We bring together all things related to scientific training
One of the longest serving members of the EMBL Course and Conference team, Nicola is an an event marketer who is always looking for ways to help up-and-coming scientists get ahead in the field...and she has a tendency to break into song.
In 2020 the EMBL Resource Development team and industry partners of the EMBL Corporate Partnership Programme will bring together academic and industrial scientists with interests in chemical biology, chemogenomic libraries, pharmacology, medicinal chemistry and bioinformatics for the EMBL Conference: Expanding the Druggable Proteome with Chemical Biology (5-7 February 2020).
We spoke to co-organiser Gerard Drewes from GSK Cellzome about how chemical biology is helping to expand the druggable proteome.
How would you define the “druggable proteome”?
This is the fraction of our >20,000 human proteins that can be functionally modulated by a drug. Drugs can be small molecules or large molecules such as therapeutic antibodies. Estimates of how many proteins are “tractable” vary widely, I think there may be around 5,000. Only a subset of these 5,000 would be “druggable” which means that modulating them with a drug will also have a therapeutic benefit.
How are advances in chemical biology helping to expand the druggable proteome?
Small molecules are still the main modality for intracellular targets. Deep pockets, typical for enzymes, are more easily tractable than shallow pockets typical for protein-protein interactions. Chemical biology has developed tools to explore different types of pockets. I am excited in particular by the potential of DNA-encoded libraries, and small fragment approaches with covalent modes of action. Some of these compounds will just be “binders” but these can be made into target degraders as PROTACs.
How can these advances help our understanding of disease biology?
If we had more chemical probes, we could use these in a standardised, controlled way to interrogate target function in cell-based models, organoids, and in some cases animal models. Yes, we have gene editing now, but that is not the same as pharmacological modulation.
We also need in vitro models that translate better to in vivo. Our old immortalised cell lines won’t do, we are going to need more work in primary cells, organoids, etc.
What are the main challenges facing scientists in this field?
Lack of standardised probe sets. Bad probe compounds, e.g. with bad selectivity, are still used and wrong conclusions drawn.
Lack of translational in vitro models.
Why is it important to bring industry and academia together to discuss this topic?
Academia brings creativity, agility, fast progress of new ideas and concepts, thinking out of the box.
Industry sometimes lacks these but knows how to develop a compound into a drug, which requires a host of technologies not readily available to academia. Also, industry requires a new generation of drug targets with better validation, and historically targets are often discovered in academia. Once a target hypothesis exists, academics and industry should ideally collaborate to figure out how to drug it.
What will be the main highlight of this conference?
Out of the posters presented, 4 were awarded a poster prize based on popular vote. Here we present the poster abstracts of four of the winners.
A computational modelling approach to characterizing postprandial glucose responses in individuals
Balazs Erdos (1), (2)*, Bart van Sloun (1), (2), Shauna O’Donovan (2), Michiel Adriaens (2), Natal van Riel (3), Ellen Blaak (4), Ilja Arts (2)
The large variability in the dynamic properties of the postprandial glucose response curves in individuals suggest that it is not sufficient to use average values or single time point measures of postprandial glycemia in order to characterize individuals’ glycemic control. Instead, approaches that are capable of capturing the dynamic events are necessary. In this study, we develop personalized computational models based on ordinary differential equations, to describe the glucose and insulin dynamics of individuals in response to an oral glucose tolerance test. We observed that these personalized models are capable of capturing a wide range of glucose and insulin dynamics including normal, prediabetic and type 2 diabetic responses as well as responses from intermediate states.
