- Email: [email protected]
The protein crystallography beamline at LNLS, the Brazilian National Synchrotron Light Source
NUCLEAR INSTRUMENTS
LMEYHoos iN PHYSICS EISBVLBR
Nuciedr
Instruments
and Methods
in Physics
Research
A 405 (1998) 159-164
The protein crystallography beamline at LNLS, the Brazilian National Synchrotron Light Source G. Olivab, E.E. Castellanob, R.C. Garrattb, P. Arruda’, A. Leite”, A. Craievich”
I. Polikarpov”.“,
a Labo~ut~~io Nocio~~~ de Lx S~~~c~otro?~. CWPq. C.P. 6192, CEP 13~~~-~~~~. Campi~ffs, SP, Bra5f ~fntituto de Fisicu de Scio Carfos, U~ftle~sfdade de SZo Paufo, C.P. 369, 13560-970 Sao Cat&s, SP, Brazil ’ Cenrro de Biologfa Molecular e Engenhariu Genhtica, ~lnf~ersidade Estadt~al de Campisas. C.P. 6109. CEP 13083-970. Campinas. SP. &&I Received 24 April 1997; received in revised form I2 August
1997
Abstract
The Brazilian National Synchrotron Light Laboratory - LNLS, will have a dedicated protein crystallography beamline. The beamline under construction indudes cylindrical mirror and bent crystal monochromator focusing the high flux of synchrotron radiation in, the horizontal plane at the position of the sample. The monochromatic radiation will be tuneable between 2.0 and 1.0 A with the optimum wavelength at 1.3-I .6 A, chosen with the aim of maximizing the photon flux from the bending magnets of the storage ring (1.37 GeV). Diffraction images will be recorded on a commercial image plate detector system with on-line readout. The beamline set-up will include cooler/chiller for the samples and biochemical lab for crystallization, heavy-metal soaks, crystal storage and mounting at 22.‘C and 4-C. will also be available. The facility, intended to serve the national and international community, is planned to be brought into operation in the second half of 1997. It is foreseen that the commissioning of the first protein crystallography beamline in Latin America will boost the number of protein structures determined locally and will increase the general interest of the molecular biology and biochemical research community of Brazil in this area. ‘(2 1998 Elsevier Science B.V. All rights reserved. PAL’S: 07.85.Qe; 29.2O.Lq Qrwurds:
Protein crystallography;
Beamline; Brazilian National Synchrotron
1. Introduction
Synchrotron radiation with its high flux and brilliance combined with tunability over a wide
*Corresponding
author.
E-mail: [email protected].
0168-9~~~98~~19.~ :(:’ 1998 Elsevier Science R.V. All Pff SO I ~~-9002(97)0 1202-3
Light Source
wavelength range make it ideal for applications in protein crystallography. The new generation of synchrotron radiation sources has shown two complementary design characteristics: high-energy rings (e.g. ESRF, APS, Spring-g) with very high flux and wide spectrum of radiation available and medium-energy rings (l-l.SGeV, CAMD, MAX II,
rights reserved.
LNLS, ALS) optimized for the soft X-ray region, with a common feature of envisaging the extensive use of insertion devices to optimize the spectrum and flux of radiation to meet the needs of specific applications. The design specifications are frequently delimited by the available feuding, particularly considering that the cost of the synchrotron radiation facility rapidly grows with the energy of the ring. Other considerations can be the regional insertion of the laboratory, with spin-offs in scientific and technological development. Still, this new generation of medium-energy rings can provide a useful X-ray photon fluxes for most protein crystallography applications. With this goal, a protein crystallography beamline has been designed and is currently being commissioned at Laboratorio National de Luz Sincrotron /LNLS (Campinas, SBo Paulo State). This is a bending-magnet beamline and if includes a cylindrical elastically bent mirror providing vertical focusing, bent crystal monochromator, focusing synchrotron radiation in horizontal plane at the position of the sample, two sets of slits and two beam monitors. Single-axis goniometer and the image plate detector will be placed on an XY translation stage resting on a 2&arm. The mirror and monochromator optics and 20-arm were designed to cover a spectral range between 1.0 and 2.0 A. All mechanical movements will be remotely controlled. In this paper a brief description of the beamline is given. The main parameters of LNLS synchrotron radiation source are given in Section 2, the X-ray optics is described in Section 3 and some of the current scientific projects in Section 4.
