- Email: [email protected]
Tetanus Toxin-Sensitive VAMP-Related Proteins Are Present in Murine Macrophages
CELLULAR IMMUNOLOGY ARTICLE NO.
169, 113–116 (1996)
0098
Tetanus Toxin-Sensitive VAMP-Related Proteins Are Present in Murine Macrophages LUCIA PITZURRA,* ORNELLA ROSSETTO,† ANNA RITA CHIMIENTI,* ELISABETTA BLASI,‡ AND FRANCESCO BISTONI* *Microbiology Section, Department of Experimental Medicine and Biochemical Sciences, University of Perugia, Via del Giochetto, 06122 Perugia, Italy; †CNR Biomembranes Center and Department of Biomedical Sciences, Universita` di Padova, 35121 Padua, Italy; and ‡Department of Biomedical Sciences, Hygiene and Microbiology Section, University of Modena, 41100 Modena, Italy Received September 15, 1995; accepted December 20, 1995
The light chain of tetanus neurotoxin (TeTx) is a zinc endopeptidase specific for VAMP/synaptobrevin (VAMP), a 120-amino-acid integral protein previously described in the small synaptic vesicles of neuronal cells. TeTx has been shown to be active also on nonneuronal cells. By SDS – PAGE and quantitative immunoblotting on proteins derived from murine macrophages (Mf) exposed to TeTx, we have shown that: (1) VAMP-related proteins are also present in Mf and (2) such proteins are sensitive to TeTx proteolytic cleavage. The demonstration that TeTx acts on VAMP-related proteins also in Mf offers a new and useful tool for molecular studies on Mf exocytosis. q 1996 Academic Press, Inc.
toxin. In particular, binding sites for TeTx have been shown on pancreatic islet cells (16), on thyroid plasma membranes (17), on rabbit kidney membranes (18), and on murine macrophages (Mf) (19). Furthermore, functional studies have proved that in vitro intoxication of murine Mf or human monocytes with TeTx occurs and provokes impairment of lysozyme (LZM) secretion (20 – 22). Particularly, using the murine Mf cell line GG2EE (GG2EE Mf), generated in vitro by v-raf/v-myc oncogenes (23), we have demonstrated that TeTx selectively inhibits IFN-ginduced, but not basal, LZM activity (21, 22). In this report we show that (i) VAMP-related proteins are present in GG2EE cells as well as in thioglycollate-elicited peritoneal Mf (thio Mf) and that (ii) such proteins are sensitive to TeTx proteolytic cleavage.
INTRODUCTION Tetanus neurotoxin (TeTx), produced by toxigenic strains of Clostridium tetani, causes the spastic paralysis of tetanus by blocking neurotransmitter release of the spinal cord inhibitory interneurons (1). TeTx is composed of two disulfide-linked chains. The heavy chain (H; 100 kDa) is responsible for neurospecific binding and cell penetration of the light chain (L; 50 kDa) (2, 3). The L chain is a zinc (Zn) endopeptidase specific for VAMP/synaptobrevin (VAMP), a 120-amino-acid integral protein present in the small synaptic vesicles (SSV) of neuronal cells (4 – 8). Initial evidence indicates that proteolytic cleavage of such VAMP accounts for the TeTx-induced blockade of neuroexocytosis (4, 8 – 10). Recent reports document the presence of VAMP-related proteins also in nonneuronal cells (11 – 15) where distinct isoforms have been identified (12, 14, 15). In addition to the extensive literature describing the discrete steps through which neuronal intoxication by TeTx occurs and its consequences, increasing evidence is provided on the susceptibility on nonneuronal cells to such bacterial
MATERIALS AND METHODS Toxin. TeTx was isolated from culture filtrates of C. tetani, as detailed elsewhere (7). TeTx was purified by immobilized-metal affinity chromatography (24) and stored at 0807C in 10 mM sodium Hepes and 50 mM NaCl, pH 7.2. Dithiothreitol (DTT) – TeTx was obtained by incubation of TeTx with 10 mM DTT for 30 min at 377C. Cells. Thio Mf were obtained from mice injected intraperitoneally 4 days earlier with 1 ml of 10% thioglycollate broth (Bacto Brewer; Difco Labs, Detroit, MI). Exudate cells were washed three times with Hanks’ balanced salt solution (HBSS; Gibco Laboratories, Grand Island, NY) and Mf were purified by plastic adherence. Briefly, Mf were plated at 2 1 106/ ml in 96-mm dishes (Nunc Inter Med, Roskilde, Denmark) and after 2 hr of adherence, the nonadherent cells were removed by washing with warm HBSS. The GG2EE cell line, generated as described elsewhere (23), was maintained in RPMI 1640 medium supple-
113
/ 6c0c$$8082
03-11-96 07:11:20
cia
AP: Cell Immuno
0008-8749/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
114
PITZURRA ET AL.
