The V-ATPase is an ATP-dependent proton pump that couples the energy released upon hydrolysis of ATP to the active transport of protons from the cytoplasm into the lumen of intracellular compartments or, if localized to the plasma membrane, from the cytoplasm to the extracellular environment. Proton transport mediated by V-ATPase participates in the acidification of intracellular organelles, and thus plays a crucial role in normal cellular processes such as receptor-mediated endocytosis, intracellular membrane trafficking, protein degradation, and neurotransmitter uptake (Forgac,2007; Jefferies et al.,2008). In addition to regulation of the pH of intracellular organelles, V-ATPase at the plasma membrane carries out cell-specific functions such as renal acidification (Karet et al.,1999), bone resorption (Toyomura et al.,2003), homeostasis of cytoplasmic pH, and sperm maturation (Pietrement et al.,2006). V-ATPase is a large multisubunit complex that is organized into two domains that operate by a rotary mechanism. The V1 domain carries out ATP hydrolysis and is composed of eight different subunits that are located on the cytoplasmic side of the membrane. The V0 domain is a membrane-embedded complex that contains six different subunits and is responsible for the translocation of protons from the cytoplasm to the lumen or extracellular space (Forgac,2007).
In addition to its role in pH homeostasis, the V0 domain of V-ATPase has been reported to directly participate in the process of membrane fusion independent of its role in acidification. Disruption of the gene that encodes the neuron-specific isoform of the a-subunit of the V0 domain in Drosophila melanogaster blocked synaptic vesicle fusion with the presynaptic membrane (Hiesinger et al.,2005). In Caenorhabditis elegans, mutations in the a-subunit of the V0 domain blocked apical secretion of Hedgehog-related cuticle proteins by epidermal cells (Liegeois et al.,2006). Mice that lack the a3 isoform exhibited defective insulin secretion in pancreatic β-cells (Sun-Wada, et al.,2006). A recent study in zebrafish also demonstrated that the a1-subunit of V-ATPase plays a key role in mediating the fusion between phagosomes and lysosomes in microglia within the brain (Peri and Nusslein-Volhard,2008).
ATP6V0C is the proteolipid c-subunit of the V0 domain, and is a membrane-spanning protein that folds into four trans-membrane helices and assembles into a hexamer to form the membrane proton channel of the V-ATPase enzyme (Hanada et al.,1991; Wilkens and Forgac,2001). In addition to its role as a V-ATPase subunit, ATP6V0C functions independently to form gap junction complexes and neurotransmitter release channels (Finbow et al.,1995; Morel,2003). Recent study has shown that mutations in the apt6v0c gene result in defects in retinoblast proliferation and survival in zebrafish, indicating that ATP6V0C plays a critical role during vertebrate eye development and maintenance (Wang et al.,2008; Nuckels et al.,2009). Here we present the specific expression pattern and expected function of atp6v0c2, a second isoform of mammalian atp6v0c, in zebrafish. The results of this study demonstrate that unlike atp6v0c, atp6v0c2 is expressed in CNS neurons, and changes in neuronal differentiation by manipulation of Notch activity result in alteration of atp6v0c2 expression in the CNS. The ATP6V0C2 protein is co-localized with the synaptic vesicle protein 2 (SV2), a protein present in presynaptic vesicles, suggesting that ATP6VOC2 is involved in neurotransmitter storage and/or release from the synaptic vesicle in neurons. In addition, the loss-of-function study suggests that ATP6V0C2 is involved in the control of neuronal excitability.
atp6v0c2 Is Specifically Expressed in the Zebrafish CNS
Previously, we performed a genome-wide analysis of the zebrafish mibta52b mutant to identify genes that were differentially regulated by Notch signaling using the serial analysis of gene expression (SAGE) procedure (Hwang et al.,2009). From this study, the zgc:77708 clone was identified as highly up-regulated in the mibta52b mutant and shown to be expressed in the developing nervous system in zebrafish (Hwang et al.,2009).
