FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2040898023> ?p ?o ?g. }

• W2040898023 endingPage "17889" @default.
• W2040898023 startingPage "17881" @default.
• W2040898023 abstract "Bacterial substrate-binding proteins exist in an equilibrium among four forms: open/substrate-free, open/substrate-bound, closed/substrate-free, and closed/substrate-bound. Ligands stabilize the closed conformation, whereas the open conformation predominates in the substrate-free species. In its closed form, the NH2-terminal and COOH-terminal domains of maltose-binding protein (MBP) are proposed to be aligned to allow residues in both domains to interact simultaneously with complementary sites on the MalF and MalG proteins of the maltodextrin uptake system or with the Tar chemotactic signal transducer. However, the initial interaction might occur with an open/substrate-bound form of the binding protein, which would then close in contact with MalFG or Tar. Ligand would help stabilize this complex. We introduced cysteines (G69C and S337C) by site-directed mutagenesis into each domain of MBP and found that they formed an interdomain disulfide cross-link that should hold the protein in a closed conformation. This mutant MBP confers a dominant-negative phenotype for growth on maltose, for maltose transport, and for maltose chemotaxis. The growth and transport defects are partially reversed when the cells are exposed to the reducing agent dithiothreitol. We conclude that the cross-linked form of MBP competes with wild-type MBP in vivo for interaction with MalFG and Tar. Bacterial substrate-binding proteins exist in an equilibrium among four forms: open/substrate-free, open/substrate-bound, closed/substrate-free, and closed/substrate-bound. Ligands stabilize the closed conformation, whereas the open conformation predominates in the substrate-free species. In its closed form, the NH2-terminal and COOH-terminal domains of maltose-binding protein (MBP) are proposed to be aligned to allow residues in both domains to interact simultaneously with complementary sites on the MalF and MalG proteins of the maltodextrin uptake system or with the Tar chemotactic signal transducer. However, the initial interaction might occur with an open/substrate-bound form of the binding protein, which would then close in contact with MalFG or Tar. Ligand would help stabilize this complex. We introduced cysteines (G69C and S337C) by site-directed mutagenesis into each domain of MBP and found that they formed an interdomain disulfide cross-link that should hold the protein in a closed conformation. This mutant MBP confers a dominant-negative phenotype for growth on maltose, for maltose transport, and for maltose chemotaxis. The growth and transport defects are partially reversed when the cells are exposed to the reducing agent dithiothreitol. We conclude that the cross-linked form of MBP competes with wild-type MBP in vivo for interaction with MalFG and Tar. INTRODUCTIONThe product of the malE gene of Escherichia coli, the periplasmic maltose-binding protein (MBP), 1The abbreviations used are: MBPmaltose-binding proteinDTTdithiothreitolPAGEpolyacrylamide gel electrophoresis. binds maltose and other maltodextrins with high affinity (1Kellermann O. Szmelcman S. Eur. J. Biochem. 1974; 47: 139-149Google Scholar). Ligand-bound MBP interacts with a membrane-bound complex of the MalF, MalG, and MalK proteins (2Froshauer S. Beckwith J. J. Biol. Chem. 1984; 259: 10896-10903Google Scholar, 3Dassa E. Hofnung M. EMBO J. 1985; 4: 2287-2293Google Scholar) to take up maltodextrins via a transmembrane channel formed by MalF and MalG (4Shuman H.A. Silhavy T.J. J. Biol. Chem. 1981; 256: 560-562Google Scholar, 5Davidson A.L. Nikaido H. J. Biol. Chem. 1991; 266: 8946-8951Google Scholar, 6Davidson A.L. Nikaido H. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 2360-2364Google Scholar). The NH2- and COOH-terminal domains of MBP appear to bind to MalG and MalF, respectively (7Treptow N.A. Shuman H.A. J. Mol. Biol. 1988; 202: 809-822Google Scholar). Certain mutations in malF and malG allow cells lacking MBP to transport maltodextrins (8Shuman H.A. J. Biol. Chem. 1982; 257: 5455-5461Google Scholar, 9Treptow N.A. Shuman H.A. J. Bacteriol. 1985; 163: 654-660Google Scholar). These mutant MalF-MalG (MalFG) complexes can stimulate ATP hydrolysis by MalK in the absence of MBP (6Davidson A.L. Nikaido H. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 2360-2364Google Scholar), suggesting that the binding of MBP to the periplasmic face of the MalFG complex normally activates MalK in the cytoplasm.Ligand-bound MBP interacts with the Tar chemotactic signal transducer to initiate an attractant response to maltose (10Hazelbauer G.L. J. Bacteriol. 