motility is presently lacking.

B5 neurons can be removed from the buccal ganglion of the pond snail Helisoma trivolvis and transferred into cell culture, where they regenerate within 1–3 days and develop large-sized growth cones, providing the opportunity to study filopodial dynamics at high resolution. ACh is found to be used in synaptic transmission between B5 neurons in vitro (Haydon and Zoran, 1989), making these identified neurons a model to study the role of ACh as a modulator of neuronal activity and filopodial motility.

The main goal of the current study was to evaluate the effect of ACh on growth cones and to identify the signaling pathway(s) activated by ACh in B5 neurons. We found that ACh decreased the input resistance (Rin), depolarized the membrane potential (RMP), increased the spiking frequency, elevated the intracellular Ca concentration ([Ca]i) in growth cones, and elongated growth cone filopodia of B5 neurons. Extracellular Ca was required for these ACh-induced changes at the growth cone. Activation of nAChRs was both necessary and sufficient in mediating ACh signals. We further found that the ACh-induced filopodial elongation can occur locally at growth cones. Taken together, this study demonstrates a modulatory role for ACh on B5 neurons, resulting in depolarization and filopodial elongation, and thereby suggesting a role for ACh in determining neuronal pathfinding and/or synaptogenesis.

METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

Animals

Freshwater pond snails, Helisoma trivolvis, were kept in aerated aquaria (10 gallons) containing filtered water under a 12 h light-dark cycle at room temperature. They were fed with organic lettuce and vegetable-based algae wafers (Hikari, Doctors Forster and Smith) once a day. Animals with a shell diameter of 15–20 mm were used for neuronal culture.

Neuronal Culture

Identified B5 neurons were isolated from the buccal ganglion of Helisoma, and plated into Falcon Petri dishes as previously described (Rehder and Kater, 1992). Briefly, neurons were plated onto poly-l-lysine (hydrobromide, MW, 70–150 kDa, 0.25 mg mL−1; Sigma, St. Louis, MO)-coated glass coverslips attached to the bottom of 35-mm cell culture dishes (Falcon 1008). B5 neurons were kept in conditioned medium at room temperature and used for experiments 24–48 h after plating. Conditioned medium was prepared by incubating two Helisoma trivolvis brains per 1 mL of Leibowitz L-15 medium (Invitrogen, Carlsbad, CA) for 4 days (Wong et al., 1981). The composition of L-15 medium was as follows (mM): 44.6 NaCl, 1.7 KCl, 1.5 MgCl2, 0.3 MgSO4, 0.14 KH2PO4, 0.4 Na2HPO4, 1.6 Na pyruvate, 4.1 CaCl2, 5 HEPES, 50 μg mL−1 gentamicin, and 0.15 mg mL−1 glutamate in distilled water, pH 7.4.

Growth Cone Image Acquisition and Analysis

Growth cones were viewed using a 100X oil immersion objective on a Sedival microscope (aus Jena, Germany). Phase-contrast images of growth cones were captured by a regular CCD camera (C-72, Dage-MTI, Michigan City, IN) and analyzed with ‘Scion Image' software (Scion Corporation, Frederick, MD). Images for all experimental conditions were taken before (−5 and 0 min) and at defined times (2, 5, 10, 15, 20, 25, 30, 40, 50, and 60 min) after drug treatment. Analysis of filopodial behavior was described previously (Trimm and Rehder, 2004). Briefly, filopodial length was analyzed by measuring the length of all individual filopodia from the tip to the edge of the central domain in one growth cone. Filopodial data were expressed as a percentage change normalized to the time point t = 0, which minimized the individual variability regarding to growth cone size and baseline filopodial length between different growth cones.

Calcium Imaging

Growth cone calcium measurement was performed as previously described (Trimm and Rehder, 2004). Briefly, B5 neurons were injected with the cell-impermeable calcium indicator dye, Fura-2 pentapotassium salt (10 mM in H2O; Molecular Probes, Eugene, OR) and used 30 min after Fura-2 injection. Growth cone calcium imaging was achieved by employing an up-right microscope (BX51 W1F, Olympus, Japan), cooled CCD camera (Andor, TILL Photonics, Germany), and calcium imaging acquisition and analysis software (Live Acquisition, TILL Photonics, Germany). Fura-2 was excited at 340 and 380 nm, and the emission ratio (340/380) was used as an indicator of growth cone [Ca]i. Growth cones were imaged for 5 min before, and up to 60 min after treatment. Image pairs were routinely obtained every 60 s and analyzed by placing a box over the central domain of the growth cone to quantify average fluorescence values. In the experiment shown in Figure 1(D) (representative example of three such experiments), the growth cone was imaged every 10 s and [Ca]i was measured simultaneously in filopodia, the central domain, and the neurite adjacent to the growth cone proper. The [Ca]i in filopodia was measured at filopodial half length for consistency. Growth cones with baseline fluorescence ratios above 0.5 indicated a higher resting level of [Ca]i and were excluded from the analysis.

