Dystrobrevin Controls Neurotransmitter Release and Muscle Ca²⁺ Transients by Localizing BK Channels in Caenorhabditis elegans

Abstract

Dystrobrevin is a major component of a dystrophin-associated protein complex. It is widely expressed in mammalian tissues, including the nervous system, in which it is localized to the presynaptic nerve terminal with unknown function. In a genetic screen for suppressors of a lethargic phenotype caused by a gain-of-function isoform of SLO-1 in Caenorhabditis elegans, we isolated multiple loss-of-function (lf) mutants of the dystrobrevin gene dyb-1. dyb-1(lf) phenocopied slo-1(lf), causing increased neurotransmitter release at the neuromuscular junction, increased frequency of Ca²⁺ transients in body-wall muscle, and abnormal locomotion behavior. Neuron- and muscle-specific rescue experiments suggest that DYB-1 is required for SLO-1 function in both neurons and muscle cells. DYB-1 colocalized with SLO-1 at presynaptic sites in neurons and dense body regions in muscle cells, and dyb-1(lf) caused SLO-1 mislocalization in both types of cells without altering SLO-1 protein level. The neuronal phenotypes of dyb-1(lf) were partially rescued by mouse α-dystrobrevin-1. These observations revealed novel functions of the BK channel in regulating muscle Ca²⁺ transients and of dystrobrevin in controlling neurotransmitter release and muscle Ca²⁺ transients by localizing the BK channel.

Introduction

Dystrobrevins are cytosolic proteins that bind to dystrophin and are considered to be a major component of a dystrophin-associated protein complex (DAPC) that links the cytoskeleton to extracellular matrix (Davies and Nowak, 2006). There are two dystrobrevin genes (αDB and βDB) in human and mice (Rees et al., 2007) but only one such gene (dyb-1) in Caenorhabditis elegans (www.wormbase.org, WS225). In mouse, knock-out of αDB causes skeletal and cardiac myopathies and impaired nitric oxide-mediated signaling (Grady et al., 1999), and knock-out of both αDB and βDB causes synaptic defects and abnormal motor behavior (Grady et al., 2006). In C. elegans, mutations of either dyb-1 or the dystrophin gene dys-1 cause muscle degeneration in a sensitized genetic background as well as behavioral and pharmacological phenotypes suggestive of enhanced cholinergic transmission (Gieseler et al., 1999; Giugia et al., 1999). Intriguingly, similar phenotypes are observed in mutants of the BK channel gene slo-1 (Carre-Pierrat et al., 2006). The relationship between SLO-1 and the DAPC is not totally clear. A recent study shows that a membrane protein known as ISLO-1 is important to SLO-1 subcellular localization in C. elegans body-wall muscle by interacting with the DAPC (Kim et al., 2009).

The BK channel is a Ca²⁺ and voltage-gated K⁺ channel expressed in many tissues, including the nervous system (Wang, 2008) and skeletal muscle (Latorre et al., 1982; Blatz and Magleby, 1984; Knaus et al., 1995). In the nervous system, the BK channel colocalizes with voltage-gated Ca²⁺ channels at axon presynaptic sites (Robitaille et al., 1993; Yazejian et al., 2000) and serves as a key negative regulator of neurotransmitter release (Robitaille et al., 1993; Hu et al., 2001; Wang et al., 2001; Raffaelli et al., 2004; Wang, 2008). The localization of the BK channel to the vicinity of voltage-gated Ca²⁺ channels allows it to be activated by Ca²⁺ microdomains, which are hemispheric sites of high [Ca²⁺] at the inner mouth of open Ca²⁺ channels (Roberts et al., 1990; Augustine et al., 2003). In mammalian striated muscle, the BK channel is enriched in the transverse tubule (t-tubule) membrane (Latorre et al., 1982; Knaus et al., 1995) with unknown function. In C. elegans body-wall muscle, which is analogous to mammalian striated muscle (Moerman and Fire, 1997), SLO-1 colocalizes with the L-type voltage-gated Ca²⁺ channel EGL-19. SLO-1 in muscle is potentially involved in regulating locomotion and egg-laying behaviors (Kim et al., 2009; Abraham et al., 2010; Chen et al., 2010a,b).

Through a genetic screen for suppressors of a behavioral phenotype caused by a gain-of-function (gf) isoform of SLO-1, we identified DYB-1 as a protein essential to SLO-1 function in vivo. Analyses of mutant phenotypes revealed that SLO-1 regulates Ca²⁺ transients in body-wall muscle cells and that DYB-1 regulates neurotransmitter release and muscle Ca²⁺ transients by localizing SLO-1 to presynaptic sites and muscle-dense body regions, respectively. These findings potentially help to understand the molecular mechanisms of neurological and muscular defects caused by deficiencies of the DAPC.

Materials and Methods

Growth and culture of C. elegans.

C. elegans hermaphrodites were grown on agar plates with a layer of OP50 Escherichia coli at room temperature (21–22°C) or inside an environmental chamber (21°C).

Strains.

