A simple and reliable approach to docking protein-protein complexes from very sparse NOE-derived intermolecular distance restraints.

A simple and reliable approach for docking protein-protein complexes from very sparse NOE-derived intermolecular distance restraints (as few as three from a single point) in combination with a novel representation for an attractive potential between mapped interaction surfaces is described. Unambiguous assignments of very sparse intermolecular NOEs are obtained using a reverse labeling strategy in which one the components is fully deuterated with the exception of selective protonation of the delta-methyl groups of isoleucine, while the other component is uniformly (13)C-labeled. This labeling strategy can be readily extended to selective protonation of Ala, Leu, Val or Met. The attractive potential is described by a 'reduced' radius of gyration potential applied specifically to a subset of interfacial residues (those with an accessible surface area > or = 50% in the free proteins) that have been delineated by chemical shift perturbation. Docking is achieved by rigid body minimization on the basis of a target function comprising the sparse NOE distance restraints, a van der Waals repulsion potential and the 'reduced' radius of gyration potential. The method is demonstrated for two protein-protein complexes (EIN-HPr and IIA(Glc)-HPr) from the bacterial phosphotransferase system. In both cases, starting from 100 different random orientations of the X-ray structures of the free proteins, 100% convergence is achieved to a single cluster (with near identical atomic positions) with an overall backbone accuracy of approximately 2 A. The approach described is not limited to NMR, since interfaces can also be mapped by alanine scanning mutagenesis, and sparse intermolecular distance restraints can be derived from double cycle mutagenesis, cross-linking combined with mass spectrometry, or fluorescence energy transfer.

Isolation, characterization, and cDNA-derived amino acid sequence of glycocyamine kinase from the tropical marine worm Namalycastis sp.

We isolated cytoplasmic glycocyamine kinase (GK) and creatine kinase (CK) from the tropical marine worm Namalycastis sp. by ammonium sulfate fractionation, gel filtration on Sephacryl S-200, and DEAE-5PW chromatography. Sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) showed that the isolated GK is highly purified and appears to be a heterodimer of two distinct subunits, alpha and beta, with molecular masses of approximately 40 kDa. The complete nucleotide sequences of the cDNAs for Namalycastis GKalpha and GKbeta were 1527 (encoding 374 amino acids) and 1579 bp (encoding 390 amino acids), respectively. The predicted amino acid sequences differ only in the N-terminal 50 residues. This is consistent with the characteristics of Neanthes GKalpha and GKbeta chains, which we have previously shown to be generated by alternative splicing. The recombinant enzymes GKalpha, GKbeta, and CK from Namalycastis were successfully expressed in Escherichia coli as maltose-binding protein fusion proteins. In contrast to the stable GKbeta enzyme, GKalpha was quite unstable, and its activity decreased remarkably with time. Thus, the N-terminal 50 residues appear to play a key role in enzyme stability. The kinetic parameters for the native GK heterodimer were similar to GKbeta, suggesting that GKalpha would have an activity similar to GKbeta if part of a heterodimer. This is the first report of precise kinetic parameters for GK. Finally, based on our results, we present a model for pluriphosphagen function in Namalycastis wherein cytoplasmic GK and CK and mitochondrial CK function together with phosphocreatine and phosphoglycocyamine to enable cells to respond quickly to a sudden large energy requirement.

Alternative splicing produces transcripts coding for alpha and beta chains of a hetero-dimeric phosphagen kinase.

Glycocyamine kinase (GK) catalyzes the reversible phosphorylation of glycocyamine (guanidinoacetate), a reaction central to cellular energy homeostasis in certain animals. GK is a member of the phosphagen kinase enzyme family and appears to have evolved from creatine kinase (CK) early in the evolution of multi-cellular animals. Prior work has shown that GK from the polychaete Neanthes (Nereis) diversicolor exits as a hetero-dimer in vivo and that the two polypeptide chains (termed alpha and beta) are coded for by unique transcripts. In the present study, we demonstrate that the GK from a congener Nereis virens is also hetero-dimeric and is coded for by alpha and beta transcripts, which are virtually identical to the corresponding forms in N. diversicolor. The GK gene from N. diversicolor was amplified by PCR. Sequencing of the PCR products showed that the alpha and beta chains are the result of alternative splicing of the GK primary mRNA transcript. These results also strongly suggest that this gene underwent an early tandem exon duplication event. Full-length cDNAs for N. virens GKalpha and GKbeta were individually ligated into expression vectors and the resulting constructs used to transform Escherichia coli expression hosts. Regardless of expression conditions, minimal GK activity was observed in both GKalpha and GKbeta constructs. Inclusion bodies for both were harvested, unfolded in urea and alpha chains, beta chains and mixtures of alpha and beta chains were refolded by sequential dialysis. Only modest amounts of GK activity were observed when alpha and beta were refolded individually. In contrast, when refolded the alpha and beta mixture yielded highly active hetero-dimers, as validated by size exclusion chromatography, electrophoresis and mass spectrometry, with a specific activity comparable to that of natural GK. The above evidence suggests that there is a preference for hetero-dimer formation in the GKs from these two polychaetes. The evolution of the alternate splicing and an additional exon in these GKs, producing alpha and beta transcripts, can be viewed as a possible compensation for a mutation(s) in the original gene, which most likely coded for a homo-dimeric protein.

A dimeric creatine kinase from a sponge: implications in terms of phosphagen kinase evolution.

