Complexin II Facilitates Exocytotic Release in Mast Cells by Enhancing Ca2+ Sensitivity of the Fusion Process

Research Article 2239 Complexin II facilitates exocytotic release in mast cells by enhancing Ca2+ sensitivity of the fusion process

Satoshi Tadokoro, Mamoru Nakanishi and Naohide Hirashima* Graduate School of Pharmaceutical Sciences, Nagoya City University, Tanabe-dori, Mizuho-ku, Nagoya 467-8603, Japan *Author for correspondence (e-mail: [email protected])

Accepted 18 February 2005 Journal of Cell Science 118, 2239-2246 Published by The Company of Biologists 2005 doi:10.1242/jcs.02338

Summary Recent studies have shown that soluble N-ethyl maleimide- distributed throughout the cytoplasm before antigen sensitive factor attachment protein receptor (SNARE) stimulation. However, the distribution of complexin II proteins are involved in exocytotic release in mast cells as changed dramatically with stimulation and it became in neurotransmitter release. However, the roles of the localized on the plasma membrane. This change in the proteins that regulate the structure and activity of SNARE intracellular distribution was observed even in the absence proteins are poorly understood. Complexin is one such of extracellular Ca2+, while exocytotic release was inhibited regulatory protein and is involved in neurotransmitter almost completely under this condition. The degranulation release, although ideas about its role are still controversial. induced by phorbol 12-myristate 13-acetate and A23187 In this study, we investigated the expression and role depended on the extracellular Ca2+ concentration, and its of complexin in the regulation of exocytotic release sensitivity to Ca2+ was decreased in knockdown cells. These (degranulation) in mast cells. We found that complexin II, results suggest that complexin II regulates exocytosis but not complexin I, is expressed in mast cells. We obtained positively by translocating to the plasma membrane and RBL-2H3 cells that expressed a low level of complexin II enhancing the Ca2+ sensitivity of fusion machinery, and found that antigen-induced degranulation was although this translocation to the plasma membrane is not suppressed in these cells. No signiﬁcant changes in the sufficient to trigger exocytotic membrane fusion. Ca2+ response or expression levels of syntaxins and synaptotagmin were observed in knockdown cells. An

Journal of Cell Science immunocytochemical study revealed that complexin II was Key words: Mast cell, Complexin, Exocytosis, SNARE, Allergy, Rat

Introduction higher order structure that is required for the fusion of synaptic It has been shown that soluble N-ethyl maleimide-sensitive vesicles. They also showed that peptides that prevent factor attachment protein receptor (SNARE) proteins play an complexin from binding to the SNARE complex inhibit evoked essential role in exocytotic release in both neuronal cells transmitter release in a squid giant synapse. Double-knockout (Sollner et al., 1993; Calakos and Scheller, 1996; Brunger, mice for complexins I and II show reduced neurotransmitter 2001) and non-neuronal secretory cells (Wheeler et al., 1996; release (Reim et al., 2001). These results suggest that Nagamatsu et al., 1999; Reed et al., 1999; Flaumenhaft, 1999). complexin acts as a positive regulator of exocytosis. However, In addition to SNARE proteins, several proteins that regulate the injection of complexin into Aplysia nerve terminal the conformation and activity of SNARE complexes are suppressed transmitter release, while the injection of anti- involved in exocytosis. Complexin (also called synaphin), a complexin antibody stimulated neurotransmitter release (Ono small soluble protein (18-19 kD), is a regulatory protein in the et al., 1998). The overexpression of complexin in PC12 mammalian brain (McMahon et al., 1995; Takahashi et al., reduced exocytotic release (Itakura et al., 1999). As these 1995; Ishizuka et al., 1995). Complexin interacts with ternary ﬁndings show, it is still unclear how complexin functions in SNARE complex and is thought to stabilize the SNARE exocytotic release, even in neuronal cells. Furthermore, it is not complex (Pabst et al., 2000; Pabst et al., 2002). Based on clear whether or not complexin is involved in exocytosis in studies of the three-dimensional structure of the complexin/ non-neuronal cells. If complexin regulates exocytosis in non- SNARE complex, it has been suggested that complexin neuronal secretory cells, it would be very interesting to stabilizes the fully assembled SNARE complex (Bracher et al., investigate how complexin regulates exocytosis in such cells. 2002; Chen et al., 2002). However, the role of complexin in As non-neuronal secretory cells are often larger and have neurotransmitter release is not yet fully understood. Using a bigger secretory granules than nerve terminals, they could squid giant synapse, Tokumaru et al. (Tokumaru et al., 2001) provide a useful experimental system for investigating the reported that complexin facilitates the association of SNAREs mechanism by which complexin regulates exocytosis. into an intermediate complex that can oligomerize to give a In the present study, we examined the expression and role 2240 Journal of Cell Science 118 (10)

