A Dominant Negative Mutant of Bacillus anthracisProtective Antigen Inhibits Anthrax Toxin Action in Vivo
2001; Elsevier BV; Volume: 276; Issue: 25 Linguagem: Inglês
10.1074/jbc.m010222200
ISSN1083-351X
AutoresYogendra Singh, Hemant Khanna, Arun P. Chopra, Varsha Mehra,
Tópico(s)Microbial Inactivation Methods
ResumoPA63, a proteolytically activated 63-kDa form of anthrax protective antigen (PA), forms heptameric oligomers and has the ability to bind and translocate the catalytic moieties, lethal factor (LF), and edema factor (EF) into the cytosol of mammalian cells. Acidic pH triggers oligomerization and membrane insertion by PA63. A disordered amphipathic loop in domain II of PA (2β2–2β3 loop) is involved in membrane insertion by PA63. Because conditions required for membrane insertion coincide with those for oligomerization of PA63 in mammalian cells, residues constituting the 2β2–2β3 loop were replaced with the residues of the amphipathic membrane-inserting loop of its homologue iota-b toxin secreted by Clostridium perfringens. It was hypothesized that such a molecule might assemble into hetero-heptameric structures with wild-type PA ultimately leading to the inhibition of cellular intoxication. The mutation blocked the ability of PA to mediate membrane insertion and translocation of LF into the cytosol but had no effect on proteolytic activation, oligomerization, or binding LF. Moreover, an equimolar mixture of purified mutant PA (PA-I) and wild-type PA showed complete inhibition of toxin activity both in vitro on J774A.1 cells and in vivo in Fischer 344 rats thereby exhibiting a dominant negative effect. In addition, PA-I inhibited the channel-forming ability of wild-type PA on the plasma membrane of CHO-K1 cells thereby indicating protein-protein interactions between the two proteins resulting in the formation of mixed oligomers with defective functional activity. Our findings provide a basis for understanding the mechanism of translocation and exploring the possibility of the use of this PA molecule as a therapeutic agent against anthrax toxin action in vivo. PA63, a proteolytically activated 63-kDa form of anthrax protective antigen (PA), forms heptameric oligomers and has the ability to bind and translocate the catalytic moieties, lethal factor (LF), and edema factor (EF) into the cytosol of mammalian cells. Acidic pH triggers oligomerization and membrane insertion by PA63. A disordered amphipathic loop in domain II of PA (2β2–2β3 loop) is involved in membrane insertion by PA63. Because conditions required for membrane insertion coincide with those for oligomerization of PA63 in mammalian cells, residues constituting the 2β2–2β3 loop were replaced with the residues of the amphipathic membrane-inserting loop of its homologue iota-b toxin secreted by Clostridium perfringens. It was hypothesized that such a molecule might assemble into hetero-heptameric structures with wild-type PA ultimately leading to the inhibition of cellular intoxication. The mutation blocked the ability of PA to mediate membrane insertion and translocation of LF into the cytosol but had no effect on proteolytic activation, oligomerization, or binding LF. Moreover, an equimolar mixture of purified mutant PA (PA-I) and wild-type PA showed complete inhibition of toxin activity both in vitro on J774A.1 cells and in vivo in Fischer 344 rats thereby exhibiting a dominant negative effect. In addition, PA-I inhibited the channel-forming ability of wild-type PA on the plasma membrane of CHO-K1 cells thereby indicating protein-protein interactions between the two proteins resulting in the formation of mixed oligomers with defective functional activity. Our findings provide a basis for understanding the mechanism of translocation and exploring the possibility of the use of this PA molecule as a therapeutic agent against anthrax toxin action in vivo. protective antigen lethal factor edema factor 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide polyacrylamide gel electrophoresis N-terminal 254 amino acids of LF Bacillus anthracis, the etiologic agent of anthrax, is a potential agent of bioterrorism (1Marshall E. Science. 