POSTER NO 1: Amanda Duncan Due

Interplay between motifs, activation domains and coregulators in the ID landscape of transcription factors

Amanda D. Due (a,b), Steffie Elkjær (a), Andreas Prestel (a,b), Nicholas Morffy (c), Lucia C. Strader (c), Norman E. Davey (d), Karen Skriver (a) and Birthe B. Kragelund (a,b)

(a) REPIN and the Linderstrøm-Lang Centre for Protein Science; (b) Structural Biology and NMR Laboratory

Department of Biology, University of Copenhagen, DK-2200 Copenhagen, Denmark

(c) Department of Biology, Duke University, Durham, NC, USA

(d) Division of Cancer Biology, The Institute of Cancer Research, 237 Fulham Road, London SW3 6JB, UK

Contact: amanda.due@bio.ku.dk

High-throughput data on transcription factor (TF) activation domains (ADs) are emerging at high speed, but mechanistic insight is still missing. We utilize a clade from the TF family NAM, ATAF1/2 and CUC2 (NAC). The family regulates a broad set of responses. Each TF is composed of a highly conserved NAC domain for DNA binding and a large intrinsically disordered region (IDR) for regulatory purposes. We ask how the interplay between conserved motifs and activation domains works within the long IDR of TFs. We further focused on Arabidopsis thaliana NAC 046 (ANAC046) to study the interplay in the context of intra- and intermolecular interactions, the latter in relation to a set of coregulators.

POSTER NO 2: Andreas Prestel

NMR experiments to determine proline cis/trans  isomer specific properties of intrinsically disordered regions

Andreas Prestel1,2, Frederik F. Theisen1,2, Nina L. Jakobsen1,2, Oline K. Nyhegn-Eriksen1,2, Karen Skriver1, Birthe B. Kragelund1,2

1 The REPIN and The Linderstrøm-Lang Centre for Protein Science, Department of Biology, University of Copenhagen, Copenhagen DK-2200, Denmark.2 Structural Biology and NMR Laboratory, Department of Biology, University of Copenhagen, Copenhagen DK-2200, Denmark

Contact: andreas.prestel@bio.ku.dk

Proline (P) is unique among the proteinogenic amino acids in that the energy difference between trans and cis XP peptidyl bond is small and therefore the cis state is significantly populated (3-40%, depending on the peptide sequence).[1] Furthermore, the activation energy for isomerization is relatively high (~20 kcal/mol) and for intrinsically disordered regions (IDRs) with x prolines there are theoretically 2x different isomers present, which are slowly interconverting.[1] Many biophysical methods are averaging the properties of this heterogeneous ensemble, while NMR has the unique power to extract isomer specific information without the need for additional labels, since the chemical shifts of nuclei in and around the XP peptidyl bond are sensitive to the isomerization state. Here we present a selection of NMR experiments to determine populations and interconversion rates as well as to extract isomer specific properties of proline containing peptides and IDRs.

POSTER NO 3: Andrew Philip Rennison

Digging into the surface dynamics of PET hydrolases

Andrew Philip Rennison and Aimilia Nousi, Marie Sofie Møller, Peter Westh Rodolphe Marie

DTU Bioengineering, DTU health Tech

Contact: anrenn@dtu.dk

PET hydrolases are an emerging class of enzymes that are being heavily researched for their use in bioprocessing polyethylene terephthalate (PET). While work has been done in studying the binding of PET oligomers to the active site of these enzymes, the dynamics of PET hydrolases binding to a bulk PET surface is an unexplored area. Here, methods were developed for total internal reflection fluorescence (TIRF) microscopy and fluorescence recovery after photobleaching (FRAP) microscopy to study the adsorption and desorption dynamics of these proteins onto a PET surface. TIRF microscopy was employed to measure both on and off rates of two of the most commonly studied PET hydrolases, PHL7 and LCC, on a PET surface. It was found that these proteins have a much slower off rates on the order of 10-3s-1, comparable to non-productive binding in enzymes such as cellulose, which we suggest demonstrates a non-specific binding mechanism. In combination with FRAP microscopy, a dynamic model is proposed in which adsorption and desorption dominates over lateral diffusion over the surface. The results of this study could have implications for the future engineering of PET hydrolases, to target them to a PET surface, or to improve interaction with their substrate.

POSTER NO 4: Ankush

What properties of biomolecular condensates influence  oxygen gradients?

Ankush Garg, Klaus Koren and Magnus Kjærgaard*

MBG, Aarhus university, Denmark

Contact: au711045@uni.au.dk

Poster abstract is not public.

POSTER NO 5: Annette Juma Nielsen

Development of a high throughput screening platform for weak interaction partners to a protein of interest utilizing mRNA-display

Annette Juma Nielsen (1), Joseph Matthew Rogers (2) & Kaare Teilum (1)

1) Structural Biology and NMR laboratory, Department of Biology, University of Copenhagen. 2) Department of Drug Design and Pharmacology, University of Copenhagen

Contact: Annette.juma@bio.ku.dk

Poster abstract is not public.

POSTER NO 6: Arriën Symon Rauh

A Molecular Dynamics Model for Studying Disordered Proteins in Crowded Environments.

Arriën Symon Rauh, Giulio Tesei and Kresten Lindorff-Larsen

Structural Biology and NMR Laboratory, Linderstrøm-Lang Centre for Protein Science, Department of Biology, Copenhagen University, Copenhagen N, 2200, Denmark.

Contact: arrien.rauh@bio.ku.dk

Intrinsically disordered proteins (IDPs) are sensitive to their crowded environment. To get a detailed understanding of how crowding influences IDP dynamics and phase separation (PS) behaviour, we introduce polyethylene glycol (PEG) in the effective residue-level coarse-grained molecular dynamics (CG MD) model CALVADOS [1,2]. We optimised model parameters for single-chain PEG properties and PEG-induced protein compaction [3,4]. Subsequently, we examined phase separation (PS) in IDPs, starting with the well-studied hnRNPA1 LCD. We observed a linear response to PEG size, enabling us to differentiate intrinsic phase separation tendencies in A1 variants [5]. Finally, we show ⍺-synuclein phase separation behaviour in line with experiments [6].

References

1.                        G. Tesei et al.  Accurate model of liquid-liquid phase behavior of intrinsically disordered proteins. PNAS. 2021. 118(40)

2.                        G. Tesei and K. Lindorff-Larsen  Improved predictions of phase behaviour of intrinsically disordered proteins by tuning the interaction range. Open Research Europe. 2022. 2:94

3.                        N. Sherck et al.  End-to-End Distance Probability Distributions of Dilute Poly(ethylene oxide) in Aqueous Solution. JACS. 2020. 142(46):19631-19641

4.                        A. Soranno et al.  Single-molecule spectroscopy reveals polymer effects of disordered proteins in crowded environments. PNAS. 2014. 111(13):4874-4879

5.                        E. Martin et al.  Valence and patterning of aromatic residues determine the phase behavior of prion-like domains. Science. 2020. 376(6478):694-699

6.                        S. Ray et al.  ⍺-Synuclein aggregation nucleates through liquid–liquid phase separation. Nature Chemistry. 2020. 12(8):705-716

POSTER NO 7: Azad Farzadfard

Thermodynamic characterization of amyloid polymorphism by Taylor dispersion analysis

Azad Farzadfard1,2, Antonin Kunka1, Thomas Oliver Mason1, Jacob Aunstrup Larsen1, Rasmus Krogh Norrild1, Elisa Torrescasana Dominguez1, Soumik Ray1, and Alexander K. Buell1*

1-                        Protein Biophysics group, Department of Biotechnology and Biomedicine, Technical University of Denmark, Søltofts Plads, Building 227, 2800, Kgs. Lyngby, Denmark

2-                        Interdisciplinary Nanoscience Center (iNANO), Aarhus University, Gustav Wieds Vej 14, 8000, Aarhus C, Denmark

Contact: seyfa@dtu.dk

Amyloid fibrils of proteins such as α-synuclein are a hallmark of neurodegenerative diseases and much research has focused on their kinetics and mechanisms of formation. The question as to the thermodynamic stability of such structures has received much less attention. Here, we present a novel experimental method to quantify amyloid fibril stability based on chemical depolymerisation and Taylor dispersion analysis. The relative concentrations of fibrils and monomer at equilibrium are determined through an in situ separation of these species through Taylor dispersion in laminar flow inside a microfluidic capillary. This method is highly sample economical, using much less than a microliter of sample per data point and its only requirement is the presence of aromatic residues because of its label-free nature. Using this method, we investigate the differences in thermodynamic stability between different fibril polymorphs of α-synuclein and quantify these differences for the first time. Importantly, we show that fibril formation can be under kinetic or thermodynamic control and that a change in solution conditions can both stabilise and destabilise amyloid fibrils. Taken together, our results establish the thermodynamic stability as a well-defined and key parameter that can contribute towards a better understanding of the physiological roles of amyloid fibril polymorphism.

