BK channels and a cGMP-dependent protein kinase (PKG) function through independent mechanisms to regulate the tolerance of synaptic transmission to acute oxidative stress at the Drosophila larval neuromuscular junction

Abstract

A cGMP-dependent protein kinase (PKG) has previously been shown to regulate synaptic transmission at the Drosophila neuromuscular junction (NMJ) during acute oxidative stress, potentially through modulation of downstream K⁺ channel kinetics; however, the specific K⁺ channels through which PKG functions remains unclear. In this study, we hypothesized that PKG may be acting on calcium-activated large-conductance Slo K⁺ channels, or BK channels. We found that genetic elimination and pharmacological inhibition of BK channel conductance increases synaptic transmission tolerance to acute H₂O₂-induced oxidative stress. Furthermore, we discovered that activation of PKG in BK channel loss-of-function (Slo⁴) mutants significantly decreases time to stimulus-induced synaptic failure, providing the first evidence of PKG and BK channels functioning independently to control synaptic transmission tolerance to acute oxidative stress.

Keywords:

• Drosophila melanogaster
• oxidative stress
• cGMP-dependent protein kinase
• neuromuscular junction
• BK channels
• iberiotoxin
• synaptic transmission

Introduction

Most organisms require oxygen for survival; however, the metabolism of oxygen produces cytotoxic by-products known as reactive oxygen species (ROS) (Liu, Fiskum, & Schubert, 2002). An overabundance of ROS can induce oxidative stress, which has been implicated in the pathophysiology of stroke (Cuzzocrea, Riley, Caputi, & Salvemini, 2001), traumatic brain injury (Juurlink & Paterson, 1998), and various neurodegenerative disorders (Federico et al., 2012). Because of its ability to withstand extreme fluctuations in oxygen levels (Haddad, 2006) and temperature (Feder, 1997; Misener, Chen, & Walker, 2001), the common fruit fly, Drosophila melanogaster, serves as an ideal model system for investigating mechanisms that control tolerance to acute stress. Previous studies have established that modulation of the cGMP-dependent protein kinase (PKG) pathway in Drosophila regulates tolerance to acute stress, including anoxia/hypoxia (Dawson-Scully et al., 2010), hyperthermia (Dawson-Scully, Armstrong, Kent, Robertson, & Sokolowski, 2007), and oxidative stress (Caplan, Milton, & Dawson-Scully, 2013).

In numerous studies, acute oxidative stress has been induced via application of hydrogen peroxide (H₂O₂), a ROS that increases intracellular calcium (Ca²⁺) to cytotoxic levels, a phenomenon called Ca²⁺ overload (Castro et al., 2004; Herson, Lee, Pinnock, Hughes, & Ashford, 1999; Josephson, Silverman, Lakatta, Stern, & Zweier, 1991; Zhang & Ma, 2008). It has been proposed that stress-induced Ca²⁺ overload increases potassium (K⁺) efflux, which disrupts ionic homeostasis via hyperpolarization of the neuron, triggering cellular damage, synaptic failure, and activation of apoptotic mechanisms (Caplan et al., 2013; Gerbino et al., 2009; Hughes, Bortner, Purdy, & Cidlowski, 1997; Money, Rodgers, McGregor, & Robertson, 2009). A previous publication from our lab found that inhibition of both PKG activity and K⁺ channel conductance increases synaptic transmission tolerance to H₂O₂-induced oxidative stress at the Drosophila neuromuscular junction (NMJ); further, the study demonstrated that 2.25 mM H₂O₂ exposure depolarizes resting membrane potential (RMP), potentially contributing to synaptic failure (Caplan et al., 2013). Findings from other studies have proposed a connection between PKG activity and K⁺ channel conductance (Dawson-Scully et al., 2010; Renger, Yao, Sokolowski, & Wu, 1999; White et al., 1993), suggesting that K⁺ channels may be downstream targets of the PKG signaling cascade; however, the specific K⁺ channels that facilitate the effects of the PKG pathway remain elusive.