N2 Bristol was used as the wild-type in all experiments. The other strains used in this study are as follows: ZW083 [zwIs101(Pslo-1::slo-1::GFP)]; ZW320 [zwIs129(Pslo-1::slo-1(gf); Pmyo-2::YFP)]; ZW331 [zwIs129(Pslo-1::slo-1(gf); Pmyo-2::YFP); dyb-1(zw11)]; ZW349 [dyb-1(zw11)]; ZW352 [zwIs101(Pslo-1::slo-1::GFP); dyb-1(zw11)]; ZW385 [zwEx132(Pdyb-1::GFP; lin-15(+)); lin-15(n765)]; ZW471 [zwEx150(Pdyb-1::αDB1; Pmyo-3::GFP); zwIs129(Pslo-1::slo-1(gf); Pmyo-2::YFP); dyb-1(zw11)]; ZW495 [zwIs132(Pmyo-3::GCaMP2; lin-15(+))]; ZW527 [zwIs132(Pmyo-3::GCaMP2; lin-15(+)); slo-1(md1745)]; ZW528 [zwIs132(Pmyo-3::GCaMP2; lin-15(+)); zwIs129(Pslo-1::slo-1(gf); Pmyo-2::YFP)]; ZW536 [zwIs132(Pmyo-3::GCaMP2; lin-15(+)); dyb-1(zw11)]; ZW581 [slo-1(md1745); dyb-1(zw11)]; ZW604 [zwIs101(Pslo-1::slo-1::GFP); zwIs135(Pdyb-1::dyb-1::mStrawberry; rol-6(+))]; ZW605 [zwEx164(Punc-47::slo-1::mStrawberry; Punc-25::GFP::unc-2; lin-15(+)); lin-15(n765)]; ZW608 [zwEx166(Pmyo-3::dyb-1; Pmyo-3::GFP); zwIs129(Pslo-1::slo-1(gf); Pmyo-2::YFP); dyb-1(zw11)]; ZW609 [zwEx167(Prab-3::dyb-1; Prab-3::GFP); zwIs129(Pslo-1::slo-1(gf); Pmyo-2::YFP); dyb-1(zw11)]; ZW610 [zwEx166(Pmyo-3::dyb-1; Pmyo-3::GFP); dyb-1(zw11)]; ZW611 [zwEx167(Prab-3::dyb-1; Prab-3::GFP); dyb-1(zw11)]; ZW612 [zwIs101(Pslo-1::slo-1::GFP); dys-1(cx18)]; ZW615 [zwIs136(Pdyb-1::dyb-1::GFP; lin-15(+)); lin-15(n765)]; ZW616 [zwIs136(Pdyb-1::dyb-1::GFP; lin-15(+)); dys-1(cx18)]; ZW622 [zwEx168(Pmyo-3::αDB1; Pmyo-3::GFP); zwIs129(Pslo-1::slo-1(gf); Pmyo-2::YFP); dyb-1(zw11)]; ZW623 [zwEx169(Prab-3::αDB1; Prab-3::GFP); zwIs129(Pslo-1::slo-1(gf); Pmyo-2::YFP); dyb-1(zw11)]; ZW624 [zwEx170(Pdyb-1::αDB1; Pmyo-2::DsRed); zwIs101(Pslo-1::slo-1::GFP); dyb-1(zw11)]; ZW625 [zwEx171(Pmyo-3::slo-1; Pmyo-2::DsRed); zwIs132(Pmyo-3::GCaMP2; lin-15(+)); slo-1(md1745)]; ZW626 [zwEx172(Prab-3::slo-1; Pmyo-2::DsRed); zwIs132(Pmyo-3::GCaMP2; lin-15(+)); slo-1(md1745)].

Mutant screen.

An integrated transgenic strain expressing Pslo-1::SLO-1(E³⁵⁰Q) and Pmyo-2::YFP in the wild-type genetic background was used for mutant screen. The Pslo-1::SLO-1(E³⁵⁰Q) transgene caused a lethargic phenotype, whereas the Pmyo-2::YFP transgene was used as a genetic marker. Synchronized L4-stage worms were treated with the chemical mutagen ethyl methanesulfonate (50 mm) for 4 h at room temperature. The F2 progeny were screened for animals that moved better than the original slo-1(gf) animals.

Behavioral assay.

Locomotion velocity and the head-bending angle were quantified using an automated worm tracking and analysis system (Chen et al., 2010a). Specifically, a single adult hermaphrodite was transferred to an agar plate without food. After allowing ∼30 s for recovery from the transfer, snapshots of the worm were taken at 15 frames/s for 30 s using a VGA FireWire camera (XCD-V60; Sony) mounted on a stereomicroscope (SMZ800; Nikon). The worm was constantly kept in the center of the view field with a motorized microscope stage (OptiScan ES111; Prior Scientific). Both the camera and the motorized stage were controlled by a custom program running in MATLAB (MathWorks).

