Nitric Oxide Potentiates cAMP-Gated Cation Current by Intracellular Acidification in Feeding Neurons of Pleurobranchaea

Abstract

A pH-sensitive cAMP-gated cation current (I_Na,cAMP) is widely distributed in neurons of the feeding motor networks of gastropods. In the sea slug Pleurobranchaea this current is potentiated by nitric oxide (NO), which itself is produced by many feeding neurons. The action of NO is not dependent on either cGMP or cAMP signaling pathways. However, we found that NO potentiation of I_Na,cAMP in the serotonergic metacerebral cells could be blocked by intracellular injection of MOPS buffer (pH 7.2). In neurons injected with the pH indicator BCECF, NO induced rapid intracellular acidification to several tenths of a pH unit. Intracellular pH has not previously been identified as a specific target of NO, but in this system NO modulation of I_Na,cAMP via pH_i may be an important regulator of the excitability of the feeding motor network.

INTRODUCTION

Many gastropod neurons express a cyclic adenosine 3′,5′-monophosphate (cAMP)–induced sodium current (I_Na,cAMP; Connor and Hockberger 1984a; Green and Gillette 1983) activated by direct cAMP gating (Sudlow et al. 1993). The current takes two forms, types 1 and 2, differentiable on the basis of their voltage dependence (Gillette and Green 1987; Huang and Gillette 1993). Observations that cell signaling pathways could target intracellular pH (pH_i) in molluscan neurons (Connor and Hockberger 1984b) was followed by finding that the type 1 I_Na,cAMP, most often found in neurons of the feeding motor network of Pleurobranchaea, was highly sensitive to small changes in pH_i, where current amplitude was increased with acidification (Green and Gillette 1988; also see Aldenhoff et al. 1983). The type 2 I_Na,cAMP form, found in pedal ganglion neurons, was entirely insensitive to pH_i (Huang and Gillette 1991, 1993).

The functional benefit of pH sensitivity of I_Na,cAMP has been a question without an obvious answer since its discovery. However, recently the pH-sensitive I_Na,cAMP was shown to be highly sensitive to potentiation by nitric oxide (NO) activity, whereas the pH-insensitive form was not (Hatcher 2003; Hatcher et al. 2006). Thus NO excited many neurons of the feeding motor network expressing both I_Na,cAMP and apparent resting levels of cAMP, but lacked detectable actions on the pedal locomotor neurons.

NO potentiation of I_Na,cAMP was not dependent on cyclic guanosine 3′,5′-monophosphate (cGMP) because it was not mimicked or blocked by cGMP analog or by inhibitors of cGMP synthesis or protein kinase G (Hatcher et al. 2006). Effects were specific to NO because they were induced by several different donors and abolished by depletion of donor through exposure to air at room temperature. NO synthase activity is considerable in the feeding network of Pleurobranchaea (Cruz et al. 1997; Floyd et al. 1998; Moroz et al. 1996).

We asked whether NO might act through altering intracellular pH. To study potential effects of NO on pH_i, we used the same well-studied, identified neuron previously used to characterize NO-induced potentiation of I_Na,cAMP (Hatcher et al. 2006), the serotonergic metacerebral giant neuron (MCG; Gillette and Davis 1977; McCrohan and Gillette 1988). This neuron is conserved in cerebral ganglion lobes of many gastropods and provides neuromodulatory excitation to the feeding motor network (Gillette and Davis 1977).

Clamping pH_i blocked the potentiating effects of NO on I_Na,cAMP. Using indicator dyes, we observed that NO induced rapid intracellular acidification. These results not only clarify a mechanism of the excitatory effects of NO on the feeding motor network of Pleurobranchaea, but also document a novel mechanism for NO action in biological systems.

METHODS

Specimens of Pleurobranchaea californica were obtained from Sea Life Supply (Sand City, CA) and maintained in Instant Ocean at 12–13°C. The cell bodies of the identified metacerebral giant neurons (MCGs) were visually identified by their large size and anterior-most position in cerebropleural ganglia, isolated by hand dissection, and stabilized in a perfusion dish against carbon fiber pins. For most experiments, the cell was perfused with 3-(N-morpholino)propanesulfonic acid (MOPS)–buffered saline (composition in mM: NaCl 420, KCl 10, MgSO₄ 25, MgCl₂ 25, CaCl₂ 10, MOPS 5, at pH 7.5) at a rate of 1 ml · min⁻¹ at 13 ± 1°C. Bicarbonate saline was made 30 mM in NaHCO₃, substituting equimolar for NaCl, and adjusted to pH 7.5.

