- Email: [email protected]
The GM2 Activator Protein, a Novel Inhibitor of Platelet-Activating Factor
Biochemical and Biophysical Research Communications 258, 256 –259 (1999) Article ID bbrc.1999.0621, available online at http://www.idealibrary.com on
The GM2 Activator Protein, a Novel Inhibitor of Platelet-Activating Factor Brigitte Rigat,* Denis Reynaud,† Natasha Smiljanic-Georgijev,* ,† and Don Mahuran* ,† ,1 *Research Institute, Hospital for Sick Children, Toronto, Ontario, Canada; and †Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada
Received March 30, 1999
The GM2 activator protein is required as a substrate-specific cofactor for b-hexosaminidase A to hydrolyze GM2 ganglioside. The GM2 activator protein reversibly binds and solubilizes individual GM2 ganglioside molecules, making them available as substrate. Although GM2 ganglioside is the strongest binding ligand for the activator protein, it can also bind and transport between membranes a series of other glycolipids, even at neutral pH. Biosynthetic studies have shown that a large portion of newly synthesized GM2 activator molecules are not targeted to the lysosome, but are secreted and can then be recaptured by other cells through a carbohydrate independent mechanism. Thus, the GM2 activator protein may have other in vivo functions. We found that the GM2 activator protein can inhibit, through specific binding, the ability of platelet activating factor (PAF) to stimulate the release of intracellular Ca 21 pools by human neutrophils. PAF is a biologically potent phosphoacylglycerol. Inhibitors for PAF’s role in the pathogenesis of inflammatory bowel disease and asthma have been sought as potential therapeutic agents. The inherent stability and protease resistance of the small, monomeric GM2 activator protein, coupled with the ability to produce large quantities of the functional protein in transformed bacteria, suggest it may serve as such an agent. © 1999 Academic Press
The GM2 activator protein (Activator) is a small (;20kDa), heat stable, protease-resistant monomeric, lysosomal protein containing a single site for Asnlinked glycosylation [reviewed in (1, 2)] and four intrachain disulfide bonds (3, 4). Its only proven in vivo function is to act as a substrate specific cofactor for lysosomal b-hexosaminidase A (Hex A) in its hydrolysis of GM2 ganglioside (GM2) (5). It first solubilizes 1 To whom correspondence should be addressed at: Research Institute, rm. 9146A, The Hospital For Sick Children, 555 University Avenue, Toronto, Ontario, Canada, M5G 1X8. Fax: 416-813-8700. E-mail: [email protected].
0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
individual molecules of ganglioside or glycolipids by interacting with both their hydrophilic oligosaccharide and their hydrophobic ceramide moieties (Fig. 1). Its specificity, i.e. strength, of binding is primarily determined by the oligosaccharide moiety of the ligand, e.g. GM2 @ GM1 . GD1a 5 GM3 5 GA2 (6, 7). The Activator can then replace its bound glycolipid in another membrane, thus acting as a general sphingolipid transport protein (6, 8). When GM2 is the ligand the 1:1 Activator:ganglioside complex can specifically interact with Hex A resulting in the hydrolysis of GM2 to GM3 [reviewed in (2, 9)]. Several groups have demonstrated that the Activator can be synthesized in bacteria and refolded to a functional form (8, 10, 11). We have shown that the biochemical properties of an Activator protein produced by a His 6-tag, bacterial expression system, followed by a refolded procedure, were similar to those measured for Activator isolated from the media of transfected CHO cells (3, 8, 12). Using the former Activator we (8) demonstrated that the activator’s lipid and oligosaccharide binding functions can be assessed independently through a fluorescence dequenching assay (13, 14). It was shown that various glycolipids inhibit the ability of the Activator to transport a selfquenching fluorescent lipid probe, octadecylrhodamine (R-18), between liposomes. Thus the rate of R-18 transport measures lipid binding, and the level by which various glycolipid inhibit this transport is a measure of oligosaccharide-binding. Using this method we determined that the relative binding strength of the Activator for the following glycolipids was, GM2 @ GT1b @ GM1 ' Gb4 ' GM3 . GA2 (8). The intra- and inter-cellular transport mechanisms for the Activator protein appear to be complex. We have shown that the mannose-6-phosphate pathway (M-6-P), common to most soluble lysosomal proteins, is likely the activator’s major intracellular, biosynthetic route to the lysosome in human fibroblasts; as well as, it serves as a high affinity re-capture pathway for the endocytosis of those Activator molecules containing
256
Vol. 258, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