Research in Drosophila has facilitated the discovery of various types of K⁺ channels, including Shab, Shaker, Shaw, and Slo channels (Frolov, Bagati, Casino, & Singh, 2012). In an attempt to isolate K⁺ channels involved downstream of PKG, we focused on Slo channels. Specifically, we targeted BK channels, or calcium-activated large-conductance Slo K⁺ channels, which have been shown to have a ten-fold increase in K⁺ conductance compared to classic voltage-gated K⁺ channels (Pallotta, Magleby, & Barrett, 1981; Stefani et al., 1997). We hypothesized that blockade of BK channels would increase synaptic transmission tolerance to H₂O₂-induced stress by reducing K⁺ efflux. Pharmacological activation of PKG using 8-bromo-cGMP has been shown to increase BK channel conductance (Alioua et al., 1998; White et al., 1993); consequently, we proposed that PKG modulates downstream BK channels in the regulation of synaptic transmission tolerance to acute oxidative stress.

In the present study, we compared synaptic transmission tolerance to 2.25 mM H₂O₂-induced acute oxidative stress at the Drosophila larval glutamatergic NMJ in w¹¹¹⁸ control larvae and BK channel loss-of-function (Slo⁴) mutants. PKG activation in w¹¹¹⁸ control and Slo⁴ larvae was induced using sildenafil citrate and 3′,5′-cyclic monophosphate 8-bromo sodium salt (8-bromo-cGMP). Activation of voltage-gated K⁺ channels in both strains was induced using dichloroacetate (DCA). Pharmacological reduction of BK channel activity in w¹¹¹⁸ control larvae was induced with iberiotoxin. K⁺ channel inhibition was achieved in w¹¹¹⁸ larvae using tetraethylammonium (TEA).

Materials and methods

Animals

Drosophila melanogaster third instar wandering larvae (≈110-h old) were utilized throughout this study. All larvae were reared on standard Bloomington stock fly food in an incubator at 25 °C under a 12:12-h light–dark cycle. Slo⁴ mutants on a w¹¹¹⁸ genetic background were used as BK channel loss-of-function mutants (Atkinson, Robertson, & Ganetzky, 1991). Slo⁴ is a gamma ray-induced chromosomal inversion on the third chromosome between 96A17 and 96F5-8 (Atkinson et al., 1991; Kadas, Ryglewski, & Duch, 2015). The Slo⁴ mutants were gifted by Dr. Gregory Macleod, but were originally obtained from Dr. Nigel Atkinson. The w¹¹¹⁸ control larvae were also gifted by Dr. Gregory Macleod, but were originally obtained from Dr. Nancy Bonnini.

Electrophysiology

All electrophysiological experiments were performed as previously described (Caplan et al., 2013). Briefly, individual w¹¹¹⁸ or Slo⁴ mutant larvae were pinned to a glass dissecting dish containing Schneider’s insect medium (Sigma, St. Louis, MO), and a longitudinal incision was made along the dorsal side of the larva in order to remove the central nervous system and internal organs. An extracellular glass suction electrode was fashioned to stimulate segmental nerves synapsing with individual muscle segments, and a 60–90 MΩ intracellular recording electrode was utilized in place of a 40 mΩ electrode. Further, the insect medium was replaced with HL3 recording saline containing 1.5 mM Ca²⁺ and 20 mM Mg²⁺ (Caplan et al., 2013; Stewart, Schuster, Goodman, & Atwood, 1996).

To quantify synaptic transmission tolerance to various treatment conditions, excitatory junction potentials (EJPs) were evoked via repetitive stimulation (0.3-ms pulses delivered suprathreshold with a frequency of 1 Hz) and time to synaptic failure (EJP amplitude <1 mV) was recorded (Caplan et al., 2013); preparations that took a longer duration of time (minutes) to reach synaptic failure in the presence of 2.25 mM H₂O₂ possessed greater tolerance for acute oxidative stress compared to those that took a shorter duration of time. Measurements from larval preparations were initially recorded in HL3 saline; after 5 min, the HL3 saline bath was replaced with HL3 saline containing 2.25 mM H₂O₂ and a drug, depending on the treatment condition. After 10 min of H₂O₂ exposure, larval preparations were tested for acute effects on characteristic EJP parameters, including differences in waveform shape and amplitude. Intracellular recordings of resting membrane potential (RMP) were measured from larval muscle 6.

