the project’s work. “What you would like to be able to do is to look at a sequence like the ß-amyloid peptide associated with Alzheimer’s and say; ‘this transcription factor might stick to this peptide if you introduce it into a cell’,” outlines Professor Schymkowitz. This would give researchers a firmer basis on which to investigate the root cause of the disease. “It would give us a structured way of formulating hypotheses about what goes on in Alzheimer’s disease. We know that the ß-amyloid peptide is toxic, but we don’t understand why,” says Professor Schymkowitz.
MANGO project The hypothesis Professor Schymkowitz explores is that this toxicity is attributable to protein interactions. Now researchers aim to extract the sequence and structural determinants of co-aggregation, with the goal of developing a bioinformatics algorithm. The basis of aggregation activity is thought to be seeding. “When you add a small amount of pre-formed aggregates to a pool of fresh monomer protein, the addition of those seeds accelerates aggregation,” explains Professor Schymkowitz. By extracting aggregates from the brains of patients and injecting them in healthy mice, researchers can trigger the aggregation process. “The question is
Behind the rules of protein aggregation
whether only the protein identical to the one in the aggregate gets pulled into the aggregate? Or is other stuff also contributing, and perhaps modulating aggregation? That would be called cross-seeding.” says Professor Schymkowitz. This aggregation activity is not found along the entire poly-peptide sequence, in fact only small segments within a protein are really prone to aggregation. The composition of a region of between 5-10 amino-acids in a protein plays a major role in determining whether that protein will eventually aggregate. “If another protein has that exact same sequence, you would expect them to co-aggregate,” outlines Professor Schymkowitz. Not many proteins have identical regions, but when crossseeding and co-aggregation events are considered, then more possibilities for aggregation arise. Professor Schymkowitz and his colleagues are using a theoretical framework to investigate this in great depth: “Our theoretical framework is focused on the amino-acid sequence, and so are the computational models that we use.” The experimental work in the project largely centres around studying peptides and proteins both in vitro and in cells to gain deeper insights into protein aggregation. There are two main ways of approaching
this work. “One is to analyse systems that represent known aggregation diseases,” says Professor Schymkowitz. If the aim is to analyse co-aggregation in Alzheimer’s disease for example, then it is possible to work just with the purified Alzheimer’s ß-amyloid peptide, which will undergo aggregation in vitro. “We can then add fragments from other proteins that share a lot of sequence similarity to critical regions of the Alzheimer’s ß-amyloid peptide and simply ask the question – do they mix? Do they aggregate together? Or does ß-amyloid simply aggregate by itself and does it ignore all the stuff that’s around it? We can analyse that in vitro,” outlines Professor Schymkowitz.
Synthetic models A second approach in the project involves creating synthetic models, where researchers essentially steer aggregation and look to induce the aggregation of bacterial proteins, or tumour proteins. This is an area in which significant progress has been achieved over the course of the project, opening up new perspectives on major contemporary problems like anti-microbial resistance, says Professor Schymkowitz. “Drug-resistant bacteria are a major concern at the moment, and we
Protein aggregation is an important factor in the development of many different diseases, now the MANGO project is investigating the underlying basis of the process. This work could both lead to new insights into neurodegenerative conditions and also open up opportunities to develop new therapies against several different diseases, as Professor Joost Schymkowitz explains. A protein is
first synthesised in the cells of our body as a chain of disordered amino-acids, before it is then folded to acquire its shape and structure. This process is not completely efficient, however, and it is thought that around half of all proteins never reach their folded state, but are simply degraded instead, which has knockon effects. “These mis-folded states start accumulating and then stick together. In a process called aggregation,” says Professor Joost Schymkowitz, the Principal Investigator of the MANGO project. There are around 20,000 proteins in the human genome, yet only 30 or so aggregate in a diseasecausing way. Now Professor Schymkowitz and his colleagues in the project aim to probe deeper into the underlying reasons behind this. “Our project is concerned with understanding the rules of engagement of protein aggregation. Which two proteins will stick together and why? What are the rules behind this process?”
10
This process of protein aggregation continues throughout our lifespan, and over time, aggregated proteins are cleared from the body. However, as we age, that clearance process becomes less efficient, and the resulting accumulation of aggregated proteins can cause health problems. “If you look at
specific part of the brain, which is where the symptoms arise,” says Professor Schymkowitz. “Alzheimer’s starts in the hippocampus, while Parkinson’s is associated more with the substantia nigra. Different parts of the brain are affected by the presence of these aggregates, as they spread from there to connected areas.”
Our project is concerned with understanding the rules of engagement of protein aggregation. Which two proteins will stick together and why?