1975; 122: 206-214Google Scholar, 11Manson M.D. Kossmann M. J. Bacteriol. 1986; 165: 34-40Google Scholar, 12Kossmann M. Wolff C. Manson M.D. J. Bacteriol. 1988; 170: 4516-4521Google Scholar, 13Gardina P. Conway C. Kossmann M. Manson M. J. Bacteriol. 1992; 174: 1528-1536Google Scholar, 14Zhang Y. Conway C. Rosato M. Suh Y. Manson M.D. J. Biol. Chem. 1992; 267: 22813-22820Google Scholar). Tar is proposed to form a homodimer (15Falke J.J. Koshland Jr., D.E. Science. 1987; 237: 1596-1600Google Scholar, 16Milligan D.L. Koshland Jr., D.E. J. Biol. Chem. 1988; 263: 6268-6275Google Scholar). Genetic (12Kossmann M. Wolff C. Manson M.D. J. Bacteriol. 1988; 170: 4516-4521Google Scholar, 13Gardina P. Conway C. Kossmann M. Manson M. J. Bacteriol. 1992; 174: 1528-1536Google Scholar, 17Wolff C. Parkinson J.S. J. Bacteriol. 1988; 170: 4509-4515Google Scholar) and structural (18Milburn M.V. Prive G.G. Milligan D.L. Scott W.G. Yeh J. Janacik J. Koshland Jr., D.E. Kim S.-H. Science. 1991; 254: 1342-1347Google Scholar) evidence indicates that the attractant L-aspartate binds at the dimer interface of the periplasmic domain of Tar, contacting residues in both subunits. Recent data from intergenic complementation studies (53Gardina, P., 1995, Attractant Binding and Signaling in the E. coli Aspartate Receptor. Ph.D. thesis, Texas A&M University.Google Scholar) is consistent with the prediction from computer modeling (19Stoddard B.L. Koshland Jr., D.E. Nature. 1992; 358: 774-776Google Scholar) that MBP also contacts both subunits of Tar.MBP is thought to function as a monomer in which the globular NH2- and COOH-terminal domains are connected by a flexible hinge (20Spurlino J.C. Lu G.-Y. Quiocho F.A. J. Biol. Chem. 1991; 266: 5202-5219Google Scholar, 21Sharff A.J. Rodseth L.E. Spurlino J.C. Quiocho F.A. Biochemistry. 1992; 31: 10657-10663Google Scholar). Short maltodextrins, like maltose and maltotriose, are buried in a cleft between the two domains when MBP closes (20Spurlino J.C. Lu G.-Y. Quiocho F.A. J. Biol. Chem. 1991; 266: 5202-5219Google Scholar). The nonreducing end of longer maltodextrins hangs out of the cleft, and bulky, nonphysiological ligands like cyclomaltoheptaose may prevent closure of the cleft almost completely when they bind to MBP (22Sharff A.J. Rodseth L.E. Quiocho F.A. Biochemistry. 1993; 32: 10553-10559Google Scholar). A bound maltodextrin forms extensive hydrophobic and hydrogen bonds with residues on each wall of the cleft to stabilize the closed conformation. To achieve the closed conformation, one domain of MBP undergoes a 35° rotation and an 8° lateral twist in the hinge relative to the other domain (21Sharff A.J. Rodseth L.E. Spurlino J.C. Quiocho F.A. Biochemistry. 1992; 31: 10657-10663Google Scholar).A working hypothesis is that when MBP assumes its closed conformation, residues in the two domains are brought into the correct spatial relationship to interact with complementary sites on the MalF-MalG complex or with the Tar dimer (14Zhang Y. Conway C. Rosato M. Suh Y. Manson M.D. J. Biol. Chem. 1992; 267: 22813-22820Google Scholar, 19Stoddard B.L. Koshland Jr., D.E. Nature. 1992; 358: 774-776Google Scholar, 20Spurlino J.C. Lu G.-Y. Quiocho F.A. J. Biol. Chem. 1991; 266: 5202-5219Google Scholar). However, since MBP (23Shilton B.H. Mowbray S.L. Protein Sci. 1995; 4: 1346-1355Google Scholar, 24Bohl E. Shuman H.A. Boos W. J. Theor. Biol. 1995; 172: 83-94Google Scholar) and other binding proteins (25Walmsley A.R. Shaw J.G. Kelly D.J. J. Biol. Chem. 1992; 267: 8064-8072Google Scholar, 26Walmsley A.R. Shaw J.G. Kelly D.J. Biochemistry. 1992; 31: 11175-11181Google Scholar, 27Wolf A. Shaw E.W. Nikaido K. Ames G.F.-L. J. Biol. Chem. 1994; 269: 23051-23058Google Scholar, 28Flocco M.M. Mowbray S.L. J. Biol. Chem. 1994; 269: 8931-8936Google Scholar) equilibrate between open and closed forms in the presence or absence of ligands, it is uncertain whether MalFG and Tar first interact with the open, closed, or some intermediate form of MBP.We have introduced mutations into a plasmid-borne malE gene to create the amino acid substitutions G69C and S337C in the normally cysteine-free MBP. Based on the crystal structure of the ligand-bound, closed form of MBP (20Spurlino J.C. Lu G.