Pharmacology

Acute oxidative stress was elicited in larval preparations via the addition of 2.25 mM H₂O₂ to HL3 saline. Addition of pharmacological agents to the larval NMJ preparation was performed as previously described (Caplan et al., 2013); however, all drugs were water-soluble at the concentrations used and thus were dissolved in ddH₂O as opposed to dimethyl sulfoxide (DMSO). A published dosage of 40 µM 8-bromo-cGMP (Caplan et al., 2013; Dawson-Scully et al., 2007) was administered to w¹¹¹⁸ control and Slo⁴ mutant larvae to activate PKG. Sildenafil citrate, another PKG activator (Broderick, Kean, Dow, Pyne, & Davies, 2004; Day, Dow, Houslay, & Davies, 2005; Francis, Turko, & Corbin, 2001), was also administered to w¹¹¹⁸ and Slo⁴ mutant larvae at the 40 µM dosage. Iberiotoxin, a BK channel inhibitor (McKay et al., 1994), was administered at the 500 pM dosage in w¹¹¹⁸ and Slo⁴ mutant larvae. Additionally, 0.25 mM DCA, a voltage-gated K⁺ channel activator (Bonnet et al., 2007; Michelakis et al., 2002), was administered to w¹¹¹⁸ and Slo⁴ mutant larvae. A published dosage of 0.25 mM TEA, a K⁺ channel inhibitor (Caplan et al., 2013), was administered to w¹¹¹⁸ larvae. Dose–response curves were utilized to determine the optimal dosage of all drugs presented in this manuscript. All drugs and chemicals used in this study were obtained from Sigma-Aldrich Inc. (Sigma-Aldrich Inc., St. Louis, MO).

Statistics

All data were analyzed using SigmaPlot (Systat Software Inc., San Jose, CA). One-way analysis of variance (ANOVA) and Holm–Sidak post hoc analysis were used to compare significant differences across groups. Statistical significance (p < .05) in the vertical bar charts was designated using letters, where different letters show statistically significant differences and the same letter indicates non-significance. The letter ‘A’ designates the highest mean, ‘B’ indicates the next highest mean, and so forth. All vertical bar charts are presented as mean ± SEM.

Results

Genetic elimination of BK channels regulates synaptic transmission tolerance to acute oxidative stress

To investigate whether inhibition of BK channel conductance would increase synaptic transmission tolerance to 2.25 mM H₂O₂-induced oxidative stress, average time to synaptic failure of BK channel loss-of-function (Slo⁴) mutants and w¹¹¹⁸ control larvae was compared; w¹¹¹⁸ control larvae underwent synaptic failure significantly earlier than Slo⁴ mutants under conditions of acute oxidative stress [one-way ANOVA, F(3, 16) = 17.76, p < .05; Figure 1(A)]. w¹¹¹⁸ and Slo⁴ mutant larvae displayed varying rates of EJP amplitude decline, as shown in Figure 1(B). Further, application of 2.25 mM H₂O₂ in both the w¹¹¹⁸ control and Slo⁴ mutant larvae did not acutely affect the characteristic shape of the EJP waveforms (Figure 1(C)). Interestingly, time to synaptic failure did not significantly differ in Slo⁴ mutants in either the presence or absence of H₂O₂ [p = .917; Figure 1(A)]. Elimination of BK channels in Slo⁴ mutants seems to have conferred resistance to H₂O₂ due to the large reduction of K⁺ channel conductance.

Figure 1. Genetic elimination of BK channels increases synaptic transmission tolerance to acute oxidative stress. (A) Average time to synaptic failure is prolonged in Slo⁴ mutants compared to w¹¹¹⁸ control larvae during H₂O₂ exposure [one-way ANOVA, F(3, 16) = 17.76, p < .05]. Vertical bar chart is shown as mean ± SEM. Different letters on bar charts indicate statistical significance, whereas the same letter indicates a nonsignificant difference. (B) Average amplitude decline of evoked EJPs in w¹¹¹⁸ controls and Slo⁴ mutants in the presence and absence of 2.25 mM H₂O₂ (n = 4–6 preparations per group). Black arrows indicate time to synaptic failure of individual w¹¹¹⁸ control larval preparations in HL3 saline. Each data point is displayed as mean ± SEM. (C) Representative waveforms of evoked EJPs from w¹¹¹⁸ control larvae 1 min before H₂O₂ exposure (black line) and 10 min after H₂O₂ exposure (gray line). (D) Representative waveforms of evoked EJPs from Slo⁴ mutant larvae 1 min before H₂O₂ exposure (black line) and 10 min after H₂O₂ exposure (gray line). H₂O₂ does not acutely affect characteristic EJP shape in both w¹¹¹⁸ control and Slo⁴ mutant larvae.

