One of the great successes of PSI Biology has been the ability to apply the powerful PSI pipelines for high-throughput structure determination to topics of interest in the wider structural biology research community. Through a program of community nominated targets, the PSI has helped researchers explore topics from ranging from protein design to translational medicine, determining more than 500 new macromolecular structures of high biological significance in the process. A few examples are given here, and many additional examples have been presented in past installments of the PSI Structural Biology Knowledgebase.
Decoding a DUF
Working with Stuart Kornfeld at Washington University School of Medicine, PSI researchers at JCSG have determined the structure and function of a new Domain of Unknown Function. DUF2233 is interesting because it is found in hundreds of bacterial proteins, but only one mammalian protein: a protein found in the Golgi that is involved in targeting proteins to the lysosome. The structure of a bacterial homolog of this protein, shown here from PDB entry 3ohg, revealed a deep active site, lined with amino acids that are conserved in the many homologous bacterial proteins (shown in brighter turquoise). Additional biochemical study revealed that these domains are involved in cleavage of sugar-phosphate compounds, such as the sugars involved in lysosomal targeting. A sulfate ion (shown here in bright yellow and red) was found in the structure in the site that may recognize phosphate in these compounds.
Click on the JSMol tab for an interactive JSmol
DUF2233 Domain (PDB entry 3ohg)
A sulfate is bound in the presumed active site, shown here colored by atom. Use the buttons to show the protein atoms with conserved amino acids shown in brighter turquoise, or a cartoon diagram with the domains colored differently.
Transition States to Translational Medicine
Vern L. Schramm at the Albert Einstein College of Medicine is an expert on using transition state analogues to inhibit key enzymes in the cell. PSI researchers at NYSGRC have worked with him for many years to uncover the structural details of these interactions and to use them to design new therapeutic drugs. The structure shown here, from PDB entry 4ffs, is one of their recent successes. It shows a new inhibitor (in spheres) that binds tightly to the active site of an enzyme from Helicobacter pylori, a bacterium involved in the development of stomach ulcers. The inhibitor blocks the enzyme, but is also highly specific. This is important in an ulcer-fighting drug, so that the drug will fight the pathogenic bacterium but spare the normal bacterial flora.
Click on the JSMol tab for an interactive JSmol
5'-Methylthioadenosine Nucleosidase (PDB entry 4ffs)
The inhibitor is shown with spheres and atomic colors, and a water molecule that may be important in the reaction is shown in turquoise. Notice how the active site, which surrounds the inhibitor, is composed of amino acids from both chains, colored blue and green. Use the buttons to show different representations of the protein and zoom in on the inhibitor.
Tails and Targeting
PSI researchers at NESG have joined up with Monica Roth at Robert Wood Johnson Medical School to uncover the structural basis of targeting of a retrovirus. The integrase enzyme of Maloney Murine Leukemia Virus (MLV) favors sites near active genes when it integrates its viral genome into the host genome. This is beneficial for the virus, since it ensures that the viral genome will be actively transcribed, but can cause big problems for the host, since these sites may cause oncogenic mutations. PSI researchers have determined the NMR structure of the domain of MLV integrase that is involved in this targeting, shown here from PDB entry 2m9u. It has a long, flexible tail, and NMR analysis revealed that many amino acids in this tail (shown in red), interact with chromatin proteins. This result allowed them to design a truncated integrase that may reduce tumor-promoting activity when MLV is used in human gene replacement therapy.Click on this image for an interactive JSmol
MLV integrase C-terminal domain (PDB entry 2m9u)
This JSmol is looping through the 20 conformations determined in the NMR structure analysis. Residues colored in red showed changes in the NMR spectrum when the protein associated with the cellular protein Brd3.
3ohg: Das, D. et al. Structure and function of the DUF2233 domain in bacteria and in the human mannose 6-phosphate uncovering enzyme. J. Biol. Chem. 288, 16789-16799 (2013).
4ffs: Wang, S. et al. A picomolar transition state analogue inhibitor of MTAN as a specific antibiotic for Helicobacter pylori. Biochem. 51, 6892-6894 (2012).
2m9u: Aiyer, S. et al. Altering murine leukemia virus integration through disruption of the integrase and BET protein family interaction. Nucl. Acids Res. 42, 5917-5928 (2014).