Structural and functional insights into hydrogen sulfide homeostasis in pathogenic bacteria

José A. Brito^1,§, Sofia S. Costa¹, Brenna J. Walsh², David P. Giedroc^2,3,§

¹ Instituto de Tecnologia Química e Biológica António Xavier, Universidade NOVA de Lisboa, Oeiras – Portugal
² Department of Chemistry, Indiana University, Bloomington, USA
³ Department of Molecular and Cellular Biochemistry, Indiana University, Bloomington, USA

^§ Corresponding authors: This email address is being protected from spambots. You need JavaScript enabled to view it. & This email address is being protected from spambots. You need JavaScript enabled to view it.

Hydrogen sulfide is an ancient molecule present in Earth’s primordial atmosphere and organisms from all Domains of Life soon evolved to utilize it in their physiology. However, H₂S can have either beneficial or toxic effects, depending on the concentration. Therefore, tight regulation of intracellular H₂S/H₂S-derived more oxidized reactive sulfur species (RSS) is paramount for survival of all organisms. In bacterial pathogens, H₂S/RSS is regarded as an important component in microbial defense mechanisms against oxidative and antibiotic stress.

The cst operon in Staphylococcus aureus encodes a nearly complete mitochondrial-like H₂S oxidation system. In addition, a cst-like operon has also been described in the human pathogen E. faecalis. Three enzymes encoded by these two operons include the coenzyme A persulfide reductase CoAPR, the multidomain persulfide dioxygenase-sulfurtransferase fusion protein CstB and the sulfide:quinone oxidoreductase SQR, which collectively protect the organism against H₂S and RSS toxicity.

Herein, we describe the X-ray crystallographic structures of full-length SaCstB (native and single cysteine substitution mutants) and the CoA-bound crystal structure of EfCoAPR. Companion cryo-EM data on these enzymes suggest a high mobility of the C-terminal rhodanese domains that may be important for catalysis. The structures of sulfite-bound mutant CstBs suggests a mechanism by which the C-terminal domain facilitates the concerted oxidation of a thiol persulfide (RSSH) to thiosulfate and thiol, without the release of the toxic sulfite intermediate.

These studies provide an enhanced understanding of the mechanisms of H₂S/RSS homeostasis encoded by the RSS-regulated cst operons in bacteria, and were possible throught an iNext-Discovery funded access proposal (PID 16108).

Keywords: hydrogen sulfide, sulfur-metabolising pathways, pathogenic bacteria, X-ray crystallography, cryo-EM

CryoEM conformational landscapes: Directly accessing macromolecular flexibility

Electron Microscopy at cryogenic temperatures is currently a very well-known approach to solve the structure of biological macromolecules with sufficiently high resolution to obtain good structural models. Part of this success was our proposal of Maximum Likelihood approaches to disentangle macromolecular structural flexibility (first implemente in the software XMIPP and then in Relion) (1). However, we then assumed that flexibility was discrete and that the user had to estimate the number of these discrete states, which clearly was not optimum. Nowadays, almost 15 years after this initial approach, we have developed new methods that allow us to consider continuous flexibility, opening the possibility to obtain conformational landscapes (2). In this context I will present recent advances in our "Zernike3D '' approach, showing how we effectively access the whole range of macromolecular flexibility present in the cryoEM images without needing further estimations from the user.

(1) Scheres et al., Nat. Met., 2007
(2) Herreros et al., Nat. Comm., 2023

The role of Instruct-ERIC in Europe

Corinna Brockhaus, Pauline Audergon, Claudia Alen Amaro, Natalie Haley, Harald Schwalbe

Instruct-ERIC, Oxford House, Parkway Court, John Smith Drive, Oxford, OX4 2JY, UK

Instruct-ERIC is a distributed infrastructure providing open access to high-end structural biology techniques to researchers from all countries to promote innovation in biomedical science in Europe. In addition to being an active partner in iNEXT-Discovery in which many Instruct centres provide access to their services, Instruct is a partner in two Horizon Europe projects: ISIDORe and canSERV. These projects offer funded access to cutting-edge services, from basic biology to clinical trials, to scientists in the field of infectious disease (ISIDORe) and cancer research (canSERV). Instruct also offers funding support towards access costs for researchers from its 16 member countries.

Moreover, Instruct offers a range of training opportunities for European scientists, enabling researchers to expand their expertise in structural biology and implement new techniques in their research.

Instruct aims to sustain and further extend funded access to structural biology techniques for European researchers through continuous participation in Horizon Europe projects as well as expansion of Instruct member countries.

