Introduction

The year 1997/8 represents yet another significant step in the development of the beamlines at the ESRF that can be used for research in the Life Sciences. In the area of macromolecular crystallography the Quadriga beamline, ID14, has made good progress with the first side station EH3 being routinely used to collect high quality intensity data and the end station EH4 having undergone successful commissioning. Station EH3 is equipped with both a 130 mm Mar-Research CCD detector and an off-line large robot controlled image plate detector (for use with very large unit cells) and a 2 x 2 CCD detector system manufactured by the Area Detector Systems Corporation has been purchased for station EH4. The construction of the second phase of the Quadriga beamline is well underway with both fixed wavelength stations (equipped with Mar-Research CCD detectors) scheduled for use early in 1999. The bending magnet beamline, BM14, has been used for a number of very successful applications of the Multiple-wavelength Anomalous Dispersion (MAD) method for deriving phase information. Plans are well advanced to move this beamline to an insertion device within the next two to three years to take advantage of an increase in brilliance and a wider accessible wavelength range. These developments will give a significant growth in the capacity to handle data collection for biological macromolecules, which ought, at least in part, to match the ever-increasing program in this area. In addition, beamlines ID9 and ID13 continue to produce new science with respect to time-resolved measurements and very small crystals and fibers respectively. A number of CRG beamlines (of which the ESRF is entitled to one third of the available beam time) have also constructed, or are planning, facilities for macromolecular data collection. Overall the future for macromolecular crystallography at the ESRF is very promising and there will be continued efforts to ensure that the beamlines are equipped with the latest technology, both hardware and software, for efficient and "user friendly" data acquisition.

 

Macromolecular Crystallography

The field of macromolecular crystallography continues to flourish and the three-dimensional structural information that is obtained underpins much of modern molecular biology and medicine. A classic example is the recent research of W. Hendrickson and his colleagues on the structure of an HIV gp120 envelope glycoprotein in complex with the CD4 receptor and a neutralizing human antibody [1]. This research reveals many aspects of how the HIV-1 virus infects cells and this, in turn, may have important implications for virology, immunology and vaccine development. It has been estimated that at present some 60-70% of all new structures being solved is based on intensity data collected at synchrotron radiation sources. The high brilliance, fine collimation, and tunable wavelength characteristics of such sources are of great importance in the collection of high quality, high resolution data from small, weakly diffracting crystals of biological macromolecules. The following paragraphs outline examples of structures which have been solved recently with data collected at the ESRF (certain other important structures are about to be published and cannot be mentioned at present) together with some recent applications of anomalous dispersion techniques.

Reference
[1] P.D. Kwong, R. Wyatt, J. Robinson, R.W. Sweet, J. Sodroski & W.A. Hendrickson. (1998), Nature, 393:648-659.

 

 

Examples of structures recently solved with data collected at the ESRF

Bovine ubiquinol-cytochrome c reductase (cytochrome bc1 complex)

The mitochondrial cytochrome bc1 complex is a respiratory multi-enzyme complex which may also function as a signal peptidase. Previous studies have characterized the transmembrane section and matrix side of the complex [1], but the present study by S. Iwata and his colleagues enables detailed information to be obtained on the intermembrane side. Thus, the functional domains containing the redox centers of cytochrome c1, the "Rieske"[2Fe-2S] protein and the whole of sub-unit 8 have been clearly identified, providing a basis for understanding the respiratory function of the complex.

X-ray data used for this structure determination were collected on beamlines ID2 and ID14/EH3 and some key features of the complex are shown in Figures 1 and 2.

Reference
[1] D. Xia, C.A. Yu, H. Kim, J.Z. Xian, A.M. Kachurin, L. Zhang, L. Yu & J. Deisenhofer. (1997), Science, 277:60-66.

Publication
S. Iwata (a), J.W. Lee (b), K. Okada (a), J.K. Lee (b), M. Iwata (a), B. Rasmussen (c), T.A. Link (d), S. Ramaswamy (e) & B.K. Jap (b). (1998), Science, 281:64-71.