mented with 10% heat-inactivated fetal calf serum, gentamicin (50 mg/ml), and L-glutamine (4 mM) (complete medium). All reagents were purchased from Gibco Laboratories. VAMP and VAMP antisera. VAMP was isolated from SSV of the rat brain cortex using a 13–18% polyacrylamide gradient sodium dodecyl sulfate–gel electrophoresis (SDS–PAGE) (25) and electroeluted as described (26). Such material was used as a source of purified SSV VAMP for immunoblotting studies or injected in vivo to produce antisera as follows. Twenty micrograms of SSV VAMP was lyophilized in 1 ml of phosphate-buffered saline (PBS) and mixed with 1 ml of complete Freund’s adjuvant and 3 mg of heat-killed Mycobacterium tuberculosis (Difco). The resulting emulsion was injected intradermally in the back of New Zealand White rabbits (27). After 4 and 12 weeks, animals were boosted by intramuscular injection of the same amount of electroeluted rat brain VAMP diluted at 1 ml with PBS and mixed with 1 ml of incomplete Freund’s adjuvant (Difco). An IgG-enriched fraction (referred to as anti-VAMP IgGs throughout) was obtained by ammonium sulfate precipitation of pooled sera (20–42% of saturation) and then dissolved in the same volume of PBS. Western blotting analysis. Proteins derived from 2 1 106 cells of either thio Mf or GG2EE Mf, exposed or not to TeTx as detailed under Results, were precipitated with 6.5% trichloroacetic acid in 1 ml of PBS containing a protease inhibitor cocktail (1 mM iodoacetamide, 0.4 mM benzamidine, 0.3 mM PMSF, 1 mg/ ml leupeptin, 2 mg/ml pepstatin, 100 mM TLCK, and 100 mM TPCK). Pellets were recovered after centrifugation (45 sec at 15,000g), dissolved in 50 ml of 6% SDS, 5 mM EDTA, 5% b-mercaptoethanol, 10 mM Tris acetate, pH 8.2, and immediately boiled for 3 min. Sample volumes corresponding to 40,000 cells were separated onto 13–18% SDS–PAGE. Proteins were then transferred onto nitrocellulose paper for 3 hr at 400 mA (28), exposed to anti-VAMP IgGs (1:200 dilution, 16 hr at 47C) and revealed with goat anti-rabbit antibodies conjugated to alkaline peroxidase (1:10,000, 2 hr at 207C; Boehringer Mannheim, Germany). RESULTS In order to evaluate the presence of VAMP-related proteins in murine Mf, we compared the electrophoretic patterns of proteins derived from thio Mf, GG2EE Mf, and SSV, under the same experimental conditions (Fig. 1). The proteins were separated by SDS–PAGE under reducing conditions on 13–18% acrylamide slab gradient gels. The Coomassie blue staining revealed the presence of a complex pattern of bands in Mf extracts (lanes 1 and 2) and several discrete bands in SSV VAMP (lane 3). The comparison among lanes re-
/ 6c0c$$8082
03-11-96 07:11:20
cia
FIG. 1. VAMP-related proteins are present in thio Mf and GG2EE Mf. Lanes 1–3 show Coomassie blue-stained SDS–PAGE of proteins derived from thio Mf (lane 1), GG2EE Mf (lane 2), and SSV (lane 3). Lanes 4–6 contain samples of SSV (lane 4), thio Mf (lane 5), and GG2EE Mf (lane 6) immunoblotted with the anti-VAMP IgGs. Migration of Mr standards is shown in the center. Arrowheads indicate the size (19 kDa) of SSV VAMP.
vealed the presence of a band of approximately 19 kDa in all three experimental groups. When parallel gels were blotted and immunostained, we observed a single protein band in all three experimental groups (lanes 4 to 6). Such a band appeared highly reactive to the antiVAMP IgGs in both Mf populations (lanes 5 and 6) and was comparable for electrophoretic mobility (19 kDa) to the VAMP protein observed in SSV extracts (lane 4). TeTx is composed of two disulfide-linked polypeptide chains. The H chain is involved in nerve binding and internalization of the L chain, which in turn is responsible for the intracellular blockade of neurotransmitter release (1). The L chain is a Zn endopeptidase specific for SSV VAMP proteins (4–8), whose activation requires reduction of the interchain disulfide bond (9). To assess the susceptibility of Mf VAMP-related proteins to TeTx, quantitative immunoblotting was performed by comparing the amounts of VAMP-related proteins in control Mf and Mf exposed to appropriately reduced TeTx (DTT–TeTx). Thus, thio Mf and GG2EE Mf (2 1 106) were incubated in media or in the presence of DTT–TeTx (1.5 mg/ml) for 2 hr. In parallel groups, pretreatment with 10 mg/ml streptolysin-O (SLO; Sclavo S.p.A, Siena, Italy) for 1 hr was also included to allow L chain free access within the cytosol of the cells, as previously described (13, 29). Then, cells were harvested and processed for immunoblotting using anti-VAMP IgGs. Quantitative scanning of the immunoblots was then performed with a dual-wavelength Shimadzu CS-630 densitometer (9). Figure 2 shows the relative intensity of the VAMP-related bands in Mf exposed or not to TeTx. We found that VAMP-related
AP: Cell Immuno
MACROPHAGE VAMP/SYNAPTOBREVIN
115
FIG. 2. TeTx recognizes and cleaves Mf-associated VAMP-related proteins. Thio Mf and GG2EE Mf, untreated (h) or exposed to DTT–TeTx (j), in the absence (control) or in the presence of SL-O, were harvested and processed for quantitative immunoblotting by antiVAMP IgGs. Data are expressed as percentage of VAMP-related proteins of DTT–TeTx-treated versus untreated Mf, taken as 100%.
