Second Workshop on Structural Molecular Biology (SMB) at SESAME

7. SYNCHROTRON RADIATION TECHNIQUES AND APPLICATIONS top

Scientists experienced with synchrotron radiation in biological applications gave tutorial lectures. These are summarized below.

7.1 Biological and Medical Applications of Synchrotron Infrared Microspectroscopy Lisa M. Miller
National Synchrotron Light Source, Brookhaven National Laboratory, Upton, NY, USA

Synchrotron infrared light is an ideal source for infrared micro-spectroscopy due to the combination of its high brightness (i.e. flux density) and broadband nature. A synchrotron source is 100-1000 times brighter than a conventional globar source and the broadband nature of the source is important for performing spectroscopy.
Unlike many x-ray based spectroscopies which were made possible by the advent of synchrotron radiation, infrared spectroscopy has been used for many years without synchrotron radiation. Although conventional IR microspectroscopy has proven extremely valuable for resolving the chemical components in biological samples, the long wavelengths of infrared radiation limit the spatial resolution that can be achieved. Existing instruments using a conventional IR source encounter a signal-to-noise (S/N) limitation when apertures confine the IR to an area of 20-30 mm in diameter.
The high brightness of the synchrotron source allows smaller regions to be probed with acceptable S/N. Indeed, aperture settings smaller than the wavelength of light can be used; though in this case, diffraction controls the available spatial resolution. Thus for a typical biological specimen, the diffraction-limited spatial resolution for primary lipid (C-H stretch), protein (amide I), and nucleic acid (P-O stretch) absorption features is approximately 3, 6, and 12 mm, respectively. This improvement in spatial resolution achieved by using a synchrotron IR source has only been realized ecently, and applications to biological systems are still in their infancy.
To date, synchrotron infrared microspectroscopy has been used for a number of biological and medical applications. Ongoing projects at the National Synchrotron Light Source at Brookhaven National Laboratory include examination of (1) the protein and mineral content in osteoporosis, (2) plaque formation and neuron degeneration in Alzheimer’s disease, (3) chemical changes in a cell during various stages of apoptosis, (4) gelatinase degradation products and pathways in model systems of extracellular matrix, and (4) the sub-millisecond folding of proteins such as cytochrome c.
The SESAME facility will be an ideal source for performing biological infrared spectroscopy and micro-spectroscopy. The characteristics of the SESAME ring are ideal for building competitive, state-of-the-art IR beamlines.It is important to emphasize that progress towards developing a biological IR facility for SESAME can begin immediately. Unlike many x-ray based spectroscopies, IR micro-spectroscopy can be performed in the laboratory using a conventional globar source. A number of commercial infrared microscope systems are currently available for ~ $100K USD. Many current systems are designed to accommodate an external IR source, so once the SESAME facility is completed, this system can be moved to the IR beamline with very little or no modification necessary. Although existing instruments using a globar source encounter are limited to a spatial resolution of 20-30 mm in diameter, this is sufficient for tissue analysis. Biological applications of IR micro-spectroscopy to tissue samples has already proven to be valuable in medicine. For example, aggregates of amyloid plaques have been identified in the brain tissue of Alzheimer’s disease patients. Spectral evidence of cervical cancer, heart disease and bone diseases such as osteoarthritis, osteoporosis, and osteogenesis imperfecta have been identified. In addition, contaminants in human tissue, such as silicone in breast tissue and narcotics in human hair have also been observed. The value of the technique has already been demonstrated and there is tremendous potential for future applications.In addition to realizing new applications for IR micro-spectroscopy, the immediate development of this technique will help to initiate a SESAME user community, spark collaborations, and provide a learning curve and experience for future SESAME scientists.

