Regulation of synaptic vesicle accumulation and axon terminal remodeling during synapse formation by distinct Ca²⁺ signaling

Animals

Zebrafish AB strain was used. Zebrafish embryos were raised at 28.5°C in embryo medium (EM) containing 17 mM NaCl, 0.27 mM CaCl₂, 0.66 mM MgSO₄ and 0.4 mM KCl.

Construction of zebrafish olfactory sensory neuron-specific expression vectors

The 1.4-kb olfactory marker protein gene (omp) promoter-driven zebrafish GAP43-EGFP (Pomp-GG), rat VAMP2-EGFP (Pomp-VG) and bovine tau-EGFP (Pomp-TG) expression vectors were previously described (Yoshida and Mishina 2005). Two oligonucleotides, 5′-CATGAGAAGAAAGTGGCAGAAAACAGGGCATGCGGTACGAGCTATTGGTCGACTGTCTTCTTAGT-3′ and 5′-CTAGACTAAGAAGACAGTCGACCAATAGCTCGTACCGCATGCCCTGTTTTCTGCCACTTTCTTCT-3′, were annealed to generate the calmodulin inhibitory peptide coding sequence. The 1.3-kb coding sequence of the zebrafish Kir2.1 gene was cloned by RT-PCR with primers, 5′-CCATGGGAAGTGTGCGGGCCAAC-3′ and 5′-TCTAGATTATATTTCAGATTCCCGCCGTAG-3′, using mRNAs prepared from adult zebrafish brains as a template. The 1.2-kb coding sequence of rat IP₃ 5-Ppase was a gift from Dr. M Iino (Furutani et al. 2006). The caCN coding sequence of Pomp-GG-caCN, Pomp-VG-caCN and Pomp-TG-caCN (Yoshida and Mishina 2005) were replaced with the coding sequences of calmodulin inhibitory peptide, zebrafish Kir2.1 and rat IP₃ 5-Ppase to yield Pomp-GG-WP, Pomp-GG-Kir2.1 and Pomp-GG-5-Ppase, Pomp-VG-WP, Pomp-VG-Kir2.1 and Pomp-VG-5-Ppase, and Pomp-TG-WP, Pomp-TG-Kir2.1 and Pomp-TG-5-Ppase, respectively.

The caCN coding sequence of Pomp-caCN (Yoshida and Mishina 2005) was replaced with the 0.7-kb coding sequence of the Gal4-VP16 fusion gene isolated from tub-GVP-Uunc (Köster and Fraser 2001) to yield Pomp-GVP. The 0.5-kb DNA fragment of 13-mer of Gal4 upstream activating sequence (UAS) sites was amplified with primers, 5′-GCGGCCGCTGTCCTCCGGGCTG-3′ and 5′-ACATGTCACAGATCCTGC-3′, using tub-GVP-Uunc as a template and cloned into pCRII-TOPO vector (Invitrogen, Carlsbad, CA, USA) to yield pCRII-U13. The 0.5-kb NotI-BspLU11I fragment carrying a 13-mer of UAS from pCRII-U13 and the 1.4-kb BspLU11I-SalI fragment carrying the VAMP2-EGFP followed by SV40 polyadenylation signal sequence from Pomp-VG were ligated into the NotI-SalI sites of pBluescript II SK+ (Stratagene, La Jolla, CA, USA) to yield U13-VG. The coding sequence of VAMP2-EGFP in U13-VG was replaced by the 1.1-kb coding sequence of a dominant-negative form of mouse type I regulatory subunit of PKA (Clegg et al. 1987) (dnPKA), the 1.1-kb coding sequence of constitutively active form of the mouse PKA α catalytic subunit gene (Orellana and McKnight 1992) (PKA*), and the 1.2-kb coding sequence of IP₃ 5-Ppase to yield U-dnPKA, U-PKA* and U-5-Ppase, respectively. The 0.7-kb AgeI-ScaI fragment carrying 5-mer of UAS sites followed by VAMP2 from U13-VG was ligated with the 4.0-kb AgeI-SmaI fragment from Pomp-VG to yield U-VG. A linker sequence, 5′-CTCGAGGCCTAGCGCTGGCGCCCGGGCCC-3′, was inserted between the XhoI and ApaI sites of U-VG to yield U-VG-linker. The 2.1-kb MluI-BssHII fragment from U-5-Ppase was blunted and inserted into the AatI site of U-VG-linker to yield U-VG-U-5-Ppase. The 6.5-kb NotI fragment from Pomp-GVP was inserted into the NotI site of U-PKA* and U-dnPKA to yield Pomp-GVP-U-PKA* and Pomp-GVP-U-dnPKA, respectively. The PKA* coding sequence of Pomp-GVP-PKA* was replaced with the coding sequence of DsRed-monomer from pDsRed-monomer-C1 (Clontech, Palo Alto, CA, USA) and that of enhanced yellow fluorescent protein (EYFP) from pEYFP-N1 (Clontech) to yield Pomp-GVP-U-RFP and Pomp-GVP-U-YFP, respectively. The VAMP2-EGFP coding sequence of U-VG-U-5-Ppase was replaced with the coding sequence of enhanced cyan fluorescent protein from pECFP-N1 (Clontech) to yield U-CFP-U-5-Ppase. The IP₃-5-Ppase coding sequence of U-CFP-U-5-Ppase was replaced with the coding sequences of EYFP and DsRed-monomer to yield U-CFP-U-YFP and U-CFP-U-RFP, respectively.