proteins of GG2EE Mf were sensitive to TeTx since the band intensity was reduced 15 or 35% in intact or SL-O-permeabilized cells, respectively, after exposure to DTT–TeTx. Similarly, VAMP-related proteins of thio Mf were susceptible to the proteolytic cleavage of TeTx, the effect being remarkable upon permeabilization with SL-O. DISCUSSION VAMP proteins are involved in the docking of neurotransmitter vesicles by binding, via a set of soluble proteins, to syntaxin, an integral protein of the presynaptic membrane (30, 31). Recent evidence indicates that (a) VAMP proteins are the molecular target for TeTx proteolytic activity in neuronal cells (9), (b) proteins of the VAMP family are ubiquitously expressed (13, 31), and (c) isoforms of VAMP (VAMP I, VAMP II, and cellulobrevin), identified outside the nervous system (12–14), are involved in the constitutive exocytosis of recycling vesicles in nonneuronal cells (14). Here we show that VAMP-related proteins are present in murine Mf and are sensitive to TeTx proteolytic cleavage. Using the in vitro-established cell line GG2EE, we have previously provided evidence that TeTx affects murine Mf by impairing their secretory (21, 22) and antitumoral functions (32). By Western blot analysis, we demonstrate here the presence of VAMP-related proteins in GG2EE Mf as well as in peritoneally elicited Mf. The appearance of a band, highly stained by anti-VAMP IgGs and showing electrophoretic mobility superimposable to that of SSV VAMP, allowed us to conclude that Mf share with neuronal cells the expression of the same and/or highly homologous VAMP proteins. These results further expand the knowledge about the wide-spectrum distribution of such proteins
/ 6c0c$$8082
03-11-96 07:11:20
cia
and raise the possibility of their involvement in Mf exocytosis. TeTx cannot usually enter nonneuronal cells because, unlike neuronal cells, they do not have a plasmalemma receptor for the holotoxin. Thus, the L chain cannot be internalized and translocated into the cytoplasm (13). To allow internalization of biologically active TeTx into Mf, we used the DTT–TeTx in combination with a SL-O cell perforation procedure, according to previously established protocols (13). By quantitative immunoblotting analysis, we show here that Mf exposure to DTT–TeTx results in a consistent reduction of VAMP band intensity. The phenomenon is evident and comparable in both Mf populations provided that permeabilization with SL-O has been performed. These results indicate that the anatomical origin of Mf does not alter the biomolecular peculiarities of such VAMP. Moreover, the finding that the L chain of TeTx exerts appreciable proteolytic activity in intact GG2EE Mf, but not in intact thio Mf, suggests that only the former Mf possess a ‘‘receptor-like’’ system capable of efficiently internalizing the toxic molecule. Experiments are in progress to clarify this point. Studies on rat models demonstrated that brain cells contain two VAMP isoforms (I and II) that comigrate in SDS–PAGE (7, 9, 25). Minor differences between the two have been identified in the amino-terminal region and, more interestingly, in one residue at the TeTx cleavage site (7, 9). This implies that only one of the two isoforms (VAMP II) is actually susceptible to the proteolytic activity of TeTx L chain. As confirmed by quantitative analysis (5, 7, 9), the cleavage of the 19kDa SSV VAMP band is incomplete, never exceeding 60–70% reduction. Similarly, the TeTx proteolytic cleavage of VAMP-related proteins is incomplete also in murine Mf. Thus, the similarity of results between
AP: Cell Immuno
116
PITZURRA ET AL.
rat brain (4–8) and the murine Mf model (present paper) strongly suggests that Mf also possess the two VAMP isoforms, one of which is insensitive to TeTx. Furthermore, the recent identification of a novel target for TeTx, namely cellulobrevin, another VAMP isoform described in nonneuronal cells with exocytosis activity (13, 15, 33), raised the possibility that not only VAMP I and II but also cellulobrevin may be present in Mf. Whether and to what extent each of those proteins may control Mf exocytosis and in turn be responsible for TeTx inhibitory effects remains an open question. Overall, our findings emphasize a close relationship between LZM exocytosis and levels of VAMP-related proteins in murine Mf, by demonstrating a parallel decrease of both upon intoxication with TeTx. As an in vitro system, highly reproducible and easy to perform, the experimental model of GG2EE Mf intoxication by TeTx may be extensively employed.
10. 11. 12. 13. 14. 15.
16. 17. 18. 19. 20. 21.