of complexin in mast cells, which are typical non-neuronal (anti-sense) and 5′-ATGGACTTCGTCATGAAGCA (sense)/5′-TTA- secretory cells. In mast cells, cross-linking of high-affinity CTTCTTGAACATGTCCTGCA-3′ (anti-sense), respectively. PCR receptors for IgE (FcεRI) by multivalent antigen activates an products were extracted from agarose gel with Gene Clean (Bio 101) intracellular signaling cascade and leads to the exocytotic and subcloned into the TA cloning vector pCRII (Invitrogen). Cloned release of granular contents (degranulation), resulting in PCR products were sequenced with a DSQ1000 DNA sequencer allergic responses (Abraham and Malaviya, 1997; Swann et (Shimadzu, Kyoto, Japan) using FITC-labeled M13 universal primer. al., 1998; Turner and Kinet, 1999; Kinet, 1999). We have previously studied the mechanism of degranulation in mast Western blotting cells and identified the protein and lipid molecules that regulate RBL-2H3 or P815 cells (5 106) were lysed with lysis buffer (10 this process (Hibi et al., 2000; Kato et al., 2002; Kato et al., mmol/l HEPES, 1% Triton-X100, 1 mmol/l EDTA, 50 mmol/l NaF, 2003). We and other groups have reported that SNARE 2.5 mmol/l p-nitrophenyl phosphate, 1 mmol/l Na3VO4, 1 mmol/l proteins are involved in degranulation in mast cells (Guo et al., PMSF, 10 µg/ml leupeptin, 10 µg/ml pepstatin A, 10 µg/ml aprotinin). 1998; Hibi et al., 2000; Paumet et al., 2000; Blank et al., 2001; After centrifugation at 23,000 g for 20 minutes, supernatant was mixed Blank and Rivera, 2004), although the active isoforms in mast with an equal volume of Laemmli sample buffer and boiled for cells are different from those in neuronal cells. However, there 5 minutes. For cellular lysate of rat brain, rat cerebrum lysate is little information available on the proteins that regulate was purchased from Transduction Laboratories. Samples were electrophoresed by SDS-PAGE and transferred to a PVDF membrane. SNARE proteins in mast cells, except for Munc18-2 (Martin- After blocking with phosphate buffer containing 5% skimmed milk, Verdeaux et al., 2003). blots were probed with primary antibody for 1 hour. As primary In this study, we found that complexin II was expressed in antibodies, anti-complexin I antibody (dilution 1:200; Santa Cruz mast cells. In experiments using complexin II knockdown cells, Biotechnology), anti-complexin II antibody (dilution 1:500; we found that complexin II positively regulated exocytotic Transduction Laboratories), anti-syntaxin-3 antibody (dilution 1:250; release in mast cells. Immunocytochemical experiments Alomone Labs, Israel), anti-syntaxin-4 antibody (dilution 1:500; revealed that complexin II changed its localization from the Santa Cruz Biotechnology), anti-synaptotagmin II antibody (dilution cytoplasm to the plasma membrane, and this translocation 1:200 Santa Cruz Biotechnology), and β-actin antibody (dilution occurred in the absence of extracellular Ca2+, while 1:4000; Sigma) were used. After being washed with 0.1% Tween 20 degranulation was inhibited almost completely. We also found in PBS, membrane was treated with anti-mouse IgG conjugated with horseradish peroxidase. Immunoreactivity was detected by enhanced that the degranulation induced by phorbol 12-myristate 13- 2+ chemiluminescence (ECL, Amersham Pharmacia) with a LAS-1000 acetate (PMA) and A23187 depended on extracellular Ca (FUJI FILM, Tokyo Japan) and analyzed by Image Gauge (FUJI 2+ concentration, and its sensitivity to Ca was decreased in FILM). knockdown cells. These results suggest that complexin II regulates exocytosis positively by translocating to the plasma membrane and enhancing the Ca2+ sensitivity of fusion Plasmid construction and transfection machinery, although the association of complexin II with Poly(A)+ RNA was obtained as described above. For the knockdown SNARE complex is not sufficient to trigger exocytotic of complexin II, 5′-GGATCCATGGACTTCGTCATGAAGCA-3′ ′ membrane fusion. (sense; BamHI site is underlined)/5 -GCGGCCGCTTACTTCTTG- Journal of Cell Science AACATGTCCTGCA-3′ (anti-sense; NotI site is underlined) was used as a primer pair. For the expression of myc-tagged complexin II, 5′- GGATCCATGGACTTCGTCATGAAGCA-3′ (sense; BamHI site Materials and Methods is underlined)/5′-GAATTCTTACTTCTTGAACATGTCCTGCA-3′ Chemicals (anti-sense; EcoRI site is underlined) was used as a primer pair. PCR PMA, 4-bromo-A23187, (±)-sulfinpyrazone, p-nitrophenyl-N-acetyl- products were extracted from agarose gel with Gene Clean (Bio 101) β-D-glucosaminide and thapsigargin were purchased from Sigma (St and subcloned into the TA cloning vector pCRII. Cloned PCR Louis, MO). Aprotinin, leupeptin, pepstatin and PMSF were obtained products were sequenced with a DSQ1000 DNA sequencer. Verified from Wako Pure Chemicals (Tokyo, Japan). All other reagents were cDNA was ligated to pcDNA3 (Invitrogen) in the antisense direction of the highest grade available commercially. to knockdown the expression of complexin II. For the expression of myc-tagged complexin II, cDNA was ligated to pCMV-Tag5 (Stratagene). RBL-2H3 cells (5 105 cells/500 µl) were electroporat- Cell culture ed in cold PBS with 40 µg of plasmid DNA at 250 V and 950 µF using A mast cell line, rat basophilic leukemia cell (RBL-2H3), was cultured Gene Pulser II (Bio-Rad) (Hibi et al., 2000). Stable clones with in Eagle’s minimal essential medium from Nissui (Tokyo, Japan) with reduced expression of complexin II were selected by G418 (500 10% fetal calf serum (Boehringer Mannheim) at 37°C in an µg/ml) and western blotting with anti-complexin II or anti-myc atmosphere of 5% CO2. Mastocytoma P815 was cultured in antibody. RPMI1640 (Gibco) supplemented with 10% fetal calf serum at 37°C in an atmosphere of 5% CO2. Assay of secreted β-hexosaminidase Degranulation of RBL-2H3 cells was monitored by measuring the RT-PCR activity of a granule-stored enzyme, β-hexosaminidase, secreted in Poly(A)+ RNA was obtained with a QuickPrep Micro mRNA cell supernatant (Amano et al., 2001). Cells were sensitized by anti- Purification Kit (Amersham Pharmacia Biotech) from 1 107 cells of DNP IgE for 30 minutes and incubated with an average of six DNP RBL-2H3 and cerebrum of Sprague-Dawley rat (6 weeks), and served groups conjugated with BSA (DNP6-BSA) for 30 minutes at 37°C. as a template for cDNA synthesis with SuperScript II RT (Gibco Aliquots of supernatant were incubated with substrate solution (2 BRL), as reported previously (Hibi et al., 2000). The primer pairs used mmol/l p-nitrophenyl-N-acetyl-β-D-glucosaminide in 100 mmol/l to amplify complexins I and II were 5′-ATGGAGTTCGTGAT- citrate, pH 4.5) for 1 hour at 37°C. After the reaction was terminated GAAACAAG-3′ (sense)/5′-TTACTTCTTGAACATGTCCTGCA-3′ with Na2CO3-NaHCO3 buffer, absorbance at 405 nm was measured. Complexin facilitates exocytosis in mast cell 2241

Release activity relative to the total β-hexosaminidase content of the cells was calculated. Total β-hexosaminidase content was determined by dissolving cells with 0.1% Triton-X100.

