Katsuaki Tomoyori

Hydrogen-bond network and pH sensitivity in transthyretin: Neutron crystal structure of human transthyretin

Transthyretin (TTR) is a tetrameric protein associated with human amyloidosis. In vitro, the formation of amyloid fibrils by TTR is known to be promoted by low pH. Here we show the neutron structure of TTR, focusing on the hydrogen bonds, protonation states and pH sensitivities. A large crystal was prepared at pD 7.4 for neutron protein crystallography. Neutron diffraction studies were conducted using the IBARAKI Biological Crystal Diffractometer with the time-of-flight method. The neutron structure solved at 2.0Å resolution revealed the protonation states of His88 and the detailed hydrogen-bond network depending on the protonation states of His88. This hydrogen-bond network is composed of Thr75, Trp79, His88, Ser112, Pro113, Thr118-B and four water molecules, and is involved in both monomer-monomer and dimer-dimer interactions, suggesting that the double protonation of His88 by acidification breaks the hydrogen-bond network and causes the destabilization of the TTR tetramer. In addition, the comparison with X-ray structure at pH 4.0 indicated that the protonation occurred to Asp74, His88 and Glu89 at pH 4.0. Our neutron model provides insights into the molecular stability of TTR related to the hydrogen-bond network, the pH sensitivity and the CH···O weak hydrogen bond.

Neutron structure analysis using the IBARAKI biological crystal diffractometer (iBIX) at J-PARC.

The IBARAKI Biological Crystal Diffractometer (iBIX), a new diffractometer for protein crystallography at the next-generation neutron source at J-PARC (Japan Proton Accelerator Research Complex), has been constructed and has been operational since December 2008. Preliminary structure analyses of organic crystals showed that iBIX has high performance even at 120 kW operation and the first full data set is being collected from a protein crystal.

Evaluation of performance for IBARAKI biological crystal diffractometer iBIX with new detectors.

The time-of-flight neutron single-crystal diffractometer iBIX at the next-generation neutron source J-PARC has been upgraded and is available for user experiments on protein samples in particular. Neutron structure analysis of a standard protein sample was carried out in order to evaluate the performance of iBIX.

Neutron and X-ray crystallographic analysis of the human α-thrombin-bivalirudin complex at pD 5.0: protonation states and hydration structure of the enzyme-product complex.

The protonation states and hydration structures of the α-thrombin-bivalirudin complex were studied by joint XN refinement of the single crystal X-ray and neutron diffraction data at resolutions of 1.6 and 2.8Å, respectively. The atomic distances were estimated by carrying out X-ray crystallographic analysis at 1.25Å resolution. The complex represents a model of the enzyme-product (EP) complex of α-thrombin. The neutron scattering length maps around the active site suggest that the side chain of H57/H was deuterated. The joint XN refinement showed that occupancies for Dδ1 and Dε2 of H57/H were 1.0 and 0.7, respectively. However, no significant neutron scattering length density was observed around the hydroxyl oxygen Oγ of S195/H, which was close to the carboxylic carbon atom of dFPR-COOH. These observations suggest that the Oγ atom of S195/H is deprotonated and maintains its nucleophilicity in the EP complex. In addition to the active site, the hydration structures of the S1 subsite and the Exosite I, which are involved in the recognition of bivalirudin, are presented.

New data reduction protocol for Bragg reflections observed by TOF single-crystal neutron diffractometry for protein crystals with large unit cells