image

Figure 1. ACh induces a transient increase in filopodial length and [Ca]i. A: Phase-contrast images of a growth cone immediately before (left), 5 min after (center), and 30 min after (right) treatment with 0.5 μM ACh. Helisoma B5 neurons were cultured for 24–48 h in vitro. Note the two white arrows pointing at the tips of two representative filopodia, highlighting their transient elongation and subsequent shortening. Scale bar, 10 μm. B: Bath application of 0.5 μM ACh resulted in a transient increase in filopodial length with the maximal response (an increase by 40.5% ± 2.6%, p < 0.001) occurring 5 min after ACh treatment, whereas growth cones receiving vehicle control did not show changes in filopodial length. Replacement of the culture medium with Ca-free solution prevented the ACh-induced filopodial elongation. C: Bath application of 0.5 μM ACh caused a transient and significant increase in the Fura-2 fluorescence emission ratio in growth cones compared to the vehicle control group (p < 0.001). The elevation of Fura-2 ratio suggested that the [Ca]i in growth cones was elevated after ACh treatment. Replacement of the culture medium with Ca-free solution prevented the ACh-induced increase in [Ca]i. D: [Ca]i in filopodia (n = 4), growth cone proper, and adjacent neurite from a representative growth cone imaged in 10 s intervals in response to 0.5 μM ACh.

Download figure to PowerPoint

Electrophysiology

Recordings from Helisoma B5 neurons in whole-cell current-clamp mode were obtained as described previously (Artinian et al., 2010; Zhong et al., 2012). Patch electrodes were pulled from borosilicate glass tube (OD 1.5 mm; ID 0.86 mm; Sutter instruments) on a Sutter instruments micropipette puller (P-87) and heat polished (Micro Forge MF-830; Narishige) with resistances of about 3–8 MΩ. Recordings were made using an Axopatch 700B amplifier (Molecular Devices) and an analog-to-digital converter (Digidata 1440). Data acquisition and analysis were performed using pClamp software version 10.0 (Molecular Devices). Current-clamp configuration was used to record membrane potential, firing properties, and input resistance (Rin). Normal saline was used as extracellular recording solution, which contained (mM): 51.3 NaCl, 1.7 KCl, 4.1 CaCl2, 1.5 MgCl2, and 5 HEPES, pH 7.3–7.4 (127 mOsm). Intracellular recording solution contained (mM): 54.4 K-aspartate, 2 MgCl2, 5 HEPES, 5 Dextrose, 5 ATP, and 0.1 EGTA (127 mOsm). Drug treatment and washout were achieved through a gravity-based perfusion system (Warner Instruments), switching among channels containing different reagents. Resting membrane potential (RMP) of spontaneous firing neurons was determined by measuring the value at the plateau of the depolarization phase before the membrane potential reached threshold. Continuous measurement of Rin was achieved by small hyperpolarizing current injection of −50 pA for 1 s and repeated every 20 s. Rin was determined by dividing the peak change in membrane potential by the magnitude of the injected current, and was then expressed as the percentage change normalized to Rin measured before treatment to remove individual variability between neurons.

Growth Cone Transection

The isolation of neuronal growth cones was achieved by severing the neurites close to the growth cone proper using a glass micropipette attached to a micromanipulator. Experiments on isolated growth cones were performed 60 min after the transection of neurites. This waiting period proved to be sufficient to restore filopodial motility to normal levels after the transient transection-induced filopodial elongation previously described in isolated growth cones (Rehder et al., 1991).

Pharmacological Agents and Ca-Free Conditions

All agents were purchased from Sigma. Acetylcholine (ACh), dimethylphenylpiperazinium (DMPP), and tubucurarine (TC) were dissolved in water to make 100, 50, and 100 mM stock solutions, respectively. For growth cone filopodia and calcium imaging experiments, stock solutions were mixed with 1 mL of conditioned medium removed from the culture dish and then gently added back around the periphery of the dish. About 1 mL medium was then pulled out and released back into the dish for three times using a pipette to facilitate the equilibration of the drugs to their final concentrations. The Ca-free solution contained (mM): 51.3 NaCl, 5.6 MgCl2, 5 HEPES, and 0.3 EGTA, pH 7.3–7.4 (127 mOsm). Extracellular Ca-free conditions were achieved by removing 1.8 mL of medium from the culture dish, adding back Ca-free solution, and repeating these steps for a total of three rinses in Ca-free solution. For electrophysiological experiments, drugs were prepared directly in the extracellular solution to achieve final concentrations, and perfused into the dish.