2. The LNLS synchrotfon radiation source The energy of electrons inside the ring was planned to be 1.15 GeV and bending-magnet magnetic field 1.37 T, which gives a critical photon wavelength of 10.0 A. The maximum electron current is 100mA and average ring diameter 28 m. Magnetic field of constructed dipoles exceeded previously planned parameters and reached 1.64 T. That allowed for increase of the energy of electrons to 1.37_GeV and changed the critical photon energy to 6.0 A. In order to additionally increase the useful
photon energy range toward shorter wavelengths, an insertion of a 6-7 T superconducting wiggler in a straight section prior to the bending magnet # 3 is planned. The critical wavelength associated with the spectrum of the photons emitted by the wiggler will be 1.3-1.5 A, close to the medium of the wavelength range covered by the beamline. The source size is 0, = 0.295 mm, crT,= 0.120 mm, a: = 0.37 mrad, c$. = 0.096 mrad at energy of the ring 1.37 GeV. 3. The optical system for protein crystallographic studies The schematic view of the protein crystallography beamline is given in Fig. 1. The beamline will be placed at the bending magnet #3 and will accept synchrotron radiation emerging at an angle of 15 to the direction of the propagation of the electrons. The X-ray optics of the beamline includes a elastically bent mirror, two sets of slits (one before mirror and one before monochromator) and triangular, bent crystal monochromator. 3. I. The mirror The mirror was made out of float-glass substrate coated with Au. To increase adhesion of gold on the glass surface, layers of Cr and Cu were initially deposited on the substrate. The ihickness of each metal layer is approximately 500 A. The mirror has a rectangular shape (800 mm long, 100 mm wide and 90 mm high). It is placed on two pairs of supporting metallic spheres lying in a horizontal plane and separated by a distance of 750 mm from one another (see Fig. 2). The mirror is elastically bent by pulling the centre of the mirror down. The radius of the curvature could be varied from 1500 to 500 m, depending on the required distance from the position of the source to the focus point. The retentivity of the mirror was characterized using a conventional Cu K, X-ray source. It was found that for 8 keV photons the critical angle of the mirror is equal to (7.8 & 0.2) mrad. The reffectivity of the mirror at glancing angle of 7 mrad is 78%. The FWHM of the focus at these conditions was roughly 0.3 mm at 2.5 m from the mirror [l J. Accurate measurements of roughness and slope error of the float-glass substrate have been
Fig. 1. Schematic design on the LNLS protein cryst~lio~r~phy beamline given in vertical and horizontal projections.
performed at ESRF [2]. Using a Micro-Profiler Wyko Topo 2D phase interferometers the rootmean-square (rms) of the mirror roughness was found to be 610 A. The rms of a slope error, equal to 45 ptrdd, was measured with a system based on a two-beam interferometer. Considering the real mirror-to-image distance at the protein crystallography beamhne { * 5 m) and the slope error of Au surface, the actual FWHM of the vertical focus is estimated to be about 0.6 mm.
vacuum and has four independent tantalum jaws which edges were machine-cut to minimize scatter. Independent movement of each of the shts allows the control of vertical and horizontal divergence of the synchrotron radiation in an independent and pre-defined way. The slits can be adjusted to give minimum counts from scattered photons on the detector while retaining the beam flux measured on ion chambers. 3.3. The nzof7athromatur
3.1.
i”fk slits
Two sets of remotely adjustable slits before and after the mirror are used. The slit system operates in
The monochromator was designed to produce a highly focused monochromatic beam and to cover the wavelength range from 1.0 to 2.0 A when
162
1. Polikarpov et al. jNucl.
instr. and Meth. in Phys. Res. A 405 (1998) (5%I64
i.__.._ _. :-eP_.