POSTER NO 8: Christian Bernsen Borg

Exploring the role of N-glycosylation as a determinant of ATG9A conformation and activity

Christian B. Borg (a), Matteo Lambrughi (a), Mattia Utichia (a,b), Sergio Esteban Echeverría (a), Elisa Fadda (c), Kenji Maeda (d), Marja Jäättelä (d,e), and Elena Papaleo (a,b)

(a): Cancer Structural Biology, Center for Autophagy, Recycling and Disease, Danish Cancer Institute, Copenhagen, Denmark

(b): Cancer Systems Biology, Section for Bioinformatics, Department of Health and Technology, Technical University of Denmark, Lyngby, Denmark

(c): Department of Chemistry, Maynooth University, Ireland

(d): Cell Death and Metabolism, Center for Autophagy, Recycling and Disease, Danish Cancer Institute, Copenhagen, Denmark

(e): Department of Cellular and Molecular Medicine, Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark

Contact: cbb@cancer.dk

Autophagy is an evolutionarily conserved cellular process in which cells degrade and recycle their own components. During autophagy organelles and proteins are sequestered by the formation of a double-membrane vesicle, the autophagosome, and delivered to the lysosome for degradation. Modulation of autophagy has been highlighted as an attractive therapeutic avenue targeting diseases associated with cellular dyshomeostasis, such as cancers. Several aspects of the autophagic pathway, however, remain elusive [1]. The membrane embedded protein ATG9A plays a pivotal role in the autophagosome formation [2]. Despite the recent structural characterization and description of the ATG9A phospholipid scrambling activity [3-5],  the molecular details underlying ATG9A function and contribution to autophagosome formation is still not understood . Cryo-EM structures showed that ATG9A is homotrimeric and exists in at least two distinct states. In addition, ATG9A possesses a single complex type (N)-linked glycosylation (N-glycosylation) at Asn99, located in a loop which is not resolved in the available cryo-EM structures [2]. The role of this N-glycosylation in ATG9A function and dynamics, such as lipid scrambling at the phagophore and trafficking is yet to be defined. Here, we are working with a cross-disciplinary approach combining structural analysis and computational modeling methods, and cell based assays to elucidate the role of the ATG9A N-glycosylation. We perform molecular dynamics (MD) simulations of ATG9A homotrimer in lipid bilayers with and without the presence of Asn99 N-glycosylation to investigate ATG9A protein structure and dynamics and use experimental cell-based assays to study the impact of ATG9A mutants abolishing N-glycosylation on protein localization and autophagy flux. Results from our MD simulations of the ATG9A homotrimer show that N-glycosylation affect the “open-to-close” transitions at the ATG9A protomer interfaces and we observe a higher occurrence of “open-like” states in ATG9AN99glyco compared to non-glycosylated ATG9A. This suggests that N-glycosylation can influence the lipid accessibility to the ATG9A central cavity. Overall, our results aid over understanding of how N-glycosylation can regulate  ATG9A, a central protein in the autophagic pathway.

1.                        Klionsky, D.J., et al., Autophagy in major human diseases. EMBO J, 2021. 40(19): p. e108863.

2.                        Young, A.R., et al., Starvation and ULK1-dependent cycling of mammalian Atg9 between the TGN and endosomes. J Cell Sci, 2006. 119(Pt 18): p. 3888-900.

3.                        Matoba, K., et al., Atg9 is a lipid scramblase that mediates autophagosomal membrane expansion. Nat Struct Mol Biol, 2020. 27(12): p. 1185-1193.

4.                        Maeda, S., et al., Structure, lipid scrambling activity and role in autophagosome formation of ATG9A. Nat Struct Mol Biol, 2020. 27(12): p. 1194-1201.

5.                        Guardia, C.M., et al., Structure of Human ATG9A, the Only Transmembrane Protein of the Core Autophagy Machinery. Cell Rep, 2020. 31(13): p. 107837.

POSTER NO 9: Emil G. P.  Stender

Characterising aggregation prone mulitvalent complex formation

Emil G. P.  Stender, PhD

Field Application Scientist at Fidabio

Contact: egps@fidabio.com

Multivalent interactions are of great importance in cellular regulation and immune response.However, the ability of multivalent molecules to form agglutination prone hetero-oligomeric networks makes it difficult to decouple the agglutination process from the direct interaction, which can skew the measured affinity and biophysical properties of the system. This makes it very challenging, if not impossible to study, these kinds of interactions using most biophysical techniques.

In this poster we will demonstrate how the Fida 1 can be used to probe the initial interaction as well as the impact of agglutination between bovine β-lactoglobulin (BLG) and a anti BLG rabbit IgG by using flow induced dispersion analysis with capillary mix. This is achieved by keeping the labelled protein of interest separate from the analyte in the sample holder and only mixing them in the capillary during the actual measurement. The adjustable mobilisation pressure then allows for control of the complex lifetime before detection allowing for quantification of interaction, aggregation, soluble agglutination, complex size and Kd.

POSTER NO 10: Fabian Schuhmann

Introducing Similarity Measures for Biological Systems (SiMBolS)

Fabian Schuhmann (1,2), Leonie Ryvkin (3), James D. McLaren (2), Luca Gerhards (2), Ilia A. Solov'yov (2)

(1) Niels Bohr International Academy, University of Copenhagen

(2) Carl von Ossietzky University Oldenburg

(3) Technical University Eindhoven

Contact: fabian.schuhmann@nbi.ku.dk

Biological processes involve movements across all measurable scales which need to be analyzed and understood to derive Nature’s reasoning and to understand the potential function of the studied object. Especially in molecular dynamics (MD) simulations, considerable resources are allocated to get a picture of the motion of a protein. Having invested heavily in the generation of the data, the analysis methods need to be precise enough to extract every last bit of information.

While one can easily compare a protein structure to a reference employing tools like, for instance, the Root Mean Square Deviation, methods need to become more involved to compare two whole trajectories. How does one spot the difference among thousands of atoms, which are all wiggling and moving in a stop-motion movie?

To solve the problem, we have gathered eight different similarity measures in an easy-to-use Python package called SiMBols. SiMBols includes the Hausdorff distance (HD), the (weak) Fréchet distance ((W)DFD), dynamic time warping (DTW), Longest Common Subsequence (LCSS), a difference distance matrix approach (DDM), Wasserstein distance (WD), and Kullback-Leibler divergence (KLD) and combines them in a unified way including basic checks to find out if the loaded data set is feasible for a certain measure. Additionally, SiMBols includes protein preprocessing routines to read simulation files and superimpose structures.

Employing Pigeon Cryptochrome 4 (6PU0) as a case study example, SiMBols will be used to compare two states of this putative biological magnetoreceptor and comparing the different similarity measures.

We will find that the different similarity measures differ in their computation time and in the fundamental research question they might answer.

POSTER NO 11: Fenne Dijkema

Live fast, die young: rock'n roll reporter Gaussia luciferase goes out with a Flash

Fenne Marjolein Dijkema

University of Copenhagen

Contact: fenne@bio.ku.dk

In nature, the protein Gaussia luciferase is used to scare off predators through an unexpected flash of light. It is produced by the 1-2 mm large copepod Gaussia princeps, which secretes the luciferases and its substrate coelenterazine into the sea water when disturbed. Although Gaussia luciferase has been used as a reporter for many years, it is only recently that it has become characterized beyond only its sequence.

We have shown that it is a protein with many unusual properties.

First of all, it auto-inactivates within the 200 first reaction cycles.

We showed that this inactivation coincides with a number of oxidations of the protein by radical oxygen species, as well as the appearance of up to three adducts of a quinoid nature per protein, derived from the reaction product coelenteramide.

Secondly, we determined the structure of the protein by NMR and found it to contain large disordered regions, making it the most dynamic catalytically active protein that we know of.

Finally, the protein displays substrate cooperativity, despite lacking multiple binding sites that could cause an allosteric effect. Combined with its dynamic structure, the most likely explanation of this is that this is an example of kinetic cooperativity, in which substrate-dependent conformational change on the same timescale as the catalytic turnover causes a change in the apparent reaction rate.

Gaussia luciferase's natural environment offers an elegant explanation for its unusual properties:

it carries out its function right after being secreted into the open ocean, where it will quickly diffuse away from its substrate. Therefore, evolutionary pressure would only apply to the light output in the first seconds of the reaction. In this way, a small protein size and rapid turnover could have been prioritized over longevity of the protein, which would be discarded anyway.

POSTER NO 12: Francesco Pesce

Design of intrinsically disordered protein variants with diverse structural properties

Francesco Pesce (1), Anne Bremer (2), Giulio Tesei (1), Jesse B. Hopkins (3), Christy R. Grace (2), Tanja Mittag (2), Kresten Lindorff-Larsen (1)

(1) Structural Biology and NMR Laboratory, The Linderstrøm-Lang Centre for Protein Science, Department of Biology, University of Copenhagen, Copenhagen, Denmark;

(2) Department of Structural Biology, St. Jude Children’s Research Hospital, Memphis, TN 38105, USA;

(3) BioCAT, Department of Physics, Illinois Institute of Technology, Chicago, IL, USA.

