The targets of the Structural GenomiX (SGX) bacterial genomics project were proteins conserved in multiple prokaryotic organisms with no obvious sequence homolog in the Protein Data Bank of known structures. The outcome of this work was 80 structures, covering 60 unique sequences and 49 different genes. Experimental phase determination from proteins incorporating Se-Met was carried out for 45 structures with most of the remainder solved by molecular replacement using members of the experimentally phased set as search models. An automated tool was developed to deposit these structures in the Protein Data Bank, along with the associated X-ray diffraction data (including refined experimental phases) and experimentally confirmed sequences. BLAST comparisons of the SGX structures with structures that had appeared in the Protein Data Bank over the intervening 3.5 years since the SGX target list had been compiled identified homologs for 49 of the 60 unique sequences represented by the SGX structures. This result indicates that, for bacterial structures that are relatively easy to express, purify, and crystallize, the structural coverage of gene space is proceeding rapidly. More distant sequence-structure relationships between the SGX and PDB structures were investigated using PDB-BLAST and Combinatorial Extension (CE). Only one structure, SufD, has a truly unique topology compared to all folds in the PDB.

The yigZ gene from Escherichia coli K12 is widely conserved among thermophiles, archaea, and pathogens including Yersinia pestis, Vibrio cholerae and Salmonella typhimurium. While the function for this protein remains uncharacterized, domain conservation patterns suggest that the yigZ gene product plays an indispensable function, making it a possible antimicrobial drug target. The closest mammalian homologue of yigZ is the gene impact (Accession ID: AAG35736). Mouse impact is an imprinted gene but little else is known of its function. Imprinted genes are expressed in a parent-of-origin-dependent manner,1 and generally have roles in differentiation, development, and regulation of cell proliferation. Aberrations in imprinted genes or their regulation have been implicated in various human diseases including Prader-Willi syndrome, Angelman syndrome, diabetes mellitus, bipolar affective disorder, and some malignant tumors.2 Knowledge of the YIGZ protein structure provides a potential template to model the structure of the IMPACT protein and provide clues regarding biochemical function. This report describes the x-ray structure determination of YIGZ to 2.8 Å resolution. Structural homology provides some insights into the function of this protein and reveals a novel polypeptide chain fold. yigZ PCR product from the E. coli genome (Accession ID: NP_4182903) was TOPO isomerase cloned into HexaHis bacterial expression plasmids with a forward primer (5′-ATGGAAAGCTGGTTAATTCCTGC-3′) and a reverse primer (5′-CTTCTTCAATCGCTAACAATTGC-3′). The final amino acid sequence of the protein is as follows: MSLMESWLIP AAPVTVVEEI KKSRFITMLA HTDGVEAAKA FVESVRAEHP DARHHCVAWV AGAPDDSQQL GFSDDGEPAG TAGKPMLAQL MGSGVGEITA VVVRYYGGIL LGTGGLVKAY GGGVNQALRQ LTTQRKTPLT EYTLQCEYHQ LTGIEALLGQ CDGKIINSDY QAFVLLRVAL PAAKVAEFSA KLADFSRGSL QLLAIEEEGG SHHHHHH Selenomethionine protein was expressed in E. coli BL21[DE3] cells. After lysis, the protein was purified via nickel ion affinity chromatography [50 mM TrisHCl, pH7.8, 500 mM NaCl, 10 mM imidazole, 10 mM methionine, 10% glycerol, 1 mM DTT with a linear gradient of 10–500mM imidazole]. The protein was then subjected to gel filtration chromatography with Superdex75 (Pharmacia). S-met protein was expressed and purified using similar procedures. Crystals were obtained within four days at 9°C via hanging drop vapor diffusion with equal volumes of protein [5 mg/ml, 1.0 mM β-mercaptoethanol (βME), 150 mM NaCl, 10 mM HEPES pH 7.5, 10 mM methionine, 10% glycerol] and reservoir [150 mM MES pH 6.5, 150 mM (NH4)2SO4, 30% PEG MME 5000 (w/v), 28.4 mM βME] solutions. Crystals were transferred to a cryoprotective solution consisting of 10% glycerol, 10% ethylene glycol, and 80% reservoir. Diffraction data were collected from frozen S-met and Se-Met crystals at the APS COM-CAT beam line, processed and reduced with MOSFLM, SCALA, and TRUNCATE.4 Four selenium atom sites were located using SnB5 and refined with SHARP.6 The SAD phased electron density map was improved by solvent flattening with SOLOMON4 and used for initial model building. Refinement was performed against the S-Met dataset with iterative cycles of manual model building with XTALVIEW/XFIT,7 REFMAC,4 and CNX.8 Quality assessment of the model was performed with PROCHECK,9 WHATCHECK,10 and SFCHECK11 (Table I). The YIGZ protein is a 23.2 kD monomer consisting of two domains separated by a six amino acid linker. The N′ terminal domain (residues 3–136) consists of a five-stranded anti-parallel β sheet with three α helices. The C′ terminal domain (residues 139–208) consists of a four-stranded anti-parallel β-sheet with two α-helices packed against one face of the β-sheet [Fig. 1(A)]. A: SPOCK20 ribbon diagram of YIGZ from Escherichia coli K12. β-strands and α-helices are shown in cyan and orange, respectively. Circle indicates region illustrated in part B. B: Putative catalytic residues and substrate binding site residues are rendered as ball-and-stick figures in yellow and grey, respectively. Figure produced with PyMOL.21 The PredAct™12 program is used to identify and rank polar residues that are conserved in sequence alignments and cluster within 5Å of one another. Ser23, His54, Glu77, and Arg104 almost certainly participate in catalysis [Fig. 1(B)]. These four residues are absolutely conserved among mammalian IMPACT proteins, which share less than 30% sequence identity overall. Lys22, Arg24, Phe25, Asp75, Gly76, Pro78, Ala82, Tyr105, Tyr106, Gly107, Leu111, Leu116, Tyr120, Asp74, and Thr81 probably contribute to substrate binding. These residues cluster on the distal side of the N′ domain within an electropositive surface charge (data not shown). A search for structural homologues of YIGZ was performed using the DALI server (http://www.ebi.ac.uk/ dali/).13 Queries with the entire protein or with the N′ terminal domain did not yield significant hits. A query with the C′ terminal domain revealed some possible functions. The four highest scoring, albeit weak, structural homologies of the C′ terminal domain include three ribosomal proteins [1DAR14 (Z-score = 9.6, 1.9 Å RMSD for 69 equivalent α-carbons), 1JQS15 (Z-score = 7.0, 2.4 Å RMSD for 66 equivalent α-carbons), and 1RIS16 (Z-score = 8.2, 2.1 Å RMSD for 67 equivalent α-carbons)] and the propeptide portion of procarboxypeptidase a2 [1AYE17 (Z-score = 6.5, 2.1 Å RMSD for 64 equivalent α-carbons)]. The ribosomal proteins include bacterial peptide elongation factor G (EF-G) and S6. EF-G and its eukaryotic counterpart, EF-2, are members of the GTPase superfamily of proteins.18 The ribosomal protein S6 from Thermus thermophilus interacts with RNA. Sequence alignments of EF-G, S6, and the C′ terminal domain suggest that this portion of the YIGZ protein may interact with nucleic acids. The C′ terminal domain of the YIGZ protein also resembles the pro-portion of procarboxypeptidase A2. Pancreatic carboxypeptidases catalyze C′ terminal exoproteolysis of alimentary proteins and esters during digestion.19 They are secreted as zymogens with a 94–95 residue N′ terminal pro-segment, which occludes the active site. The A2 isoform has a preference for aliphatic and aromatic C-terminal residues and has a clear specificity for bulkier aromatic C-terminal residues. While the structure of the YIGZ protein from E. coli K12 has been determined, its biochemical and biological functions remain unclear. Structural evidence and sequence homology of the YIGZ protein suggest that the C′ terminal domain binds nucleic acids, but essentially shed no light on the function(s) supported by the balance of the polypeptide chain. Functional studies of the YIGZ protein or a member of the IMPACT families may be accelerated by our X-ray crystallographic results. We thank the staff at the Advanced Photon Source and COM-CAT, for their help during data collection. The atomic coordinates (code: 1VI7) have been deposited in the Protein Data Bank (http://www.rcsb.org/).