(1) TiFN, Wageningen, The Netherlands, (2) Maastricht Centre for Systems Biology (MaCSBio), Maastricht University, Maastricht, The Netherlands, (3) Dept. of Biomedical Engineering, Eindhoven University of Technology, Eindhoven, The Netherlands, (4) Dept. of Human Biology, NUTRIM School of Nutrition and Translational Research in Metabolism, Maastricht University, Maastricht, The Netherlands
Predict nephrotoxicity associated with cisplatin-based chemotherapy in testicular cancer patients
In industrialized countries, testicular cancer (TC) is the most common solid tumor in men between 20 and 40 years old and besides being one of the most treatable types of cancer, the long-term side-effects of chemotherapy are worrisome, since they are largely irreversible. Their severity is normally related to the total amount of chemotherapy received, which makes that an important factor to a successful treatment. The standard treatment for TC is 3 cycles of cisplatin, etoposide and bleomycin (BEP), being that the number of cycles can vary between 4-5 or more if the prognosis of the patient is intermediate or poor. Some of the late side-effects include nephrotoxicity, which can be measured by the drop in glomerular filtration rate after the patient follows chemotherapy. Materials and Methods: Integrative machine learning models were built using a dataset of 400 Danish individuals in order to identify clinical and/or genomics features and classify patients at higher risk of developing nephrotoxicity given a treatment of BEP-cycles. Results: First, only clinical features, such as age at the time of treatment, dose of cisplatin, patient’s prognosis, and number of cycles, were considered, and relevant features were selected to use in the classifier (AUC 0.66, SD 0.02). The classifier was then optimized by adding genomics markers, which helped improving the prediction (AUC 0.75, SD 0.02). Conclusions: Therefore, it is proposed a machine learning algorithm which, by helping predicting nephrotoxicity in advance, can benefit to improve chemotherapy efficacy in TC patients. These data driven models can also be applicable to other cancers, such as ovarian, bladder, and lung cancer where more elderly patients are at risk of nephrotoxicity and identification upfront will have direct clinical implications.
Poster currently not available
(1) Technical University of Denmark, Denmark, (2) Copenhagen University Hospital, Denmark, (3) University of Chinese Academy of Sciences, China
Loss of N-glycanase 1 alters transcriptional and translational regulation
Petra Jakob (1), William Mueller (1), Sandra Clauder-Münster (1), Han Sun (2), Sonja Ghidelli-Disse (3), Diana Ordonez (1), Markus Boesche (3), Markus Bantscheff (3), Paul Collier (1), Bettina Haase (1), Vladimir Benes (1), Malte Paulsen (1), Peter Sehr (1), Joe Lewis (1), Gerard Drewes (3), Lars Steinmetz (1)
N-Glycanase 1 (NGLY1) deficiency is an ultra-rare, complex and devastating neuromuscular disease. Patients display multi-organ symptoms including developmental delays, movement disorders, seizures, constipation and lack of tear production. NGLY1 is a deglycosylating protein involved in the degradation of misfolded proteins retrotranslocated from the endoplasmic reticulum (ER). NGLY1-deficient cells have been reported to exhibit decreased deglycosylation activity and an increased sensitivity to proteasome inhibitors. We show that the loss of NGLY1 causes substantial changes in the RNA and protein landscape of K562 cells and results in downregulation of proteasomal subunits, consistent with its processing of the transcription factor NFE2L1. We employed the CMap database to predict compounds that can modulate NGLY1 activity. Utilizing our robust K562 screening system, we demonstrate that the compound NVP-BEZ235 (Dactosilib) promotes degradation of NGLY1-dependent substrates, concurrent with increased autophagic flux, suggesting that stimulating autophagy may assist in clearing aberrant substrates during NGLY1 deficiency.
(1) EMBL Heidelberg, Germany, (2) Stanford University, United States of America, (3) Cellzome, Germany
Data integration for prediction of weight loss in clinically controlled dietary trials
Rikke Linnemann Nielsen (1), Marianne Helenius (1), Sara Garcia (1), Henrik Munch Roager (2), Derya Aytan (3), Lea Benedicte Skov Hansen (1), Mads Vendelbo Lind (2), Josef Vogt (1), Marlene Danner Dalgaard (1), Martin I Bahl (3), Cecilia Bang Jensen (1), Rasa Muktupavela (1), Christina Warinner (4), Vincent Appel (5), Rikke Gøbel (5), Mette B Kristensen (2), Hanne Frøkjær (6), Morten H Sparholt (7), Anders F Christensen (7), Henrik Vestergaard (5), Torben Hansen (5), Karsten Kristiansen (6), Susanne Brix Pedersen (1), Thomas Nordahl Petersen (3), Lotte Lauritzen (2), Tine Rask Licht (3), Oluf Pedersen (5), Ramneek Gupta (1)
Diet is a key strategy in weight loss management. Advances in omics technologies research allow analyses of determinants of clinical interventions outcomes. We have previously reported diet-induced weight loss in non-diabetic middle-aged Danes in two clinically controlled dietary trials where the content of whole grain or gluten was changed. However, it remains elusive how predictable weight loss is at the individual level. We here classify weight loss responders and non-responders from the whole grain and gluten trials by integrating multi-omics data (host genetics, gut microbiome, urine metabolome) together with physiology and anthropometrics into random forest models. The most predictive models for weight loss included features of diet, gut microbial species and urine metabolites (ROC-AUC:0.84-0.88, model only with diet type ROC-AUC:0.62). Furthermore, we demonstrate that a model ensemble is robust to missing information of microbiome and metabolome profiles given features of physiology (including postprandial response), host genetics and transit-time (ROC-AUC:0.72).
