Menu
Institut de Biologie StructuraleGrenoble / France

Contact person(s) related to this article / FARIAS ESTROZI Leandro

Combining EM and X-ray data

Phasing X-ray data of symmetric oligomeric proteins using EM reconstructions

This project aims to validate the combined use of EM data with low resolution X-ray diffraction data for the determination of small symmetric oligomeric structures.

Negative stain EM images of symmetric oligomeric proteins will be used to obtain their 3D reconstruction. The internal symmetry of the oligomer, which can be obtained from the crystallographic self-rotation function, will be imposed in order to improve the reconstruction procedure. The 3D reconstructions will then be used as MR probes to phase the low resolution X-ray data obtained from crystals of these proteins. Finally, phase extension will be carried out by exploiting the non crystallographic symmetry which arises from the internal symmetry of the protein.

This approach has already proven successful for large structures, i.e. viruses (Dodson, 2001). We will use crystals of three proteins, oligomers ranging from 90 to 200 kDa. For two of them the structure is already determined and will be used to confirm the validity of the method, optimize the strategy and eventually estimate the limits (particles size) of the method. For the third protein, we wish to solve its structure using this approach.

Previously (Trapani et al., 2006), we demonstrated the usefulness of combining multiple approaches for difficult molecular replacement projects. However, due to the lack of overlap between our X-ray diffraction and electronic microscopy data we failed to phase the X-ray data using the electronic microscopy reconstruction directly as a molecular replacement probe. Therefore, a major effort will be made in order to obtain complete diffraction data at very low resolution.

Contacts:
- Jorge Navaza : jorge.navaza@ibs.fr
- Stefano Trapani : stefano.trapani@ibs.fr

Collaborators:
- Information Génomique Et Structurale, CNRS, Marseille

References

Dodson, E. J. (2001) Using electron-microscopy images as a model for molecular replacement. Acta Cryst. D57, 1405-1409.

Trapani, S., Abergel. C., Gutsche, I., Horcajada, C., Fita, I., Navaza, J. (2006) Combining experimental data for structure determination of flexible multimeric macromolecules by molecular replacement. Acta Cryst. D62, 467-475.

Combining X-ray and EM data of small symmetric oligomeric proteins

Phasing X-ray data of symmetric oligomeric proteins using EM reconstructions

This project aims to validate the combined use of EM data with low resolution X-ray diffraction data for the determination of small symmetric oligomeric structures.

Negative stain EM images of symmetric oligomeric proteins will be used to obtain their 3D reconstruction. The internal symmetry of the oligomer, which can be obtained from the crystallographic self-rotation function, will be imposed in order to improve the reconstruction procedure. The 3D reconstructions will then be used as MR probes to phase the low resolution X-ray data obtained from crystals of these proteins. Finally, phase extension will be carried out by exploiting the non crystallographic symmetry which arises from the protein internal symmetry.

This approach has already been proven successful for large structures, i.e. viruses (Dodson, 2001). We will use crystals from three proteins, corresponding to 90 to 200 kDa oligomers. For two of them the structure is already determined and will be used to confirm the validity of the method, optimize the strategy and eventually estimate the limits (particles size) of the method. For the third protein, we wish to solve its structure using this approach.

In a previous study (Trapani et al., 2006), we have already demonstrated the usefulness of combining multiple approaches for difficult molecular replacement projects. However, due to the lack of overlap between our X-ray diffraction and electronic microscopy data we failed in phasing the X-ray data using directly the electronic microscopy reconstruction as a molecular replacement probe. Therefore, a major effort will be done in order to obtain complete diffraction data at very low resolution.

Building oligomeric MR probes by fitting atomic-resolution structures into EM reconstructions

An alternative approach constists in building an oligomeric model by fitting atomic resolution structures of homologous monomers, if available, into the low resolution EM reconstruction. In the case of symmetric oligomers, the molecular symmetry can be exploited during the fitting procedure. Also, orientation constraints derived from the analysis of the crystallographic self-rotation and cross-rotation functions can be imposed to improve the model quality (Trapani et al., 2006). The model is then used as a probe for MR.

