Superoxide dismutases (SODs) are the major regulators of oxidative stress and therefore the first line of defense to protect organisms against metabolic- and environmentally-induced reactive oxygen species (ROS). Human mitochondrial manganese SOD (MnSOD) expression is modulated to prevent ROS-based damage, promote redox homeostasis and maintain proper cell signaling. Our research goal is to understand the molecular basis of how MnSOD uses coupled proton-electron transfers to dismute superoxide. For this, the 3D arrangement of all atoms is needed, most importantly the position of protons. Our recent technical advancements with neutron crystallography at Oak Ridge National Laboratory have overcome the limitations of X-ray crystallography – revealing proton positions with high detail while also allowing control of the metal electronic state. In this research project, MnSOD neutron maps reveal the proton relays to the active site metal and the protonation states of metal-bound ligands. Our results demonstrate the transfer of protons to the bound active site solvent that is triggered by the reduction of the active site manganese. This proton transfer involves unusual active site amino acid pKas, at least five low barrier hydrogen bonds, glutamine tautomerization and a water bridge in the active site channel.

Copper-containing nitrite reductases (CuNiRs) that convert NO₂ to NO are of central importance in nitrogen-based energy metabolism [1]. These metalloenzymes, like all redox enzymes, are very susceptible to radiation damage from the intense synchrotron radiation by X-rays, that are used to obtain structures at high resolution. Understanding the chemistry that underpins the enzyme mechanisms in these systems usually requires atomic resolutions of better than 1.2 Å. The damage-free structure of the resting state of one of the most studied CuNiRs was obtained by X-ray free-electron laser (XFEL) and neutron crystallography, which allows direct comparison of neutron, XFEL structural data [2] and atomic resolution X-ray structural data used to obtain the most accurate (atomic resolution with unrestrained SHELX refinement) structure.

It was demonstrated that Asp_CAT (Asp₉₈) and His_CAT(His₂₅₅) are deprotonated in the resting state of CuNiRs at pH values close to the optimum for activity (Fig.1).

References

[1] Zumft, W. G. (1997). Microbiol. Mol. Biol. Rev. 61, 533

[2] Halsted, T.P., Yamashita, K., Gopalasingam, C. C., Shenoy, R.T, Hirata, K., Ago, H., Ueno, G., Blakeley, M.P., Eady, R R.; Antonyuk, S.V., Yamamoto, M., Hasnain, S. S. (2019). IUCrJ 6, 761

Acknowledgement; Moulin M.; Haertlain,M.; Blakeley, M.P.; Halsted, T.P.; Yamashita, K.; Gopalasingam, C,;. C.;Hirata, K; Ago, H.; Ueno, G.; Eady, R R.; Yamamoto, M. and Hasnain S.S

Hydrogen (H) atoms often play essential roles in enzymatic reactions as they are responsible for the reversible protonation of active site residues and for the organization of the solvent network. More generally, hydrogens are also necessary for the establishment of H-bonds which, in turn, stabilize interactions between macromolecules and between macromolecules and their ligands. Although H atoms represent a large fraction of the total atomic content of macromolecules their direct visualization is not straightforward. Even at (sub-) atomic resolution (<1.2 Å), X-ray macromolecular crystallography (MX), the most common technique for structural determination, affords the localization of only a small percentage of H atoms as their contribution to the total scattering is minimal owing to their low electron content. Differently from MX, neutron macromolecular crystallography (NMX) relies on the interaction between neutrons and atomic nuclei. With this technique the visualization of H atoms is possible even at modest resolution (2.0 - 2.5 Å). In fact, NMX maps indicate the nuclear positions of H atoms while MX maps show the positions of valence-electron density for H atoms shifted along the bond vector. Recently, single particle cryo-electron microscopy (cryo-EM) achieved atomic resolution protein structure determination [1, 2] . Nakane et al. determined apoferritin and the GABA A receptor structures at 1.22 and 1.7 Å resolutions, respectively. H density peaks can be seen even at 1.7 Å, unlikely in MX experiments. Interestingly, cryo-EM/electron diffraction experiments inform on both nuclear and electron localization of H atoms.