Cloning of dyb-1.

dyb-1(zw11) was used for single nucleotide polymorphism-based genetic mapping (Davis et al., 2005). After mapping the mutation to a small interval, the candidate gene was identified by testing whether mutants of candidate genes would complement with zw11 in suppressing the lethargic phenotype of slo-1(gf). Full-length dyb-1 cDNA was obtained by RT-PCR. Molecular lesions of dyb-1 mutants were identified by sequencing the cDNA or genomic DNA of dyb-1 gene prepared from three randomly picked mutant alleles.

Analysis of expression pattern and subcellular localization.

The expression pattern of dyb-1 was determined by expressing GFP under the control of a 3.1-kb dyb-1 promoter (Pdyb-1::GFP, wp747). The plasmid was injected into the lin-15(n765) strain using a lin-15 rescue plasmid as a transformation marker. The transgenic animals were photographed using a Carl Zeiss Axiovert 200M fluorescence microscope (40× objective) with an apotome device (Carl Zeiss) for optical sectioning.

Subcellular localization of DYB-1 was determined by expressing DYB-1 with GFP tagged to its C terminus under the control of Pdyb-1 (Pdyb-1::DYB-1::GFP, wp1132). SLO-1 subcellular localization was examined with an integrated transgenic strain expressing Pslo-1::SLO-1::GFP (wp5) (Chen et al., 2010a). To determine whether DYB-1 and SLO-1 are colocalized, mStrawberry-tagged DYB-1 (Pdyb::DYB-1::mStrawberry, wp1130) was expressed in the transgenic strain expressing Pslo-1::SLO-1::GFP. To determine whether SLO-1 colocalized with UNC-2 in neurons, SLO-1::mStrawberry and UNC-2::GFP (Saheki and Bargmann, 2009) fusion proteins were expressed under the control of Punc-47 and Punc-25, respectively. Epifluorescence of the fusion proteins in transgenic animals was visualized and photographed with a Nikon TE2000-U inverted microscope (100× objective) and a Peltier cooled digital camera (F-view II; Olympus).

Recording of postsynaptic currents.

Evoked postsynaptic currents (ePSCs) and miniature postsynaptic currents (mPSCs) were recorded from the C. elegans neuromuscular junction (NMJ) using an established technique (Richmond et al., 1999; Liu et al., 2007). The recording pipette solution contained the following (in mm): 120 KCl, 20 KOH, 5 Tris, 0.25 CaCl₂, 4 MgCl₂, 36 sucrose, 5 EGTA, and 4 Na₂ATP, pH 7.2. Two external solutions with different [Ca²⁺] (5 and 0.5 mm) were used. The external solution with the higher [Ca²⁺] contained the following (in mm): 140 NaCl, 5 KCl, 5 CaCl₂, 5 MgCl₂, 11 dextrose, and 5 HEPES, pH 7.2. This solution was modified by reducing CaCl₂ to 0.5 mm and increasing NaCl to 145 mm to make the external solution with the lower [Ca²⁺].

Ca²⁺ imaging.

An integrated Pmyo-3::GCaMP2 transgene was crossed into different mutant backgrounds. Ca²⁺ transients were monitored by imaging GCaMP2 fluorescence in body-wall muscle cells of dissected worms as described previously (Liu et al., 2011).

Western blot.

To examine SLO-1::GFP protein level expressed in worms, worms of mixed stages were homogenized in a lysis buffer containing 2% SDS/100 mm NaCl/10% glycerol/50 mm Tris, pH 6.8. Soluble protein extracts were separated on 4–12% SDS-PAGE gels and probed with GFP (Invitrogen) and α-tubulin (Santa Cruz Biotechnology) antibodies. Anti-mouse IgG HRP (Santa Cruz Biotechnology) was used as the secondary antibody for detection by enhanced chemiluminescence (Pierce).

Data analysis.

The averaged locomotion speed and head-bending angle were determined using an automated worm tracking and analysis system (Chen et al., 2010a,b). Briefly, 13 marker points were placed at equal intervals along the spline of each worm, which was imaged at 15 frames/s, and the angle supplementary to the angle formed by the two straight lines connecting the marker points 1 and 2 and the marker points 2 and 3 was determined. Both the full-amplitude dorsal/ventral head-bending angle and the root mean square of head-bending angles were determined. The mean locomotion speed was calculated based on the distance traveled over time.

Amplitude and frequency of mPSCs were analyzed using MiniAnalysis (Synaptosoft). A detection threshold of 10 pA was used in initial automatic analysis, followed by visual inspections to include missed smaller events (5 pA or larger) and to exclude false events resulting from baseline fluctuations. For each recording, we adjusted the position of the stimulating electrode to obtain the largest ePSCs and used an average of the two largest values to derive an amplitude measurement for that recording.

For Ca²⁺ imaging data, all the muscle cells within the camera imaging field were chosen as separate regions of interest for quantification of fluorescence intensity changes over successive frames. Fluorescence intensity was first plotted as absolute signal intensity over time using the NIS-Elements software (Nikon) and then converted to F/F₀ using MATLAB (MathWorks) running a custom module. The frequency of Ca²⁺ transient peaks (defined as F/F₀ ≥ 0.05) was quantified for each cell. The mean frequency of all cells in the imaging field per preparation except for the slo-1(gf) group was treated as one sample. For the slo-1(gf) group, only data from cells with Ca²⁺ transients were included in the analysis.