(a) Dept. of Biochemistry, University of Uppsala , (Sweden)
(b) Life Sciences Division, Lawrence Berkeley National Laboratory, University of California (USA)
(c) EMBL Grenoble Outstation (France)
(d) Uniklinikum Frankfurt, Zbc Bioichemie I, Molecular Bioenergetics, Frankfurt/Main (Germany)
(e) Dept. of Molecular Biology, Biomedical Center, Uppsala (Sweden)

 

 

 

Stat3ß homodimer bound to DNA

The Stat proteins (signal transducers and activators of transcription) are a family of eukaryotic transcription factors that mediate the response of the cell to extracellular signals provided by a large number of cytokines and growth factors. When activated by cell surface receptors or associated kinases, the Stat proteins dimerize, translocate to the cell nucleus and bind to specific promoter sequences of their target genes. A team at the EMBL (Grenoble Outstation) led by C. Müller have elucidated one of the first crystal structures of a Stat protein in complex with its DNA binding site. All X-ray data were collected from crystals frozen at 100K using beamlines ID2, BM14 and ID14/EH3. Figure 3 is a ribbon diagram of the Stat3ß homodimer DNA complex. The structure provides insights into the various steps by which Stat proteins transmit the response signal directly from the cell membrane to the target gene in the nucleus.

Publication
S. Becker (a), B. Groner (b) & C.W. Müller (a). (1998), Nature, 394:145-151.

(a) EMBL Grenoble Outstation, (France)
(b) Institute for Experimental Cancer Research, Tumor Biology Center, Freiburg i Br. (Germany)

 

 

The complex between a monoclonal antibody fragment Fab and HIV-1 capsid protein, p24

In the mature Human Immunodeficiency Virus (HIV, the causative agent of AIDS) the capsid protein p24 forms the conical shell surrounding the nucleocapsid within the viral membrane. The shell is formed in a maturation step, following specific HIV protease cleavage of the Gag polyprotein of which p24 is the central part between the matrix protein and the nucleocapsid protein. p24 has a clear structural role, but is also important in assembly and maturation and due to its specific interaction with cyclophilin A (a proline cis-trans isomerase). S. Cusack and his colleagues have solved the structure of a p24:Fab complex using only two data sets. Initially the microfocus beamline ID13 enabled the collection of a partial data set (80% complete) to 3.8 Å resolution from very thin crystals (30 µm thick) with an associated twinning problem. A second data set to 3.0 Å was then collected from a slightly larger crystal on ID14/EH3 using a Mar-Research CCD detector. The structure reveals for the first time the complete p24 molecule, which indeed comprises two helical domains linked by a flexible peptide. More surprisingly, as seen in Figure 4, p24 in the crystal structure forms head-to-tail dimers. The monoclonal Fab fragment recognizes an epitope on the C-terminal domain.

Publication
C. Berthet-Colominas (a), S. Monaco (a), A. Novelli (b), F. Mallet (b) & S. Cusack (a).

(a) EMBL Grenoble Outstation (France)
(b) BioMérieux, Marcy L'Etoile (France)

 

 

Ribosomal protein S7

Protein S7 is one of the components found in the ribosome, the ribonucleotide complex that performs the central role in protein biosynthesis. It is located at the head of the small (30S) ribosomal subunit and faces into the decoding center where the message carried by the mRNA is used to synthesize the corresponding protein. S7 is one of the primary 16S rRNA-binding proteins responsible for initiating the assembly of the head of the 30S sub-unit. The structure was solved using multi-wavelength data collected on BM14 [1]. The molecule comprises a helical hydrophobic core and a ß-ribbon arm extending from the core. This structure has now been used to simulate the interaction of the S7 protein with the three-dimensional model structure of the 16S ribosomal RNA [2] as shown in Figure 5.

Publications
[1] H. Hosaka (a), A. Nakagawa (a), I. Tanaka (a), N. Harada (b), K. Sano (b), M. Kimura (b), M. Yao (c, d) & S. Wakatsuki (c). (1997), Structure, 5:1199-1208.

(a) Div. of Biological Sciences, Graduate School of Science, Hokkaido University , Sapporo (Japan)
(b) Lab. of Biochemistry, Faculty of Agriculture, Kyushu University, (Japan)
(c) ESRF
(d) MAC Science, Kohokuku, Yokohama, Kanagawa (Japan)

[2] I. Tanaka (a), A. Nakagawa (a), H. Hosaka (a), S. Wakatsuki (b), F. Mueller (c) & R. Brimacombe (c). (1998), RNA, 4:542-550.

(a) Grad. Sch. of Science, Hokkaido University, Sapporo (Japan)
(b) ESRF
(c) Max-Planck-Institut für Molekulare Genetik, Berlin, (Germany)

 

 

Troponin C in complex with a troponin I fragment

Troponin plays a central role in the calcium dependent regulation of muscle contraction. It is a complex of three subunits, TnC, TnI and TnT, and is associated with the actin filaments in the muscle fibers. The crystal structure of the TnC subunit, with two Ca2+ bound, in complex with the

N-terminal fragment of TnI has been solved by a combination of multi-wavelength anomalous dispersion and isomorphous replacement techniques using data collected on the protein crystallography beamlines at the ESRF. The structure of the complex, Figure 6, suggests a mechanism for the way that troponin regulates the mechanism of muscle contraction. A unique -helical segment of TnI, which binds to the N-lobe of TnC in its calcium bound conformation, may play a key regulatory role.