Microinjection of DNA into zebrafish embryos

Expression vectors were linearized by SpeI or SacII and dissolved in 100 mM KCl containing 0.05% phenol red. About 0.2–0.5 nL of the DNA solution at the concentrations of 50–100 ng/μL was injected into the cytoplasm of 1- to 4-cell embryos.

Microscopy

Zebrafish embryos expressing reporter genes in the olfactory placode at 28–29 hpf were screened by a fluorescent microscope. At 36, 60 and 84 hpf, the embryos were mounted in 1% low-melting point agarose in EM containing 0.02% 3-aminobenzoic acid ethyl ester (tricaine, Sigma, St. Louis, MO, USA) with their head toward the glass bottom of a dish. Embryos were imaged by Radiance 2100 Confocal Scanning System (Bio-Rad, Hercules, CA, USA) using Nikon 60× water lens [numerical aperture, 1.00]: zoom setting, 5.0; Z-step, 1–2 μm; image size, 512 × 512 pixels.

Drug treatments

Zebrafish embryos stably carrying omp promoter-driven tau-EGFP transgene or those injected with Pomp-GG, Pomp-TG or Pomp-VG were dechorionated and soaked in EM containing 1% ethanol or 0.1% dimethyl sulfoxide with either 1 μM calmidazolium chloride (Calbiochem, La Jolla, CA, USA), 1 μM U-73122 (Calbiochem), 1 μM U-73433 (Calbiochem) or 10 μM nifedipine (Sigma) from 30 hpf to 84 hpf. Treated embryos were mounted in 1% low-melting point agarose in EM at 36, 60 and 84 hpf for microscopy.

Image processing

All quantitative measurements were made on the computer screen using the NIH Image 1.62 in a blind manner with respect to drug treatment, expression vectors injected and developmental stages. For measurements of axon terminal morphology and VAMP2-EGFP puncta, z series optical sections were projected by the brightest point method and smoothened by the rank filter method. The outline of an axon terminal was traced from the first branch point or from the point where axonal width was three or more times wider than that of the axon shaft. The trace and the enclosed region were defined as the perimeter and area of the axon terminals, respectively. The complexity was given by dividing the square of the perimeter by the area. For the quantitative measurements of VAMP2-EGFP punctum, the z-projected images with VAMP2-EGFP signals in the linear range of fluorescent intensity below saturation (ranging from 68 to 255 in the NIH Image software) were used. The VAMP2-EGFP punctum was defined as an area larger than five pixels where the intensity of VAMP2-EGFP signals was five or more times stronger than that of non-varicose and non-punctate axonal shaft on the same axon. Axons longer than 10 μm were used for measurements. Data represent mean ± SEM. Statistical significance was evaluated by one-way or two-way anova. When the interaction was significant, unpaired t-test or Fisher’s post hoc test was employed.

Ca²⁺/calmodulin inhibitor suppressed both synaptic vesicle accumulation and axon terminal remodeling

We hypothesized that Ca²⁺/calmodulin signaling regulates the axon terminal differentiation of zebrafish olfactory sensory neurons. To test the hypothesis, we first examined the effect of calmidazolium, an inhibitor of calmodulin, on synaptic vesicle accumulation and axon terminal remodeling. Zebrafish embryos at 30 hpf were soaked in the medium containing calmidazolium and were incubated in the presence of the drug thereafter. Treatment of zebrafish embryos with 1 μM calmidazolium caused no apparent changes in gross morphology as well as axon extension estimated by terminal reach values that are distances between center of the nasal pit and axon terminals of individual olfactory sensory neurons (Yoshida et al. 2002) (Fig. S1). We thus examined the effect of calmidazolium at the concentration of 1 μM in subsequent experiments.