7.2 Small Angle X-ray Scattering as a Tool to Investigate Protein Structures

Boris Batterman; Cornell University/SSRL/LBNL

Structures in protein crystallography are normally derived from single crystal Bragg reflections, usually in the tens of thousands for relatively complex unit cells containing many thousands of atoms. Changes in individual molecules not ordered in crystalline arrays are important factors in understanding biological function. Any assemblage of atoms not in an ordered array will scatter x-rays diffusely. That is they will not produce the sharp Bragg reflections which are easy to measure.
At very low angles of scattering, this diffuse scattering can be fairly strong and can be measured using intense synchrotron x-rays, even in samples which have a miniscule amount of individual molecules of interest. The variation of the scattered intensity with respect to angle in this low angle regime gives a measure of the aggregate size and shape of the individual units. For example, one can measure the radius of gyration of a molecule, and distinguish between a folded or unfolded protein.
The ability to study the folding process in real time of a linear chain of amino acids allows one to follow a random coil in its process to a biologically active folded state. A group at CHESS, Cornell University under the direction of Lois Pollack performed such an experiment. The process was triggered in molecules of cytochrome-c by creating a rapid pH-jump by the injection of buffer solution in a specially designed flow cell. The procedure could allow tracking the folding process on a millisecond time scale. It was shown that the folding was a two step process in which the initial folding was completed in less than 1/2 millisecond
Reference: Pollack., Tate,M,W.,Darton,N.C., Knight,J.B., Gruner,S.N., Eaton, W.A., and Austin, R.H. Compactness of the Denatured State of a Fast-Folding Protein Measured by Submillisecond Small- Angle X-ray Scattering. Proc. Natl. Acad. Sci. 96, 10115-10117 (1999)
Another example of Small Angle Scattering in a biological study is given below.

Small angle X-ray solution scattering studies of recombinant copper-metallothionein from Saccharomyces Cerevisiae

Zehra Sayersx, Patricia Brouillon*, Dimitri I. Svergun, Piotr Zielenkiewicz + & Michel Koch
xSabanci University, Faculty of Engineering and Natural Sciences, Orhanli, 81474 Tuzla-Istanbul, Turkey, European Molecular Biology Laboratory, Hamburg Outstation, EMBL c/o DESY, Notkestrasse 85, D-22603 Hamburg, Federal Republic of Germany. + Institute of Biochemistry and Biophysics, Polish Academy of Sciences, ul. Pawinskiego, 02-106 Warszawa, Poland. * Present address: SPI-BIO, rue Buisson aux Fraises, F-91741 Massy Cedex.
Metallothioneins are low molecular mass cysteine rich proteins that bind heavy metals in a wide range of organisms. MTs also play a direct role in cellular defense against oxidative stress and in radiation resistance, and appear to be involved in abnormal metal metabolism in inherited diseases such as leukemia and Menke's disease. In baker's yeast Saccharomyces cerevisiae, copper-metallothionein (Cu-MT), a class II metallothionein, takes part in copper homeostasis and is a major factor in protection against Cu-induced toxicity. Cu-MT has been isolated from yeast as a 53 residue protein (truncated Cu-MT) lacking the first eight amino acids from the N-terminal coded for in the CUP1 locus. Metal binding studies indicated that the isolated protein contains 8 Cu(I) ions bound to the 12 cysteines and that in vitro Ag(I) ions with may also be bound with the same stoichiometry. The truncated form of yeast Cu-MT, overexpressed in E. coli, was also shown to contain 8 ± 2 Cu(I) ions per molecule (Sayers et al., 1993).
We carried out small angle X-ray solution scattering measurements to determine structural parameters of recombinant yeast Cu-MT and model calculations were performed for comparison with the model based on NMR data (Peterson et al., 1996). Small angle X-ray solution scattering experiments were carried out at the EMBL Hamburg Outstation, on the double-focusing monochromator-mirror camera X33 in HASYLAB on the storage ring DORIS of the Deutsches Elektronen Synchrotron (DESY) using quadrant delay line readout detectors and standard data acquisition and evaluation systems. The observation range was 0.05 <s<0.4 nm-1, where s = 4p sin /l, 2 is the scattering angle and l the wavelength (0.15 nm).
The radii of gyration (Rg) for Cu-MT together with those for BSA and lysozyme and the forward scattering I(0) intensities are given in table 1. A comparison of the Rg values reveals that, although the molecular mass of Cu-MT is less than half of that of lysozyme, the radii of gyration of the two proteins are comparable, reflecting an extended structure for Cu-MT in solution. The I(0) value yields a molecular mass of 5.2 kDa for Cu-MT relative to BSA and lysozyme. The maximum dimension of the MT was found to be 4.5±0.5 nm (which also practically coincides with that of lysozyme) and the distance distribution function is typical for an elongated particle. The scattering curve is neatly (c=1.1) fitted by the scattering from an ellipsoid with half axes 2.75:1.04:0.43 nm. These data, taken together, indicate that the conformation of MT in solution is extremely anisometric.
The scattering pattern expected from the 3D structure based on the recent NMR data was calculated for comparison with X-ray measurements. This structure was based on measurements on native and Ag(I) substituted protein and the coordinates given in the PDB lack the last 13 residues from the C-terminal which appear to be disordered (entry 1aoo). The agreement between experimental and calculated curves was found to be poor (c= 2.8) and is likely to be due to the core of the NMR structure which is more compact than what is expected from the X-ray scattering pattern. The sequence-structure matching, carried out for Cu-MT, using threading indicates that the sequence fits into the structure of erythroid transcription factor (PDB entry 1gat) and the best agreement (c= 0.88) with the experimental results was obtained for the case where the C-terminal was fully extended (Sayers et al. 1999).