2000; 289: 382-383Crossref PubMed Scopus (3) Google Scholar). The toxic action has been attributed to anthrax toxin produced by the bacterium. The anthrax toxin can be resolved into three distinct protein components: protective antigen (PA),1lethal factor (LF), and edema factor (EF). The combination of EF and PA (an edema toxin) produces skin edema, whereas LF and PA (a lethal toxin) are lethal to animals (2Stanley J.L. Smith H. J. Gen. Microbiol. 1961; 26: 49-66Crossref PubMed Google Scholar). The three proteins are individually non-toxic (2Stanley J.L. Smith H. J. Gen. Microbiol. 1961; 26: 49-66Crossref PubMed Google Scholar). Whereas EF is a calcium- and calmodulin-dependent adenylate cyclase that acts by increasing the intracellular cAMP levels in eukaryotic cells (3Leppla S.H. Proc. Natl. Acad. Sci. U. S. A. 1982; 79: 3162-3166Crossref PubMed Scopus (756) Google Scholar), LF is a Zn2+-dependent metalloprotease (4Klimpel K.R. Arora N. Leppla S.H. Mol. Microbiol. 1994; 13: 1093-1100Crossref PubMed Scopus (263) Google Scholar) that leads to an increase in IL-1 and TNF-α production by susceptible cells (5Hanna P.C. Acosta D. Collier R.J. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 10198-10201Crossref PubMed Scopus (269) Google Scholar) and cleaves several mitogen-activated protein kinase kinases (MKK 1, 2 and 3) (6Duesbery N.S. Webb C.P. Leppla S.H. Gordon V.M. Klimpel K.R. Copeland T.D. Ahn N.G. Oskarsson M.K. Fukasawa K. Paull K.D. Vande Woude G.F. Science. 1998; 280: 734-737Crossref PubMed Scopus (885) Google Scholar, 7Vitale G. Pellizzari R. Recchi C. Napolitani G. Mock M. Montecucco C. Biochem. Biophys. Res. Commun. 1998; 248: 706-711Crossref PubMed Scopus (361) Google Scholar, 8Pellizzari R. Guidi-Rontani C. Vitale G. Mock M. Montecucco C. FEBS Lett. 1999; 462: 199-204Crossref PubMed Scopus (254) Google Scholar).According to the current model of anthrax toxin action, PA binds to an as yet unknown cell surface receptor and gets proteolytically activated by cell surface protease furin to PA63. This allows oligomerization and binding of LF/EF. The toxin complex is internalized by receptor-mediated endocytosis and is exposed to acidic pH inside the endosome. This change in pH triggers both membrane insertion by PA63 and translocation of LF/EF into the cytosol (recently reviewed in 9).Membrane insertion and channel formation are brought about by a large 2β2–2β3 loop (amino acid residues 302–325) in the domain II of PA (10Benson E.L. Huynh P.D. Finkelstein A. Collier R.J. Biochemistry. 1998; 37: 3941-3948Crossref PubMed Scopus (162) Google Scholar). The loop shows a conserved pattern of alternating hydrophilic and hydrophobic amino acid residues similar to that observed inClostridium perfringens iota-b toxin andStaphylococcus aureus α-hemolysin (11Petosa C. Collier R.J. Klimpel K.R. Leppla S.H. Liddington R.C. Nature. 1997; 385: 833-838Crossref PubMed Scopus (676) Google Scholar). PA also has been shown to possess a high degree of homology with the iota-b toxin secreted by C. perfringens (12Perelle S. Gibert M. Boquet P. Popoff M.R. Infect. Immun. 