M-6-P from extracellular fluids. Additionally we found that there exists a second lower affinity re-capture pathway that requires the activator’s native protein structure, is carbohydrate independent, and likely does not involve its ability to bind glycosphingolipids in the plasma membrane. Finally we found that there is a major pool of newly synthesized precursor Activator proteins which do not contain a M-6-P targeting signal and make-up the secreted forms of the protein in normal human fibroblasts (12). Similar results were reported from the study of the biosynthesis of Activator in human epidermis (15). Given that there is a major pool of Activator proteins that are secreted, other non-lysosomal functions may exist for the Activator in vivo, e.g. (16, 17). Since gangliosides on the plasma membrane of some cells have been associated with intracellular calcium release (18 – 20) we examined the effect on intracellular calcium concentrations of extracellular Activator added to a suspension of human neutrophils. We found that while Activator on its own produced no effect, the presence of the Activator partially blocked the release of intracellular calcium stores by the exogenously added PAF (platelet activating factor). PAF is a biologically potent phosphoacylglycerol which exhibits a wide variety of physiological and pathophysiological effects in various cells and tissues. We then determined that the Activator inhibits PAF by directly binding the compound, with a strength of binding greater that of GM1 ganglioside, but less than GM2. MATERIALS AND METHODS Cell preparation and measurement of intracellular calcium. Neutrophils were isolated from heparinized venous blood collected from healthy volunteers as previously reported (21). Neutrophils (1 mL, 10 7 cells) were loaded with a 3 mM solution of a fluorescent dye, INDO-1 AM, used to calculate [Ca 21] i, as previously reported (21, 22). For each measurement 2 3 10 6 cells were used in 1 mL of cell medium in a temperature-controlled (37°C) plastic cuvette with continuous stirring. Fluorescence was measured in a Perkin-Elmer fluorescence spectrophotometer as previously reported (22). The wild type Activator or a non-functional truncated form (see below), 0, 320, or 800 pmol, was included in the various samples to be assayed. After the assay mix had equilibrated, 2 pmol of PAF was added and the increase in [Ca 21] i calculated. The experiment was repeated using a second preparation of human neutrophils with the identical results. Activator purification and the determination of ligand binding. The GM2 activator protein was purified from an E. coli expression system that allowed synthesis, purification, and refolding of a His 6mature (S 32-I 193) GM2 activator fusion protein as previously reported (8). A non-functional (both as a co-factor for Hex A and in R-18 binding), truncated form of the Activator also produced using the His 6-system (8), was used as a negative control in the above calcium release experiments. The ability of 5 mg (0.26 nmol) of the Activator protein to bind 1.3 nmol of GM2 ganglioside, PAF, N-formyl-Met-Leu-Phe (fMLP), or leukotriene B 4 (LTB 4) was assessed at pH 5 using a fluorescence dequenching assay. The latter two compounds were included as negative controls as they, like PAF also induce the release of intracellular stores of Ca 21. This assay first evaluates the ability of the
FIG. 1. Fluorescence measurement of the [Ca 21] i of human neutrophils before (base line) and after (peak) the addition of PAF. Each panel indicates an identical experiment done in the presence of increasing amounts of functional Activator.
Activator to transport a self-quenching fluorescence dye, rhodamine conjugated to a saturated C-18 hydrocarbon chain (R-18), between labeled and unlabeled liposomes (phosphatidylcholine containing large unilamellar vesicles, PC-LUV s). Then the protein’s ability to bind various ligands is assessed by using each test compound as a specific inhibitor of the transport process (8).
RESULTS The addition of extracellular Activator (functional or non-functional) to a suspension of INDO-1 AM-loaded cells did not result in a change in [Ca 21] i (data not shown). When 2 pmol of PAF was added in the absence of the functional Activator the [Ca 21] i was increased by 401 nM. However, if the mixtures contained the functional Activator, a significant decrease in the release of intracellular calcium pools into the cytosol in response to PAF was noted (Fig. 1). The PAF-response was inhibited by ;28% or ;75% when 320 or 800 pmol of Activator was present (Fig. 1), respectively, but not at all in the presence of 800 pmol of the non-functional (see methods section) truncated Activator (data not shown).
257
Vol. 258, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Fluorescence dequenching assay using 5 mg (0.26 nmol) of functional Activator and 1.3 nmol of the indicated ligands (GM2 AP, is the Activator protein alone) as potential inhibitors. The degree of inhibition is directly related to the affinity of the Activator for the ligand.