Contact: francesco.pesce@bio.ku.dk

Intrinsically disordered proteins (IDPs) perform a wide range of functions in biology, suggesting that the ability to design IDPs could help expand the repertoire of proteins with novel functions. Designing IDPs with specific structural or functional properties has, however, been difficult, in part because determining accurate conformational ensembles of IDPs generally requires a combination of computational modelling and experiments. Motivated by recent advancements in efficient physics-based models for simulations of IDPs, we have developed a general algorithm for designing IDPs with specific structural properties. We demonstrate the power of the algorithm by generating variants of naturally occurring IDPs with different levels of compaction and that vary more than 100 fold in their propensity to undergo phase separation, even while keeping a fixed amino acid composition. We tested experimentally designs of variants of the low-complexity domain of hnRNPA1 and find high accuracy in our computational prediction, both in terms of single-chain compaction and propensity to undergo phase separation. We analyze the sequence features that determine changes in compaction and propensity to phase separate and find an overall good agreement with previous knowledge for naturally occurring sequences. Our general, physics-based method enables the design of disordered sequences with specified conformational properties. Our algorithm thus expands the toolbox for protein design to include also the most flexible proteins and will enable the design of proteins whose functions exploit the many properties afforded by protein disorder.

POSTER NO 13: Frans Zdyb

Procrustes likelihood for deep generative models of protein structure

Frans Zdyb, Christian B. Thygesen, Ola Rønning,  Christian Skjødt Steenmans, Anders Bundgård Sørensen, Kanti V. Mardia, John T. Kent, Thomas Hamelryck

University of Copenhagen, Copenhagen, Denmark

Contact: frzd@di.ku.dk

Deep generative models of proteins either use rotation-invariant representations, such as dihedral angles, or 3D coordinates combined with rotation invariance. We propose a multi-scale likelihood that includes both. We use a sine bivariate von Mises (SBVM) likelihood for the former, and a Procrustes likelihood that features latent rotations and standard deviations of the 3D positions for the latter.

POSTER NO 14: Freia Buus

NMR Spectroscopy Analysis of the Phase-Separation of Two Highly Charged Disordered Proteins

Freia Stryhn Buus1,2, Nicola Galvanetto3,4, Jacob H. Martinsen1,2, Andreas Prestel1, Benjamin Schuler3,4, Birthe B. Kragelund1,2.

1REPIN, 2Structural Biology and NMR Laboratory, The Linderstrøm-Lang Centre for Protein Science, Department of Biology, University of Copenhagen, Denmark, 3Department of of Biochemistry, University of Zurich, Zurich, Switzerland, 4Department of Physics, University of Zurich, Zurich, Switzerland.

Contact: freia.buus@bio.ku.dk

Prothymosin-a and the linker histone H1.0 interaction can drive phase-separation.

The interaction between the two human disordered proteins, linker histone H1.0 (H1) and the nuclear chaperone, prothymosin-a (ProTa), is notable for its highly disordered and dynamic nature (1). Both proteins are recognized as polyelectrolytes, being highly positively and negatively charged, respectively. Recent findings have revealed an intriguing facet of this interaction – its capacity to drive biomolecular phase separation under conditions of high protein concentration and low salt levels, as illustrated in the phase diagram to the right (2). Techniques such as smFRET and nsFCS revealed that the dense phase, which is 1000 times more concentrated than the dilute phase, consists of a network of interactions with a bulk viscosity 300 times that of water. Moreover, in the dense phase, these disordered proteins remain highly dynamic, with rapidly changing configurations (2). We propose that the interaction of H1:ProTa serves as a suitable model system for advancing our understanding of how intrinsic disorder and polyelectrolyte properties contribute to the formation of condensates. This study represents a preliminary step towards utilizing the power of NMR spectroscopy to explore deeper into this phenomenon.

POSTER NO 15: Giulio Tesei

Conformational ensembles of the human intrinsically disordered proteome

Giulio Tesei, Anna Ida Trolle, Nicolas Jonsson, Johannes Betz, Frederik E. Knudsen, Francesco Pesce, Kristoffer E. Johansson, and Kresten Lindorff-Larsen

Structural Biology and NMR Laboratory & the Linderstrøm-Lang Centre for Protein Science, Department of Biology, University of Copenhagen, Copenhagen, Denmark

Contact: giulio.tesei@bio.ku.dk

Intrinsically disordered regions (IDRs) constitute about one third of the human proteome and play important roles in biological processes. While lacking well-defined 3D structures, IDRs adopt highly diverse conformational ensembles which are determined by the amino acid sequence, yet not captured by existing structure prediction methods. In this study, we investigated sequence-ensemble-function relationships of IDRs by constructing conformational ensembles of 28,058 human IDRs using CALVADOS, an efficient molecular model. Through bioinformatics analyses, we explored the relationship between the compaction of the IDRs and the biological function and cellular localization of the proteins. Further, we gained insights into how sequence features relate to chain compaction and analyzed the evolutionary conservation of conformational properties across homologous IDRs. Our work exemplifies how our database of conformational ensembles, which we made freely available, can be used to formulate hypotheses on the biological role and evolution of human IDRs.

POSTER NO 16: Hossein Mohammad-Beigi

Ubiquitination of α-synuclein by Nedd4 ligases studied by flow-induced dispersion analysis

Hossein Mohammad-Beigi, Julian Koch, Rune Busk Damgaard, Alexander K. Buell

Protein Biophysics, Department of Biotechnology and Biomedicine, Technical University of Denmark, Lyngby, Denmark

Contact: hosmbe@dtu.dk

Poster abstract is not public.

POSTER NO 17: Ida K.S. Meitil

Classification of bacterial surface polysaccharide polymerases

Ida K.S. Meitil, Garry P. Gippert, Kristian Barrett, Cameron J. Hunt, Bernard Henrissat

DTU Bioengineering

Contact: idamei@dtu.dk

Poster abstract is not public.

POSTER NO 18: Ida Sjøgaard

Regulation of the stress responsive transcription factor DREB2A through formation of a ternary complex between the repressor RCD1 and the MED25 subunit?

Ida M. Sjøgaard, Frederik F. Theisen, Bjørn W. Nordsteen, Andreas Prestel, Steffie Elkjær, Birthe B. Kragelund, & Karen Skriver

Department of Biology, UCPH

Contact: ida.zobbe@bio.ku.dk

The Arabidopsis thaliana transcription factor DREB2A interacts with subunit 25 of the transcriptional regulator complex, Mediator, and positive and negative co-regulators to guide transcriptional activity of target genes in response to abiotic stress signals. DREB2a binds to the ACID-domain

of MED25 through two separate short linear motifs (SLiMs) engaging a novel ACID-binding motif (ABS) and the known RCD1-binding motif (RIM).

Here, we examine the bivalent ABS-RIM binding region as a molecular switch to integrate signals from coregulators through formation of ternary complexes with the MED25-ACID domain.

POSTER NO 19: Ina Afra Surkamp

A single molecule approach reveals how enzyme activity is correlated to diffusion

Ina Surkamp, Min Zhang, Nikos Hatzakis

Copenhagen University, University Hamburg

Contact: ina.surkamp@studium.uni-hamburg.de

This study aims to establish a correlation between the hydrolysis activity of lipase and its diffusion behavior at a lipid-aqueous interface. Lipase are a type of enzyme that takes part in digestion and also has an important role in the industry. In this sudy, Lipase of the fungus thermomyces lanuginosus (TLL) is examined. By single particle tracking the lipase are anayzed individually. Compared to conventional methods, that only give averaged results, we can see the heterogenious activity which are otherwise masked. We usee bulk activity measurements as a supplyment to support our results. The recorded tracks are treated with a three state hidden markov analysis to find the different diffusion states of lipase. First, the wild type TLL and an inactive mutatnt are compared, with a very clear distinction [2]. The inactive variant shows shorter step length and a more confined diffusion. Subsequently, two situations 1) incorporate product in the preparation of the surface and 2) observing the lipase over time, when product should naturally occur, are compared. Here, the diffusion and the activity cannot be aligned so perfectly. We can observe a maximum in the diffusion coefficient where there is no such behavior in the activity. Within limits, they can be reconciled through the step length. In this experiment, many outliers can be seen. Hypothetically, the exceptions could be attributed to a different structure that Triglyceride can form.

POSTER NO 20: Jacob Aunstrup Larsen

Amyloid Φ-value analysis, illuminating the transition state

Jacob Aunstrup Larsen (1),

Abigail Barclay (2), (3),

Nicola Vettore (1),

Kresten Lindorf-Larsen (3),

Alexander Kai Buell (1)

1: Section for Protein Chemistry & Enzyme Technology, DTU Bioengineering, Technical University of Denmark, 2800 Kgs. Lyngby, Denmark

2: Niels Bohr Institute, University of Copenhagen, 2100 Copenhagen, Denmark

3: Structural Biology and NMR Laboratory, University of Copenhagen, 2100 Copenhagen, Denmark

Contact: j-a-l@live.dk

Background: It was previously expected, that the problem of amyloid related diseases would be solved by the introduction of high-resolution Cryo-EM structures of fibrils. The first high-resolution structures of pathological amyloids were published in the mid-2010s, and structural information of amyloids is becoming increasingly abundant. Unfortunately, this progress has not translated into significant development of therapeutic agents. It is clear that understanding the microscopic steps of the aggregation reaction will play an important role in understanding the amyloid phenomenon and further identify develop therapeutic targets and strategies [1].

Aims: We sought to develop an experimental framework able to discern the change in the energy-landscape (energy barrier of elongation and thermodynamic stability) of the amyloid elongation reaction upon mutagenesis. We expected such information would make it possible to estimate the fraction of formed contacts of a particular residue within the transition state, similar to classical Φ-value analysis [2]. We expected that molecular dynamics simulations, restrained by the experimental data, could provide the first estimate of the transition state ensemble of the amyloid elongation reaction.