Statistical Analysis

All data were expressed as mean ± SEM. For growth cone filopodial analysis and calcium imaging results, a repeated-measures ANOVA was employed for testing overall statistical significance between conditions (SPSS statistical software, SPSS, Chicago, IL). The Tukey test was used for post hoc analysis of preplanned comparisons. An unpaired Student's t test or paired t test was used for testing statistical significance between individual time points depending on the experimental conditions. For electrophysiological data analysis, the significance of effects was evaluated by one-way ANOVA and Tukey's post hoc test using ORIGIN DATA ANALYSIS AND GRAPHING software (OriginLab, Northampton, MA). Significant differences are indicated as *p < 0.05, **p < 0.01, and ***p < 0.001.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

ACh Elongates Neuronal Growth Cone Filopodia and Elevates [Ca]i in Growth Cones

To investigate the effect of ACh on growth cone filopodial dynamics, we used identified B5 neurons extracted from buccal ganglia of the freshwater snail Helisoma trivolvis and investigated their morphology in response to various treatments. Experiments were performed after neurons had been cultured for 24–48 h, at which time these neurons had extended well developed neurites tipped by motile growth cones (Welshhans and Rehder, 2005; Tornieri and Rehder, 2007). ACh had a significant overall effect on filopodial length (F3,53 = 21.67, p < 0.001; repeated-measures ANOVA) [Fig. 1(B)]. Bath application of 0.5 μM ACh led to a transient but significant increase in filopodial length compared with the vehicle-only control condition (Tukey's post hoc, p < 0.001) [Fig. 1(A,B)]. The ACh-induced filopodial elongation started as early as 2 min after treatment, and reached its maximal response (an increase by 40.5% ± 2.6%, n = 15) at 5 min following ACh application. After that, filopodial length slowly decreased in the continued presence of ACh and fully returned to the baseline levels 25 min after treatment.

In previous studies, filopodial elongation had been shown to be elicited by transient increases in the intracellular calcium concentration ([Ca]i) in growth cones (Rehder and Kater, 1992; Van Wagenen and Rehder, 1999; Cheng et al., 2002). Therefore, we next measured [Ca]i in growth cones using the calcium indicator Fura-2. ACh had a significant overall effect on [Ca]i (F3,51 = 76.94, p < 0.001; repeated-measures ANOVA). 0.5 μM ACh treatment caused an immediate and significant elevation in [Ca]i compared to the vehicle-control group (Tukey's post hoc, p < 0.001) [Fig. 1(C)], as indicated by the fluorescence emission ratio at excitation wavelengths of 340 and 380 nm. The Fura-2 ratio was elevated significantly from the baseline level of 0.32 ± 0.01 (n = 17) to a maximal ratio of 1.18 ± 0.05 at 8 min after ACh application (t16 = −18.03, p < 0.001; paired t test), indicating a strong increase in [Ca]i in growth cones. [Ca]i decreased and returned to a plateau level slightly above resting levels by 35 min. To investigate the location and time course of the ACh-induced elevation in [Ca]i with increased time resolution, we next acquired fura-2 images every 10 s, instead of every 60 s. The ACh-induced elevation in [Ca]i occurred throughout the entire growth cone and [Ca]i increased simultaneously in filopodia, the central domain, and the adjacent neurite [Fig. 1(D)]. Taken together, ACh induced an elevation in growth cone [Ca]i and an increase in filopodial length in B5 neurons with a similar time course.

We showed previously that Ca-dependent filopodial elongation can be triggered either by Ca influx or by release of Ca from intracellular stores (Rehder and Kater, 1992; Welshhans and Rehder, 2007). Since studies of Xenopus spinal neurons identified the importance of extracellular Ca in ACh-induced growth cone turning behaviors, we next investigated whether extracellular Ca was also required for the ACh-induced filopodial elongation. The culture medium was first replaced with Ca-free solution, and B5 neurons were then stimulated with ACh. Replacement with Ca-free solution caused a slight and long-term reduction in filopodial length by 10% starting at five minutes (t24 = −4.24, p < 0.001; two sample t test; data not shown), compared to control. Interestingly, the effect of 0.5 μM ACh on filopodial elongation was fully blocked in Ca-free solution compared to ACh by itself (Tukey's post hoc, p < 0.001) [Fig. 1(B)]. Correspondingly, Ca imaging studies on growth cones revealed that the increase in [Ca]i by 0.5 μM ACh was eliminated in Ca-free solution as well (Tukey's post hoc, p < 0.001) [Fig. 1(C)]. Taken together, these data suggested that ACh caused an increase in [Ca]i via Ca influx, which, in turn, resulted in filopodial elongation.

