Purification of the coenzyme B12-dependent isobutyryl-CoA mutase (ICM) from Streptomyces cinnamonensis gave a protein of ∼65 kDa by SDS-polyacrylamide gel electrophoresis, whose gene icmAwas cloned using sequences derived from tryptic peptide fragments. The gene encodes a protein of 566 residues (62,487 Da), with 43–44% sequence identity to the large subunit of methylmalonyl-CoA mutase (MCM) from S. cinnamonensis and Propionibacterium shermanii. Targeted disruption of the icmA gene yielded an S. cinnamonensis mutant devoid of ICM activity. The IcmA protein is ∼160 residues shorter than the large subunit of the bacterial MCMs, corresponding to a loss of the entire C-terminal coenzyme B12 binding domain. The sequence of the (β/α)8-barrel comprising residues A1–A400 in P. shermanii MCM is highly conserved in IcmA. The protein was produced in Streptomyces lividans and Escherichia coli with an N-terminal His6 tag (His6-IcmA), but after purification His6-IcmA showed no ICM activity. In the presence of coenzyme B12, protein from S. lividans and S. cinnamonensisof ∼17 kDa by SDS-polyacrylamide gel electrophoresis could be selectively eluted with His6-IcmA from a Ni2+affinity column. After purification, this small subunit showed no ICM activity but gave active enzyme when recombined with coenzyme B12 and IcmA or His6-IcmA.

The extracellular interferon γ receptor α-chain comprises two immunoglobulin-like domains, each with fibronectin type-III topology, which are responsible for binding interferon γ at the cell surface. The epitopes on the human receptor recognized by three neutralizing antibodies, A6, γR38 and γR99, have been mapped by homolog scanning mutagenesis. In this way, a loop connecting β-strands C and C′ in the N-terminal domain was identified as a key component of the epitopes bound by A6 and γR38, whereas γR99 binds to the C-terminal domain in a region including strands A and B and part of the large C′E loop. The epitope for A6 was confirmed in a crystal structure of a complex between a recombinant N-terminal receptor domain and the Fab fragment from A6, determined by X-ray diffraction to 2.8 Å resolution. The antibody-antigen interface buries 1662 Å2 of protein surface, including 22 antibody residues from five complementarity determining regions, primarily through interactions with the CC′ surface loop of the receptor. The floor of the antigen binding cavity is formed mainly by residues from CDR L3 and CDR H3 while a surrounding ridge is formed by residues from all other CDRs except L2. Many potential polar interactions, as well as 13 aromatic side-chains, four in VL, six in VH and three in the receptor, are situated at the interface. The surface of the receptor contacted by A6 overlaps to a large extent with that contacted by interferon-γ, in the ligand-receptor complex. However, the conformation of this epitope is very different in the two complexes, demonstrating that conformational mobility in a surface loop on this cytokine receptor permits steric and electrostatic complementarity to two quite differently shaped binding sites.

The catabolism of branched chain amino acids, especially valine, appears to play an important role in furnishing building blocks for macrolide antibiotic biosynthesis. To determine for the first time the importance of valine dehydrogenase (vdh) in polyether antibiotic biosynthesis, the vdh gene from Streptomyces cinnamonensis has been cloned and sequenced. The enzyme (Mr 37 718 Da) has been produced in large amounts in an active form in the E. coli cytoplasm using a T7 RNA-polymerase expression system. Upon inactivation of the gene in S. cinnamonensis by a double-crossover mechanism, a hyg::vdh mutant was isolated that was devoid of vdh activity. Upon growth in chemically defined media, as well as a complex medium optimised for monensin production, the mutant and wild-type grew equally well and reached the same levels of monensin production. In both strains a valine transaminase activity could be detected that provides an alternative route for converting valine into 2-oxoisovaleric acid. The results show that vdh is not essential for normal growth of S. cinnamonensis, and its inactivation does not significantly affect normal levels of monensin production in this strain.