Publication
D.G. Vassylyev (a), S. Takeda (a), S. Wakatsuki (b), K. Maeda (a) & Y. Maeda (a). (1998), Proc. Natl. Acad. Sci. (USA), 95:4847-4852.

(a) International Institute for Advanced Research, Central Research Labs., Matsushita Electric Industrial Co. Ltd., Kyoto (Japan).
(b) ESRF

 

The use of anomalous dispersion techniques

The first major task in a crystal structure determination is to obtain single crystals suitable for X-ray analysis. The second is to obtain amplitude and phase information for the individual reflections in the X-ray diffraction pattern so that an image can be synthesized. The amplitudes can be obtained directly from the diffraction intensities, but the phase information normally has to be accessed by indirect means. In this regard anomalous dispersion methods, exploiting the wavelength tunability of a synchrotron source, are playing a more and more important role. It is expected that at least 30% of all new structures solved in the future will use the MAD technology. An important aspect that has enhanced the applicability of the method has been the development of molecular biological techniques whereby the sulphur atoms of methionine residues can be substituted by selenium during expression of the protein or enzyme. The presence of the "heavy atom" selenium and a suitable choice of wavelength give rise to anomalous and dispersive differences that can be measured with precision.

 

 

Structures solved by MAD on BM14, January - June 1998

Beamline BM14 at the ESRF was explicitly designed for multi-wavelength measurements and has had some notable successes over the last year. Table 1 lists groups which successfully collected MAD data on BM14 during the period January ­ June 1998. A distinction is drawn between groups where data collection resulted in a high quality electron density map and those where a model had already been built and the structure was being refined. The table also gives the number of heavy atom sites per molecule, the molecular weight of the protein, and the number of molecules per asymmetric unit.

 

ModE protein from E. coli

The potential of BM14 for extremely rapid de novo structure determination using multi-wavelength anomalous dispersion techniques is illustrated by the recently determined structure of the ModE protein from E. coli carried out in collaboration with the group of W.N. Hunter (University of Dundee, UK). ModE is involved in the regulation of uptake of molybdenum by cells and a selenomethionyl derivative of the protein (6 Se per 56 kDa) had been crystallized in the orthorhombic system with unit cell dimensions, a = 81.6 Å, b = 127.2 Å, c = 63.0 Å and space group P21212. A three-wavelength MAD experiment was undertaken with data collected using the Princeton-ESRF image intensified CCD detector. To ensure as complete as possible measurement of Friedel/Bijvoet pairs, 135° of data in 1.0° oscillations were collected at each of the wavelengths, a total of 405° of data. Despite the large amount of data and the extremely high speed of the data collection (one oscillation image every 20 s), it was possible to process the data "on the fly", and show that they were of high quality. Determination of the Se heavy atom positions was affected using Direct Method techniques and phasing was undertaken with the program MLPHARE of the CCP4 program suite [1]. The resulting electron density map, a section of which is shown in Figure 7, was fully interpretable in terms of the polypeptide chain. The time difference between the start of the data collection and the production of the map was a mere 7 hours. The ModE protein, together with a similar problem solved in 11 hours, currently sets the standard for speed of structure determination on BM14, although it is easy to imagine that this could be lowered considerably for cases of suitably diffracting crystals of higher symmetry.

Reference
[1] Collaborative Computing Project, Number 4. (1994), Acta Cryst., D50:760-763.

 

Fe-Ni hydrogenases

Another use of anomalous dispersion techniques is represented by the work of E. Garcin and colleagues at the IBS, Grenoble, France, and concerns the structures of Fe-Ni hydrogenases [1]. Hydrogenases are found in a large number of bacteria and consume hydrogen according to the reaction: H2 <=> 2H+ + 2e-. In the case of sulphate reducing bacteria the electrons produced are then used in the conversion of sulphite (SO32-) to sulphur in the form of hydrogen sulphide. The [Fe-Ni] hydrogenase isolated from Desulfovibrio gigas comprises two domains [2]. The smaller has a chain of three iron-sulphur aggregates (two [4Fe-4S] clusters and an intermediate [3Fe-4S] cluster) which are involved in electron transfer, and the larger domain contains the active site. Earlier studies had shown the presence of two metal atoms at the active site, one of which was known to be a nickel atom. Intensity data were collected on either side of the Fe K-edge on BM14, and combined with a data set collected at 0.98 Å. The subsequent computation of anomalous difference, double difference and dispersive difference electron density maps, unambiguously showed that the second element of the active site was an iron atom. The presence of the iron was unexpected and a study of the enzyme in both the oxidized and reduced forms has enabled a rational explanation for the catalytic mechanism of this type of hydrogenase.