Data graphing and statistical analyses.

Graphing and statistical analyses were performed with Origin (version 8.5; OriginLab). Either unpaired t test or one-way ANOVA (followed by Bonferroni's post hoc test) was used for statistical comparisons. p < 0.05 is considered statistically significant. All values are expressed as mean ± SEM.

Results

dyb-1 mutants suppressed the lethargic phenotype caused by slo-1(gf)

To identify proteins important to BK channel function in vivo, we performed a genetic screen for mutants that suppressed the lethargic phenotype of a worm strain expressing SLO-1(gf) under the control of slo-1 promoter (Pslo-1) (Chen et al., 2010b). SLO-1(gf) was created by mutating glutamate 350 to glutamine, which causes a 40 mV shift toward more hyperpolarized potentials in the SLO-1 conductance–voltage relationship (Chen et al., 2010b). Among 25 mutants isolated from a screen of 24,000 haploid genomes, seven mutants belonged to the dystrobrevin gene dyb-1 (locus F47G6.1, www.wormbase.org, WS225) based on complementation tests and rescue experiments. Although there are two dystrobrevin genes (α and β) with multiple splice variants in human (Peters et al., 1998; Holzfeind et al., 1999), the C. elegans genome has only one dystrobrevin gene with probably a single isoform (www.wormbase.org, WS225). DYB-1 shares 45.5 and 46.9% sequence identity with human αDB1 and βDB, respectively (Fig. 1). Three mutants were randomly chosen for sequencing, and all of them were found to be putative nulls (Fig. 1). Subsequent analyses were performed with the dyb-1(zw11) allele.

Figure 1.

dyb-1 encodes a homolog of mammalian dystrobrevins. Shown is the alignment of predicted amino acid sequences of DYB-1, human αDB1 (accession number NM_001390), and human βDB (accession number AF022728). Positions of molecular lesions of three dyb-1 alleles are indicated. Two alleles have mutations leading to premature stop codon (marked by *) and one allele (marked by an arrow) disrupts the splice acceptor site of the fourth intron, leading to the removal of exon 5 and a frame shift after Y171, which was shortly followed by a stop codon.

DYB-1 was coexpressed and colocalized with SLO-1 in both neurons and muscle cells

The isolation of dyb-1 mutants as suppressors of the slo-1(gf) lethargic phenotype suggests that DYB-1 might be required for SLO-1 function in vivo and have similar expression and subcellular localization patterns as SLO-1. Expression of GFP under the control of the dyb-1 promoter (Pdyb-1) resulted in GFP epifluorescence in many neurons and several muscles, including body-wall muscle (Fig. 2A). This expression pattern of dyb-1 is indistinguishable from that of slo-1 (Chen et al., 2010a,b). When mStrawberry-tagged full-length DYB-1 (DYB-1::mStrawberry) was coexpressed with GFP-tagged full-length SLO-1 (SLO-1::GFP) under the controls of their respective promoters, the two fusion proteins appeared as puncta with colocalization at muscle dense body regions and along the dorsal nerve cord (Fig. 2B,C). The two fusion proteins also showed overlapping expression in the nerve ring (Fig. 2D).

Figure 2.

DYB-1 and SLO-1 were coexpressed and colocalized in neurons and muscle cells. A, Expression of GFP under the control of Pdyb-1 resulted in GFP epifluorescence in many neurons in the nerve ring (NR), ventral cord (VC), and tail (not labeled), as well as several types of muscles, including body-wall muscle (BM), vulval muscle (VM), stomatointestinal muscle (SIM), and anal depressor muscle (ADM). B–D, SLO-1::GFP and DYB-1::mStrawberry colocalized in body-wall muscle cells (B), the dorsal nerve cord (C), and the nerve ring (D) when they were expressed under the control of their native promoters. E, SLO-1::mStrawberry and UNC-2::GFP, which were expressed in GABAergic neurons, colocalized in the ventral nerve cord. * indicates the soma of a GABAergic motoneuron. Scale bars, 10 μm.

To determine whether the SLO-1::GFP puncta in the dorsal nerve cord corresponded to synaptic sites, we coexpressed mStrawberry-tagged SLO-1 (SLO-1::mStrawberry) and GFP-tagged UNC-2 (UNC-2::GFP) in GABAergic neurons under the controls of the unc-47 (vesicular GABA transporter) and unc-25 (glutamic acid decarboxylase) promoters, respectively, and analyzed their localization in the ventral nerve cord (SLO-1::mStrawberry signal in the dorsal cord was too weak to be imaged). UNC-2 is an N/P/Q-type (Ca_V2) voltage-gated Ca²⁺ channel (Bargmann, 1998) and is localized to presynaptic sites in motoneurons (Saheki and Bargmann, 2009). We found that the two fusion proteins colocalized in the ventral nerve cord (Fig. 2E). The separate observations of SLO-1 colocalization with DYB-1 and UNC-2 at the nerve cords collectively suggest that SLO-1 and DYB-1 colocalized at presynaptic sites in motoneurons.