The mechanism for the above inhibition of the PAF response in human neutrophils was probed. Since the Activator is known to bind some negatively charged gangliosides, it was possible that the Activator was directly binding PAF. To determine if this was the case we utilized our fluorescence dequenching assay. Using 5 mg of Activator and 1.3 nmol of GM2, PAF, fMLP, or LTB 4 , it was found that only GM2 and PAF inhibited the activator’s transfer of R-18 between liposomes (Fig. 2). PAF inhibited R-18 transport by 34% as compared to 77% for the same amount of GM2. Thus PAF inhibits the activator’s transport function less than does an identical amount of GM2, but more than GM1 (25% (8)). DISCUSSION During the past two decades, studies describing the chemistry and biology of PAF have been extensive. This potent phosphoacylglycerol exhibits a wide variety of physiological and pathophysiological effects in various cells and tissues. PAF acts, through specific receptors and a variety of signal transduction systems, to elicit diverse biochemical responses (23). PAF’s role in inflammatory response is particularly important. The movement of polymorphonuclear leukocytes (PMNs) from the mainstream of blood to the extravascular space is a characteristic feature of the inflammatory response. This process requires that the PMN initially contacts the endothelium, then adheres firmly to the vessel wall, and finally migrates out of the microvasculature. Each of these events requires signals or pro-inflammatory molecules that direct the PMN to the potential site of inflammation. These molecules include histamine, which appears to be of importance in the initial recruitment of PMNs, LTB 4, which promotes PMN adhesion, and PAF, which may contribute to both the adhesion process as well as the migration through the endothelial barrier. Thus, PAF
and cytokines, such as interleukins, tumor necrosis factor, and others, are thought to play a role in the inflammatory process involving gastrointestinal disorders such as Crohn’s disease, ulcerative colitis, ischemic colitis, or antibiotic-associated colitis (24). PAF is also one of the chemical mediators that may participate in the inflammatory process underlying asthma. It has been shown to produce a prolonged increase in airway responsiveness in several species, including humans. In humans the increase in bronchial reactivity peaks 3 days after inhalation and can last as long as several weeks. PAF may induce this response by recruiting eosinophils into the airways and activating them in the airway walls, with the subsequent release of basic proteins. This process causes eosinophils, which are rich sources of PAF, to release their contents and sets up a self-perpetuating cycle (25). Because of PAF’s important role in the pathogenesis of inflammatory bowel disease and asthma, inhibition of PAF by specific antagonists, mediators, or other agents may have a potential therapeutic benefit in treatment and management of these inflammatory diseases. During an experiment aimed at determining if exogenously added Activator protein could cause neutrophils to release calcium, which used PAF as a positive control, we discovered that; while the Activator did not directly affect this process, it inhibited PAF’s ability to do so (Fig. 1). The effects of other compounds that elicit Ca 21 release by these cells, i.e. fMLP and LTB4, were not inhibited by the presence of Activator in the assay (data not shown). Further experiments using a fluorescence dequenching assay (8), demonstrated that the Activator protein binds PAF, but not fMLP or LTB4 (Fig. 2). Thus, the mechanism by which the Activator inhibits the PAF-response is through direct binding of the compound, presumably preventing PAF’s interaction with its receptors. Since cells can endocytose the Activator by a mechanism that is not dependent on either its carbohydrate moiety or whether or not it has formed a glyco-lipid complex (12), the Activator could serve to remove PAF from extracellular fluids, bypassing PAF receptors. Other than being lipids coupled with hydrophilic head-groups that each contain a negatively charged component (phosphate or NeuAc), the structures of PAF and GM2 ganglioside are not very similar. It was thus surprising that the Activator bound GM2 . PAF . GM1. However, the Activator has recently also been shown to activate ADP ribosylation factor-dependent phospholipase D (17). This enzyme catalyzes the hydrolysis of phosphatidylcholine to generate lipid mediators that control a wide range of physiological and pathological processes. This study did not identify the Activator’s mechanism of action; e.g., Activator binding to phosphatidylcholine was not assessed. Given our results with PAF, the ability of the Activator to bind phosphatidylcholine would not be surprising.
258
Vol. 258, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Most PAF inhibitors are analogs which act by blocking the PAF receptors (23, 26). Because of receptorheterogeneity, conventional agonists may only inhibit or have a differential effect on a subset of PAFreceptors. Furthermore there are issues of the overall toxicity of the agonist and its ability to gain access to the various cellular PAF-receptors. A naturally occurring protein normally secreted by different human cell types (12, 15), that directly binds secreted PAF, and is heat stable and protease resistant, such as the Activator, would have none of these drawbacks. Thus, theoretically if the Activator protein could be delivered in a concentrated form directly to the site of inflammation, it could be useful in the treatment and management of these diseases. In addition the future determination of the three dimensional structure of the Activator at the atomic level, through X-ray crystallography studies, may suggest changes that could be made in its primary structure to improve its specificity towards PAF and increase its pH optimum for ligand-binding. ACKNOWLEDGMENTS This work was supported by a grant from the Medical Research Council of Canada to D.M. We also thank Dr. C. R. Pace-Asciak for his advice concerning the Ca 21 release assay.