Further, application of 2.25 mM H₂O₂ to both w¹¹¹⁸ and Slo⁴ mutant larvae affected resting membrane potential (RMP) and, consequently, EJP amplitude (Figure 2). RMP depolarization occurred more rapidly in w¹¹¹⁸ and Slo⁴ larvae exposed to 2.25 mM H₂O₂ compared to HL3 controls (Figure 2(A)); a corresponding decrease in EJP amplitude was observed in the same larval preparations (Figure 2(B)). Hyperpolarized RMPs corresponded to higher EJP amplitudes, whereas depolarized RMPs corresponded to lower EJP amplitudes.

Figure 2. Drosophila muscle resting membrane potential (RMP) and EJP amplitude decline during acute oxidative stress. (A) RMP from Drosophila larval muscle 6 was recorded in w¹¹¹⁸ control larvae and Slo⁴ mutants in the presence and absence of 2.25 mM H₂O₂ (n = 6 per group). RMP depolarization occurs more rapidly in w¹¹¹⁸ and Slo⁴ mutant larvae exposed to 2.25 mM H₂O₂ compared to HL3 control larvae. (B) In the same larval preparations, EJP amplitude was also recorded (n = 6 per group). Loss of RMP tracks EJP amplitude decline across all treatments.

PKG modulation of downstream K⁺ channels mediates synaptic transmission tolerance to acute stress

Genetic elimination of BK channel conductance and pharmacological activation of the PKG pathway were combined to determine the potential mechanistic action of PKG on downstream BK channels. Slo⁴ mutants were exposed to oxidative stress (2.25 mM H₂O₂) and treated with 40 µM of PKG activators 8-bromo-cGMP and sildenafil citrate. Application of 40 µM 8-bromo-cGMP significantly reduced time to synaptic failure in Slo⁴ mutants compared to mutants only treated with H₂O₂ [one-way ANOVA, F(9, 53) = 13.24, p < .001; Figure 3], as did application of 40 µM sildenafil citrate [p < .001; Figure 3]. Contrary to our original hypothesis, these results suggest that PKG modulates K⁺ channels other than BK channels to regulate neuromuscular communication during H₂O₂-induced stress. Activation of PKG using 8-bromo-cGMP in w¹¹¹⁸ control larvae exposed to acute oxidative stress did not significantly alter synaptic transmission tolerance compared to w¹¹¹⁸ larvae only treated with H₂O₂ [p = .651; Figure 3]. Additionally, activation of PKG using sildenafil citrate did not significantly alter time to synaptic failure during acute oxidative stress in w¹¹¹⁸ larvae [p = 0.313; Figure 3]. These results suggest that activation of PKG does not exacerbate the deleterious effects of H₂O₂ in w¹¹¹⁸ control larvae.

Figure 3. PKG functions independently of BK channels to protect synaptic function from H₂O₂ exposure. PKG activation using 40 µM 8-bromo-cGMP [one-way ANOVA, F(9, 53) = 13.24, p < .001] and 40 µM sildenafil citrate [p < .001] in Slo⁴ mutant larvae significantly decreases time to synaptic failure during acute oxidative stress. Activation of PKG with 40 µM 8-bromo-cGMP [p = .651] and 40 µM sildenafil citrate [p = .313] does not alter time to synaptic failure in w¹¹¹⁸ control larvae. Activation of voltage-gated K⁺ channels using 0.25 mM DCA in w¹¹¹⁸ control larvae does not alter time to synaptic failure during acute oxidative stress [p = .561]. Application of 0.25 mM DCA to Slo⁴ mutant larval preparations decreases time to synaptic failure during oxidative stress [p < .05]. Vertical bar chart is shown as mean ± SEM. n = 5–9 preparations per group.

To further confirm that PKG acts on K⁺ channels other than BK channels, we manipulated K⁺ channel kinetics using DCA, a voltage-gated K⁺ channel activator. Administration of 0.25 mM DCA and 2.25 mM H₂O₂ to Slo⁴ mutant larvae led to a significant reduction in time to synaptic failure compared to Slo⁴ H₂O₂ control larvae [p < .05; Figure 3], suggesting that non-BK K⁺ channels also modulate synaptic transmission during H₂O₂ stress. In w¹¹¹⁸ larvae, 0.25 mM DCA did not significantly alter synaptic transmission tolerance to acute stress [p = .561; Figure 3]. Taken together, these results suggest that the PKG pathway functions independently of BK channels in the regulation of synaptic transmission tolerance to acute oxidative stress, potentially through the modulation of other downstream voltage-gated K⁺ channels.