ACKNOWLEDGMENTS We are indebted to Professor C. Montecucco and Dr. G. P. Schiavo for discussing and reviewing the manuscript and to Eileen Mahoney for editorial assistance. This work was supported by MURST (40%) Progetto Nazionale Controllo della Patogenicita` Microbica, Italy.
22. 23. 24.
REFERENCES
25. 1. Simpson, L. L. ‘‘Botulinum Neurotoxin and Tetanus Toxin,’’ Academic Press, San Diego, 1989. 2. Ahnert-Higler, G., Weller, U., Dauzenroth, M. E., Habermann, E., and Gratzel, M., FEBS Lett. 242, 245, 1989. 3. Krieglstain, K., Henschen, A., Weller, U., and Habermann, E., Eur. J. Biochem. 188, 39, 1990.
26. 27. 28.
4. Montecucco, C., and Schiavo, G., Trends Biochem. Sci. 18, 324, 1993.
29.
5. Patarnello, T., Bargelloni, L., Rossetto, O., Schiavo, G., and Montecucco, C., Nature 364, 581, 1993. [Letter]
30.
6. Schiavo, G., Rossetto, O., Benefati, F., Poulain, B., and Montecucco, C., Ann. N. Y. Acad. Sci. 71, 65, 1994.
31.
7. Schiavo, G., Poulain, B., Rossetto, O., Benefati, F., Tauc, L., and Montecucco, C., EMBO J. 10, 3577, 1992.
32.
8. Schiavo, G., Rossetto, O., and Montecucco, C., Scienze 304, 40, 1993.
33.
9. Schiavo, G., Benefati, F., Poulain, B., Rossetto, O., Polverino
/ 6c0c$$8082
03-11-96 07:11:20
cia
de Laureto, P., Das Gupta, B. R., and Montecucco, C., Nature 359, 832, 1992. Poulain, B., Rossetto, O., Deloye, F., Schiavo, G., Tauc, L., and Montecucco, C., J. Neurochem. 61, 1175, 1993. Baumert, M., Maycox, P. R., Navone, F., De Camilli, P., and Jahn, R., EMBO J. 8, 379, 1989. Cain, C. C., Trimble, W. S., and Lienhard, G. E., J. Biol. Chem. 267, 1168, 1992. Galli, T., Chilcote, T., Mundigl, O., Binz, T., Niemann, H., and De Camilli, P., J. Cell Biol. 125, 1015, 1994. Ralston, E., Beushausen, S., and Ploug, T., J. Biol. Chem. 269, 15403, 1994. McMahon, H. T., Uskaryov, Y. A., Edelmann, L., Link, E., Binz, T., Niemann, H., Jahn, R., and Su¨dhof, T. C., Nature 364, 346, 1993. Eisenbarth, G. S., Shimizu, K., Bowring, M. A., and Wells, A., Proc. Natl. Acad. Sci. USA 79, 5066, 1982. Ledley, F. D., Lee, G., and Kohn, L. D., J. Biol. Chem. 252, 4049, 1977. Habermann, E., and Allbus, V., J. Neurochem. 46, 1219, 1986. Blasi, E., Pitzurra, L., Burhan Fuad, A. M., Marconi, P., and Bistoni, F., Scand. J. Immunol. 32, 289, 1990. Ho, J. L., and Klempner, M. S., J. Infect. Dis. 157, 925, 1985. Pitzurra, L., Marconi, P., Bistoni, F., and Blasi, E., Infect. Immun. 57, 2452, 1989. Pitzurra, L., Blasi, E., Puliti, M., and Bistoni, F., Infect. Immun. 61, 3605, 1993. Blasi, E., Mathieson, B. J., Varesio, L., Cleveland, S. L., Borchert, P. A., and Rapp, U., Nature 318, 667, 1985. Rossetto, O., Schiavo, G., Polverino de Laureto, P., Fabbiani, S., and Montecucco, C., Biochem. J. 285, 9, 1992. Huttner, W. B., Schiebler, W., Greengard, P., and De Camilli, P., J. Cell Biol. 96, 1374, 1983. Hunkapiller, M. W., Lujan, E., Ostrander, F., and Hood, L. E., Methods Enzymol. 91, 227, 1983. Vaitujaitis, J. L., Methods Enzymol. 73, 46, 1981. Towbin, H., Staehelin, T., and Gordon, J., Proc. Natl. Acad. Sci. USA 76, 4350, 1979. Ahnert-Higler, G., Bader, M. F., Bhakdi, S., and Gratzl, M., J. Neurochem. 52, 1751, 1989. Bennet, K., Calakos, N., and Scheller, R. H., Science 257, 255, 1992. Sollner, T., Whiteheart, S. W., Brunner, M., Erdjument-Bromage, H., Geromanos, S., Tempst, P., and Rothman, J. E., Nature 362, 31, 1993. Pitzurra, L., Puliti, M., Fuad, M. A. B., Bistoni, F., and Blasi, E., FEMS Immunol. Med. Microbiol. 7, 289, 1993. Link, E., McMahon, H. T., Fisher von Mollard, G., Yamasaki, S., Nieman, H., Su¨dhof, T. C., and Jahn, R., J. Biol. Chem. 268, 18423, 1993.