REFERENCES 1. Fu¨rst, W., and Sandhoff, K. (1992) Biochim. Biophys. Acta 1126, 1–16. 2. Mahuran, D. J. (1998) Biochem. Biophys. Acta 1393, 1–18. 3. Xie, B., Rigat, B., Smiljanic-Georgijev, N., Deng, H., and Mahuran, D. J. (1998) Biochemistry 37, 814 – 821. 4. Schu¨tte, C. G., Lemm, T., Glombitza, G. J., and Sandhoff, K. (1998) Protein Sci. 7, 1039 –1045. 5. Meier, E. M., Schwarzmann, G., Fu¨rst, W., and Sandhoff, K. (1991) J. Biol. Chem. 266, 1879 –1887. 6. Conzelmann, E., Burg, J., Stephan, G., and Sandhoff, K. (1982) Eur. J. Biochem. 123, 455– 464.
7. Conzelmann, E., and Sandhoff, K. (1979) Hoppe-Seyler’s Z. Physiol. Chem. 360, 1837–1849. 8. Smiljanic-Georgijev, N., Rigat, B., Xie, B., Wang, W., and Mahuran, D. J. (1997) Biochim. Biophys. Acta 1339, 192–202. 9. Sandhoff, K., Harzer, K., and Fu¨rst, W. (1995) in The Metabolic Basis of Inherited Disease (Scriver, C. R., Beaudet, A. L., Sly, W. S., and Valle, D., Eds.), Vol. 2, 7 Ed., pp. 2427–2441, McGrawHill, New York. 10. Wu, Y. Y., Lockyer, J. M., Sugiyama, E., Pavlova, N. V., Li, Y.-T., and Li, S.-C. (1994) J. Biol. Chem. 269, 16276 –16283. 11. Klima, H., Klein, A., Van Echten, G., Schwarzmann, G., Suzuki, K., and Sandhoff, K. (1993) Biochem. J. 292, 571–576. 12. Rigat, B., Wang, W., Leung, A., and Mahuran, D. J. (1997) Biochemistry 36, 8325– 8331. 13. Kuwana, T., Mullock, B. M., and Luzio, J. P. (1993) Biochem. Soc. Trans. 21, 299 –300. 14. Kuwana, T., Mullock, B. M., and Luzio, J. P. (1995) Biochem. J. 308, 937–946. 15. Glombitza, G. J., Becker, E., Kaiser, H. W., and Sandhoff, K. (1997) J. Biol. Chem. 272, 5199 –5207. 16. Mundel, T. M., Heid, H. W., Mahuran, D. J., Kriz, W., and Mundel, P. (1999) J. Am. Soc. Nephrol. 10, 435– 43. 17. Nakamura, S., Akisue, T., Jinnai, H., Hitomi, T., Sarkar, S., Miwa, N., Okada, T., Yoshida, K., Kuroda, S., Kikkawa, U., and Nishizuka, Y. (1998) Proc. Natl. Acad. Sci. USA 95, 12249 – 12253. 18. Benfenati, F., Fuxe, K., and Agnati, L. F. (1991) Neurochem. Int. 19, 271–279. 19. Hilbush, B. S., and Levine, J. M. (1991) Proc. Natl. Acad. Sci. USA 88, 5616 –5620. 20. Goldenring, J. R., Otis, L. C., Yu, R. K., and DeLorenzo, R. J. (1985) J. Neurochem. 44, 1229 –1234. 21. Laneuville, O., Renaud, D., Grinstein, S., Nigam, S., and PaceAsciak, C. R. (1993) Biochem. J. 295, 393–397. 22. Reynaud, D., and Pace-Asciak, C. R. (1997) Prostaglandins Leukot. Essent. Fatty Acids 56, 9 –12. 23. Chao, W., and Olson, M. S. (1993) Biochem. J. 292, 617–29. 24. Kubes, P. (1993) Canadian Journal of Physiology & Pharmacology 71, 88 –97. 25. Barnes, P. J. (1993) J. Allergy Clin. Immuno. 92, 187–189. 26. Izumi, T., and Shimizu, T. (1995) Biochim. Biophys. Acta 1259, 317–33.
