Localization of SNARE proteins in the brain and corpus allatum of Bombyx mori

Soluble N-ethylmaleimide-sensitive factor attachment protein receptors (SNAREs) make up the core machinery that mediates membrane fusion. SNAREs, syntaxin, synaptosome-associated protein (SNAP), and synaptobrevin form a tight SNARE complex that brings the vesicle and plasma membranes together and is essential for membrane fusion. The cDNAs of SNAP-25, VAMP2, and Syntaxin 1A from Bombyx mori were inserted into a plasmid, transformed into Escherichia coli, and purified. We then produced antibodies against the SNAP-25, VAMP2, and Syntaxin 1A of Bombyx mori of rabbits and rats, which were used for immunohistochemistry. Immunohistochemistry results revealed that the expression of VAMP2 was restricted to neurons in the pars intercerebralis (PI), dorsolateral protocerebrum (DL), and central complex (CX) of the brain. SNAP-25 was restricted to neurons in the PI and the CX of the brain. Syntaxin 1A was restricted to neurons in the PI and DL of the brain. VAMP2 co-localized with SNAP-25 in the CX, and with Syntaxin 1A in the PI and DL. VAMP2, SNAP-25, and Syntaxin 1A are present in the CA. Bombyxin-immunohistochemical reactivities (IRs) of brain and CA overlapped with VAMP2-, SNAP-25, and Syntaxin 1A-IRs. VAMP2 and Syntaxin 1A are present in the prothoracicotropic hormone (PTTH)-secretory neurons of the brain.


Introduction
Membrane fusion is a key process in all living organisms and is essential for exocytosis, including the release of neuropeptides and neurotransmitters (Chernomordik and Kozlov 2005;Han et al. 2017;Houy et al. 2013).
The efficient and controllable fusion of biological membranes is known to be driven by the cooperative action of soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE) proteins, which constitute the central components of the eukaryotic fusion machinery responsible for the fusion of synaptic vesicles with the plasma membrane (Pelham 2001;Südhof and Rothman 2009;Shin 2014;Urbina and Gupton 2020). During exocytosis, vesicle-associated v-SNARE (VAMP/synaptobrevin) and target-cell-associated t-SNAREs (syntaxin and SNAP) assemble into a core trans-SNARE complex. This complex plays a versatile role at various stages of exocytosis, ranging from priming to fusion pore formation and expansion, resulting in the release or exchange of the vesicle content. The best-studied SNARE complex is the one that mediates the fast exocytosis of neuronal cells, being formed by three proteins: VAMP2/synaptobrevin-2, syntaxin-1, and SNAP-25.
In Drosophila melanogaster, the SNARE proteins VAMP/ synaptobrevin, SNAP-25, and syntaxin are core components of the apparatus that mediates the release of neurotransmitters (Megighian et al. 2013). SNARE complexes are present in vivo and have been isolated directly from Drosophila head extracts (Littleton et al. 1998). VAMP2, SNAP-25, and Syntaxin 1A mutants reduce or eliminate the evoked release of Mako Sasao and Tomohide Uno contributed equally to this work. neurotransmitters (Kawasaki and Ordway 2009;Rao et al. 2001;Schulze et al. 1995;Deitcher et al. 1998; Barrecheguren et al. 2017).
Neuropeptides play important roles in the development, physiology, and behavior of both vertebrates and invertebrates (Nassel and Winther 2010;Hoyer and Bartfai 2012;Xu et al. 2020;Zeng et al. 2020). In particular, insect neuropeptides regulate insect-specific phenomena, such as metamorphosis, molting, feeding, development, and ecdysis. These neuropeptides are synthesized mainly in the brain and are secreted into the hemolymph by neuroendocrine organs, such as the corpus allatum (CA) (Roller et al. 2008;Tobe and Pratt 1974). There are no reports on the molecular mechanisms that drive neuropeptide-filled secretory vesicle fusion in the insect brain via SNARE proteins and the neuropeptides secreted by CA.
The neurosecretory hormone bombyxin is an insulinlike peptide found in the order Lepidoptera (butterflies and moths). Bombyxin was first identified in the silkworm, Bombyx mori (Nagasawa et al. 1986), and later in the tobacco hornworm, Manduca sexta (Nijhout and Grunert 2002;Van de Velde et al. 2007). Bombyxin stimulates cell division and is a growth factor in the wing imaginal disks of Precis coenia and M. sexta (Nijhout and Grunert 2002;Nijhout et al. 2007).
In all insects, prothoracicotropic hormone (PTTH) acts on the prothoracic glands (PGs), which initiate ecdysone synthesis (Nagata et al. 2005). Ecdysone is converted to 20-hydroxyecdysone, which acts on numerous target tissues and induces the expression of genes related to the molting process (Thummel 2002).
Therefore, the present study aimed to clarify the localization of SNARE proteins (VAMP2, SNAP-25, and Syntaxin 1A) and investigate the relationship between SNARE proteins and bombyxin, or between PTTH localization in the brain and the CA of B. mori. Antibodies against B. moriderived SNARE proteins (VAMP2, SNAP-25, and Syntaxin 1A) were used to identify SNARE protein-expressing cells, locate their regional distribution, and determine their colocalization status with bombyxin and PTTH.