1993; 61: 5147-5156Crossref PubMed Google Scholar).Translocation of LF or EF to the cytosol is believed to occur through a channel formed by insertion of heptameric PA63 into the membrane (11Petosa C. Collier R.J. Klimpel K.R. Leppla S.H. Liddington R.C. Nature. 1997; 385: 833-838Crossref PubMed Scopus (676) Google Scholar). The formation of ion-conductive channels by PA63 has been demonstrated in both artificial lipid membranes (13Koehler T.M. Collier R.J. Mol. Microbiol. 1991; 5: 1501-1506Crossref PubMed Scopus (77) Google Scholar) and in CHO-K1 cells (14Milne J.C. Collier R.J. Mol. Microbiol. 1993; 10: 647-653Crossref PubMed Scopus (132) Google Scholar). Acidic pH triggers oligomerization, membrane insertion by PA63, and translocation of LF into the cytosol of mammalian cells (10Benson E.L. Huynh P.D. Finkelstein A. Collier R.J. Biochemistry. 1998; 37: 3941-3948Crossref PubMed Scopus (162) Google Scholar, 15Milne J.C. Furlong D. Hanna P.C. Wall J.S. Collier R.J. J. Biol. Chem. 1994; 269: 20607-20612Abstract Full Text PDF PubMed Google Scholar, 16Singh Y. Klimpel K.R. Arora N. Sharma M. Leppla S.H. J. Biol. Chem. 1994; 269: 29039-29046Abstract Full Text PDF PubMed Google Scholar). In this paper, we show that a mutant PA protein, in which amino acid residues comprising the 2β2–2β3 loop of PA (PA-I) were substituted with the residues of the amphipathic loop of the homologous iota-b toxin, is defective in its ability to insert into the membrane and completely inhibits the lethal effect of the wild-type toxin at equimolar concentrations.RESULTS AND DISCUSSIONPrior work showed that proteolytic cleavage of PA at the sequence164RKKR167 in solution or on the surface of mammalian cells results in the removal of the N-terminal 20-kDa fragment (PA20) that leads to heptamer formation (11Petosa C. Collier R.J. Klimpel K.R. Leppla S.H. Liddington R.C. Nature. 1997; 385: 833-838Crossref PubMed Scopus (676) Google Scholar). The heptamer has been assumed to insert into membranes at acidic pH (15Milne J.C. Furlong D. Hanna P.C. Wall J.S. Collier R.J. J. Biol. Chem. 1994; 269: 20607-20612Abstract Full Text PDF PubMed Google Scholar). Acidic pH inside the endosome leads to insertion of PA63 into the membrane by forming a β-barrel composed of an amphipathic 2β2–2β3 loop consisting of an alternating arrangement of hydrophilic and hydrophobic amino acids (11Petosa C. Collier R.J. Klimpel K.R. Leppla S.H. Liddington R.C. Nature. 1997; 385: 833-838Crossref PubMed Scopus (676) Google Scholar).Because acidic pH is necessary for both oligomerization and membrane insertion by PA63 (15Milne J.C. Furlong D. Hanna P.C. Wall J.S. Collier R.J. J. Biol. Chem. 1994; 269: 20607-20612Abstract Full Text PDF PubMed Google Scholar), we investigated the functional significance of the correlation between the conditions required for both of the events to occur. Mutant PA protein (PA-I) was produced in which residues constituting the 2β2–2β3 loop were replaced with the corresponding residues of iota-b toxin, a closely related toxin secreted byC. perfringens (12Perelle S. Gibert M. Boquet P. Popoff M.R. Infect. Immun. 1993; 61: 5147-5156Crossref PubMed Google Scholar), so that the alternating arrangement of hydrophilic and hydrophobic amino acids was retained. PA and PA-I were purified from the culture supernatant of B. anthracis (Fig.1 A). PA-I was tested by immunoblot analysis for reactivity against anti-PA polyclonal antibodies (Fig. 1 B). The results suggested that PA-I was purified to more than 90% homogeneity and was reactive to anti-PA rabbit polyclonal antibodies like wild-type PA. The typical yield of the proteins was 10 mg/l. PA-I did not aggregate