-Y. Quiocho F.A. J. Biol. Chem. 1991; 266: 5202-5219Google Scholar), 2J. C. Spurlino and F. A. Quiocho, unpublished data. we anticipated that cysteines at these positions could form an interdomain disulfide bridge. We have determined that this cysteine-substituted MBP does form such an intramolecular cross-link. Growth experiments, transport measurements, and chemotaxis assays strongly suggest that this cross-linking occurs in the periplasm in vivo. The double mutant MBP does not function in maltose transport or chemotaxis, and it inhibits, in a dose-dependent fashion, the function of wild-type MBP in transport and chemotaxis. This result demonstrates that the cross-linked protein, which we infer is in the closed conformation, retains biological activity and competes with wild-type MBP for interaction with MalFG and Tar in vivo.DISCUSSIONThe specificity of interaction of bacterial substrate-binding proteins with membrane-protein complexes responsible for transport and chemoreception has focused on the large conformational changes of the binding proteins as they go from their open to closed forms. An attractive model is that docking of the binding proteins with their membrane counterparts involves residues in both the NH2-terminal and COOH-terminal domains of the binding proteins. In the open form of the binding protein these residues are not in the right spatial relationship to interact simultaneously with their docking partners. The large conformational change that occurs as the binding protein assumes a closed form would then bring these residues into the correct juxtaposition to interact with the membrane partner. For MBP, this model is supported by structural (20Spurlino J.C. Lu G.-Y. Quiocho F.A. J. Biol. Chem. 1991; 266: 5202-5219Google Scholar, 21Sharff A.J. Rodseth L.E. Spurlino J.C. Quiocho F.A. Biochemistry. 1992; 31: 10657-10663Google Scholar), genetic (7Treptow N.A. Shuman H.A. J. Mol. Biol. 1988; 202: 809-822Google Scholar, 12Kossmann M. Wolff C. Manson M.D. J. Bacteriol. 1988; 170: 4516-4521Google Scholar, 14Zhang Y. Conway C. Rosato M. Suh Y. Manson M.D. J. Biol. Chem. 1992; 267: 22813-22820Google Scholar), and computer modeling (19Stoddard B.L. Koshland Jr., D.E. Nature. 1992; 358: 774-776Google Scholar) studies.Recent work from several laboratories (25Walmsley A.R. Shaw J.G. Kelly D.J. J. Biol. Chem. 1992; 267: 8064-8072Google Scholar, 26Walmsley A.R. Shaw J.G. Kelly D.J. Biochemistry. 1992; 31: 11175-11181Google Scholar, 27Wolf A. Shaw E.W. Nikaido K. Ames G.F.-L. J. Biol. Chem. 1994; 269: 23051-23058Google Scholar, 28Flocco M.M. Mowbray S.L. J. Biol. Chem. 1994; 269: 8931-8936Google Scholar) has suggested that binding proteins are in equilibrium between closed and open forms in the absence or presence of ligand. Rather than triggering closure of the binding proteins, as had originally been proposed (45Quiocho F.A. Phil. Trans. R. Soc. London B. 1990; 326: 341-351Google Scholar), recent data suggest that hydrogen bonding and hydrophobic stacking interactions between the ligand and the cleft-exposed faces of the two domains shift the equilibrium in favor of the closed conformation for the two binding proteins, MBP and the lysine-arginine-ornithine-binding protein, for which both open and closed structures have been solved (20Spurlino J.C. Lu G.-Y. Quiocho F.A. J. Biol. Chem. 1991; 266: 5202-5219Google Scholar, 21Sharff A.J. Rodseth L.E. Spurlino J.C. Quiocho F.A. Biochemistry. 1992; 31: 10657-10663Google Scholar, 46Oh B.-H. Pandit J. Kang C.-H. Nikaido K. Gokcen S. Ames G.F.-L. Kim S.-H. J. Biol. Chem. 1993; 268: 11348-11355Google Scholar). Techniques used to detect the closed/substrate-free form of periplasmic binding proteins include fluorescence changes of the C4-dicarboxylate-binding protein (DctP) of Rhodobacter capsulatus (25Walmsley A.R. Shaw J.G. Kelly D.J. J. Biol. Chem. 1992; 267: 8064-8072Google Scholar, 26Walmsley A.R. Shaw J.G. Kelly D.J. Biochemistry. 1992; 31: 11175-11181Google Scholar); trapping by conformation-specific antibodies with the histidine-binding protein (HisJ) of Salmonella typhimurium (27Wolf A. Shaw E.W. Nikaido K. Ames G.F.-L. J. Biol. Chem. 1994; 269: 23051-23058Google Scholar); and crystallization of the closed/substrate-free form of the galactose/glucose-binding protein (MglB) of S. typhimurium (28Flocco M.M. Mowbray S.L. J. Biol. Chem. 1994; 269: 8931-8936Google Scholar).The ligand-free form of binding proteins is not functionally inert. In a reconstituted histidine transport system employing purified HisJ and right-side-out membrane vesicles containing HisQMP, it was found that excess substrate-free HisJ inhibited transport (47Prossnitz E. Gee A. Ames G.F.