I_Na,cAMP responses to iontophoretic injection of cAMP were measured under two-electrode voltage clamp as described previously (Hatcher et al. 2006; Sudlow and Gillette 1997). Briefly, sharp electrodes were pulled from borosilicate glass (WPI). A single-barreled capillary filled with 3 M KCl served as a voltage-sensing electrode. A second, double-barreled, electrode acted as the current-passing electrode (filled with 3 M KCl) in one half and a cAMP iontophoresis capillary in the other (filled with 200 mM cAMP and 20 mM Tris, adjusted to pH 7.5 with KOH). Iontophoretic injection of cAMP (5 s duration; 5–200 nA) was accomplished with negative current from a constant-current source (Model 260, WPI) that allowed precise, repeatable quantities of cAMP to be injected intracellularly.

Injection amplitudes were kept relatively low to avoid current saturation while yielding a robust I_Na,cAMP response. Under our conditions, cAMP injection occurs with a constant transport number of 0.1 over the wide range of injection currents used here (Sudlow and Gillette 1997). The kinetics of I_Na,cAMP is largely determined by intracellular cAMP diffusion and phosphodiesterase activity (Huang and Gillette 1991). I_Na,cAMP measures in the presence of diethylamine/nitric oxide (DEA/NO) were begun around 10 min after addition of NO donor, at which time responses were nearly maximal and relatively constant in the absence of blocking treatment.

Earlier work in neurons of the mollusk Archidoris showed that pressure-injected cAMP to values of 0.1–1 mM could acidify pH_i in the range of 0.1 pH unit, an effect likely mediated by cAMP-dependent kinase (Connor and Hockberger; 1984b). Such an effect could add an interesting dimension to I_Na,cAMP functioning, but was not visible in these present studies. A previous study by Hatcher et al. (2006) did not find effects of NO on pH-sensitive I_Na,cAMP to be dependent on cAMP beyond a simple gating role for the current, perhaps because iontophoretic injections were smaller (Sudlow and Gillette 1997) and were not previously found to cause significant changes in pH_i themselves (Green and Gillette 1988).

Intracellular recording and voltage clamping were performed with an Axoclamp 2B amplifier (Axon Instruments, Union City, CA) and data were recorded digitally with Biopac hardware and the accompanying Acknowledge software package (Biopac Systems, Goleta, CA). All recordings were performed at a holding potential of −50 mV unless otherwise indicated.

For fluorescence measurements, glass microelectrodes were pulled to resistances in the range 5–15 MΩ when filled with 200 mM KCl, 10 mM MOPS, and the fluorescent pH indicator BCECF [2′,7′-bis(2-carboxyethyl)-5,6-carboxyfluorescein; Sigma], pressure injected from an electrode containing 1 mM dye. Following injection, the electrode was withdrawn and the cell was penetrated with an electrode filled with 3 M KCl for electrical recording via an Axoclamp 2A amplifier and Clampex 8.1 software (Axon Instruments).

Fluorescence measurements were made using an InCyt dual-wavelength fluorescence imaging and photometry system (sampling rate 10 per min; Intracellular Imaging) with a Sutter Lambda 10C optical filter changer (Sutter Instrument) and InCyt Im2 acquisition software. For BCECF measurements, a calibration curve for the 488:436 nm fluorescence ratio was generated using a series of MOPS-saline standards, pH 6.5 to 8. These standards proved reliably stable over time. A slow negative (i.e., acidic) drift occurred in 10 pH recordings, averaging 0.007 ± 0.002 pH units · min⁻¹ (mean ± SD), suggesting that slow dye bleaching or leakage was leading to a greater signal contribution by autofluorescence. This artifact did not interfere with observation of the more rapid dynamic changes caused by NO. Resting membrane potentials in these experiments were −45 to −55 mV.

MOPS was pressure injected as a 1 M solution, adjusted to pH 7.2 with KOH. Injection volumes were estimated at ≥10% of cell volume; thus injections are likely to have raised cell buffering power to fourfold or more that of the resting values (cf. Table 2 in Ahmed and Connor 1980). Injection sequences were composed of multiple small injections where pressures were maintained until a small polarizing membrane response indicated injection success. Subsequent injections were made approximately 1 min after membrane potential recovered. Dye injections were performed similarly. The NO donor diethylamine/nitric oxide (DEA/NO, 1–2 mM; Axxora LLC) was applied in the perfused saline. When measured using a calibrated NO electrode (WPI), freshly prepared 1 mM DEA/NO in Pleurobranchaea saline at this temperature yielded a value of 430 nM of NO, with first-order exponential decay of t_1/2 = 418 s (Hatcher et al. 2006). Usually 1–2 half-lives were passed by the time perfusion occurred, so actual NO concentrations may have been in the range of 100–200 nM.

