Methods development for structural biology
Co-workers: Jonathan Brooks-Bartlett, Katharina Jungnickel, Charles Bury
Our research focuses on improving methods for structural biology and particularly for Macromolecular Crystallography (MX) to enable problems not previously accessible to structure solution to be tackled. This work currently includes studies on 100K and room temperature (RT) radiation damage, modelling the 3-D distribution of absorbed dose during an MX experiment, and the accurate quantitative analysis of the trace elements in proteins using microbeam Proton Induced X-ray Emission (microPIXE).
Radiation damage (Figs 1a and b) to the sample is an inherent problem when utilising ionising X-rays in MX, even during cryocrystallographic (100K) data collections. We seek to understand the physical and chemical basis of the damage, and to find mitigation strategies to optimise our data collections and thus maximise the biological information obtained. For instance, we have carried out a series of studies on the addition of radioprotectants to protein crystals, and have identified a method to analyse the B-factors of structures deposited in the Protein Data Bank for signs of radiation damage.
Following our previous landmark measurement of 30 MGy as the experimental dose limit for crystals held at 100K, we are now characterising the diffraction intensity decay in more detail so that we can predict crystal lifetimes for a wide range of experimental conditions and data collection strategies. A crucial component of this work is modelling the absorbed dose distribution in the crystal (Figure 2), and our software RADDOSE-3D, which can be accessed at raddo.se.
It is under continuous development and was recently extended to cover SAXS experiments. The next stage is its incorporation into data collection pipelines at synchrotrons.
The RADDOSE-3D software, in conjunction with a new pipeline for analysing electron density differences over multiple consecutively collected diffraction datasets, has enabled us to characterise the radiation damage to crystals of both a DNA/protein complex and a large protein/RNA complex, the tryptophan RNA-binding attenuation protein (TRAP) (Fig 3). From these studies we have concluded that both DNA and RNA are far less radiation sensitive than protein.
Our method for accurately determining the concentration of trace elements bound in liquid or crystalline protein samples by proton induced X-ray emission (PIXE) is finding on-going application. It has recently been further developed to be a high-throughput technique using a non-contact array jet printer to deposit protein drops. We have now carried out a study of 29 proteins provided by a structural genomics consortium in the USA and found unexpected metals present in them, some of which are functional. This reveals that unambiguous identification of metals in proteins is a vital component of their characterisation
RNA protects a nucleoprotein complex against radiation damage
Charles S. Bury, John E. McGeehan, Alfred A. Antson, Ian Carmichael, Markus Gerstel, Mikhail B. Shevtsov, Elspeth F. Garman Acta Crystallographica D (2016) (in press)
Identifying and quantifying radiation damage at the atomic level.
Markus Gerstel, Charlotte M. Deane, Elspeth F. Garman Journal of Synchrotron Radiation (2015) 22, 201-212
Developments in X-ray Crystallographic Structure Determination of Biological Macromolecules.
Elspeth F. Garman. Science (2014) 343, 1102-1108
Predicting the X-ray lifetime of protein crystals
Oliver B. Zeldin, Sandor Brockhauser, John Bremridge, James Holton and Elspeth Garman. PNAS (2013) 110, 20551-6
RADDOSE-3D: time- and space-resolved modelling of dose in macromolecular crystallography
Oliver B. Zeldin, Markus Gerstel and Elspeth Garman Journal of Applied Crystallography (2013) 46, 1225-1230.
Elemental analysis of proteins by microPIXE.
Elspeth Garman and Geoff Grime (2005) Progress in Biophysics and Molecular Biology 89/2, 173-205
Figure 1: A crystal of DNA in a cryo-loop at 100 K before (eft) and after (right) irradiation with X-rays at Diamond Light Source (dose approx. 70 MGy). The crystal has gone black due to radiation damage.
Figure 2: Dose modelling in RADDOSE-3D (www.raddo.se) for a data collection strategy with a Gaussian profile X-ray beam. The crystal centre has been displaced from the goniometer rotation axis (causing the doughnut shaped contours) to spread the dose more evenly and reduce the very high dose at the centre if no offset is used.
Figure 3: TRAP (tryptophan attenuation protein) complex showing (left) the contents of one crystallographic asymmetric unit and (right) a face on view of the complex.