Novel tools to analyze macromolecular heterogeneity and increase resolution by cryo-EM

Electron Microscopy at cryogenic temperatures is currently a very well-known approach to solve the structure of biological macromolecules with sufficiently high resolution to obtain good structural models. Part of this success was our proposal of Maximum Likelihood approaches to disentangle macromolecular structural flexibility (first implemente in the software XMIPP and then in Relion) (1). However, we then assumed that flexibility was discrete and that the user had to estimate the number of these discrete states, which clearly was not optimum. Nowadays, almost 15 years after this initial approach, we have developed new methods that allow us to consider continuous flexibility, opening the possibility to obtain conformational landscapes (2). In this context I will present recent advances in our "Zernike3D '' approach, showing how we effectively access the whole range of macromolecular flexibility present in the cryoEM images without needing further estimations from the user.

Structure determination of the magnesium ion free and bound KRas G12C+GDP complex using NMR data driven molecular dynamics simulations

Márton Gadanecz ^1,2, Zsolt Fazekas ^1,2, Gyula Pálfy, Dóra K. Menyhárd ¹ and András Perczel ¹

¹ Laboratory of Structural Chemistry and Biology, ELKH-ELTE Protein Modelling Research Group, Institute of Chemistry, Eötvös Loránd University, Pázmány Péter sétány 1/A, 1117 Budapest, Hungary
² Hevesy György PhD School of Chemistry, Eötvös Loránd University, Pázmány P. stny. 1/A, Budapest, H-1117 Hungary

In this work, catalytically significant states of the G12C oncogenic variant of K-Ras, those of  Mg²⁺-free and Mg²⁺-bound GDP-loaded forms, have been determined  using NMR data driven molecular dynamics (MD) simulations. There are multiple Mg²⁺-bound G12C K-Ras-GDP structures deposited in the Protein Data Bank (PDB) so this system was used as a reference, while the structure of the Mg²⁺-free, but GDP-bound state of the RAS-cycle has not been determined previously. Due to the high flexibility of the switch-I and switch-II regions – which happen to be the catalytically most significant segments also - only chemical shift information can be  collected for the most important regions of both systems.  Chemical Shift-Rosetta software was applied for to derive an “NMR ensemble” based on the measured chemical shift values, which however does not contain any non-protein components of the complex, such as ligands, cofactors or ions. To overcome this limitation, we tested different methodologies, where we aimed to retain the structural information from the CS-Rosetta ensemble while re-introducing missing, but structurally crucial non-protein components into the models. We have carried out Monte Carlo Multiple Minimum (MCMM) simulations, constrained according to the best CS-Rosetta models, but have encountered unexpected difficulties. We then developed torsional restraint set for all backbone torsions of the studied systems based on the CS-Rosetta ensembles for GROMACS MD simulations, overriding force-field-based parametrization with knowledge-based steering in the presence of the reinserted cofactors. This protocol resulted in chemically meaningful, complete models for both systems that also retained the structural features and heterogeneity defined by the measured chemical shift values, and allowed the detailed comparison of Mg²⁺-bound and Mg²⁺-free states of G12C K-Ras-GDP.

Accelerating structure-enabled drug discovery with high-throughput crystallographic fragment screening

D. Fearon, The COVID Moonshot Consortium, ASAP Discovery Consortium

Fragment-based drug discovery is a well-established method for the identification of chemical starting points which can be developed into clinical drugs. Historically, crystallographic fragment screening using traditional methods was low-throughput and time consuming. However, thanks to advances in synchrotron capabilities and the introduction of dedicated facilities, such as the XChem platform at Diamond, there has been substantial improvements in throughput and integration between sample preparation, data collection and hit identification.

The XChem team and collaborators have identified numerous fragments which bind to various antiviral drug discovery targets and leveraged our high-throughput platform to rapidly drive the hit-to-lead process to deliver pre-clinical candidates for SARS-CoV-2 in under 2 years.

Structure determination of the magnesium ion free and bound KRas G12C+GDP complex using NMR data driven molecular dynamics simulations

Márton Gadanecz ^1,2, Zsolt Fazekas ^1,2, Gyula Pálfy, Dóra K. Menyhárd ¹ and András Perczel ¹

¹ Laboratory of Structural Chemistry and Biology, ELKH-ELTE Protein Modelling Research Group, Institute of Chemistry, Eötvös Loránd University, Pázmány Péter sétány 1/A, 1117 Budapest, Hungary
² Hevesy György PhD School of Chemistry, Eötvös Loránd University, Pázmány P. stny. 1/A, Budapest, H-1117 Hungary