We next identified zebrafish-expressed sequence tags (ESTs) with sequence similarity to zgc:77708 sequences, and found that zgc:77708 encodes an ATP6V0C-like protein (BC065849) that is highly homologous to zebrafish ATP6V0C (BC152275). We designated this gene atp6v0c2. A comparison of primary protein structures indicated that the deduced protein sequences of atp6v0c and atp6v0c2 were highly homologous, and both clones also showed high sequence similarity with other vertebrate atp6v0c sequences, including human (Fig. 1).
In order to investigate the specific role of ATP6V0C2 in zebrafish development, we examined its expression in developing zebrafish embryos. In contrast to the expression pattern of atp6v0c, which has been shown to be expressed in the retinal pigment epithelium and ganglion cells in the eye, the telencephalon, and the pigment cells and mucous cells of the fish body (Wang et al.,2008; Nuckels et al.,2009), atp6v0c2 is expressed specifically in the developing nervous system (Fig. 2). Expression of atp6v0c2 was initially detected maternally, and ubiquitous zygotic expression was detected at the 30%-epiboly stage (Fig. 2A, B). At 11 hpf (hours post fertilization), atp6v0c2 expression was not detectable anywhere in the embryo (Fig. 2C), but by 14 hpf, atp6v0c2 expression was detected in the presumed primary neurons positioned bilaterally along the anteroposterior axis of the neural tube (Fig. 2D, E). Expression of atp6v0c2 in the central nervous system was more prominent in the brain and spinal cord of 18-hpf embryos (Fig. 2F). By 24 hpf, atp6v0c2 expression was apparent in the telencephalic cluster (tc), epiphysial cluster (ec), nucleus of the tract of the post optic commissure (nTPOC), the nucleus of the medial longitudinal fasciculus (nMLF) in the fore- and midbrain, and spinal cord (sc) (Fig. 2G). In addition, all cranial ganglions, including the trigeminal ganglion and seven distinct rhombomeres in the hindbrain (Figs. 2G, 3F), also exhibited expression of atp6v0c2. To investigate whether atp6v0c2-expressing cells were proliferating precursors, we treated embryos at 24 hpf with the thymidine analog, BrdU, and labeled the embryos with an anti-BrdU antibody. We then examined atp6v0c2 expression by in situ hybridization. The majority of the atp6v0c2-expressing cells were BrdU− and located at the lateral margin of the spinal cord, indicating that they were not proliferating precursors in the nervous system (Fig. 2H). Consistent with this data, in situ hybridization of 24-hpf embryos with an atp6v0c2 RNA probe and subsequent labeling with anti-Hu antibody, which is a marker of post-mitotic neurons, revealed that atp6v0c2-expressing cells are Hu+ post-mitotic neurons (Fig. 2I).
Because atp6v0c2 showed neuron-specific expression in the zebrafish CNS, we next compared the temporal and spatial expression patterns of atp6v0c2 with those of HuC, which is a post-mitotic pan-neuronal marker in zebrafish (Kim et al.,1997; Park et al.,2000a). By the 3-somite stage (11 hpf), expression of HuC was detected in the primary neurons, including Rohon-Beard cells (Fig. 3A, arrowheads) and primary motor neurons (Fig. 3A, arrow) in the spinal cord, as has been reported previously (Kim et al.,1997). However, atp6v0c2 mRNA transcripts were not detected in the 3-somite stage embryos (Fig. 3B), and were not expressed in the spinal cord until 14 hpf (Fig. 2D). By 18 hpf, there was a high level of HuC expression in the cranial ganglions extending from the hindbrain rhombomeres (Fig. 3C, arrowheads), whereas a relatively weak level of atp6v0c2 mRNA transcript expression was detected in the same region (Fig. 3D, arrowheads). By 24 hpf, similar levels of HuC and atp6v0c2 expression were detected in all cranial ganglions, including the trigeminal ganglion and seven distinct rhombomeres in the hindbrain. However, the number of neuronal cells expressing ATP6V0C was substantially fewer than those expressing HuC in the hindbrain cranial ganglions (Fig. 3E, F, arrowheads). In situ hybridization of 24-hpf embryos with an atp6v0c2 RNA probe and subsequent labeling with anti-Hu antibody revealed that atp6v0c2 is expressed in a subset of Hu+ post-mitotic neurons in the trigeminal ganglion of the developing brain (Fig. 3G, H). However, in situ hybridization in 3-dpf embryos with atp6v0c2 and HuC RNA probes revealed that atp6v0c2 is eventually expressed in the majority of neurons (Fig. 3I, J).