ACh Depolarizes Membrane Potential and Reduces Input Resistance in a Dose-Dependent Manner

ACh has been shown to modulate neuronal activity in various systems, including neurons in the pond snails Helisoma trivolvis and Lymnaea stagnalis (Bahls, 1987; Perry et al., 1998). Given that we had measured a significant Ca influx in response to ACh, we next tested whether ACh treatment might have caused the increase in [Ca]i and the subsequent elongation of filopodia by regulating the electrical activity of B5 neurons. We used the patch clamp recording technique in the whole-cell current clamp configuration to investigate the role of ACh on neuronal electrical activity, and injected a small negative current (−50 pA) for 1 s every 20 s to monitor input resistance (Rin) continuously. B5 neurons fire spontaneous action potentials (APs) with a resting membrane potential (RMP) close to −40 mV [Fig. 2(A)] (Artinian et al., 2010). To test the effects of ACh on the electrical properties of B5 neurons, we stimulated neurons with various concentrations of ACh ranging from 10 nM to 100 μM using a perfusion system. ACh began to depolarize the RMP at a concentration of 100 nM and its effect saturated at 10 μM with a depolarizing response of + 23.0 ± 0.6 mV (n = 6). Meanwhile, Rin started to decrease at 30 nM ACh and this response also saturated at 10 μM ACh, at which Rin was reduced to 10.4% ± 1.3% (n = 6). Although we observed a gradual increase in the firing frequency at lower concentrations, ranging from 10 nM to 1 μM, higher concentrations of ACh (>3 μM) depolarized the membrane potential to a level that resulted in neuronal silencing [Fig. 2(A)]. ACh caused a concentration-dependent depolarization of RMP with an estimated half maximal effective concentration (EC50) of 1.7 μM [Fig. 2(B)] and a reduction in Rin with an EC50 of 0.4 μM [Fig. 2(C)], based on the Hill equation. This result suggested that B5 neurons are tuned to dynamically respond to small changes in the concentration of ACh, but that they would become less responsive to synaptic inputs in the continued presence of relatively higher concentrations of ACh. Because the firing frequency could not be studied at higher concentrations of ACh, we instead quantified RMP and Rin in all following electrophysiology experiments.

image

Figure 2. ACh depolarizes the membrane potential and reduces Rin in a dose-dependent manner. A: Representative recordings of changes of the membrane potential and Rin in response to various concentration of ACh treatment (10 nM to 10 μM). Negative current injection (−50 pA) was applied for 1 s of each 20 s recording trial. Note: Whereas lower concentrations of ACh (<1 μM) resulted in a progressive increase in spiking frequency, higher concentration (10 μM) caused neuronal silencing at a depolarized RMP. B: Dose-dependent curve showing that ACh caused a depolarization of RMP with an estimated EC50 of 1.7 μM. C: Dose-dependent curve suggesting that ACh progressively reduced Rin (changes of the membrane potential in response to the negative current injection; −50 pA, 1 s) with an estimated EC50 of 0.4 μM.

Download figure to PowerPoint

nAChR Agonist DMPP Elongates Filopodia and Elevates Growth Cone [Ca]i

To further investigate the mechanism by which ACh induced filopodial elongation, we considered the possibility that ACh may signal through nAChRs, which are widely expressed in molluscan nervous systems and whose activation mediates a significant portion of ACh-associated effects (Bahls, 1987; Perry et al., 1998; Elliott and Vehovszky, 2000). DMPP is a prominent agonist of nAChRs with little selectivity between neuronal nAChR subtypes. Bath application of 5 μM DMPP caused a significant increase in filopodial length when compared with the vehicle control (F1,25 = 33.17, p < 0.001; repeated-measures ANOVA) [Fig. 3(A)]. The maximal response of DMPP on filopodial length occurred 10 min after drug treatment (an increase by 33.8% ± 3.5%, n = 13). Interestingly, the peak response was not significantly different between the groups treated with ACh (0.5 μM) and DMPP (5 μM) (t26 = −1.59, p = 0.12; unpaired t test), suggesting that the effect of ACh on filopodial elongation was indeed mediated through activation of nAChRs.

image

Figure 3. nAChR agonist DMPP elongates filopodia and elevates [Ca]i. A: Bath application of 5 μM DMPP led to a significant increase in filopodial length followed by a gradual reduction toward baseline (p < 0.001 as compared to vehicle control), and the maximal response (an increase by 33.8% ± 3.5%) occurred 10 min after DMPP treatment. B: Bath application of 5 μM DMPP resulted in a significant increase in Fura-2 ratio (p < 0.001 as compared to vehicle control), indicating a pronounced increase in [Ca]i, which also peaked at around 10 min.

Download figure to PowerPoint

In addition to its effect on filopodial elongation, DMPP also affected [Ca]i. Bath application of 5 μM DMPP induced a quick and significant increase in the Fura-2 ratio (F1,21 = 73.61, p < 0.001; repeated-measures ANOVA) as compared to vehicle control, where no changes in the ratio were observed (n = 12) [Fig. 3(B)]. The peak ratio induced by 5 μM DMPP appeared 7 min after treatment (1.10 ± 0.10, n = 11). Taken together, both direct ACh application and treatment with the nAChR agonist DMPP produced similar effects on filopodia and [Ca]i, supporting the hypothesis that ACh acted on nAChRs to regulate growth cone filopodial dynamics.