This study demonstrates conclusively that tissues of the sponge Tethya aurantia contain significant creatine kinase (CK) activity. This CK was purified and analyzed with respect to a number of physico-chemical properties. Size exclusion chromatography and denaturing gel electrophoresis analyses showed that this enzyme is dimeric. The sequences of several Lys-C endoproteinase peptides from Tethya CK are consistent with this enzyme being a member of the phosphagen kinase family and a true CK. CK in higher organisms exists in a variety of quaternary structure forms--dimer, octamer and large monomer consisting of a three contiguous CK domains. The present results indicate that CK evolved very early in metazoan evolution and that the dimeric structure preceded other subunit association forms.

Evolution of phosphagen kinase VII. Isolation of glycocyamine kinase from the polychaete Neanthes diversicolor and the cDNA-derived amino acid sequences of alpha and beta chains.

Glycocyamine kinase (GK) was isolated from the marine polychaete Neanthes diversicolor by gel filtration, DEAE-cellulose chromatography, butyl-Toyopearl hydrophobic chromatography, and chromatofocusing. The GK was eluted as a single peak on the latter three chromatographies, and the molecular mass for the native GK was estimated to be about 80 kDa. The SDS-PAGE showed that the isolated GK consists of two distinct subunits in equal proportion, alpha and beta chains, with molecular masses of 42.2 and 43.8 kDa, respectively. The present results suggest that the Neanthes GK has a heterodimeric structure. The cDNAs for alpha and beta chains of Neanthes GK were amplified by PCR and their cDNA-derived amino acid sequences were determined. The alpha and beta chains are composed of 374 and 390 amino acids, and the molecular masses were calculated to be 42,392 and 43,966 Da, respectively, in good agreement with the apparent masses on SDS PAGE. The beta chain has a characteristic N-terminal extension of 15 amino acids, and all of the sequence differences between alpha and beta chains were restricted in the N-terminal region of 50 residues. The overall sequence identity was 92%. The occurrence of heterodimeric nature in Neanthes GK is of great interest from the evolutionary point of view, because the heterodimeric structure is only known for creatine kinase MB-isozyme specific for mammalian heart muscle among phosphagen kinases.

Identification by NMR of the binding surface for the histidine-containing phosphocarrier protein HPr on the N-terminal domain of enzyme I of the Escherichia coli phosphotransferase system.

The interaction between the approximately 30 kDa N-terminal domain of enzyme I (EIN) and the approximately 9.5 kDa histidine-containing phosphocarrier protein HPr of the Escherichia coli phosphoenolpyruvate:sugar phosphotransferase system has been investigated by heteronuclear magnetic resonance spectroscopy. The complex is in fast exchange, permitting us to follow the chemical shift changes of the backbone NH and 15N resonances of EIN upon complex formation by recording a series of 1H-15N correlation spectra of uniformly 15N-labeled EIN in the presence of increasing amounts of HPr at natural isotopic abundance. The equilibrium association constant derived from analysis of the titration data is approximately 1.5 x 10(5) M(-1), and the lower limit for the dissociation rate constant is 1100 s(-1). By mapping the backbone chemical shift perturbations on the three-dimensional solution structure of EIN [Garrett, D. S., Seok, Y.-J., Liao, D.-I., Peterkofsky, A., Gronenborn, A. M., & Clore, G. M. (1997) Biochemistry 36, 2517-2530], we have identified the binding surface of EIN in contact with HPr. This surface is primarily located in the alpha domain and involves helices H1, H2, and H4, as well as the hinge region connecting helices H2 and H2'. The data also indicate that the active site His 15 of HPr must approach the active site His 189 of EIN along the shallow depression at the interface of the alpha and alpha/beta domains. Interestingly, both the backbone and side chain resonances (assigned from a long-range 1H-15N correlation spectrum) of His 189, which is located at the N-terminus of helix H6 in he alpha/beta domain, are only minimally perturbed upon complexation, indicating that His 189 (in the absence of phosphorylation) does not undergo any significant conformational change or change in pK(a) value upon HPr binding. On the basis of results of this study, as well as a previous study which delineated the interaction surface for EI on HPr [van Nuland, N. A. J., Boelens, R., Scheek, R. M., & Robillard, G. T. (1995) J. Mol. Biol. 246, 180-193], a model for the EIN/HPr complex is proposed in which helix 1 (residues 16-27) and the helical loop (residues 49-53) of HPr slip between the two pairs of helices constituting the alpha domain of EIN. In addition, we suggest a functional role for the kink between helices H2 and H2' of EIN, providing a flexible joint for this interaction to take place.

Evolution of phosphagen kinase. VI. Isolation, characterization and cDNA-derived amino acid sequence of lombricine kinase from the earthworm Eisenia foetida, and identification of a possible candidate for the guanidine substrate recognition site.

Lombricine kinase (LK) from the body wall muscle of the earthworm Eisenia foetida was purified to homogeneity. The enzyme was shown to be a dimer consisting of 40 kDa subunits. The cDNA-derived amino acid sequence of 370 residues of Eisenia LK was determined. The validity of the sequence was supported by chemical sequencing of internal tryptic peptides. This is the first reported lombricine kinase amino acid sequence. Alignment of Eisenia LK with those of creatine kinases (CKs), arginine kinases (AKs) and glycocyamine kinase (GK) suggested a region displaying remarkable amino acid deletions (referred to GS region), as a possible candidate for guanidine substrate recognition site. A phylogenetic analysis using amino acid sequences of all four phosphagen kinases indicates that CK, GK and LK probably evolved from a common immediate ancestor protein.