These studies were extended to the [Fe-Ni-Se] hydrogenase from Desulfomicrobium baculatum. In this enzyme the intermediate [3Fe-4S] aggregate in the small sub-unit is replaced by a [4Fe-4S] cluster, and at the active site one of the cysteine residues bound to the Ni is replaced by a seleno-cysteine. Figure 8 shows an anomalous difference Fourier synthesis computed with all data between 8.0 Å and 2.15 Å and using model phases; data were collected on beamline D2AM (a French CRG beamline at the ESRF) at a wavelength of 0.98 Å. The synthesis clearly shows the presence of the three [4Fe-4S] moieties and the Fe-Ni active site, but there also appears to be another mononuclear metal site. The possibility that this extra site corresponds to a Zn or Mn atom was eliminated by collecting data either side of the K-edges of these metals and observing no signal in anomalous difference syntheses; the site has since been shown to be an additional iron atom. The presence of the selenium atom was also confirmed from anomalous difference maps computed using data collected on either side of the Se K-edge.

These studies illustrate the elegant way that the wavelength tunability characteristic of a synchrotron source can be exploited to help to unravel the complicated chemistry involved in catalysis reactions involving metalloproteins.

References
[1] E. Garcin. (1998), Doctoral Thesis, Université Joseph Fourier, Grenoble, France.
[2] A. Volbeda, C. Piras, M.H. Charon, E.C. Hatchikian, M. Frey, & J.C. Fontecilla-Camps. (1995) Nature, 373:580-587.

Publication
A. Volbeda (a), E. Garcin (a), C. Piras (a), A.I. de Lacey (a), V.M. Fernandez, E.C. Hatchikian (a), M.C. Frey (a) & J.-C. Fontecilla-Camps (a). (1997), J. Amer. Chem. Soc., 118:12989-12996.

(a) Laboratoire de Cristallographie et de Cristallogenèse des Protéines, Institut de Biologie Structurale, Grenoble (France)

 

 

 

Fiber Diffraction : Micro Small-Angle Scattering on Biopolymers

Micro X-ray small angle scattering (SAXS) methods have been developed at the ID13 beamline in order to study the large scale organization of biopolymers above the level of the unit cell. At the present stage, a limit in the s-value (s = 1/d) of about 0.05 nm-1 can be reached using a beam of about 3 µm diameter [1], but additional improvements in the resolution can be expected as instrument development progresses. Compared to electron diffraction techniques, cutting by a microtome can be largely avoided and samples can be studied in a hydrated state.

In 1997, experiments on biopolymers concentrated on carbohydrates. This class of molecules includes both sugars and polysaccharides such as starch and cellulose [2]. Starch is the main source of stored energy in higher plants and is also the major carbohydrate nutrient for humans and animals. Cellulose is the most abundant biopolymer since it is used as a structural material by plants.

Figure 9a shows the SAXS pattern obtained by reflection from a small region (approximately 4 µm) within a single starch granule ( 80 µm). The structure is lamellar (9 nm thick lamellae). The SAXS pattern agrees in principle with a model of a superhelix built up from amylopectin helices as the basic building blocks, Figure 9b [3,4]. There is very good agreement for the observed weak four-point pattern (Figure 9c) [5] with a model based on super-helical lamellae with cumulative distortions along the z-axis. Future work will consider the stability of such structures both under genetic modification of the parent plant, and following different processing treatments.

Figure 10a shows the SAXS pattern of a 20 µm thick, single cellulose fibre extracted from flax stems [6]. The continuous streak can be related to the arrangement of crystalline microfibrils along the fibre axis. Analysis suggests that the distribution of the tilt angles of these microfibrils is ± 3.5°. This value is remarkably low compared to synthetic polymers. The small beam size allows scanning SAXS experiments across the fibre diameter. Figure 10b shows that the distribution of tilt angles is constant across the fibre. Experiments on fibre bundles resulted in significantly higher values, presumably due to a misalignment of individual fibres. This result opens the way to investigate texture effects of the crystalline microfibrils in cell walls and fibres.