,_
_ __,,_. _..,._..._. . .”
..
Fig. 2. Design of the X-ray mirror and the mirror vacuum chamber.
using either Si(1 1 1) or Ge(1 1 1) crystal (Fig. 3). The monochromator crystal has a classical triangular shape [3] and, after elastic deformation, a nearly cylindrical curvature. The crystal is asymmetrically cut to reduce the lateral size of the monochromatic beam. The 1.0 mm thick triangular-shaped (250 mm long and 50 mm on its base) crystal is clamped at its base and the apex rests upon an eccentric cylinder whose rotation controls the elastic bending of the crystal [4]. This rotation is produced via a translation stage connected to the eccentric cylinder by a thin wire which is wound around the axis. The
tilting of the crystal, which is necessary to control the vertical position of the focus of the monochromatic beam, is based on a cantilever using flexural pivots, allowing for backlash-free small tilt variation. This variation is also controlled by a translation stage connected to the cantilever by a thin wire. Both translation stages have stepping motors controlled by a computer. The cantilever is mounted on a plateau, whose rotation is produced via a translation stage, using the same scheme as the other movements. The compact housing of the monochromator and the very simple mechanical parts inside it, make it ultra-high-vacuum compatible [4].
1. Polikarpoc et al. /Nucl. Instr. and Meth. in PhJx Rex A 405 (I 998) 159-164
163
asymmetry cited above we will be able to cover the wavelength range 1.2 A I i I 2.0 A in a non-dispersive mode. Crystals cut at a smaller asymmetry angles will be available to work at a dispersionless conditions at shorter wavelengths. The monochromatic radiation will be tuneable between 2 and 1 A with the optimum wavelength at 1.6 A. The spectral density of the synchrotron radiation at LNLS storage ring will be about two and a half times higher at the wavelength 1.6 A as compared to 1.3 A. This can be further optimized by use of softer X-rays, however absorption in capillary, protein crystal and air might cause problems. 3.4. Detector system
Fig. 3. Design of the triangular bent crystal monochromator and its vacuum chamber. A part of Zj-arm is also shown.
The monochromator crystal is placed inside a high-vacuum chamber which is maintained below lo-’ mbar (without baking) using 120 l/s ion pump. The monochromator chamber is isolated from the mirror chamber (with a pressure below lo-* mbar) by a 125 urn thick Be window and it has a 100 urn thick Be exit window. The crystal (silicon (1 1 1) or germanium (1 1 1)) is asymmetrically cut at 7.25” and will be used in a condensing mode. This asymmetry angle corresponds to asymmetry factor b = 4 for the 1.3 A radiation and h = 2.86 for the 1.6 A radiation. To accept maximum possible divergence it will work in a dispersive mode. When necessary (for example, for multiwavelength anomalous diffraction/MAD data collection) we will be in a position to switch to a non-dispersive mode of data collection, when the source and the image are on the Rowland circle and the energy resolution is essentially limited by the Darwin width of the rocking curve. With angles of
Diffraction images will be recorded on a 345 mm diameter image plate system (MarResearch, I-Iamburg) to high resolution (up to 1.2 A resolution at 1.3 A wavelength of synchrotron light) and stored on a hard disk. Data acquisition will be controlled with a Pentium-Pro-based microcomputer and data processing will be performed on a dedicated Silicon Graphics Solid Impact Indigo 2 system. The image plate detector will be placed on a X Y translation stage allowing for precise translation of the detector system in the directions perpendicular and along the incident monochromatic beam. The XY translation stage rests on a 28-arm which is essentially a granite optical table which slides on an air-cushion, driven by a remote-controlled stepping motor. This will allow for multiple/single isomorphous replacement diffraction experiments with optimized anomalous scattering (MIRAS/ SIRAS) as well as multiwavelength anomalous diffraction (MAD) data collection (e.g. on Sm and Eu L-edges and to record anomalous contribution from iodine and uranium atoms).