In protein crystallography, high backgrounds are caused by incoherent scattering from the hydrogen atoms of protein molecules and hydration water. In addition, the scattering intensity from large unit-cell crystals is very small, which makes it difficult to improve the signal-to-noise ratio. In the case of time-of-flight (TOF) single-crystal neutron diffractometry, the measured spectra cover four-dimensional space including X, Y, and TOF in addition to intensity. When estimating the integrated intensity, 3D background domains in the vicinity of peaks should be clearly classified. In conventional 1D or 2D background evaluation, the evaluation is applied for individual peaks assigned using peak searches; however, it is quite difficult to classify the 3D background domain in TOF protein single-crystal neutron diffraction experiments. We undertook the development of a data reduction protocol for measurements involving large biomacromolecules. At the initial stage of the reduction protocol, appropriate 3D background estimation and eliminations were applied over the entire range of X, Y, and TOF bins. The histograms were then searched for peaks and indexed, and the individually assigned peaks were finally integrated with an effective profile function in the TOF direction. Three-dimensional deconvolution procedures for overlapping peaks associated with large unit cells were implemented as necessary. This data reduction protocol may lead to the improvement of signal-to-noise ratios to enable TOF spectral analysis of large unit-cell protein crystals.

Neutron diffractometer covering protein crystals with large unit cell at J-PARC

The structural information of hydrogen atoms and hydration waters obtained by neutron protein crystallography is expected to contribute to elucidation of protein function and its improvement. However, many proteins, especially membrane proteins and protein complexes, have larger molecular weight and then unit cells of their crystals have larger volume, which is out of range of measurable unit cell volume for conventional diffractometers. Therefore, our group had designed the diffractometer which can cover such crystals with large unit cell volume (target lattice length: 250 Å). This diffractometer is dedicated for protein single crystals and has been proposed to be installed at J-PARC (Japan Proton Accelerator Research Complex). Larger unit cell volume causes a problem to separate spots closer to each other in spatial as well as time dimension in diffraction images. Therefore, our proposed diffractometer adopts longer camera distance (L2 = 800mm) and selects decoupled hydrogen moderator as neutron source which has shorter pulse width. Under the conditions that L1 is 33.5m, beam divergence 0.4° and crystal edge size 2mm, this diffractometer is estimated to be able to resolves spots diffracted from crystals with a lattice length of 220 Å in each axis at d-space of 2.0 Å. In order to cover large neutron detecting area due to long camera distance, novel large-area detector (larger than 300mm ⨯ 300mm) with a spatial resolution of better than 2.5mm is under development. More than 40 these detectors plan to be installed, providing the total solid angle coverage of larger than 33%. For neutron guide, ellipsoidal supermirror is considered to be adopted to increase neutron flux at the sample position. The final gain factor of this diffractometer is estimated to be about 20 or larger as compared with BIX-3/4 diffractometers operated in the research reactor JRR-3 at JAEA (Japan Atomic Energy Agency) [1,2].

Neutron diffraction study to elucidate reaction mechanism of RNase A

Ribonuclease A (RNase A) is a pyrimidine-specific endoribonuclease that claves and hydrolyzes single-stranded RNA in two distinct steps. The mechanism of the cleavage reaction catalyzed by RNase A involves two key histidine residues, His12 and His119. It is important to know the protonation states of them in order to understand the hydrolysis mechanism of RNase A. Neutron protein crystallography is a powerful technique for solving the problems. In previous reports, the protonation states of them for RNase A complexed with phosphate ion, uridine vanadate and phosphate free one have been investigated by neutron diffraction analysis [1-3]. In this study, neutron diffraction analysis of phosphate free RNase A has been carried out with high resolution and completeness data set in order to clarify the protonation states of two active site histidine residues, and to elucidate the detailed mechanism of the cleavage reaction. Neutron diffraction data of bovine pancreatic RNase A were collected by IBARAKI biological crystal diffractometer iBIX in J-PARC. The structure was determined by joint neutron and X-ray structure refinement. The final values of Rcryst and Rfree were 19.5% and 22.0%, respectively, for completeness of 86.7% to a resolution of 1.4Å. The structure with high reliability and good data statistics could be obtained by comparing with the already-reported one [3]. We calculated |Fo|-|Fc| neutron scattering length density map after omitting Dδ1 and Dε2 of His12 and His119 in order to confirm the protonation states of them. These omit maps indicated that His12 is completely singly protonated and Hi119 is doubly protonated. The protonation states of them are consistent with those in the first step of the putative mechanism of catalysis by RNase A. We could also observed a D atom of water molecule which is hydrogen bonded to Nε2 of His12.