Zebrafish embryos were injected with omp promoter-driven VAMP2-EGFP expression vector (Pomp-VG, Fig. 1a) to monitor the synaptic vesicle accumulation. VAMP2-EGFP punctate areas in the axon terminal of olfactory neurons in mock-treated embryos markedly increased between 36 and 60 hpf, then slightly increased during next 24 h (Fig. 1b) as described (Yoshida and Mishina 2005). VAMP2-EGFP punctate areas in the axon terminals of calmidazolium-treated embryos were smaller than those of mock-treated embryos at 60 and 84 hpf (t-test, p = 0.0021 and 0.0089, respectively) (Fig. 1c and d). These results suggest that the calmodulin inhibitor suppresses the accumulation of synaptic vesicles in axon terminals.

Zebrafish embryos were next injected with omp promoter-driven GAP43-EGFP expression vector (Pomp-GG, Fig. 2a) and the labeled axon terminals of olfactory sensory neurons were imaged by a confocal microscope at 60 and 84 hpf to monitor axon terminal remodeling. Area, perimeter and complexity values significantly decreased from 60 to 84 hpf in mock-treated embryos (t-test, p < 0.001, = 0.0028 and = 0.010, respectively) (Fig. 2b and d) as described (Yoshida and Mishina 2005). On the other hand, perimeter and complexity values were unchanged during development in calmidazolium-treated embryos (t-test, p = 0.23 and 0.59, respectively), while area values were slightly decreased in calmidazolium-treated embryos (t-test, p = 0.040) (Fig. 2c and d). At 84 hpf, area, perimeter and complexity values were all significantly larger in calmidazolium-treated embryos than in mock-treated embryos (t-test, p < 0.001 for all parameters). These results suggest that suppression of Ca²⁺/calmodulin by calmidazolium prevented the remodeling of axon terminal from large and complex shapes to small and simple shapes between 60 and 84 hpf.

Olfactory sensory neuron-specific expression of calmodulin inhibitory peptide suppressed both synaptic vesicle accumulation and axon terminal remodeling

To restrict the effect of calmodulin suppression in olfactory sensory neurons, we employed the double-cassette vector strategy in living zebrafish embryos (Yoshida and Mishina 2003). The amino acid sequence of zebrafish calmodulins is identical to that of human calmodulins (Friedberg and Taliaferro 2005). Thus, we used the calmodulin-binding domain of mammalian smooth muscle myosin light chain kinase as a calmodulin inhibitory peptide (WP) (Török and Trentham 1994). The omp promoter-driven expression vector Pomp-VG-WP (Fig. 1a) was injected into zebrafish embryos to express VAMP2-EGFP and WP in the same olfactory neurons. Olfactory sensory neurons in zebrafish embryos injected with Pomp-VG served as controls. VAMP2-EGFP punctate areas in the axon terminals of embryos injected with Pomp-VG-WP were significantly smaller than those of mock-treated embryos at 60 and 84 hpf (t-test, p = 0.0014 and 0.047) (Figs 1e and S2[link]). These results suggest that the inhibition of calmodulin in olfactory sensory neurons suppressed synaptic vesicle accumulation in the axon terminals.

We then examined the axon terminal remodeling using double-cassette vector Pomp-GG-WP for GAP43-EGFP and WP (Fig. 2a). The axon terminals of olfactory sensory neurons in control embryos injected with Pomp-GG were remodeled to simple shapes from 60 to 84 hpf (t-test; p < 0.001 for axon terminal area, perimeter and complexity) (Fig. 2e and g) as previously reported (Yoshida and Mishina 2005). In contrast, the axon terminals of olfactory sensory neurons in embryos injected with Pomp-GG-WP showed no morphological changes between 60 and 84 hpf (t-test; p = 0.83, 0.63 and 0.63 for axon terminal area, perimeter and complexity, respectively) (Fig. 2f and g). Area, perimeter and complexity values were significantly larger in Pomp-GG-WP-injected embryos than in control Pomp-GG-injected embryos at 84 hpf (t-test, p < 0.001, < 0.001 and = 0.0013, respectively) (Fig. 2g). These results suggest that the inhibition of calmodulin in olfactory sensory neurons suppressed the morphological remodeling of their axon terminals.

