Martin Lindberg

Structure of the IgG-binding regions of streptococcal protein G.

The gene encoding the IgG‐binding protein G from Streptococcus G148 was isolated by molecular cloning. A subclone containing a 1.5‐kb insert gave a functional product in Escherichia coli. Protein analysis of affinity‐purified polypeptides revealed two gene products, both smaller than protein G spontaneously released from streptococci, but with identical IgG‐binding properties. The complete nucleotide sequence of the insert revealed a repeated structure probably evolved through duplications of fragments of different sizes. The deduced amino acid sequence revealed an open reading frame extending throughout the insert, terminating in a TAA stop codon. Analysis of the two gene products by N‐terminal amino acid determination suggests that two different TTG codons are recognized in E. coli for initiation of translation to yield the two products. Based on these results several truncated gene constructions were expressed and analysed. The results suggest that the C‐terminal part of streptococcal protein G consists of three IgG‐binding domains followed by a region which anchors the protein to the cell surface. Structural and functional comparisons with streptococcal M protein and staphylococcal protein A have been made.

Cloning and expression of the gene for a fibronectin-binding protein from Staphylococcus aureus.

The gene encoding the fibronectin‐binding protein (FNBP) from Staphylococcus aureus strain 8325‐4 was isolated from a gene bank in pBR322. The original clone, containing a 6.5‐kb insert, gave a functional product present in the periplasm of Escherichia coli. Analysis of polypeptides isolated after affinity chromatography on fibronectin‐Sepharose followed by ion‐exchange chromatography revealed two gene products, 87 and 165 kd in mol. wt. The amino acid compositions of these two polypeptides and a native FNBP from S. aureus strain Newman were very similar. Antibodies raised against the native FNBP from strain Newman precipitated the 125I‐labelled 165‐kd polypeptide, and unlabeled 165‐ and 87‐kd polypeptides as well as native FNBP inhibited the immunoprecipitation reactions. The region of the fnbp‐gene encoding the fibronectin‐binding activity has been identified and subcloned in an expression vector based on the staphylococcal protein A gene. The resulting product in E. coli is an extracellular fusion protein consisting of two IgG‐binding domains of protein A followed by a fibronectin‐binding region. The fusion protein binds to fibronectin and completely inhibits the binding of fibronectin to intact cells of S. aureus.

A second IgG-binding protein in Staphylococcus aureus

Most strains of Staphylococcus aureus express IgG-binding activity and this binding has been considered to be solely mediated by protein A. However, the existence of a second gene in S. aureus strain 8325-4 encoding an IgG-binding polypeptide was recently reported. This novel IgG-binding polypeptide was found after panning a shotgun phage display library, made from chromosomal DNA, against immobilized human IgG. The complete gene (sbi) encoding this novel IgG-binding protein, denoted protein Sbi, has now been cloned and sequenced. Analysis of other S. aureus strains showed that this gene is not unique for strain 8325-4. The protein consists of 436 amino acids and exhibits an immunoglobulin-binding specificity similar to protein A. Furthermore, it is shown that Sbi is highly expressed in strain Newman 4, which shows that IgG-binding activity in S. aureus can be mediated by proteins other than protein A.

Gene fusion vectors based on the gene for staphylococcal protein A

Two plasmid vectors, containing the gene coding for staphylococcal protein A and adapted for gene fusion, have been constructed. These vectors will allow fusion of any gene to the protein A gene, thus giving hybrid proteins which can be purified, in a one-step procedure, by IgG affinity chromatography. As an example of the practical use of such vectors, the protein A gene has been fused to the lacZ gene of Escherichia coli. E. coli strains containing such plasmids produce hybrid proteins with both IgG binding and beta-galactosidase activities. The hybrid protein(s) can be immobilized on IgG-Sepharose by its protein A moiety with high efficiency without losing its enzymatic activity and they can be eluted from the column by competitive elution with pure protein A. The fused protein(s) also binds to IgG-coated microtiter wells which means that the in vivo product can be used as an enzyme conjugate in ELISA tests.