Pharmacological inhibition of K⁺ channel conductance regulates neurotransmission during acute oxidative stress

To further examine the role of BK channels in protecting synaptic transmission during acute oxidative stress, low doses of iberiotoxin were administered to w¹¹¹⁸ control larval preparations. Application of 500 pM iberiotoxin increased synaptic transmission tolerance in w¹¹¹⁸ larvae exposed to 2.25 mM H₂O₂ compared to w¹¹¹⁸ H₂O₂ control larvae [one-way ANOVA, F(3, 19) = 9.449, p < .05; Figure 4]. Thus, acute pharmacological inhibition of BK channel conductance in w¹¹¹⁸ larvae has similar protective effects compared to genetic elimination of BK channel conductance.

Figure 4. Pharmacological blockade of BK channel conductance protects synaptic transmission from acute stress. Pharmacological inhibition of BK channel conductance using 500 pM iberiotoxin significantly increases time to synaptic failure in w¹¹¹⁸ control larvae exposed to 2.25 mM H₂O₂ compared to w¹¹¹⁸ H₂O₂ controls [one-way ANOVA, F(3, 19) = 9.449, p < .05]. Nonselective K⁺ channel inhibition using 0.25 mM TEA also increases synaptic transmission tolerance to acute oxidative stress in w¹¹¹⁸ control larvae [p < .05]. Vertical bar chart is shown as mean ± SEM. n = 5–6 preparations per group.

To explore the effects of nonselective K⁺ channel inhibition on synaptic transmission, we applied TEA to w¹¹¹⁸ larvae exposed to 2.25 mM H₂O₂. We found that administration of 0.25 mM TEA increased synaptic transmission tolerance during acute oxidative stress, similar to the effects of 500 pM iberiotoxin in w¹¹¹⁸ larval preparations [p < .05; Figure 4]. These results suggest that pharmacological inhibition of both BK channels and classic voltage-gated K⁺ channels protects synaptic transmission from the harmful effects of acute oxidative stress.

Iberiotoxin does not possess major off-target effects in Slo⁴ mutants

Although iberiotoxin has been shown to inhibit Drosophila BK channels in GH3 cells (McKay et al., 1994), its effects have not been observed at the Drosophila NMJ; to address this concern, we tested for potential off-target effects by administering 500 pM iberiotoxin to Slo⁴ mutants. Because Slo⁴ mutants lack BK channels, a low dosage of iberiotoxin should not significantly increase synaptic transmission tolerance to acute oxidative stress; consistent with this hypothesis, 500 pM iberiotoxin did not significantly alter time to synaptic failure in Slo⁴ mutants during H₂O₂ exposure [one-way ANOVA, F(2, 16), p = .736; Figure 5], indicating that iberiotoxin specifically modulates BK channels at this dosage and does not possess any major off-target effects using our synaptic transmission assay.

Figure 5. Iberiotoxin does not possess major off-target effects using this experimental model. Administration of BK channel inhibitor iberiotoxin does not significantly alter time to synaptic failure in Slo⁴ mutant larvae [one-way ANOVA, F(2, 16), p = .736], suggesting that iberiotoxin does not possess any major off-target effects at the Drosophila larval NMJ. Vertical bar chart is shown as mean ± SEM. n = 4–6 preparations per group.

Discussion

Previous work demonstrated that PKG regulates synaptic transmission tolerance to acute oxidative stress at the Drosophila NMJ, possibly through modulation of K⁺ channel conductance (Caplan et al., 2013). The present study served to further investigate which K⁺ channels might be functioning downstream of the PKG pathway. Findings from the present study provide evidence of PKG and BK channel regulation of synaptic function during acute oxidative stress. Genetic and pharmacological inhibition of BK channels led to increases in synaptic transmission tolerance to 2.25 mM H₂O₂-induced oxidative stress; in addition, nonselective inhibition of K⁺ channels in w¹¹¹⁸ larvae also increases time to synaptic failure during acute oxidative stress. The results indicate that PKG and BK channels function independently to regulate synaptic transmission during acute stress; in other words, PKG acts on downstream K⁺ channels other than BK channels to control neuromuscular communication during H₂O₂ exposure.