-L. J. Biol. Chem. 1989; 264: 5006-5014Google Scholar). Also, the best fit to the data for maltose transport in Escherichia coli as a function of the concentration of maltose and periplasmic MBP (31Manson M.D. Boos W. Bassford Jr., P.J. Rasmussen B.A. J. Biol. Chem. 1985; 260: 9727-9733Google Scholar) was obtained when the theoretical model considered that the ligand-bound and the ligand-free forms of MBP compete for interaction with MalFG (23Shilton B.H. Mowbray S.L. Protein Sci. 1995; 4: 1346-1355Google Scholar, 24Bohl E. Shuman H.A. Boos W. J. Theor. Biol. 1995; 172: 83-94Google Scholar).It has usually been assumed that binding proteins initiate interaction with membrane transport proteins or chemotactic signal transducers when the binding proteins are in the closed conformation. However, the available data do not, in our view, rule out a priori that a binding protein, already associated with ligand, may have to assume an open form that can then “clamp down” on the membrane components. Ligand would then stabilize this complex. One scenario requiring such a mechanism would be that an element on a membrane protein might sterically interfere with binding of the closed form of MBP. This element might be accommodated in, or even contribute to, the ternary complex. In the case of transport, such an “intrusive” element could even facilitate subsequent opening of the binding protein to allow substrate release. Wolf et al. (48Wolf A. Shaw E.W. Oh B.-H. De Bondt H. Joshi A.K. Ames G.F.-L. J. Biol. Chem. 1995; 270: 16097-16106Google Scholar) demonstrated that mutant HisJ proteins defective in closing fail to interact normally with the HisQMP membrane transport complex. However, although this result is consistent with a requirement for the closed form in binding to HisQMP, the defect in closing could also interfere with a “clamping down” process of the type just described.Interdomain disulfide cross-bridges have been used before to study the properties of sulfate-binding protein (49Jacobson B.L. He J.J. Vermersch P.S. Lemon D.D. Quiocho F.A. J. Mol. Biol. 1992; 223: 27-30Google Scholar) and galactose/glucose-binding protein (50Careaga C.L. Sutherland J. Sabeti J. Falke J.J. Biochemistry. 1995; 34: 3048-3055Google Scholar) in vitro. MBP containing introduced cysteines at residues 69 and 337 (G69C and S337C) spontaneously forms such cross-bridges in the periplasmic space. This circumstance has allowed us to test whether such a cross-linked binding protein can bind to membrane transport and chemotaxis components in vivo.Besides G69C/S337C MBP, we constructed two other double mutant MBPs: S233C/P298C and S233C/A301C. In either mutant protein, if cross-bridges formed they would link the NH2-terminal and COOH-terminal domains of MBP at the other end of the cleft from the 69-337 cross-bridge (Fig. 1). We confirmed the presence of the S233C, P298C, and A301C substitutions by DNA sequencing, but no bands with a mobility different from that of wild-type MBP were observed during SDS-PAGE analysis (data not shown). We inferred that the cross-linking did not occur. We learned later that the 233 and 301 residues may be too far apart in the closed form of MBP to cross-link efficiently. 4F. A. Quiocho, personal communication. The 233 and 298 residues should be close enough to allow cross-linking,4 but the proline to cysteine substitution at residue 298 may distort local secondary structure, thus precluding cross-linking between Cys233 and Cys298. The distortion cannot be global, however, since the P298C and S233C/P298C proteins retain nearly wild-type function (data not shown).G69C/S337C MBP does not function in maltose transport (Fig. 5, Table II) or maltose chemotaxis (Fig. 7). The slow growth and residual transport observed in strains expressing G69C/S337C MBP probably depend on the fraction of the protein that remains noncross-linked (Fig. 3). Reduction of the disulfide bridge in vivo restores transport function to G69C/S337C MBP (Fig. 6, Table II), demonstrating that the two introduced cysteine residues per se do not seriously impair MBP function. Until the structure of the cross-linked form of the protein is determined, we cannot be certain what exact relationship its structure bears to that of the closed form of ligand-bound wild-type MBP.The conclusion that there is a good match between the structure of the cross-linked form of MBP and the closed, ligand-bound form of wild-type MBP is supported by the observation that G69C/S337C MBP is negatively co-dominant for growth on maltose and for maltose chemotaxis (Figs. 8, 9, 10). This result establishes that the cross-linked protein does not have a null phenotype, because it competes with wild-type MBP for MalFG and Tar. The specificity of the competition is demonstrated by the finding that the negative dominance of G69C/S337C MBP is greater when the amount of wild-type MBP is reduced by malE signal-sequence mutations (Fig. 9, Fig. 10).Three lines of evidence strongly suggest that G69C/S337C MBP forms an interdomain disulfide bridge in the periplasm. 