DMPP Mimics the Effect of ACh on Electrical Activity

We next tested whether the activation of nAChRs was sufficient to explain the effects of ACh on the electrical activity seen above. 5 μM DMPP caused a depolarization of RMP by + 4.7 ± 0.4 mV (n = 11) and a reduction in Rin to 46.4% ± 3.8% (n = 12) [Fig. 4(A)]. To study the concentration dependency of DMPP on electrical properties, we next tested one lower (1 μM) and two higher concentrations (10 and 50 μM), respectively. 1 μM DMPP induced a much smaller depolarization (+ 1.5 ± 0.2 mV, n = 10) and a reduction in Rin (74.6% ± 5.1%, n = 9) compared to the 5 μM DMPP group, whereas 10 and 50 μM DMPP had stronger effects on both RMP (10 μM: a depolarization by + 6.5 ± 0.7 mV, n = 10; 50 μM: + 13.2 ± 1.6 mV, n = 9) and Rin (10 μM: a reduction to 37.0% ± 5.4%, n = 9; 50 μM: 24.9% ± 2.4%, n = 6) [Fig. 4(B,C)]. In three out of nine cases, 50 μM DMPP caused silencing of B5 neurons [data not shown], a phenomenon that was similar to what we observed with higher concentrations of ACh (>3 μM). Taken together, activation of nAChRs by DMPP mimicked the effects of ACh on electrical activity of B5 neurons, namely depolarization of RMP and reduction of Rin in a dose-dependent manner.

image

Figure 4. DMPP treatment results in a depolarization of RMP and a reduction in Rin in a dose-dependent manner. A: Examples of a firing B5 neuron before and after treatment with 5 μM DMPP. Note that 5 μM DMPP induced a small depolarization of RMP and a decrease in Rin. B: DMPP induced a depolarization of RMP in a dose-dependent manner. C: DMPP caused a dose-dependent reduction in Rin.

Download figure to PowerPoint

Inhibition of nAChRs by TC Blocks DMPP-Induced Filopodial Elongation and Increase in [Ca]i

If ACh indeed activated nAChRs to regulate growth cone motility, the blockade of nAChRs should inhibit the ACh-induced filopodial elongation. We tested this hypothesis by using the classical nAChR antagonist, tubocurarine (TC). TC is known to antagonize functional responses mediated by nAChRs in various organisms (Haydon and Zoran, 1989; Zheng et al., 1994). To activate nAChRs selectively and to avoid the potential activation of other types of AChRs, the specific nAChR agonist DMPP was used in this set of experiments. Following a 10-min pretreatment with 100 μM TC, 5 μM DMPP was bath applied into the dish. Whereas TC on its own did not affect filopodial dynamics (F1,25 = 2.83, p = 0.107; repeated-measured ANOVA; as compared to vehicle control, data not shown), pretreatment with TC eliminated the DMPP-induced increase in filopodial length (Tukey's post hoc, p < 0.001, 100 μM TC + 5 μM DMPP compared to 5 μM DMPP alone; overall effect: F2,37 = 33.85, p < 0.001; repeated-measures ANOVA) [Fig. 5(A,B)]. The maximal filopodial elongation normally observed at 10 min following 5 μM DMPP treatment was fully blocked in the group pretreated with 100 μM TC (100 μM TC + 5 μM DMPP: increase by 0.4% ± 1.3%, n = 14 vs. 5 μM DMPP: 33.8% ± 3.5%, n = 13; t25 = −0.28, p < 0.001; unpaired t test) [Fig. 5(B)]. Instead, the TC + DMPP group maintained filopodial length close to baseline levels throughout the post-treatment period. Furthermore, preincubation with 100 μM TC also fully inhibited the DMPP-induced elevation of [Ca]i (Tukey's post hoc, p < 0.001, 100 μM TC + 5 μM DMPP compared to 5 μM DMPP alone; F2,33 = 81.80, p < 0.001; repeated-measures ANOVA), whereas TC by itself did not have an effect on [Ca]i at growth cones (F1,24 = 0.08, p = 0.779; repeated-measured ANOVA; as compared to vehicle control, data not shown) [Fig. 5(C)]. These data strongly suggested that the activation of nAChRs was necessary for ACh to elicit its effect on growth cone filopodia, and that the increase in [Ca]i played a key role in filopodial elongation induced by nAChR activation.

image

Figure 5. Inhibition of nAChRs by TC blocks the DMPP-induced filopodial elongation and increase in [Ca]i. A: Phase-contrast images of a growth cone immediately before (left), 10 min after 100 μM TC (center left), 5 min after addition of 5 μM DMPP in the presence of TC (center right), and 30 min after addition of 5 μM DMPP in the presence of TC (right). Note that these treatments did not result in obvious changes of filopodial length. Scale bar, 10 μm. B: The DMPP-induced filopodial elongation was significantly blocked by the treatment with 100 μM TC (p < 0.001 as compared to 5 μM DMPP alone). C: Treatment with 100 μM TC significantly inhibited the DMPP-induced increase in [Ca]i in growth cones (p < 0.001 as compared to 5 μM DMPP alone).