4. Applications Protein crystallography in Brazil is showing significant development, with established research groups and new ones being created at different academical institutions. The limiting factor for this developments has been the difficulties in finding funding for the expensive investment costs in X-ray
164
equipment for protein crystallography applications. With the protein crystallography beamline at LNLS, it is expected that these new groups will have access to efficient data-collection facilities in a national laboratory intended to provide full technical support for its usage. Therefore, it is foreseen that the commissioning of the first protein crystallography beamline in Latin America will boost the number of protein structures determined locally and will increase the general interest of the national and regional molecular biology and biochemical research community in this area. Several of the research projects in structural biology currently being developed are focused on problems that are locally important and socially relevant. One important area is structure-based drug and vaccine design against tropical diseases as Chagas’ Disease, leishmaniasis, malaria and schistossomosis. The selected enzyme targets in these systems include glycolytic enzymes, proteins involved in host receptor recognition, proteases and redox enzymes. Other areas of interest are structure determination of enzymes, aimed at the understanding of the molecular mechanisms of activity and allosteric regulation; plant-seed storage proteins and plant lectins with important properties of cell recognition and regulation; snake toxins; S-endotoxins from Baccilus thurinyiensis with specificity against classes of insects; bacterial enzymes capable of degrading organochlo~nated herbicides and pesticides; structure of proteins involved in regulation of granulocytes in mammals; and a human salivary c1amylase inhibitor from wheat, with implications in diagnosis of pancreatic disorders and other forms of hyperamylasemias, diabetes control and obesity.
5. Conclusions The protein CrystaIlography beamline described above, designed for standard protein crystallography applications, will be installed at a bendingmagnet port of the LNLS electron storage ring. The critical phot:n wavelength of the LNLS bending magnet is 6 A. In vrder to have higher flux of X-ray photons below 6 A an insertion of 6-7 T superconducting wiggler in a straight section of a storage ring prior to bending magnet is planned for the
near future. In this case, the beamline could be shifted around the source point to accept radiation from the insertioa device, which critical wavelength will be 1.3-1.5 A. The beamline set-up will also provide cryo-cooling facilities for cryo-crystallography applications. A biochemistry laboratory for crystal crystallization, heavy-metal soaks, mounting and packing, etc., and a cold room will also be available. The computing facilities for beamline operation, data collection and evaluation are also part of the beamline. To efficiently optimize data collection by external users, a full technical support by trained personnel will be provided. As the LNLS storage ring is effectively a soft X-ray source, the use of longer wavelengths (2-3 A) in MAD experiments exploiting L and M edges of heavy-metal complexes will be envisaged in the design of a second dedicated protein crystallography beamline at LNLS. The LNLS protein crystallography beamline will be commissioned and brought into operation immediately after the electron storage ring will be operated at full power - by the end of 1997. All these facilities will be set up to serve the national and international user community and beamtime allocation will be based on the scientific quality of the proposals, as reviewed by a scientific board. Any information concerning beamline usage can be obtained at the following e-mail addresses: [email protected] or [email protected]. Acknowtedgements We thank FAPESP for the financial support to construct the beamline with computing and biochemistry facilities (Grants # 94/0711-l and #96/2285-5). Thanks are also due to CNPq and PADCT for the funding of the research projects associated to the beamline. References F. Vincentin. LNLS Technical Communication CTOl/95, 1994. 0. Hignette, unpublished results, 1996. M. Lemmonier, R. Fourme. F. Rousseaux, R. Kahn, Nucl. Instr. and Meth. 152 (1978) 173-t 77. L.A. Bernardes. H. Tolentino. A.R.D. Rodrigues, A.F. Craievich, L. Torriani. Rev. Sci. Instr. 63 (1992) 1065-1067.