Trimeric model of glycogen synthase (148kDa).
The monomeric structure from A. tumefaciens (AtGS) was fitted into a 3D EM reconstruction of the P. abyssi enzyme (PaGS) imposing 3-fold symmetry and orientational constraints derived from the crystallographic self- and cross-rotation function of PaGS. The trimeric model thus obtained could be successfully used as a MR probe to solve the PaGS crystal structure. A trimeric model obtained by imposing only 3-fold symmetry during the EM fitting procedure did not lead to an effective MR probe (Trapani et al., 2006).

Contacts:
- Jorge Navaza : jorge.navaza@ibs.fr
- Stefano Trapani : stefano.trapani@ibs.fr

Collaborators:
- Information Génomique Et Structurale, CNRS, Marseille

References

Dodson, E. J. (2001) Using electron-microscopy images as a model for molecular replacement. Acta Cryst. D57, 1405-1409.

Trapani, S., Abergel. C., Gutsche, I., Horcajada, C., Fita, I., Navaza, J. (2006) Combining experimental data for structure determination of flexible multimeric macromolecules by molecular replacement. Acta Cryst. D62, 467-475.

Combining EM and X-ray diffraction data for structure determination

Biological processes frequently require formation of macromolecular complexes. To understand the intricate interactions between the components of these complexes, it is desirable to know their three-dimensional (3D) structure.

Some macromolecular complexes have been produced in homogeneous form, crystallized and analyzed at high resolution (2-3 Å or better) by X-ray crystallography (XR). However, most of them are too large or too unstable to crystallize, in which case only individual components of such complexes could be analyzed by XR. On the other hand, electron microscopy (EM) allows 3D reconstructions of whole macromolecular complexes, but is limited to resolutions of 5-30 Å.

URO: a suite of programs to fit X-ray structures into EM maps

It is in principle possible to obtain a high-resolution interpretation of the EM data by fitting XR models of individual components into an EM reconstruction of the whole structure. The first successful applications of such a procedure relied on manual fitting, but recent methodological developments, such as the software URO developed by us (Navaza et al., 2002), improved considerably the fitting procedure and the information that can be gained from it.

The URO methodology is inspired from the molecular replacement technique, adapted to take into account phase information and the symmetry imposed during the EM reconstruction. Calculations are performed in reciprocal space, which enables the selection of big volumes of the EM maps, thus avoiding the bias introduced when defining boundaries of the target density. Examples of fitting obtained with URO are shown in the first two figures here below.

We are currently developing UROX, a graphical interface for URO, to allow the user to adjust interactively the fitted structures and get real-time monitoring of the quality of the fit. A screenshot of the current version of UROX can be found in the third figure here below.

Contacts:
- Jorge Navaza : jorge.navaza@ibs.fr
- Xavier Siebert : siebert@vms.cnrs-git.fr

Exemples of structures fitted using URO (Mathieu et al., 2001)

UROX screenshot. The left panel shows the graphical interface displaying the GroEl atomic structure fitted in an EM map.(*) The correlation between the X-ray model and the EM map is calculated on-the-fly and shown on the left bottom corner. The right panel shows the wizard that allows the user to conveniently enter various information such as the symmetry of the map, the resolution, etc.

(*) data from Helen Saibil’s presentation at the EMBO workshop Combination of Electron Microscopy and X-ray Crystallography in Structure Determination, organized by our group in October 2005: http://www.vms.cnrs-gif.fr/EMXRAY/index.html

References

Mathieu, M., Petitpas, I., Navaza, J., Lepault, J., Kohli, E., Pothier, P., Prasad, B. V., Cohen, J. and Rey, F. A. (2001) Atomic structure of the major capsid protein of rotavirus: implications for the architecture of the virion, EMBO J. 20(7), 1485-1497.

Navaza, J., Lepault, J., Rey, F. A., Alvarez-Rua, C. and Borge, J. (2002) On the fitting of model electron densities into EM reconstructions: a reciprocal-space formulation. Acta Cryst. D58, 1820-1825.