Methods: Studying the PI3K-SH3 amyloid system, we have measured changes in the energy-landscape of polymerization upon hydrophobic deletions of fifteen residues. We used depolymerization by chemical denaturation and Quartz Crystal Microbalance to probe the free energy of polymerization and energy barrier of elongation (calculated from relative elongation rates of mutants) and calculated distinct Φ-values. We have simulated the transition state MD-simulations, restrained by the produced Φ-values, to determine the transition state ensemble.

Results: Our work maps the energy landscape of the elongation reaction of PI3K-SH3 at an unprecedented level. Our work represents the largest single dataset of the effect of point mutations on the thermodynamic stability of an amyloid fold. We are able to quantifiably illuminate the structural features of the transition state. By restrained MD-simulations, we determine the transition state ensemble of the amyloid elongation reaction. Combined, we have identified key sequence-segments and residues stabilizing the transition from free monomer to the aggregated amyloid state.

Conclusion: In conclusion, the developed protocol provides a strong foundation for studying the amyloid elongation reaction with residue-level resolution. Providing a stepping-stone for thorough biophysical study of the aggregation mechanisms of amyloids.

References:

[1] Chiti, F. & Dobson, C. M. Protein Misfolding, Amyloid Formation, and Human Disease: A Summary of Progress Over the Last Decade. Annu Rev Biochem 86, 27–68 (2017).

[2] Fersht, A. R. & Satoshi, S. Phi-value analysis and the nature of protein-folding transition states. Proc Natl Acad Sci U S A 101, 7976–81 (2004).

POSTER NO 21: Jing Zhao

Discovery of de novo molecular glues

Jing Zhao,  Joseph M. Rogers

Department of drug design and pharmacology, University of Copenhagen

Contact: ntz138@sund.ku.dk

Protein-protein interactions (PPIs) play pivotal roles in life processes. The network of PPIs in the human cell is vast. Aberrant PPIs lead to various diseases, including cancer. Therefore, targeting PPIs is an essential strategy for the development of new drugs. The number of PPI inhibitors that are either approved or under late-stage clinical investigation has expanded over the years. However, strengthening, or even creating interactions between proteins is another fascinating mechanism.

Molecular glues are proximity-inducing molecules that can promote the dimerization of two proteins, leading to diverse biological functions such as protein folding, localization, and degradation. However, rational discovery of molecular glues is difficult due to the lack of understanding of the PPIs they stabilize. Certain natural products such as cyclosporin and rapamycin can act as molecular glues. Cyclosporin is a macrocyclic peptide. Inspired by the chemical backbone of cyclosporine, the discovery of cyclic peptide molecular glue is a new avenue of drug modality.

The RaPID system is a systematic screening approach that has been used to discover binding cyclic peptides de novo. RaPID can quickly and efficiently screen trillions of DNA-tagged cyclic peptides and identify the highest affinity molecule. Here, we developed and applied the RaPID system to find de novo molecular glues, and after 10 rounds of selection, we obtained a ranked list of cyclic peptides with the highest binding affinity.

POSTER NO 22: Kira Devantier

Viral Small Hydrophobic Protein form Pentameric Pores in the Membrane

Kira Devantier1,2, Trine L. Toft-Bertelsen3, Andreas Prestel2, Cagla Sahin4,5, Katrine Bugge2,

Mette M. Rosenkilde1 & Birthe B. Kragelund2

1: Molecular Pharmacology Laboratory (MolPharm), Department of Biomedical Science, University of Copenhagen, Copenhagen, Denmark. 2: Structural Biology and NMR laboratory

(SBiNLab), The Linderstrøm-Lang Centre for Protein Science and Integrative Structural Biology at University of Copenhagen (ISBUC), Department of Biology, University of Copenhagen,

Copenhagen, Denmark.3: Department of Neuroscience, Faculty of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark. 4: Department of Microbiology,

Tumor and Cell Biology, Karolinska Institutet, Stockholm, Sweden. 5: Department of Biology, University of Copenhagen, Denmark.

Contact: kira@sund.ku.dk

The small hydrophobic protein (SH) from Mumps virus is an integral membrane protein, involved in immune evasion during viral infections. Studies have shown that SH is able to transport Cl- ions across a membrane, suggesting the protein could belong to the family of viroporins. Given that SH is a virulence factor, it has been suggested as a possible drug target. Here we ask: What is the structure and oligomeric state of SH? To be used as a basis for structure based drug-design.

Through multiple biophysical techniques we show that SH has the ability to homo-oligomerize with a preferred pentameric state. Using a divide-and-conquer approach we show that the transmembrane domain of SH consists of single pass helix, while the extramebrane region lacks the characteristics of defined secondary structure, and which self assembles in reconstituted environments. By truncating SH at the C-terminus, we demonstrate the key role of the extramembrane C-terminal region in the formation of higher state oligomerisation. The combination of these experimental data of wildtype SH and non-functionel variants, contribute to a proposed model of the SH pentameric structure.

POSTER NO 23: Kristine Degn

Framework to Study Missense Mutations in Proteins

Kristine Degn (1),  Ludovica Beltrame (1), Matteo Arnaudi (1,2), Mattia Utichi (1,2), Simone Scrima (2), Pablo Sánchez-Izquierdo (1,2), Francesca Maselli (2,3), Karolina Krzesińska (1), Terézia Dorčaková (1), Jordan Safer (4), Katrine Meldgård (1), Philip

(1) Cancer Systems Biology, Section for Bioinformatics, Department of Health and Technology, Technical University of Denmark, 2800, Lyngby, Denmark

(2) Cancer Structural Biology, Danish Cancer Institute, 2100, Copenhagen, Denmark

(3) Institute of Molecular Bioimaging and Physiology, National Research Council (IBFM-CNR), Via F. Cervi 93, Segrate-Milan, 20054, Milan, Italy

(4) Bioinformatic and Computational Biology group, the Center for Development of Therapeutics, Broad Institute of MIT and Harvard, 415 Main St, Cambridge, MA 02142, USA

(5) Department of Science, Technology and Society, Scuola Universitaria IUSS, Istituto Universitario di Studio Superiori, Piazza della Vittoria 15, 27100 Pavia, Italy

(6) Department of Clinical Genetics and Genomics, Sahlgrenska university hospital, Sweden

(7) Department of Laboratory Medicine, Institute for Biomedicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden

(8) Institute of Clinical Medicine, Dept. Oncology and Pathology, Lund University, Sweden

(9) Department of Clinical Research, Copenhagen University Hospital Amager and Hvidovre, Copenhagen, Denmark

Contact: krde@dtu.dk

The understanding of genomic variants in diseases such as cancer is continually deepening due to the adoption of advanced sequencing methods within clinical settings. Detection of genomic variations has resulted in numerous variants being classified as Variants of Uncertain Significance (VUS) or being associated with conflicting evidence, thereby presenting notable obstacles in the interpretation and utilization of their detection. Here, we introduce a framework, MAVISp (Multi-layered Assessment of VarIants by Structure for proteins), and one of its supporting tools, PDBminer, to study these variants in protein structures. MAVISp is a modular structure-based framework for variant interpretation collectively aiming to elucidate the intricate reasoning a particular variant may be benign or pathogenic. Accordingly, we continually develop the framework, analyze proteins, and have over 170 proteins and approximately 70000 variants available on our web server as of November 2023. The collective aim of the framework is to advance the interpretation of genomic variants for clinical application and disease research. In addition, we also present PDBminer, a supporting tool in the MAVISp framework aiming at making structure identification and selection easier, faster, and less error-prone. The computational methods used in the MAVISp framework rely on the structure selected for investigation. The sources of these structures are typically experimental structures available at the Protein Data Bank (PDB) and high-quality in-silico model structures, such as AlphaFold-generated models. PDBminer curates a ranking of available structures that can be queried and visualized.

POSTER NO 24: Lovisa Majtorp

Exploring the Utilization Mechanism of  β-Mannooligosaccharides in Roseburia hominis A2-183

Lovisa Majtorp, Abhishek Bhattacharya, Mathias Wiemann, Simon Birgersson, Henrik Stålbrand

Lund University

Contact: lovisa.majtorp@biotechnology.lu.se

Poster abstract is not public.

POSTER NO 25: Macarena Gomez de Salazar

Reverse engineering of the synaptic tagging and capture mechanisms

Macarena Gomez de Salazar, Jerusheny Henry, Victoria Twiddy, Magnus Kjaergaard

Department of Molecular Biology and Genetics, PROMEMO and DANDRITE , Aarhus university

Contact: macarena.gsh@mbg.au.dk

Memories rely on connections between neurons that to be strengthened require the delivery of plasticity-related proteins (PRP)  from a range of different signaling pathways. PRPs can be locally produced and recruited only to activated synapses. This process is called “synaptic tagging”. In this project we aim to identify the biophysical characteristics that PRPs should have. To address this,  we engineered artificial peptides, Synthetic PRPs (SynPRPs), to test mechanisms of activity-dependent targeting. We demonstrated in vitro that these SynPRPs are phosphorylated by CaMKII and bind postsynaptic density protein 95 (PSD95). In mouse hippocampal cultured neurons, we demonstrated that synaptic clusters of SynPRPs + PSD95 increase their synaptic localization after chemical long term potentiation (cLTP). Moreover, we observed neuronal activity changes when we record with a multielectrode array (MEA) that suggest that SynPRPs may interfere with synaptic plasticity. Overall, our results help to further understand “synaptic tagging”.