References:

Peterson, C. W., Narula, S. S. & Armitage I. A. (1996) 3D solution structure of copper and silver-substituted yeast metallothioneins, FEBS Lett. 379, 85-93.
Pountney, D. L., Schauwecker, I., Zarn, J. & Vasak, M. (1994) Formation of Cu8-metallothionein in vitro: evidence for the existence of two Cu(I)4-thiolate clusters, Biochemistry 33, 9699-9705.
Sayers, Z., Brouillon, P., Vorgias, C. E., Nolting, H. F., Hermes, C. & Koch, M. H. J. (1993) Cloning and expression of Saccharomyces cerevisiae copper-metallothionein gene in Escherichia coli & characterization of the recombinant protein, Eur. J. Biochem. 212, 521-528.
Sayers, Z., Brouillon, P., Svergun, D.I., Zielenkiewicz, P. and Koch, M.H.J. (1999). Biochemical and Structural Characterization of Recombinant Copper-Metallothionein from Saccharomyces Cerevisiae. Eur. J. Biochem. 262, 858-865.

Table 1. Structural parameters of Cu-MT determined from Guinier Plots.

7.3 X-Ray Absorption Spectroscopy and Structural Biology
James E. Penner-Hahn University of Michigan

Physical basis of x-ray absorption spectroscopy. X-Ray absorption spectroscopy (XAS) refers to the structured absorption that is found on the high energy side of an x-ray absorption “edge” (the “edge” refers to the abrupt increase in absorption cross-section that occurs when the incident x-ray energy matches the binding energy of a core electron). X-ray excitation results in ejection of a core electron. Interference of this photoelectron as it scatters off of the potential of the surrounding atoms gives rise to structured absorption that can be analyzed to determine the local environment of the absorbing atom. This information, which arises from absorption well above the absorption edge (ca. 50 to ca. 1000 eV beyond the edge), is often referred to as extended x-ray absorption fine structure (EXAFS). In addition to EXAFS, useful structural information is also encoded in the details of the absorption edge. This spectral region (within ca. 50 eV of the absorption edge) is sometimes referred to as the x-ray absorption near edge structure, or XANES region. The energy of the absorption edge can be related to the oxidation state of the metal; higher oxidation states tend to have higher edge energies. In addition, the detailed shape of a XANES spectrum is sensitive to the geometry of the absorbing site. This is typically used as a fingerprint to distinguish between possible geometries. Finally, there of often bound-state transitions (e.g., 1s®3d) that provide direct access to information about the electronic structure of the absorbing atom.

Comparison of XAS to other methods in structural biology. Unlike many other spectroscopies used in structural biology (most notably NMR spectroscopy), XAS does not depend on the presence of particular nuclei or particular spin states, and is thus always observable. Information that can be obtained from x-ray absorption is, however, relatively limited compared to that obtained from protein crystallography or from multidimensional NMR spectroscopy. In particular, EXAFS structural details are limited to near vicinity of absorbing atom (“near vicinity” depends on the details of the absorbing site, but is seldom more than 6 Å from the absorbing atom). Moreover, EXAFS does not, except in very special circumstances, provide anything other than radial structure; three-dimensional structural information is generally not available. EXAFS can be used to identify the neighbors to an absorbing atom to the nearest row of periodic table, and is in this regard comparable to x-ray crystallography. EXAFS can be used to determine the coordination number to approximately 20% or, that is, to ± 1. This information can often be extremely important, as a complement to crystallographic structure determination, since EXAFS can be used to identify the presence of ligands that are disordered (and thus invisible) in a crystal structure. The information that is most well defined by EXAFS is the bond lengths of the nearest, and often next nearest, neighbors to the absorbing atom. The accuracy of EXAFS bond lengths is ca. 0.02 Å, while the precision can be as good as 0.004 Å. This is an order of magnitude better than typically is possible using protein crystallography. Most importantly, EXAFS structural information can be obtained under all conditions, regardless of the state of matter. In particular, it is not necessary to have a crystalline protein in order to determine the local structure of the metal site.