Poster currently not available
(1) Technical University of Denmark, Denmark, (2) University of Copenhagen, National Food Institute, Technical University of Denmark, Denmark, (3) National Food Institute, Technical University of Denmark, Denmark, (4) Harvard University, United States of America, (5) The Novo Nordisk Foundation Center for Basic Metabolic Research, University of Copenhagen, Denmark, (6) University of Copenhagen, Denmark, (7) Bispebjerg University Hospital, Denmark
Out of the 248 posters presented, 2 stood out from the rest and were awarded a poster prize based on popular vote. Here we present the abstracts and posters of the winners.
CalQTrace: Simultaneous Calculation and Quantification of 100,000 immune activation Traces at single-cell resolution using CNN
Authors: Liliana Barbieri (1), Kseniya Korobchevskaya (2), Azeem Ahmad (3), Huw Colin-York (1), Aurelien Barbotin (4), Glykeria Karanika (1), Loic Peters (5), Isabela Pedroza-Pacheco (4), Angela Lee (1), Lena Cords (1), Anish Priyadarshi (3), Dominic Waithe (6), Jana Kohler (6), Christoffer Lagerholm (6), Balpreet Singh Ahluwalia (3), Marco Fritzsche (2)
Quantification of immune cell activation is essential to the understanding of their effector function. Tracing activation signatures like cellular calcium release and the expression of surface markers in response to activation signals allows the classification of the course of immune cell activation from early triggering events to late differentiation. However, robust quantitative platforms for such measurements represent a major challenge, restricting the analysis to small single-cell population or more recently to cell ensembles with high-dimensional parameter analysis tools. Here, we introduce a combination of a convolutional neural network-based CalQTrace (Calculation and Quantification of Trace) software, together with a Graphical User Interface, and an optical high-throughput light-sheet platform, allowing the simultaneous fully automated quantitation of immune cell activation traces of >100,000 live immune cells. CalQTrace enables user-independent statistically robust classification and quantification of multiple fluorescent activation markers including calcium, CD25+/- expression, and cell viability tracking single cells in space and time within a 5 mm x 5 mm large-field-of-view, opening-up unprecedented insights into physiological activation tracing in living immune cells.
(1) MRC Human Immunology Unit, University of Oxford, United Kingdom (2) Kennedy Institute for Rheumatology, University of Oxford, United Kingdom (3) The Arctic University of Norway, Norway (4) University of Oxford, United Kingdom (5) University College London, United Kingdom (6) Weatherall Institute of Molecular Medicine, University of Oxford, United Kingdom
Poster currently not available
Bleaching-insensitive STED microscopy with exchangeable fluorescent probes
Authors: Christoph Spahn (1), Florian Hurter (1), Mathilda Glaesmann (1), Jonathan Grimm (2), Luke Lavis (2), Hans-Dieter Barth (1), Marko Lampe (3), Mike Heilemann (1)
Photobleaching affects image quality and resolution in fluorescence microscopy, and thus limits the extractable information. This is in particular relevant for super-resolution microscopy where typically high laser intensities are used. In order to minimize photobleaching, we repurposed the use of exchangeable fluorescent probes, as used in single-molecule localisation microscopy methods such as Point Accumulation for Imaging in Nanoscale Tomography (PAINT) , for STED microscopy. We demonstrate pseudo-permanent labeling of target structures and constant exchange of photobleached fluorophores. This concept allows for whole-cell, 3D, multi-color and live-cell STED microscopy . Using transiently binding hydrophobic dyes and fluorophore-labeled major minor groove binders [3, 4], we visualised the nanostructure of chromatin, cell membranes and organelles in bacterial and mammalian cells in 3D. To expand the range of targets, we employed oligonucleotide-labeled antibodies that transiently bind fluorophore-labeled oligonucleotides, as used in single-molecule super-resolution imaging with DNA-PAINT , and demonstrate multi-color STED imaging.