Three single-chain antibody fragments that recognize the extracellular human interferon gamma receptor alpha-chain (IFN gamma R), and inhibit the binding of human IFN gamma, have been produced in Escherichia coli. These fragments are derived from murine anti-receptor monoclonal antibodies, and comprise the variable heavy (VH) domain linked to the variable light (VL) chain through a 15 amino acid linker [(GGGGS)3]. Using surface plasmon resonance technology (BIAcore), the soluble proteins were shown to retain a high affinity for recombinant IFN gamma R, and by radioimmunoassay to possess a high inhibitory activity towards IFN gamma-binding to human Raji cells. The antibody fragments most likely recognize epitopes that overlap the cytokine binding site on the receptor surface. Attempts to dissect further the antibodies to isolated VH- and VL-chains and to synthetic linear and cyclic peptides derived from the individual complementarity determining regions failed to afford fragments with significant IFN gamma R binding affinity. Nevertheless, these native-like variable region fragments and petidomimetics derived from them are of interest in the design of novel IFN gamma R antagonists.

The extracellular interferon gamma receptor alpha-chain (IFN gamma R) is believed to comprise two discrete approximately 110 amino acid immunoglobulin-like domains, perhaps similar to those seen in the crystal structure of the extracellular human growth hormone receptor [De Vos, A. M., Ultsch, M., & Kossiakoff, A. (1992) Science 255, 306-312], a distant relative in the cytokine receptor superfamily. In accord with this idea, we show that these IFN gamma R immunoglobulin-like domains can be produced separately in a soluble form with a native-like fold. The N-terminal domain (residues 1-108), with a Cys105 to Ser105 mutation, was produced at a high level, in a soluble form, as a thioredoxin-interferon gamma receptor fragment fusion protein in the cytoplasm of Escherichia coli. Upon extraction, the receptor Cys60-Cys68 disulfide bond formed spontaneously, to generate a native-like structure directly without the need for refolding. Cleavage of the fusion protein by enterokinase released the receptor fragment (approximately 12 kDa), which was recognized by several neutralizing antibodies with affinities, measured using surface plasmon resonance technology, that were essentially indistinguishable from those seen with the full length extracellular IFN gamma R produced in eukaryotic cells. Circular dichroism and 1D 1H nuclear magnetic resonance spectra indicated that the receptor fragment adopts a folded state, with mainly beta-sheet and reverse turn secondary structure. The second membrane-proximal Ig-like domain of the IFN gamma R (residues 90-229) was produced, albeit less efficiently, and characterized in a similar way. The production of these two independently folded proteins provides experimental support for the two domain organization of the IFN gamma R and opens new avenues for structural studies on these Ig-like molecules by NMR and crystallographic methods.

The extracellular portion of the human interferon-gamma receptor (IFN-gamma R) is predicted to adopt two Ig-like domains each with Ig superfamily type C2 topology. Surface plasmon resonance technology has been used here to compare the equilibrium and kinetic constants for binding reactions between these and several mutant domains, and neutralising anti-IFN-gamma R monoclonal antibodies. The biosensor measurements provide a sensitive method for monitoring the effects of mutations on the functional epitopes recognized by the neutralising antibodies. Thus, using recombinant native-like proteins made in E. coli, the ten N-terminal residues of the receptor were found not to be essential for domain folding, nor for recognition by the antibodies A6, D2 and gamma R38. In a similar way, residues in the interdomain region were found to play an important functional role in the epitope recognized by the antibody gamma R99.

Murine monoclonal antibodies (mAbs) are described that recognize the extracellular human interferon gamma receptor alpha-chain (IFN gamma R) and inhibit the binding to it of interferon gamma. The inhibitory activities (IC50s) of these mAbs, quantified by radioimmunoassay using native receptor on human Raji cells, lie in the range 0.5-24 nM, whereas their relative affinities for the immobilised recombinant extracellular receptor, determined using surface plasmon resonance technology, are in the range 0.6-40.9 nM. Nine mAbs derived from one immunization, were shown by variable region cDNA sequencing to be clonally related, with mAb A6 from this group showing the highest affinity for the receptor. Another two mAbs, gamma R38 and gamma R99, derived from a separate immunization, are clonally unrelated to each other and to those in the A6 family. From the V-region sequences, the L-chains of mAbs A6, gamma R38 and gamma R99 were shown to belong to the V kappa 34C, V kappa 34C and V kappa 1 families, whereas the H-chains belong to the 3069, J606 and J558 families, respectively. The mAbs A6 and gamma R38 recognize overlapping epitopes on the N-terminal Ig-like domain of the IFN gamma R, whereas the gamma R99 epitope is located largely in the membrane proximal Ig-like domain. Sequence comparisons with Ig structures solved by X-ray diffraction allowed deductions concerning likely CDR canonical conformations. These studies provide essential information for crystallographic and mutagenesis experiments aimed at understanding the molecular basis of the interactions of these mAbs with the extracellular IFN gamma R.