AP: Cell Immuno
169, 113–116 (1996)
0098
Tetanus Toxin-Sensitive VAMP-Related Proteins Are Present in Murine Macrophages LUCIA PITZURRA,* ORNELLA ROSSETTO,† ANNA RITA CHIMIENTI,* ELISABETTA BLASI,‡ AND FRANCESCO BISTONI* *Microbiology Section, Department of Experimental Medicine and Biochemical Sciences, University of Perugia, Via del Giochetto, 06122 Perugia, Italy; †CNR Biomembranes Center and Department of Biomedical Sciences, Universita` di Padova, 35121 Padua, Italy; and ‡Department of Biomedical Sciences, Hygiene and Microbiology Section, University of Modena, 41100 Modena, Italy Received September 15, 1995; accepted December 20, 1995
The light chain of tetanus neurotoxin (TeTx) is a zinc endopeptidase specific for VAMP/synaptobrevin (VAMP), a 120-amino-acid integral protein previously described in the small synaptic vesicles of neuronal cells. TeTx has been shown to be active also on nonneuronal cells. By SDS – PAGE and quantitative immunoblotting on proteins derived from murine macrophages (Mf) exposed to TeTx, we have shown that: (1) VAMP-related proteins are also present in Mf and (2) such proteins are sensitive to TeTx proteolytic cleavage. The demonstration that TeTx acts on VAMP-related proteins also in Mf offers a new and useful tool for molecular studies on Mf exocytosis. q 1996 Academic Press, Inc.
toxin. In particular, binding sites for TeTx have been shown on pancreatic islet cells (16), on thyroid plasma membranes (17), on rabbit kidney membranes (18), and on murine macrophages (Mf) (19). Furthermore, functional studies have proved that in vitro intoxication of murine Mf or human monocytes with TeTx occurs and provokes impairment of lysozyme (LZM) secretion (20 – 22). Particularly, using the murine Mf cell line GG2EE (GG2EE Mf), generated in vitro by v-raf/v-myc oncogenes (23), we have demonstrated that TeTx selectively inhibits IFN-ginduced, but not basal, LZM activity (21, 22). In this report we show that (i) VAMP-related proteins are present in GG2EE cells as well as in thioglycollate-elicited peritoneal Mf (thio Mf) and that (ii) such proteins are sensitive to TeTx proteolytic cleavage.
INTRODUCTION Tetanus neurotoxin (TeTx), produced by toxigenic strains of Clostridium tetani, causes the spastic paralysis of tetanus by blocking neurotransmitter release of the spinal cord inhibitory interneurons (1). TeTx is composed of two disulfide-linked chains. The heavy chain (H; 100 kDa) is responsible for neurospecific binding and cell penetration of the light chain (L; 50 kDa) (2, 3). The L chain is a zinc (Zn) endopeptidase specific for VAMP/synaptobrevin (VAMP), a 120-amino-acid integral protein present in the small synaptic vesicles (SSV) of neuronal cells (4 – 8). Initial evidence indicates that proteolytic cleavage of such VAMP accounts for the TeTx-induced blockade of neuroexocytosis (4, 8 – 10). Recent reports document the presence of VAMP-related proteins also in nonneuronal cells (11 – 15) where distinct isoforms have been identified (12, 14, 15). In addition to the extensive literature describing the discrete steps through which neuronal intoxication by TeTx occurs and its consequences, increasing evidence is provided on the susceptibility on nonneuronal cells to such bacterial
MATERIALS AND METHODS Toxin. TeTx was isolated from culture filtrates of C. tetani, as detailed elsewhere (7). TeTx was purified by immobilized-metal affinity chromatography (24) and stored at 0807C in 10 mM sodium Hepes and 50 mM NaCl, pH 7.2. Dithiothreitol (DTT) – TeTx was obtained by incubation of TeTx with 10 mM DTT for 30 min at 377C. Cells. Thio Mf were obtained from mice injected intraperitoneally 4 days earlier with 1 ml of 10% thioglycollate broth (Bacto Brewer; Difco Labs, Detroit, MI). Exudate cells were washed three times with Hanks’ balanced salt solution (HBSS; Gibco Laboratories, Grand Island, NY) and Mf were purified by plastic adherence. Briefly, Mf were plated at 2 1 106/ ml in 96-mm dishes (Nunc Inter Med, Roskilde, Denmark) and after 2 hr of adherence, the nonadherent cells were removed by washing with warm HBSS. The GG2EE cell line, generated as described elsewhere (23), was maintained in RPMI 1640 medium supple-
113
/ 6c0c$$8082
03-11-96 07:11:20
cia
AP: Cell Immuno
0008-8749/96 $18.00 Copyright q 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
114
PITZURRA ET AL.