Intracellular Ca2+ measurement RBL-2H3 cells were loaded with Fura2/AM (Molecular Probes; Eugene, OR) for 30 minutes at 37°C and washed twice with HEPES- buffered saline (140 mmol/l NaCl, 5 mmol/l KCl, 1 mmol/l CaCl2, 0.6 mmol/l MgCl2, 0.1% glucose, 0.1% BSA, 0.1 mg/ml sulfinpyrazone and 10 mmol/l HEPES, pH 7.4). For experiments with different 2+ extracellular Ca concentration, concentration of CaCl2 in HEPES- buffered saline was varied from 0 to 2 mmol/l. Cells were sensitized with anti-DNP IgE for 30 minutes and stimulated with DNP6-BSA (100 ng/ml). For the stimulation with Ca2+ ionophore and phorbol ester, A23187 (1 µmol/l) and PMA (50 ng/ml) were used instead of DNP6-BSA. The fluorescence intensities with excitation at 340 and 360 nm were measured, and the ratio (F340/F360) and Ca2+ concentration were calculated by a spectrofluorometer linked to a personal computer (RF-5300PC; Shimadzu, Japan), following a procedure described previously (Grynkiewicz et al., 1985). During measurement of the intracellular Ca2+ concentration, cells were kept at 37°C.

Immunocytochemistry Cells (1 105) were plated in a ZOG-3 glass-bottom chamber (Elekon Science; Chiba, Japan) and incubated for 18 hours. After cells were washed with PBS, they were fixed with PBS containing 4% paraformaldehyde and permeabilized with PBS containing 0.2% Fig. 1. Expression of complexin II in RBL-2H3 cells. (A) RT-PCR Triton-X100. After blocking with 5% BSA in PBS, cells were products amplified with specific primer pairs for complexins I and II incubated with anti-complexin II antibody (dilution 1:100; were electrophoresed in agarose gel (1.5%). Far left and far right lanes Transduction Laboratories) or anti-myc antibody (9E10, dilution 1:50; are for the 100 bp ladder marker (M). PCR products amplified with Santa Cruz Biotechnology) at 4°C overnight. Cells were treated with primer pairs of complexin I (CPX I) and II (CPX II) using templates fluorescence-labeled secondary antibody (FITC-conjugated anti- derived from rat brain (positive control) and RBL-2H3 cells are shown. mouse IgG) after they were washed three times with PBS for 10 The expected size of the product is about 400 bp for both isoforms. A minutes. Fluorescent images were taken with a confocal laser single band in the lane for complexin II of RBL-2H3 cells was scanning microscope (Zeiss, LSM-510) with a 63 objective lens detected, and the product was identified as complexin II by DNA

Journal of Cell Science (Plan-Apochromat 63/1.4 oil). Samples were excited at 488 nm with sequencing. (B) Western blot analysis for complexin. Cell lysates were an Ar laser and fluorescence was observed with an LP560 filter (Kato prepared from RBL-2H3 cells, P815 cells and rat cerebrum and were et al., 2002). electrophoresed by SDS-PAGE. After the samples were transferred to a PVDF membrane, blots were probed with primary antigens specific for complexins I and II. Blots were visualized with anti-mouse IgG conjugated with horseradish peroxidase using chemiluminescence Results methods. Complexin II was detected at about 19 kD (lower panel), but Expression of complexin in mast cells complexin I was not detected (upper panel) in mast cells. The expression of complexin in RBL-2H3 cells (a mast cell (C) Intracellular distribution of complexin II. Complexin II was detected with anti-complexin II antibody and visualized with FITC- line) was investigated by RT-PCR. As there are two isoforms conjugated anti mouse IgG. Fluorescent images were collected with a of complexin, complexins I and II, we performed PCR with confocal laser scanning microscope. Complexin II was detected in the primers specific to complexins I and II. The PCR products cytoplasm and the nucleus. amplified with the primer pairs for complexins were electrophoresed, and a clear band at the expected size was detected only in the lane for complexin II (Fig. 1A). The band To investigate the intracellular distribution of complexin II was excised from agarose gel and the extracted PCR product in mast cells, we carried out immunocytochemical experiments was sequenced. The sequence was identical to the cDNA using an anti-complexin II antibody and an FITC-labeled sequence of rat complexin II. When RT-PCR was carried out secondary antibody. Complexin II resided throughout the using mRNA from rat brain as a template, expressions of cytoplasm and nucleus, and no localization in a distinct complexin I and II were detected. The expression of complexin organelle was observed. However, the distribution of was confirmed at the protein expression level by western fluorescence signal seemed to be punctate rather than uniform blotting. As shown in Fig. 1B, the expression of complexin II (Fig. 1C). was detected at about 19 kD (lower panel), while the expression of complexin I was not detected (upper panel). We also investigated the expression of complexin II in mastocytoma, Characterization of complexin II-knockdown cell P815. As in the case of RBL-2H3, complexin II but not I was To investigate the role of complexin II in exocytotic release detected in P815 (Fig. 1B). in mast cells, we obtained RBL-2H3 cells that expressed a 2242 Journal of Cell Science 118 (10)