Insect cultures
We raised the hybrids (Kinshu × Showa or Daizo, p50 strains, National BioResource Program, Fukuoka, Japan) of B. mori. Twenty larvae in one container were reared on an artificial diet (Silkmate 2 M, Nosan Co., Yokohama, Japan) at 25 °C and subjected to a 16-h light/8-h dark cycle with a relative humidity of 70%. Fourth-instar larvae (5 days old) were used in all experiments. For immunoblotting, 12-15 brains were isolated and extracted. Five micrograms of protein was used for immunoblotting. For immunohistochemistry, 12-15 brains were used.

Purification of B. mori VAMP2, SNAP25, and Syntaxin 1A and the production of antiserum
The cDNA fragments containing the coding sequences of B. mori VAMP2, SNAP-25, and Syntaxin 1A were generated using reverse-transcription PCR (RT-PCR), and then separately subcloned into pTA2 (cat. no. TAK-101, Osaka, Japan). DNA sequences encoding the His-tag were added to the amino-terminal using PCR. The plasmid pCR2.2 was digested using BamHI and EcoRI. The digested fragments containing the SNAREs of B. mori were then independently inserted between the BamHI and EcoRI sites of pGEX6P-2. The plasmid pGEX6P2 expresses the target protein as a GST-fusion protein. The identity of the clones containing B. mori SNAREs was confirmed using sequence analyses by using an ABI Prism 377 DNA Sequencer (Artisan Technology Group, IL 61,822, USA).
The expression of B. mori SNARE proteins in E. coli (BL21 strain) was assessed and their purification performed as previously described (Uno et al. 2014). The GST proteins, which were independently expressed using pGEX6P2, were purified using a glutathione S-Sepharose column. After digestion using the PreScission protease, the His-tagged protein was purified using a His-accept column.
Antisera were generated in a rabbit or rat by the injection of a 1:1 (v/v) mixture of each purified protein of B. mori (1 mg) and Freund's complete adjuvant. The rabbits and rats received three booster injections at 2-week intervals. The sera were isolated and tested for the presence of anti-B. mori SNARE antibodies using western blotting. Antisera against PTTH, and bombyxin was obtained as previously described (Uno et al. 2007).

Western immunoblotting
One microgram of the purified proteins (GST-BVAMP2., GST-BSNAP-25, GST-BSyntaxin 1A) or 5 μg of brain extract from 12-15 brains was transferred to a polyvinylidene difluoride membrane after SDS-PAGE, which was performed according to the method of Laemmli (Laemmli 1970), using a 4.5% stacking gel/15% separating gel (16 mA for 60 min). The membrane was blocked in Blocking One solution [60 min at room temperature (RT)], and then incubated for 60 min at RT with the appropriate primary antibody, as follows: anti-SNARE serum (1:2000) in Trisbuffered saline (TBS; 50 mM Tris-HCl and 50 mM NaCl, pH 8.0) containing Blocking One solution. The membrane was then washed three times with TBS [including 0.05% Tween-20 (v/v)], followed by incubation for 60 min at RT with the secondary antibody, and peroxidase-conjugated goat anti-rabbit IgG (1:2000). The membrane was washed three times with TBS plus Tween-20, and the proteins were detected using a peroxidase-staining DAB kit.
The insect head sections were washed (at RT) in distilled water and in phosphate-buffered saline (PBS; 137 mM NaCl, 2.7 mM KCl, 10 mM Na 2 HPO 4 , and 2 mM KH 2 PO 4 , pH 7.4) containing 0.3% Triton-X100 (PBS-Tr), blocked (30 min at RT) with an antibody dilution buffer (PBS-Tr containing 1.5% goat serum), and incubated overnight at 4 °C in dilution buffer with the primary antibodies, anti-SNARE rabbit serum (1:500), and anti-bombyxin or anti-PTTH mouse IgG (1:500). After rinsing three times for 10 min each with PBS-Tr at RT, the sections were incubated for 1.5 h at RT with the secondary antibody (7.5 μg/mL), donkey anti-mouse IgG (H + L)-CF555, or goat anti-rabbit IgG (H + L)-CF488. After washing in PBS-Tr, the sections were mounted in Aqua-Poly/Mount medium and examined using a BX50 microscope equipped with BX-FLA reflected light fluorescence and WIG and NIBA mirror/filter units. NIBA is excited in blue and detected in green, and WIG is excited in green and detected in red. Three to five individuals were used for each immunocytochemical experiment. The excitation and emission wavelength ranges used in the WIG mirror/filter units were 520-550 nm and more than 580 nm, respectively. The excitation and emission wavelength ranges used in the NIBA mirror/filter unit were 470-490 nm and 515-550 nm, respectively.
In the control experiments, the primary antibodies were replaced with pre-immune rabbit serum. No significant staining intensity above the background intensity level was observed.