References
[3] G.T. Oostergetel & E.F.J. van Bruggen. (1993), Carbohydrate Polymers, 21: 7-12.
[4] D.J. Gallant, B. Bouchet & P.M. Baldwin. (1997), Carbohydrate Polymers, 32:1777-191.

Publications
[1] C. Riekel (a), P. Engström (b) & C. Martin (c), (1998), J. of Macromol. Sci. - Phys., B37:587-599.

(a) ESRF
(b) KCK, Chalmers University of Technology, Göteborg (Sweden)
(c) Dept. of Polymer Physics, University of Keele (UK)

[2] A.M. Donald. (1994), "Physics of foodstuffs" in "Reports on Progress in Physics", 57:1083-1133.

[5] T.A. Waigh (a), A.M. Donald (a), F. Heidelbach (b), C. Riekel (b) & M.J. Gidley (c), (1998), Biopolymers, in press.

(a) Polymers & Colloids, Dept. of Physics, Cavendish Laboratory, Cambridge (UK)
(b) ESRF
(c) Unilever Research Colworth, Sharnbrook, Bedford, (UK)

[6] M. Müller (a), C. Czihak (b), G. Vogl (c), P. Fratzl (d), H. Schober (e) & C. Riekel (a). (1998), Macromolecules, 31:3953 ­ 3957.

(a) ESRF
(b) ILL Grenoble & Institut für Materialphysik der Universität Wien (Austria)
(c) Institut für Materialphysik der Universität Wien (Austria)
(d) Erich Schmid Institut of the Austrian Academy of Sciences, University of Leoben (Austria)
(e) ILL, Grenoble (France).

 

 

 

Progress on the Medical Beamline

Research activities on the Medical Beamline, ID17, cover three main areas, angiography, computed tomography, and micro-beam radiation therapy. Although some of these activities are currently more advanced at other European synchrotron sources, for example, the angiography program at DESY in Hamburg, considerable progress has been made over the past year. Part of this success has been due to a fruitful collaboration between a group of physicians and physicists attached to the Centre Hospitalier Universitaire (CHU) of Grenoble and the ESRF beamline team.

 

 

Angiography

Coronary angiography is an established procedure for examining the arteries of patients with cardiovascular diseases. However, the procedure is invasive and efforts are therefore underway to find some other means of visualizing the coronary arteries with far less invasive methods. Magnetic Resonance Imaging (MRI) and Fast Computed Tomography (CT) are also aiming at this goal, but they are both currently limited by the movement of the heart arteries during the measurement period.

The synchrotron method is based on the energy subtraction of two images acquired with X-ray energies either side of the K absorption edge of a contrast agent such as iodine. The logarithmic subtraction of the two images enables the location of the iodine to be isolated, thus allowing the physician to detect stenoses in the arteries. Only synchrotron radiation can provide the monochromatic flux necessary to visualize arteries of 1.0 mm diameter after intra-venous injection of the contrast agent.

This joint project between the CHU and ESRF teams is entering the final phase of commissioning; the instrumentation is complete and tests have been performed with a coronary artery phantom, Figure 11 under real imaging conditions [1]. The flux was set so that each pair of images would result in an absorbed dose of 20 mGy. The iodine concentration in the arteries was 10 mg/cm3. An attenuation comparable to that of a typical patient was obtained by adding pork meat, bones and plexiglass around the phantom. The phantom was placed in the patient chair and images recorded with the same instrumentation that will be used for a patient, (e.g. same monochromator, patient positioning system and data acquisition system).

In the most recent studies carried out in September 1998, angiograms were recorded from pigs. A typical example is shown in Figure 12. These very successful studies were an important prerequisite for the medical program involving human patients and the first patient run is scheduled early in 1999.

 

 

Computed tomography applied to brain pathologies

Nowadays, considerable effort in medical research is concentrated in extending the usefulness of medical imaging techniques from anatomic mapping to functional and quantitative information such as blood volume. Despite giving excellent spatial and temporal resolution, the Computed Tomography (CT) scanning method has seen comparatively few recent developments. Nevertheless, it remains the only quantitative technique available with the same scale everywhere (Hounsfield units) and leading to absolute concentrations. Such properties justify investigating the use of the method with synchrotron radiation [2] and a CT system for clinical studies in brain pathology is part of the medical beamline project [3]. The main question to be studied is whether it is possible to obtain brain tomograms which selectively map the absolute concentration of contrast agents (for example, gadolinium and iodine) at clinically acceptable doses by K-edge subtraction measurements with a synchrotron X-ray source. Concentration measurement of intravenous contrast agents, i.e. accessing blood distribution, is one of the central objectives of this program to characterize cerebral blood volume. Such a method should advance the state of the art of diagnosis, particularly for brain tumors, and lead to improvements in basic knowledge in the field of neurosciences.