Pharmacological inhibition of phospholipase C perturbed synaptic vesicle accumulation but not axon terminal remodeling

The finding that calmodulin is essential for both synaptic vesicle accumulation and axon terminal remodeling raises a question whether a sole Ca²⁺ signaling that activates calmodulin orchestrates the two representative events of presynaptic differentiation or different Ca²⁺ sources are responsible for respective events. Major Ca²⁺ sources would be Ca²⁺ release from intracellular stores or Ca²⁺ influx from the extracellular fluid. Ca²⁺ release from the endoplasmic reticulum (ER) is triggered by IP₃, which is produced by large PLC family isozymes including PLC-β, PLC-γ and PLC-δ (Berridge 1984) by hydrolysis of the phosphotidylinositol 4,5-bisphosphate. Thus, we examined the effect of pharmacological suppression of PLCs on VAMP2-EGFP punctum formation. Zebrafish embryos at 30 hpf were soaked in the medium containing 1 μM U-73122, a potent PLC inhibitor with broad specificity toward all PLCs (Smith et al. 1990; Knox et al. 2004) or U-73433, an inactive analogue of U-73122, and were incubated in the presence of the drug thereafter. The drug treatment caused no apparent changes in gross morphology of zebrafish embryos (Fig. 3a). Treatment with U-73122 suppressed VAMP2-EGFP punctum formation in the axon terminals at 60 hpf [post hoc test, p < 0.05 (vs. mock treatment) and p < 0.05 (vs. U-73433 treatment)] (Fig. 3b). These results suggest that suppression of PLCs by U-73122 prevented the synaptic vesicle accumulation at 60 hpf, suggesting that IP₃ signaling is important for synaptic vesicle accumulation. On the other hand, the treatment with U-73122 had little effect on the complexity value of Pomp-GG injected embryos at 84 hpf (Fig. 3c) suggesting no apparent effect of PLC inhibition on the axon terminal remodeling.

Olfactory sensory neuron-specific expression of IP₃ 5-Ppase suppressed synaptic vesicle accumulation

We further examined the effect of IP₃ 5-Ppase, which selectively hydrolyzes IP₃ to inhibit Ca²⁺ release from ER (Majerus 1992; Laxminarayan et al. 1994). Olfactory sensory neuron-specific expression of IP₃ 5-Ppase using Pomp-VG-5-Ppase significantly suppressed the formation of VAMP2-EGFP puncta (Figs 3d and S3[link]). VAMP2-EGFP punctate areas in the axon terminals of embryos injected with Pomp-VG-5-Ppase were significantly smaller than those of control embryos injected with Pomp-VG at 60 and 84 hpf (t-test, p = 0.0095 and 0.026) (Fig. 3d). On the other hand, expression of IP₃ 5-Ppase did not affect the morphological remodeling of olfactory neuron axon terminals (Figs 3e and S3[link]). These results suggest that the inhibition of Ca²⁺ release from ER in olfactory sensory neurons selectively suppressed VAMP2-EGFP punctum formation in the axon terminals.

Voltage-gated Ca²⁺ channel blocker suppressed axon terminal remodeling

As the suppression of IP₃-mediated Ca²⁺ release from ER exerted little effect on axon terminal remodeling, we next examined the effects of pharmacological inhibition of Ca²⁺ influx. When bath-applied after 30 hpf, L-type Ca²⁺ channel blocker nifedipine at the concentration of 10 μM caused no apparent changes in gross morphology of zebrafish embryos (Fig. 4a). The nifedipine treatment had little effect on VAMP2-EGFP punctate area at 60 hpf during early stage of synapse formation (t-test, p = 0.12) (Fig. 4b). On the other hand, the complexity values at 84 hpf were significantly larger in nifedipine-treated embryos than in mock-treated embryos (t-test, p < 0.001) (Fig. 4c). These results suggest that neuronal activity-induced Ca²⁺ signaling regulates axon terminal remodeling.