In this work, catalytically significant states of the G12C oncogenic variant of K-Ras, those of  Mg²⁺-free and Mg²⁺-bound GDP-loaded forms, have been determined  using NMR data driven molecular dynamics (MD) simulations. There are multiple Mg²⁺-bound G12C K-Ras-GDP structures deposited in the Protein Data Bank (PDB) so this system was used as a reference, while the structure of the Mg²⁺-free, but GDP-bound state of the RAS-cycle has not been determined previously. Due to the high flexibility of the switch-I and switch-II regions – which happen to be the catalytically most significant segments also - only chemical shift information can be  collected for the most important regions of both systems.  Chemical Shift-Rosetta software was applied for to derive an “NMR ensemble” based on the measured chemical shift values, which however does not contain any non-protein components of the complex, such as ligands, cofactors or ions. To overcome this limitation, we tested different methodologies, where we aimed to retain the structural information from the CS-Rosetta ensemble while re-introducing missing, but structurally crucial non-protein components into the models. We have carried out Monte Carlo Multiple Minimum (MCMM) simulations, constrained according to the best CS-Rosetta models, but have encountered unexpected difficulties. We then developed torsional restraint set for all backbone torsions of the studied systems based on the CS-Rosetta ensembles for GROMACS MD simulations, overriding force-field-based parametrization with knowledge-based steering in the presence of the reinserted cofactors. This protocol resulted in chemically meaningful, complete models for both systems that also retained the structural features and heterogeneity defined by the measured chemical shift values, and allowed the detailed comparison of Mg²⁺-bound and Mg²⁺-free states of G12C K-Ras-GDP.

Accelerating protein chemical shift assignment by deep learning for visual spectra analysis, structure and shift prediction

Piotr Klukowski¹, Roland Riek¹, Peter Güntert^1,2,3

¹ Laboratory of Physical Chemistry, ETH Zurich, Vladimir-Prelog-Weg 2, 8093 Zurich, Switzerland
² Institute of Biophysical Chemistry, Goethe University Frankfurt, Max-von-Laue-Str. 9, 60438 Frankfurt am Main, Germany
³ Department of Chemistry, Tokyo Metropolitan University, 1-1 Minami-Osawa, Hachioji, 192-0397 Tokyo, Japan

Chemical shift assignments are required for most protein NMR studies and often demand most of the measurement and analysis time. Here, we present a hybrid automated approach for protein chemical shift assignment that allows to reduce the number of spectra to be measured and the time to analyze them. It is based on machine learning for visual spectra analysis with ARTINA (1,2), structure prediction with AlphaFold2 (3), chemical shift prediction with UCBShift (4), and automated assignment with FLYA (5). Results from more than 10’000 assignment calculations with 100 proteins show that a small number of spectra suffices to establish the backbone and side-chain assignments. In conjunction with AlphaFold2 structures, the five 3D spectra 15N-NOESY, 13C-NOESY, CBCAcoNH, HCCH-TOCSY, and CCH-TOCSY yield on average better assignments than if ARTINA is run with all available spectra but without AlphaFold2 structures. NOESY spectra are particularly valuable for automated assignment. To be beneficial, structures should have an accuracy better than 2 Å.

This new version of ARTINA, which is available for use at the open NMRtist web server (2), offers users the general possibility to load, in addition to NMR spectra, 3D structures, manually or otherwise prepared peak lists, chemical shifts, distance restraints and torsion angle restraints as additional input for the ARTINA shift assignment and ARTINA structure determination applications. The new version of NMRtist also provides quality scores for peak picking, shift assignments, and structure calculations that are computed purely on the basis of the input data without recourse to manually obtained reference results.

1. Klukowski, P. et al. Nat. Commun. 13, 6151 (2022).
2. Klukowski, P. et al. Bioinformatics 39, btad066 (2023).
3. Jumper, J. et al. Nature 596, 583–589 (2021).
4. Li, J. et al. Chem. Sci. 11, 3180–3191 (2020).
5. 5. Schmidt, E. et al. Am. Chem. Soc. 134, 12817–12829 (2012).

...

AlphaFill: Enriching AlphaFold models with co-factors, small molecules and metal ions

Artificial Intelligence algorithms implemented in AlphaFold and the like are having a transformative effect on structural biology research: accurate predictions of 3D protein sequences can now be generated from the corresponding amino acid sequence only. The AlphaFold protein structure prediction database (AFDB) has millions of such predictions publicly available, and more are likely to come. However, biochemical interpretation of these predictions is limited to amino acids only. That is, co-factors, small molecules and/or metal ions, which are required for protein function or structural integrity are lacking.

To address this limitation we developed AlphaFill: an algorithm that enriches AlphaFold models with co-factors, small molecules and (metal) ions. Such compounds are transplanted from homologous experimental structures available in the PDB-REDO databank. Application of the AlphaFill algorithm to the core AFDB data has created some 586 thousands enriched models with over 12 million fitted compounds. The algorithm was validated against transplants obtained with 100% sequence identity.