RESULTS

NO donors potentiate the I_Na,cAMP response to iontophoretically injected cAMP in neurons of the feeding motor network (Hatcher et al. 2006), including the MCG neurons (Fig. 1A). In a total of 16 MCG neurons the NO donor DEA/NO enhanced the I_Na,cAMP response on average by 85% (24% SE) over control, predonor measures (P < 0.001; two-tailed t-test). Intracellular acidification of a few tenths of pH unit was previously shown to have a similar effect (Green and Gillette 1988), which is recapitulated in the records of Fig. 1B. In the present study, bicarbonate-containing salines increased I_Na,cAMP amplitude in four neurons by 71.7% (6.3% SE). Thus intracellular acidification mimics the enhancing effect of NO donors on I_Na,cAMP.

Fig. 1.

Potentiation of the cyclic adenosine 3′,5′-monophosphate (cAMP)–induced sodium current (I_Na,cAMP) response amplitude to injected cAMP by nitric oxide (NO) donor and intracellular acidification, and NO-induced depolarization. A: diethylamine/NO (DEA/NO, 1 mM) augmented I_Na,cAMP response amplitude by 78% relative to control measures in saline alone taken just prior to addition of the NO donor. B: saline containing 30 mM bicarbonate mimicked the effect of NO donor, increasing I_Na,cAMP amplitude by 41%. Such bicarbonate salines acidify Pleurobranchaea neurons by several tenths of a pH unit (Green and Gillette 1988). Holding potentials were −50 mV. Horizontal bars in records indicate 5 s, 100 nA iontophoretic injections of cAMP.

A likely causal linkage of NO and pH_i was supported by finding in each of four cells tested that intracellular injection of 1 M MOPS buffer at pH 7.2, in volumes estimated near 0.1× soma volume, eliminated the potentiating effect of NO on I_Na,cAMP amplitude [Fig. 2; average change 10% (5% SE); P = 0.30, two-tailed t-test]. Similar intracellular injections of BCECF (see following text) did not block NO-induced depolarization or NO effects on pH_i. Thus a nonspecific effect of simple pressure injection was unlikely to be a significant factor in the pH_i response. The blockade of NO effects by intracellular pH buffering was significant relative to the positive effects of NO donor on the 16 uninjected MCGs (P = 0.0003; Fisher's exact test, two-sided) and led to the following assessment of the effects of NO on pH_i.

Fig. 2.

Intracellular injection of the pH buffer MOPS suppresses NO potentiation of I_Na,cAMP. A: I_Na,cAMP responses to injected cAMP in normal saline (left), after MOPS injection and in presence of NO donor. Holding potentials were −50 mV. B: a typical dose–response relation for cAMP injection and I_Na,cAMP response, showing blockade of NO effects over a broad range of injection amplitude.

In six of eight MCG somata injected with the pH-indicator dye BCECF, initial pH_i lay close to an averaged 7.0 (0.08 SE; values for the remaining two cells were not attained). Bath application of DEA/NO caused strong and rapid intracellular acidification in all eight cells (Fig. 3). Peak response amplitudes were attained within 13–16 min of donor application and ranged from 0.16 to 0.46 pH units (mean 0.22; SE 0.04; n = 8 cells). Decay of the pH response was monitored for seven cells following washout of DEA/NO and lasted ≤26 min with an average half-time to estimated full recovery of 6.5 min (SE 0.72). The minutes-long time course of acidification and recovery of pH_i was comparable to observations of effects of NO and its slow washout on I_Na,cAMP amplitudes (cf. Fig. 4 in Hatcher et al. 2006). The recovery of pH_i was consistent with the slow recovery times previously observed following acidification in mollucan neurons (Ahmed and Connor 1980; Thomas and Meech 1982).

Fig. 3.

NO stimulated rapid intracellular acidification in isolated metacerebral giant neuron (MCG) somata, as measured with BCECF fluorescence. Bath application of the NO donor, DEA/NO (1 mM, 10 min, solid bar) to a soma in saline at pH 7.5 caused an acidification of intracellular pH (pH_i) >0.3 pH unit, peaking around 15 min after donor application and decaying slowly over minutes following washout.