DYB-1 was required for SLO-1 function in both neurons and muscle cells

To determine whether DYB-1 contributes to SLO-1 function in neurons, muscles, or both, we expressed wild-type DYB-1 in the slo-1(gf);dyb-1 double mutant using either the muscle-specific myo-3 promoter (Pmyo-3) or the pan-neuronal rab-3 promoter (Prab-3) and analyzed its effects on animal behaviors. Expression of wild-type DYB-1 in either neurons or muscle cells primarily reinstated the locomotion and egg-laying defects of slo-1(gf) (Fig. 3A,B), suggesting that DYB-1 is required for SLO-1 function in both neurons and muscle cells. In addition, we analyzed a head-bending phenotype associated with slo-1(lf) mutant (Kim et al., 2009; Chen et al., 2010a,b). The head-bending angle was increased to a similar degree in the dyb-1 mutant, slo-1(lf), and the double-mutant, slo-1(lf);dyb-1 (Fig. 3C,D). The head-bending phenotype of the dyb-1 mutant could be rescued by expressing wild-type DYB-1 in muscle cells but not neurons (Fig. 3C,D), suggesting that DYB-1 functions together with SLO-1 in muscle to regulate the head-bending angle.

Figure 3.

dyb-1 mutant suppressed locomotion and egg-laying phenotypes of slo-1(gf) and phenocopied slo-1(lf) in head-bending behavior. A, B, dyb-1 mutant suppressed the locomotion (A) and egg-laying (B) defects caused by slo-1(gf), which could be reversed by expressing wild-type DYB-1 in either neurons (Prab-3) or body-wall muscle cells (Pmyo-3). Twenty to 30 worms per group in A and 10 worms per group in B. C, D, Comparisons of full-amplitude head-bending angle and root mean square (RMS) head-bending angle. The head-bending angle was similarly increased in dyb-1 mutant, slo-1(lf), and the double mutant. This phenotype of dyb-1 mutant could be rescued by expressing wild-type DYB-1 in muscle but not neurons. Ten worms per group. *p < 0.01, significantly different between the indicated groups (A, B) or from wild type (C, D) (one-way ANOVA with Bonferroni's post hoc tests).

DYB-1 was required by SLO-1 to regulate neurotransmitter release

SLO-1 is a potent negative regulator of neurotransmitter release at the C. elegans NMJ (Wang et al., 2001; Liu et al., 2007). To determine whether DYB-1 is required for this function of SLO-1, we analyzed the effect of dyb-1 on ePSCs at the NMJ at two different extracellular Ca²⁺ concentrations (5 and 0.5 mm). The higher [Ca²⁺] is more suitable for determining whether dyb-1 may reverse an inhibitory effect of slo-1(gf) on neurotransmitter release, whereas the lower [Ca²⁺] is more suitable for testing whether dyb-1 could increase ePSC amplitude as slo-1(lf) does (Liu et al., 2007; Chen et al., 2010b). At 5 mm [Ca²⁺]_o, the inhibitory effect of slo-1(gf) on the ePSC amplitude was reversed by dyb-1, whereas dyb-1 alone did not increase ePSC amplitude (Fig. 4A), which resembles slo-1(lf) and is probably attributable to the limited capacity of the readily releasable pool of synaptic vesicles (Wang et al., 2001). At 0.5 mm [Ca²⁺]_o, ePSC amplitude was increased to a similar degree in dyb-1 and slo-1(lf) as well as in the slo-1(lf);dyb-1 double mutant (Fig. 4B). These observations suggest that DYB-1 is required by SLO-1 to control neurotransmitter release.

Figure 4.

DYB-1 was required by SLO-1 to control neurotransmitter release at NMJs. A, dyb-1 mutant counteracted the inhibitory effect of slo-1(gf) on ePSCs at 5 mm [Ca²⁺]_o. B, The ePSC amplitude of dyb-1 mutant was significantly larger than that of the wild type but similar to that of slo-1(lf) or the slo-1(lf);dyb-1 double mutant. C, The frequency of mPSCs was increased, whereas the mean amplitude did not change in slo-1(lf), dyb-1, and slo-1(lf);dyb-1 compared with the wild type. D, Responses of body-wall muscle cells to exogenous ACh (100 μm) and GABA (100 μm) were unaltered in the mutants compared with the wild type. The holding potential for all recordings was −60 mV. *p < 0.05, significantly different from the wild type (one-way ANOVA with Bonferroni's post hoc test). The number of samples analyzed is indicated inside each column.

To test whether the abnormal ePSCs in slo-1 and dyb-1 mutants resulted from altered sensitivity of postsynaptic receptors, we recorded the mPSCs at 0.5 mm [Ca²⁺]_o and compared the frequency and amplitude of mPSCs between the wild-type and the mutants. We found that the frequency of mPSCs was increased, whereas the amplitude of mPSCs did not change in slo-1(lf), dyb-1, and slo-1(lf);dyb-1 (Fig. 4C). We also examined the sensitivities of postsynaptic acetylcholine (ACh) and GABA receptors by puffing exogenous ACh and GABA directly onto the muscle cells. We found that the amplitude of ACh- and GABA-induced currents was similar between wild-type worms and the mutants (Fig. 4D). These observations suggest that SLO-1 and DYB-1 regulated neuromuscular transmission through a presynaptic mechanism.