in solution and behaved similar to wild-type PA.Purified PA-I was tested for its ability to lyse J774A.1 macrophage cells in combination with LF. Whereas wild-type PA lysed 50% of the cells at a concentration of 0.04 μg/ml in 3 h (Fig.2 A), PA-I was completely non-toxic to J774A.1 cells at the highest concentration tested (100 μg/ml) (Fig. 2 A). The data suggest that PA-I is inactive in exhibiting a lethal effect to macrophage cells as compared with wild-type PA. A more sensitive assay to measure the toxic activity of PA is to study the PA-dependent inhibition of protein synthesis in combination with LF-(1–254)·TR·PE-(398–613) (16Singh Y. Klimpel K.R. Arora N. Sharma M. Leppla S.H. J. Biol. Chem. 1994; 269: 29039-29046Abstract Full Text PDF PubMed Google Scholar). The fusion protein is comprised of the N-terminal 254 amino acids of LF (LF-(1–254); LFn) fused to the ADP-ribosylating domain of Pseudomonas aeruginosa exotoxin (16Singh Y. Klimpel K.R. Arora N. Sharma M. Leppla S.H. J. Biol. Chem. 1994; 269: 29039-29046Abstract Full Text PDF PubMed Google Scholar). Cytotoxicity in this assay is measured by the inhibition of protein synthesis catalyzed by Pseudomonas exotoxin and resulting from the ADP-ribosylation of elongation factor 2 (22Arora N. Klimpel K.R. Singh Y. Leppla S.H. J. Biol. Chem. 1992; 267: 15542-15548Abstract Full Text PDF PubMed Google Scholar). Whereas wild-type PA (0.1 μg/ml) showed 90% inhibition in protein synthesis when administered in combination with LF-(1–254)·TR·PE-(398–613) (1 ng/ml) to CHO-K1 cells as measured by the percent incorporation of [3H]leucine, no inhibition in protein synthesis was detected with PA-I when used even at a concentration of 100 μg/ml (Fig. 2 B). The marked inhibition of the biological activity of the mutant PA protein confirmed the functional significance of 2β2–2β3 loop of PA for the biological activity of anthrax toxin.Figure 2Cytotoxicity assay with PA and mutant PA proteins. A, J774A.1 cells were cultured in 96-well plates in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum and incubated with LF (1 μg/ml) in combination with varying concentrations of PA and PA-I for 3 h at 37 °C. At the end of the experiment, toxicity was determined by MTT assay as described under "Experimental Procedures." B, CHO-K1 cells were incubated with varying concentrations of PA or PA-I together with LF-(1–254)·TR·PE-(398–613) (1 ng/ml) at 37 °C for 3 h. At the end of the incubation, medium was replaced with medium containing [3H]leucine (1 μCi/ml) and the extent of protein synthesis was measured as described under "Experimental Procedures." Results are expressed as percent of [3H]leucine incorporation by viable cells in the absence of added proteins. Counts equivalent to 12,500 were considered as 100% [3H]leucine incorporation.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Several previous studies have shown that cleavage at the sequence164RKKR167 by trypsin/furin is a prerequisite for anthrax toxin action (17Singh Y. Chaudhary V.K. Leppla S.H. J. Biol. Chem. 1989; 264: 19103-19107Abstract Full Text PDF PubMed Google Scholar) and leads to the formation of SDS-resistant oligomers by PA63 at acidic pH (15Milne J.C. Furlong D. Hanna P.C. Wall J.S. Collier R.J. J. Biol. Chem. 1994; 269: 20607-20612Abstract Full Text PDF PubMed Google Scholar). Analysis of the mutant PA protein for sensitivity toward trypsin showed that the mutant was equally susceptible to trypsin as wild-type PA (TableI). The mutation introduced did not affect the ability of PA-I to bind LFn on cell surface (Table I) and, therefore, did not alter the ability of PA to bind to the receptor.Table ICharacteristics of mutant PA proteinExperiments with cultured cellsExperiments with Fischer 344 ratsPAPA-I 1-aSequence at 2β2–2β3 loop: 302TVGVSISAGYQNGFTGNITTSAG324.PALFPA-IPA-DTTD 1-bTTD is the time to death of Fischer 344 rats after administration of proteins.