We previously demonstrated that Notch signaling is required to limit the formation of early-born primary neurons and maintain proliferative neural precursors (Park and Appel,2003). Because atp6v0c2 is expressed specifically in neurons, we speculated that its expression is regulated by Notch signaling. To investigate the regulation of atp6v0c2 expression by Notch signaling, we compared the distribution of atp6v0c2 transcripts in embryos with different levels of Notch activity. In wild-type embryos, atp6v0c2 transcripts were detected in a very limited region of the brain and spinal cord (Fig. 4A), and in the lateral margin of the spinal cord (Fig. 4B). In mutant embryos homozygous for the mind bomb(mib), a protein that encodes a ubiquitin ligase necessary for efficient Notch signaling (Itoh et al.,2003), the populations of atp6v0c2-expressing cells were substantially enlarged in the brain and spinal cord (Fig. 4C, D). In the spinal cord of these embryos, atp6v0c2-expressing cells were detected in positions normally occupied by proliferative cells, and the majority of the spinal cord cells were atp6v0c2+(Fig. 4D). In contrast, forced expression of a constitutively active form of Notch1a (Notch1aac; Scheer and Campos-Ortega,1999; Scheer et al.,2001) just before the beginning of neurogenesis effectively inhibited atp6v0c2 expression in the brain and spinal cord (Fig. 4E, F). Altogether, these data suggest that changes in neuronal differentiation by manipulation of Notch activity results in alteration of atp6v0c2 expression in the zebrafish CNS.
ATP6V0C2 Function Is Dispensable for Neurogenesis
A recent study reported that apt6v0c mutant zebrafish were microphthalmic, and their eyes possessed a thin, hypopigmented retinal pigmented epithelium (RPE) (Nuckels et al.,2009). BrdU incorporation experiments and TUNEL staining revealed that apt6v0c function was essential for the exit of retinoblasts from the cell cycle, as well as for the sustained proliferation of retinal stem cells (Nuckels et al.,2009). In order to determine whether atp6v0c2 has a similar function in neurogenesis during CNS development, we designed an antisense morpholino oligonucleotide (MO) that would interfere with the splicing of apt6v0c2 RNA and generate alternative transcripts that were expected to be nonfunctional (see Supp. Fig. 1A, which is available online). Injection of atp6v0c2 MOs led to a specific reduction in the synthesis of normal length atp6v0c2 RNA transcripts (Supp. Fig. 1B). Because atp6v0c2 MOs interfere with splicing between exon2 and exon3 and generate an mRNA split by intron2, whole mount RNA in situ hybridization did not detect any atp6v0c2 expression in the embryos injected with atp6v0c2 MOs (Supp. Fig. 1C).
We next examined neurogenesis in embryos injected with atp6v0c2 MOs and synthetic atp6v0c2 mRNAs, respectively, by whole mount in situ RNA hybridization with HuC (Fig. 5A–C). In contrast to apt6v0c, the loss-of- or gain-of-atp6v0c2 function did not affect neuronal differentiation. Embryos injected with atp6v0c2 MOs and synthetic apt6v0c2 mRNA both exhibited normal expression of HuC(Fig. 5A–C). Consistent with these data, the number of Isl+ sensory neurons (Fig. 5D, E) and Znp-1+ motor axons (Fig. 5F, G, arrows) were not affected in the spinal cords of atp6v0c2 MO-injected embryos. Altogether, these data suggest that ATP6V0C2 might have a function distinct from that of ATP6V0C in the development of the CNS.