Structural and putative regulatory genes involved in cellulose synthesis in Rhizobium leguminosarum bv. trifolii.

Six genes involved in cellulose synthesis in Rhizobium leguminosarum bv. trifolii were identified using Tn5 mutagenesis. Four of them displayed homology to the previously cloned and sequenced Agrobacterium tumefaciens cellulose genes celA, celB, celC and celE. These genes are organized similarly in R. leguminosarum bv. trifolii. In addition, there were strong indications that two tandemly located genes, celR1 and celR2, probably organized as one operon, are involved in the regulation of cellulose synthesis. The deduced amino acid sequences of these genes displayed a high degree of similarity to the Caulobacter crescentus DivK and PleD proteins that belong to the family of two-component response regulators. This is to our knowledge the first report of genes involved in the regulation of cellulose synthesis. Results from attachment assays and electron microscopic studies indicated that cellulose synthesis in R. leguminosarum bv. trifolii is induced upon close contact with plant roots during the attachment process.

A unipolarly located, cell-surface-associated agglutinin, RapA, belongs to a family of Rhizobium-adhering proteins (Rap) in Rhizobium leguminosarum bv. trifolii.

The phage-display cloning technique was used to find rhizobial proteins that bind to receptors located on the bacterial cell surface. The aim was to clone the gene(s) encoding rhicadhesin, a universal rhizobial adhesion protein, and/or other cell-surface-binding proteins. Four such Rhizobium-adhering proteins (Rap) were revealed in Rhizobium leguminosarum bv. trifolii strain R200. The binding is mediated by homologous Ra domains in these proteins. One member of the Rap protein family, named RapA1, is a secreted calcium-binding protein, which are also properties expected for rhicadhesin. However, the size of the protein (24 kDa instead of 14 kDa) and its distribution among different rhizobia (present in only Rhizobium leguminosarum biovars and R. etli instead of all members of Rhizobiaceae argue against RapA1 being rhicadhesin. Protein RapA1 consists of two homologous Ra domains and agglutinates R200 cells by binding to specific receptors located at one cell pole during exponential growth. Expression of these cell-surface receptors was detected only in rhizobia that produce the RapA proteins. The authors propose that the homologous Ra domains, found to be present also in other proteins with different structure, represent lectin domains, which confer upon these proteins the ability to recognize their cognate carbohydrate structures.

DNA homology between the arsenate resistance plasmid pSX267 from Staphylococcus xylosus and the penicillinase plasmid pI258 from Staphylococcus aureus.

A 29.5-kb plasmid, pSX267, from Staphylococcus xylosus DSM 20267 was found to code for arsenate, arsenite, and antimony (III) resistance. The isolated plasmid was transformed into S. aureus, where the same resistances were expressed. It was of special interest to see whether pSX267 showed any DNA sequence homology with the well-studied penicillinase plasmid from S. aureus pI258, also conferring arsenate, arsenite, and antimony III resistance. By the use of the Southern blotting technique, it was found that DNA sequence homology exists in the region of arsenate, arsenite, and antimony resistance, in addition to the region where the origin of replication, the incompatibility, and the replication A function were mapped on pI258. This finding was confirmed by electron microscope heteroduplex analysis, which allowed a correlation between the genetic and physical maps of pI258 and pSX267. Duplex DNA was formed at the arsenate operon of pI258, with a length of 2.6 kb, and at the incompatibility and replication A region, comprising a length of 2.5 kb. Adjacent to this latter region, two small regions of DNA homology were present, with lengths of 0.2 and 0.27 kb. Both plasmids share approximately 20% DNA sequence homology. The DNA homology of the arsenate, arsenite, and antimony III resistance coding regions between pI258 and pSX267 indicate that these plasmid-determined resistance markers are highly conserved and distributed among different staphylococcal species.