In streptomycetes, the conversion of succinyl-coenzyme A (CoA) into methylmalonyl-CoA, catalyzed by methylmalonyl-CoA mutase, most likely represents an important source of building blocks for polyketide antibiotic biosynthesis. In this work, the structural gene for methylmalonyl-CoA mutase from Streptomyces cinnamonensis was cloned by using a heterologous gene probe encoding the mutase from Propionibacterium shermanii. A 5,732-bp fragment was sequenced, within which four open reading frames were identified on one DNA strand. The two largest (mutA and mutB) overlap by 1 nucleotide and encode proteins of 616 and 733 residues showing high amino acid sequence similarities to each other and to methylmalonyl-CoA mutases from P. shermanii and mammalian sources. The transcriptional start of the mutA-mutB message, determined by S1 mapping, coincides with the first nucleotide of the translational start codon. Evidence that these two open reading frames encode a functional mutase in S. cinnamonensis was obtained by subcloning and expression in Streptomyces lividans TK64. The mutA and mutB gene products were detected in Western blots (immunoblots) with mutase-specific antibodies and by direct detection of mutase activity with a newly developed assay method. The methylmalonyl-CoA mutase was unable to catalyze the conversion of isobutyryl-CoA into n-butyryl-CoA, another closely related adenosylcobalamin-dependent rearrangement known to occur in S. cinnamonensis.

The Streptomyces glaucescens genome frequently undergoes gross genomic rearrangement events which result in the deletion of extremely large segments of chromosomal DNA. The structure and origin of the DNA forming the novel junctions arising from five of these deletion events are described. Only one junction proved to be the result of a relatively simple event; the remainder were more complex, with one involving DNA which originated from at least five distinct loci. In three of the investigated cases, DNA sequences present in the junctions appeared to have resulted from the duplication of previously unique sequences, suggesting that duplication of chromosomal segments may be an important factor in genetic instability. The nucleotide sequences surrounding these junctions and their respective wild-type termini were determined.

Genetic instability inStreptomyces glaucescens characteristically involves the occurrence of gross genomic rearrangements including high-level sequence amplification and extensive deletion. We investigated the relationship of the unstablemelC andstrS loci and a 100 kb region of the chromosome which frequently gives rise to intense heterogeneous DNA amplification. Standard chromosome walking using a cosmid bank in conjunction with a “reverse-blot” procedure enabled us to construct a contiguous genomicBamHI map of the unstable region exceeding 900 kb. The unstable genes and the amplifiable region (AUD locus) are physically linked within a 600 kb segment of the chromosome. The previously characterized deletions which affect these loci are merely components of much larger deletions ranging from 270 to over 800 kb which are polar in nature, effecting the sequential loss of thestrS andmelC loci. The more extensive deletions terminate either adjacent to, or in the vicinity of DNA reiterations at the AUD locus. Additionally, a deletion junction fragment and the corresponding deletion ends were cloned and analysed at the sequence level.

Certain chromosomal markers inStreptomyces glaucescens behave unstably, being lost at high frequency as a result of extensive genomic deletion. Additionally, mutant strains possessing such deletions frequently display intense DNA amplification. With the help of a wild-type cosmid library we investigated the structure of the amplified DNA sequences (ADS) and the corresponding wild-type amplifiable units of DNA (AUD). The reiterations were heterogeneous in location, copy number and sequences involved and originated predominantly from a single 100 kb region of the chromosome called the AUD locus. All strains bearing reiterations possessed associated deletions which terminated either close to or at the ADS. The termini of four AUDs were sequenced in order to gain more knowledge about these heterogeneous amplifications. In three of the four cases investigated small, interrupted homologies were found bordering the AUDs. With the help of orthogonal-field-alternation gel electrophoresis (OFAGE) we were able to visualize a tandem reiteration of more than 1500 kb in length.