259
The GM2 Activator Protein, a Novel Inhibitor of Platelet-Activating Factor Brigitte Rigat,* Denis Reynaud,† Natasha Smiljanic-Georgijev,* ,† and Don Mahuran* ,† ,1 *Research Institute, Hospital for Sick Children, Toronto, Ontario, Canada; and †Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada
Received March 30, 1999
The GM2 activator protein is required as a substrate-specific cofactor for b-hexosaminidase A to hydrolyze GM2 ganglioside. The GM2 activator protein reversibly binds and solubilizes individual GM2 ganglioside molecules, making them available as substrate. Although GM2 ganglioside is the strongest binding ligand for the activator protein, it can also bind and transport between membranes a series of other glycolipids, even at neutral pH. Biosynthetic studies have shown that a large portion of newly synthesized GM2 activator molecules are not targeted to the lysosome, but are secreted and can then be recaptured by other cells through a carbohydrate independent mechanism. Thus, the GM2 activator protein may have other in vivo functions. We found that the GM2 activator protein can inhibit, through specific binding, the ability of platelet activating factor (PAF) to stimulate the release of intracellular Ca 21 pools by human neutrophils. PAF is a biologically potent phosphoacylglycerol. Inhibitors for PAF’s role in the pathogenesis of inflammatory bowel disease and asthma have been sought as potential therapeutic agents. The inherent stability and protease resistance of the small, monomeric GM2 activator protein, coupled with the ability to produce large quantities of the functional protein in transformed bacteria, suggest it may serve as such an agent. © 1999 Academic Press
The GM2 activator protein (Activator) is a small (;20kDa), heat stable, protease-resistant monomeric, lysosomal protein containing a single site for Asnlinked glycosylation [reviewed in (1, 2)] and four intrachain disulfide bonds (3, 4). Its only proven in vivo function is to act as a substrate specific cofactor for lysosomal b-hexosaminidase A (Hex A) in its hydrolysis of GM2 ganglioside (GM2) (5). It first solubilizes 1 To whom correspondence should be addressed at: Research Institute, rm. 9146A, The Hospital For Sick Children, 555 University Avenue, Toronto, Ontario, Canada, M5G 1X8. Fax: 416-813-8700. E-mail: [email protected].
0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
individual molecules of ganglioside or glycolipids by interacting with both their hydrophilic oligosaccharide and their hydrophobic ceramide moieties (Fig. 1). Its specificity, i.e. strength, of binding is primarily determined by the oligosaccharide moiety of the ligand, e.g. GM2 @ GM1 . GD1a 5 GM3 5 GA2 (6, 7). The Activator can then replace its bound glycolipid in another membrane, thus acting as a general sphingolipid transport protein (6, 8). When GM2 is the ligand the 1:1 Activator:ganglioside complex can specifically interact with Hex A resulting in the hydrolysis of GM2 to GM3 [reviewed in (2, 9)]. Several groups have demonstrated that the Activator can be synthesized in bacteria and refolded to a functional form (8, 10, 11). We have shown that the biochemical properties of an Activator protein produced by a His 6-tag, bacterial expression system, followed by a refolded procedure, were similar to those measured for Activator isolated from the media of transfected CHO cells (3, 8, 12). Using the former Activator we (8) demonstrated that the activator’s lipid and oligosaccharide binding functions can be assessed independently through a fluorescence dequenching assay (13, 14). It was shown that various glycolipids inhibit the ability of the Activator to transport a selfquenching fluorescent lipid probe, octadecylrhodamine (R-18), between liposomes. Thus the rate of R-18 transport measures lipid binding, and the level by which various glycolipid inhibit this transport is a measure of oligosaccharide-binding. Using this method we determined that the relative binding strength of the Activator for the following glycolipids was, GM2 @ GT1b @ GM1 ' Gb4 ' GM3 . GA2 (8). The intra- and inter-cellular transport mechanisms for the Activator protein appear to be complex. We have shown that the mannose-6-phosphate pathway (M-6-P), common to most soluble lysosomal proteins, is likely the activator’s major intracellular, biosynthetic route to the lysosome in human fibroblasts; as well as, it serves as a high affinity re-capture pathway for the endocytosis of those Activator molecules containing
256
Vol. 258, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
