Award details

Diamond professorial fellowship for imaging chromosomes by coherent X-ray diffraction

Reference	BB/H022597/1
Principal Investigator / Supervisor	Professor Ian Robinson
Co-Investigators / Co-Supervisors
Institution	University College London
Department	London Centre for Nanotechnology
Funding type	Research
Value (£)	1,643,968
Status	Completed
Type	Fellowships
Start date	01/01/2011
End date	31/12/2015
Duration	60 months

Abstract

The CXD imaging will be made at I-13 in the forward ('small-angle') scattering geometry as originally used by Miao and also by Nishino. I-13 has a 250m distance from the source, so any disturbance of its wavefront by the front-end windows and collimating lenses or mirrors will be minimised. The coherence-defining aperture, clean-up slits, sample and beamstop will all be placed in vacuum to avoid parasitic scattering. The frozen sample will need an additional cryoshield to prevent sublimation. A temporary setup in air will be used while the cryo-vacuum system is under development; air dried samples will be used for this or else a 'cryostream' cooling system. The samples will be less than 5 microns in size, which is the field of view planned. The Medipix detector has 55 micron pixels, so it will have to be placed at least 1.8m away when used with 8keV X-rays. Lensless imaging, as in MX crystallography, requires phases to be determined for the recorded diffraction data. From there the image is achieved with a simple Fourier transform (FT). In Nishino's work and most of the recent work of the PI, this requires three essential steps. As pointed out by Sayre in 1953, the data must be oversampled with respect to the Shannon sampling frequency of the object; only then is the inversion problem overdetermined. This in turn implies the object being imaged must be 'compact' so that a 'support' or molecular envelope can be defined for use as a real-space constraint on the iterative FT phasing algorithm. Last, but not least, local minima must be avoided in a large-dimensional space, for which the Fienup Hybrid Input-Output (HIO) algorithm is superb. The craft of efficiently phasing the data follows from strategic use of computational tricks and alternation and recycling of methods, as is standard in MX crystallography. The PI's group has more than 10 years of experience with these phasing methods and several notable successes.

Summary

The chromosome is the repository of all genetic material in eukaryotes. Humans have 22 separate chromosomes in their karyotype plus the special X/Y sex-determining chromosome. One third of the life cycle of the chromosome, metaphase, is involved with mitosis, whereby the safe transmission of all the genetic material to progeny cells is undertaken; the remainder of the time, interphase, is when the genes are are unpacked and transcribed to operate the cellular machinery and copied to make the daughter chromosomes. Mitosis is an elaborate combination of cellular processes of which the protective packaging of the DNA into safe organized bundles - the chromosomes - is one of the fundamental steps. The first levels of organization of chromatin, the double helix, the nucleosome (DNA-histone protein complex) and the 'beads on a string' (BoS), packed into 30nm fibres with histone H1, have represented major breakthroughs in biology resulting in Nobel prizes. These levels of structure, summarized in Fig 1, are now relatively well understood, although there are still several possible models discussed for the fibre, and possibly some interesting diversity also. It is the next level of structure of the metaphase chromosome where the organisation becomes complicated and the research proposed here will begin to have impact. To protect the genes in transit though mitosis, they are packed together tightly into the familiar X-shaped chromosome pairs that separate once the cell division begins. The 30nm fibres are presumably coiled up in a regular superstructure at the next level within the chromatids; this coiling is the structure we intend to image by X-ray methods. This packing is achieved by scaffolding proteins which also protect the DNA mechanically from forces that could damage the genes. The optical refractive index of this dense complex is high enough that chromosomes are most easily visualized in metaphase. The high density also means that metaphase chromosomes can be handled under the optical microscope with micromanipulation tools that are familiar to cytogeneticists. We plan to use these same tools for sample preparation for the X-ray imaging experiments. This proposal plans to undertake a full 3D imaging of the chromosome at the 30nm resolution level. The metaphase chromosome is of necessity a compact object devoid of hanging strands or loops that would interfere with mitosis. This works well with the use of a 'support' constraint in phasing methods partly developed by the PI. The chomosome electron density is high for a biological substance because of the tightly-packed phosphorus (and counterions associated with the DNA). Staining methods might even be developed to enhance the X-ray contrast. The well-defined boundary will enable support-seeking methods such as 'shrinkwrap' to work effectively. Because they are readily manipulated, as described above, individual chromosomes can be isolated and mounted on pins or fiducialised membranes for measurement. Yet the chromosome is not expected to be a highly reproducible structure like the ribosome or certain viruses that can be solved by MX or the newer method of serial crystallography with an X-ray free-electron laser. In fact, much of the interest in cytogenetics lies in the differences between chromosomes of individuals and between copies from the same individual. All this can be contemplated and attempted in the context of the HRC.