We have begun a small scale structural genomics project aimed at obtaining fold information for the set of sequence families comprising eukaryotic intracellular signaling domains, and exploiting that information to understand biological function and mechanism.The SMART database (http://smart.embl-heidelberg.de/smart) is the primary target list for the project.The practical issues of obtaining soluble and crystallizable representatives for each sequence family will be discussed.The question of how much can be learned about signaling mechanisms from elucidating the structure of one representative protein per sequence family will be considered.Examples illustrating issues of principle and practice will be drawn from recent work on VHS and other domains.

Background: Quorum sensing is the mechanism by which bacteria control gene expression in response to cell density. Two major quorum-sensing systems have been identified, system 1 and system 2, each with a characteristic signaling molecule (autoinducer-1, or Al-1, in the case of system 1, and Al-2 in system 2). The luxS gene is required for the Al-2 system of quorum sensing. LuxS and Al-2 have been described in both Gram-negative and Gram-positive bacterial species and have been shown to be involved in the expression of virulence genes in several pathogens.Results: The structure of the LuxS protein from three different bacterial species with resolutions ranging from 1.8 Angstrom to 2.4 Angstrom has been solved using an X-ray crystallographic structural genomics approach. The structure of LuxS reported here is seen to have a new alpha-beta fold. In all structures, an equivalent homodimer is observed. A metal ion identified as zinc was seen bound to a Cys-His-His triad. Methionine was found bound to the protein near the metal and at the dimer interface.Conclusions: These structures provide support for a hypothesis that explains the in vivo action of LuxS. Specifically, acting as a homodimer, the protein binds a methionine analog, S-ribosylhomocysteine (SRH). The zinc atom is in position to cleave the ribose ring in a step along the synthesis pathway of Al-2.