Production of antibodies against the SNARE proteins of B. mori
To make sure that antibodies recognize the purified proteins we raised them against, we tried immunoblotting using the purified proteins ( Fig. 1 lanes 1-9). The antibodies produced against VAMP2, SNAP-25, and Syntaxin 1A of B. mori specifically recognized the protein band corresponding to the position of the purified partial protein of B. mori (Fig. 1). In the two control experiments (i.e., the addition of pre-immune serum instead of the primary antibody and the addition of the antigen and primary antibody together), the bands were not detected (Fig. 1). To ascertain that these antibodies are not cross-reactive with any other/unknown protein in the tissue analyzed, we tried immunoblotting using brain extract ( Fig. 1 lanes 10-18). Western immunoblots for VAMP2, SNAP-25, and Syntaxin 1A showed one band in the brain (Fig. 1, lanes 10, 13, and 16).

SNARE proteins are present in specific neurons in the B. mori brain
Anti-VAMP2-stained subcellular regions in individual neurons in the pars intercerebralis (PI), central complex (CX), and dorsolateral protocerebrum (DL) (Fig. 2a and d). Two broad areas in CX, four to eight neurons in PI, and one to four neurons in DL were detected ( Fig. 2a and d).
Anti-SNAP25-stained subcellular regions in individual neurons in the PI and CX ( Fig. 2b and g). Two broad areas in CX and one to two neurons in PI were detected ( Fig. 2b  and g).  1 Immunoblot analysis of the anti-SNARE protein antibodies (SNAP25, VAMP2, and Syntaxin 1A). Lanes 1-3 were loaded with GST-BVAMP2; lanes 4-6 with GST-BSNAP-25; lanes 7-9 with GST-BSyntaxin 1A; lanes 10-18 with brain extract. Lanes 1 and 10, anti-VAMP2, antiserum staining; lanes 4 and 13, anti-SNAP-25 anti-serum staining; lanes 7 and 16, anti-Syntaxin 1A antiserum staining. In lanes 2, 5, 8, 11, 14, and 17, pre-immune serum was used as primary antibody. Lanes 3, 6, 9, 12, 15, and 18 contain an antibody-positive antigen, which was used as the primary antibody (control)  e and h). The section was stained with different antibodies together. SNAP-25-IRs colocalized with VAMP2-IRs (c). Syntaxin 1A-IRs co-localized with VAMP2-IRs (f) and did not co-localize with SNAP-25-IRs (i). Scale bar, 100 μm. A schematic of the B. mori brain (j). CA, corpus allatum Anti-Syntaxin 1A detected a restricted area in a set of neurons in the PI and DL (Fig. 2e and h). Four to eight neurons per section in PI and one to four neurons per section in DL were detected ( Fig. 2b and g). The approximate total number of neurons in the insect was 10 5 -10 6 (Mizunami et al. 2004).
The double-labeling experiments showed that SNAP-25-immunohistochemical reactivities (IRs) in the CX were present in VAMP2-IRs (Fig. 2c). Syntaxin 1A-IRs in the PI and DL overlapped with VAMP2-IRs (Fig. 2f) and did not overlap with the areas of SNAP-25-IR (Fig. 2i).
As a control for the binding specificity, the anti-VAMP2 antibody or anti-SNAP-25 antibody was pre-incubated with an excess amount of antigen before immunological staining (Fig. 3a-f). No significant staining was observed following preabsorption.
The bleed-through of fluorescence was assessed by imaging the single-stained samples through both filter/mirror units (WIG and NIBA). No significant staining was observed ( Fig. 3g-l).

SNARE proteins are present in the CA
Anti-VAMP2, anti-SNAP-25, and anti-Syntaxin 1A detected a restricted area in the CA section (Fig. 4). Further double-labeling experiments showed that SNAP-25-IRs and Syntaxin 1A-IRs were present in VAMP2-IRs (Fig. 4c, f, and i). This study is the first to report a possible relationship between SNAREs and neurosecretion in the CA of insects.

