A Novel Alternatively Spliced Variant of Synaptotagmin VI Lacking a Transmembrane Domain

Synaptotagmins are a family of membrane proteins that are characterized by a single transmembrane region and tandem C2 domains and that are likely to regulate constitutive and/or regulated vesicle traffic. We have shown that a subclass of synaptotagmins (III, V, VI, and X) forms homo- and heterodimers through an evolutionarily conserved cysteine motif at their N termini (Fukuda, M., Kanno, E., and Mikoshiba, K. (1999) J. Biol. Chem. 274, 31421–31427). In this study, we identified a novel alternatively spliced variant of synaptotagmin (Syt) VI that lacks the N-terminal 85 amino acids including the transmembrane region (thus designated as Syt VIΔTM). Because it lacks the cysteine motif responsible for self-dimerization, Syt VIΔTM could not associate with Syt VI even in the presence of Ca2+. Despite lacking the transmembrane region, Syt VIΔTM can associate with the plasma membrane through the C-terminal 29 amino acids. In adult mouse brain, two closely comigrating bands at M r ∼50,000, which closely corresponded to the molecular weight of recombinant Syt VIΔTM, were detected by anti-Syt VI antibody. These immunoreactive bands were found in both soluble and membrane fractions of mouse brain, indicating that they are membrane-associated proteins (Syt VIΔTM), but not transmembrane proteins (Syt VI). Expression of Syt VI and Syt VIΔTM in PC12 or COS-7 cells indicated that the two molecules have a distinct subcellular distribution: Syt VIΔTM is present in the cytosol or is associated with the plasma membrane or internal membrane structures, whereas Syt VI is localized to the endoplasmic reticulum and/or Golgi-like perinuclear compartment. These results suggest that Syt VI and Syt VIΔTM may play distinct roles in vesicular trafficking.

Synaptotagmins are a family of membrane proteins that are suggested to be involved in regulated and/or constitutive vesicle traffic. All members share a short amino terminus, a single transmembrane region, and tandem C2 domains (named C2A and C2B) (reviewed in Refs. [1][2][3][4]. To date, a number of other molecules that contain tandem C2 domains have also been identified, including rabphilin 3A, Doc2, and synaptotagmin B/K, which are also suggested to be involved in vesicle traffic (reviewed in Refs. 1 and 5). However, the synaptotagmin family proteins are apparently distinguished from other tandem C2 domain proteins in that they have a single transmembrane region (1,5). At least 12 synaptotagmin isoforms have been identified in rat or mouse (6 -17, 48); three in electric ray (18); and only a single isoform corresponding to vertebrate synaptotagmin (Syt) 1 I in Drosophila (19), Caenorhabditis elegans (20), Aplysia (21), and squid (22). Although the roles of Syt I in Ca 2ϩ -regulated synaptic vesicle exocytosis and endocytosis have been well examined (Refs. 22-26 and reviewed in Refs. [1][2][3][4], isoforms that are involved in vesicle traffic other than secretory vesicle exocytosis have yet to be determined. It is also unknown whether synaptotagmin mRNAs are alternatively spliced because only a single isoform of synaptotagmins has been reported in rat or mouse to date. In the accompanying article (48), we cloned mouse Syt I-XI cDNAs and found that Syts III, V, VI, and X form stable homoand/or heterodimers via a conserved cysteine motif at the N terminus through disulfide bonds. In this study, we have identified a novel alternatively spliced variant of Syt VI (designated as Syt VI⌬TM) that lacks the N-terminal domain including the transmembrane region (amino acids 1-85), which falls outside the classical synaptotagmin category described above (1)(2)(3)(4). Because it lacks the conserved cysteine motif at the N terminus, Syt VI⌬TM did not interact with Syts III, V, VI, and X even in the presence of Ca 2ϩ . Expression of Syt VI and Syt VI⌬TM in PC12 or COS-7 cells indicated that the two molecules show distinct subcellular distribution: Syt VI is mainly localized to the endoplasmic reticulum (ER) and/or Golgi-like perinuclear compartment, whereas Syt VI⌬TM is localized to the plasma membrane, internal membrane structures, and cytosolic fraction. On the basis of these results, we discuss the functional differences between these two Syt VI proteins and a possible role for Syt VI in constitutive vesicle traffic, especially ER-to-Golgi or Golgi-to-ER vesicle transport.