 Sharanov and Hochstrasser, PNAS 103 (50), 18911-18916 (2006)
 Spahn et al., Nano Letters 19 (1), 500-505 (2019)
 Lukinavičius et al., Nature Communications 6, 8497 (2015)
 Spahn et al., Scientific Reports 8, 14768 (2018)
 Schnitzbauer et al., Nature Protocols 12(6), 1198-1228 (2017)
110 researchers came together at the EMBL Advanced Training Centre in Heidelberg, Germany for 3,5 days of talks, posters and networking. Here we present the work of 4 scientists who received best poster awards at the conference by popular vote.
Engineering portability of the CcaSR light switch for the control of biofilm formation in Pseudomonas putida
Authors: Angeles Hueso-Gil (1), Ákos Nyerges (2), Csaba Pál (2), Belén Calles (1), Victor de Lorenzo (1)
Two of the technical challenges faced by contemporary microbiology involve controlling gene expression using light and regulating bacterial biofilm formation, determined by the intracellular levels of the secondary messenger c-di-GMP. CcaSR system is one of the light switches repeatedly used for transcription induction in Escherichia coli. This two-component system represented a good candidate for its adaptation to Pseudomonas putida. Previous attempts have tried to use this microorganism as chassis for the implementation of new pathways, being biofilm formation an important function to control. To this end, we unified CcaSR components in one single construct and randomly mutagenized their regulatory regions to find a clone with a balanced expression of the system key parts inside P. putida. The combination of this novel mutagenization process with a proper screening, which included a first sorting of the libraries and the later isolation of colonies, lead us to a clone with a much improved induction by green light. The selected variant had a notable capacity in response to green light. Finally, optimized CcaSR was used to control the expression of super-efficient variant of PleD, a diguanylate cyclase of Caulobacter which allowed a tight control of c-di-GMP levels, and therefore, of biofilm production.
Genetic code expansion is a powerful tool to study and control protein function with single-residue precision. It is widely used to e.g. perform labeling for microscopy or to photocontrol proteins. This is achieved by introducing an orthogonal tRNA/synthetase suppressor pair into the host, to recode a stop codon to incorporate a noncanonical amino acid (ncAA) into the nascent chain. This technique is codon-specific, but it cannot select specific mRNAs, so naturally occurring stop codons could be suppressed leading to potential interference with housekeeping translation. Nature avoids cross-talk between cellular processes by confining specific functions into organelles. We aimed to design an organelle dedicated to protein engineering, but as translation is a complex process requiring hundreds of factors to work together, membrane-encapsulation would not be feasible. Inspired by the concept of phase separation we hypothesized that such an organelle could instead be designed membraneless. Phase separation can generate high local concentrations of proteins and RNAs in cells and has recently gained attention owing to its role in the formation of specialized organelles such as nucleoli or stress granules. Despite being membraneless and constantly exchanging with the cytoplasm/nucleoplasm, these organelles still perform complex tasks, such as transcription. We combined phase separating proteins with microtubule motor proteins to generated orthogonally translating organelles in living cells that contain an RNA-targeting system, the stop codon suppression machinery and ribosomes. These large organelles enable site- and mRNA-specific ncAA incorporation, decoding one specific codon exclusively in the mRNA of choice. Our results demonstrate a simple yet effective approach to the generation of semi-synthetic eukaryotic cells containing artificial organelles to harbor two
distinct genetic codes, providing a route towards customized orthogonal translation and protein engineering.