POSTER NO 26: Maciej Gielnik

De novo design of intrinsically disordered desiccation chaperones

Maciej Gielnik, Magnus Kjærgaard

Department of Molecular Biology and Genetics, Aarhus University, Aarhus, Denmark

Contact: maciejgielnik@mbg.au.dk

De novo protein design is an emerging branch of protein engineering which creates functional proteins without the specified sequence or structure. Protein structures catalyze every biochemical reaction, however the function of desiccation chaperones has been ascribed exclusively to intrinsically disordered proteins (IDPs). Here we test a new approach for de novo design of functional IDP chaperone to preserve activity of a model enzyme, citrate synthase (CS), during lyophilization. We designed and expressed a panel of synthetic IDPs with different physico-chemical properties and tested them as chaperones. CS lyophilized alone loose enzymatic activity, however addition of IDP chaperone AavLEA1 preserves natural activity of CS. Synthetic IDPs containing basic amino acids show no- to little chaperone activity, while acidic, polyampholyte and amphiphilic sequences preserve more than 50% of CS activity. The results suggest desiccation chaperones might have evolved from random polypeptide chains and it is possible to de novo design functional IDPs. In the next step we will combine physico-chemical properties of the best performing IDPs, test their chaperone activity and iteratively repeat the process until we do not see any improvement in chaperone function. De novo designed IDP chaperones can be used to increase shelf life of biopharmaceuticals and make them available in developing countries without the need for cold chain logistics.

POSTER NO 27: Mathias Jensen

Structural diversity and signatures of Akkermansia muciniphila β-N-acetylhexosaminidases

Mathias Jensen1, Bashar Shouker1,2, Jens Preben Morth1, Maher Abou Hachem1

1- Section of Protein Chemistry & Enzyme Technology, Department of Biotechnology & Biomedicine, DTU, 2800 Kongens Lyngby, Denmark.

2- Division of Biotechnology, Department of Chemistry, Lund University, 22100 Lund, Sweden.

Contact: matjen@dtu.dk

Akkermansia muciniphila is a prevalent member of the human gut microbiota (HGM)1. The abundance of A. muciniphila has been correlated positively with various health benefits, including lean body mass, lower inflammation and lower insulin resistance2. This bacterium colonizes the outer mucus layer, which lines host enterocytes in the lower gut, and grows on mucin, the main structural scaffold of the gut mucosa as a sole carbon and nitrogen source. Details of the mucolytic apparatus of A. muciniphila remain underexplored. Recently, the A. muciniphila decapping machinery, which confers removal of fucosyl- and sialyl unit that decorate the terminal epitopes in mucin has been described showing that this decapping is crucial for to initiate mucin deconstruction3.

We analyzed the Carbohydrate Active Enzyme (CAZy) repertoire encoded by A. muciniphila, which showed that glycoside hydrolase family 20 (GH20) is the most represented family in A. muciniphila. Thus, 11 GH20 sequences populating 5 phylogenetic clusters that do not harbour any biochemically described enzymes. Here, we report a previously not described A. mucinphila GH20 featuring a modular architecture with a putative carbohydrate binding module (CBM) from a previously unknown family. Crystallographic evidence revealed the binding of a β-N-acetylgalactosamine in this putative CBM. Comparison of GH characterized orthologues from A. muciniphila revealed common structural features, and more interestingly key differences in loops around the active site and appended domain, which may confer potential substrate binding. These findings highlight the structural signatures that underpins the diversity within GH20, as a potential adaptation to target diverse human β-N-acetylhexosamine-rich O-glycans.

1.                        Derrien, M., Collado, M. C., Ben-Amor, K., Salminen, S. & De Vos, W. M. The mucin degrader Akkermansia muciniphila is an abundant resident of the human intestinal tract. Appl. Environ. Microbiol. 74, 1646–1648 (2008).

2.                        Cani, P. D., Depommier, C., Derrien, M., Everard, A. & de Vos, W. M. Akkermansia muciniphila: paradigm for next-generation beneficial microorganisms. Nat. Rev. Gastroenterol. Hepatol. 19, 625–637 (2022).

3.                        Shuoker, B. et al. Sialidases and fucosidases of Akkermansia muciniphila are crucial for growth on mucin and nutrient sharing with mucus-associated gut bacteria. Nat. Commun. 14, 1833 (2023).

Funding: Independent Research Fund Denmark: Technology and Production Research Project proposal, Ministry of Higher Education and Scientific Research of Iraq.

POSTER NO 28: Matteo Lambrughi

LipidDyn: a tool to investigate cellular membranes and proteins

Matteo Lambrughi1, Simone Scrima1,2, Sergio Esteban Echeverría1, Laura Kappel Christensen1, Agnes Syshøj Lorenzen1, Anne Marie Nørrelykke Rossen1, Elisabeth Corcelle-Termeau3, Mads Møller Foged3, Knut Kristoffer Bundgaard Clemmensen3, Marja Jäättelä3, Ken

1Cancer Structural Biology, Center for Autophagy, Recycling and Disease, Danish Cancer Society Research Center, 2100, Copenhagen, Denmark

2Cancer Systems Biology, Section for Bioinformatics, Department of Health and Technology, Technical University of Denmark, 2800, Lyngby, Denmark

3Cell Death and Metabolism, Center for Autophagy, Recycling and Disease, Danish Cancer Society Research Center, 2100, Copenhagen, Denmark

Contact: matteolambrughi@gmail.com

Abstract

Cellular processes rely on the concerted interplay of proteins with membranes. Cellular membranes are formed from different lipids depending on the subcellular localization. The lipid compositions are sensitive to the cellular environment, and their alterations are linked to cancer. However, we have sparse knowledge about the lipid compositions of subcellular membranes and their functional roles. Increasing the knowledge of these mechanisms is paramount to understanding their contributions to cellular processes and cancer.

Molecular Dynamics (MD) simulations are a powerful tool to complement experimental approaches, such as mass spectrometry-based lipidomics of cellular compartments. Lipidomics data provide a massive amount of information that can be utilized to design biologically relevant membrane models for simulations. Simulations allow to investigate complex membrane systems and study their dynamics at timescales that are difficult to reach by experimental approaches. Open challenges in the field are combining lipidomics data of membranes and effectively rationalizing the collected simulation data. We developed LipidDyn, a pipeline to integrate lipidomics data and simulations better and monitor membrane and protein biophysical properties from MD simulations [1].

Here, we present the critical improvements in the updated version of LipidDyn pipeline, which introduces automated integration of organelle lipidomics data into the design of membrane models for MD simulations via the Lipid2MD tool, bridging a critical gap in biological membrane modeling. Furthermore, we broadened the range of supported analysis of structural and biophysical properties in the pipeline, including modules for calculating membrane curvatures, individual lipid contributions, protein-lipid interactions, and lipid transbilayer motions.  enhancing the fidelity and applicability of MD simulations in lipid research. With these advancements, LipidDyn equips a versatile and powerful toolkit to enhance the design of MD simulations and support comprehensive investigations. LipidDyn is available, free of charge, under the GNU General Public License from https://github.com/ELELAB/LipidDyn.

Funding

E.P. group is supported by Danmarks Frie Forskningsfond, Natural Science, Project 1 (102517), NovoNordisk Fonden Bioscience and Basic Biomedicine (NNF20OC0065262) and is part of the Center of Excellence for Autophagy, Recycling, and Disease (CARD), which is supported by Danmarks Grundforskningsfond (DNRF125).

Reference

1 Scrima S, Tiberti M, Campo A, Corcelle-Termeau E, Judith D, Foged MM, Bundgaard Clemmensen KK, Tooze S, Jäättelä M, Maeda K, Lambrughi M. and Papaleo E. Unraveling membrane properties at the organelle-level with LipidDyn, Comput Struct Biotechnol J. 2022 30;20:3604-3614

POSTER NO 29: Michael Pichler

Sialidases from Akkermansia muciniphila, including a member of a novel family, mediate the removal of all sialic acid mucin caps and their sharing with the mucus associated community

Michael Pichler1, Bashar Shuoker1,2,, Chunsheng Jin3, Hiroka Sakanaka1, Haiyang Wu4, Ana Martínez Gascueña4, Jining Liu5, Tine Sofie Nielsen1, Jan Holgersson5, Eva Nordberg Karlsson2, Nathalie Juge4, Sebastian Meier6, Jens Preben Morth1,8, Niclas G. Karls

1Department of Biotechnology and Biomedicine, Technical University of Denmark, Lyngby, 2800, Denmark

2Biotechnology, Department of Chemistry, Lund University, Lund, Sweden. 3Proteomics Core Facility at Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden. 4Quadram Institute Bioscience, Norwich, United Kingdom. 5Department of Laboratory Medicine, Institute of Biomedicine, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden. 6Department of Chemistry, Technical University of Denmark, Kgs Lyngby, Denmark.

Contact: mijap@dtu.dk

Mucin O-glycans display a wide array of structural complexity and diversity, primarily distinguished by their fucosylated and sialylated terminal epitopes. The terminal capping confers resistance to microbial attacks and provides nutrients and adhesion site for mucus-adapted bacteria, thereby playing a key role in host-microbiome symbiosis and pathogenesis.