V-ATPase has been reported to function in neurotransmitter storage in synaptic vesicle in neurons (Yamagata and Parsons,1989). In addition, the V0 domain of V-ATPase has been shown to play a direct role in synaptic vesicle fusion, which is required for the late step of synaptic vesicle exocytosis (Hiesinger et al.,2005). Therefore, we next examined the cellular localization of ATP6V0C2 in neurons to explore whether its function is related to neurotransmitter storage and/or synaptic vesicle exocytosis by membrane fusion. To accomplish this, we generated a DNA construct that expressed atp6v0c2 fused to red fluorescent protein (RFP) under the control of the neuron-specific HuC promoter (Park et al.,2000b) and injected it into zebrafish embryos at the one-cell stage. The injected embryos were then labeled with an antibody against synaptic vesicle protein 2 (SV2), a protein present in presynaptic vesicles (Buckley and Kelly,1985). In embryos injected with the HuC:apt6v0c2-rfp DNA construct, we observed that the neuronal cells expressed the ATP6V0C2-RFP fusion protein in their cell bodies and axons (Fig. 6A, B). At high magnification, we found that SV2 and RFP were co-localized in the axons (Fig. 6C, D, arrowheads), suggesting that ATP6V0C2 is located in the membranes of presynaptic vesicles. Altogether, these data suggest that ATP6V0C2 function is dispensable for neurogenesis but may be important for proper neurotransmitter storage and/or synaptic vesicle exocytosis.
ATP6V0C2 Function Is Required for the Proper Function of Neurons
Because the results of the loss-of-function experiment with atp6v0c2 MO indicated that ATP6V0C2 function is dispensable for neuronal differentiation, we hypothesized that ATP6V0C2 is required for the post-developmental function of neurons, excitability of the neuronal cells after they are differentiated. In order to investigate whether the neurons from the morphant retain excitability, we dissociated cells from 2 dpf Tg (HuC:egfp) embryos (Park et al.,2000b) injected with atp6v0c2 MO, and performed Ca2+ imaging of the dissociated EGFP+ neurons with high KCl-induced depolarization stimulation. Under normal conditions, high levels of extracellular K+ depolarize excitable cells, leading to the activation of voltage-gated Ca2+ channels in the plasma membrane and subsequent intracellular Ca2+ increases. As shown in Figure 7, neurons from wild-type Tg(HuC:egfp) embryos showed robust increases in intracellular Ca2+ upon repeated extracellular application of 100 mM KCl (Fig. 7A, C). Conversely, depolarization-evoked Ca2+ influx was completely inhibited in neurons from Tg(HuC:egfp) embryos injected with atp6v0c2 MO (Fig. 7B, C). These data suggest that ATP6V0C2 function is required for neuronal excitability.
In neurons, V-ATPases are present in the membranes of synaptic vesicles, which are the organelles in which neurotransmitters are stored, and the proton gradient generated inside the synaptic vesicles by these enzymes is used to accumulate neurotransmitters (Yamagata and Parsons,1989). In addition to its function in acidification, the V0 domain of V-ATPase plays an important role in neurotransmitter release by forming the fusion pore necessary for membrane fusion and neurotransmitter release (Morel,2003; Hiesinger et al.,2005). Among the proteins that contain the V0 domain, ATP6V0C, which is a membrane-spanning protein that forms the membrane proton channel of the V-ATPase, functions to form gap junction complexes and neurotransmitter release channels (Finbow et al.,1995).