low level of complexin II. After selection with G418, four to Ca2+ influx from the extracellular medium. Fig. 2C shows independent clones with a low expression of complexin II the timecourse of the intracellular Ca2+ concentration after (knockdown cells) were picked up by western blotting antigen stimulation in wild-type cells. Fig. 2D shows the analysis. All these clones expressed complexin II at less than timecourse of the average Ca2+ concentration in four 50% of the level in wild-type cells, and they all exhibited independent knockdown clones. No significant difference was similar behaviors in this study. Fig. 2A shows western observed in either phase of the Ca2+ increase between wild- blotting of cell lysate derived from four knockdown clones. type and knockdown cells. This suggests that the reduction in Expression levels of complexin II in knockdown clones (kd- the expression of complexin II had no effect on events 1, 2, 3 and 4) were 42, 46, 40 and 38% of those in wild-type upstream of Ca2+ influx, such as the activation of IgE receptors, cells, respectively. The expression levels of syntaxin-3 and tyrosine phosphorylation or opening of Ca2+ channels on the syntaxin-4, which are SNARE proteins expressed in RBL- Ca2+ store and plasma membrane. 2H3 cells and thought to be involved in degranulation, were The exocytotic release of complexin knockdown cells was not significantly changed in knockdown cells (Fig. 2A). In investigated by quantifying β-hexosaminidase secreted in addition, the expression of synaptotagmin II, which is the medium. Fig. 3A shows the timecourse of degranulation most abundant isoform of synaptotagmin and regulates induced by antigen stimulation. A significant inhibition exocytotic release in RBL (Baram et al., 1999), was not of degranulation was observed in complexin II knockdown affected (Fig. 2A). The expression of complexin II was cells. There was no difference in the total amount of β- attenuated in knockdown cells, but its intracellular hexosaminidase between wild-type and knockdown cells. distribution was not changed, and complexin II was Because the cellular response of RBL-2H3 sometimes varies distributed throughout the cytoplasm and the nucleus, as in individual cells, a selected clone by antibiotics and Western observed in wild-type cells (Fig. 2B). blotting could show impaired degranulation regardless of the As degranulation is triggered by the elevation of the expression level of complexin II. Thus, we stimulated cells intracellular Ca2+ concentration due to Ca2+ influx from the with a Ca2+ ionophore A23187 and PMA, which mimic the extracellular solution into mast cells, we examined the Ca2+ activation of mast cells by bypassing events before Ca2+ influx. mobilization evoked by antigen in complexin II-knockdown In this stimulation condition, degranulation was induced with cells. It has been well established that antigen stimulation of minimum effects of clonal variance. Furthermore, we can mast cells causes an initial Ca2+ increase due to release from confirm the notion that complexin II regulates processes intracellular Ca2+ stores followed by a sustained increase due downstream of Ca2+ influx. As shown in Fig. 3B, degranulation was inhibited in knockdown cells, as observed in the case of antigen stimulation (Fig. 3A). These results suggest that complexin II positively regulates exocytosis in mast cells.

Translocation of complexin II after stimulation Journal of Cell Science To elucidate the mechanism by which complexin II regulates exocytosis in mast cells, the intracellular localization of complexin II before and after antigen stimulation was observed. Before stimulation, complexin II was distributed throughout the cell as described in Fig. 1. However, upon stimulation, the distribution of complexin II changed dramatically, and it was translocated to the plasma membrane (left images

Fig. 2. Characterization of complexin II-knockdown cells. (A) Western blot analysis for complexin II, syntaxins 3, syntaxin 4 and synaptotagmin II in four knockdown clones (kd-1 to kd-4). Expression of complexin II was reduced in knockdown cells, while expression of syntaxins and synaptotagmin II was not affected. (B) Intracellular distribution of complexin II in a kd cell. The fluorescence image was obtained as described in Fig. 1C. (C,D) Timecourse of the intracellular Ca2+ concentration in wild-type (C) and kd cells (D), respectively. Cells were sensitized with IgE and loaded with Fura2-AM, and stimulated with antigen (DNP-BSA) at the time indicated by an arrow. Average fluorescence intensity ratios (F340/F360) were plotted against time. Values are obtained from four independent preparations of wild-type cells and four independent knockdown clones [mean±s.e.m. (n=4)]. No significant differences in the Ca2+ response were detected between wild-type and kd cells. Complexin facilitates exocytosis in mast cell 2243

20 release (%)

0 010015 time (s)

B 40

20 release (%)

0 01052015 time (s)