μgμgμgμgToxicity1-cToxicity was determined by dye oxidation assay (19) and inhibition of protein synthesis (16).+−40−−−Survived−8−−SurvivedTrypsin cleavage 1-dPA and PA-I (1 mg/ml) were incubated with trypsin (1 μg/ml) for 30 min at 22 °C and analyzed on 10% SDS-PAGE followed by staining with Coomassie Blue. The extent of cleavage was estimated visually and compared to that of wild-type PA.100%100%408−−60LF binding 1-eBinding of LF to PA and PA-I was measured by incubating trypsin-nicked PA (2 μg/ml) and PA-I (2 μg/ml) with CHO-K1 cells at 4 °C for 2 h. Cells were then washed and incubated with 35S-labeled LFn for 1 h. After another washing step, cell-associated radioactivity was measured. Counts equivalent to 4500 dpm were taken as 100%.100%97%40840−SurvivedOligomerization 1-fTrypsin-nicked PA and PA-I were incubated with CHO-K1 cells for 2 h at 4 °C. After washing, cells were treated with buffer of pH 5.0 and incubated at 37 °C for 30 min. Cells were then solubilized in SDS sample buffer, analyzed on 4–25% SDS-PAGE followed by immunoblotting, and detected using chemiluminescence detection kit.++408−40701-a Sequence at 2β2–2β3 loop: 302TVGVSISAGYQNGFTGNITTSAG324.1-b TTD is the time to death of Fischer 344 rats after administration of proteins.1-c Toxicity was determined by dye oxidation assay (19Quinn C.P. Singh Y. Klimpel K.R. Leppla S.H. J. Biol. Chem. 1991; 266: 20124-20130Abstract Full Text PDF PubMed Google Scholar) and inhibition of protein synthesis (16Singh Y. Klimpel K.R. Arora N. Sharma M. Leppla S.H. J. Biol. Chem. 1994; 269: 29039-29046Abstract Full Text PDF PubMed Google Scholar).1-d PA and PA-I (1 mg/ml) were incubated with trypsin (1 μg/ml) for 30 min at 22 °C and analyzed on 10% SDS-PAGE followed by staining with Coomassie Blue. The extent of cleavage was estimated visually and compared to that of wild-type PA.1-e Binding of LF to PA and PA-I was measured by incubating trypsin-nicked PA (2 μg/ml) and PA-I (2 μg/ml) with CHO-K1 cells at 4 °C for 2 h. Cells were then washed and incubated with 35S-labeled LFn for 1 h. After another washing step, cell-associated radioactivity was measured. Counts equivalent to 4500 dpm were taken as 100%.1-f Trypsin-nicked PA and PA-I were incubated with CHO-K1 cells for 2 h at 4 °C. After washing, cells were treated with buffer of pH 5.0 and incubated at 37 °C for 30 min. Cells were then solubilized in SDS sample buffer, analyzed on 4–25% SDS-PAGE followed by immunoblotting, and detected using chemiluminescence detection kit. Open table in a new tab Purified PA63 forms SDS and boiling-resistant oligomers when exposed to acidic pH on mammalian cells (15Milne J.C. Furlong D. Hanna P.C. Wall J.S. Collier R.J. J. Biol. Chem. 1994; 269: 20607-20612Abstract Full Text PDF PubMed Google Scholar). We, thus, determined whether PA-I retained the ability to form SDS-resistant oligomers in solution as well as when incubated with mammalian cells. PA-I was equally as effective in oligomerizing in solution as was wild-type PA (Table I). Electron microscopy data confirmed the formation of heptamers by PA-I as well as wild-type PA (Fig. 3,A and B). These results suggest that PA-I retained the ability to perform intermolecular interaction leading to oligomerization.Figure 3Electron microscopy of the oligomeric 63-kDa fragments of PA and PA-I. Samples (40 μg/ml) were negatively stained with 1% uranyl formate and analyzed by electron microscopy.A, PA; B, PA-I.View Large Image Figure ViewerDownload Hi-res image Download (PPT)We next examined the ability of PA-I to insert into the plasma membrane of CHO-K1 cells at acidic pH. Membrane insertion was tested by measuring the ability of PA-I to form ion-conductive channels in the plasma membrane of mammalian cells. PA63 forms ion-conductive selective channels in artificial membranes and plasma membrane of mammalian at acidic pH (13Koehler T.M. Collier R.J. Mol. Microbiol. 