Olfactory sensory neuron-specific expression of Kir2.1 channel prevented axon terminals from morphological remodeling

It is known that expression of inward rectifying potassium channel Kir2.1 results in a significant hyperpolarization and inhibits neuronal firing and calcium spiking (Yu et al. 2004; Hua et al. 2005). We thus examined the effect of blocking of neuronal firing that triggers Ca²⁺ influx by olfactory sensory neuron-specific expression of Kir2.1. VAMP2-EGFP punctate areas in the axon terminals of embryos injected with Pomp-VG-Kir2.1 were comparable to those of control Pomp-VG-injected embryos at 36 hpf (t-test, p = 0.090) and 60 hpf (t-test, p = 0.46) (Figs 4d and S4[link]). On the other hand, at 84 hpf, VAMP2-EGFP punctate areas were smaller in Pomp-VG-Kir2.1-injected embryos (t-test, p = 0.005). The axon terminals of olfactory sensory neurons in embryos injected with Pomp-GG-Kir2.1 showed no morphological changes between 60 and 84 hpf (t-test, p = 0.52) (Figs 4e and S4[link]). The complexity values were significantly larger in Pomp-GG-Kir2.1-injected embryos than in control Pomp-GG-injected embryos at 60 and 84 hpf (t-test, p = 0.021 and p < 0.001, respectively). These results suggest that the activity blockade of olfactory sensory neurons exerted little effect on the accumulation of synaptic vesicles in the axon terminal at the early stage of synapse formation but strongly prevented the morphological remodeling of axon terminals and suppressed synaptic vesicle accumulation at the late stage of synapse formation.

IP₃ signaling is upstream of PKA signaling

Present investigation together with previous results (Yoshida and Mishina 2005) suggests that VAMP2-EGFP punctum formation in the axon terminal of olfactory sensory neurons requires both PKA and IP₃-mediated Ca²⁺ signaling. To examine the interaction of two signaling genetically, we manipulated both IP₃ signaling and PKA signaling in olfactory sensory neurons employing Gal4-UAS system (Scheer and Campos-Ortega 1999). We first constructed omp promoter-driven Gal4-VP16 vector (Pomp-GVP, Fig. 6a), Pomp-GVP linked with Gal4-binding UAS-driven DsRed (Pomp-GVP-U-RFP), Pomp-GVP linked with UAS-driven EYFP (Pomp-GVP-U-YFP), and UAS-driven ECFP linked with UAS-driven EYFP (U-CFP-U-YFP) or UAS-driven DsRed (U-CFP-U-RFP) (Fig. 5a). ECFP positive cells in embryos coinjected with Pomp-GVP-U-RFP and U-CFP-U-YFP were both DsRed and EYFP positive (Fig. 5b and c). Moreover, DsRed intensity in the olfactory neurons of embryos injected with Pomp-GVP-U-RFP and U-CFP-U-YFP, with Pomp-GVP-U-YFP and U-CFP-U-DsRed, and with Pomp-CFP-RFP carrying the omp promoter-driven ECFP linked with the omp promoter-driven DsRed were comparable (Fig. 5c). Thus, the Gal4-UAS system enabled the co-expression of ECFP, EYFP and DsRed in the same olfactory sensory neurons. Expression levels of these molecules by Gal4-UAS vectors were comparable to those by omp promoter-driven double-cassette vectors.

We then constructed Pomp-GVP linked with UAS-driven dominant negative PKA (Pomp-GVP-U-dnPKA) or UAS-driven constitutively active PKA (Pomp-GVP-U-PKA*), UAS-driven VAMP2-EGFP (U-VG), and U-VG linked with UAS-driven IP₃ 5-Ppase (U-VG-U-5-Ppase) to express two effector molecules in a single olfactory sensory neuron labeled with a reporter (Fig. 6a). Olfactory sensory neurons coinjected with Pomp-GVP and U-VG served as controls. Both expression of dnPKA by co-injecting Pomp-GVP-U-dnPKA with U-VG and expression of IP₃ 5-Ppase by co-injecting Pomp-GVP with U-VG-U-5-Ppase, suppressed VAMP2-EGFP punctum formation at 60 hpf [post hoc test, p < 0.01 (vs. control) and p < 0.05 (vs. control)] (Fig. 6b). Expression of both IP₃ 5-Ppase and dnPKA by co-injecting Pomp-GVP-U-dnPKA and U-VG-U-5-Ppase also suppressed VAMP2-EGFP punctum formation (post hoc test, p < 0.01 vs. control). However, VAMP2-EGFP punctate areas in axon terminals were comparable among olfactory neurons expressing dnPKA, IP₃ 5-Ppase, and both. Moreover, VAMP2-EGFP punctate area of olfactory sensory neurons injected with both Pomp-GVP-U-PKA* and U-VG-U-5-Ppase was comparable to that of controls, suggesting that suppressive effect of IP₃ 5-Ppase on VAMP2-EGFP punctate area was rescued to the level of controls by additive expression of PKA*. These results suggest that IP₃-mediated Ca²⁺ signaling regulates synaptic vesicle accumulation upstream of PKA.