Fig. 3. Degranulation in complexin II knockdown cells. (A) Wild- Journal of Cell Science type or knockdown cells were sensitized with IgE and stimulated with antigen (100 ng/ml DNP6-BSA). The quantity of β- hexosaminidase in the supernatant is expressed as a percentage of total β-hexosaminidase. Values were obtained from four independent preparations of wild-type cells () and four independent knockdown clones () [means.e.m. (n=4)]. (B) Cells were stimulated with A23187 (1 µM) and PMA (50 ng/ml). Values were obtained and plotted as mentioned above. Fig. 4. Translocation of complexin II after stimulation. Complexin II in RBL-2H3 was visualized with anti-complexin II antibody or anti- myc antibody, using FITC-conjugated anti-mouse IgG, as in Fig. 1C. (A) Distribution of complexin II before antigen stimulation. The left image shows the distribution of complexin II in wild-type cells using in Fig. 4A,B). This translocation to the plasma membrane was anti-complexin II antibody. The right image shows the distribution of clear at about 3 minutes after stimulation. The distribution of myc-tagged complexin II in cells transfected with myc-tagged fluorescence signal became more punctate. Similar results were complexin II using anti-myc antibody. (B) Distribution of complexin obtained in RBL-2H3 cells expressing myc-tagged complexin II II at 5 minutes after antigen stimulation. Complexin II was with anti-myc antibody (right images in Fig. 4A,B). Next, we translocated to the plasma membrane. Left and right images show the investigated this translocation in the absence of extracellular distribution of complexin II and myc-tagged complexin II, Ca2+. Translocation to the plasma membrane was observed even respectively. (C) Distribution of complexin II at 5 minutes after 2+ in the absence of extracellular Ca2+ (Fig. 4C). Under this antigen stimulation in the absence of extracellular Ca . Complexin II was translocated to the plasma membrane. (D) Timecourse of condition, exocytotic release was inhibited almost completely 2+ (Fig. 4D). Stimulation with thapsigargin in the absence of antigen-induced degranulation in the absence of extracellular Ca . Values were obtained from four independent preparations of wild- extracellular Ca2+ gave similar results (Fig. 4E,F). These results 2+ type cells [mean±s.e.m. (n=4)]. (E) Distribution of complexin II at 5 suggest that the transient elevation of the intracellular Ca minutes after thapsigargin (50 nmol/l) stimulation in the absence of concentration is sufficient for translocation to the plasma extracellular Ca2+. Complexin II was translocated to the plasma membrane, and neither sustained Ca2+ increase due to the influx membrane. (F) Timecourse of thapsigargin-induced degranulation in of extracellular Ca2+ nor the activation of IgE receptors was the absence of extracellular Ca2+. Values are obtained from four necessary. independent preparations of wild-type cells [mean±s.e.m. (n=4)]. 2244 Journal of Cell Science 118 (10) Ca-dependence of degranulation on extracellular concentrations when cells were stimulated in various calcium concentration extracellular Ca2+ concentrations. Using these estimated Experiments with complexin-knockout mice (Reim et al., intracellular Ca2+ concentrations, we re-plotted Ca2+ 2001) suggested that complexin is involved in the calcium- dependency of degranulation (Fig. 5B). Data were fitted with dependence of exocytotic release in neuronal cells. Therefore, the Hill equation: 2+ we compared the Ca -dependence of degranulation in wild- Release = Rmax [Ca2+]n/(Kdn + [Ca2+]n) , type and knockdown cells. Cells were stimulated with the 2+ calcium ionophore A23187 and PMA. The calcium ion where Rmax is maximum release, [Ca ] is intracellular 2+ concentration in extracellular solution was changed from concentration of Ca , Kd is apparent dissociation constant and nominally free to 2 mmol/l. In Fig. 5, the Ca2+-dependent n is the Hill coefficient. By nonlinear regression analysis, Rmax, component of degranulation activity, which was calculated by Kd and n were determined. Maximum release (Rmax) of wild- subtracting the value of degranulation at 0 mmol/l, is shown. type and knockdown cells predicted by the fit were 37.0 ± 1.4% In both wild-type and knockdown cells, degranulation and 33.6 ± 5.4%, respectively. Kd and n values for wild-type µ decreased as the extracellular Ca2+ concentration decreased, cells were 0.28 ± 0.01 M and 3.1 ± 0.50. For knockdown cells, µ but the dose-response curve of knockdown cells was shifted the Kd and n were 0.56 ± 0.13 M and 1.7 ± 0.27, respectively. 2+ toward a higher concentration of Ca2+ (Fig. 5A). To compare These values suggest that the Ca sensitivity of degranulation the Ca2+ sensitivity of Ca2+-dependent degranulation between was reduced in knockdown cells. wild-type and knockdown cells, degranulation activity shown in Fig. 5A was re-plotted against intracellular Ca2+ Discussion concentration (Fig. 5B). As shown in an inset of Figure 5B, intracellular Ca2+ concentration reached a plateau at about 400 We found that complexin II, but not complexin I, was expressed seconds after stimulation. The timecourse of intracellular Ca2+ in mast cells. Although these two isoforms are highly concentration change induced by PMA and A23187 did not homologous (84% amino acid sequence identity) and localized differ between wild-type and knockdown cells. Therefore, we in the brain, they are distributed differently in the brain (Yamada considered an average concentration from 400 seconds to 20 et al., 1999). The expression of complexin II in mast cells is minutes after stimulation as representative of intracellular consistent with the observation that complexin II is expressed in Ca2+ concentration, and estimated the intracellular Ca2+ peripheral tissues, while complexin I is restricted to the brain and spinal cord (Takahashi et al., 1995). Recent studies have suggested that complexin II is related to neurological diseases A such as Huntington’s disease and schizophrenia (Morton et al., 2001; Eastwood et al., 2001). Morton and Edwardson reported 40 that complexin II was progressively depleted in a mouse model of Huntington’s disease, while the expression of complexin I and 30 SNARE proteins remained unchanged (Morton and Edwardson, ) 2001). In addition, the expression of mutant huntingtin blocks % Journal of Cell Science (

e exocytosis in PC12 cells through the depletion of complexin II, s

a 20

e but not complexin I (Edwardson, 2003). Complexin II knockout l e

r mice survive to maturity, but abnormalities in hippocampal long- 10 term potentiation (Takahashi et al., 1999) and cognitive function (Glynn et al., 2003) have been observed. Therefore, complexin II plays an essential role in neurological function (Glynn et al., 0 2003). In mast cells, complexin II was distributed throughout the 0.050.1 0.2 0.3 0.5 1.0 2.0 cytoplasm and nucleus (Fig. 1C). Because complexin II is a [Ca]ex (mM) small (18–19 kD) and soluble protein, it is not unusual that it B 2+ 2+ 2.0 Fig. 5. Effects of the extracellular Ca concentration on Ca - 1.0 mM 1.0 ) dependent degranulation. Mast cells were stimulated with PMA and 1.8 0.5 mM M

0.6 µ 2+ (

A23187 at various extracellular Ca concentrations. (A) o

1.6 0.4 n i

0.25 mM i t 40 ] a Degranulation activity at 20 minutes after stimulation is plotted

r 0.1 mM 1.4 0.2 a

C 2+ 0.05 mM 0.1 [ against extracellular Ca concentration. Values are expressed as the 1.2 0 mM percentage of total β-hexosaminidase as shown in Fig. 3, but β-