Our research is focused on the modelling and refinement of H atoms by using different experimental data (cryo-EM, neutron and electron diffraction) integrated in a common framework, in order to provide new insights in biological processes such as enzyme mechanisms. New features in the crystallographic refinement package REFMAC5 [3], one of the flagships of the scientific CCP4 computational suite, have been developed and will be presented. CCP4 Monomer Library [4] has been implemented for more accurate H atom positions derived from neutron data analysis [5] and Quantum Mechanics (QM) calculations. Recent developments in REFMAC5 and relative tools for the refinement of structural models obtained by neutron diffraction data will also be presented.

[1] Nakane, T., Kotecha, A., Sente, A., McMullan, G., Masiulis, S., Brown, P. M. G. E., Grigoras, I. T., Malinauskaite, L., Malinauskas, T., Miehling, J., Uchański, T., Yu, L., Karia, D., Pechnikova, E. V., De Jong, E., Keizer, J., Bischoff, M., McCormack, J., Tiemeijer, P., Hardwick, S. W., Chirgadze, D. Y., Murshudov, G., Aricescu, A. R. & Scheres, S. H. W. (2020). Nature 587, 152.

[2] Yip, K. M., Fischer, N., Paknia, E., Chari, A. & Stark, H. (2020). Nature 587, 157.

[3] Kovalevskiy, O., Nicholls, R. A., Long, F., Carlon, A. & Murshudov, G. N. (2018). Acta Cryst. D74, 215.

[4] Vagin, A. A., Steiner, R. A., Lebedev, A. A., Potterton, L., McNicholas, S., Long, F. & Murshudov, G. N. (2004). Acta Cryst. D60, 2184.

[5] Allen, F. H. & Bruno, I. J. (2010). Acta Cryst. B66, 380.

Finding the Goldilocks zone for chemical crystallography via Laue single-crystal neutron diffraction – what have we learned from KOALA to improve KOALA 2.0?

A.J. Edwards, R.O. Piltz

Australian Centre for Neutron Scattering, Australian Nuclear Science and Technology Organization,

New Illawarra Rd, Lucas Heights, N.S.W., Australia

Alison.Edwards@ansto.gov.au

KOALA is a single-crystal Laue neutron diffractometer standing at the end of guide position of the supermirror guide TG3 at the OPAL reactor, ANSTO. The instrument was initially modelled closely on VIVALDI[1], an instrument available in the user program at the ILL from 2001-2010. The elegantly simple concept of the instrument employs a cylindrical neutron sensitised image plate detector which is used to record a series of diffraction images from a suitable number of crystal positions to provide a sufficient data set from which valid model parameters can be derived to answer questions regarding material properties which cannot be adequately derived from more readily available methods, most particularly X-ray diffraction and more recently the hybrid methodology of quantum crystallography.

Our initial practice with the instrument adhered largely to that shared with us by the scientists at the ILL. This early experience[2] was the commencement of a steep learning curve which has, with a very limited number of other instruments brought single-crystal neutron diffraction into greater use in chemistry and chemical crystallography in the second decade of the 21st century. Key developments have been (i) the installation of an Oxford Cryosystems COBRA™ nitrogen cryostream which facilitates handling of oxygen and moisture sensitive compounds (which encompass a significant fraction of the proposals received for the instrument) and (ii) the development of a user accessible data reduction for the diffraction images. From the first proposal round for the instrument in 2009 exciting chemistry was proposed for experiments which exceeded the nominal maximum primitive unit cell volume for the recording of useful diffraction images. A simple work around for this has been to reduce the resolution of the images by manipulation of the temperature at which they are recorded – in order to obtain data against which a model may be refined.

More commonly though, it is observed that crystals for which the unit cell volume is relatively large tend, where they can be grown to a size sufficient for Laue neutron diffraction, to have a mosaic spread which limits the resolution of the pattern observed without manipulating the temperature to further reduce the resolution.

With careful attention to experimental detail and the availability of discretionary beam-time access it has been possible to undertake studies of important new materials in timeframes which have resulted in the publication of the single-crystal neutron diffraction study with the chemistry it underpins, rather than as a stand alone paper reporting only the neutron study result. It is of particular importance to note that in the case of hydride containing compounds, it can be critical to prove the location of the hydride via neutron diffraction and even a low resolution study can provide the necessary proof. In consequence of their publication with the chemistry, papers from KOALA are now submitted to and published in journals of the highest standing [4-7].