M-6-P from extracellular fluids. Additionally we found that there exists a second lower affinity re-capture pathway that requires the activator’s native protein structure, is carbohydrate independent, and likely does not involve its ability to bind glycosphingolipids in the plasma membrane. Finally we found that there is a major pool of newly synthesized precursor Activator proteins which do not contain a M-6-P targeting signal and make-up the secreted forms of the protein in normal human fibroblasts (12). Similar results were reported from the study of the biosynthesis of Activator in human epidermis (15). Given that there is a major pool of Activator proteins that are secreted, other non-lysosomal functions may exist for the Activator in vivo, e.g. (16, 17). Since gangliosides on the plasma membrane of some cells have been associated with intracellular calcium release (18 – 20) we examined the effect on intracellular calcium concentrations of extracellular Activator added to a suspension of human neutrophils. We found that while Activator on its own produced no effect, the presence of the Activator partially blocked the release of intracellular calcium stores by the exogenously added PAF (platelet activating factor). PAF is a biologically potent phosphoacylglycerol which exhibits a wide variety of physiological and pathophysiological effects in various cells and tissues. We then determined that the Activator inhibits PAF by directly binding the compound, with a strength of binding greater that of GM1 ganglioside, but less than GM2. MATERIALS AND METHODS Cell preparation and measurement of intracellular calcium. Neutrophils were isolated from heparinized venous blood collected from healthy volunteers as previously reported (21). Neutrophils (1 mL, 10 7 cells) were loaded with a 3 mM solution of a fluorescent dye, INDO-1 AM, used to calculate [Ca 21] i, as previously reported (21, 22). For each measurement 2 3 10 6 cells were used in 1 mL of cell medium in a temperature-controlled (37°C) plastic cuvette with continuous stirring. Fluorescence was measured in a Perkin-Elmer fluorescence spectrophotometer as previously reported (22). The wild type Activator or a non-functional truncated form (see below), 0, 320, or 800 pmol, was included in the various samples to be assayed. After the assay mix had equilibrated, 2 pmol of PAF was added and the increase in [Ca 21] i calculated. The experiment was repeated using a second preparation of human neutrophils with the identical results. Activator purification and the determination of ligand binding. The GM2 activator protein was purified from an E. coli expression system that allowed synthesis, purification, and refolding of a His 6mature (S 32-I 193) GM2 activator fusion protein as previously reported (8). A non-functional (both as a co-factor for Hex A and in R-18 binding), truncated form of the Activator also produced using the His 6-system (8), was used as a negative control in the above calcium release experiments. The ability of 5 mg (0.26 nmol) of the Activator protein to bind 1.3 nmol of GM2 ganglioside, PAF, N-formyl-Met-Leu-Phe (fMLP), or leukotriene B 4 (LTB 4) was assessed at pH 5 using a fluorescence dequenching assay. The latter two compounds were included as negative controls as they, like PAF also induce the release of intracellular stores of Ca 21. This assay first evaluates the ability of the
FIG. 1. Fluorescence measurement of the [Ca 21] i of human neutrophils before (base line) and after (peak) the addition of PAF. Each panel indicates an identical experiment done in the presence of increasing amounts of functional Activator.
Activator to transport a self-quenching fluorescence dye, rhodamine conjugated to a saturated C-18 hydrocarbon chain (R-18), between labeled and unlabeled liposomes (phosphatidylcholine containing large unilamellar vesicles, PC-LUV s). Then the protein’s ability to bind various ligands is assessed by using each test compound as a specific inhibitor of the transport process (8).
RESULTS The addition of extracellular Activator (functional or non-functional) to a suspension of INDO-1 AM-loaded cells did not result in a change in [Ca 21] i (data not shown). When 2 pmol of PAF was added in the absence of the functional Activator the [Ca 21] i was increased by 401 nM. However, if the mixtures contained the functional Activator, a significant decrease in the release of intracellular calcium pools into the cytosol in response to PAF was noted (Fig. 1). The PAF-response was inhibited by ;28% or ;75% when 320 or 800 pmol of Activator was present (Fig. 1), respectively, but not at all in the presence of 800 pmol of the non-functional (see methods section) truncated Activator (data not shown).
257
Vol. 258, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Fluorescence dequenching assay using 5 mg (0.26 nmol) of functional Activator and 1.3 nmol of the indicated ligands (GM2 AP, is the Activator protein alone) as potential inhibitors. The degree of inhibition is directly related to the affinity of the Activator for the ligand.