What are the rules behind this process? post-mortem tissue of older people, you often find all sorts of aggregates. It seems like we accumulate aggregates as we age,” explains Professor Schymkowitz. Every type of dementia is essentially an aggregation-associated disease, characterised by the aggregation of different proteins. “They start aggregating in a
By analysing the rules of engagement between proteins, Professor Schymkowitz hopes to gain new insights into how these conditions develop and progress. This could in the long run provide the basis for the development of novel, more effective therapies, a major motivating factor behind
EU Research
Proteins contain aggregation prone regions (indicated in red) that are normally buried inside the core of well folded molecules (the blue and green shapes). When mutations or changes in circumstances lead the proteins to loose their proper organisation, these aggregation prone region my become exposed, causing the molecules to stick together. When two different proteins contain similar aggregation prone regions, they can co-aggregate, causing a cascade.
www.euresearcher.com
11
MANGO
The determinants of cross-seeding of protein aggregation: a Multiple TANGO
Project Objectives
The key question that I aim to address in this proposal is how the beta-interactions of the amino acids in the aggregate spine determine the trade-off between the specificity of aggregation versus cross-seeding. To this end, I will determine the energy difference between homotypic versus heterotypic interactions and how differences in sequence translate into energy gaps. Moreover, I will analyse the sequence variations of aggregation prone stretches in natural proteomes to understand the danger of widespread co-aggregation.
Project Funding
Programme Funding: Horizon 2020 / Sub Programme Area: ERC-2014-CoG / Project Reference: 647458 / From 01.06.2015 to 31.05.2020 / Budget: EUR 1 995 523 / Contract type: ERC-COG.
Contact Details
Project Coordinator, Professor Joost Schymkowitz Switch Laboratory VIB‐KU Leuven Center for Brain & Disease Research Department of Cellular and Molecular Medicine KULeuven T: +32 (0)16 37 25 70 E: joost.schymkowitz@switch.vib-vub.be W: https://www.kuleuven.be/ samenwerking/lind/doc/2013-02-06-lindcv-joost-schymkowitz.pdf196857_en.html Khodaparast L, Khodaparast L, Gallardo R, Louros NN, Michiels E, Ramakrishnan R, et al. Aggregating sequences that occur in many proteins constitute weak spots of bacterial proteostasis. Nature communications. 2018;9:866. Gallardo, R., M. Ramakers, F. De Smet, F. Claes, L. Khodaparast, L. Khodaparast, J. R. Couceiro, T. Langenberg, M. Siemons, S. Nystrom, L. J. Young, R. F. Laine, L. Young, E. Radaelli, I. Benilova, M. Kumar, A. Staes, M. Desager, M. Beerens, P. Vandervoort, A. Luttun, K. Gevaert, G. Bormans, M. Dewerchin, J. Van Eldere, P. Carmeliet, G. Vande Velde, C. Verfaillie, C. F. Kaminski, B. De Strooper, P. Hammarstrom, K. P. Nilsson, L. Serpell, J. Schymkowitz and F. Rousseau (2016). “De novo design of a biologically active amyloid.” Science 354(6313).
Professor Joost Schymkowitz
need to look at new ways of killing them,” he stresses. The idea here is that if aggregation can kill neurons in the brain, then maybe it can be redirected to have a similar impact on drug-resistant bacterial proteins. “This approach has proved successful. You can cause massive aggregation in bacteria, and it has lethal effects,” continues Professor Schymkowitz.
“We’re trying to learn from the underlying processes behind aggregation diseases, so that we can look into applying them to our benefit.” A lot of progress has been made in terms of possible applications, yet there is still more to learn about the underlying nature of protein aggregation, which will remain an important topic of research at the VIB Switch laboratory
Every type of dementia is essentially an aggregation-associated disease, characterised by the aggregation of different proteins. They start aggregating in a specific part of the brain, which is where the symptoms arise This work has sparked a lot of interest, and some of the technology has already been patented and licensed to Aelin Therapeutics, a spin-off company founded in 2017. This is something Professor Schymkowitz plans to investigate further in the future. “We’re still exploring the application-side of things. We’re working on a number of examples in disease contexts,” he says. A lot of success has been achieved in applying this approach to bacterial infections and tumours, and Professor Schymkowitz says it could potentially be applied on a wider basis in future. “This synthetic system is a technology. So if we can induce the aggregation of bacterial proteins, or tumour proteins, this could be a means of causing problems for a pathogen, or a particular disease process,” he explains.
in Leuven, where Professor Schymkowitz is based. Researchers are continuing to gather data on the determinants and sequence of protein aggregation. “We are still very much engaged in understanding the details of what happens structurally. So we’re looking at what happens when these protein segments meet - which mismatches stop the interactions, which are tolerated, and why,” outlines Professor Schymkowitz. The goal is to develop an effective bioinformatics algorithm, which could greatly accelerate medical research. “The computational approach is faster than working experimentally. We want to have a good, precise predictor, so that experimentalists can focus on a narrower range of proteins,” explains Professor Schymkowitz.
The image shows cell expressing a target protein and synthetic aggregates that are designed to enter the cell and cause the target protein to aggregate and loose its function. The background is an transmission electron micrograph of fibrillar aggregates of a synthetic aggregate that target a tumor specific protein.
Professor Joost Schymkowitz is a group leader at the Switch Laboratory of the KU Leuven-VIB Centre for Brain & Disease Research. He gained his PhD in 2001 from the University of Cambridge, working on protein folding at the Centre for Protein Engineering under the supervision of Laura Itzhaki, before moving to the European Molecular Biology Laboratory in Heidelberg.
12
EU Research