No abstract

No abstract

The interaction of the hydrogenase maturation endopeptidase HycI with its substrate, the precursor of the large subunit, was studied. Replacement of conserved amino-acid residues in HycI, which have been shown to bind a cadmium ion from the crystallization buffer in crystals of HybD (endopeptidase for hydrogenase 2), abolished or strongly reduced processing activity. Atomic absorption spectroscopy of purified HycI and HybD proteins showed the absence of nickel. In vitro processing assays showed that the reaction requires nickel to be bound to the precursor and the protease does not have a function in nickel delivery to the substrate. Radioactive labelling of cells with 63 Ni, devoid of endopeptidase, resolved several forms of the precursor which are possibly intermediates in the maturation pathway. It is concluded that the endopeptidase uses the metal in the large subunit of [NiFe]-hydrogenases as a recognition motif.Keywords: endopeptidase; hydrogenase maturation; nickel; processing.Genetic analysis of the maturation of active [NiFe]-hydrogenases revealed that the activities of at least seven gene products are involved in the process. These are the products of the so-called hyp genes [1,2], which include a specific chaperone (HypC) [3], a proposed nickel-donating GTPase (HypB) [4,5] and an endopeptidase that proteolytically removes a C-terminal extension from the precursor form of the large hydrogenase subunit (reviewed in [6]). The endopeptidases specific for hydrogenases 3 (HycI) and 2 (HybD) have been purified and an in vitro processing system has been established for the maturation of hydrogenase 3 [7,8].Major progress in the understanding how the endopeptidase recognizes its substrate came from the determination of the crystal structure of HybD [9]. It crystallizes as a complex with a cadmium ion from the crystallization buffer; the metal is pentacoordinated by three amino-acid side chains (Glu16, Asp62 and His93) and a water molecule in a pseudo-tetragonal arrangement. We proposed that the cadmium site marks the nickel binding site by which the protease interacts with the nickel-containing precursor of the large subunit. A model has been postulated which implies that iron and nickel are ligated separately to the apoprotein and are assembled into the complete heterobinuclear metal center after cleavage of the C-terminus [6]. Alternatively, the cadmium (i.e. nickel)-binding site may indicate that the endopeptidase is a nickel-dependent metalloprotease or that it is involved in metal transfer and incorporation into the hydrogenase large subunit. The observations that the precursor of the large subunit which accumulates in a protease HycI-deficient strain contains nickel and that cleavage appears to require prior nickel incorporation [8] argue against those functions but do not rule them out completely. The reason is that both in vivo and in vitro the precursor of the large subunit occurs in multiple folding states [3] and cannot be converted stoichiometrically into the mature form. In this communication, ...

Creatine and its phosphorylated form play a central role in the energy metabolism of muscle and nerve tissues. l-Arginine:glycine amidinotransferase (AT) catalyses the committed step in the formation of creatine. The mitochondrial and cytosolic forms of the enzyme are believed to derive from the same gene by alternative splicing. We have expressed recombinant human AT in Escherichia coli with two different N-termini, resembling the longest two forms of the enzyme that we had isolated recently from porcine kidney mitochondria as a mixture. The enzymes were expressed with N-terminal histidine tags followed by factor Xa-cleavage sites. We established a new method for the removal of N-terminal fusion peptides by means of an immobilized snake venom prothrombin activator. We identified cysteine-407 as the active-site residue of AT by radioactive labelling and isolation of labelled peptides, and by site-directed mutagenesis of the protein.

Inosamine-phosphate amidinotransferases catalyze two nonconsecutive transamidination reactions in the biosynthesis of the streptomycin family of antibiotics. L-Arginine:inosamine-phosphate amidinotransferase StrB1 from Streptomyces griseus (StrB1) was cloned as an N-terminal hexa-histidine fusion protein, purified by affinity chromatography, and crystallized, and its crystal structure was solved by Patterson search methods at 3.1 A resolution. The structure is composed of five betabeta alphabeta-modules which are arranged circularly into a pseudo-5-fold symmetric particle. The three-dimensional structure is closely related to the structure of human L-arginine:glycine amidinotransferase (AT), but five loops (the 40-, 170-, 220-, 250-, and 270-loop) are organized very differently. The major changes are found in loops around the active site which open the narrow active site channel of AT to form an open and solvent-exposed cavity. In particular, module II of StrB1 is AT-like but lacks a 10-residue alpha-helix in the 170-loop. The concomitant reorganization of neighboring surface loops that surround the active site, i.e., the 40-loop and the 270-loop, results in an arrangement of loops which allows an unrestricted access of substrates to the cavity. However, the residues which are involved in substrate binding and catalysis are conserved in AT and StrB1 and are at equivalent topological positions, suggesting a similar reaction mechanism among amidinotransferases. The binding site for L-arginine had been deduced from its complex with AT. Molecular modeling revealed a possible binding mode for the second substrate scyllo-inosamine 4-phosphate.