Staphylococcus aureus expresses a cell surface protein that binds both IgG and β2-glycoprotein I

The existence of a second IgG-binding protein, protein Sbi, in Staphylococcus aureus has been reported previously. Later data indicated that protein Sbi also bound another serum component. This component has now been affinity-purified on immobilized protein Sbi and identified as beta2-glycoprotein I (beta2-GPI), also known as apolipoprotein H. The minimal beta2-GPI-binding domain was identified by shotgun phage display and the binding was shown to be mediated by a region of 57 amino acids, clearly separated from the IgG-binding domain. It is also shown that protein Sbi, and thus the beta2-GPI-binding activity, is expressed on the staphylococcal cell surface at levels varying between strains.

M-like proteins of Streptococcus dysgalactiae

ABSTRACT Streptococcus dysgalactiae is one of the most important bacterial species isolated from bovine mastitis. To identify potential virulence factors of this species we prepared chromosomal DNA from strain 8215 and constructed a phage display library. By affinity selection of the library against fibrinogen (Fg), we isolated and characterized a gene, called demA, encoding a protein with the molecular mass of ∼58 kDa, called DemA, displaying both plasma protein binding properties and sequence similarities with the M and M-like proteins of other streptococcal species. Purified recombinant DemA protein was found to completely inhibit Fg-binding to cells ofS. dysgalactiae. A continued sequence analysis revealed that the demA gene was preceded by an open reading frame (dmgA) coding for a putative protein, called DmgA, with high similarities to the Mga proteins of Streptococcus pyogenes. By additional cloning, the correspondingdmgA and demA genes from another strain, called Epi9, were isolated and analyzed. These genes, called dmgBand demB, respectively, revealed a high degree of similarity to the corresponding genes in strain 8215. Increased binding of Fg by cells of strain Epi9, grown in an atmosphere with 10% CO2, was correlated to an enhanced transcription of thedemB gene as shown in a Northern blot. Strain 8215 did not respond to CO2, which could be explained by a nonfunctionaldmgA gene due to insertion of an insertion sequence element. Based on sequence similarities of the described proteins to Mga, M, and M-like proteins and the response to elevated level of CO2, we suggest that the dmg anddem genes are members of a regulon similar to the describedmga regulon in S. pyogenes, which encodes several virulence factors in this species.

A Monoclonal Antibody Enhances Ligand Binding of Fibronectin MSCRAMM (Adhesin) from Streptococcus dysgalactiae

A monoclonal antibody 3A10, generated from a mouse immunized with the Streptococcus dysgalactiae fibronectin (Fn) binding protein FnbA, was isolated, and its effect on ligand binding by the antigen was examined. The epitope for 3A10 was localized to a previously unidentified Fn binding motif (designated Au) just N-terminal of the repeat domain which represents the primary ligand binding site on FnbA. Fn binding to Au was enhanced by 3A10 rather than inhibited. This effect was demonstrated in two different assays. First, in the presence of 3A10 the Au-containing proteins and synthetic peptide more effectively competed with bacterial cells for binding to Fn. Second, 3A10 dramatically increased the binding of biotin-labeled forms of the Au-containing proteins to Fn immobilized on a blotting membrane. Pure 3A10 IgG did not recognize the antigen by itself, and Fn was required for the immunological interaction between the antibody and the epitope. This induction effect of Fn was shown in both Western blot and enzyme-linked immunosorbent assay in which immobilized Au-containing molecules were probed with 3A10 in the presence of varying concentrations of Fn. Specificity analyses of 3A10 revealed that the monoclonal also recognized a ligand binding motif in a Streptococcus pyogenes Fn binding MSCRAMM but not the corresponding motifs in two related adhesins from Staphylococcus aureus and S. dysgalactiae. Furthermore, 3A10 stimulated Fn binding by S. pyogenes cells. These results together with subsequent biophysical studies presented in the accompanying paper (House-Pomepeo, K., Xu, Y., Joh, D., Speziale, P., and Höök, M.(1996) J. Biol. Chem. 271, 1379-1384) indicate that the ligand binding sites of Fn binding MSCRAMMs have little or no secondary structure. However, on binding to Fn, they appear to undergo a structural rearrangement resulting in a defined structure rich in β sheet and expressing a ligand-induced binding site for antibodies such as 3A10.