EXPERIMENTAL PROCEDURES
Materials-ExTaq and AmpliTaq DNA polymerases were obtained from Takara Biomedicals and Perkin-Elmer, respectively. Polyclonal and monoclonal antibodies (M2) against the FLAG peptide were purchased from Zymed Laboratories Inc. and Sigma, respectively. Horseradish peroxidase-conjugated anti-T7 tag antibody and anti-His 6 antibody were from Novagen and Roche Molecular Biochemicals, * This work was supported in part by grants from the Science and Technology Agency of Japan. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The nucleotide sequence(s) reported in this paper has been submitted to the GenBank TM /EBI Data Bank with accession number(s) AB026809 and AB026810.
§ To whom correspondence should be addressed. respectively. All other chemicals were commercial products of reagent grade. Solutions were made in deionized water.
cDNA Cloning of Alternatively Spliced Variants of Mouse Synaptotagmin VI-Alternatively spliced variants of Syt VI were screened by reverse transcriptase-polymerase chain reaction (PCR) from mouse cerebellum cDNAs (8) using two sets of primers designed on the basis of rat sequences (14): N-terminal half, 5Ј-GCATGAGCGGAGTTTGG-3Ј (sense; amino acids 1-5) and 5Ј-CGAATTCAGTAGCGTACTGGATGT-CCT-3Ј (antisense; amino acids 351-357); and C-terminal half, 5Ј-CG-GATCCGCCGCCAAGAGCTGTGGGAA-3Ј (sense; amino acids 227-233) and 5Ј-GAATGAAATCACAACCG-3Ј (antisense; amino acids 510 -511 and 3Ј-noncoding regions). Reactions were carried out in the presence of Perfect Match PCR Enhancer (Stratagene) for 30 cycles, each consisting of denaturation at 94°C for 1 min, annealing at 50°C for 2 min, and extension at 72°C for 2 min. For the N-terminal half of Syt VI, a second reaction was run to highlight the difference in size between three splice variants using internal sense (5Ј-CGGGATCCAT-GAGCGGAGTTTGGGGGGCCG-3Ј, amino acids 1-8) and antisense (5Ј-TGATCTTCACAGCTGCCT-3Ј, amino acids 129 -135) primers. The cycling conditions were denaturation at 94°C for 1 min, annealing at 55°C for 1 min, and extension at 72°C for 1 min for 20 cycles. The PCR products, purified from an agarose gel by a MicroSpin column (Amersham Pharmacia Biotech), were directly inserted into the pGEM-T Easy vector (Promega). Both strands were completely sequenced by the Ther-moSequenase premixed cycle sequencing kit (Amersham Pharmacia Biotech) using a Hitachi SQ-5500 DNA sequencer.
Glass-bottomed dishes (35-mm dish; Mattek Corp., Ashland, MA) were coated with collagen type IV (Becton Dickinson Labware). PC12 cells were cultured on these dishes in Dulbecco's modified Eagle's medium containing 10% horse serum and 10% fetal bovine serum at 37°C and 5% CO 2 . Transfection was done using LipofectAMINE Plus reagent (Life Technologies, Inc.) according to the manufacturer's instructions. Transfection of various pEF-T7 (or FLAG)-Syt constructs into COS-7 cells was carried out by the DEAE-dextran method, and the expressed proteins were analyzed by immunoprecipitation following immunoblot-  ting as described (48).
Subcellular Fractionation of COS-7 Cells-COS-7 cells transfected with pEF-Syt constructs (5 ϫ 10 5 cells/10-cm dish) were harvested 3 days after transfection and homogenized in 1 ml of 0.32 M sucrose, 1 mM EDTA, 0.1 mM phenylmethylsulfonyl fluoride, 10 M leupeptin, 10 M pepstatin A, 1 mM ␤-mercaptoethanol, and 5 mM Tris-HCl, pH 7.5, in a glass-Teflon Potter homogenizer with 10 strokes at 900 -1000 rpm. The homogenate was centrifuged at 1000 ϫ g for 10 min at 4°C. The supernatant was further centrifuged at 100,000 ϫ g for 1 h at 4°C to precipitate membrane fractions. Equal proportions of supernatant and membrane fractions were subjected to 10% SDS-polyacrylamide gel electrophoresis (PAGE) and analyzed by immunoblotting as described (48). Subcellular fractionation of adult mouse olfactory bulb was similarly performed.