1991; 5: 1501-1506Crossref PubMed Scopus (77) Google Scholar, 14Milne J.C. Collier R.J. Mol. Microbiol. 1993; 10: 647-653Crossref PubMed Scopus (132) Google Scholar). Earlier studies have correlated the ability of PA to insert into membranes with the extent of86Rb+ released at acidic pH (16Singh Y. Klimpel K.R. Arora N. Sharma M. Leppla S.H. J. Biol. Chem. 1994; 269: 29039-29046Abstract Full Text PDF PubMed Google Scholar). Cells preloaded with 86Rb+ were incubated with trypsin-nicked PA or PA-I at 4 °C and placed into acidic medium, and release of 86Rb+ was measured. Incubation of trypsin-nicked wild-type PA (2 μg/ml) induced release of 70%86Rb+, whereas trypsin-nicked PA-I did not cause leakage of 86Rb+ (Fig.4). The result suggests that PA-I is unable to form ion-conductive channels and that integrity of the 2β2–2β3 loop is essential for proper membrane insertion by PA63.Figure 4Ion channel formation by PA and PA-I.CHO-K1 cells preloaded with 86Rb+ were incubated with trypsin-nicked PA or PA-I at 4 °C for 2 h and subjected to acidic pH shock. The leakage of86Rb+ into the medium was determined as described under "Experimental Procedures." Results are expressed as percent of 86Rb+ associated with cells in the absence of added proteins.View Large Image Figure ViewerDownload Hi-res image Download (PPT)To test the ability of PA-I to translocate LF into the cytosol of mammalian cells, we employed the previously described in vitro translocation assay (20Wesche J. Elliott J.L. Falnes P.φ. Olsnes S. Collier R.J. Biochemistry. 1998; 37: 15737-15746Crossref PubMed Scopus (171) Google Scholar) that uses the in vitrotranscribed and translated LFn labeled with [35S]methionine. We measured the ability of PA-I to translocate radiolabeled LFn across the plasma membrane of CHO-K1 cells upon treatment with low pH buffer. As shown in Fig.5, whereas wild-type PA translocated 45% of the bound LFn at pH 5.0, PA-I showed no translocation of LFn. Insignificant translocation of LFn was observed with wild-type PA at pH 7.0 consistent with the earlier reports that showed that acidic pH is a prerequisite for the translocation event to occur (20Wesche J. Elliott J.L. Falnes P.φ. Olsnes S. Collier R.J. Biochemistry. 1998; 37: 15737-15746Crossref PubMed Scopus (171) Google Scholar). The results suggest that the mutant PA protein is inactive in translocation. The results confirm the previous propositions that membrane insertion by PA63 is a prerequisite for the translocation of LF into the cytosol but do not, in any way, suggest that LF passes through the lumen of the channel formed by the 2β2–2β3 loop.Figure 5In vitro translocation assay.CHO-K1 cells were chilled in ice for 15 min and incubated with trypsin-nicked PA or PA-I for 2 h at 4 °C. After washing with phosphate-buffered saline, the cells were further incubated for 1 h with 35S-labeled LFn at 4 °C. After another washing step, the pH of the medium was either maintained at 7.0 or lowered to 5.0. The cells were either lysed directly or treated with Pronase E to remove surface bound proteins. Cell-associated radioactivity was then counted. Results