(1) Johannes Gutenberg University Mainz, Germany (2) Institute of Molecular Biology, Germany (3) EMBL Heidelberg, Germany
Metabolic perceptrons for neural computing in biological systems
Amir Pandi (1), Mathilde Koch (1), Peter Voyvodic (2), Paul Soudier (1), Jerome Bonnet (2), Manish Kushwaha (1), Jean-Loup Faulon(1)
Synthetic biological circuits are promising tools for developing sophisticated systems for medical, industrial, and environmental applications. So far, circuit implementations commonly rely on gene expression regulation for information processing using digital logic. Here, we present a new approach for biological computation through metabolic circuits designed by computer-aided tools, implemented in both whole-cell and cell-free systems. We first combine metabolic transducers to build an analog adder, a device that sums up the concentrations of multiple input metabolites. Next, we build a weighted adder where the contributions of the different metabolites to the sum can be adjusted. Using a computational model fitted on experimental data, we finally implement two four-input of metabolite combinations by applying model-predicted weights to the metabolic perceptron. The perceptron-mediated neural computing introduced here lays the groundwork for more advanced metabolic circuits for rapid and scalable multiplex sensing.
The bottom up recreation of cellular processes into synthetic compartments has, in recent years, emerged as an exciting line of research with which to study biological processes in a controlled environment. However, the interior of a living cell is a difficult milieu to mimic in bottom-up synthetic cells, as it is an environment crowded with high concentrations of many different biomacromolecules. In this work, we describe the development of a powerful new tool to more accurately emulate the cell cytosol in discrete coacervate-based protocells. The coacervate core utilized herein not only provides an inherently crowded and highly charged microenvironment, but has also been chemically modified to interact specifically with recombinantly expressed proteins. Our method leverages the well-established binding of His-tagged proteins to Ni2+-nitrilotriacetic acid, which ensures that macromolecules are taken up in a highly efficient, yet gentle manner, thus preserving biological activity. The straightforward method allowed for both control over the amount taken up and an increased local concentration. Moreover, the engineered uptake of proteins was then employed to study two key aspects: the effect of the Ni-NTA interaction on the diffusivity of incorporated proteins, and the enhancement in activity of an encapsulated two-enzyme cascade. This direct and targeted method of protein uptake into a discrete, membrane bound platform is a significant step forward for synthetic cells, and will enable the engineering of highly complex enzyme and signaling networks with increasingly life-like properties.
Poster currently not available
Eindhoven University of Technology, The Netherlands
Biodiversity – in all its forms and interactions – is the variety of life on Earth. Climate change is exacerbating biodiversity loss, and vice versa. Ahead of the EMBO | EMBL Symposium ‘The Organism and its Environment’ (1–4 March 2020), we talk to Scientific Organiser and EMBL Director General Edith Heard about the impact the environment has on biodiversity and the role of research in solving global challenges.
Does the environment play a large role in the creation of biological diversity?
Biodiversity is the variety of life on Earth. This life, in all its shapes and sizes, occurs in the context of ecosystems: it relies on and interacts with other organisms and the physical environment. Biodiversity represents the collective ‘knowledge learned’ by evolving species over millions of years, about how to survive the vastly varying environmental conditions Earth has and is experiencing. These varying environmental conditions cause natural variations in biodiversity, as well as genetic and epigenetic changes, within and between species over time. Today, scientists are trying to understand the basis of these natural variations, as they will allow us to understand how life evolves.
But biodiversity is also a measure of the health of any ecosystem. Recent trends in biodiversity loss show very clearly that humans are destroying ecosystems on a massive scale. According to the Director General of the World Wildlife Fund (WWF), increased pollution, deforestation, climate change and other manmade factors have created a “mind-blowing” crisis. The WWF Living Planet Report 2018 (WWF LPR, 2018) also states that freshwater fish populations have declined by more than 80% on average since 1970 and half of the world’s shallow water coral reefs have been lost in the last 30 years (WWF LPR, 2018). Alongside this, deforestation of tropical rainforests means we are currently losing more than 100 species of plants and animals a day (Holley, 2017). In short, human’s influence on the environment greatly impacts biodiversity and we are currently burning the library of life.
How can you determine the effect of the environment on an organism?
The environment can affect an organism in a multitude of ways. The impact can be transient or longer term; within an individual or across generations. The environment can also lead to molecular, cellular, physiological or behavioural changes. For example, the expression of genes in an organism can be influenced by the external environment, such as where the organism develops or factors associated with where it is located. Gene expression could also be influenced by an organism’s internal environment, including hormones or metabolism. Finally, the genome itself – genetic factors that vary between individuals in natural populations – could also influence gene expression.