Production and Purification of Polyclonal Antibody against Syt VI-New Zealand White rabbits were immunized with the purified Syt VI C2A domain fused to glutathione S-transferase (amino acids 151-281) (29) by subcutaneous injection with RIBI adjuvant at intervals of 28 days. Antisera were collected after the third booster injection. Crude IgG fractions were obtained by adding an equal amount of saturated ammonium sulfate following centrifugation at 9000 rpm for 15 min at 4°C (GRX-220 high speed refrigerated centrifuge, TOMY, Saitama, Japan). The precipitates were dissolved in a minimum volume of phosphate-buffered saline (PBS) and then extensively dialyzed against PBS for 1 night. Since this crude IgG weakly recognized Syts I-III, V, IX, and X expressed in COS-7 cells (48), the cross-reactive component was removed by incubation with glutathione-Sepharose (wet volume 1 ml; Amersham Pharmacia Biotech) coupled to the glutathione S-transferase-Syt I-III, V, IX, and X C2A domain fusion proteins (Ͼ1 mg each) (30). Then, the anti-Syt VI antibody was affinity-purified by exposure to antigen bound to Affi-Gel 10 beads (Bio-Rad) according to the manufacturer's instructions. The protein concentration was determined by the Bio-Rad protein assay kit using bovine serum albumin as a reference. Immunoblotting was performed as described previously (48).
Immunocytochemistry-Three days after transfection, PC12 cells were washed twice with PBS and then fixed in 4% paraformaldehyde in 0.1 M sodium phosphate buffer for 20 min at room temperature, followed by washing with 0.1 M glycine. The fixed cells were permeabilized with 0.3% Triton X-100 in PBS for 2 min and immediately washed with blocking solution (1% bovine serum albumin and 0.1% Triton X-100 in PBS) three times for 5 min. The cells were incubated in blocking solution for 1 h at room temperature and then incubated with anti-BiP (Grp78, 1:200 dilution; Stressgen Biotech Corp.) and anti-TGN38 (1:500 dilution; Transduction Laboratories) mouse monoclonal antibodies and/or anti-Syt VI polyclonal antibody (0.38 g/ml) for 1 h at room temperature. Primary antibodies were washed out with blocking solution three times for 5 min; and then the cells were incubated with appropriate secondary antibodies, anti-Alexa 488 rabbit or anti-Alexa 568 mouse (Molecular Probes, Inc.), for 1 h at room temperature. After washing out the secondary antibodies with blocking solution five times for 5 min, immunoreactivity was analyzed using a fluorescence microscope (TE300, Nikon) attached to a laser confocal scanner unit (CSU 10, Yokogawa Electric Corp.) and a HiSCA CCD camera (C6790, Hamamatsu Photonics). Images were pseudo-colored and superimposed using Adobe Photoshop software (Version 4.0).

Identification of a Novel Alternatively Spliced Variant of
Synaptotagmin VI-To date, 12 isoforms of synaptotagmin (Syts I-XI and Srg1) have been reported in rat and mouse (6 -17, 48), but no splice variants have yet been described. To examine whether synaptotagmin genes produce splice variants, we divided the coding region of synaptotagmins into two parts (from the N terminus to the end of the C2A domain and from the C2A domain to the C terminus) and amplified them separately by reverse transcriptase-PCR using custom-designed oligonucleotides. The PCR products were purified from an agarose gel, subcloned into the pGEM-T Easy vector, and completely sequenced. We obtained several cDNAs containing insertions or deletions as compared with the original sequences previously reported ( Fig. 1) (data not shown). Among them, Syt VI has at least four splice variants, one major (ϳ85%) and three minor, in adult mouse cerebellum (Fig. 1). Sequence analysis of the four splice variants indicated that the longest PCR prod- In contrast, Syt VI was always localized to the membrane fractions independent of the presence of 1 M NaCl even when the blots were overexposed (middle and lower panels), but the localization of Syt VI⌬TM protein produced by alternative initiation of translation was also sensitive to NaCl concentration (arrowhead). The positions of M r markers (ϫ10 Ϫ3 ) are shown on the left. ucts (Fig. 1, asterisk) contained 47-base pair insertions just upstream of the transmembrane region of Syt VI ( Fig. 2A,  arrowhead). However, since the content of this isoform was quite low and the insertion might be an unspliced intron, we did not examine this form of Syt VI further. The second longest Syt VI cDNA corresponds to the rat Syt VI previously described (14), and the two shorter forms were produced by alternative splicing around the transmembrane region, which resulted in a frameshift of the protein translation ( Fig. 2A). We therefore designated the middle and shortest Syt VI splice variants as Syt VI⌬TM1 and Syt VI⌬TM2, respectively. Syt VI⌬TM2 was generated by the splicing out of 83 base pairs from Syt VI, and Syt VI⌬TM1 included another exon containing 66 base pairs at the same position. This frameshift generated new 5Ј-in-frame stop codons in the Syt VI⌬TM1 and Syt VI⌬TM2 cDNAs ( Fig.  2A, asterisk). However, the sequence around the putative first methionine residue of Syt VI⌬TM1 and Syt VI⌬TM2 (methionine at position 86 of Syt VI) did not correspond well to the Kozak sequence (28).