) 30 1.0 2+ 0 100 200300400 500 600 12001400 hexosaminidase at [Ca ]ex=0 was subtracted. Each point was % ( time (s)

e obtained from four independent preparations of wild-type cells ( ) s

a and four independent knockdown clones () [mean±s.e.m. (n=4)]. e 20 l 2+ e (B) Degranulation activity is re-plotted against intracellular Ca r concentration, which was estimated from Ca2+ concentration at 10 plateau phase after stimulation using Fura-2 as shown in the inset. (Inset) Timecourses of intracellular Ca2+ concentration in wild-type cells induced by PMA and A23187 at various extracellular Ca2+ 0 concentrations, indicated on each line. Cells were stimulated at the 0.10.2 0.3 0.5 1.0 time indicated by the arrow. Ca2+ concentration was converted from [Ca]in (µM) ratio values and shown as a second ordinate. Complexin facilitates exocytosis in mast cell 2245 was found in the nucleus, and it might have some function there. postsynaptic currents (EPSC) of hippocampal neurons from No localization of complexin II in a distinct organelle was complexin knockout mice. They investigated the Ca2+ observed but the distribution was not uniform. The reason for dependence of synchronous evoked transmitter release and this non-uniform distribution is not known. As shown in Fig. 4B, estimated the Kd value for extracellular Ca2+ concentration using the punctate distribution became clearer after stimulation and the Hill equation. They found that the Kd of double knockout bright spots were observed on the plasma membrane. mice for complexins I and II increased twice as much as that of In experiments using complexin II knockdown cells, we found single knockout mice for complexin II. that complexin II positively regulated exocytotic release. These As we could not obtain data at higher concentrations of inhibitory effects were not due to inhibition of the expression of intracellular Ca2+, due to cellular damage caused by long syntaxin (Fig. 2A) or antigen-induced Ca2+ mobilization (Fig. exposure to high concentrations of Ca2+, the dose-response curve 2D), which suggests that complexin II plays a role in the process for knockdown clones does not seem to be saturated (Fig. 5B). of membrane fusion between secretory granules and the plasma Nonlinear regression analysis allows us to estimate values membrane. related to intracellular Ca2+ dependency, which is extremely To investigate the mechanism of the regulation induced by important for understanding the mechanism of calcium- complexin II, we observed its intracellular distribution after dependent secretion, although lack of measured values at antigen stimulation. Before stimulation, complexin II resided in saturated phase did not provide an ideal condition for curve- the cytoplasm; however, it was translocated to the plasma fitting using the Hill equation. membrane after stimulation (Fig. 4B). This translocation was Complexin itself does not have an apparent binding site with observed at 3 minutes after stimulation and degranulation also Ca2+ and its binding to the SNARE complex is not affected by became clear in the same time window, suggesting that this Ca2+ (Pabst et al., 2000). Therefore, it is unlikely that complexin translocation is closely related to degranulation. As complexin II is the Ca2+ sensor that regulates Ca2+-dependent exocytosis. selectively associates with the ternary SNARE complex, but not As complexin II is translocated to the plasma membrane after with monomeric SNARE proteins, this translocation might be stimulation, there are at least three steps that may be sensitive to due to the association of complexin II with SNARE complex on the Ca2+ concentration: the translocation of complexin II to the the plasma membrane that is formed by stimulation. The plasma membrane; the association of complexin II with the distribution of fluorescence signal became more punctate, and SNARE complex on the plasma membrane; and the induction clear spots appeared on the plasma membrane after stimulation. of exocytotic membrane fusion. Considering the finding that This spot-like structure might reflect the sites of exocytotic exocytotic release requires a sustained increase in the membrane fusion. Interestingly, the translocation to the plasma intracellular Ca2+ concentration due to Ca2+ influx, it is probable membrane occurred even in the absence of extracellular Ca2+ that the latter two steps have major contributions to the Ca2+- (Fig. 4C). Furthermore, since stimulation by thapsigargin in the sensitivity of degranulation induced by ionophore and PMA. absence of extracellular Ca2+ caused the translocation of Change in not only Kd but also the Hill coefficient suggests that complexin II (Fig. 4E), a transient increase in intracellular Ca2+ complexin II regulates the conformation of Ca2+-binding sites of is sufficient for translocation, and a signal through the IgE the Ca2+ sensor. So far, synaptotagmin is the most likely receptor is not required. When cells were stimulated without candidate for the Ca2+ sensor. As synaptotagmin has two Ca2+- Journal of Cell Science extracellular Ca2+, degranulation was inhibited almost binding sites, complexin II might regulate the conformation or completely (Fig. 4D,F). This suggests that the translocation to binding ability of these sites. In mast cells, synaptotagmins II, the plasma membrane of complexin II is not enough to induce III, V and IX are expressed (Baram et al., 2001), and Baram et membrane fusion between secretory granules and the plasma al. (Baram et al., 1999) reported that synaptotagmin II, which is membrane, and Ca2+ influx from the extracellular medium a major isoform of synaptotagmin, negatively regulates Ca2+- triggers membrane fusion. Transient increase in Ca2+ might be dependent exocytosis in RBL-2H3 cells. As the expression level necessary to form a ternary SNARE complex, to which of synaptotagmin II was not affected in knockdown cells, as complexin II binds. Once the SNARE complex is formed, shown in Fig. 2A, a study on the interaction of synaptotagmin complexin II can associate with the SNARE complex, and Ca2+ II with the SNARE complex in the presence of complexin II influx from the extracellular medium is not required. This notion should shed light on the mechanism of the reduction of Ca2+ is consistent with the model that the Ca2+-independent sensitivity in complexin II knockdown mast cells. association of complexin with the SNARE complex puts the Pabst et al. (Pabst et al., 2000) investigated the association of complex in a metastable state, which is essential for efficient, complexin with the SNARE complex using different isoforms of fast Ca2+-dependent neurotransmitter release (Chen et al., 2002; syntaxin. They showed that complexin binds SNARE complexes Marz and Hanson, 2002). These observations, that complexin II comprised of SNAP-25, VAMP 2 and syntaxin 3 as efficiently was translocated to the plasma membrane but translocation itself as it binds with complexes comprised of SNAP-25, VAMP 2 and did not trigger degranulation, were observed for the first syntaxin 1. However, complexin did not bind to the SNARE time and provide useful information for understanding the complex containing syntaxin 4. This finding, together with the mechanism of exocytosis regulation, not only in mast cells but result that complexin is involved in degranulation, suggests that also in neuronal cells. syntaxin 3 plays a major role as t-SNARE in mast cells. This is Interestingly, knockdown of complexin II affected the Ca2+ also supported by the observation that Munc18-2, which sensitivity of degranulation (Fig. 5). While maximum release associates with syntaxin 3 but not syntaxin 4, is involved in was not significantly changed, Kd and the Hill coefficient degranulation in mast cells (Martin-Verdeaux et al., 2003). suggested that the Ca2+ sensitivity was clearly reduced in Very recently, Abderrahmani et al. (Abderrahmani et al., knockdown cells. A similar reduction of Ca2+ sensitivity was 2004) showed that complexin I, but not complexin II, regulates observed by Reim et al. (Reim et al., 2001) in excitatory secretion in pancreatic β cells. They also knocked down the 2246 Journal of Cell Science 118 (10)