mented with 10% heat-inactivated fetal calf serum, gentamicin (50 mg/ml), and L-glutamine (4 mM) (complete medium). All reagents were purchased from Gibco Laboratories. VAMP and VAMP antisera. VAMP was isolated from SSV of the rat brain cortex using a 13–18% polyacrylamide gradient sodium dodecyl sulfate–gel electrophoresis (SDS–PAGE) (25) and electroeluted as described (26). Such material was used as a source of purified SSV VAMP for immunoblotting studies or injected in vivo to produce antisera as follows. Twenty micrograms of SSV VAMP was lyophilized in 1 ml of phosphate-buffered saline (PBS) and mixed with 1 ml of complete Freund’s adjuvant and 3 mg of heat-killed Mycobacterium tuberculosis (Difco). The resulting emulsion was injected intradermally in the back of New Zealand White rabbits (27). After 4 and 12 weeks, animals were boosted by intramuscular injection of the same amount of electroeluted rat brain VAMP diluted at 1 ml with PBS and mixed with 1 ml of incomplete Freund’s adjuvant (Difco). An IgG-enriched fraction (referred to as anti-VAMP IgGs throughout) was obtained by ammonium sulfate precipitation of pooled sera (20–42% of saturation) and then dissolved in the same volume of PBS. Western blotting analysis. Proteins derived from 2 1 106 cells of either thio Mf or GG2EE Mf, exposed or not to TeTx as detailed under Results, were precipitated with 6.5% trichloroacetic acid in 1 ml of PBS containing a protease inhibitor cocktail (1 mM iodoacetamide, 0.4 mM benzamidine, 0.3 mM PMSF, 1 mg/ ml leupeptin, 2 mg/ml pepstatin, 100 mM TLCK, and 100 mM TPCK). Pellets were recovered after centrifugation (45 sec at 15,000g), dissolved in 50 ml of 6% SDS, 5 mM EDTA, 5% b-mercaptoethanol, 10 mM Tris acetate, pH 8.2, and immediately boiled for 3 min. Sample volumes corresponding to 40,000 cells were separated onto 13–18% SDS–PAGE. Proteins were then transferred onto nitrocellulose paper for 3 hr at 400 mA (28), exposed to anti-VAMP IgGs (1:200 dilution, 16 hr at 47C) and revealed with goat anti-rabbit antibodies conjugated to alkaline peroxidase (1:10,000, 2 hr at 207C; Boehringer Mannheim, Germany). RESULTS In order to evaluate the presence of VAMP-related proteins in murine Mf, we compared the electrophoretic patterns of proteins derived from thio Mf, GG2EE Mf, and SSV, under the same experimental conditions (Fig. 1). The proteins were separated by SDS–PAGE under reducing conditions on 13–18% acrylamide slab gradient gels. The Coomassie blue staining revealed the presence of a complex pattern of bands in Mf extracts (lanes 1 and 2) and several discrete bands in SSV VAMP (lane 3). The comparison among lanes re-
/ 6c0c$$8082
03-11-96 07:11:20
cia
FIG. 1. VAMP-related proteins are present in thio Mf and GG2EE Mf. Lanes 1–3 show Coomassie blue-stained SDS–PAGE of proteins derived from thio Mf (lane 1), GG2EE Mf (lane 2), and SSV (lane 3). Lanes 4–6 contain samples of SSV (lane 4), thio Mf (lane 5), and GG2EE Mf (lane 6) immunoblotted with the anti-VAMP IgGs. Migration of Mr standards is shown in the center. Arrowheads indicate the size (19 kDa) of SSV VAMP.
vealed the presence of a band of approximately 19 kDa in all three experimental groups. When parallel gels were blotted and immunostained, we observed a single protein band in all three experimental groups (lanes 4 to 6). Such a band appeared highly reactive to the antiVAMP IgGs in both Mf populations (lanes 5 and 6) and was comparable for electrophoretic mobility (19 kDa) to the VAMP protein observed in SSV extracts (lane 4). TeTx is composed of two disulfide-linked polypeptide chains. The H chain is involved in nerve binding and internalization of the L chain, which in turn is responsible for the intracellular blockade of neurotransmitter release (1). The L chain is a Zn endopeptidase specific for SSV VAMP proteins (4–8), whose activation requires reduction of the interchain disulfide bond (9). To assess the susceptibility of Mf VAMP-related proteins to TeTx, quantitative immunoblotting was performed by comparing the amounts of VAMP-related proteins in control Mf and Mf exposed to appropriately reduced TeTx (DTT–TeTx). Thus, thio Mf and GG2EE Mf (2 1 106) were incubated in media or in the presence of DTT–TeTx (1.5 mg/ml) for 2 hr. In parallel groups, pretreatment with 10 mg/ml streptolysin-O (SLO; Sclavo S.p.A, Siena, Italy) for 1 hr was also included to allow L chain free access within the cytosol of the cells, as previously described (13, 29). Then, cells were harvested and processed for immunoblotting using anti-VAMP IgGs. Quantitative scanning of the immunoblots was then performed with a dual-wavelength Shimadzu CS-630 densitometer (9). Figure 2 shows the relative intensity of the VAMP-related bands in Mf exposed or not to TeTx. We found that VAMP-related
AP: Cell Immuno
MACROPHAGE VAMP/SYNAPTOBREVIN
115
FIG. 2. TeTx recognizes and cleaves Mf-associated VAMP-related proteins. Thio Mf and GG2EE Mf, untreated (h) or exposed to DTT–TeTx (j), in the absence (control) or in the presence of SL-O, were harvested and processed for quantitative immunoblotting by antiVAMP IgGs. Data are expressed as percentage of VAMP-related proteins of DTT–TeTx-treated versus untreated Mf, taken as 100%.