All AlphaFill models are freely available through alphafill.eu: a new resource to support scientists interpreting the AlphaFold models while creating new hypotheses and designing experiments. AlphaFill also allows users to 'fill' there own structure models no matter if they are computational or experimental.

We will present the 'What and how' of AlphaFill, including judging the quality of enriched models, and will give a sneak preview of new developments to come.

High-resolution cryo-EM analysis of macromolecular complexes: when chemistry and biology meet

Centre for Integrative Biology, Department of Integrated Structural Biology, IGBMC, Illkirch, France.

Cryo electron microscopy (cryo-EM) is currently moving forward at high pace towards the high-resolution analysis of macromolecular complexes. Two key factors contributing towards that are (i) new-generation cryo electron microscopes that include improved energy filters and direct electron detectors (Fréchin et al., 2022), and (ii) new tools for advanced image processing including 3D classification methods to sort different structural states and refine sub-regions of the complexes of interest by focused refinements (Barchet et al., 2023). Using latest-generation instrumentation and focused classification and refinement methods we have now succeeded in crossing the 2 Å resolution barrier on a particularly difficult complex to analyse, the human ribosome, which is much less stable than its bacterial counterpart, with many flexible regions such as the 40S ribosomal subunit difficult to visualize at high resolution. Focused refinements and multibody refinements allowed obtaining a 1.9 Å resolution structure revealing numerous chemical modifications of the ribosomal RNA (rRNA), ions such as Zn²⁺, K⁺ and Mg²⁺ including water molecules of the Mg-octahedral coordination. Chemical modifications of the rRNA of the human ribosome comprise hundreds of nucleotides with 2’-O-methylations, pseudo-uridines and various base-specific modifications. They are involved in human protein synthesis dysregulations such as cancer (NAR Cancer, 2020) and other diseases but their role therein is unknown. The structure now reveals their precise mode of interaction, including series of 2’-O-Methylation and pseudo-uridine sites that could not be visualized in our previous study at 2.9 Å resolution (Natchiar et al., 2017). The integrated analysis of chemical modification, which also includes cutting-edge mass spectrometry of enzyme-induced rRNA fragments, provides unprecedented mechanistic insights into the translation mechanism in humans and paves the way to understanding the role of rRNA chemical modifications.

We will also present some highlights on the high-resolution structural analysis of a DNA-dependent bacteriophage and on new technological developments in super-resolution imaging including spectral demixing, which will offer new synergies in the context of integrated structural biology projects.

This project was run at the CBI, which hosts the national and European infrastructures FRISBI, Instruct-ERIC and iNEXT-Discovery. Within their framework, access can be provided to users to run experiments on protein production, single particle cryo-EM (sample optimization and high-resolution data collection) and FIB & tomography for the analysis of cellular samples: https://frisbi.eu  http://instruct-eric.com  https://inext-discovery.eu

References

S. Holvec, C. Barchet, A. Lechner, L. Fréchin, S. Nimali T. DeSilva, I. Hazemann, P. Wolff, O. von Loeffelholz & B. P. Klaholz. Structural role of chemical modifications and ions in the RNA architecture of the fully assembled mature human 80S ribosome. Submitted.
C. Barchet, L. Fréchin, S. Holvec, I. Hazemann, O. von Loeffelholz & B. P. Klaholz. Focused classifications and refinements in high-resolution single particle cryo-EM analysis. Under revision.
L. Fréchin, S. Holvec, O. von Loeffelholz, I. Hazemann & B. P. Klaholz. J. Struct. Biol., 2023, 215, 107905.
L. Andronov, R. Genthial, D. Hentsch, B. P. Klaholz. Commun Biol, 2022, 1-13
V. Marcel et al., Y. Motorin, B. P. Klaholz, A. Viari, J-J. Diaz. NAR Cancer, 2020, 2, 4
S. K. Natchiar, A. G. Myasnikov, H. Kratzat, I. Hazemann & B. P. Klaholz. Nature, 2017, 551, 472-477.
O. von Loeffelholz et al. & B. P. Klaholz. Curr. Opin. Struct. Biol., 2017, 46, 140-148.
B. P. Klaholz. Open Journal of Statistics, Special Issue on Multivariate Statistical Analysis, 2015, 5, 820-836.
H. Khatter, A. G. Myasnikov, K. Natchiar & B. P. Klaholz. Nature, 2015, 520, 640-5.