The information that can be obtained from XANES spectra (oxidation state, electronic structure) can sometimes be obtained from other spectral probes (magnetic resonance, UV-visible absorption etc.). However, XANES again has the advantage of being always detectable. Such information is generally not accessible from NMR or crystallography.

Utility of XAS in the Structural Biology Context

Considering the strengths and weaknesses of the different techniques used in structural biology, and the relative information content of XAS and other structural biology probes, it is possible to define three important niches for XAS.

Characterization of unknown sites. NMR is of little use for defining metal site structure, since NMR structures are generally based on proton-proton distances. Moreover, NMR is (at least at present) limited to proteins of ca. 40 kDa or less. Crystallography is, of course, limited to proteins that can be crystallized into diffraction quality crystals. Although great strides have been improving crystallization techniques, there remain many interesting proteins which have not been crystallized. Consequently, there are many proteins for which XAS is the only available structural probe. For metalloproteins, where the most interesting structural questions concern the metal site, XAS can provide unique information. Current estimates are that approximately 1/3 of all proteins are metalloproteins, and that the human genome contains approximately 30,000 genes. There are, then, at least 10,000 candidates for XAS structural characterization. The majority of these proteins have not been crystallized. In addition to these proteins, there are also numerous examples of enzymes that have been crystallized in their resting state, but for which no structural information is available regarding the intermediates in the enzymatic reaction. Once again, EXAFS and XANES can be used to determine the structure of intermediate species (often the most interesting forms of an enzyme from a biochemical perspective).

Determination of structural details that are unavailable by crystallography. Even when a protein can be crystallized, it is often not possible to determine the structure of the metal site with sufficient accuracy to permit detailed mechanistic analysis. In such cases, there is much to be gained from combining EXAFS, which gives more accurate bond lengths, with crystallography, which gives detailed three-dimensional structure. The two methods together provide an enhanced description of a metalloprotein. A second important advantage of XAS is that it can provide information about the oxidation state of the metal(s) in a protein. Given the sensitivity of many metalloproteins to x-ray induced photoreduction, direct determination of the oxidation state of a crystal is highly desirable. One can thus imagine that, in the future, synchrotron beamlines will be set up to permit measurement of diffraction data (to define the three dimensional structure) and after completion of the diffraction experiment, measurement of XAS data (to define metal bond lengths and oxidation state). With cryocrystallography, these could be performed on a single crystal. Once again, the potential demand for synchrotron time is quite large, as XAS could prove useful for approximately 1/3 of all crystals that are examined crystallographically.

As a probe of molecular dynamics. In a few very special cases, XAS and diffraction data have been combined to probe the dynamics of a metal site. One of the best examples of such a study is an investigation of the “blue” copper site in plastocyanin. Crystallography had shown that the Cu active site had a methionine sulfur atom as a nearest neighbor. Variable temperature EXAFS was able to show that this sulfur atom did not make any significant contribution to the overall EXAFS spectrum. The explanation for these apparently inconsistent results is that the motion of the Cu and the S atoms is nearly uncorrelated. Such insight, which bears directly on the nature of the bonding between the Cu and the S, could not have been gained from either XAS or diffraction alone. Such studies, while important, are likely to remain a small part of the overall demand for XAS in structural biology.

All three classes of experiment described above have relatively modest demands on the flux, flux density, and brightness of x-ray beams. In particular, most current XAS studies of metalloproteins are not limited by the quality of x-ray beams. Rather, they are limited by sample homogeneity, sample stability (a problem for the most intense synchrotron beams) and by detector quality. Since the hard x-ray beams from the SESAME superconducting wigglers are comparable in performance up to about 20 KeV with bending magnets at the NSLS and wigglers on SPEAR2 and DORIS, SESAME will be a competitive source for such studies.

A word about in vivo studies. One of the most exciting advances in XAS in recent years has been the development of methods for focusing x-ray beams to small (sub micrometer) sizes. This has made it possible to measure spatially resolved XAS spectra for a variety of systems. To date, most such applications have focused on geological samples. However, there are numerous opportunities to extend such studies to biological systems. In such cases, one obtains the same information outlined above, but with the added advantage of being able to probe in situ (for example in a living cell). In such a way, one can produce a two (or even three) dimensional map of the distribution of chemical species in biological tissue. The ultimate impact of such studies on biology remains to be determined. History suggests, however, that every advance in imaging modality has led to large (and often unanticipated) advances in biological understanding.
MicroXAS spectroscopy is an experiment that depends on x-ray brightness, since this is what permits one to focus a large number of x-ray photons into a small spot. Much of the current microXAS work is performed at NSLS X-26 and at SPEAR2. SESAME will have brightness comparable to these facilities, and thus will be competitive for spot sizes down to a few micrometers. The third generation synchrotrons have much brighter beams, and thus will be superior for smaller spot sizes. Nevertheless, there should remain a wide range of samples for which SESAME can provide important microstructural data. The primary limitation to many of these studies is sample preparation and experiment design, not synchrotron capabilities.