Here we report the expression pattern of atp6v0c2, the c-subunit of the V-ATPase V0 complex in zebrafish. atp6v0c2 was cloned as an up-regulated gene identified in a genome-wide analysis of the mib mutant to search for genes that are differentially regulated by Notch signaling in neurons (Hwang et al.,2009). Up-regulation of atp6v0c2 suggests that its expression is regulated by Notch signaling. We postulated that atp6v0c2 plays an important role in neurotransmitter storage and/or release from the synaptic vesicle in neurons on the basis of three observations. First, unlike atp6v0c, which is expressed in retinal pigment epithelium and ganglion cells in the eye, the telencephalon, and in pigment cells and mucous cells in zebrafish (Wang et al.,2008; Nuckels et al.,2009), atp6v0c2 is specifically expressed in post-mitotic neurons in the developing zebrafish CNS. Second, whereas the onset of HuC expression in the primary motor neurons and Rohon-Beard cells (primary sensory neurons) occurs at the 3-somite stage (11 hpf), atp6v0c2 mRNA transcripts were first detected in the spinal cord of zebrafish embryos at the 10-somite stage. By 24 hpf, both HuC and atp6v0c2 were expressed at similar levels in all the cranial ganglions, but only a subset of HuC+ neurons expressed atp6v0c2, suggesting that atp6v0c2 is expressed in relatively more mature neurons. Third, expression of the ATP6V0C2-RFP fusion protein in neuronal cells in the zebrafish spinal cord showed co-localization of ATP6V0C2 with SV2, a protein present in presynaptic vesicles (Buckley and Kelly,1985). Altogether, these data suggest that atp6v0c2 might play an important role in neurotransmitter storage and/or release from the synaptic vesicle in mature neurons.
Neuronal excitability is highly sensitive to intra- and extracellular pH; thus, regulation of pH in the nervous system is critical for the normal function of neurons. Previous study revealed that transient or persistent changes in intracellular pH can modulate neuronal excitability and the activity of ion channels, such as voltage-dependent Ca2+ channels and Ca2+-activated K+ channels (Peers and Green,1991; Tombaugh and Somjen,1997; Bonnet and Wiemann,1999; Kiss and Korn,1999). A recent study also showed that Slc4a10, a member of the SLC4 bicarbonate transporter family, is involved in control of excitability of hippocampal CA3 neurons through the regulation of internal pH (Jacobs et al.,2008).
In addition to regulating the pH of intracellular organelles, V-ATPases are expressed at the plasma membrane, where they carry out cell-specific functions, such as renal acidification and homeostasis of cytoplasmic pH (Forgac,2007). The Ca2+ imaging data obtained from atp6v0c2 morpholino injection experiments suggests that in addition to its possible role in neurotransmitter storage/release in synaptic vesicles, ATP6V0C2 is involved in the control of neuronal excitability by contributing to the regulation of intracellular pH.
The mutant strain used in this study was mind bomb (mibta52b; Itoh et al.,2003). We also used two transgenic lines, Tg(hsp70l:Gal4)1.5kca4 and Tg(UAS: myc-Notch1a-intra)kca3, to induce Notch signaling in response to heat shock (Scheer and Campos–Ortega,1999).
Heat-Induced Gene Expression
To induce the expression of constitutively active Notch1a, Tg(hsp70:GAL4) and Tg(UAS:Notch1aac-myc) adults were mated and the embryos were collected and raised at 28.5°C. At 7.5 hpf, the embryos were transferred to embryo medium (EM; 15 mM NaCl, 0.5 mM KCl, 1 mM CaCl2, 1 mM MgSO4, 0.15 mM KH2PO4, 0.05 mM NH2PO4, and 0.7 mM NaHCO3) at 39.0°C for 30 min and then returned to EM at 28.5°C until the three-somite stage (11 hpf). Approximately, one-fourth of the embryos were predicted to inherit both transgenes, which was confirmed by anti-Myc immunocytochemistry (data not shown).
Morpholino Design and Injection
A morpholino oligonucleotide (5′- GACTTACTTGTACAGAGTGATCTT G-3′) was designed to block the splicing between exon2 and exon3 of the atp6v0c2. The morpholino oligonucleotide was dissolved in 1× Danieau solution at a concentration of 20 μg/μl, and was then further diluted with distilled water. Ten nanograms of morpholino oligonucleotide was injected into 1–2-cell-stage embryos.
Whole-Mount In Situ Hybridization
Anti-sense riboprobes were transcribed from cDNA for zebrafish atp6v0c2 and HuC (Kim et al.,1997). In situ hybridization was performed as described previously (Hauptmann and Gerster,2000). Photos were taken using a differential interference contrast microscope (Axioskop, Carl Zeiss).