Towards comprehensive analysis of large crystallographic fragment screening campaigns

T. Krojer

MAX IV Laboratory, Lund University, PO Box 118, S-221 00 Lund, Sweden

Keywords: FragMAX, fragment screening, batch processing

The throughput of macromolecular X-ray crystallography experiments has surged over the last decade. Increases in the availability of high-intensity X-ray beams, fast detectors, and high levels of automation have permitted this extraordinary improvement in productivity. These advancements have allowed for the establishment of the FragMAX facility and several other specialized centres for crystal-based fragment screening, which enable preparation and collection of hundreds of single-crystal diffraction datasets per day. In addition, crystal structure determination has become significantly easier due to the availability of user-friendly software packages, which support users with different levels of experience from data processing to model building and structure refinement. However, simultaneous analysis of hundreds of related crystal structures, such as those present in fragment screening or structure-based drug design programs, remains a formidable problem because all major software suites adhere to the prevalent idea of "one project equals one structure". Moreover, fragment screening generates an abundance of meta-data that must be tracked for subsequent analysis and PDB deposition, but such functionality is currently not integrated in the available software packages.

FragMAXdb and FragMAXapp are two applications developed at MAX IV Laboratory to overcome this issue by facilitating comprehensive project management and parallel processing of hundreds of datasets from crystallographic screening campaigns. This presentation will cover their implementation, current functionalities, and highlight recent advancements. It will also outline potential future developments because there is an unmet need for more generic systems that can also support newer approaches such as (time-resolved) serial crystallographic studies. Such advancements will be essential if we are to realize the full potential of the enormous throughput of modern synchrotron beamlines and allow structural biologists to devote their valuable time to structure analysis rather than data management and processing. Expansion of existing and creation of new tools will not be simple but will greatly boost protein crystallographers' output.

The inner dealings of a tetrameric serine protesase:  calculations based on cryo-EM structures

A. J. Kiss-Szemán, I. Jákli, A. Perczel, V. Harmat, D. K. Menyhárd
Laboratory of Structural Chemistry and Biology, Institute of Chemistry, Eötvös Lórand University, Budapest, Hungary

Acyl-aminoacyl peptidase (AAP) is a homotetrameric serine-protease enzyme.  Its primary function is the removal of N-acylated amino acids from the N-terminus of proteins in preparation of their proteosomal degradation.  Through this, it plays a key role in the maturation and degradation processes of various proteins and peptides, functioning as a key enzyme of the protein quality control apparatus. As such, AAP has been indicated as a possible pharmacological target in various forms of cancer. AAP is the site of an unexpected drug-drug interaction too: carbapenem antibiotics were shown to interfere with a widely used anticonvulsant, valproate, via both forming crucial interactions with the enzyme in a competitive manner.

Recently we have determined the first structure of a mammalian AAP using cryo-EM, that of porcine liver AAP (pAAP). The structure is unique in the sense that the bacterial orthologues of AAP and other members of the S9 oligopeptidase family are monomeric, dimeric or hexameric in their active form. The tetrameric association seen in case of pAAP – confirmed as the functional form of the human variant too – is guided by amyloidogenic β-edges and mammalian-specific inserts and creates a toroid-shaped quaternary structure. The secluded nature of the substrate binding site, buried and protected by tetramerization, and the structural heterogeneity of the catalytic triad seems to be the key features potentiating the enzyme toward carbapenem antibiotics. In a subsequent work we have shown that binding of antibiotics requires the reorganization of the serine protease active site of pAAP which also leads to the irreversible inhibition of the enzyme. The cryo-EM structure of the pAAP-meropenem complex provides the first glimpse at the mechanisms by which antibiotics might produce side effects in human physiology – it is the first structure showing complex formation between a carbapenem antibiotic and (a very close homologue of) a human enzyme.

Protein Folding, Misfolding & Aggregation Studied by NMR Spectroscopy

Philipp Neudecker^1,2

¹Institut für Physikalische Biologie, Heinrich-Heine-Universität, 40225 Düsseldorf, Germany.
²IBI-7 (Strukturbiochemie) and JuStruct, Forschungszentrum Jülich, 52425 Jülich, Germany.

Protein folding, misfolding and aggregation are tightly interconnected processes involving a variety of conformational states with very different stability and lifetimes. NMR spectroscopy is particularly well suited to characterize the structure and dynamics of proteins on the relevant time scale (milliseconds to real time). We used CPMG relaxation dispersion NMR spectroscopy to elucidate the kinetics, thermodynamics and structural changes of the folding pathway of the Fyn SH3 A39V/N53P/V55L domain. The atomic-resolution three-dimensional solution structure of the 2% populated, on-pathway folding intermediate provides a detailed characterization of the non-native interactions stabilizing an aggregation-prone intermediate under native conditions and insight into how such an intermediate can derail folding and initiate fibril formation [1].