7.4 Macromolecular crystallography at SESAME – an Overview

Peter Kuhn SSRL/SLAC/Stanford University

The goal of understanding life has evolved into a large interdisciplinary effort, which integrates information that extends from experimental results at the atomic and molecular level to studies of organelle, cellular and tissue organization and function. There is an increasingly strong interplay among areas that include chemistry, structural biology, biochemistry, genomics, physics, informatics and computational sciences. Atomic level information provides the means through which biological function, and malfunction that leads to disease, will be understood.

Macromolecular crystallography has provided the vast majority of information on three-dimensional biological structure and will play an even greater role in the future, especially on larger and more complex systems. Information relating structure to function has also led to the development and successes of new approaches to drug discovery (structure aided drug design) and forms the basis of structural and functional genomics.

The unique properties of synchrotron radiation (SR) have been shown to play a seminal role in enabling these advances, as first demonstrated by two important classes of protein crystallography studies more than two decades ago;

the tunability of SR allows for a completely new approach based upon multiple wavelength phasing (MAD) to solve the classical phase problem, and
the highly intense and collimated beam enables study of much smaller crystals and to a significantly higher resolution, all in a much shorter time.

Results from synchrotron-based studies have given insights on the molecular level for hundreds of new structures of biological macromolecules, which span a range of biological structure and function.
It has been shown that the SESAME super-conducting wiggler will deliver radiation into a 100-200 micron (at the edge) protein crystal that is comparable to that provided by a bend magnet at NSLS and the best wigglers at SPEAR2 and DORIS. These characteristics enable a scientific program in the Region that will be highly competitive with structural biology programs at synchrotrons around the world. It can address problems of particular interest to the Region’s health and environmental situation as well as of international scientific interest. The design of a SMB Resource at SESAME can benefit from the experiences at other synchrotron sources such as the European Synchrotron Radiation Facility and the Stanford Synchrotron Radiation Laboratory, for which the existing programs were described by Anastassis Perrakis (NKI Amsterdam) and Peter Kuhn (Stanford University) respectively.

7.5 Biological Studies with EXAFS and IR Techniques

Using Synchrotron Radiation for Spectroscopic and Microspectroscopic Studies of the Mode of Action of Metalloproteinases
Irit Sagi, Department of Structural Biology, The Weizmann Institute of Science.

Metastatic disease is responsible for the majority of cancer-related deaths, either directly due to tumor involvement of critical organs, or indirectly due to complications of therapy to control tumor growth and spread. Invasion and metastasis are exceedingly complex processes, and their genetic and biological determinants remain incompletely understood. Malignant tumors express high levels of matrix metalloproteinases (MMPs), which facilitate tumor cell invasion and metastasis by removing the physical barriers to invasion through degradation of the extracellular matrix (ECM) components.

A large amount of experimental evidence points to the role of MMP in tumor formation and metastasis in cancer processes, and specifically in associated human brain tumors, showing that malignant tumors express MMPs more frequently or at higher levels than cells of normal tissues. Major constituents of the basement membranes are the metalloproteins collagen IV and gelatinases A and B (designated as MMP-2 and MMP-9, respectively). Their role in tumor metastasis and angiogeneis is intriguing and recently they have been a target for drug design to block tumor growth and metastasis. We have used X-ray absorption spectroscopy (XAS) to characterize the molecular structures of the metallo-catalytic site of MMP-2 and MMP-9, in their latent and activated forms and when bound to various inhibitors (natural and synthetic). Our results show that XAS can be used to screen Enzyme-Inhibitor interactions at the catalytic sites of MMPs. In addition, we have developed an innovative IR microscopy assay to examine the degradation pathways of MMPs on relevant ECM surfaces and fibers. In this assay we utilize a high-flux, bright IR beam generated from a synchrotron source. Using IR microscopy imaging we were able to detect the mode of MMPs degradation and their directionality in ECM like surfaces. This work involves an international collaboration with Lisa Miller's group in the USA and extensive use of various synchrotron radiation sources.

Next Page