LMEYHoos iN PHYSICS EISBVLBR
Nuciedr
Instruments
and Methods
in Physics
Research
A 405 (1998) 159-164
The protein crystallography beamline at LNLS, the Brazilian National Synchrotron Light Source G. Olivab, E.E. Castellanob, R.C. Garrattb, P. Arruda’, A. Leite”, A. Craievich”
I. Polikarpov”.“,
a Labo~ut~~io Nocio~~~ de Lx S~~~c~otro?~. CWPq. C.P. 6192, CEP 13~~~-~~~~. Campi~ffs, SP, Bra5f ~fntituto de Fisicu de Scio Carfos, U~ftle~sfdade de SZo Paufo, C.P. 369, 13560-970 Sao Cat&s, SP, Brazil ’ Cenrro de Biologfa Molecular e Engenhariu Genhtica, ~lnf~ersidade Estadt~al de Campisas. C.P. 6109. CEP 13083-970. Campinas. SP. &&I Received 24 April 1997; received in revised form I2 August
1997
Abstract
The Brazilian National Synchrotron Light Laboratory - LNLS, will have a dedicated protein crystallography beamline. The beamline under construction indudes cylindrical mirror and bent crystal monochromator focusing the high flux of synchrotron radiation in, the horizontal plane at the position of the sample. The monochromatic radiation will be tuneable between 2.0 and 1.0 A with the optimum wavelength at 1.3-I .6 A, chosen with the aim of maximizing the photon flux from the bending magnets of the storage ring (1.37 GeV). Diffraction images will be recorded on a commercial image plate detector system with on-line readout. The beamline set-up will include cooler/chiller for the samples and biochemical lab for crystallization, heavy-metal soaks, crystal storage and mounting at 22.‘C and 4-C. will also be available. The facility, intended to serve the national and international community, is planned to be brought into operation in the second half of 1997. It is foreseen that the commissioning of the first protein crystallography beamline in Latin America will boost the number of protein structures determined locally and will increase the general interest of the molecular biology and biochemical research community of Brazil in this area. ‘(2 1998 Elsevier Science B.V. All rights reserved. PAL’S: 07.85.Qe; 29.2O.Lq Qrwurds:
Protein crystallography;
Beamline; Brazilian National Synchrotron
1. Introduction
Synchrotron radiation with its high flux and brilliance combined with tunability over a wide
*Corresponding
author.
E-mail: [email protected].
0168-9~~~98~~19.~ :(:’ 1998 Elsevier Science R.V. All Pff SO I ~~-9002(97)0 1202-3
Light Source
wavelength range make it ideal for applications in protein crystallography. The new generation of synchrotron radiation sources has shown two complementary design characteristics: high-energy rings (e.g. ESRF, APS, Spring-g) with very high flux and wide spectrum of radiation available and medium-energy rings (l-l.SGeV, CAMD, MAX II,
rights reserved.
LNLS, ALS) optimized for the soft X-ray region, with a common feature of envisaging the extensive use of insertion devices to optimize the spectrum and flux of radiation to meet the needs of specific applications. The design specifications are frequently delimited by the available feuding, particularly considering that the cost of the synchrotron radiation facility rapidly grows with the energy of the ring. Other considerations can be the regional insertion of the laboratory, with spin-offs in scientific and technological development. Still, this new generation of medium-energy rings can provide a useful X-ray photon fluxes for most protein crystallography applications. With this goal, a protein crystallography beamline has been designed and is currently being commissioned at Laboratorio National de Luz Sincrotron /LNLS (Campinas, SBo Paulo State). This is a bending-magnet beamline and if includes a cylindrical elastically bent mirror providing vertical focusing, bent crystal monochromator, focusing synchrotron radiation in horizontal plane at the position of the sample, two sets of slits and two beam monitors. Single-axis goniometer and the image plate detector will be placed on an XY translation stage resting on a 2&arm. The mirror and monochromator optics and 20-arm were designed to cover a spectral range between 1.0 and 2.0 A. All mechanical movements will be remotely controlled. In this paper a brief description of the beamline is given. The main parameters of LNLS synchrotron radiation source are given in Section 2, the X-ray optics is described in Section 3 and some of the current scientific projects in Section 4.