proteins of GG2EE Mf were sensitive to TeTx since the band intensity was reduced 15 or 35% in intact or SL-O-permeabilized cells, respectively, after exposure to DTT–TeTx. Similarly, VAMP-related proteins of thio Mf were susceptible to the proteolytic cleavage of TeTx, the effect being remarkable upon permeabilization with SL-O. DISCUSSION VAMP proteins are involved in the docking of neurotransmitter vesicles by binding, via a set of soluble proteins, to syntaxin, an integral protein of the presynaptic membrane (30, 31). Recent evidence indicates that (a) VAMP proteins are the molecular target for TeTx proteolytic activity in neuronal cells (9), (b) proteins of the VAMP family are ubiquitously expressed (13, 31), and (c) isoforms of VAMP (VAMP I, VAMP II, and cellulobrevin), identified outside the nervous system (12–14), are involved in the constitutive exocytosis of recycling vesicles in nonneuronal cells (14). Here we show that VAMP-related proteins are present in murine Mf and are sensitive to TeTx proteolytic cleavage. Using the in vitro-established cell line GG2EE, we have previously provided evidence that TeTx affects murine Mf by impairing their secretory (21, 22) and antitumoral functions (32). By Western blot analysis, we demonstrate here the presence of VAMP-related proteins in GG2EE Mf as well as in peritoneally elicited Mf. The appearance of a band, highly stained by anti-VAMP IgGs and showing electrophoretic mobility superimposable to that of SSV VAMP, allowed us to conclude that Mf share with neuronal cells the expression of the same and/or highly homologous VAMP proteins. These results further expand the knowledge about the wide-spectrum distribution of such proteins
/ 6c0c$$8082
03-11-96 07:11:20
cia
and raise the possibility of their involvement in Mf exocytosis. TeTx cannot usually enter nonneuronal cells because, unlike neuronal cells, they do not have a plasmalemma receptor for the holotoxin. Thus, the L chain cannot be internalized and translocated into the cytoplasm (13). To allow internalization of biologically active TeTx into Mf, we used the DTT–TeTx in combination with a SL-O cell perforation procedure, according to previously established protocols (13). By quantitative immunoblotting analysis, we show here that Mf exposure to DTT–TeTx results in a consistent reduction of VAMP band intensity. The phenomenon is evident and comparable in both Mf populations provided that permeabilization with SL-O has been performed. These results indicate that the anatomical origin of Mf does not alter the biomolecular peculiarities of such VAMP. Moreover, the finding that the L chain of TeTx exerts appreciable proteolytic activity in intact GG2EE Mf, but not in intact thio Mf, suggests that only the former Mf possess a ‘‘receptor-like’’ system capable of efficiently internalizing the toxic molecule. Experiments are in progress to clarify this point. Studies on rat models demonstrated that brain cells contain two VAMP isoforms (I and II) that comigrate in SDS–PAGE (7, 9, 25). Minor differences between the two have been identified in the amino-terminal region and, more interestingly, in one residue at the TeTx cleavage site (7, 9). This implies that only one of the two isoforms (VAMP II) is actually susceptible to the proteolytic activity of TeTx L chain. As confirmed by quantitative analysis (5, 7, 9), the cleavage of the 19kDa SSV VAMP band is incomplete, never exceeding 60–70% reduction. Similarly, the TeTx proteolytic cleavage of VAMP-related proteins is incomplete also in murine Mf. Thus, the similarity of results between
AP: Cell Immuno
116
PITZURRA ET AL.
rat brain (4–8) and the murine Mf model (present paper) strongly suggests that Mf also possess the two VAMP isoforms, one of which is insensitive to TeTx. Furthermore, the recent identification of a novel target for TeTx, namely cellulobrevin, another VAMP isoform described in nonneuronal cells with exocytosis activity (13, 15, 33), raised the possibility that not only VAMP I and II but also cellulobrevin may be present in Mf. Whether and to what extent each of those proteins may control Mf exocytosis and in turn be responsible for TeTx inhibitory effects remains an open question. Overall, our findings emphasize a close relationship between LZM exocytosis and levels of VAMP-related proteins in murine Mf, by demonstrating a parallel decrease of both upon intoxication with TeTx. As an in vitro system, highly reproducible and easy to perform, the experimental model of GG2EE Mf intoxication by TeTx may be extensively employed.
10. 11. 12. 13. 14. 15.
16. 17. 18. 19. 20. 21.
ACKNOWLEDGMENTS We are indebted to Professor C. Montecucco and Dr. G. P. Schiavo for discussing and reviewing the manuscript and to Eileen Mahoney for editorial assistance. This work was supported by MURST (40%) Progetto Nazionale Controllo della Patogenicita` Microbica, Italy.