Download figure to PowerPoint

nAChR Antagonist TC Blocks DMPP-Induced Depolarization and Decreases in Rin

We next investigated whether the electrical responses induced by DMPP could also be blocked by pretreatment with TC. 100 μM TC on its own did not have an effect on either membrane potential (a depolarization by + 0.2 ± 0.4 mV, n = 9) or Rin (a reduction to 107.0% ± 4.0%, n = 9) [Fig. 6(A)]. After perfusion with 100 μM TC, B5 neurons were treated with a solution containing 100 μM TC and 5 μM DMPP. As shown in Figure 6(B), 5 μM DMPP failed to depolarize the RMP of B5 neurons in the presence of 100 μM TC (100 μM TC + 5 μM DMPP: + 0.4 ± 0.2 mV, n = 15 vs. 5 μM DMPP, + 4.7 ± 0.4 mV, n = 11; p < 0.001; Tukey's post hoc). Moreover, the DMPP-induced decrease in Rin was completely blocked in the presence of 100 μM TC (100 μM TC + 5 μM DMPP: 108.6% ± 3.8%, n = 15 vs. 5 μM DMPP, 46.4% ± 3.8%, n = 12; p < 0.001; Tukey's post hoc) [Fig. 6(C)]. Hence, the effects of DMPP further indicated that the modulation of the electrical activity of B5 neurons was mediated via activation of nAChRs.

image

Figure 6. nAChR antagonist TC blocks the DMPP-induced depolarization and decrease in Rin. A: Example of a firing B5 neuron before (control, left), after pretreatment with TC (100 μM, center), after subsequent addition of DMPP (5 μM, right). Note that no obvious changes in membrane potential and hyperpolarizing responses to negative current injection (−50 pA, 1 s) were seen. B: Quantification of changes in the membrane potential from experiments such as shown in A. The depolarizing response induced by 5 μM DMPP was significantly inhibited when B5 neurons were pretreated with 100 μM TC (p < 0.001 as compared to 5 μM DMPP), whereas 100 μM TC on its own had no effect on RMP. C: Quantification of normalized Rin from experiments such as shown in A. The effect of DMPP (5 μM) on Rin was fully blocked when B5 neurons were pretreated with 100 μM TC (p < 0.001 as compared to 5 μM DMPP), whereas TC by itself had no effect on Rin.

Download figure to PowerPoint

ACh Acts Locally at the Growth Cone to Elongate Filopodia

Growth cones have been demonstrated to possess some degree of autonomous function, and contain most of the machinery required for proper responses to extrinsic stimulation (Kater et al., 1994; Gomez and Zheng, 2006). Therefore, we next tested whether ACh acted at the growth cone proper, or, alternatively, whether the ACh-induced changes in filopodial dynamics were the result of ACh acting on another region of the neuron, such as the cell body. To explore this possibility, we physically isolated growth cones from the remaining neuron by transection of the adjacent neurites using a microknife [Fig. 7(A)]. Such isolated growth cones have been shown to survive for up to 24 h and maintain many features seen in growth cones that are connected to their parent neuron (Rehder et al., 1991). In this series of experiments, 0.5 μM ACh was bath applied to dishes containing isolated growth cones. Filopodia responded to the ACh treatment with filopodial elongation within 2 min, and reached their maximal length (an increase by 34.9% ± 3.5%, n = 14) 5 min after drug treatment, followed by a gradual decrease in filopodial length towards baseline [Fig. 7(A,B)]. The effect of ACh on filopodial length was significant when compared to the control group, in which only solvent was applied (F1,25 = 45.57, p < 0.001; repeated-measures ANOVA) [Fig. 7(B)]. This result supported the hypothesis that ACh acted locally at the growth cone to regulate filopodial dynamics.

image

Figure 7. ACh causes a transient increase in filopodial length on physically isolated growth cones. A: Phase-contrast images showing a cultured B5 neuron and a physically isolated growth cone generated by neurite transection (left, the isolated growth cone highlighted in dashed box and the transection site marked by a white arrowhead), and the same growth cone magnified immediately before (Pre), 10 min after (center), 40 min after treatment with 0.5 μM ACh (right). Note that filopodial elongation was clearly visible at 10 min after drug treatment, as indicated by the white arrows. Scale bar, 10 μm. B: Bath application of 0.5 μM ACh induced a transient and significant increase in filopodial length on isolated growth cones (p < 0.001 as compared to vehicle control), and the maximal response (length increase by 34.9% ± 3.5%) occurred 5 min after the treatment with ACh.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES

ACh has been shown to affect numerous functions in developing neurons in vitro, including neurite outgrowth, growth cone guidance, synapse formation and synaptic transmission (Zheng et al., 1994; Lauder and Schambra, 1999; Woodin et al., 2002; Rudiger and Bolz, 2008). Here, we describe a role for cholinergic modulation at the growth cone, where ACh affects an important component of growth cone motility, namely filopodial dynamics. We report that ACh induced a transient but significant filopodial elongation in B5 neurons of the buccal ganglion of Helisoma trivolvis. The signaling cascade resulting in filopodial elongation is via opening of nAChRs, a depolarization of the membrane potential, and a subsequent elevation of [Ca]i in growth cones. Moreover, by performing experiments on physically isolated growth cones, we demonstrated that ACh can act as a local signal at the growth cone to elongate filopodia.