2. The LNLS synchrotfon radiation source The energy of electrons inside the ring was planned to be 1.15 GeV and bending-magnet magnetic field 1.37 T, which gives a critical photon wavelength of 10.0 A. The maximum electron current is 100mA and average ring diameter 28 m. Magnetic field of constructed dipoles exceeded previously planned parameters and reached 1.64 T. That allowed for increase of the energy of electrons to 1.37_GeV and changed the critical photon energy to 6.0 A. In order to additionally increase the useful
photon energy range toward shorter wavelengths, an insertion of a 6-7 T superconducting wiggler in a straight section prior to the bending magnet # 3 is planned. The critical wavelength associated with the spectrum of the photons emitted by the wiggler will be 1.3-1.5 A, close to the medium of the wavelength range covered by the beamline. The source size is 0, = 0.295 mm, crT,= 0.120 mm, a: = 0.37 mrad, c$. = 0.096 mrad at energy of the ring 1.37 GeV. 3. The optical system for protein crystallographic studies The schematic view of the protein crystallography beamline is given in Fig. 1. The beamline will be placed at the bending magnet #3 and will accept synchrotron radiation emerging at an angle of 15 to the direction of the propagation of the electrons. The X-ray optics of the beamline includes a elastically bent mirror, two sets of slits (one before mirror and one before monochromator) and triangular, bent crystal monochromator. 3. I. The mirror The mirror was made out of float-glass substrate coated with Au. To increase adhesion of gold on the glass surface, layers of Cr and Cu were initially deposited on the substrate. The ihickness of each metal layer is approximately 500 A. The mirror has a rectangular shape (800 mm long, 100 mm wide and 90 mm high). It is placed on two pairs of supporting metallic spheres lying in a horizontal plane and separated by a distance of 750 mm from one another (see Fig. 2). The mirror is elastically bent by pulling the centre of the mirror down. The radius of the curvature could be varied from 1500 to 500 m, depending on the required distance from the position of the source to the focus point. The retentivity of the mirror was characterized using a conventional Cu K, X-ray source. It was found that for 8 keV photons the critical angle of the mirror is equal to (7.8 & 0.2) mrad. The reffectivity of the mirror at glancing angle of 7 mrad is 78%. The FWHM of the focus at these conditions was roughly 0.3 mm at 2.5 m from the mirror [l J. Accurate measurements of roughness and slope error of the float-glass substrate have been
Fig. 1. Schematic design on the LNLS protein cryst~lio~r~phy beamline given in vertical and horizontal projections.
performed at ESRF [2]. Using a Micro-Profiler Wyko Topo 2D phase interferometers the rootmean-square (rms) of the mirror roughness was found to be 610 A. The rms of a slope error, equal to 45 ptrdd, was measured with a system based on a two-beam interferometer. Considering the real mirror-to-image distance at the protein crystallography beamhne { * 5 m) and the slope error of Au surface, the actual FWHM of the vertical focus is estimated to be about 0.6 mm.
vacuum and has four independent tantalum jaws which edges were machine-cut to minimize scatter. Independent movement of each of the shts allows the control of vertical and horizontal divergence of the synchrotron radiation in an independent and pre-defined way. The slits can be adjusted to give minimum counts from scattered photons on the detector while retaining the beam flux measured on ion chambers. 3.3. The nzof7athromatur
3.1.
i”fk slits
Two sets of remotely adjustable slits before and after the mirror are used. The slit system operates in
The monochromator was designed to produce a highly focused monochromatic beam and to cover the wavelength range from 1.0 to 2.0 A when
162
1. Polikarpov et al. jNucl.
instr. and Meth. in Phys. Res. A 405 (1998) (5%I64
i.__.._ _. :-eP_.