Bromodeoxyuridine (BrdU) Labeling and Immunohistochemistry
Manually-dechorionated embryos were labeled with BrdU (Roche, Nutley, NJ) by incubation for 20 min on ice in a solution of 10 mM BrdU and 15% DMSO in EM at 24 hpf. The embryos were then placed in EM, incubated for 20 min at 28.5°C, and fixed using 4% paraformaldehyde in PBS. The embryos were processed for in situ RNA hybridization to detect the atp6v0c2, treated for 1 hr with 2 M HCl, and then processed for anti-BrdU immunohistochemistry. For immunohistochemistry, the following primary antibodies were used: mouse anti-BrdU (G3G4, 1:1,000, Developmental Studies Hybridoma Bank (DSHB), Iowa City, IA), mouse anti-SV2 (1:100, DSHB), mouse anti-Znp-1 (1:1,000, DSHB), mouse anti-Islet (39.4D5, 1:100, DSHB), and mouse anti-HuC/D (16A11, 1:20, Molecular Probes, Eugene, OR). For fluorescent detection of antibody labeling, we used Oregon Green 488-conjugated goat anti-mouse, Alexa Fluor 594-conjugated goat anti-rabbit IgG, and Alexa Fluor 594-conjugated goat anti-mouse IgG (1:500, Molecular Probes). Photos were taken using a confocal laser scanning microscope (LSM 510 Pascal, Carl Zeiss, Thornwood, NY).
Preparation of Dissociated Zebrafish Neurons
Tg(HuC:egfp) embryos (Park et al.,2000b) were collected at 2 dpf and placed in a Petri dish on ice to anesthetize them. Embryos were then chopped into several pieces, collected in 1.5-ml Eppendorf tubes, and washed in Ringer's solution (116 mM NaCl, 2.9 mM KCl, 1.8 mM CaCl2, and 5 mM HEPES, pH 7.2) for 20 min at 4°C. The embryos were then washed three times with 1 ml of Dulbecco's PBS (D-PBS, WelGENE) and resuspended in 100 μl of D-PBS. Thirty microliters of Liberase Blendzyme 3 solution (Roche) was added and the embryos were incubated at 29°C for 30 min with occasional pipetting. Next, 1 ml of Trypsin solution (HyClone, Logan, UT) was added and the samples were incubated at 29°C for 15 min. The trypsinized cells were applied to a Cell Strainer (40 μm, BD Falcon, Rockville, MD) to remove any undissociated mass. Dissociated cells were collected by centrifugation at 300g and 4°C for 5 min, and then washed twice with D-PBS containing 10% FBS (Hyclone). Finally, dissociated cells were suspended in Leibovitz L-15 medium (Sigma, St. Louis, MO) with 1% FBS (Andersen, 2001).
Ca2+ Imaging Experiments
Ca2+ imaging experiments were carried out based on the protocol established by Bang et al., with slight modifications (Bang et al.,2007). Briefly, dissociated cells from Tg(HuC-egfp) embryos were loaded with 5 M Fura-2AM and 1% pluronic F-127 for 1 hr, and then resuspended in the bath solution (140 mM NaCl, 5 mM KCl, 2 mM CaCl2, 2 mM MgCl2, 10 mM HEPES, pH 7.4). Images of Fura-2 loaded neurons were captured with a cooled CCD camera (Retiga-SRV, Q-imaging Corp., Burnaby, BC, Canada) at an excitation wavelength alternating between 340 and 380 nm. The ratio of fluorescence intensity of the two wavelengths in each experiment was analyzed using MetaFluor (Molecular Devices, Sunnyvale, CA). Values from different experiments were normalized to the baseline of the ratio of 340/380 nm. GFP fluorescence-positive neurons were selected for recording Ca Ca2+ levels. For the depolarizing stimulus, 100 mM NaCl was substituted with 100 mM KCl in the bath solution.
This study was supported by a grant of the Korean Health Technology R&D Project, Ministry for Health, Welfare & Family Affairs (code A084909), Republic of Korea.