are expressed as percent 35S-labeled LFn translocated of the total bound to cells.View Large Image Figure ViewerDownload Hi-res image Download (PPT)We next investigated whether mixing of the mutant PA protein and wild-type PA at varying ratios resulted in alterations in the cytotoxic activity of the wild-type toxin. When the mutant and wild-type PA were present at equimolar concentrations, complete inhibition in protein synthesis of CHO-K1 cells was observed (Fig.6). A significant inhibition could be detected when the ratio of PA-I to PA was 1:4. These data suggest that the PA-I inhibits wild-type PA-mediated cellular intoxication.Figure 6Inhibition of toxic activity of PA in combination with LF -(1–254)·TR·PE-(398–613). CHO-K1 cells were incubated with PA-I or PA-D mixed with varying concentrations of wild-type PA at 37 °C for 3 h in combination with LF-(1–254)·TR·PE-(398–613). At the end of 3 h, cells were incubated with medium containing [3H]leucine (1 μCi/ml) for 1 h at 37 °C. At the end of the experiment, the amount of [3H]leucine incorporation was measured. Results are expressed as percent of [3H]leucine incorporated by viable cells in the absence of added proteins.View Large Image Figure ViewerDownload Hi-res image Download (PPT)It has been reported previously that a mutant form of PA, in which the furin cleavage site 164RKKR167 was deleted (PA-D), blocked the anthrax toxin effect in vitroand in vivo (17Singh Y. Chaudhary V.K. Leppla S.H. J. Biol. Chem. 1989; 264: 19103-19107Abstract Full Text PDF PubMed Google Scholar). Because this molecule cannot be cleaved with trypsin or furin, does not form oligomers, and can only inhibit the action of wild-type PA by competing for cellular receptor, this molecule was employed as a control in our experiments to investigate whether inhibition in the cytotoxic activity was due to the competition for binding to the receptor. PA-D failed to inhibit the cytotoxic effect when mixed with wild-type PA in all the ratios tested (Fig. 6). It has been shown previously that PA-D, rather than wild-type PA, inhibits the lethal toxin activity when present at a 10-fold excess concentration (17Singh Y. Chaudhary V.K. Leppla S.H. J. Biol. Chem. 1989; 264: 19103-19107Abstract Full Text PDF PubMed Google Scholar). Typically, a substantial excess of mutant protein is required to inhibit the binding of an active ligand to cell surface receptors, and PA-I more likely inhibits the action of anthrax toxin by interacting with wild-type PA to form an inactive hetero-heptameric complex and thus, is a more potent inhibitor of anthrax toxin action. This model is consistent with the ability of purified PA-I to inhibit wild-type PA-mediated cytotoxic activity when the ratio of PA-I to PA is 1:4. Purification of active homoheptamers of PA and PA-I to homogeneity was not successful due to the presence of lower order oligomers as well.To confirm the hypothesis that wild-type PA and PA-I might assemble to form non-functional oligomeric structures, the trypsin-nicked proteins (2 μg/ml each) were mixed together at neutral pH and incubated with CHO-K1 cells preloaded with 86Rb+ at 4 °C. After 2 h, the cells were washed to remove unbound proteins and incubated with isotonic buffer of pH 5.0 or 7.0 for 30 min at 37 °C. Whereas wild-type PA released 62% of the radiolabel from cells, equimolar mixture containing PA and PA-I showed insignificant release of 86Rb+ (Fig.7). The results suggest that there is complete inhibition of channel-forming ability of PA by PA-I. Indeed, the capacity of PA-I to dramatically alter the channel-forming ability of wild-type PA provides evidence that