ACh and nAChRs

nAChRs appeared to be both necessary and sufficient for ACh to elicit its effects on growth cone filopodial dynamics, intracellular calcium, and neuronal electrical properties. Treatment with DMPP, a nonselective nAChR agonist, fully mimicked the effects of ACh on membrane depolarization, reduction in Rin, elevation of [Ca]i, and filopodial elongation. Meanwhile, TC, a well-known antagonist of nAChRs, completely inhibited all DMPP-induced responses. These results not only supported the specificity of our pharmacological approach, but also provided convincing evidence that ACh acted through activation of nAChRs to produce these effects. Taken together, our results add a novel role to the critical functions of nAChRs in cholinergic modulation of cellular properties in both vertebrates and invertebrates (Fu et al., 1998; Clementi et al., 2000; Woodin et al., 2002; Cobb and Davies, 2005).

nAChRs belong to a large Cys-loop family of ligand-gated ion channels including the 5-HT3, glycine, and GABAA receptors (Le Novere and Changeux, 1995). nAChRs exist as pentameric complexes assembled either from five copies of a single subunit or by a composition of five different subunits (Clementi et al., 2000). Although the molecular identities of nAChRs are well studied in mammalian systems, much less information is available in gastropods. Van Nierop and Smit identified nAChR subunits in the pond snail Lymnaea stagnalis, a closely related species (van Nierop et al., 2005; van Nierop et al., 2006), and cloned a total of twelve subunits of nAChR. Considering that ACh treatment in this study caused the depolarization of RMP and an elevation in [Ca]i in Helisoma B5 neurons, it is likely that the effect of ACh was contributed by the classic, cation-selective nAChRs. A characterization of Helisoma nAChRs at the molecular level is beyond the scope of this study but will extend our understanding of cholinergic modulation in the Helisoma nervous system in the future. Because the effects of ACh on growth cones and electrophysiological properties could be fully mimicked by the nAChR agonist, DMPP, we did not pursue any potential involvement of muscarinic ACh receptors in regulating growth cone filopodial length and [Ca]i.

ACh and Cell Excitability

Cholinergic modulation of cell excitability has been studied in some detail in gastropods. Salivary gland cells in Helisoma respond to ACh by a long-lasting depolarization followed by hyperpolarization (Bahls, 1987). Moreover, ACh is thought to be the core neurotransmitter to control neuronal activity in the feeding central pattern generator in Lymnaea, such as protraction phase premotor interneurons N1L and N1M (Elliott and Vehovszky, 2000). Cholinergic projections to the Lymnaea proesophagus modulate foregut contractile activity (Perry et al., 1998). Furthermore, ACh acts on an ionotropic receptor sensitive to nicotinic antagonists to evoke an afterdischarge in Aplysia bag cell neurons (White and Magoski, 2012).

We demonstrated that ACh induced rapid and significant changes in electrical activity in B5 neurons. Both treatment with ACh and the nAChR agonist, DMPP, caused a reduction in Rin in a dose-dependent manner, suggesting the opening of nAChRs. The activation of nAChRs, in turn, induced the depolarization of the membrane potential. Relatively lower doses of ACh or DMPP resulted in an increase in spiking frequency, whereas higher concentrations caused cell silencing at a depolarized membrane potential.

ACh, Ca, and Growth Cone Motility

We report that both stimulation with ACh and activation of nAChRs by DMPP resulted in a quick and pronounced increase in [Ca]i in growth cones. Interestingly, the elevation of [Ca]i and the increase in filopodial length occurred with a very similar time course. Considering the critical role of spiking activity in regulating [Ca]i within neurons (Spitzer, 2006), and that causality has been demonstrated between an elevation in [Ca]i and filopodial elongation in B5 neurons (Rehder and Kater, 1992), the present study is consistent with the hypothesis that ACh functions through an increase in [Ca]i to elongate growth cone filopodia. After the replacement of culture medium with a Ca free solution, both the filopodial elongation and the elevation of [Ca]i induced by ACh were blocked, suggesting that Ca influx plays a critical role in ACh-signaling at the growth cone. In fact, other neurotransmitters and neuromodulators have been demonstrated to signal through [Ca]i to regulate growth cone functions (Henley and Poo, 2004; Gomez and Zheng, 2006). For example, the gaseous messenger nitric oxide affects growth cone filopodial dynamics in Helisoma B5 neurons via a Ca-dependent mechanism (Van Wagenen and Rehder, 2001; Trimm and Rehder, 2004; Welshhans and Rehder, 2005; Welshhans and Rehder, 2007). An increase in [Ca]i is able to reduce neurite outgrowth rate in Helisoma neurons (Cohan, 1992). In addition, [Ca]i is required for glutamate, netrin-1, and myelin-associated glycoprotein to guide growth cone turning in cultured Xenopus spinal neurons (Zheng et al., 1996; Ming et al., 2001). Upon the elevation of [Ca]i at the growth cone, multiple Ca-mediated signaling pathways could be activated to translate external signals into cytoskeletal changes. Our lab previously showed that calmodulin and the Ca-dependent phosphatase calcineurin are acting downstream of Ca to elongate filopodia (Cheng et al., 2002). Moreover, calmodulin-dependent protein kinase II is found to mediate ACh-induced chemoattraction in growth cone guidance (Zheng et al., 1994).

