Search

Multiprotein complexes, rather than individual proteins, make up a large part of the biological macromolecular machinery of a cell. Understanding the structure and organization of these complexes is critical to understanding cellular function. Chemical cross-linking coupled with mass spectrometry is emerging as a complementary technique to traditional structural biology methods and can provide low-resolution structural information for a multitude of purposes, such as distance constraints in computational modeling of protein complexes. In this review, we discuss the experimental considerations for successful application of chemical cross-linking-mass spectrometry in biological studies and highlight three examples of such studies from the recent literature. These examples (as well as many others) illustrate the utility of a chemical cross-linking-mass spectrometry approach in facilitating structural analysis of large and challenging complexes.

Phage viruses that infect prokaryotes integrate their genome into the host chromosome; thus, microbial genomes typically contain genetic remnants of both recent and ancient phage infections. Often phage genes occur in clusters of atypical G+C content that reflect integration of the foreign DNA. However, some phage genes occur in isolation without other phage gene neighbors, probably resulting from horizontal gene transfer. In these cases, the phage gene product is unlikely to function as a component of a mature phage particle, and instead may have been co-opted by the host for its own benefit. The product of one such gene from Salmonella enterica serovar Typhimurium, STM3605, encodes a protein with modest sequence similarity to phage-like lysozyme (N-acetylmuramidase) but appears to lack essential catalytic residues that are strictly conserved in all lysozymes. Close homologs in other bacteria share this characteristic. The structure of the STM3605 protein was characterized by X-ray crystallography, and functional assays showed that it is a stable, folded protein whose structure closely resembles lysozyme. However, this protein is unlikely to hydrolyze peptidoglycan. Instead, STM3605 is presumed to have evolved an alternative function because it shows some lytic activity and partitions to micelles.

Protein domain family YabP (PF07873) is a family of small protein domains that are conserved in a wide range of bacteria and involved in spore coat assembly during the process of sporulation. The 62-residue fragment of Dsy0195 from Desulfitobacterium hafniense, which belongs to the YabP family, exists as a homodimer in solution under the conditions used for structure determination using NMR spectroscopy. The structure of the Dsy0195 homodimer contains two identical 62-residue monomeric subunits, each consisting of five anti-parallel beta strands (β1, 23-29; β2, 31-38; β3, 41-46; β4, 49-59; β5, 69-80). The tertiary structure of the Dsy0195 monomer adopts a cylindrical fold composed of two beta sheets. The two monomer subunits fold into a homodimer about a single C2 symmetry axis, with the interface composed of two anti-parallel beta strands, β1-β1' and β5b-β5b', where β5b refers to the C-terminal half of the bent β5 strand, without any domain swapping. Potential functional regions of the Dsy0195 structure were predicted based on conserved sequence analysis. The Dsy0195 structure reported here is the first representative structure from the YabP family.

We have determined the solution NMR structure of SACOL2532, a putative GCN5-like N-acetyltransferase (GNAT) from Staphylococcus aureus. SACOL2532 was shown to bind both CoA and acetyl-CoA, and structures with and without bound CoA were determined. Based on analysis of the structure and sequence, a subfamily of small GCN5-related N-acetyltransferase (GNAT)-like proteins can be defined. Proteins from this subfamily, which is largely congruent with COG2388, are characterized by a cysteine residue in the acetyl-CoA binding site near the acetyl group, by their small size in relation to other GNATs, by a lack of obvious substrate binding site, and by a distinct conformation of bound CoA in relation to other GNATs. Subfamily members are found in many bacterial and eukaryotic genomes, and in some archaeal genomes. Whereas other GNATs transfer the acetyl group of acetyl-CoA directly to an aliphatic amine, the presence of the conserved cysteine residue suggests that proteins in the COG2388 GNAT-subfamily transfer an acetyl group from acetyl-CoA to one or more presently unidentified aliphatic amines via an acetyl (cysteine) enzyme intermediate. The apparent absence of a substrate-binding region suggests that the substrate is a macromolecule, such as another protein, or that a second protein subunit providing a substrate-binding region must combine with SACOL2532 to make a fully functional N-acetyl transferase.

The solution structure of MTH1175, a 124-residue protein from the archaeon Methanobacterium thermoautotrophicum has been determined by NMR spectroscopy. MTH1175 is part of a family of conserved hypothetical proteins (COG1433) with unknown functions which contains multiple paralogs from all complete archaeal genomes and the archaeal gene-rich bacterium Thermotoga maritima. Sequence similarity indicates this protein family may be related to the nitrogen fixation proteins NifB and NifX. MTH1175 adopts an alpha/beta topology with a single mixed beta-sheet, and contains two flexible loops and an unstructured C-terminal tail. The fold resembles that of Ribonuclease H and similar proteins, but differs from these in several respects, and is not likely to have a nuclease activity.