The mechanism for the above inhibition of the PAF response in human neutrophils was probed. Since the Activator is known to bind some negatively charged gangliosides, it was possible that the Activator was directly binding PAF. To determine if this was the case we utilized our fluorescence dequenching assay. Using 5 mg of Activator and 1.3 nmol of GM2, PAF, fMLP, or LTB 4 , it was found that only GM2 and PAF inhibited the activator’s transfer of R-18 between liposomes (Fig. 2). PAF inhibited R-18 transport by 34% as compared to 77% for the same amount of GM2. Thus PAF inhibits the activator’s transport function less than does an identical amount of GM2, but more than GM1 (25% (8)). DISCUSSION During the past two decades, studies describing the chemistry and biology of PAF have been extensive. This potent phosphoacylglycerol exhibits a wide variety of physiological and pathophysiological effects in various cells and tissues. PAF acts, through specific receptors and a variety of signal transduction systems, to elicit diverse biochemical responses (23). PAF’s role in inflammatory response is particularly important. The movement of polymorphonuclear leukocytes (PMNs) from the mainstream of blood to the extravascular space is a characteristic feature of the inflammatory response. This process requires that the PMN initially contacts the endothelium, then adheres firmly to the vessel wall, and finally migrates out of the microvasculature. Each of these events requires signals or pro-inflammatory molecules that direct the PMN to the potential site of inflammation. These molecules include histamine, which appears to be of importance in the initial recruitment of PMNs, LTB 4, which promotes PMN adhesion, and PAF, which may contribute to both the adhesion process as well as the migration through the endothelial barrier. Thus, PAF
and cytokines, such as interleukins, tumor necrosis factor, and others, are thought to play a role in the inflammatory process involving gastrointestinal disorders such as Crohn’s disease, ulcerative colitis, ischemic colitis, or antibiotic-associated colitis (24). PAF is also one of the chemical mediators that may participate in the inflammatory process underlying asthma. It has been shown to produce a prolonged increase in airway responsiveness in several species, including humans. In humans the increase in bronchial reactivity peaks 3 days after inhalation and can last as long as several weeks. PAF may induce this response by recruiting eosinophils into the airways and activating them in the airway walls, with the subsequent release of basic proteins. This process causes eosinophils, which are rich sources of PAF, to release their contents and sets up a self-perpetuating cycle (25). Because of PAF’s important role in the pathogenesis of inflammatory bowel disease and asthma, inhibition of PAF by specific antagonists, mediators, or other agents may have a potential therapeutic benefit in treatment and management of these inflammatory diseases. During an experiment aimed at determining if exogenously added Activator protein could cause neutrophils to release calcium, which used PAF as a positive control, we discovered that; while the Activator did not directly affect this process, it inhibited PAF’s ability to do so (Fig. 1). The effects of other compounds that elicit Ca 21 release by these cells, i.e. fMLP and LTB4, were not inhibited by the presence of Activator in the assay (data not shown). Further experiments using a fluorescence dequenching assay (8), demonstrated that the Activator protein binds PAF, but not fMLP or LTB4 (Fig. 2). Thus, the mechanism by which the Activator inhibits the PAF-response is through direct binding of the compound, presumably preventing PAF’s interaction with its receptors. Since cells can endocytose the Activator by a mechanism that is not dependent on either its carbohydrate moiety or whether or not it has formed a glyco-lipid complex (12), the Activator could serve to remove PAF from extracellular fluids, bypassing PAF receptors. Other than being lipids coupled with hydrophilic head-groups that each contain a negatively charged component (phosphate or NeuAc), the structures of PAF and GM2 ganglioside are not very similar. It was thus surprising that the Activator bound GM2 . PAF . GM1. However, the Activator has recently also been shown to activate ADP ribosylation factor-dependent phospholipase D (17). This enzyme catalyzes the hydrolysis of phosphatidylcholine to generate lipid mediators that control a wide range of physiological and pathological processes. This study did not identify the Activator’s mechanism of action; e.g., Activator binding to phosphatidylcholine was not assessed. Given our results with PAF, the ability of the Activator to bind phosphatidylcholine would not be surprising.
258
Vol. 258, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
Most PAF inhibitors are analogs which act by blocking the PAF receptors (23, 26). Because of receptorheterogeneity, conventional agonists may only inhibit or have a differential effect on a subset of PAFreceptors. Furthermore there are issues of the overall toxicity of the agonist and its ability to gain access to the various cellular PAF-receptors. A naturally occurring protein normally secreted by different human cell types (12, 15), that directly binds secreted PAF, and is heat stable and protease resistant, such as the Activator, would have none of these drawbacks. Thus, theoretically if the Activator protein could be delivered in a concentrated form directly to the site of inflammation, it could be useful in the treatment and management of these diseases. In addition the future determination of the three dimensional structure of the Activator at the atomic level, through X-ray crystallography studies, may suggest changes that could be made in its primary structure to improve its specificity towards PAF and increase its pH optimum for ligand-binding. ACKNOWLEDGMENTS This work was supported by a grant from the Medical Research Council of Canada to D.M. We also thank Dr. C. R. Pace-Asciak for his advice concerning the Ca 21 release assay.