Using H₂O₂-induced stress as a model to probe effects of PKG pathway manipulation on neurotransmission in Drosophila

H₂O₂ has been used to model oxidative stress in various neurological disorders, including Alzheimer’s disease (Behl, Davis, Lesley, & Schubert, 1994; Miyata & Smith, 1996), Parkinson’s disease (Jenner, 2003), and amyotrophic lateral sclerosis (Nani, Cifra, & Nistri, 2010). H₂O₂ has also been shown to modulate neurotransmission in multiple electrophysiological paradigms, including the frog NMJ (Giniatullin & Giniatullin, 2003), rat striatal neurons (Smith, Herson, Lee, Pinnock, & Ashford, 2003), and the Drosophila NMJ (Caplan et al., 2013). Thus, studying H₂O₂ exposure at the Drosophila NMJ serves as an ideal model to study novel targets of the PKG signaling cascade, enabling future work in the development of protective pathways that can counteract the deleterious effects of oxidative stress.

Role of BK channels in ionic homeostasis and synaptic transmission

A stimulus-induced rise in [Ca²⁺]_i has been shown to activate BK channels in Drosophila and mammalian neurons (Bian, Favre, & Moczydlowski, 2001; DiChiara & Reinhart, 1995; Nimigean & Magleby, 1999; Piskorowski & Aldrich, 2002), resulting in BK channel-dependent efflux of [K⁺]_i, repolarization of the membrane, and reduction of Ca²⁺ influx (Faber & Sah, 2003; Lancaster & Nicoll, 1987; Vetri, Choudhury, Pelligrino, & Sundivakkam, 2014). BK channels also modulate various neurophysiological processes, including neurotransmitter release, interspike interval, and spike frequency (Faber & Sah, 2003). Consequently, BK channels have been shown to regulate many essential physiological processes in both fruit flies and mammals. For instance, BK channels have been implicated in several human ailments, such as urinary incontinence (Lawson, 2000), neuromuscular abnormalities (Lehmann & Jurkat, 1999), hypertension (Yang et al., 2013), and epilepsy (Mantegazza, Curia, Biagini, Ragsdale, & Avoli, 2010). In the present study, BK channels were found to mediate synaptic transmission tolerance to acute oxidative stress at the Drosophila larval NMJ.

Potential downstream targets of the PKG pathway

Our results demonstrate that activation of PKG in Slo⁴ mutants significantly reduces synaptic transmission tolerance to acute oxidative stress. TEA was used as a nonselective K⁺ channel inhibitor in the present study. Interestingly, it has been shown that low doses of TEA inhibit Drosophila slo channels expressed in Xenopus oocytes (Shen et al., 1994) as well as voltage-gated K⁺ channels, including K₃ channels, in Drosophila muscle (Gorczyca & Wu, 1991). Our results demonstrate that application of 0.25 mM TEA to w¹¹¹⁸ controls protects synaptic transmission from oxidative stress, likely due to inhibition of both BK channels and classic voltage-gated K⁺ channels. Moreover, we propose that PKG functions independently of BK channel conductance in the modulation of synaptic transmission during acute stress.

Notably, application of DCA to Slo⁴ mutants significantly decreased time to synaptic failure during acute oxidative stress. A previous publication similarly utilized DCA to activate voltage-gated K⁺ channels in Drosophila (Dawson-Scully et al., 2010); however, this study did not directly demonstrate DCA activation of voltage-gated K⁺ channels in neurons or muscles. Recently, it has been shown that mature flies fed DCA display acute increases in ROS production and oxidative stress (Pandey et al., 2014). Therefore, the DCA-induced decrease in time to synaptic failure in Slo⁴ mutants could potentially be attributed to DCA activating K⁺ channels, exacerbating oxidative stress, or modulating other metabolic processes.