Many of the proteins forming amyloid fibrils in neurodegenerative diseases such as Amyloid-beta (Abeta) or the human Prion Protein (huPrP) are partially or fully disordered in their monomeric state and therefore readily characterized by NMR spectroscopy in solution. The truncation variant huPrP(23-144), which has been reported to cause a Gerstmann-Sträussler-Scheinker-like disease with amyloid deposits in the brain, is devoid of any stable secondary structure in solution [2,3]. In addition to causing Transmissible Spongiform Encephalopathies, huPrP is also a high-affinity receptor for Abeta oligomers that has been suggested to contribute to Abeta toxicity in Alzheimer's Disease, but may also play a protective role by sequestration of Abeta oligomers. Comparison of the solution NMR spectra of the monomers and the solid-state NMR spectra of the large Abeta-huPrP heteroaggregates reveals that the N-terminus of huPrP becomes immobilized in the complex without adopting a regular secondary structure. By contrast, the Abeta oligomer preparation represents a heterogeneous mixture of beta-strand-rich assemblies, of which some have the potential to evolve and elongate into different fibril polymorphs [3].

1. Neudecker, P. Robustelli, A. Cavalli, P. Walsh, P. Lundström, Arash Zarrine-Afsar, S. Sharpe, M. Vendruscolo & L. E. Kay: Structure of an Intermediate State in Protein Folding and Aggregation, Science 336, 362-366 (2012)
2. S. Rösener, L. Gremer, E. Reinartz, A. König, O. Brener, H. Heise, W. Hoyer, P. Neudecker & D. Willbold: A D-enantiomeric peptide interferes with heteroassociation of amyloid-beta oligomers and prion protein, J. Biol. Chem. 293, 15748-15764 (2018)
3. A. S. König, N. S. Rösener, L. Gremer, M. Tusche, D. Flender, E. Reinartz, W. Hoyer, P. Neudecker, D. Willbold & H. Heise: Structural details of amyloid β oligomers in complex with human prion protein as revealed by solid-state MAS NMR spectroscopy, J. Biol. Chem. 296, 100499 (2021)

Antirepressor specificity is shaped by highly efficient dimerization of the repressors in regulation of staphylococcal pathogenicity islands

Kinga Nyíri^1,2,*, Enikő Gál^1,2, Máté Laczkovich^1,2, Beáta G. Vértessy^1,2

¹ Department of Applied Biotechnology and Food Science, Faculty of Chemical Technology and Biotechnology, Budapest University of Technology and Economics, Műegyetem rkp. 3., H-1111 Budapest, Hungary, ² Institute of Enzymology, Research Centre for Natural Sciences, Hungarian Academy of Sciences, Budapest, Magyar tudósok krt 2. H-1117, Hungary, ^* To whom correspondence should be addressed. Email: This email address is being protected from spambots. You need JavaScript enabled to view it.

Pathogenicity islands in Staphylococci are regulated by Stl repressors that form strong dimers. It has been recently shown that SaPIbov1-Stl dimers are separated during the activation of the Staphylococcus aureus pathogenicity island (SaPI) transcription via helper phage proteins. To understand the mechanism of this regulation, a quantitative analysis of the dimerization characteristics is required. Due to the highly efficient dimerization process, such an analysis has to involve specific solutions that permit relevant experiments to be performed.

In the present work, we focused on two staphylococcal Stls associated with high biomedical interest, namely Stl proteins of Staphylococcus aureus bov1 and Staphylococcus hominis ShoCI794_SEPI pathogenicity islands. Exploiting the interactions of these two Stl proteins with their antirepressor-mimicking interaction partners allowed precise determination of the Stl dimerization constant in the subnanomolar range.

Consequently only an intricate, fine-tuned interaction network between repressor:antirepressor complexation can adeptly compete with the highly efficient dimerization of the Stl repressors. In addition we showed at the first time that the strong interaction within the Stl repressor dimer dramatically interferes with the measurement of repressor:antirepressor equilibrium binding constant. Our results thus propose a generally applicable experimental approach to study other similar complexes.

Cryo-EM structures of tau filaments from human brain

The assembly of microtubule-associated protein tau into abundant filamentous inclusions underlies many neurodegenerative diseases called tauopathies. The groups of Sjors Scheres & Michel Goedert established atomic structure determination of amyloid filaments by electron cryo-microscopy (cryo-EM), and used this method to determine the structures of tau filaments extracted from the brains of individuals with Alzheimer’s disease, Pick’s disease, chronic traumatic encephalopathy (CTE), corticobasal degeneration (CBD), progressive supranuclear palsy (PSP) and other tauopathies. Specific molecular conformations of tau characterize different diseases, providing the basis for a structure-based classification of tauopathies. Using cryo-EM, I show that tau filaments from Amyotrophic Lateral Sclerosis/Parkinsonism-Dementia complex (ALS/PDC) and Subacute Sclerosing Panencephalitis (SSPE) adopt the same structures as observed in CTE, suggesting that neurodegenerative diseases that are caused by environmental factors share similar molecular mechanisms of ilament formation, with neuroinflammation possibly playing an important role.