1) Oxidizing agents other than atmospheric oxygen were not required to produce the cross-linked protein. 2) Singly cysteine-substituted mutant proteins and the double mutant protein formed nonspecific cross-links to many proteins when cells were lysed under nonreducing conditions, but only the double mutant protein was cross-linked in periplasmic fractions prepared by osmotic shock. 3) The purified G69C/S337C protein yielded the same band of about 45 kDa apparent molecular mass during SDS-PAGE that was seen in osmotic shock fluid. This last result rules out the possibility that the more slowly migrating form of G69C/S337C MBP is cross-linked to another membrane or periplasmic protein.Because the purified double mutant protein migrates at a molecular mass close to that of wild-type MBP in both native and denaturing gels (Fig. 4), the disulfide cross-link appears to be intramolecular, as we predicted, rather than intermolecular. The slight differences in mobility of the double mutant protein on native gels may reflect the presence of two conformations of MBP, even after reduction. One might predict the protein could assume the following forms: 1) an open, reduced conformation without maltose bound; 2) a closed, reduced form with maltose bound; 3) a closed, oxidized form with maltose bound; 4) a closed, oxidized form without maltose bound. Any or all of these forms could migrate somewhat differently on a native gel. Ligand-bound (closed) and ligand-free (open) forms of lysine-arginine-ornithine-binding protein are resolved by high pressure liquid chromatography (51Nikaido K. Ames G.F.-L. J. Biol. Chem. 1992; 267: 20706-20712Google Scholar). Furthermore, maltose bound to some fraction of the purified double mutant MBP could prevent the reduced form of the protein from opening, favoring rapid reoxidation and reformation of the cross-link. The separation of mutant, but not wild-type, liganded and unliganded forms may reflect that the cysteine substitutions alter the equilibrium thermodynamics or the kinetics of MBP opening and closing. Clearly, more work will be required to resolve these alternatives.An important, and still unanswered, question is whether the cross-linked G69C/S337C MBP must harbor maltose in its substrate-binding cleft in order to compete effectively with wild-type MBP. Jacobson et al. (49Jacobson B.L. He J.J. Vermersch P.S. Lemon D.D. Quiocho F.A. J. Mol. Biol. 1992; 223: 27-30Google Scholar) showed that an interdomain cross-linked form of G46C/S129C sulfate-binding protein has reduced interdomain flexibility and exhibits a much slower dissociation of sulfate. However, the Cys69-Cys337 cross-bridge is at one end of the cleft (Fig. 1), and it is possible that limited movement between domains still occurs at the other end of the cleft, allowing maltose to enter and leave the binding site. We wanted to make an MBP with two cross-bridges, one at each end of the cleft, which presumably would be truly “locked shut.” Unfortunately, as described above, we have not yet identified another pair of cysteine residues that could provide this second cross-link.Cross-linked MBP was observed in shock fluid prepared from cells grown in glycerol-minimal medium (Fig. 3), in which periplasmic maltose concentrations should be low or nonexistent. Also, preliminary investigations indicate that strains expressing G69C/S337C MBP show increased methylation of Tar in cells grown in minimal glycerol medium with or without maltose, whereas cells producing wild-type MBP show increased methylation of Tar only when they are grown with maltose. 5Y. Zhang, unpublished data. Thus, the double mutant protein appears to be able to induce an attractant signal via Tar without its having been exposed, at least intentionally, to maltose. We plan to measure the kinetics of maltose binding of purified cross-linked G69C/S337C MBP. It will be more difficult, although worthwhile, to determine rigorously whether cross-linked G69C/S337C MBP isolated from cells grown in maltose-free or maltose-replete medium is bound to maltose.We think that the cross-linked MBP will be a useful tool for further investigations of maltose transport and chemotaxis, both in vivo and in vitro. An advantage of our system for such studies is that any decrease in maltose affinity caused by chemotaxis-defective or transport-defective malE mutations should not influence the outcome, thus simplifying the analysis.Treptow and Shuman (7Treptow N.A. Shuman H.A. J. Mol. Biol. 1988; 202: 809-822Google Scholar) reported the isolation of malE missense mutations causing specific defects in maltose transport. Some of these have already been shown to be co-dominant (52Hor L.