expression of complexin I and showed a reduction in secretion inhibits store-operated calcium currents and exocytotic membrane fusion in induced by nutrients. As the SNARE complex formed by RBL-2H3 cells. Biochemistry 42, 11808-11814. Kinet, J. P. (1999). The high-affinity IgE receptor (Fc epsilon RI): from syntaxin I, VAMP and SNAP25 is required for exocytotic release physiology to pathology. Annu. Rev. Immunol. 17, 931-972. in pancreatic β cells (Wheeler et al., 1996; Regazzi et al., 1996), Martin-Verdeaux, S., Pombo, I., Iannascoli, B., Roa, M., Varin-Blank, N., complexin I serves as a positive regulator of SNARE-like Rivera, J. and Blank, U. (2003). Evidence of a role for Munc18-2 and complexin II in mast cells, although different isoforms of microtubules in mast cell granule exocytosis. J. Cell Sci. 116, 325-334. Marz, K. E. and Hanson, P. I. (2002). Sealed with a twist: complexin and the SNARE and complexin act in these cells. The involvement of synaptic SNARE complex. Trends Neurosci. 25, 381-383. complexin in degranulation provides an insight into the McMahon, H. T., Missler, M., Li, C. and Südhof, T. C. (1995). Complexins: mechanism of allergy and the development of anti-allergy drugs. cytosolic proteins that regulate SNAP receptor function. Cell 83, 111-119. Morton, A. J. and Edwardson, J. M. (2001). Progressive depletion of complexin II in a transgenic mouse model of Huntington’s disease. J. Neurochem. 76, 166-172. References Morton, A. J., Faull, R. L. and Edwardson, J. M. (2001). Compartmental loss Abderrahmani, A., Niederhauser, G., Plaisance, V., Roehrich. M.-E., of NADPH diaphorase in the neuropil of the human striatum in Huntington’s Lenain, V., Coppola, T., Regazzi, R. and Waeber, G. (2004). Complexin I disease. Brain Res Bull. 56, 111-117. regulates glucose-induced secretion in pancreatic β-cells. J. Cell Sci. 117, Nagamatsu, S., Watanabe, T., Nakamichi, Y., Yamamura, C., Tsuzuki, K. 2239-2247. and Matsushima, S. (1999). alpha-soluble N-ethylmaleimide-sensitive factor Abraham, S. N. and Malaviya, R. (1997). Mast cells in infection and immunity. attachment protein is expressed in pancreatic beta cells and functions in insulin Infect. Immun. 65, 3501-3508. but not gamma-aminobutyric acid secretion. J. Biol. Chem. 274, 8053-8060. Amano, T., Furuno, T., Hirashima, N., Ohyama, N. and Nakanishi, M. Ono, S., Baux, G., Sekiguchi, M., Fossier, P., Morel, N. F., Nihonmatsu, I., (2001). Dynamics of intracellular granules with CD63-GFP in rat basophilic Hirata, K., Awaji, T., Takahashi, S. and Takahashi, M. (1998). Regulatory leukemia cells. J. Biochem. 129, 739-744. roles of complexins in neurotransmitter release from mature presynaptic nerve Baram, D., Adachi, R., Medalia, O., Tuvim, M., Dickey, B. F., Mekori, Y. A. terminals. Eur. J. Neurosci. 10, 2143-2152. and Sagi-Eisenberg, R. (1999). Synaptotagmin II negatively regulates Ca2+- Pabst, S., Hazzard, J. W., Antonin, W., Südhof, T. C., Jahn, R., Rizo, J. and triggered exocytosis of lysosomes in mast cells. J. Exp. Med. 189, 1649-1658. Fasshauer, D. (2000). Selective interaction of complexin with the neuronal Baram, D., Mekori, Y. A. and Sagi-Eisenberg, R. (2001). Synaptotagmin SNARE complex. Determination of the binding regions. Biol. Chem. 275, regulates mast cell functions. Immunol. Rev. 179, 25-34. 19808-19818. Blank, U. and Rivera, J. (2004). The ins and outs of IgE-dependent mast-cell Pabst, S., Margittai, M., Vainius, D., Langen, R., Jahn, R. and Fasshauer, exocytosis. Trends Immunol. 25, 266-267. D. (2002). Rapid and selective binding to the synaptic SNARE complex Blank, U., Cyprien, B., Martin-Verdeaux, S., Paumet, F., Pombo, I., Rivera, suggests a modulatory role of complexins in neuroexocytosis. J. Biol. Chem. J., Roa, M. and Varin-Blank, N. (2001). SNAREs and associated regulators 277, 7838-7848. in the control of exocytosis in the RBL-2H3 mast cell line. Mol. Immunol. 38, Paumet, F., le Mao, J., Martin, S., Galli, T., David, B., Blank, U. and Roa, 1341-1345. M. (2000). Soluble NSF attachment protein receptors (SNAREs) in RBL-2H3 Bracher, A., Kadlec, J., Betz, H. and Weissenhorn, W. (2002). X-ray structure mast cells: functional role of syntaxin 4 in exocytosis and identification of a of a neuronal complexin-SNARE complex from squid. J. Biol. Chem. 277, vesicle-associated membrane protein 8-containing secretory compartment. J. 26517-26523. Immunol. 