Expression of Syt VI and Syt VI⌬TM in COS-7 Cells-To elucidate whether Syt VI⌬TM1 and Syt VI⌬TM2 cDNAs are translated into proteins, we expressed C-terminal His 6 -tagged Syt VI, Syt VI⌬TM1, or Syt VI⌬TM2 (Fig. 2B) in COS-7 cells. Total cell homogenates were subjected to SDS-PAGE, followed by immunoblot analysis using anti-His 6 antibody. Two immunoreactive bands (apparent M r ϭ 60,000 and 50,000) were detected in the cells transfected with pEF-Syt VI-His; one band (apparent M r ϭ 50,000) was detected in the cells transfected with pEF-Syt VI⌬TM1-His or pEF-Syt VI⌬TM2-His; but no bands were seen in control cells (Fig. 3). The apparent M r values of the two bands were almost identical to the calculated M r values of 58,300 (Syt VI-His) and 49,400 (Syt VI⌬TM-His), respectively. This result indicates that the Syt VI⌬TM1 and Syt VI⌬TM2 cDNAs are indeed translated into the same protein (hereafter simply designated as Syt VI⌬TM protein), and Syt VI cDNA is most likely to have alternative initiation of translation, which produces the two proteins (Syt VI and Syt VI⌬TM), although we could not completely rule out the possibility that the M r 50,000 band seen in Fig. 3 (lane 2) was a degradation product of Syt VI-His.
Association of Syt VI⌬TM with Membrane Fractions-We initially thought that Syt VI⌬TM protein would be present in the cytosol because it lacks the transmembrane region as well as putative palmitoylation sites (cysteines at positions 68 and 84 of Syt VI) (31,32). However, the subcellular fractionation of COS-7 cells transiently expressing pEF-Syt VI-His or pEF-Syt VI⌬TM2-His indicated that both proteins were membrane-associated, although small amounts of Syt VI⌬TM protein (Ͻ5%) were present in the soluble fraction (Fig. 4A). The membraneassociated Syt VI⌬TM protein was easily released from the membrane by incubation with buffer containing 1 M NaCl, whereas Syt VI protein was tightly associated with the membrane fractions even in the presence of 1 M NaCl (Fig. 4B). Interestingly, Syt VI⌬TM protein derived from pEF-Syt VI-His was also sensitive to NaCl concentration (Fig. 4B, lower panel,  arrowhead). Thus, we concluded that Syt VI is an integral membrane protein, whereas Syt VI⌬TM is a peripheral membrane protein.
To delineate which region of Syt VI⌬TM is responsible for the membrane association, we produced five deletion mutants of Syt VI⌬TM (Fig. 5A). Deletion of the N-terminal region up to 137 amino acids (Syt VI⌬N1-N4) had almost no effect on the membrane association, but deletion of only 29 amino acids at the C terminus (Syt VI⌬C1) resulted in reduced membrane association (Ͼ60% of the protein present in the soluble fraction) (Fig. 5B). About 40% of Syt VI⌬C1 protein probably associates with phospholipids through the two C2 domains that are thought to bind negatively charged phospholipids (8,14,29,33).  (48), we showed that Syt VI can form stable homodimers through the N-terminal cysteine residue (at position 12) in cells and that substitution of this residue for alanine abolishes homodimer formation. Because Syt VI⌬TM lacks this cysteine residue, it could not form an SDS-resistant homodimer (Figs. 3 and 4A) or a ␤-mercaptoethanol-sensitive homodimer (Fig. 6). In the next set of experiments, we sought to determine whether Syt VI⌬TM functions in concert with or independently of Syt VI because it has been reported that Syt I or II self-dimerizes via C2B domains in the presence of Ca 2ϩ (34 -38). To this end, we used a T7-and FLAG-Syt coexpression assay as described previously (48). When pEF-T7-Syt VI and pEF-FLAG-Syt III, V, VI, or X were coexpressed in COS-7 cells, T7-Syt VI and FLAG-Syt VI proteins tightly associated with each other through disulfide bonds (Fig. 7, lanes 11-20), consistent with our previous article (48). This association was independent of the presence of 250 M Ca 2ϩ . In contrast, when pEF-T7-Syt VI⌬TM and pEF-FLAG-Syt III, V, VI, or X were coexpressed, T7-Syt VI⌬TM essentially did not associate with FLAG-Syt III, V, VI, or X, even in the presence of 250 M Ca 2ϩ . Only a weak interaction of FLAG-Syt X with T7-Syt VI⌬TM could be detected in the presence of 250 M Ca 2ϩ (Fig. 7, lanes  1-10). Furthermore, when coexpressed with pEF-T7-Syt VI, pEF-FLAG-Syt VI, and pEF-FLAG-Syt VI⌬TM2 in COS-7 cells, T7-Syt VI protein was preferentially coprecipitated with FLAG-Syt VI, but not FLAG-Syt VI⌬TM, even in the presence of Ca 2ϩ (data not shown).