-I. Shuman H.A. J. Mol. Biol. 1993; 233: 659-670Google Scholar). Mutations in malE that cause transport defects can be introduced into the G69C/S337C mutant, and the ability of the triple mutant protein to inhibit transport mediated by wild-type MBP can be determined. If the triple mutant protein no longer inhibits, then the transport-defective mutation probably blocks MBP-MalFG binding. If the triple mutant protein still inhibits, then the mutation may interfere with the ability of MBP to initiate transmembrane signaling by MalFG rather than binding. Cross-linked MBP may also prove useful for measuring MBP-stimulated activities of the MalFGK transport complex in vitro (5Davidson A.L. Nikaido H. J. Biol. Chem. 1991; 266: 8946-8951Google Scholar, 6Davidson A.L. Nikaido H. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 2360-2364Google Scholar). Can, for example, binding of cross-linked MBP to MalFG stimulate ATP hydrolysis by MalK, and if so, how many rounds of ATP hydrolysis does the interaction trigger?We have accumulated a set of malE mutations that disrupt maltose chemotaxis (14Zhang Y. Conway C. Rosato M. Suh Y. Manson M.D. J. Biol. Chem. 1992; 267: 22813-22820Google Scholar), but analogous to the situation with the transport-defective malE mutations, we do not know whether the mutations interfere with binding of MBP to Tar or with generation of the chemotactic signal after the mutant MBP has bound to Tar. We can create triple-mutant MBP species similar to those described above for analyzing transport and ask if the third mutation eliminates the negative dominance for chemotaxis imposed by G69C/S337C MBP. We can also determine whether the triple mutant protein can stimulate increased methylation of Tar in the absence of maltose. The lack of an in vitro assay for MBP binding to Tar, largely because of the apparent low affinity of MBP for Tar (estimated in vivo at 250 µM; 31Manson M.D. Boos W. Bassford Jr., P.J. Rasmussen B.A. J. Biol. Chem. 1985; 260: 9727-9733Google Scholar), requires us to resort to another method to examine the structure/function relationships of the MBP-Tar interaction. We think that studying the effects of particular mutations on the negative-dominant phenotype of G69C/S337C MBP constitutes such a viable alternative. INTRODUCTIONThe product of the malE gene of Escherichia coli, the periplasmic maltose-binding protein (MBP), 1The abbreviations used are: MBPmaltose-binding proteinDTTdithiothreitolPAGEpolyacrylamide gel electrophoresis. binds maltose and other maltodextrins with high affinity (1Kellermann O. Szmelcman S. Eur. J. Biochem. 1974; 47: 139-149Google Scholar). Ligand-bound MBP interacts with a membrane-bound complex of the MalF, MalG, and MalK proteins (2Froshauer S. Beckwith J. J. Biol. Chem. 1984; 259: 10896-10903Google Scholar, 3Dassa E. Hofnung M. EMBO J. 1985; 4: 2287-2293Google Scholar) to take up maltodextrins via a transmembrane channel formed by MalF and MalG (4Shuman H.A. Silhavy T.J. J. Biol. Chem. 1981; 256: 560-562Google Scholar, 5Davidson A.L. Nikaido H. J. Biol. Chem. 1991; 266: 8946-8951Google Scholar, 6Davidson A.L. Nikaido H. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 2360-2364Google Scholar). The NH2- and COOH-terminal domains of MBP appear to bind to MalG and MalF, respectively (7Treptow N.A. Shuman H.A. J. Mol. Biol. 1988; 202: 809-822Google Scholar). Certain mutations in malF and malG allow cells lacking MBP to transport maltodextrins (8Shuman H.A. J. Biol. Chem. 1982; 257: 5455-5461Google Scholar, 9Treptow N.A. Shuman H.A. J. Bacteriol. 1985; 163: 654-660Google Scholar). These mutant MalF-MalG (MalFG) complexes can stimulate ATP hydrolysis by MalK in the absence of MBP (6Davidson A.L. Nikaido H. Proc. Natl. Acad. Sci. U. S. A. 1992; 89: 2360-2364Google Scholar), suggesting that the binding of MBP to the periplasmic face of the MalFG complex normally activates MalK in the cytoplasm.Ligand-bound MBP interacts with the Tar chemotactic signal transducer to initiate an attractant response to maltose (10Hazelbauer G.L. J. Bacteriol. 1975; 122: 206-214Google Scholar, 11Manson M.D. Kossmann M. J. Bacteriol. 1986; 165: 34-40Google Scholar, 12Kossmann M. Wolff C. Manson M.D. J. Bacteriol. 1988; 170: 4516-4521Google Scholar, 13Gardina P. Conway C. Kossmann M. Manson M. J. Bacteriol. 1992; 174: 1528-1536Google Scholar, 14Zhang Y. Conway C. Rosato M. Suh Y. Manson M.D. J. Biol. Chem. 1992; 267: 22813-22820Google Scholar). Tar is proposed to form a homodimer (15Falke J.J. Koshland Jr., D.E. Science. 1987; 237: 1596-1600Google Scholar, 16Milligan D.L. Koshland Jr., D.E. J. Biol. Chem. 1988; 263: 6268-6275Google Scholar). Genetic (12Kossmann M. Wolff C. Manson M.D. J. Bacteriol. 