Impact Summary

A major outcome of the planned work will be a new way of looking inside the chromosome that could have widespread impact in medicine. Knowledge of the coiling and packing architecture within the metaphase chromosome could have unimaginable medical implications, for example new classes of genetic defect in the population and consequential new therapies. The Harwell Research Complex (HRC) has a wide-platform activity in Correlative Microscopy, called OCTOPUS, of which one 'leg' is X-ray imaging. Coherent Diffractive Imaging (CDI) is just one part of this activity, but potentially high-impact. Other groups at Diamond and in the UK have interest in phase contrast imaging, which, like CDI, needs the high coherence of a 3rd generation synchrotron source. Another technique is X-ray microscopy, both scanning and transmission, which will also be a big activity at Diamond. CDI is more specialized and less developed than the other imaging techniques. However the greatest impact will come about by strong interactions between these communities, which will be greatly facilitated by our on-site presence at Harwell. Dave Clarke, of the Harwell Central Laser Facility (CLF), is interested in developing a common sample mounting for Correlative Microscopy and a system of fiducials. Since optical microscopy is the starting point and frame of reference in chromosome biology, we will adopt this system, which will immediately benefit the X-ray imaging methodology. Microscopy generally will have greater impact when this is possible to apply as many imaging modes as possible to the same sample set. Similarly, we would expect to interact strongly with the users and developers of the just-ordered JEOL 200kV TEM in the HRC. X-ray to electron Correlative Microscopy would be very new and of high impact. We will develop the cryofreezing capabilities as part of the project with particular attention to encouraging the formation of vitreous ice, which is desirable to avoid spoiling the images with ice crystals (which show strong phase contrast at their boundaries). The cryofreezing methods we will need to develop can learn from the experience of Baumeister (Munich), who has developed the methodology for TEM. The CCP4 software distribution activity will be located also in the HRC. It is hoped this will be extended to cover inter-operability with imaging data. There is considerable potential impact in the development of common file formats, phasing algorithms, image segmentation methods, tomographic reconstruction etc. We have already had useful interactions with Robert Attwood of Diamond's I-12 imaging beamline, who is an expert on imaging software methods, and would hope to involve him in common activities through the HRC. Lensless imaging using X-rays (i.e. using the phasing of CXD instead of a lens) is the method to be exploited in the current proposal. The distribution of the once- developed methods by a professional organization will bring impact to all parties. In the future, there lies the exciting prospect of CDI enabling general purpose 3D imaging of biological samples, limited as always by the radiation damage. There is a lot of current excitement about collecting entire diffraction patterns in a single shot of X-rays at free-electron laser (XFEL) sources; it is estimated that samples will be immune to radiation damage at atomic resolution for 10-20 femtoseconds, which is an achievable pulse length. The further development of CDI through this proposal will have impact on XFEL science.

Committee	Research Committee C (Genes, development and STEM approaches to biology)
Research Topics	Structural Biology
Research Priority	X – Research Priority information not available
Research Initiative	Fellowship - Professorial Fellowship (PF) [1996-2009]
Funding Scheme	X – not Funded via a specific Funding Scheme

I accept the terms and conditions of use (opens in new window)

export PDF file

back to list new search