Syt VI⌬TM Is Not Associated with Syt VI in Vitro-In the accompanying article
Expression of Syt VI⌬TM in Mouse Brain-While this work was in review, Butz et al. (39) reported that Syt VI protein was highly expressed in olfactory bulb, but was expressed at rather low levels in other brain regions. We produced a specific antibody against Syt VI and confirmed their results (data not shown). The anti-Syt VI antibody recognized two closely comigrating bands with apparent M r values of 50,000 in mouse olfactory bulb (Fig. 8, upper panel, lane 1), consistent with the results of Butz et al. (39). The apparent M r values of these two bands closely corresponded to the M r of recombinant Syt VI⌬TM expressed in COS-7 cells, but were apparently lower than that of recombinant Syt VI as well as the calculated M r of Syt VI. A subcellular fractionation study indicated that the two immunoreactive bands were present in both the soluble and membrane fractions (Fig. 8, upper panel, lanes 2 and 3) and were further released from the membrane fraction by 1 M NaCl  treatment (upper panel, lane 4), whereas Syt I, an integral membrane protein of synaptic vesicles, was not released from the membrane even in the presence of 1 M NaCl (lower panel). In addition, these two bands could not form a ␤-mercaptoethanol-sensitive homodimer as reported previously (48) (data not shown). These results indicated that the two immunoreactive bands detected by the anti-Syt VI antibody were membraneassociated proteins, most likely Syt VI⌬TM. The slight difference in M r between the two bands may be due to the difference  1-10) or pEF-T7-Syt VI (lanes [11][12][13][14][15][16][17][18][19][20] and pEF-FLAG-Syt III (lanes 1, 6, 11, and 16), pEF-FLAG-Syt V (lanes 2, 7, 12, and 17), pEF-FLAG-Syt VI (lanes 3, 8, 13, and 18), pEF-FLAG-Syt X (lanes 4, 9, 14, and 19), or control vector (lanes 5, 10, 15, and 20) were cotransfected into COS-7 cells by the DEAE-dextran method (48). Expressed T7-Syt VI or T7-Syt VI⌬TM proteins were immunoprecipitated in the presence of 2 mM EGTA or 250 M Ca 2ϩ by anti-T7 tag antibody-conjugated agarose (Novagen) and analyzed by immunoblotting as described (48). Coprecipitated FLAG-Syt constructs were detected by anti-FLAG antibody (upper panel). The same blots were stripped and reprobed with horseradish peroxidase-conjugated anti-T7 tag antibody (lower panel). Although a Ͼ20-fold greater amount of T7-Syt VI⌬TM than T7-Syt VI was loaded, T7-Syt VI⌬TM essentially did not associate with FLAG-Syts III, V, VI, and X even in the presence of 250 M Ca 2ϩ due to the lack of the conserved N-terminal cysteine motif responsible for stable dimerization (48). In contrast, T7-Syt VI strongly associated with FLAG-Syt III, V, VI, or X through the disulfide bond as described (48). Differential Distribution of Syt VI and Syt VI⌬TM in PC12 Cells-In the final set of experiments, we compared the subcellular localization of Syt VI and Syt VI⌬TM transiently expressed in PC12 cells. Syt VI⌬TM protein was mostly associated with the plasma membrane and was also found throughout cell body (Fig. 9A). This plasma membrane association required the C-terminal 29 amino acids (Fig. 9C), but not the N-terminal 137 amino acids (Fig. 9B). As judged from the subcellular fractionation study in Figs. 4 and 5, Syt VI⌬TM protein in the cell body is associated with the internal membrane structures, and a small part of Syt VI⌬TM is present in the cytosol. In contrast, Syt VI protein was localized to the tubule-like structures (Fig. 9D), Golgi-like perinuclear compartment (Fig. 9J), or both (Fig. 9G). The weak signals around the plasma membrane may be Syt VI⌬TM protein that is generated by alternative initiation of translation from pEF-Syt VI because N-terminal T7-or FLAG-tagged Syt VI was hardly detected around the plasma membrane when Syt VI protein was visualized by anti-T7 (or FLAG) antibody (data not shown). This tubule-like structure