1988; 170: 4516-4521Google Scholar, 13Gardina P. Conway C. Kossmann M. Manson M. J. Bacteriol. 1992; 174: 1528-1536Google Scholar, 17Wolff C. Parkinson J.S. J. Bacteriol. 1988; 170: 4509-4515Google Scholar) and structural (18Milburn M.V. Prive G.G. Milligan D.L. Scott W.G. Yeh J. Janacik J. Koshland Jr., D.E. Kim S.-H. Science. 1991; 254: 1342-1347Google Scholar) evidence indicates that the attractant L-aspartate binds at the dimer interface of the periplasmic domain of Tar, contacting residues in both subunits. Recent data from intergenic complementation studies (53Gardina, P., 1995, Attractant Binding and Signaling in the E. coli Aspartate Receptor. Ph.D. thesis, Texas A&M University.Google Scholar) is consistent with the prediction from computer modeling (19Stoddard B.L. Koshland Jr., D.E. Nature. 1992; 358: 774-776Google Scholar) that MBP also contacts both subunits of Tar.MBP is thought to function as a monomer in which the globular NH2- and COOH-terminal domains are connected by a flexible hinge (20Spurlino J.C. Lu G.-Y. Quiocho F.A. J. Biol. Chem. 1991; 266: 5202-5219Google Scholar, 21Sharff A.J. Rodseth L.E. Spurlino J.C. Quiocho F.A. Biochemistry. 1992; 31: 10657-10663Google Scholar). Short maltodextrins, like maltose and maltotriose, are buried in a cleft between the two domains when MBP closes (20Spurlino J.C. Lu G.-Y. Quiocho F.A. J. Biol. Chem. 1991; 266: 5202-5219Google Scholar). The nonreducing end of longer maltodextrins hangs out of the cleft, and bulky, nonphysiological ligands like cyclomaltoheptaose may prevent closure of the cleft almost completely when they bind to MBP (22Sharff A.J. Rodseth L.E. Quiocho F.A. Biochemistry. 1993; 32: 10553-10559Google Scholar). A bound maltodextrin forms extensive hydrophobic and hydrogen bonds with residues on each wall of the cleft to stabilize the closed conformation. To achieve the closed conformation, one domain of MBP undergoes a 35° rotation and an 8° lateral twist in the hinge relative to the other domain (21Sharff A.J. Rodseth L.E. Spurlino J.C. Quiocho F.A. Biochemistry. 1992; 31: 10657-10663Google Scholar).A working hypothesis is that when MBP assumes its closed conformation, residues in the two domains are brought into the correct spatial relationship to interact with complementary sites on the MalF-MalG complex or with the Tar dimer (14Zhang Y. Conway C. Rosato M. Suh Y. Manson M.D. J. Biol. Chem. 1992; 267: 22813-22820Google Scholar, 19Stoddard B.L. Koshland Jr., D.E. Nature. 1992; 358: 774-776Google Scholar, 20Spurlino J.C. Lu G.-Y. Quiocho F.A. J. Biol. Chem. 1991; 266: 5202-5219Google Scholar). However, since MBP (23Shilton B.H. Mowbray S.L. Protein Sci. 1995; 4: 1346-1355Google Scholar, 24Bohl E. Shuman H.A. Boos W. J. Theor. Biol. 1995; 172: 83-94Google Scholar) and other binding proteins (25Walmsley A.R. Shaw J.G. Kelly D.J. J. Biol. Chem. 1992; 267: 8064-8072Google Scholar, 26Walmsley A.R. Shaw J.G. Kelly D.J. Biochemistry. 1992; 31: 11175-11181Google Scholar, 27Wolf A. Shaw E.W. Nikaido K. Ames G.F.-L. J. Biol. Chem. 1994; 269: 23051-23058Google Scholar, 28Flocco M.M. Mowbray S.L. J. Biol. Chem. 1994; 269: 8931-8936Google Scholar) equilibrate between open and closed forms in the presence or absence of ligands, it is uncertain whether MalFG and Tar first interact with the open, closed, or some intermediate form of MBP.We have introduced mutations into a plasmid-borne malE gene to create the amino acid substitutions G69C and S337C in the normally cysteine-free MBP. Based on the crystal structure of the ligand-bound, closed form of MBP (20Spurlino J.C. Lu G.-Y. Quiocho F.A. J. Biol. Chem. 1991; 266: 5202-5219Google Scholar), 2J. C. Spurlino and F. A. Quiocho, unpublished data. we anticipated that cysteines at these positions could form an interdomain disulfide bridge. We have determined that this cysteine-substituted MBP does form such an intramolecular cross-link. Growth experiments, transport measurements, and chemotaxis assays strongly suggest that this cross-linking occurs in the periplasm in vivo. The double mutant MBP does not function in maltose transport or chemotaxis, and it inhibits, in a dose-dependent fashion, the function of wild-type MBP in transport and chemotaxis. This result demonstrates that the cross-linked protein, which we infer is in the closed conformation, retains biological activity and competes with wild-type MBP for interaction with MalFG and Tar in vivo." @default.
• W2040898023 created "2016-06-24" @default.
• W2040898023 creator A5006384935 @default.
• W2040898023 creator A5024215680 @default.
• W2040898023 creator A5055696213 @default.
• W2040898023 creator A5078325329 @default.