ACh has been shown to regulate cell movements, cell proliferation, and neuronal differentiation in various developing central nervous systems (Lauder and Schambra, 1999). Although the roles played by ACh in neuronal development in vivo are yet unclear, evidence for a functional role of ACh in developing neurons came from the studies of cultured embryonic Xenopus spinal neurons (Zheng et al., 1994). ACh was found to act as a chemoattractive guidance cue that elicited growth cone turning behavior towards the source of ACh release. Here, we extended this research by studying the effect of ACh on growth cone filopodia. Filopodia on growth cones are essential for growth cone guidance, and filopodial elongation increases the area that a growth cone can sample during pathfinding (Kater and Rehder, 1995; Rehder et al., 1996). A transient elongation of filopodia, as seen in this study in response to stimulation with ACh, could play a critical role in decision-making at the growth cone. Longer filopodia would encounter cues located 10–20 μm ahead of the advancing growth cone proper, could result in a change in the direction of growth towards or away from the cue depending on the signal content, and ultimately, upon contact of an appropriate cellular target, could transform a growth cone into a presynaptic structure. In our experiments, ACh was bath-applied, mimicking a general, extrasynaptic stimulation of B5 neurons. To determine the location at which ACh acted to elicit filopodial elongation, we physically isolated growth cones and demonstrated that they responded to ACh treatment in a similar fashion as intact growth cones did, suggesting that ACh can regulate growth cone motility at the growth cone proper. We did not attempt to produce gradients of ACh across growth cones to determine if this would result in asymmetrical filopodial elongation, as might be expected to precede growth cone turning. In fact, filopodial asymmetry on growth cones of Xenopus neurons was found to precede growth cone turning in responses to glutamate gradients (Zheng et al., 1996). Additionally, our lab reported that both transient changes of growth cone filopodial dynamics (Van Wagenen and Rehder, 1999) and decreases in nerve growth speed (Trimm and Rehder, 2004) can be triggered by nitric oxide, a phenomenon we described as “slow-down and search” behavior. Here, the ACh-induced change in filopodial dynamics may serve as a first response of an extending neurite to ACh encountered during pathfinding.

Although the sources of ACh release and physiological concentrations reached in the Helisoma buccal ganglia are unknown, several studies in which ACh release was measured suggest that the concentrations used in our in vitro study are comparable to in vivo conditions. The ACh concentration detected in the vicinity of magnocellular basal forebrain neurons using the nAChR-rich patches prepared from rat myotubes as focal ACh sensors was between 480 nM to > 50 μM (Allen and Brown, 1996). This concentration range of ACh matched the concentrations used in the current study and resulted in significant electrophysiological and morphological responses in B5 neurons. B5 neurons are responsive to ACh and are themselves cholinergic (Haydon and Zoran, 1989). In vitro, the site of ACh release in Helisoma neuron B5 is thought to be mainly confined to the distal neurites, but rarely detected from the soma. ACh release from neurites and growth cones has been reported in different neurons (Hume et al., 1983; Allen and Brown, 1996; Zakharenko et al., 1999; Yao et al., 2000), suggesting that growing cholinergic axons might influence other extending axons expressing nAChRs to regulate their developmental status. Moreover, ACh has been reported to act in an autocrine fashion on presynaptic terminal of Xenopus motoneurons (Fu et al., 1998), suggesting the possibility that ACh release from axonal terminals might directly affect growing axons via an autoreceptive feedback mechanism.

In conclusion, our results provide novel insights into the effects of ACh on developing neurons. ACh affected growth cone filopodial dynamics through the activation of TC-sensitive nAChRs by depolarizing membrane potential and increasing [Ca]i. While such effect could be caused by presynaptic stimulation, supporting the well-known literature of effects of electrical activity on neurite outgrowth and growth cone motility, the finding that ACh can act locally at a growth cone suggested that ACh might also act at extrasynaptic receptors to modulate growth cone motility, neuronal pathfinding, and possibly synaptogenesis directly at the level of the growth cone.

Acknowledgments

The authors thank Dr. Chun Jiang for expert advice on the electrophysiological experiments and their analysis.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. METHODS
  5. RESULTS
  6. DISCUSSION
  7. REFERENCES