,_
_ __,,_. _..,._..._. . .”
..
Fig. 2. Design of the X-ray mirror and the mirror vacuum chamber.
using either Si(1 1 1) or Ge(1 1 1) crystal (Fig. 3). The monochromator crystal has a classical triangular shape [3] and, after elastic deformation, a nearly cylindrical curvature. The crystal is asymmetrically cut to reduce the lateral size of the monochromatic beam. The 1.0 mm thick triangular-shaped (250 mm long and 50 mm on its base) crystal is clamped at its base and the apex rests upon an eccentric cylinder whose rotation controls the elastic bending of the crystal [4]. This rotation is produced via a translation stage connected to the eccentric cylinder by a thin wire which is wound around the axis. The
tilting of the crystal, which is necessary to control the vertical position of the focus of the monochromatic beam, is based on a cantilever using flexural pivots, allowing for backlash-free small tilt variation. This variation is also controlled by a translation stage connected to the cantilever by a thin wire. Both translation stages have stepping motors controlled by a computer. The cantilever is mounted on a plateau, whose rotation is produced via a translation stage, using the same scheme as the other movements. The compact housing of the monochromator and the very simple mechanical parts inside it, make it ultra-high-vacuum compatible [4].
1. Polikarpoc et al. /Nucl. Instr. and Meth. in PhJx Rex A 405 (I 998) 159-164
163
asymmetry cited above we will be able to cover the wavelength range 1.2 A I i I 2.0 A in a non-dispersive mode. Crystals cut at a smaller asymmetry angles will be available to work at a dispersionless conditions at shorter wavelengths. The monochromatic radiation will be tuneable between 2 and 1 A with the optimum wavelength at 1.6 A. The spectral density of the synchrotron radiation at LNLS storage ring will be about two and a half times higher at the wavelength 1.6 A as compared to 1.3 A. This can be further optimized by use of softer X-rays, however absorption in capillary, protein crystal and air might cause problems. 3.4. Detector system
Fig. 3. Design of the triangular bent crystal monochromator and its vacuum chamber. A part of Zj-arm is also shown.
The monochromator crystal is placed inside a high-vacuum chamber which is maintained below lo-’ mbar (without baking) using 120 l/s ion pump. The monochromator chamber is isolated from the mirror chamber (with a pressure below lo-* mbar) by a 125 urn thick Be window and it has a 100 urn thick Be exit window. The crystal (silicon (1 1 1) or germanium (1 1 1)) is asymmetrically cut at 7.25” and will be used in a condensing mode. This asymmetry angle corresponds to asymmetry factor b = 4 for the 1.3 A radiation and h = 2.86 for the 1.6 A radiation. To accept maximum possible divergence it will work in a dispersive mode. When necessary (for example, for multiwavelength anomalous diffraction/MAD data collection) we will be in a position to switch to a non-dispersive mode of data collection, when the source and the image are on the Rowland circle and the energy resolution is essentially limited by the Darwin width of the rocking curve. With angles of
Diffraction images will be recorded on a 345 mm diameter image plate system (MarResearch, I-Iamburg) to high resolution (up to 1.2 A resolution at 1.3 A wavelength of synchrotron light) and stored on a hard disk. Data acquisition will be controlled with a Pentium-Pro-based microcomputer and data processing will be performed on a dedicated Silicon Graphics Solid Impact Indigo 2 system. The image plate detector will be placed on a X Y translation stage allowing for precise translation of the detector system in the directions perpendicular and along the incident monochromatic beam. The XY translation stage rests on a 28-arm which is essentially a granite optical table which slides on an air-cushion, driven by a remote-controlled stepping motor. This will allow for multiple/single isomorphous replacement diffraction experiments with optimized anomalous scattering (MIRAS/ SIRAS) as well as multiwavelength anomalous diffraction (MAD) data collection (e.g. on Sm and Eu L-edges and to record anomalous contribution from iodine and uranium atoms).