• W2040898023 creator A5081856697 @default.
• W2040898023 date "1996-07-01" @default.
• W2040898023 modified "2023-10-13" @default.
• W2040898023 title "Maltose-binding Protein Containing an Interdomain Disulfide Bridge Confers a Dominant-negative Phenotype for Transport and Chemotaxis" @default.
• W2040898023 cites W1486598550 @default.
• W2040898023 cites W1486683074 @default.
• W2040898023 cites W1490002405 @default.
• W2040898023 cites W1492554871 @default.
• W2040898023 cites W1511766554 @default.
• W2040898023 cites W1513453126 @default.
• W2040898023 cites W1528987874 @default.
• W2040898023 cites W1529506325 @default.
• W2040898023 cites W1553074357 @default.
• W2040898023 cites W1571047021 @default.
• W2040898023 cites W1578737142 @default.
• W2040898023 cites W1581832566 @default.
• W2040898023 cites W1583331377 @default.
• W2040898023 cites W1588507030 @default.
• W2040898023 cites W1594147709 @default.
• W2040898023 cites W1596062696 @default.
• W2040898023 cites W1596933866 @default.
• W2040898023 cites W1630052702 @default.
• W2040898023 cites W1684524094 @default.
• W2040898023 cites W1714769843 @default.
• W2040898023 cites W1938292149 @default.
• W2040898023 cites W1967810228 @default.
• W2040898023 cites W1976862382 @default.
• W2040898023 cites W1977036272 @default.
• W2040898023 cites W1978530172 @default.
• W2040898023 cites W2002288024 @default.
• W2040898023 cites W2010310396 @default.
• W2040898023 cites W2020690016 @default.
• W2040898023 cites W2030503112 @default.
• W2040898023 cites W2034953645 @default.
• W2040898023 cites W2042119968 @default.
• W2040898023 cites W2048783040 @default.
• W2040898023 cites W2068109629 @default.
• W2040898023 cites W2068296787 @default.
• W2040898023 cites W2072788975 @default.
• W2040898023 cites W2082497364 @default.
• W2040898023 cites W2082928276 @default.
• W2040898023 cites W2089972372 @default.
• W2040898023 cites W2095292926 @default.
• W2040898023 cites W2100837269 @default.
• W2040898023 cites W2101755633 @default.
• W2040898023 cites W2115895870 @default.
• W2040898023 cites W2133131542 @default.
• W2040898023 cites W2156783211 @default.
• W2040898023 cites W2161252555 @default.
• W2040898023 cites W2169191525 @default.
• W2040898023 cites W2170299585 @default.
• W2040898023 cites W2888511473 @default.
• W2040898023 cites W4251838592 @default.
• W2040898023 cites W4296765338 @default.
• W2040898023 doi "https://doi.org/10.1074/jbc.271.30.17881" @default.
• W2040898023 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/8663400" @default.
• W2040898023 hasPublicationYear "1996" @default.
• W2040898023 type Work @default.
• W2040898023 sameAs 2040898023 @default.
• W2040898023 citedByCount "19" @default.
• W2040898023 countsByYear W20408980232014 @default.
• W2040898023 countsByYear W20408980232015 @default.
• W2040898023 countsByYear W20408980232016 @default.
• W2040898023 countsByYear W20408980232019 @default.
• W2040898023 countsByYear W20408980232020 @default.
• W2040898023 crossrefType "journal-article" @default.
• W2040898023 hasAuthorship W2040898023A5006384935 @default.
• W2040898023 hasAuthorship W2040898023A5024215680 @default.
• W2040898023 hasAuthorship W2040898023A5055696213 @default.
• W2040898023 hasAuthorship W2040898023A5078325329 @default.
• W2040898023 hasAuthorship W2040898023A5081856697 @default.
• W2040898023 hasBestOaLocation W20408980231 @default.
• W2040898023 hasConcept C100776233 @default.
• W2040898023 hasConcept C104317684 @default.
• W2040898023 hasConcept C105702510 @default.
• W2040898023 hasConcept C123894998 @default.
• W2040898023 hasConcept C12554922 @default.
• W2040898023 hasConcept C127716648 @default.
• W2040898023 hasConcept C170493617 @default.
• W2040898023 hasConcept C181199279 @default.
• W2040898023 hasConcept C185592680 @default.
• W2040898023 hasConcept C27256138 @default.
• W2040898023 hasConcept C2778197599 @default.
• W2040898023 hasConcept C2780676925 @default.
• W2040898023 hasConcept C40767141 @default.
• W2040898023 hasConcept C54166955 @default.
• W2040898023 hasConcept C55493867 @default.
• W2040898023 hasConcept C86803240 @default.
• W2040898023 hasConcept C95444343 @default.
• W2040898023 hasConceptScore W2040898023C100776233 @default.
• W2040898023 hasConceptScore W2040898023C104317684 @default.
• W2040898023 hasConceptScore W2040898023C105702510 @default.

• next