Phase Transition from Liquid to Solid (amyloids): from the origin to the end of life

Proteins may aggregate into amyloid fibrils. This phase transition is associated with neurodegenerative diseases. However also functional amyloids exist. The structure activity relationship studies of both Alzheimer’ and Parkinson-related amyloids as well as functional amyloids will be presented. While the latter amyloids are reversible, the disease-associated ones comprise the so called mechanism of secondary nucleation which is a positive feed back loop accelerating the aggregation beyond control. In addition, since the structures have not been evolved they are highly environment-dependent yielding many distinct structural polymorphs like a chameleon. These concepts are illustrated based on physico chemical experiments and high resolution structural work including NMR and cryo EM. Finally, hypothesis that peptide amyloids may have been played an important role in the origin of life is investigated.

Using integrated structural biology to determine the specificity in RNA-protein interactions

The regulation of gene expression is majorly steered on the level of proteins interacting with nucleic acids. An essential contribution is found in the formation of regulatory complexes between proteins and RNAs. Seeing the unexpected large number of approximately 2000 human RNA-binding proteins (RBPs) interacting with the limited chemical space of nucleotides, we are still at the beginning of understanding the hidden rules of specificity between RNAs and proteins. As some of the most basic concepts RBPs by default use multiple specialized RNA-binding domains (RBDs) to increase specificity. Also, RNAs represent specificity as targets via complex folds and intrinsic flexibility, which are often impossible to predict from their sequences.

High-resolution structures of RNA-protein complexes (RNPs) have and will yield relevant insights into the molecular parameters for specificity. Cryo-EM has tremendously enlarged the space for RNP structure determination of macromolecular complexes. However, large RNAs remain challenging in light of their dynamic heterogeneity, and the same is true for intrinsically disordered proteins that have recently emerged as RBPs. Some of those bottlenecks may be overcome with the integration of multiple methods, including structure prediction via molecular modeling and simulations. Modern approaches may thus, on the longer run, also help in the rational design of proteins and RNAs to specifically interact with each other.

In this lecture I will give an overview of challenges in analyzing RNPs on a structural level as well as a short introduction into combined experimental approaches applicable to this end. I will use examples from our own lab’s previous and current projects, including different types of RBP, RBDs, and RNA elements. A focus will be on the interaction of highly specialized proteins with regulatory regions in mRNAs, which we investigate using X-ray crystallography; but more extensively with the solution methods NMR spectroscopy and small-angle X-ray scattering (SAXS), supported by solid wet-lab biochemistry and biophysics. The lecture will provide insight into the complementary utilization of methods to comprehensively study RNPs starting from structural information.

Common structural features of wild type and variant ATTR amyloid fibrils extracted from different patients

Matthias Schmidt, Maximilian Steinebrei, Julian Baur, Ute Hegenbart, Stefan Schönland, Marcus Fändrich

Introduction:

Wild type transthyretin-derived amyloid is the major component of non-hereditary transthyretin amyloidosis. Its accumulation in the heart of elderly patients is life threatening. A variety of genetic variants of transthyretin can lead to hereditary transthyretin amyloidosis, which shows different clinical symptoms, like age of onset and pattern of organ involvement. However, in the case of non-hereditary and in several cases of hereditary ATTR amyloidosis fibril deposits are located primarily in heart tissue and cause cardiac failures.

Objectives:

The goal of this project is the analysis of wild type and variant ATTR amyloid fibril structures from the heart of different patients with non-hereditary or hereditary transthyretin amyloidosis and comparison with previous published ATTR fibril structures.

Materials & methods:

Fibrils were extracted from the amyloidotic tissue of an explanted heart of wild type and variant ATTR patient. Cryo electron microscopy with subsequent structure modelling and mass spectrometry was used to study these fibrils.

Results:

A density map of ex vivo ATTR fibrils with resolutions of 2.78 Å (wild type), 3.18 Å (V20I), 2.37 Å (G47E) and 2.99 Å (V122I) from different patients were reconstructed. The structures are formed by stacked N- and C-terminal fragments of transthyretin forming an amyloid fibril which was confirmed with mass spectrometry. All structures show a similar spearhead like shape in their cross-sectional view, formed by the same N- and C-terminal fragments with some minor differences. The N-terminal fragment has an extensive hydrophobic core. The C-terminal fragment on the other hand contains a big cavity surrounded by polar and charged amino acids and is probably filled with water.