Akkermansia muciniphila is a prevalent mucolytic specialist strongly associated with host metabolic health. Mucin O-glycans are heavily sialylated, however the specificities of A. muciniphila sialidases acting on mucin sialo-capped O-glycans have remained unexplored on mucin O-glycans.                

The present study explored the preferences of A. muciniphila sialidases for mucin-conjugated and free O-glycans. Employing this powerful and physiologically more relevant substrate with 160 assigned unique O-glycan structures, we uncovered that the decapping repertoire of A. muciniphila exhibit a broad range of enzyme selectivities. Strikingly, while some enzymes are mono-specific towards a single glycan motif, others are highly promiscuous, which is modulated by subtle changes in active size loops. Collectively, this sialidase arsenal confer the decapping of all mucin sialyl epitopes. Similar to other GH33 sialidases, the A. muciniphila sialidases from this family displayed broad but variable preferences for sialo-motifs. Interestingly we report the discovery and characterization of a novel inverting sialidase family, which is strictly targeting the sialyl-T antigen amongst all other sialo-epitopes. This enzyme exhibits unique structural features consistent with its strict specificity and is the founding member of GH181. Our findings bring novel insights into the initiation of mucin O-glycan degradation by A. muciniphila sialidases.

POSTER NO 30: Milena Lalic

Phosphorylation of the AB domain in PPAR⍺ is enhanced by the DNA binding domain

Milena Lalic, Elisabeth Thomsen, Lukas Bauer, Birthe Brandt Kragelund

Contact: milena.lalic@bio.ku.dk

Poster abstract is not public.

POSTER NO 31: Min Zhang

Direct Observation of the Formation of Amyloid Spherulites in Real-time by Super-resolution Microscopy

Min Zhang, Henrik D. Pinholt, Xin Zhou, Søren S.R. Bohr, Vito Foderà, Nikos Hatzakis

University of Copenhagen

Contact: min.zhang@chem.ku.dk

The misfolding of native structural proteins into amyloid fibrils is associated with a spectrum of neurodegenerative diseases such as Alzheimer’s and Parkinson’s. Recent studies revealed fibrils not to be the only protein aggregate species, but rather the existence of amyloid-like spherical aggregates, named spherulites, ranging from few micrometers to millimeters in diameter. Conventional spectrometric methods reporting the average growth rates are mostly used to investigate the aggregation mechanism, while microscopy only readouts of final structures, consequently masking the morphological and growth heterogeneity of the aggregates. To solve this, we recently developed a new super resolution method (REPLOM) based on single molecule localization microscopy, and successfully observe directly the formation of insulin spherulites and quantify the existence and abundance of diverse morphologies as well as their heterogeneous growth kinetics of each structure. Our results revealed insulin spherulite growth is not exclusively isotropic, but it may also occur anisotropically. Combined with Machine learning, we associated growth rates to specific morphological transitions and provided energy barriers and the energy landscape for each aggregation morphology. Our method can be extended to more medically- relevant proteins, such as α-synuclein or Aβ peptide. In those cases, our approach may provide unprecedented hitherto information on transient intermediate species, which are nowadays recognized as the cause of the disease progression, both in terms of energetics and morphology.

POSTER NO 32: Nicolas Jonsson

Decoding the molecular mechanisms of loss-of-function variants in the human proteome

Nicolas Jonsson, Matteo Cagiada, and Kresten Lindorff-Larsen

University of Copenhagen

Contact: nicolas.jonsson@bio.ku.dk

Proteins play a critical role in cellular function by interacting with other biomolecules. Identifying functional sites and their mechanism is crucial to understanding how the protein works, how its function can be affected, and how disease may occur. In previous work, we developed a computational tool called the Functional Site Model (FuSiL) to identify functional residues and understand their role in the molecular mechanisms of the protein. In this work, we introduce a new version of the model, called re:FuSiL, which uses deep learning representations to improve computational performance while keeping the accuracy of the original implementation. We ran re:FuSiL predictions on all human proteins and extensively analysed the annotated proteome data, providing insight into how mutations affect the molecular mechanisms behind loss-of-function in the human proteome, and quantifying the number of residues involved in structural and functional roles. Finally, by analysing over fifty thousand annotated Mendelian clinical variants, we show how most pathogenic variants in the dataset have loss-of function (LOF) as the underlying mechanism behind the emergence of disease.

POSTER NO 33: Nicolas Sebastian Gonzalez Foutel

Effect of protein condensates on kinase signaling

Nicolas S. Gonzalez-Foutel (1),  Magnus Kjaergaard (1-3)

1. Department of Molecular Biology and Genetics, Aarhus University

2. The Danish Research Institute for Translational Neuroscience (DANDRITE), Nordic EMBL Partnership for Molecular Medicine

3. Center for Proteins in Memory - PROMEMO, Danish National Research Foundations

Contact: nfoutel@mbg.au.dk

Macromolecular condensates are formed by spontaneous and reversible self-interactions of proteins and nucleic acids i.e. scaffolds, followed by their phase separation inside the cell. These condensates can recruit enzymes and substrates, known as clients, boosting or quenching biochemical reactions and therefore acting as switches in cellular signaling pathways. Phosphorylation by protein kinases has been shown to occur inside these condensates, but little is known about their functional consequences and how this could affect the kinase signaling network. We hypothesize that condensates change phosphorylation rates and broaden substrate usage, and that a slow diffusion inside the condensates is the dominant effect modifying these reactions. To study this, we target both Protein kinase A and substrates of different quality (kcat/KM) and length, to synthetic intrinsically disordered protein condensates formed by repetitive octapeptides. We then measured phosphorylation kinetics by 32P radioactivity assays. Preliminary results show that condensates reduce phosphorylation rates irrespective of any substrate. This decrease is more pronounced for lengthier kinase substates regardless of their quality, which are recruited and kinetically trapped in a slow diffusion regime. However, given the same substrate quality, the reduction in phosphorylation rates is less noticeable for shorter substrates, where their diffusion rates across the condensates are faster. This work provides better understanding on how condensates affect kinase reactions and might help to elucidate generalizable effects of condensates.

POSTER NO 34: Nina Louise Jacobsen

Targeting disorder in the human Interleukin-4 receptor

Nina L. Jacobsen, Joseph M. Rogers, Christina Mølck, Flemming H. Larsen, Birthe B. Kragelund

REPIN and Structural Biology and NMR Laboratory, Department of Biology, University of Copenhagen, Denmark (NLJ, BBK)

Department of Drug Design and Pharmacology, University of Copenhagen, Denmark (JMR)

LEO Pharma A/S, Ballerup, Denmark (NLJ, CM, FHL)

Contact: nina.jacobsen@bio.ku.dk

Poster abstract is not public.

POSTER NO 35: Oana Antonescu

Structural insights into short linear motifs in autophagy: phospho-regulation and flanking regions

Oana Antonescu (1), Matteo Arnaudi (1,2), Laura Mattioli (1), Burcu Aykac Fas (1), Valentina Sora (1), Mukesh Kumar (1), Matteo Tiberti (1), Flavie Strappazzon (3), Matteo Lambrughi (1), Vladimir Rogov (4,5), Emiliano Maiani (1), Elena Papaleo (1,2)

(1) Cancer Structural Biology, Danish Cancer Society Research Center, Copenhagen, Denmark

(2) Cancer Systems Biology, Section for Bioinformatics, Department of Health and Technology, Technical University of Denmark, Lyngby, Denmark

(3) IRCCS Fondazione Santa Lucia, Rome, Italy

(4) Institute of Pharmaceutical Chemistry, Goethe University Frankfurt, Frankfurt, Germany

(5) Structural Genomics Consortium, Buchmann Institute for Life Sciences, Goethe University Frankfurt, Frankfurt, Germany

Contact: ona@cancer.dk

The interaction between intrinsically disordered proteins and their binding partners can be mediated by sequence motifs called Short Linear Motifs (SLiMs). Key features of SLiMs include their sequence degeneracy and fine-tuned regulation by post-translational modifications (PTMs), such as phosphorylation. SLiMs, which bind with low affinity to their partners in their unmodified variants, can become stronger or weaker binders upon PTMs (1). Besides the core SLiM sequence, residues in the flanking regions can also modulate binding affinity and specificity to their partners (2).

In this work, we integrated biomolecular simulations, in silico high-throughput mutational scans, biophysical measurements and cellular assays to unveil the structural details of phospho-regulation and contribution of flanking regions in a class of SLiMs crucial for autophagy, i.e. the LC3 interacting regions (LIRs). We focused our efforts on three essential autophagy adaptors (AMBRA1, optineurin and NDP52/CALCOCO2), proteins that contain LIR motifs and interact with LC3/GABARAP proteins to fulfill their function.

In AMBRA1, we have detected and characterized a new LIR motif, that can partially affect the interaction with LC3B. Furthermore, we showed that AMBRA1 phosphorylation at multiple sites around the previously known LIR motif can regulate the binding affinity to LC3B. In optineurin, we provided new insights into the structural mechanisms and conformational changes promoted by phosphorylation upstream of its LIR motif. In addition, we characterized the contribution of residues in the LIR flanking region to the interaction with LC3B. Finally, we have discovered a novel potential helical flanking region in the C-terminal region of the noncanonical LIR motif of NDP52/CALCOCO2, that contributes to the interaction with its selective binding partner LC3C.