22. 23. 24.
REFERENCES
25. 1. Simpson, L. L. ‘‘Botulinum Neurotoxin and Tetanus Toxin,’’ Academic Press, San Diego, 1989. 2. Ahnert-Higler, G., Weller, U., Dauzenroth, M. E., Habermann, E., and Gratzel, M., FEBS Lett. 242, 245, 1989. 3. Krieglstain, K., Henschen, A., Weller, U., and Habermann, E., Eur. J. Biochem. 188, 39, 1990.
26. 27. 28.
4. Montecucco, C., and Schiavo, G., Trends Biochem. Sci. 18, 324, 1993.
29.
5. Patarnello, T., Bargelloni, L., Rossetto, O., Schiavo, G., and Montecucco, C., Nature 364, 581, 1993. [Letter]
30.
6. Schiavo, G., Rossetto, O., Benefati, F., Poulain, B., and Montecucco, C., Ann. N. Y. Acad. Sci. 71, 65, 1994.
31.
7. Schiavo, G., Poulain, B., Rossetto, O., Benefati, F., Tauc, L., and Montecucco, C., EMBO J. 10, 3577, 1992.
32.
8. Schiavo, G., Rossetto, O., and Montecucco, C., Scienze 304, 40, 1993.
33.
9. Schiavo, G., Benefati, F., Poulain, B., Rossetto, O., Polverino
/ 6c0c$$8082
03-11-96 07:11:20
cia
de Laureto, P., Das Gupta, B. R., and Montecucco, C., Nature 359, 832, 1992. Poulain, B., Rossetto, O., Deloye, F., Schiavo, G., Tauc, L., and Montecucco, C., J. Neurochem. 61, 1175, 1993. Baumert, M., Maycox, P. R., Navone, F., De Camilli, P., and Jahn, R., EMBO J. 8, 379, 1989. Cain, C. C., Trimble, W. S., and Lienhard, G. E., J. Biol. Chem. 267, 1168, 1992. Galli, T., Chilcote, T., Mundigl, O., Binz, T., Niemann, H., and De Camilli, P., J. Cell Biol. 125, 1015, 1994. Ralston, E., Beushausen, S., and Ploug, T., J. Biol. Chem. 269, 15403, 1994. McMahon, H. T., Uskaryov, Y. A., Edelmann, L., Link, E., Binz, T., Niemann, H., Jahn, R., and Su¨dhof, T. C., Nature 364, 346, 1993. Eisenbarth, G. S., Shimizu, K., Bowring, M. A., and Wells, A., Proc. Natl. Acad. Sci. USA 79, 5066, 1982. Ledley, F. D., Lee, G., and Kohn, L. D., J. Biol. Chem. 252, 4049, 1977. Habermann, E., and Allbus, V., J. Neurochem. 46, 1219, 1986. Blasi, E., Pitzurra, L., Burhan Fuad, A. M., Marconi, P., and Bistoni, F., Scand. J. Immunol. 32, 289, 1990. Ho, J. L., and Klempner, M. S., J. Infect. Dis. 157, 925, 1985. Pitzurra, L., Marconi, P., Bistoni, F., and Blasi, E., Infect. Immun. 57, 2452, 1989. Pitzurra, L., Blasi, E., Puliti, M., and Bistoni, F., Infect. Immun. 61, 3605, 1993. Blasi, E., Mathieson, B. J., Varesio, L., Cleveland, S. L., Borchert, P. A., and Rapp, U., Nature 318, 667, 1985. Rossetto, O., Schiavo, G., Polverino de Laureto, P., Fabbiani, S., and Montecucco, C., Biochem. J. 285, 9, 1992. Huttner, W. B., Schiebler, W., Greengard, P., and De Camilli, P., J. Cell Biol. 96, 1374, 1983. Hunkapiller, M. W., Lujan, E., Ostrander, F., and Hood, L. E., Methods Enzymol. 91, 227, 1983. Vaitujaitis, J. L., Methods Enzymol. 73, 46, 1981. Towbin, H., Staehelin, T., and Gordon, J., Proc. Natl. Acad. Sci. USA 76, 4350, 1979. Ahnert-Higler, G., Bader, M. F., Bhakdi, S., and Gratzl, M., J. Neurochem. 52, 1751, 1989. Bennet, K., Calakos, N., and Scheller, R. H., Science 257, 255, 1992. Sollner, T., Whiteheart, S. W., Brunner, M., Erdjument-Bromage, H., Geromanos, S., Tempst, P., and Rothman, J. E., Nature 362, 31, 1993. Pitzurra, L., Puliti, M., Fuad, M. A. B., Bistoni, F., and Blasi, E., FEMS Immunol. Med. Microbiol. 7, 289, 1993. Link, E., McMahon, H. T., Fisher von Mollard, G., Yamasaki, S., Nieman, H., Su¨dhof, T. C., and Jahn, R., J. Biol. Chem. 268, 18423, 1993.
AP: Cell Immuno