Conclusion:

We compare the structure of ATTRwt fibrils with variant ATTR fibril structures (V30M, G47E, V122I and V20I) extracted from hearts and eye of different patients (Steinebrei, M. et al. (2022) Nat. Commun.; Schmidt, M. et al. (2019) Nat. Commun.; Iakovleva, I. et al. (2021) Nat. Commun.). All studied structures show a remarkably similar cross-sectional shape of the N- and C-terminal fragments with only some minor differences. This demonstrates common features for ATTR fibrils despite differences in mutations and patients. Additionally, the N- and C-terminal fragments found in all cases points to that amyloid formation starts with complete or partly unfolded native TTR peptides, followed by fibril formation of the full-length ATTR peptides and subsequent proteolytic cleavage at the accessible region between amino acid 44 and 58.

RNA structural biology by NMR spectroscopy

Harald Schwalbe, Institute for Organic Chemistry and Chemical Biology, Center for Biomolecular Magnetic Resonance (BMRZ), Instruct-ERIC

Structural determination of RNA is difficult due to the intrinsic structural dynamics of RNA.

In this contribution, we will present work of joint MD/NMR methods to generate structural models describing the conformational heterogeneity of RNA (Oxenfarth et al, under evaluation).

The structural dynamics allows targeting RNAs by low molecular weight ligands.

Examples derived from the viral RNAs of SARS-CoV-2 will be discussed.

In-Cell NMR as a tool in Cellular Structural Biology

Remarkable developments in cellular in situ methods, including cryo-electron tomography (cryoET), cross-linking mass spectrometry (CLMS), Förster resonance energy transfer (FRET), nuclear magnetic resonance (NMR) and electron paramagnetic resonance (EPR) spectroscopies have clearly shown that the future of Structural Biology is in the cell. Together, these methods delivered unprecedented insights into the structure & function of proteins, membranes and metabolites directly in intact specimens, with current efforts geared towards improving their robustness, sensitivity, and overall ease-of-use. In my talk, I discuss the role of in-cell NMR spectroscopy in the era of Cellular Structural Biology and how integrated approaches with other methods offer enticing possibilities for future applications. In particular, I highlight recent discoveries that in-cell NMR enabled us to make and how they (re)shaped our understanding of fundamental processes in cellular biology.

In-cell structural biology using NMR: overview and latest developments to depict IDPs at 310K

Abstract:

In-cell structural biology by NMR is appealing in many regards: It proposes, among others, to investigate conformational equilibria or ligand binding/processing in very relevant conditions, i.e in cells [1,2]. We will briefly describe the passed and present experimental conditions exploited in the field, and give an overview of the contributions and limits of in-cell NMR.

The approach comes with a number of challenges, among which i) the many difficulties in sample production, and ii) important signal losses due to promiscuous, transient interactions with cellular entities, which, in turn, urges to use (too) high concentrations of the studied proteins. We will show how we and others are trying to facilitate in-cell NMR studies, using new production methods in situ, new labeling schemes, and better adapted pulse sequences.

References:

[1] Theillet F.-X. In-Cell Structural Biology by NMR: The Benefits of the Atomic-Scale. Chem. Rev. 122 (2022). https://doi.org/10.1021/acs.chemrev.1c00937

[2] Theillet F.X., Luchinat E. In-cell NMR: Why and how? Prog Nucl Reson Magn Spectr. 132-133 (2022). https://doi.org/10.1016/j.pnmrs.2022.04.002

Non-canonical structural motifs of nucleic acids

Viktor Viglasky and Lukas Trizna
Department of Biochemistry, Institute of Chemistry, Faculty of Sciences, Pavol Jozef Šafárik University, 04001 Košice, Slovakia

Nucleic acids can adopt various secondary structures, depending mainly on their sequence. These structural forms play a key role in the regulation of gene expression in cells due to their unique properties. A major challenge is to identify such sequences in the genome in order to predict their formation and to find a specific target molecule to either facilitate or eliminate occurrence of such structures. An alternative approach to search for putative sequences has recently been developed in our laboratory [1,2]. In addition, non-canonical motifs are contained in many aptamers that are applied in biomedical research as sensing molecules and have recently been used to functionalize next-generation nanoparticles.

Our recent advances in the structural stability of tetrahelical non-canonical forms of nucleic acids will be presented, as well as an entirely new motif predicted by us. In addition, DNA nanocircles and three-way junctions consisting of noncanonical motifs will be described.

References:

1. Trizna L, Osif B, Víglaský V. G-QINDER Tool: Bioinformatically Predicted Formation of Different Four-Stranded DNA Motifs from (GT)n and (GA)n Repeats. Int J Mol Sci. 2023, 24, 7565
2. Víglaský V. Hidden Information Revealed Using the Orthogonal System of Nucleic Acids.

Int J Mol Sci. 2022, 23, 1804.