The combined computational and experimental framework developed in these studies can be used to characterize other LIR-containing proteins and build a platform for structural biology of (phospho-modulated) short linear motifs in disordered proteins.

(1) Kliche J, Ivarsson Y. Orchestrating serine/threonine phosphorylation and elucidating downstream effects by short linear motifs. Biochem J. 2022 Jan 14;479(1):1-22. doi: 10.1042/BCJ20200714.

(2) Bugge K, Brakti I, Fernandes CB, Dreier JE, Lundsgaard JE, Olsen JG, Skriver K, Kragelund BB. Interactions by Disorder - A Matter of Context. Front Mol Biosci. 2020 Jun 16;7:110. doi: 10.3389/fmolb.2020.00110.

POSTER NO 36: Per Hägglund

Anastellin impacts on the processing of extracellular matrix fibronectin and stimulates release of cytokines from coronary artery smooth muscle cells

1Department of Biomedical Sciences, University of Copenhagen, Copenhagen, Denmark

2Department of Chemistry, University of Copenhagen, Copenhagen, Denmark

3Interdisciplinary Nanoscience Center (iNANO); Department of Molecular Biology and Genetics, Aarhus

Fibronectin (FN) play a key role in cell adhesion, migration and differentiation. FN assembly in the extracellular matrix (ECM) is modulated by Anastellin, a fragment of the first type III module in FN.

The influence of Anastellin on the assembly of ECM laid down by human coronary artery smooth muscle cells (HCASMC) was examined by immunohistochemistry (IHC). Culture supernatants from HCASMC were subjected to proteomics analysis to evaluate if Anastellin induced changes in the levels of specific secreted proteins.

A decrease in the extent of FN fibrils was observed for cells exposed to Anastellin using IHC with two FN-specific antibodies. Proteomics analysis revealed that the levels of several other ECM-associated proteins are also modulated in cells exposed to Anastellin. Several cytokines, including interleukin 6 were also upregulated in cells exposed to Anastellin.

Based on these observations we conclude that Anastellin impacts on the structure of FN fibrils in ECM from HCASMC and modulates secretion of cytokines and proteins involved in ECM processing.

Contact: pmh@sund.ku.dk

POSTER NO 37: Peter Røgen

Sequence-Similar Protein Domain Pairs with Structural or Topological Dissimilarity

Peter Røgen

DTU Compute

Contact: prog@dtu.dk

Traditional methods for structural alignment of protein structures report a price or length of the motion required to bring one structure onto the other structure. These methods do not check if there is room for this motion, if it causes steric clashes, or more severely, if it changes the topology or threading of the compared protein backbone curves. We apply ProteinAlignmentObstruction[1] to distinguish such cases.

In this work we check 966546 structural alignments of CATH4.2 domain pairs with at least 30% sequence identity for significant structural or topological dissimilarity. Alternate threading is found in 0.35% of the alignments and relatively independent of sequence identity and structural dissimilarity as measured, e.g., by RMSD. Inside a sequence family often one or a few domains take part in most of the aligned pairs with alternate threading. E.g., 513 domains have either essential topological obstructions or untangling motion >140Å to at least half of their sequence-based structural alignments.

[1] P. Røgen, Quantifying steric hindrance and topological obstruction to protein structure superposition, Algorithms Mol. Biol. (2021) 16:1

POSTER NO 38: Robin Dorau

Discovery and Engineering of Fungal Cutinases with Activity on Polyethylene Terephthalate (PET)

William Brinch-Pedersen (1), Malene Billeskov Keller (1), Robin Dorau (2), Bijoya Paul (1,2), Kenneth Jensen (2), Kim Borch (2) and Peter Westh (1)

(1) Technical University of Denmark (DTU)

(2) Novozymes A/S

Contact: rdo@novozymes.com

Polyethylene terephthalate (PET) is a polymer contributing significantly to environmental pollution and resource depletion. Enzymatic recycling of PET could mitigate negative impacts by conserving resources and promoting a sustainable future. PET hydrolases efficiently break down PET under mild conditions into the building blocks terephthalic acid and ethylene glycol, which can in turn be converted into new high-quality materials. To this date, more than 80 PET hydrolases have been reported, from which most are of bacterial origin and belong to the alpha/beta hydrolase family (InterPro: IPR029059), such as IsPETase from Ideonella sakaiensis. In contrast to that, fungal cutinases have not been studied as intensively. Thermomyces (Humicola) isolens cutinase (HiC) and Fusarium solani cutinase (FsC) are the only true fungal PET hydrolases described in scientific literature so far, with HiC being most efficient. Both enzymes belong to the cutinase/acetylxylan esterase family (InterPro: IPR000675) and are also alpha/beta hydrolases, but smaller. To address the question of whether we are missing important parts of natural diversity with potential activity against PET, we explored diversity around HiC and FsC by cloning 24 novel cutinases.

POSTER NO 39: Sanchari Banerjee

Structural studies on fungal Auxiliary Activity (AA7) flavin dehydrogenases

Sanchari Banerjee, Simone Turella, Jens Preben Morth, Maher Abou Hachem

Department of Biotechnology and Biomedicine, Technical University of Denmark, Lyngby, Denmark

Contact: sancban@dtu.dk

Poster abstract is not public.

POSTER NO 40: Steffie Elkjær

Intrinsic Disorder Content: A Key Player in Transcription Factor Function

The REPIN and The Linderstrøm-Lang Centre for Protein Science;

Structural Biology and NMR Laboratory, Department of Biology, University of Copenhagen, Copenhagen DK-2200, Denmark

Intrinsic disorder (ID)-associated interactions allow more complex binding mechanism; hub interactions, competition, allostery, multivalency and condensate formation. So far, many of these mechanisms are not quite understood. To obtain insight into and help understand the underlying biomolecular mechanism, kinetics is crucial.

Contact: steffie.elkjaer@bio.ku.dk

POSTER NO 41: Søren Brander

Identification of a novel class of N-terminal histidine methyl transferases that acts on LPMOs

Tanveer S. Batth (1)

Jonas L. Simonsen (1,2)

Cristina Hernández-Rollán (3)

Søren Brander (4)

Jens Preben Morth (2)

Katja S. Johansen (4)

Morten H. H. Nørholm (3)

Jakob B. Hoof (2)

Jesper V. Olsen (1)

(1) The Novo Nordisk Foundation Center for Protein Research, University of Copenhagen Denmark

(2) Department of Biotechnology and Biomedicine, Technical University of Denmark, Kongens Lyngby, Denmark

(3) The Novo Nordisk Foundation Center for Biosustainability, Technical University of Denmark, Kongens Lyngby, Denmark

(4) Department of Geosciences and Natural Resource Management, University of Copenhagen, Frederiksberg, Denmark

Contact: sbd@ign.ku.dk

In this study, we have identified the elusive gene encoding the N-terminal histidine methyl transferase (NHMT) that is solely responsible for N-terminal histidine methylation of LPMOs in A. nidulans. This was accomplished through a combination of in-depth proteomics and CRISPR/Cas9 facilitated gene knockouts. We identified close to 7000 A. nidulans protein groups from the initial proteomics analysis, generating the largest dataset of filamentous fungal proteins to date. Quantitative proteomics analysis enabled us to narrow down the list of potential NHMT candidates to 22 genes that were subsequently selected for CRISPR/Cas9 knockouts. We developed a targeted proteomics assay to determine the effect of these knockouts and ultimately identified the putative NHMT candidate

The NHMT sequence encodes a seven-β-strand fold that is shared with the human histidine methyltransferases METTL9 and METTL18 as well as the human N-terminal methyltransferase METTL13. NHMT also contains an integral membrane region with a potentially novel seven-transmembrane with little structural similarities when compared to known membrane proteins the PDB database. The transmembrane region is unique to filamentous fungi and genes with similar sequence similarity are found exclusively as single copies in the genomes of filamentous fungi, suggesting distinct phylogeny of the enzyme.

The discovery of NHMT enabled co-expression with LPMOs in K. phaffii (P. pastoris), leading to the demonstration of recombinant N-terminal histidine protein methylation in a heterologous host that is not a filamentous fungus.

The presented work is also reported in Nat Commun 14, 4202 (2023)

POSTER NO 42: Sören von Bülow

AlphaFold-guided simulations of multi-domain proteins

Sören von Bülow, Giulio Tesei, Fan Cao, Kresten Lindorff-Larsen

Structural Biology and NMR Laboratory, Department of Biology, University of Copenhagen, Ole Maaløes Vej 5, DK-2200 Copenhagen N, Denmark

Contact: soren.bulow@bio.ku.dk

Coarse-grained molecular simulations capture processes on time and length scales inaccessible to atomistic simulations. The CALVADOS hydrophobicity scale model has shown excellent performance in the prediction of the single-chain and phase coexistence properties of intrinsically disordered proteins but does not correctly preserve the structure of folded domains. AF-CALVADOS presents an extension to the CALVADOS model for folded and multi-domain proteins. Here, native interactions within and between domains are restrained in a protein-specific and automated way, guided by AlphaFold 2 predictions on the protein structure and interactions. The implementation enables extremely fast simulations of multi-domain protein.