SLO-1 and DYB-1 controlled Ca²⁺ transients in body-wall muscle cells

Because SLO-1 colocalizes with the L-type voltage-gated Ca²⁺ channel EGL-19 (Kim et al., 2009) and EGL-19 is the predominant voltage-gated Ca²⁺ channel in body-wall muscle (Jospin et al., 2002), SLO-1 could potentially be activated by Ca²⁺ entering through EGL-19 and, in turn, regulate intracellular [Ca²⁺]. To examine this possibility, we expressed the calcium indicator GCaMP2 in body-wall muscle under the control of Pmyo-3 and imaged Ca²⁺ transients as described previously (Liu et al., 2011). Ca²⁺ transients were observed in all muscle cells examined in wild-type and slo-1(lf) worms, with the frequency being much higher in the slo-1(lf) mutant than the wild type (Fig. 5). In slo-1(gf) worms, however, Ca²⁺ transients were detected in only approximately one-third of the preps with reduced frequency (Fig. 5). These observations suggest that a physiological function of SLO-1 is to inhibit muscle Ca²⁺ transients. Expression of wild-type SLO-1 in either muscle cells or neurons reversed the effect of slo-1(lf) on Ca²⁺ transients (Fig. 5), suggesting that SLO-1 regulates muscle Ca²⁺ transients through both presynaptic and postsynaptic effects.

Figure 5.

The frequency of Ca²⁺ transients in body-wall muscle cells was altered by mutations of slo-1 and dyb-1. A, Representative traces of Ca²⁺ transients from wild-type worms and various mutants. B, Quantification of the frequency of Ca²⁺ transients. The frequency of muscle Ca²⁺ transients was increased in slo-1(lf), which was restored by expressing wild-type SLO-1 in either muscles (Pmyo-3::SLO-1) or neurons (Prab-3::SLO-1). The frequency of Ca²⁺ transients was decreased in slo-1(gf), which was reversed by dyb-1 mutant. dyb-1 mutant showed a similar increase in the frequency of Ca²⁺ transients as slo-1(lf) did. *p < 0.05, significantly different from the wild type (one-way ANOVA with Bonferroni's post hoc test). The number of samples analyzed is indicated inside each column.

To determine whether DYB-1 also regulates muscle Ca²⁺ transients by affecting SLO-1, we analyzed Ca²⁺ transients in the dyb-1 mutant and slo-1(gf);dyb-1 double mutant. The frequency of muscle Ca²⁺ transients was increased to a similar degree in the dyb-1 mutant as in the slo-1(lf) mutant; furthermore, the dyb-1 mutant counteracted the effect of slo-1(gf) on muscle Ca²⁺ transients (Fig. 5). These observations suggest that DYB-1 is required by SLO-1 to regulate Ca²⁺ transients.

DYB-1 was required for SLO-1 subcellular localization in neurons and body-wall muscle cells

DYB-1 could contribute to SLO-1 function through several potential mechanisms. We found that DYB-1 did not affect channel functional properties when it was coexpressed with SLO-1 in Xenopus oocytes (data not shown). In contrast, SLO-1::GFP puncta appeared diffuse and dim in the ventral and dorsal nerve cords (Fig. 6A,E) and were essentially absent in body-wall muscle cells (Fig. 6B) of the dyb-1 mutant. The apparent SLO-1::GFP mislocalization did not result from a gross destabilization of presynaptic sites or muscle-dense bodies because the presynaptic marker RIM (Koushika et al., 2001) (Fig. 6C,E) and the dense body marker vinculin (Francis and Waterston, 1985) (Fig. 6D) were normally localized in the dyb-1 mutant, nor did it result from a decrease of SLO-1::GFP protein level, which was similar between the wild-type and dyb-1 mutant (Fig. 6F). These observations suggest that DYB-1 is required for proper SLO-1 subcellular localization in neurons and muscle cells and that the observed behavioral, synaptic, and muscle Ca²⁺ transient phenotypes of the dyb-1 mutant were likely caused by SLO-1 mislocalization in these cells.

Figure 6.

SLO-1 was mislocalized in both body-wall muscle cells and neurons of dyb-1 mutant. A, SLO-1::GFP in the dorsal cord and ventral cord appeared as puncta in the wild type but was diffusely distributed in dyb-1 mutant. B, SLO-1::GFP in muscle was localized to dense body regions in the wild type but undetectable in dyb-1 mutant. C, D, The presynaptic marker RIM (C) and dense body marker vinculin (D) were normally localized in dyb-1 mutant. E, Quantification of the density of dorsal cord puncta shown in A and C. *p < 0.0001, significantly different from the wild type (unpaired t test). F, Western blot shows that the level of SLO-1::GFP total protein was comparable between the wild-type and dyb-1 mutant. α-Tubulin was blotted to show equal loading of the protein samples. Scale bars, 10 μm.