164, 5850-5857. Brunger, A. T. (2001). Structure of proteins involved in synaptic vesicle fusion Reed, G. L., Houng, A. and Fitzgerald, M. L. (1999). Human platelets contain in neurons. Curr. Opin. Struct. Biol. 11, 163-173. SNARE proteins and a Sec1p homologue that interacts with syntaxin 4 and Calakos, N. and Scheller, R. H. (1996). Synaptic vesicle biogenesis, docking, is phosphorylated after thrombin activation: implications for platelet secretion. and fusion: a molecular description. Physiol. Rev. 76, 1-29. Blood 93, 2617-2626. Chen, X., Tomchick, D. R., Kovrigin, E., Arac, D., Machius, M., Südhof, T. Regazzi, R., Sadoul, K., Meda, P., Kelly, R. B., Halban, P. A. and Wollheim, 2+ Journal of Cell Science C. and Rizo, J. (2002). Three-dimensional structure of the C. B. (1996). Mutational analysis of VAMP domains implicated in Ca - complexin/SNARE complex. Neuron 33, 397-409. induced insulin exocytosis. EMBO J. 15, 6951-6959. Eastwood, S. L., Cotter, D. and Harrison, P. J. (2001). Cerebellar synaptic Reim, K., Mansour, M., Varoqueaux, F., McMahon, H. T., Südhof, T. C., protein expression in schizophrenia. Neuroscience 105, 219-229. Brose, N. and Rosenmund, C. (2001). Complexins regulate a late step in Edwardson, J. M., Wang, C. T., Gong, B., Wyttenbach, A., Bai, J., Jackson, Ca2+-dependent neurotransmitter release. Cell 104, 71-81. M. B., Chapman, E. R. and Morton, A. J. (2003). Expression of mutant Sollner, T., Bennett, M. K., Whiteheart, S. W., Scheller, R. H. and Rothman, huntingtin blocks exocytosis in PC12 cells by depletion of complexin II. J. J. E. (1993). A protein assembly-disassembly pathway in vitro that may Biol. Chem. 278, 30849-30853. correspond to sequential steps of synaptic vesicle docking, activation, and Flaumenhaft, R., Croce, K., Chen, E., Furie, B. and Furie, B. C. (1999). fusion. Cell 75, 409-418. Inhibition of calpain blocks platelet secretion, aggregation, and spreading. J. Swann, P. G., Odom, S. and Rivera, J. (1998). Protein kinase C and early mast Biol. Chem. 274, 2492-2501. cell signals. In Signal Transduction in Mast Cells and Basophils (ed. E. Razin Glynn, D., Bortnick, R. A. and Morton, A. J. (2003). Complexin II is essential and J. Rivera), pp. 152-170. New York: Springer-Verlag. for normal neurological function in mice. Hum. Mol. Genet. 12, 2431-2434. Takahashi, S., Yamamoto, H., Matsuda, Z., Ogawa, M., Yagyu, K., Grynkiewicz, G., Poenie, M. and Tsien, R. Y. (1985). A new generation of Taniguchi, T., Miyata, T., Kaba, H., Higuchi, T., Okutani, F. et al. (1995). Ca2+ indicators with greatly improved fluorescence properties. J. Biol. Chem. Identification of two highly homologous presynaptic proteins distinctly 260, 3440-3448. localized at the dendritic and somatic synapses. FEBS Lett. 368, 455-460. Guo, Z., Turner, C. and Castle, D. (1998). Relocation of the t-SNARE SNAP- Takahashi, S., Ujihara, H., Huang, G. Z., Yagyu, K. I., Sanbo, M., Kaba, H. 23 from lamellipodia-like cell surface projections regulates compound and Yagi, T. (1999). Reduced hippocampal LTP in mice lacking a presynaptic exocytosis in mast cells. Cell 94, 537-548. protein: complexin II. Eur. J. Neurosci. 11, 2359-2366. Hibi, T., Hirashima, N. and Nakanishi, M. (2000). Rat basophilic leukemia Tokumaru, H., Umayahara, K., Pellegrini, L. L., Ishizuka, T., Saisu, H., cells express syntaxin-3 and VAMP-7 in granule membranes. Biochem. Betz, H., Augustine, G. J. and Abe, T. (2001). SNARE complex Biophys. Res. Commun. 271, 36-41. oligomerization by synaphin/complexin is essential for synaptic vesicle Ishizuka, T., Saisu, H., Odani, S. and Abe, T. (1995). Synaphin: a protein exocytosis. Cell 104, 421-432. associated with the docking/fusion complex in presynaptic terminals. Turner, H. and Kinet, J. P. (1999). Signalling through the high-affinity IgE Biochem. Biophys. Res. Commun. 213, 1107-1114. receptor Fc epsilonRI. Nature 402, B24-B30. Itakura, M., Misawa, H., Sekiguchi, M., Takahashi, S. and Takahashi, M. Wheeler, M. B., Sheu, L., Ghai, M., Bouquillon, A., Grondin, G., Weller, U., (1999). Transfection analysis of functional roles of complexin I and II in the Beaudoin, A. R., Bennett, M. K., Trimble, W. S. and Gaisano, H. Y. (1996). exocytosis of two different types of secretory vesicles. Biochem. Biophys. Res. Characterization of SNARE protein expression in beta cell lines and Commun. 265, 691-696. pancreatic islets. Endocrinology 137, 1340-1348. Kato, N., Nakanishi, M. and Hirashima, N. (2002). Transbilayer asymmetry Yamada, M., Saisu, H., Ishizuka, T., Takahashi, H. and Abe, T. (1999). of phospholipids in the plasma membrane regulates exocytotic release in mast Immunohistochemical distribution of the two isoforms of synaphin/complexin cells. Biochemistry 41, 8068-8074. involved in neurotransmitter release: localization at the distinct central nervous Kato, N., Nakanishi, M. and Hirashima, N. (2003). Cholesterol depletion system regions and synaptic types. Neuroscience 93, 7-18.