POSTER NO 43: Thea Klarsø Schulze

Modelling the cellular abundance of protein variants

Thea K. Schulze and Kresten Lindorff-Larsen

University of Copenhagen

Contact: thea.schulze@bio.ku.dk

Multiplexed assays of variant effects (MAVEs) make it possible to probe the functional effects in vivo of all possible single amino acid residue substitutions in a protein in a single experiment. The accumulation of data from such experiments has enabled large-scale analyses of protein variant effects and created the opportunity to train supervised models directly against experimental data to learn sequence-function relationships, including how missense variants affect protein stability and abundance. We have used data obtained by variant abundance by massively parallel sequencing (VAMP-seq), a MAVE technique that quantifies the steady-state cellular abundance of protein variants, to create a model for predicting the impact of missense variation on cellular abundance across proteins. We used data reporting on the effects of ca. 32,000 missense variants on the abundance in six proteins as training data for a deep learning model. The model consists of (i) a self-supervised convolutional neural network on protein structures, (ii) a supervised model for effects on protein stability and abundance, and (iii) a downstream model that describes the experimental process. Our model predicts variant abundance with state-of-the-art accuracy. Moreover, we found a baseline model based only on residue solvent-exposure to also perform well in the prediction task, highlighting how simple structural considerations can be used to explain variant effects observed in large-scale abundance assays.

POSTER NO 44: Thibault Viennet

Mechanistic importance of disordered tails in enzyme regulation by phosphorylation

T. Viennet, D. Dempsey, N. Chu, H. Bae, J. Li, S. Matosin, R. Gregory, P. Cole, H. Arthanari

Dana-Farber Cancer Institute, Cancer Biology, Boston, USA

Harvard Medical School, Biological Chemistry and Molecular Pharmacology

Aarhus University, Department of Chemistry and iNANO

Contact: thibault@chem.au.dk

Intrinsically disordered regions of proteins make up for ca. 33-50% of the human proteome and are generally associated with disease due to their high prevalence in signaling, regulation and control functions. Many cell cycle enzymes contain so-called tails, 30-70 residues terminal disordered regions that are often overlooked by structural biologists because their high flexibility impedes their study by X-ray crystallography or cryo-electron microscopy. However, they have important mechanistic roles regulated by post-translational modifications such as phosphorylation. We have invested significant effort in investigating structural and mechanistic roles of these tails using a combination of protein semi-synthesis, structural biology, enzymology, and cell biology. We investigate regulation of a range of enzymes including PTEN, METTL1, AKT and SHP2. Activation of the kinase AKT seems to depend on phospho-induced remodeling of the interactions between the regulatory PH domain and the kinase domain. PTEN auto-inhibition depends on direct binding of the phosphorylated tail near the active site of the phosphatase. And METTL1 is inactivated by phosphorylation of a residue of the tail that is inherently part of the methyl transferase catalytic pocket. SHP2 regulation relies on phospho-induced alteration of interactions between SH2 and phosphatase domains.

POSTER NO 45: Tobias Tandrup

The molecular basis for alginate oligosaccharide cleavage by Bacteroides ovatus Polysaccharide Lyase family 38

Tobias Tandrup 1*, José Pablo Rivas Fernandez 2,3, Mikkel Madsen 1, Mette E. Rønne 1,4, Agnes B. Petersen 4, Leesa J. Klau 4, Anne Tøndervik 5, Casper Wilkens 6, Finn L. Aachmann 4, Carme Rovira 2,3, Birte Svensson 1

1 Enzyme and Protein Chemistry, Department of Biotechnology and Biomedicine, Technical University of Denmark, DK-2800 Kongens Lyngby, Denmark.

2 Departament de Química Inorgánica i Orgánica (Secció de Química Orgánica) and Institut de Química Teorica i Computacional (IQTCUB), Universitat de Barcelona, ES-08028 Barcelona, Spain 3 Institució Catalana de Recerca i Estudis Avancats (ICREA), ES-08010 Barcelona, Spain.

4 Norwegian Biopolymer Laboratory (NOBIPOL), Department of Biotechnology and Food Science, NTNU Norwegian University of Science and Technology, NO-7491 Trondheim, Norway.

5 Department of Biotechnology and Nanomedicine, SINTEF AS, NO-7491 Trondheim, Norway.

6 Structural Enzymology, Department of Biotechnology and Biomedicine, Technical University of Denmark, DK-2800 Kongens Lyngby, Denmark.

Contact: tobtan@dtu.dk

Alginate is an anionic linear polysaccharide and a major component of the cell wall of brown algae, that holds an important role in the marine biosphere. Alginate and alginate oligosaccharides (AOS) have several applications as viscosifier, texturizer, and for encapsulation in food, pharma, and biomedicine industrial sectors. Alginate is organized in blocks of β-D-mannuronic acid (M) and α-L-guluronic acid (G) or alternating MG, giving the polysaccharide characteristic properties, which vary with the structure, molecular size, and M/G ratio. Alginate decomposition into AOSs is orchestrated by enzymes called alginate lyases, belonging to several polysaccharide lyase (PL) families. Similar for all alginate active PL families, catalysis is employed by a β-elimination mechanism of cleavage between uronic acid residues with formation of an unsaturated uronic acid at the non-reducing chain end. The structural folds associated with alginate lyases so far are the parallel β-helix, the β-jelly roll, the (α/α)6-toroid and the (α/α)6-toroid + antiparallel β-sheet.

Here, we present a PL38 from a novel Bacteroides ovatus strain (BoPL38) of the human gut microbiota, enabling bacterial growth on alginate. BoPL38 is to our knowledge the first PL38 alginate lyase confirmed through biochemical characterization and was found active on all three types of AOS. We present crystallographic analysis of complexed structures obtained with the three AOSs, revealing an (α/α)7-barrel with an active site architecture capable of binding oligo-M, oligo-G, and oligo-MG saccharides through distortion. Using site-directed mutagenesis we uncover the molecular basis of substrate cleavage and through NMR spectroscopy we identify the formation of unsaturated chain ends and the mode-of-action. Finally, we present QM/MM simulations of the reaction coordinate, revealing mechanistic details for both M-M and G-G decomposition.   

This work is supported by Independent Research Fund Denmark | Natural Sciences, Novo Nordisk Foundation, the Norwegian Research Council, Carlsberg Foundation, DanScatt and DTU. Diffraction data were collected at synchrotron radiation facilities MAXIV and ESRF.

POSTER NO 46: Yu Wang

Enzymatic degradation of starch granules by interfacial catalysis

Yu  Wang 1, Yu Tian 2, Stefan Jarl Christensen 3,Yuyue Zhong 2, Georges Feller 4, Xinxun Liu 5, Klaus Herburger 6, Peter Westh 7, Andreas Blennow 2, Marie S. Møller 8, Birte Svensson 1

1: Enzyme and Protein Chemistry, DTU Bioengineering, Technical University of Denmark, Denmark; 2: Department of Plant and Environmental Sciences, University of Copenhagen, Denmark; 3: Enzyme Technology, DTU Bioengineering; 4: Center for Protein Engineering-InBioS, University of Liège, Belgium; 5: College of Food Science and Engineering,  Nanjing University of Finance and Economics,  China; 6: Institute of Biosciences, University of Rostock, Germany; 7: Interfacial Enzymology, DTU Bioengineering; 8: Applied Molecular Enzyme Chemistry, DTU Bioengineering

Contact: yuawa@dtu.dk

Enzymatic modification of starch granules occurs naturally through heterogenous catalysis during biosynthesis and degradation. Thus, in Nature mobilization and utilization of storage starch in seeds and tubers during germination as well as during digestion by enzymes from the gut microbiota rely on intimate binding of enzymes onto starch granules.

We investigated, inspired by cellulase-crystalline cellulose interfacial kinetics , how different amylolytic enzymes degrade waxy maize starch (WMS) granules by using a combination of conventional Michaelis-Menten kinetics having substrate in excess, with inverse Michaelis-Menten kinetics having enzyme in excess, and a Langmuir isotherm binding to determine kinetic parameters (kcat and Km) as well as the densities of attack sites and enzyme binding sites.

POSTER NO 47: Zhiyu Huang

Crystallographic studies of novel and model histidine brace proteins

Zhiyu Huang

University of Copenhagen

Contact: zh@chem.ku.dk

Poster abstract is not public.

POSTER NO 48: Zongxin Guo

Diverse roles of the metal binding domains and transport mechanism of copper transporting P-type ATPases

Zongxin Guoa, Fredrik Oräddb, Viktoria Bågenholma, Christina Grønberga, Jian Feng Mac, Peter Ottd, Yong Wange, Magnus Anderssonb, Per Amstrup Pedersenf, Kaituo Wanga & Pontus Gourdona,g,

a                         Department of Biomedical Sciences, Copenhagen University, Copenhagen, Denmark

b                         Department of Chemistry, Umeå University, Umeå, Sweden

c Institute of Plant Science and Resources, Okayama University, Japan

d                         Medical Department of Hepatology and Gastroenterology, Aarhus University Hospital, Skejby, Denmark

e                         College of Life Sciences, Zhejiang University, Zhejiang, China

f                          Department of Biology, University of Copenhagen, Copenhagen, Denmark

g                         Department of Experimental Medical Science, Lund University, Lund, Sweden

Contact: zongxin@sund.ku.dk

Poster abstract is not public.