Experiments are carried out by using a fan-shaped X-ray beam of 1.5 mm height and 120 mm width and a subject rotating around a vertical axis. Attenuation is measured with a linear array germanium detector with 432 elements each of 0.35 mm width. This technique uses the same absorption edge subtraction method as is used for coronary angiography at the K-edges of iodine (33.17 keV) and gadolinium (50.24 keV). The resulting subtracted image is a quantitative map of the iodine or gadolinium distribution with a higher contrast resolution than a conventional CT image [2].

Preliminary experiments carried out on phantoms [4] show that 1 mg/cm3 (± 10%) of iodine or gadolinium is quantifiable in a 6 mm diameter detail and 2 mg/cm3 (± 10%) in a 3 mm diameter detail with geometry and dosimetry parameters close to conventional CT (surface dose < 50 mGy). At the present time, effort is aimed at validating and optimizing the experiment and protocols before starting a clinical program. As an example, Figure 13 shows images obtained above and below the iodine K-edge from a sheep's head (bought at a butcher's shop!). The subtracted image reveals several tubes filled with solutions of different concentrations of iodine (5.0, 2.5, 1.0, 0.5, 0.25, 0.125 mg/cm3). Measured values were found to be 5.01, 2.54, 1.14, 0.46, 0.26, and 0.28 mg/cm3 respectively, which apart from the result for the lowest concentration, are in good agreement with the expected values.

At the present time, in vivo measurements in rats bearing glioma are carried out at the medical beamline in order to quantify cerebral blood volume and to study angiogenesis and blood brain barrier rupture in tumor growth. All the results are correlated to MRI and histology at the CHU, Grenoble. Such preclinical studies are of great interest since they provide quantitative information about the distribution of contrast agents for MRI or X-rays.

 

 

Microbeam radiation therapy

It is well known that the severity of normal tissue damage after the irradiation of tumors with therapeutically effective absorbed doses diminishes as the irradiated volume is decreased. It has recently been shown that when the radiation field is made up of arrays of synchrotron X-ray beams with cross sections that are microscopic, in at least one dimension (of the order of 25-50 µm), the threshold absorbed doses for tissue necrosis in the normal rat brain are 10-100 times higher than for the case of continuous field irradiation. This is thought to be due to the fact that the spatial micro-fractionation of the absorbed dose prevents the destruction of the vasculature, since the radiation damage to microscopic volumes can be rapidly repaired by surrounding tissues. This led to the hypothesis that crossfiring of microbeam arrays on a lesion could be used as a new radiotherapy method, Microbeam Radiation Therapy (MRT), with less normal tissue damage than conventional techniques using seamless radiation fields [5, 6]. Experiments on rats bearing 9L gliosarcoma have lent some credence to the MRT concept [7]. If proved true, MRT would be of particular value for treatment of some brain tumors in children.

MRT requires X-ray beams with microscopic dimensions with very well defined geometric margins. The only source that can be used today to generate suitable micro-beams is a synchrotron X-ray source. This prompted a group of European medical and medical physics scientists to propose the development of a beamline for MRT research at the ESRF. As a result a small animal irradiation facility for preclinical studies was commissioned as part of the medical beamline, and is now in frequent use.

The extremely well defined margins of synchrotron X-ray micro-beams, that are a prerequisite for a successful development of MRT, are clearly illustrated in Figure 14, one result of a study of loco-regional cytopathological effects of microplanar irradiation of the transiently immobilized pupa stage of the fruit fly, Drosophila [8]. The radiogenic lesions resulting from 4000 Gy in the compound eye of a adult fly exposed at 48 hours of pupal development are easily discernible in a semi-thin section prepared for transmission electron microscopy. Note the remarkably precise delimited aisles, carved through the retinal epithelium by the micro-planar beams. Even where half of ommatidia are destroyed, the remaining part of the organelle seems unaffected. Spaces of 75 micrometers between the necrotic area exhibit no signs of subacute effects. The precision of the cutting is revealing particularly for consideration of using synchrotron X-ray microbeams in functional neurological radiosurgery.