these two species can interact to form dysfunctional hetero-oligomeric structures. A dominant negative mutant of VacA toxin secreted by Helicobacter pylori has recently been reported that inhibits the vacuolating activity of wild-type toxin when present at a 5-fold-less concentration than wild-type VacA (23Vinion-Dubiel A.D. McClain M.A. Czajkowsky D.M. Iwamoto H. Ye D. Cao P. Schraw W. Szabo G. Blanke S.R. Shao Z. Cover T.L. J. Biol. Chem. 1999; 274: 37736-37742Abstract Full Text Full Text PDF PubMed Scopus (112) Google Scholar).Figure 7Inhibition of channel-forming activity of PA by PA-I. CHO-K1 cells, preloaded with86Rb+, were incubated with trypsin-cleaved PA and PA-I mixed in equimolar ratios at neutral pH for 2 h at 4 °C. After washing twice with cold phosphate-buffered saline, the cells were subjected to acidic pH shock as described under "Experimental Procedures." The leakage of86Rb+ into the medium was then determined. Results are expressed as percent of 86Rb+associated with cells in the absence of added proteins.View Large Image Figure ViewerDownload Hi-res image Download (PPT)Identification of such a dominant negative inhibitor might be valuable for treatment of anthrax toxin action. Animal experiments were thus initiated to test the efficacy of PA-I to act as a dominant negative inhibitor of lethal toxin action in vivo. Whereas wild-type lethal (40 μg of PA + 8 μg of LF) resulted in the death of male Fischer 344 rats in ∼60 min (Table I), a 1:1 mix containing wild-type PA and PA-I (40 μg of PA + 40 μg of PA-I + 8 μg LF) protected rats, and no symptoms were evident even after 48 h. Equimolar ratio of wild-type PA and PA-D resulted in the death of rats within 70 min (Table I). Taken together, these data confirm that PA-I can act as a dominant negative and a potent inhibitor of anthrax toxin actionin vivo.Use of B. anthracis as a bioweapon has become the bane of the defense establishments in various countries (1Marshall E. Science. 2000; 289: 382-383Crossref PubMed Scopus (3) Google Scholar). Keeping in view the potent activity of PA-I both in vitro and in vivo it has the potential to be used as therapeutic agent for use in neutralizing anthrax toxin action in individuals infected withB. anthracis. Bacillus anthracis, the etiologic agent of anthrax, is a potential agent of bioterrorism (1Marshall E. Science. 2000; 289: 382-383Crossref PubMed Scopus (3) Google Scholar). The toxic action has been attributed to anthrax toxin produced by the bacterium. The anthrax toxin can be resolved into three distinct protein components: protective antigen (PA),1lethal factor (LF), and edema factor (EF). The combination of EF and PA (an edema toxin) produces skin edema, whereas LF and PA (a lethal toxin) are lethal to animals (2Stanley J.L. Smith H. J. Gen. Microbiol. 1961; 26: 49-66Crossref PubMed Google Scholar). The three proteins are individually non-toxic (2Stanley J.L. Smith H. J. Gen. Microbiol. 1961; 26: 49-66Crossref PubMed Google Scholar). Whereas EF is a calcium- and calmodulin-dependent adenylate cyclase that acts by increasing the intracellular cAMP levels in eukaryotic cells (3Leppla S.H. Proc. Natl. Acad. Sci. U. S. A. 1982; 79: 3162-3166Crossref PubMed Scopus (756) Google Scholar), LF is a Zn2+-dependent metalloprotease (4Klimpel K.R. Arora N. Leppla S.H. Mol. Microbiol. 1994; 13: 1093-1100Crossref PubMed Scopus (263) Google Scholar) that leads to an increase in IL-1 and TNF-α production by susceptible cells (5Hanna P.C. Acosta D. Collier R.J. Proc. Natl. Acad. Sci. U. S. A. 1993; 90: 10198-10201Crossref PubMed Scopus (269) Google Scholar) and cleaves several mitogen-act
Referência(s)