4. Applications Protein crystallography in Brazil is showing significant development, with established research groups and new ones being created at different academical institutions. The limiting factor for this developments has been the difficulties in finding funding for the expensive investment costs in X-ray
164
equipment for protein crystallography applications. With the protein crystallography beamline at LNLS, it is expected that these new groups will have access to efficient data-collection facilities in a national laboratory intended to provide full technical support for its usage. Therefore, it is foreseen that the commissioning of the first protein crystallography beamline in Latin America will boost the number of protein structures determined locally and will increase the general interest of the national and regional molecular biology and biochemical research community in this area. Several of the research projects in structural biology currently being developed are focused on problems that are locally important and socially relevant. One important area is structure-based drug and vaccine design against tropical diseases as Chagas’ Disease, leishmaniasis, malaria and schistossomosis. The selected enzyme targets in these systems include glycolytic enzymes, proteins involved in host receptor recognition, proteases and redox enzymes. Other areas of interest are structure determination of enzymes, aimed at the understanding of the molecular mechanisms of activity and allosteric regulation; plant-seed storage proteins and plant lectins with important properties of cell recognition and regulation; snake toxins; S-endotoxins from Baccilus thurinyiensis with specificity against classes of insects; bacterial enzymes capable of degrading organochlo~nated herbicides and pesticides; structure of proteins involved in regulation of granulocytes in mammals; and a human salivary c1amylase inhibitor from wheat, with implications in diagnosis of pancreatic disorders and other forms of hyperamylasemias, diabetes control and obesity.
5. Conclusions The protein CrystaIlography beamline described above, designed for standard protein crystallography applications, will be installed at a bendingmagnet port of the LNLS electron storage ring. The critical phot:n wavelength of the LNLS bending magnet is 6 A. In vrder to have higher flux of X-ray photons below 6 A an insertion of 6-7 T superconducting wiggler in a straight section of a storage ring prior to bending magnet is planned for the
near future. In this case, the beamline could be shifted around the source point to accept radiation from the insertioa device, which critical wavelength will be 1.3-1.5 A. The beamline set-up will also provide cryo-cooling facilities for cryo-crystallography applications. A biochemistry laboratory for crystal crystallization, heavy-metal soaks, mounting and packing, etc., and a cold room will also be available. The computing facilities for beamline operation, data collection and evaluation are also part of the beamline. To efficiently optimize data collection by external users, a full technical support by trained personnel will be provided. As the LNLS storage ring is effectively a soft X-ray source, the use of longer wavelengths (2-3 A) in MAD experiments exploiting L and M edges of heavy-metal complexes will be envisaged in the design of a second dedicated protein crystallography beamline at LNLS. The LNLS protein crystallography beamline will be commissioned and brought into operation immediately after the electron storage ring will be operated at full power - by the end of 1997. All these facilities will be set up to serve the national and international user community and beamtime allocation will be based on the scientific quality of the proposals, as reviewed by a scientific board. Any information concerning beamline usage can be obtained at the following e-mail addresses: [email protected] or [email protected]. Acknowtedgements We thank FAPESP for the financial support to construct the beamline with computing and biochemistry facilities (Grants # 94/0711-l and #96/2285-5). Thanks are also due to CNPq and PADCT for the funding of the research projects associated to the beamline. References F. Vincentin. LNLS Technical Communication CTOl/95, 1994. 0. Hignette, unpublished results, 1996. M. Lemmonier, R. Fourme. F. Rousseaux, R. Kahn, Nucl. Instr. and Meth. 152 (1978) 173-t 77. L.A. Bernardes. H. Tolentino. A.R.D. Rodrigues, A.F. Craievich, L. Torriani. Rev. Sci. Instr. 63 (1992) 1065-1067.