REFERENCES 1. Fu¨rst, W., and Sandhoff, K. (1992) Biochim. Biophys. Acta 1126, 1–16. 2. Mahuran, D. J. (1998) Biochem. Biophys. Acta 1393, 1–18. 3. Xie, B., Rigat, B., Smiljanic-Georgijev, N., Deng, H., and Mahuran, D. J. (1998) Biochemistry 37, 814 – 821. 4. Schu¨tte, C. G., Lemm, T., Glombitza, G. J., and Sandhoff, K. (1998) Protein Sci. 7, 1039 –1045. 5. Meier, E. M., Schwarzmann, G., Fu¨rst, W., and Sandhoff, K. (1991) J. Biol. Chem. 266, 1879 –1887. 6. Conzelmann, E., Burg, J., Stephan, G., and Sandhoff, K. (1982) Eur. J. Biochem. 123, 455– 464.
7. Conzelmann, E., and Sandhoff, K. (1979) Hoppe-Seyler’s Z. Physiol. Chem. 360, 1837–1849. 8. Smiljanic-Georgijev, N., Rigat, B., Xie, B., Wang, W., and Mahuran, D. J. (1997) Biochim. Biophys. Acta 1339, 192–202. 9. Sandhoff, K., Harzer, K., and Fu¨rst, W. (1995) in The Metabolic Basis of Inherited Disease (Scriver, C. R., Beaudet, A. L., Sly, W. S., and Valle, D., Eds.), Vol. 2, 7 Ed., pp. 2427–2441, McGrawHill, New York. 10. Wu, Y. Y., Lockyer, J. M., Sugiyama, E., Pavlova, N. V., Li, Y.-T., and Li, S.-C. (1994) J. Biol. Chem. 269, 16276 –16283. 11. Klima, H., Klein, A., Van Echten, G., Schwarzmann, G., Suzuki, K., and Sandhoff, K. (1993) Biochem. J. 292, 571–576. 12. Rigat, B., Wang, W., Leung, A., and Mahuran, D. J. (1997) Biochemistry 36, 8325– 8331. 13. Kuwana, T., Mullock, B. M., and Luzio, J. P. (1993) Biochem. Soc. Trans. 21, 299 –300. 14. Kuwana, T., Mullock, B. M., and Luzio, J. P. (1995) Biochem. J. 308, 937–946. 15. Glombitza, G. J., Becker, E., Kaiser, H. W., and Sandhoff, K. (1997) J. Biol. Chem. 272, 5199 –5207. 16. Mundel, T. M., Heid, H. W., Mahuran, D. J., Kriz, W., and Mundel, P. (1999) J. Am. Soc. Nephrol. 10, 435– 43. 17. Nakamura, S., Akisue, T., Jinnai, H., Hitomi, T., Sarkar, S., Miwa, N., Okada, T., Yoshida, K., Kuroda, S., Kikkawa, U., and Nishizuka, Y. (1998) Proc. Natl. Acad. Sci. USA 95, 12249 – 12253. 18. Benfenati, F., Fuxe, K., and Agnati, L. F. (1991) Neurochem. Int. 19, 271–279. 19. Hilbush, B. S., and Levine, J. M. (1991) Proc. Natl. Acad. Sci. USA 88, 5616 –5620. 20. Goldenring, J. R., Otis, L. C., Yu, R. K., and DeLorenzo, R. J. (1985) J. Neurochem. 44, 1229 –1234. 21. Laneuville, O., Renaud, D., Grinstein, S., Nigam, S., and PaceAsciak, C. R. (1993) Biochem. J. 295, 393–397. 22. Reynaud, D., and Pace-Asciak, C. R. (1997) Prostaglandins Leukot. Essent. Fatty Acids 56, 9 –12. 23. Chao, W., and Olson, M. S. (1993) Biochem. J. 292, 617–29. 24. Kubes, P. (1993) Canadian Journal of Physiology & Pharmacology 71, 88 –97. 25. Barnes, P. J. (1993) J. Allergy Clin. Immuno. 92, 187–189. 26. Izumi, T., and Shimizu, T. (1995) Biochim. Biophys. Acta 1259, 317–33.
259