DYB-1 subcellular localization did not depend on dystrophin

Dystrophin has been implicated in controlling dystrobrevin expression in mouse (Ohlendieck and Campbell, 1991) and SLO-1 subcellular localization in C. elegans (Kim et al., 2009; Chen et al., 2010a). We found that SLO-1::GFP was mislocalized in body-wall muscle but normally localized in the dorsal nerve cord in the dystrophin null mutant dys-1(cx18) (Fig. 7A,B,E). However, DYB-1 expression and subcellular localization were unaltered in both muscle cells and neurons of the dys-1 mutant (Fig. 7C–E). These observations suggest that in neurons DYB-1 localizes SLO-1 through a mechanism that is independent of DYS-1.

Figure 7.

Effects of dys-1 mutant on SLO-1 and DYB-1 subcellular localization. A, B, SLO-1::GFP was mislocalized in body-wall muscle but not neurons of the dystrophin mutant dys-1(cx18). C, D, DYB-1::GFP was normally localized in both muscle cells and neurons of dys-1 mutant. E, Quantification of SLO-1::GFP and DYB-1::GFP puncta density in the dorsal nerve cord. Scale bars, 10 μm.

Mouse αDB1 partially rescued dyb-1 mutant phenotype

The relatively high level of sequence homology between DYB-1 and mammalian dystrobrevins (Fig. 1) raised the possibility that a mammalian dystrobrevin may be able to substitute DYB-1 with respect to SLO-1 function in vivo. Therefore, we analyzed the effect of mouse αDB1 on various phenotypes of the dyb-1 mutant. Targeted expression of αDB1 in neurons but not muscle cells of the slo-1(gf);dyb-1 double mutant essentially reinstated behavioral phenotypes of slo-1(gf), including reduced locomotion speed and increased egg retention in the uterus (Fig. 8A,B). Expression of αDB1 in slo-1(gf);dyb-1 double mutant also essentially restored the small ePSC amplitude of slo-1(gf) (Fig. 8C). Furthermore, expression of αDB1 under the control of Pdyb-1 essentially rescued dyb-1-induced SLO-1::GFP mislocalization in the dorsal cord but had no effect in body-wall muscle (Fig. 8D,E). The differential rescuing effects of mouse αDB1 in neurons and muscle cells suggest that DYB-1 likely mediates SLO-1 localization by interacting with different proteins in neurons and muscle cells and that αDB1 may interact with the related protein(s) in neurons but not muscle cells.

Figure 8.

Mouse αDB1 partially rescued the neuronal phenotypes of dyb-1 mutant. A, B, Expression of αDB1 in neurons (Prab-3::αDB1) but not muscles (Pmyo-3::αDB1) partially reinstated the defective locomotion (A) and egg-laying (B) phenotypes of slo-1(gf) in slo-1(gf);dyb-1 mutant. The datasets for slo-1(gf) and slo-1(gf);dyb-1 in Figures 3 were replotted here for comparison. C, Expression of αDB1 in slo-1(gf);dyb-1 mutant under the control of Pdyb-1 essentially restored the small ePSC amplitude of slo-1(gf). The datasets for slo-1(gf) and slo-1(gf);dyb-1 in Figure 4 were replotted here for comparison. The holding potential used for recording ePSCs was −60 mV. D, E, Expression of αDB1 under the control of Pdyb-1 partially rescued SLO-1::GFP localization in the dorsal nerve cord but did not have an effect in body-wall muscle. *p < 0.05, significantly different between the indicated groups (one-way ANOVA with Bonferroni's post hoc test). Scale bars, 10 μm.

Discussion

The present study shows that DYB-1 plays pivotal roles in controlling muscle Ca²⁺ transients and synaptic exocytosis in C. elegans by localizing the BK channel to subcellular domains containing a high density of voltage-gated Ca²⁺ channels. These observations are consistent with the previous reports that dystrobrevin is localized to axon terminals (Ueda et al., 2000) and that mutations of either dystrobrevin or the BK channel may contribute to synaptic defects and muscular dystrophy (Grady et al., 1999; Wang et al., 2001; Carre-Pierrat et al., 2006; Grady et al., 2006).

Ca²⁺ mishandling has been suggested as the basis of muscle degeneration in Duchenne muscular dystrophy (DMD) patients (Duncan, 1978). It is thought that sarcolemma damage or enhanced activity of Ca²⁺ leak channels leads to elevation of intracellular [Ca²⁺], which activates Ca²⁺-dependent proteases such as calpains and eventually leads to cell death (Deconinck and Dan, 2007; Allen et al., 2010). However, measurements of Ca²⁺ concentration in skeletal muscle from DMD patients and mdx mice have produced controversial results (Constantin et al., 2006). Through Ca²⁺ imaging in live worms, we found that SLO-1 is a potent regulator of muscle Ca²⁺ transients in C. elegans body-wall muscle and that its mislocalization in dyb-1 mutant led to a large increase in Ca²⁺ transient frequency. Given that SLO-1 colocalizes with the L-type voltage-gated Ca²⁺ channel EGL-19 at dense body regions in body-wall muscle (Kim et al., 2009), Ca²⁺ microdomains formed at the inner mouth of EGL-19 in response to action potentials likely activate colocalized SLO-1, which in turn terminates further Ca²⁺ entry by facilitating membrane repolarization. This function of SLO-1 in regulating muscle Ca²⁺ transients might be an important contributing factor to the muscular degenerative changes related to slo-1, dyb-1, and dys-1 mutants.