Having achieved a more routine applicability of neutron diffraction in chemical crystallography, we reached a point where elctronic components of KOALA had exceeded their serviceable lifespans and contemplation of replacing this aspect of the instrument led us to realise that reworking the existing mechanical elements with new electronics posed significant challenges and would cost a large fraction of the potential cost of building a new instrument. We are fortunate that the decision was reached to design a new instrument which is allowing us to optimise key design elements to yield maximum flexibility of the instrument across all of its possible applications in chemistry, physics, materials science and crystallography. The instrument is currently under construction and should be available for users in the second half of 2022.

[1] C Wilkinson, JA Cowan, DAA Myles, F Cipriani, GJ Mclntyre, (2002), Neutron News, 13, 37-41.

[2] AJ Edwards, (2011) Australian Journal of Chemistry 64, 869-872

[3] RO Piltz,(2018) Journal of Applied Crystallography 51, 635-645 and 963-965

[4] M Garçon, C Bakewell, GA Sackman, AJP White, RI Cooper, AJ Edwards, MR Crimmin (2019) Nature 574 , 390-393

[5] SJ Bonyhady, D Collis, N Holzmann, AJ Edwards, RO Piltz, G Frenking, A Stasch C Jones, (2018) Nature Comms 9 , 3079

[6] JAB Abdalla, A Caise, CP Sindlinger, R Tirfoin, AL Thompson, AJ Edwards, S Aldridge, (2017) Nature chemistry 9, 1256-1262

[7] R Chen, G Qin, S Li, AJ Edwards, RO Piltz, I Del Rosal, L Maron, D Cui, J Cheng Angewandte Chemie 132, 11346-11351

Keywords: neutron diffraction; Laue diffraction; chemical crystallography; instrumentation; structure

Single-crystal neutron scattering experiments have the advantage compared to X-ray experiments that it is possible to get positions for hydrogen atoms – typically replaced by deuterons in neutron protein crystallography. The hydrogens are important because they constitute approximately half of the atoms in a protein and determine the directionality of hydrogen bonds, which are key to the structure and function [1]. Therefore, neutron crystallographic experiments give important additional information to the model. However, adding all hydrogens by hand to the model is a tedious and error-prone work and most software add hydrogens at positions suggested by a statistical analysis of neutron structures and not based on the measured data. Moreover, it is important to decide for each hydrogen atom whether its position is supported by the experimental data or not.

To solve these problems, we developed an automatised procedure that places all hydrogen atoms of a protein based on local integration of the neutron 2mF_o – DF_c map. For each putative hydrogen atom, we search for the highest integrated value of the nuclear scattering length density within a sphere of the covalent radius of hydrogen. For some hydrogens, the position is dictated by the positions of the surrounding heavy atoms. However, many hydrogens can be anywhere on a circle (e.g. OH, SH and NH₃ groups) and we search all possible positions systematically for the highest integrated density. Likewise, we consider possible flips of Asn and Gln residues, we consider six possible states of His residues, and we consider alternative protonation states of Asp, Glu, Lys, Tyr and Cys residues. The method is calibrated to available neutron structures and the number of favourable hydrogen bonds are evaluated.

[1] Engler, N., Ostermann, A., Niimura, N., & Parak, F. G. (2003). Hydrogen atoms in proteins: positions and dynamics, Proc. Natl. Acad. Sci. U.S.A., 100(18), 10243-10248.

Harnessing the spin dependence of the neutron scattering cross section for hydrogen, Dynamic Nuclear Polarization (DNP) is a potentially powerful technique for neutron diffraction measurements, especially for biological systems. Polarizing the neutron beam and aligning the proton spins in a polarized sample modulates and tunes the coherent and incoherent neutron scattering cross-sections of hydrogen [1], in ideal cases maximizing the scattering from - and visibility of - hydrogen atoms in the sample while simultaneously minimizing the incoherent background to zero (see Figure 1).

ORNL has developed a prototype system for the purpose of performing proof-of-concept Neutron Macromolecular Crystallography measurements which highlight the potential of DNP [2]. We will describe DNP concepts, experimental design, labelling strategies and the most recent results, as well as considering future prospects for data collection and analysis that these techniques enable.