The high threshold absorbed doses for cellular and tissue necrosis observed in the normal rat brain after micro-beam irradiations have been verified using ESRF X-ray beams. Since MRT's advantage over other radiotherapy techniques would be an increased normal tissue tolerance as a result of the spatial microfractionation of the absorbed dose distribution in the normal tissues, studies of normal tissue tolerance have so far been prioritised. Recently experiments designed to investigate whether or not 12 days old suckling rats tolerate micro-beam doses in the range 100-600 Gy have been performed. The doses were delivered to the posterior fossa of the rats and these animals are now being studied for late complications, such as abnormal development of body and brain, or behavioural changes including neurological signs. Since this is a study of long term effects it is premature to draw definite conclusions. However, early observations seem to support the MRT idea.

This article was written by Per Spanne who tragically lost his life in the Swissair crash in September 1998. Per was one of the originators of the MRT technique and it is intended to continue this project as a collaboration between the ESRF and J.L. Laissue (Univ. of Berne) and colleagues.

References
[2] F.A. Dilmanian, X.Y. Wu, E.C. Parsons, B. Ren, J. Kress, T.M. Button, L.D. Chapman, J.A. Coderre, F. Giron, D. Greenberg, D.J. Krus, Z. Liang, S. Marcovici, M.J. Petersen, C.T. Roque, M. Shleifer, D.N. Slatkin, W.C. Thomlinson, K. Yamamoto & Z. Zhong. (1997), Phys. Med. Biol., 42:371-387.
[5] D.N. Slatkin, P. Spanne, F.A. Dilmanian, and M. Sandborg. (1992), Medical Physics 196:1395-1400.
[6] D.N. Slatkin, P. Spanne, F.A. Dilmanian, J.O. Gebbers, and J.A. Laissue. (1995), Proc Natl Acad Sci. (USA), 92:8783-8787.

Publications
[1] J.-F. Le Bas (a), F. Esteve (a), B. Bertrand (a), G. Le Duc (b), A.-M. Charvet (c), H. Elleaume (a).

(a) Centre Hospitalier Universitaire, Unité d'Imagerie par Résonance Magnétique, Grenoble (France)
(b) ESRF
(c) University Joseph Fourier, Grenoble (France)

[3] A.M. Charvet (a), J.F. Le Bas (b), H. Elleaume (b), C. Schulze (c), P. Suortti (c) and P. Spanne (d). Medical applications of synchrotron radiation at the ESRF in "Les comptes-rendus de l'Ecole Internationale de Physique Enrico Fermi", pp. 355-377 (eds: E. Buratini and A. Balerna, IOS press, Amsterdam, 1996).

(a) University Joseph Fourier, Grenoble (France)
(b) Centre Hospitalier Universitaire, Unité IRM, Grenoble (France)
(c) ESRF
(d) Medical Department, Brookhaven National Laboratory, New York (USA)

[4] A.M. Charvet (a), C. Lartizien (b), F. Estève (c), G. Le Duc (b), A. Collomb (a), H. Elleaume (d), S. Fiedler (b), A. Thompson (b), T. Brochard (b), U. Kleuker (b), H. Steltner (b), P. Spanne (b), P. Suortti (b) & J.F. Le Bas (c). Synchrotron Radiation Computed Tomography Applied to the Brain: Phantom Studies at the ESRF Medical Beamline. In "Medical applications using synchrotron radiation" (eds: Springer Verlag, Tokyo, 1998), in press.

(a) University Joseph Fourier, Department. of Physics, Grenoble (France)
(b) ESRF
(c) University Joseph Fourier and Centre Hospitalier Universitaire, Grenoble (France)
(d) Centre Hospitalier Universitaire, Grenoble (France)

[7] J.L. Laissue (a), G. Geiser (a), P. Spanne (b), F.A. Dilmanian (c), J.-O. Gebbers (d), M. Geiser (a), X.Y. Wu (c), M.S. Makar (c), P.L. Micca (c), M.M. Nawrocky (c), D.D. Joel (c) and D.N. Slatkin (c). (1998), Intl. J. Cancer, in press.

(a) University of Bern (Switzerland)
(b) ESRF
(c) Brookhaven National Laboratory (USA)
(d) Kantonspital, Luzern (Switzerland)

[8] P.M. Schweizer (a), P. Spanne (b), J.A. Laissue (c), and U. Jauch (a), (1998), Microplanar synchrotron X-ray beam irradiation effects in Drosophila melanogaster, to be submitted.