Contraction of mammalian skeletal muscle requires the functions of both L-type voltage-gated Ca²⁺ channels in the sarcolemma and ryanodine receptors (RyRs) in the sarcoplasmic reticulum (SR) membrane (Rios and Brum, 1987; Takeshima et al., 1994), which is similar to C. elegans body-wall muscle (Liu et al., 2011). Although RyR-mediated Ca²⁺ release from the SR serves as the primary source of Ca²⁺ for skeletal muscle contraction, the activation of RyRs is coupled to depolarization-induced conformational changes of voltage-gated Ca²⁺ channels through protein/protein interactions occurring in t-tubules (Ríos et al., 1992; Schneider, 1994). Because the BK channel is also enriched or localized in t-tubules (Latorre et al., 1982; Knaus et al., 1995), whether or not it colocalizes with voltage-gated Ca²⁺ channels to regulate Ca²⁺ concentrations in mammalian skeletal muscle is worthy of investigation.

The similar locomotion and muscle Ca²⁺ transient phenotypes of dyb-1 and slo-1 mutants suggest that DYB-1 likely regulates muscle Ca²⁺ transients mainly through SLO-1. Nevertheless, we cannot exclude the potential involvement of other mechanisms. For example, Ca²⁺ mishandling in dystrophic mammalian skeletal muscle has been attributed to weakened membrane integrity and increased opening of other ion channels (e.g., store-operated cation channels and inositol triphosphate receptors) (Constantin et al., 2006). It would be interesting to determine whether these kinds of mechanisms are also implicated in the regulatory effect of DYB-1 on worm muscle Ca²⁺ transients.

The DAPC has also been implicated in synaptic transmission. For example, αDB1 is required for ACh receptor clustering at postsynaptic sites in mouse skeletal muscle (Pawlikowski and Maimone, 2008). In C. elegans, mutations of a DAPC component (DYS-1, DYB-1, or STN-1/syntrophin) result in increased cholinergic transmission as suggested by behavioral and pharmacological assays (Gieseler et al., 1999; Giugia et al., 1999). The present study shows that dystrobrevin controls neurotransmitter release by localizing SLO-1 to presynaptic sites, which potentially allows SLO-1 to be functionally coupled with UNC-2, the predominant voltage-gated Ca²⁺ channel at the C. elegans NMJ (Richmond et al., 2001; Saheki and Bargmann, 2009). Given that mouse αDB1 is also localized to axon terminals in mammals (Ueda et al., 2000) and may partially substitute DYB-1 in neurons with respect to SLO-1 function in C. elegans, it would be interesting to see whether dystrobrevins play a similar role in mammalian brain.

Dystrobrevins are mainly known to function as a component in the DAPC. However, there is evidence that dystrobrevins may also function independently of dystrophin. In mice lacking dystrophin, the assembly of the DAPC is disrupted in skeletal muscle but the dystrobrevin–syntrophin complexes are still formed in the brains (Blake et al., 1999). Although DYB-1 was required for SLO-1 localization in neurons as well as body-wall muscle cells, DYS-1 was dispensable for normal SLO-1 localization in neurons, suggesting that, at least in neurons, DYB-1 localizes SLO-1 through a mechanism independent of DYS-1. This conclusion is consistent with the data showing that mouse αDB1 was able to partially rescue SLO-1 mislocalization in neurons but not muscle cells in dyb-1 mutant. It remains to be determined what differences exist in the protein partners interacting with DYB-1 to mediate SLO-1 localization in neurons and body-wall muscle cells.

αDB and βDB are widely expressed in mammalian tissues. It has been suggested that mutations of dystrobrevins may underlie some behavioral and cognitive defects (Rees et al., 2007). However, relatively little is known about the functions of dystrobrevins. Patients with αDB deficiency may display severe congenital muscular dystrophy with ophthalmoplegia (Jones et al., 2003). Studies with mice have shown that knock-out of αDB causes mild muscular dystrophy and cardiomyopathy, abnormal distribution and reduced level of ACh receptors at the NMJs, and reduced level of neuronal nitric oxide synthase at the sarcolemma, and knock-out of βDB causes reduction in the number of GABA_Aα1 receptors in cerebellar Purkinje cells (Rees et al., 2007). Our analyses with C. elegans revealed new physiological functions of dystrobrevins and raised an interesting question as to whether dystrobrevin-dependent BK channel localization is involved in DAPC-associated diseases.