Crystal structures of 2OG oxygenases involved in ribosomal protein hydroxylation

Post-translational modifications play diverse biological functions. Hydroxylation of collagen proteins has long been a recognised post-translational modification in eukaryotes. In the case of collagen, hydroxylation of prolyl residues, by 2-oxoglutarate and iron dependent enzymes (2OG oxygenases), in collagen proteins allows for the stabilisation of the collagen triple helix structure through conformational restraint and through the addition of a hydrogen bond donor. Additionally, hydroxylation of lysine side chains of collagen is required for cross-linking collagen (and possibly other proteins) in the extra-cellular matrix. Post-translational prolyl hydroxylation also plays a pivotal role in transcriptional regulation of the hypoxic response, as catalyzed by the hypoxia inducible factor / HIF prolyl hydroxylases (PHDs or EGLN enzymes). Recently, ribosomal protein hydroxylation catalyzed by 2OG- and Fe(II)-dependent oxygenases has been found to be a highly conserved post-translational modification in eukaryotes and prokaryotes (Ge et al and Loenarz et al). We present several crystal structures of 2OG oxygenases involved in ribosomal protein hydroxylation.

When protein crystal dehydration helps

A typical protein crystal contains 30-60% solvent. For a naked crystal, this solvent is distributed between solvent shells, where water and solvent molecules make specific interactions with the crystalline protein, and solvent channels filled with disordered solvent molecules. This internal solvent map of the crystal can be modified by placing the crystal in a dehydrating environment. This may in turn induce changes to the crystal lattice and affect mosaicity, resolution and quality of diffraction data. A dehydrating environment can be generated around a crystal in several ways with various degrees of precision and complexity. In this study we have used the HC1 device (Maatel) to mount crystals an air stream of known relative humidity - a precise yet hassle-free approach to altering crystal hydration. We set out to analyse a range of different crystals to establish usable protocols that will allow one to explore to crystal hydration space, either by preparing samples before synchrotron beamtime or by undertaking the experiments during beamtime. Our results, considered in the light of the literature surrounding crystal dehydration, provide guidance for when dehydration can help diffraction.

I04-1, a future 'fragment screening' beamline facility at Diamond Light Source

I04-1 is one of the six macromolecular crystallography (MX) beamlines at Diamond Light Source (DLS), the third generation synchrotron light source in the UK. It was built and delivered in 2010 as a stable and reliable fixed-wavelength MX station. It is currently preparing to release its user programme for exploiting fragment screening using X-ray crystallography in structural medicinal chemistry projects. For this purpose, the beamline has been going through several upgrades in order to achieve unattended high-throughput ligand crystallography. The new developments are aiming at improving the flux, stability and reliability of the beamline and its auto-alignment. In parallel, a peripheral laboratory is being set up to provide a facility for medium throughput compound soaking. Jointly with the Structural Genomics Consortium (SGC), a semi-automatic crystal soaking and harvesting scheme, which will provide hundreds of MX samples per day, is being tested at DLS. The beamline can currently process 400 crystals per day. However, the recent upgrades and automation should further improve that throughput. In this presentation, we will summarise the current specifications of the beamline and its new features, the development of a peripheral laboratory for compounds soaking and underline the remaining work.

Virtual Screening of Histone Lysine Demethylase(JMJD2) identifies new inhibitors

The JmjC domain-containing proteins are hydroxylases that confer posttranslational modifications on histone tails, by removing methylation marks on methylated lysine residues. This serves to either promote or repress gene transcription. The JMJD2A-D family members include the enzyme Jumonji domain 2C (JMJD2C), which specifically demethylates di- and trimethylated histone H3 at Lys 9 or Lys 36.[1] Dysregulation of JMJD2C has been implicated in prostate, colonic, and breast cancer as the demethylase can modify the expression levels of oncogenes.[2] The goal of the present study was to identify potent and selective small-molecule inhibitors of JMJD2C, to be used as chemical biology tools to further investigate the role of JMJD2C in cell proliferation and survival. Using high-resolution crystal structures of the JMJD2 subfamily members as templates, we have performed a small molecule virtual docking screen. From the ~3 million molecules that were docked, this experiment identified 21 compounds as possible leads. These compounds were tested against JMJD2C in enzymatic assays and here we report an overall hit rate of 76%, with 8 compounds demonstrating an IC50 of 176μM to 1.18μM. A molecule containing a salicylate core was selected as a candidate for optimization and thus far we have completed several rounds of iterative target-specific compound docking, hybrid molecule design, compound synthesis and in vitro characterization. Notably, our method demonstrated a substantial increase in potency when we linked two docked fragments together and further derivatized this new scaffold, through which we have successfully derived a 65nM inhibitor of JMJD2C. A compound representing the inhibitor scaffold has been co-crystallized with JMJD2A to a resolution of 2.4 Å. In the crystal structure each asymmetric unit contains two JMJD2A monomers, each bound to a single inhibitor molecule. This complex-structure superposes well with the docked pose for the hybrid series of compounds. We are now focusing our efforts on identifying an inhibitor that is selective for the JMJD2 family over other JmjC domain-containing proteins.