precisely overlapped that stained by anti-BiP antibody, a marker for the ER (Fig. 9, E and F, in yellow); and the perinuclear signals overlapped well with that of TGN38, a marker for the trans Golgi network (Fig. 9, H-K). Similar results were obtained in COS-7 cells (data not shown), suggesting that the subcellular localization of Syt VI and Syt VI⌬TM is probably independent of the cell type. However, higher resolution studies are necessary to determine the exact localization of Syt VI and Syt VI⌬TM proteins and to understand their function in vesicle traffic. DISCUSSION Synaptotagmins are classified as a family of tandem C2 domain proteins that contain a short amino terminus and a single transmembrane region (1)(2)(3)(4). However, this study, which searched for alternatively spliced variants of Syt VI, demonstrated the presence of a novel isoform lacking the transmembrane region (Syt VI⌬TM), which therefore falls outside the classical synaptotagmin category. Syt VI⌬TM protein is produced by two different alternatively splicing events that occur just around the transmembrane region and also by alternative protein translation from the Syt VI mRNA originally reported (Fig. 3). Syt VI⌬TM was shown to be predominantly expressed in adult mouse brain by reverse transcriptase-PCR ( Fig. 1) and immunoblot analyses (Fig. 8). We also identified a similar variant of Syt V, which is spliced out around the transmembrane region. 2 Thus, it will be interesting to examine in the future whether this type of alternative splicing is a unique event within a subclass of synaptotagmins (III, V, VI, and X).
In the previous article (48), we showed that a subclass of synaptotagmins (III, V, VI, and X) forms hetero-and homodimers through a conserved N-terminal cysteine motif. Since Syt VI⌬TM lacks this motif, it is unlikely to associate with Syt VI in cells. This was also confirmed by in vitro binding experiments (Fig. 7) and differential distribution of Syt VI and Syt VI⌬TM transiently expressed in PC12 or COS-7 cells (Fig.  9), suggesting that the two molecules may have different functions.
Recently, it was reported that the cytoplasmic domain of Syt I produced by proteolysis (40) is involved in fibroblast growth factor-1 release from NIH 3T3 cells (41)(42)(43). Since Syt VI mRNA is more widely distributed than Syt I mRNA (14) and since Syt VI⌬TM protein is mostly localized around the plasma membrane (Fig. 9A), it is possible that Syt VI⌬TM might be involved in fibroblast growth factor-1 release, especially in non-neuronal tissues. The association of Syt VI⌬TM protein 2 M. Fukuda, unpublished observations. with the plasma membrane requires a C-terminal tail. The target site of Syt VI⌬TM is under investigation, but one report has indicated that the presynaptic protein neurexin I␣ can bind the C termini of synaptotagmin family proteins (44). Further work is required to elucidate this point.
In ER-to-Golgi transport, Ca 2ϩ is an essential component at a stage between vesicle docking and the actual membrane fusion event (45)(46)(47). The optimal Ca 2ϩ concentration required for ER-to-Golgi transport was reported to be 0.01-0.1 M (45). This range corresponds well with the EC 50 values for the Ca 2ϩdependent phospholipid binding measurement of the Syt VI C2A domain in vitro (29). In addition, Syt VI was localized to the ER and/or Golgi-like perinuclear compartment in PC12 cells (Fig. 9, D-K) and COS-7 cells, but other synaptotagmin isoforms did not show such distribution when transiently expressed in PC12 cells. 2 Thus, it is tempting to speculate that Syt VI is a specific synaptotagmin isoform that may be involved in this Ca 2ϩ -dependent event.
In summary, we have identified a novel alternatively spliced variant of Syt VI (named Syt VI⌬TM) that shows a distinct subcellular localization compared with Syt VI. Our findings suggest that such alternative splicing can produce divergent synaptotagmin isoforms.