Action potential dynamics as potential mechanism for protective effect of K⁺ channel conductance inhibition

In the current study, genetic inhibition of BK channel conductance using Slo⁴ mutants and pharmacological inhibition using iberiotoxin led to increased synaptic transmission tolerance to acute oxidative stress. Inhibition of K⁺ channel conductance using TEA resulted in similar extensions in time to synaptic failure. Inhibiting K⁺ channel conductance results in reduced K⁺ efflux, which has been shown to prevent the initiation of cellular apoptotic cascades (Bortner, Hughes, & Cidlowski, 1997). Additionally, blocking K⁺ channel conductance depolarizes neurons, broadening the action potential to release neurotransmitters; this increased transmitter release may delay rundown of synaptic transmission, or increase time to synaptic failure during acute oxidative stress. Previously published work provides support for this potential mechanism by demonstrating that pharmacological inhibition of Shaker K⁺ channels in Slo⁴ mutants increases neurotransmitter release and excitatory postsynaptic potential amplitude compared to wild-type controls at the Drosophila NMJ (Lee, Ueda, & Wu, 2008; Lee, Ueda, & Wu, 2014). Consistent with this finding, pharmacological inhibition of BK channel conductance leads to an increase in transmitter release within the mammalian hippocampus (Griguoli, Sgritta, & Cherubini, 2016; Raffaelli, Saviane, Mohajerani, Pedarzani, & Cherubini, 2004).

Additionally, the relationship between synaptic failure and resting membrane potential has been previously studied. A group found that RMP in Drosophila larval body wall muscles is directly correlated with EJP amplitude in electrophysiological recordings utilizing HL3 saline; consistent with our findings, they demonstrated that hyperpolarized RMPs were correlated with higher EJP amplitudes, whereas depolarized RMPs were correlated with lower EJP amplitudes (Krans, Parfitt, Gawera, Rivlin, & Hoy, 2010). Thus, synaptic transmission failure could be partly attributed to changes in RMP.

Using BK channel inhibitors to confer neuroprotection

Recent studies have indicated that activation of BK channel conductance protects neurons from stroke-induced cell death (Chi et al., 2010; Gribkoff et al., 2001; Li et al., 2014). However, the drugs utilized in these studies (chlorpromazine, BMS-204352, and Baifuzi) possess major off-target effects; for instance, CPZ inhibits dopamine D2 receptors (Carpenter & Koenig, 2008), reduces serotonergic activity (Croll, Baker, Khabarova, Voronezhskaya, & Sakharov, 1997), and modulates Ca²⁺ channels (McNaughton, Green, & Randall, 2001), Cl⁻ channels (Quamme, 1997), and nicotinic receptor channels (Benoit & Changeux, 1993). Thus, protection of neurons from stroke may not be specifically attributable to BK channel activation.

A previous study assessed the protective effect of BK channel blockade in a human neuronal cell line SH-SY5Y during H₂O₂-induced oxidative stress and CoCl₂-induced hypoxia mimetic insults; the results revealed that specific BK channel activators, isopimaric acid and NS1619, exacerbated cytotoxicity, whereas inhibition of BK channels using TEA and tetrandrine provided a neuroprotective effect (Zhang, 2016). Because increases in [Ca²⁺]_i activate BK channels, K⁺ channel blockers may be utilized in protecting neurons from oxidative stress. The findings from Zhang (2016) are consistent with our results, demonstrating that inhibition of BK channel and voltage-gated K⁺ channel conductance protects synaptic transmission during acute oxidative stress. Further confirming the role of BK channel inhibition in neuroprotection, a recent study provided evidence of BK channel inhibitors, including iberiotoxin, charybdotoxin, paxilline, and TEA, protecting neurons from stroke-induced insults (Chen et al., 2013). Consequently, the results from Chen et al. (2013) indicate that optimizing the dosage range of specific BK channel blockers (iberiotoxin) and K⁺ channel inhibitors (TEA) may assist in the development of effective treatments for stroke-induced cell death.

Iberiotoxin has been used to inhibit BK channel conductance in mammalian systems (McKay et al., 1994). BK channels expressed in Xenopus oocytes have been shown to be insensitive to iberiotoxin up to 100 nM (Meera, Wallner, Song, & Toro, 1997). Thus, there is no direct evidence of low doses of iberiotoxin inhibiting Drosophila BK channels; however, 200 nM charybdotoxin has been shown to inhibit the BK channel encoded K⁺ current (I^CF current) at the Drosophila NMJ (Kadas et al., 2015). Despite its highly similar amino acid sequence and mechanism of action, iberiotoxin has been shown to be approximately ten times more effective at inhibiting rat BK channels compared to charybdotoxin (Candia, Garcia, & Latorre, 1992). If low doses of iberiotoxin also inhibited non-BK K⁺ channels, we might expect to observe an increase in time to synaptic failure in Slo⁴ mutants using our assay; contrarily, our results demonstrate that application of 0.5 nM iberiotoxin to the Drosophila NMJ does not significantly modify time to synaptic failure in Slo⁴ mutants. Thus, 0.5 nM iberiotoxin may specifically act on BK channels as opposed to other K⁺ channels at the Drosophila NMJ. Iberiotoxin was used in this study because of its structural similarity to charybdotoxin; unlike charybdotoxin, however, iberiotoxin has not demonstrated any off-target effects at the larval NMJ (Candia et al., 1992; Kadas et al., 2015).