Untangling the impact of genetic and environmental variation can be very challenging and for a long time, scientists have tended to focus on minimising variations in the environment in order to understand how changes in genotype affect phenotype. This, alongside the deeply embedded “one genotype = one phenotype” metaphor, has meant that environmentally induced phenotypic variation has been ignored in favour of ‘‘more useful and precise’’ study of genetic polymorphisms. This is despite the fact that from as far back as the early 1900s, scientists have known that the phenotype of an individual depends on the interaction between its genotype and environmental cues! Today, we finally have the power to consider the impact of the environment on phenotype. We can make precise measurements at the molecular, cellular and organism scales in controlled environments that can be varied and we can sequence genomes at the same time.
We can also take human data paired with environmental data – for example in the context of some of EMBL’s research interests such as infectious disease and microbiomes – to understand the quantitative effects of the environment and its influence on human biology. Pioneering projects such as Tara Oceans have also allowed us to research the interactions between organisms and the environment by generating reference data, discovering emergent ecological principles and developing predictions about how ecosystems will be affected by a changing environment. Understanding how organisms exist together and in changing environments is of fundamental importance for our understanding of biological principles and our knowledge of life.
What challenges are currently being faced in this field?
Understanding the behaviour of individual molecules, cells or whole organisms is already challenging. Understanding how the environment influences an organism – or populations of organisms – represents a whole new scale in complexity. This is an area that I think EMBL could uniquely contribute to in the future. It will be necessary to shift from researching organisms mainly in the laboratory to studying them in their environment. We will also need to ensure the rapid development of technologies and tools to meet these scientific needs. Alongside this, we need new approaches to integrate large, complex data sets and make sense of them. To rise up to this challenge, we need theory. We are now in a unique position to address the dynamics and complexity of living matter across multiple scales and in the context of changing environment. But we need theory to address societal and planetary issues too. We must aim for a rate of scientific discovery that outpaces the rate of calamity such as biodiversity loss, ecosystem degradation, epidemics and climate change.
What can be done to prepare for the future with regard to biological diversity, the organism and its environment?
Research, research and more research! Environmental problems such as the hole in the ozone layer or acid rain were solved by sound scientific approaches. We need to learn from these past scientific and societal successes. Today the ever-increasing numbers of new technologies are allowing us to collect, measure and store data at unprecedented scales. We also need to bring ecologists, zoologists, population geneticists and environmental experts together to address these research questions. Together we can apply cutting-edge technology with rigour, attract new scientific talent and disseminate knowledge to global communities.
What inspired you to organise this symposium?
As a geneticist and epigeneticist, I have explored the intersection between genotype and the environment and how that produces a phenotype. From observing many areas of research – ranging from social insects such as bees and ants, to plant vernalisation and variations between identical twins – I felt that the time is ripe to bring together scientists from many different areas. I also wanted this to be a symposium that would attract scientists from different areas to EMBL.
At EMBL, we want to understand the molecular basis of life. Until now, EMBL has been known for exploring genomic, molecular, structural and cell biology at the level of individual organisms. Looking ahead, we want to study organisms in the context of their physical and biological environments – not just in isolation. In order to truly understand life on Earth, we need to study organisms in nature, not just in the lab. One way to understand life at the molecular level will be to try to bring relevant ecosystems back to the lab, to measure and perturb them under controlled conditions. The speakers we’ve invited are experts from many different areas of biology or ecology, and will bring exciting new perspectives to our research.
What is the greatest benefit of this symposium for the scientific community?
The symposium is an opportunity to address how organisms are influenced by a changing environment. It will bring together different research disciplines and go beyond pure genetic or ecological perspectives of phenotypic variation. Geneticists, molecular biologists, evolutionary biologists and ecologists do not necessarily meet under ordinary circumstances. This meeting will enable such interactions and cross-fertilisation.
What will be the main highlight of the symposium?
Today we are in a unique position to address the complexity and dynamics of life at multiple scales, from molecules to ecosystems. We also need to consider the idea that change – including in the environment – is not necessarily a bad thing. After all, without change, evolution could not occur and none of the amazing biodiversity of life on our planet would exist! I hope that a highlight of this symposium will be some wonderful new insights into evolutionary processes.
Holley D., (2017). General Biology II, Organisms and Ecology. Indianapolis: Dog Ear Publishing, 898.