DISCUSSION

The results show that NO promotes intracellular acidification in neurons whose excitatory state is markedly sensitive to pH_i. This demonstration is significant in providing a functional answer to a long-standing question concerning the likely import of the marked pH sensitivity of I_Na,cAMP in gastropod neurons (Aldenhoff et al. 1983; Green and Gillette 1988).

NO/pH_i regulation of type 1 I_Na,cAMP

Until now, no clearly relevant functional context for the pH sensitivity of I_Na,cAMP has been available. We previously documented two forms of I_Na,cAMP. One is the pH-sensitive type 1 form found in many neurons of the feeding motor network in the opisthobranch gastropods Pleurobranchaea (Green and Gillette 1983, 1988), Navanax and Aplysia (unpublished results), and in the pulmonate snails Helix (Aldenhoff et al. 1983) and Lymnaea (McCrohan and Gillette 1988). This current is a highly sensitive indicator of pH_i, responding with measurable amplitude change to ΔpH_i of 0.03 unit and closely following induced changes in pH_i (Green and Gillette 1988). Two unpublished observations in excised patches indicate a population of pH-sensitive I_Na,cAMP channels (Sudlow and Gillette). In contrast, the type 2 form of I_Na,cAMP, so far found mainly in neurons of the locomotor network (Huang and Gillette 1993), is relatively insensitive both to changing pH_i and to the potentiating effects of NO (Hatcher et al. 2006).

Demonstration of NO-induced acidification in the MCG provides a mechanism for potentiation of type 1 I_Na,cAMP by NO in the same neuron (Hatcher et al. 2006). It also provides a likely pathway for the ability of exogenous NO to activate the feeding motor network (Moroz et al. 1995). The MCG itself is a neuromodulatory excitor of the feeding network (Gillette and Davis 1977). It lacks chemical markers (unpublished) characteristic of nitrergic neurons (Cruz et al. 1997; Floyd et al. 1998) and is thus a more likely NO target than source. NO has roles in activation of the feeding motor network in diverse gastropod species (Elphick et al. 1995; Moroz et al. 1993; Park et al. 1998) but its potentiation of I_Na,cAMP has as yet been documented only in Pleurobranchaea.

Significance

The activity of NOS in Pleurobranchaea's CNS is high, comparable to that of mammalian brain when assayed in vitro (Moroz et al. 1996), and localized to numerous neurons in the feeding and locomotion motor networks (Cruz et al. 1997; Floyd et al. 1998b). The presence of NOS in the feeding motor network and the ability of NO to activate the feeding motor rhythm suggest a normal role in regulating the excitability of the feeding neurons, acting on I_Na,cAMP through pH_i. Mechanisms of NO action in altering pH_i remain to be explored.

These results indicate a novel action for NO in modulating neuronal excitability through intracellular acidification. A similar, although slower, acidification response to NO donors was observed in cultured hippocampal neurons, leading to apoptosis (Vincent et al. 1999). However, effects of NO in the cultured neurons were protein kinase G dependent (Maiese et al. 1993), unlike for the MCG neurons of Pleurobranchaea (Hatcher et al. 2006). An extensive literature search has otherwise failed to turn up similar results. Thus NO effects on pH_i in this system are likely to reflect a new signaling pathway.

Intracellular H⁺ may interact in complex ways with other signaling pathways and metabolic pathways in the cell to add further dimension to cell signaling and neuronal function. Injections of cAMP in neurons of the mollusk Archidoris were found to acidify pH_i ≤0.1 pH unit, an effect likely mediated by cAMP-dependent kinase (Connor and Hockberger 1984b). Ca²⁺ influx during spike activity can acidify neurons through activation of Ca²⁺/H⁺ exchange mechanisms (Ahmed and Connor 1980; Thomas 2009), which in fine neuritic processes can cause swings in pH_i up to several tenths of a unit (Schwiening and Willoughby 2002). Such pH shifts can modify many aspects of pH-sensitive ion channels and metabolic processes.

In this system, at levels of cell and circuit NO-induced acidification is liable to amplify effects of neuromodulator-induced cAMP. Thus our emerging picture of the functional role of NO in the feeding motor network of Pleurobranchaea is one in which it acts as a gain-enhancing mechanism, boosting the cellular responses to neuromodulator-induced fluctuations of cAMP. Such gain control will affect the activity state of the feeding network in large part through broad modulation of pH-sensitive I_Na,cAMP in its neuronal elements.

GRANTS

This work was supported by National Institute of Neurological Disorders and Stroke Grant R01 NS-26838 and National Science Foundation Grant IOS 08-43621.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).