SNARE proteins are present in the bombyxin-secretory neurons of the brain
Anti-bombyxin detected four to eight neurons per section in the PI area (Fig. 5a, d, and g). Bombyxin is produced by four pairs of PI neurosecretory cells of the brain. (Mizoguchi et al. 1987). Double-labeling experiments showed that bombyxin-IRS were present in VAMP2-, SNAP-25-, and Syntaxin 1A-IRs (Fig. 5c, f and i, arrow).

Fig. 3
Control and bleedthrough fluorescence. The effect of antibody preabsorption (a-f) and filter specificity (g-l). The anti-SNAP-25 antibody (a-c) or anti-VAMP2 antibody (d-f) was pre-incubated with an excess amount of antigen before immunological staining. No significant staining was observed following preabsorption (b and d). The bleed-through of fluorescence was assessed by imaging the single-stained samples through both filter/mirror units (WIG and NIBA). No significant staining of VAMP2-IR at WIG filter was observed (h). No significant staining of SNAP25-IR at NIBA filter was observed (k). Scale bar, 100 μm

SNARE proteins are present in the bombyxin-secretory neurons of CA
Double-labeling experiments showed that bombyxin-IRs are present in VAMP2-, SNAP-25-, and Syntaxin 1A-IRs (Fig. 6c, f and i).

VAMP2 is present in the PTTH-secretory neurons of the brain
PTTH is released from the CA into the hemolymph (O'Brien et al. 1988). Anti-PTTH detected one to two neurons in the dorsolateral protocerebrum area (Fig. 7a, d, and g). PTTH is produced by two pairs of DL neurosecretory cells of the  (a, d, and g). Anti-VAMP2 (b), anti-SNAP-25 (e), and anti-Syntaxin 1A (h) detected subcellular regions in individual neurons in the brain. Bombyxin-IRs co-localized with VAMP2-, SNAP-25, and Syntaxin 1A-IRs (c, f and i, arrow). Scale bar, 100 μm brain (Ishizaki and Suzuki 1994). Double-labeling experiments showed that PTTH-IRs co-localized in VAMP2-IRs and Syntaxin 1A-IRs in the brain ( Fig. 7c and i, arrow) and did not co-localize in the SNAP25-IR neurons of the brain (Fig. 7f). PTTH-IRs were not present in SNAP25-IRs in the brain, so we did not perform double-labeling experiments to show whether PTTH-IRs were present in SNAP25-IRs in the CA. PTTH-IRs did not overlap with VAMP2-IRs and Syntaxin 1A-IRs in the secretory neurons of the CA (Fig. 7l  and o). Therefore, these results indicate that VAMP2 and Syntaxin 1A are present in the PTTH-secretory neurons of the brain.