Localization of BK channel-specific neuroprotection at the Drosophila NMJ

Although this study provided evidence of increased synaptic transmission tolerance to H₂O₂ exposure through BK channel inhibition, we did not localize the effect to presynaptic motor neurons or postsynaptic muscles. Expression of BK channels in presynaptic boutons at the Drosophila larval NMJ has been shown (Zhou et al., 1999); however, expression patterns of BK channels in postsynaptic muscles are not well-characterized via antibody expression. Despite this, various morphological and electrophysiological studies strongly suggest that BK channels are expressed in both larval (Komatsu, Singh, Rathe, & Wu, 1990; Lee et al., 2008; Singh & Wu, 1990) and adult muscles (Elkins & Ganetzky, 1988). Future work will elucidate the mechanisms through which muscle-specific BK channels function to protect synaptic transmission from acute oxidative stress.

Quantifying synaptic transmission in the presence of physiological stress at the larval NMJ is an effective way to identify protective genes and pharmacological agents

Many groups have utilized the Drosophila larval NMJ to identify drugs or genes that mediate cell death, or synaptic failure. For instance, results from a previous study indicate that a prior heat-shock protects synaptic transmission during subsequent heat-induced stress and that this effect is largely mediated by genes encoding Hsp70 (Karunanithi, Barclay, Robertson, Brown, & Atwood, 1999). Additionally, pharmacological inhibition of the PKG pathway was found to increase synaptic transmission tolerance to acute heat stress at the larval NMJ (Dawson-Scully et al., 2007; Krill & Dawson-Scully, 2016). Recently, it was discovered that pseudopterosin A (PsA), a natural product, increases synaptic transmission tolerance to acute oxidative stress at the Drosophila NMJ (Caplan, Zheng, Dawson-Scully, White, & West, 2016). Thus, the Drosophila larval NMJ is a well-established model for studying neurotransmission during physiological stress and serves as a starting point for identifying neuromodulatory drugs that protect synaptic function from acute stress.

Conclusions

In this study, we provide evidence of increased synaptic transmission tolerance to acute oxidative stress through genetic elimination and acute pharmacological inhibition of BK channel conductance. Additionally, we demonstrated that BK channels function independently of PKG-modulated K⁺ channels to protect synaptic transmission during H₂O₂ exposure. Consistent with these results, we found that acute activation of voltage-gated K⁺ channel conductance reduced time to synaptic failure during oxidative stress in Slo⁴ mutants. These results suggest that reduction of the overall conductance of BK channels, non-BK voltage-gated K⁺ channels, and PKG-modulated K⁺ channels prolongs synaptic function during acute stress at the Drosophila NMJ. Thus, development of specific K⁺ channel modulators may emerge as novel therapies for the treatment of oxidative stress-induced physiological insults.

Acknowledgements

We would like to thank Dr. Gregory Macleod for providing fly stocks and critical feedback. We also thank Dr. William Ja, Dr. Colin Hughes, Stephanie Kelly, Karlis Justs, and anonymous reviewers for providing valuable insights.

This article is dedicated to Professor Harold Atwood, who spent a career building a body of knowledge on the integration of morphological and physiological processes that modulate neuromuscular junction synaptic transmission. He also created a vast field of study based on the numerous investigators he trained through the decades of his works. On a personal note, I strive to create a scientific and social environment like the one I experienced as a Ph.D. student. As most young people, I mistakenly assumed that having such an environment was an easy thing to implement; but I see now that skill and expertise is also required to create an environment that built, what most of us remember as, the Atwood Lab. I will continue to strive for this with my own laboratory.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This research was support by a CTP (Compound Transfer Program) grant from Pfizer (WI225058 to K.D.-S.).