(a) University of Zurich (Switzerland)
(b) ESRF
(c) University of Bern (Switzerland)

 

Conclusion

A third RF acceleration unit has been successfully built and put into operation on the ESRF storage ring. It was designed entirely by ESRF staff in order to provide and ideal match to the operation constraints and specificities, and to fit into the ESRF environment, taking into account the experience gained on the existing RF units. Performing such design and construction work has been very beneficial in broadening the knowledge and expertise of the RF group personnel. The project was completed within its budget and on schedule, and was managed so as to minimise inconvenience to ESRF users. This was a challenging issue for a machine which delivers 5600 hours of X-ray beam per year.

 

 

 

X-ray imaging : very high resolution phase micro-tomographic imaging of samples of mouse bone

 

The study of bone architecture is particularly important in research on osteoporosis. For this purpose, Synchrotron Radiation Computed Micro-tomography (SR CMT) is particularly attractive since it may provide three-dimensional images of bone samples with high signal to noise ratios [1]. Due to the monochromaticity of the synchrotron radiation beams that can be achieved at the ESRF, these images may be interpreted as accurate maps of the three-dimensional distribution of linear attenuation coefficients within the sample. Furthermore the characteristics of synchrotron radiation allow both a spatial resolution higher than when using conventional X-ray sources and additional new features, such as phase imaging; such features are very useful for weakly absorbing samples.

The parallel beam micro-tomography station developed at ID19 was used in the present work. The detector consists of a screen converting X-rays to light (a Gd02S2:Tb screen or YAG:Ce crystal [2]), using light optics and the FRELON CCD camera developed by the ESRF Detector Group. The choice of the optics allows pixel sizes in the recorded images within the range, 10 µm down to 0.7 µm. The sample is the metatarsal bone of a seventeen day old fetal mouse prepared for a space experiment (sample from Dr. J. Van Loon). Due to the low absorption content of the sample, the images were recorded using the phase contrast mode, at 18 keV, with the two-dimensional detector set at 35 cm from the sample. The size of the cube volume elements in the three-dimensional image is 1.8 µm. Figure 15a represents a three-dimensional display of the embryonic bone surface obtained by ray-tracing. Figures 15b, 15c and 15d show three orthogonal 1.8 µm thick slices of the sample. From a detailed inspection of the volume, the characteristic features of bone growth may be identified. Although the size of the mineralized cartilage network is very small in this sample, it can be clearly visualized.

These images illustrate the potentialities of CMT with synchrotron radiation for the analysis of trabecular bone architecture. The spatial resolution achievable is especially well suited to experiments using rats or mice with a view to understanding the mechanisms of osteoporosis or evaluating the effects of various treatments.

Publications

[1] F. Peyrin (a, b), P. Cloetens (a), W. Ludwig (a), M. Salome (a, b), J. Van Loon (c), J.P. Veldhuÿzen (c), J. Baruchel (a).

(a) ESRF
(b) CREATIS, INSA 502, Villeurbanne, (France)
(c) Dutch Experiment Support Centre, Acta ­ 0CB, Amsterdam (The Netherlands)

[2] A. Koch, C. Raven, P.Spanne & A, Snigirev. (1998), J. Opt. Soc. Amer. A, in press.

ESRF

Although macromolecular crystallography currently dominates the life sciences program, other sections continue to develop. One area where development has been slower than might have been anticipated is that of small-angle scattering. Although innovative experiments have been undertaken with respect to the problem of muscle contraction, relatively few applications on macromolecules in solution, or their complexes, have been attempted. This has been partly due to the lack of a dedicated beamline in the area of soft-condensed matter, but the completion of ID14, may well enable ID2 to be dedicated to such studies. Facilities for X-ray Absorption Spectroscopy (XAFS) have been significantly increased during the year and ID26 (ultra-dilute samples and very fast XAFS) is now available to the biological community. Spectra of samples with relatively high concentrations (in excess of 1.0 mM), which took several hours to record at other synchrotron radiation sources, can now be recorded in minutes and there is the possibility of recording spectra from samples where the concentration is as low as 50 µM. Several other beamlines can also be used for Life Sciences applications, including ID19 (Imaging and X-ray Topography), ID22 (X-ray Fluorescence Microprobe), ID21 (X-ray Microscopy), and the Medical Research beamline, ID17. There are opportunities at the ESRF to undertake complementary experiments on a particular problem or to use the individual techniques themselves on a wide range of different biological problems. These opportunities are at the disposal of the European community and effective usage of them should enable Europe to stay at the forefront of world structural biology.

This section concerning the Life Sciences program at the ESRF does not attempt a comprehensive description of the numerous scientific highlights. A few examples have been chosen which demonstrate the quality of the research being undertaken, indicate future opportunities in the area of medical research, and which will hopefully inspire new generations of users.