Discussion
The expression of VAMP2 was restricted to neurons in the pars intercerebralis (PI), dorsolateral protocerebrum (DL), and central complex of the brain (CX). SNAP-25 was restricted to neurons in the PI and the CX of the brain. Syntaxin 1A was restricted to neurons in the PI and DL of the brain.
VAMP2 co-localized with SNAP-25 in the CX and with Syntaxin 1A in the PI and DL. VAMP2, SNAP-25, and Syntaxin 1A are present in the CA.
Bombyxin-IRs of brain and CA are present in VAMP2-, SNAP-25-, and Syntaxin 1A-IRs. VAMP2 and Syntaxin 1A are present in the PTTH neurons of the brain.
VAMP2-IRs and SNAP-25-IRs are in the center of the Bombyx brain. In the brains of all insects, the CX is a unique midline neuropil. The CX plays a key role in controlling spatial orientation and navigation. In addition to sensory-motor integration, Drosophila CX is required for spatial working memory during navigation tasks (Honkanen et al. 2019;Neuser et al. 2008). To control these diverse behaviors, neurotransmitters in the brain may be actively released via VAMP2 and SNAP-25. The CX is a focus of neuropeptidecontaining interneurons in the insect brain (Nassel and Homberg 2006). The transport of these neuropeptides in the CX may be related to VAMP2 and SNAP-25. Further study is necessary to clarify the importance of SNAREs in the CX, for example, whether SNARE mutants of B. mori alter diverse behaviors.
The expression of VAMP2 is restricted to neurons in the PI, DL, and CX of the Bombyx brain. The PI and DL of insects play important roles in neurosecretion. These regions send major neural processes to the corpora cardiaca and CA, the main neurohemal organ (Hartenstein 2006). VAMP2-IRs may regulate the fusion of secretory vesicles in these regions.
SNAP-25-IRs overlapped with VAMP2-IRs in the PI and CX of B. mori, but did not overlap in DL. In DL, other SNAP proteins may mediate the fusion of secretory vesicles. The SNAP-25 protein subfamily in insects consists of SNAP-25, SNAP-24, SNAP-29, and SNAP-47 (Kádková et al. 2019). These SNAP proteins may undergo membrane fusion in place of SNAP-25 in the DL. An antibody against Syntaxin 1A recognizes a restricted set of neurons in the PI and DL. Syntaxin 1A is a critical component of the SNARE complex and is thought to be essential for synaptic vesicle fusion, but in the central region, the other type of syntaxin may undergo vesicle fusion and release neurotransmitters. Many neuropeptides and neurotransmitters are synthesized in the PI and DL. Syntaxin 1A-IRs co-localize with VAMP2-IRs in the PI and DL. VAMP2 and Syntaxin 1A may mediate membrane fusion of neuropeptide-or neurotransmitter-filled secretory vesicles in the PI and DL.
Anti-VAMP2, anti-SNAP-25 and anti-Syntaxin 1A detected a restricted set of neurons in the CA section. Neuropeptides are synthesized in specific neurosecretory cells, transported along the axons, and secreted as complexes from the CA into the hemolymph. SNAREs may mediate the exocytosis of neuropeptides in CA.
Insulin and ILPs regulate numerous functions in insects, including growth, development, carbohydrate metabolism, and female reproduction (Nassel and Vanden Broeck 2016). SNARE proteins are also involved in insulin secretion by the β-cells of the pancreas (Hou et al. 2009;Wang and Thurmond 2009). We attempted to clarify the relationship between SNAREs and bombyxin secretion using doublestaining immunohistochemistry of the brain and CA of B. mori. SNAP-25-, VAMP2-, and Syntaxin 1A-IRs were co-localized with bombyxin-IRs in the CA and brain. Insulin-like peptides are synthesized in specific neurosecretory cells, transported along the axons, and secreted as complexes from the CA into the hemolymph. SNAP-25, VAMP2, and Fig. 7 Localization of SNAP-25-IRs, VAMP2-IRs, and Syntaxin 1A-IRs and PTTH in the brain and CA of Bombyx mori. Anti-PTTH detected subcellular regions in individual neurons in the brain (a, d, and g) and the CA (j and m). Anti-VAMP2 (b), anti-SNAP-25 (e), and anti-Syntaxin 1A (h) detected subcellular regions in individual neurons in the brain. Anti-VAMP2 (k) and anti-Syntaxin 1A (n) detected subcellular regions in individual neurons in the CA. PTTH-IRs co-localized with VAMP2-IRs and Syntaxin 1A-IRs in the brain (c and i, arrow) and are not present in the SNAP25-IRs of the brain (f). PTTH-IRs are not present in VAMP2-IRs and Syntaxin 1A-IRs in the CA (l and o). Scale bar, 100 μm (a-i) and 50 μm (j-o) Syntaxin 1A may mediate the membrane fusion of secretory vesicles containing bombyxin in the brain and CA. Further experiments are needed to determine whether SNARE mutants of B. mori alter bombyxin secretion.
PTTH is a pivotal regulator of the molting process and metamorphosis. PTTH was identified as a product of two pairs of neurons in the brain. The secretion of PTTH exhibits diurnal rhythms that are probably subject to regulation by a central clock. There is no report to clarify functional proteins to regulate PTTH secretion.
PTTH-IRs are present in VAMP2-IRs and Syntaxin 1A-IRs neurons in the brain, but did not overlap with SNAP-25-IRs. In CA, PTTH-IRs were not co-localized with VAMP2-IRs or Syntaxin-IR. In the brain, VAMP2 and Syntaxin 1A-IRs mediate the fusion of PTTH-filled secretory vesicles. The other synaptobrevin/VAMP may function as a v-SNARE in CA.
The release of PTTH into the hemolymph oscillates and is regulated by the circadian rhythm (Vafopoulou et al. 2012(Vafopoulou et al. , 2007. In Lepidoptera, PTTH-IRs resides in close proximity to period-IRs in the brain (Sauman and Reppert 1996). Serotonin or pigment dispersing factor in the clock cells may induce the release of PTTH. Further studies are needed to clarify how circadian-rhythm-dependent PTTH is regulated by VAMP2-mediated fusion in the brain.
PTTH and bombyxin may be related to SNARE proteins. Regulation of PTTH and bombyxin secretion is connected with the development of chemical compounds.