A new biological neutron diffractometer (iBIX) in J-PARC

The present talk describes novel powerful imaging for X-ray fluorescence (XRF) and X-ray diffraction (XRD). So far, the scanning-type imaging has been widely used in those techniques. Though recent progress in high-spatial-resolution imaging using synchrotrons is significant, there have been a clear limit; because of the step-scan, the imaging requires a long measuring time. In many scientific applications, X-ray imaging that are much more rapid, e.g., capable of high-speed resolution have been demanded. It is possible to do X-ray imaging without performing any scans. Here, the method uses quite a wide beam, which illuminates the whole sample surface in a low-angle-incidence arrangement (0.5~3 deg). The detector used is a CCD camera working at 30 fr./sec, equipped with a collimator inside, and the distance between the sample surface and the detector is set extremely close, in order to enhance both spatial resolution and efficiency. Note that the imaging is done with one shot. In the case of XRF imaging, distinguishing elements are required and, therefore, most of the experiments were performed with monochromatic or quasi-monochromatic X-rays. The procedure for XRD imaging uses a combination of exposure and incident X-ray energy scan (or just tuning). Since the present experiment employs a fixed small-angle incidence and also a fixed diffraction angle of around 90 deg, the diffraction plane here is inclined at about 45 deg from the surface of the specimen. By scanning the energy of the incident X-rays, one obtains a diffraction peak, which corresponds to the lattice spacing. Further instrumental details and many applications will be presented. References [1] K.Sakurai, Spectrochimica Acta B54, 1497 (1999) [2] K.Sakurai and H.Eba, Anal. Chem. 75, 355 (2003)

Development of data-processing software for a TOF single-crystal neutron diffractometer at J-PARC

1 a new basic life science fields as well as applied industries. To achieve the performance mentioned as above, the diffractometer will be installed on a coupled moderator which has more intense peak and integrated intensity but wider pulse shape than a decoupled moderator. It is expected that some neighbor Bragg spots will overlap partially each other along the time-axis. The overlapping of Bragg spots along the time-axis should be considered for the optimization of design parameters and It is necessary to deconvolute the overlapped spots in order to obtain a data set that has a quality good enough to identify hydrogen atoms in biological macromolecules. The three original simulation programs of TOF diffraction data with designed parameters of the diffractometer were developed to obtain information of spot-overlapping, completeness of Bragg spots and spot profiles along time-axis. The consideration of important designed parameters (divergence of incident neutron beam to a sample crystal, the distance between sample and detector surface and the best detector configuration) focused on biological macromolecular, the strategy of data collection and de-convoluting overlapped spots will be reported based on the results of simulation by using the simulation programs mentioned as above.

Optics and shielding of IBARAKI Biological Crystal Diffractometer (iBIX) in J-PARC

1 a new basic life science fields as well as applied industries. To achieve the performance mentioned as above, the diffractometer will be installed on a coupled moderator which has more intense peak and integrated intensity but wider pulse shape than a decoupled moderator. It is expected that some neighbor Bragg spots will overlap partially each other along the time-axis. The overlapping of Bragg spots along the time-axis should be considered for the optimization of design parameters and It is necessary to deconvolute the overlapped spots in order to obtain a data set that has a quality good enough to identify hydrogen atoms in biological macromolecules. The three original simulation programs of TOF diffraction data with designed parameters of the diffractometer were developed to obtain information of spot-overlapping, completeness of Bragg spots and spot profiles along time-axis. The consideration of important designed parameters (divergence of incident neutron beam to a sample crystal, the distance between sample and detector surface and the best detector configuration) focused on biological macromolecular, the strategy of data collection and de-convoluting overlapped spots will be reported based on the results of simulation by using the simulation programs mentioned as above.