Publication highlights

Go inside our research

Explore a selection of research cases studies from the past five years.

Read now
A Crick researcher reading a scientific paper on a screen.

Intro

Researchers at the Crick are tackling the big questions about human health and disease, and new findings are published every week.

Our faculty have picked some of the most significant papers published by Crick scientists, all of which are freely available thanks to our open science policy.

Highlights

Filter by year of publication

Neutrophils (shown in brown) a type of immune cells helping breast cancer cells to grow in the lung.

Changes in circulating immune cells may be able to reveal the presence of breast cancer

Research led by a team of scientists at Francis Crick Institute and clinicians at Imperial College London investigated whether changes in certain circulating immune cells (neutrophils) were detectable in newly diagnosed patients with breast cancer. The team recruited women that, after routine mammograms and subsequent biopsy, were diagnosed with breast cancer. Their disease was very early stage and asymptomatic. The researchers collected blood before treatment, isolated and analysed circulating neutrophils (one of the more abundant immune cells in blood) and compared it to neutrophils from age matched healthy volunteers.

The results showed that different cancer specific activities in the cells were detectable in circulating neutrophils from early cancer patients compared to healthy volunteers. These activities were not detected in patients with benign breast disease. This study only included a limited number of patients, but it represents proof-of-concept evidence suggesting that disruption to neutrophils occurs very early in the disease. Defining these disruptions could represent not only a way to understand how they contribute to tumour progression, but also could be exploited as biomarkers for early disease.

Circulating neutrophils from patients with early breast cancer have distinct subtype-dependent phenotypes

Published in Breast Cancer Research

Published

Modelling and enhancing migration of hiPSC-derived myogenic progenitors

Cell therapies to treat severe muscular dystrophies are inefficient. Major hurdles include the limited ability to expand mature myogenic cells in vitro, as well as the minimal migration capacity of myogenic cells upon transplantation, which inhibits dispersal into affected tissues. Researchers in the Tedesco lab have used directed iPSC differentiation, single-cell profiling, microfluidics and 3D tissue engineering to show that hiPSC-derived muscle satellite stem cells, which may be useful in cell therapies for muscular dystrophy, can have their in-vivo migration enhanced through activating the NOTCH and PDGF pathways, via treatment with DLL4 and PDGF-BB.

Assessing and enhancing migration of human myogenic progenitors using directed iPS cell differentiation and advanced tissue modelling

Published in EMBO Molecular Medicine

Published

Mechanical disengagement of the cohesin ring

Mechanical disengagement of the cohesin ring

In healthy cell division, the replicated DNA forms sister chromatids that must remain connected until separation later in the process. It’s only then that X-shaped chromosomes must be segregated symmetrically: each sister chromatid (one half of the X) is pulled to the opposite edges of the dividing cell by microtubules - protein filaments that generate force – to give rise to two daughter cells with an equal amount of genetic material. A ring-shaped protein called cohesin physically links sister chromatids and, like an elastic band, resists the forces generated by microtubules. Not only is the absence of cohesion lethal, but mutations in it can lead to cancer and incurable developmental disorders.

In this research by the Molodtsov and Uhlmann groups, the force that the cohesin complex can withstand is revealed. Using optical tweezers, the researchers pulled apart the DNA molecules tied by cohesin, showing that one cohesin ring is capable of embracing two DNAs and can resist up to 20 piconewtons of force, and when it breaks, it always opens at its weakest point: the hinge domain. These findings reveal that 40 cohesins are sufficient to oppose the tension generated in mitosis, whilst larger forces release the sisters. For the first time, this work lifts the veil on cohesin’s physical properties, bringing us closer to understanding how it is dysregulated in disease.

Mechanical disengagement of the cohesin ring

Published in Nature Structural and Molecular Biology

Published

Bone marrow micro-environment in leukemia

A study led by the Bonnet lab looks at how acute myeloid leukemia cells interact with and alter bone marrow. The team has produced an omics repository of potential biomarkers for different bone marrow cell populations.

Integrated OMICs unveil the bone-marrow microenvironment in human leukemia

Published in Cell Reports

Published

An asymmetric pattern of PAR proteins

Going with the flow: how to polarise a cell

Cell polarisation is a fundamentally important ordering process that breaks the internal symmetry of a cell by establishing a preferential axis. The Goehring lab used the nematode worm C. elegans to study why the polarity protein PAR-3 needs to aggregate to efficiently move to the front of the worm embryo. Contrary to previous theories, they found that the size of molecule aggregates did not directly affect PAR-3 movement. Instead, what matters is how tightly these molecules stick to the membrane. This discovery challenges existing ideas about cell transport mechanisms and highlights the role of membrane stability in cellular processes. Defects in cell polarisation can disrupt numerous processes, so developing a systems-level understanding may enable new therapies for developmental defects and cancer.

Design principles for selective polarization of PAR proteins by cortical flows

Published in The Journal of Cell Biology

Published

Watching tissues developing in real time

During development, multicellular organisms undergo stereotypical patterns of tissue growth in space and time, but how this is orchestrated remains unclear, largely due to the difficulty of observing and quantitating this process in a living organism. The Tapon and Salbreux labs used live imaging and computational methods to quantitatively analyse developmental growth in the fruit fly adult abdominal epidermis. Abdominal growth is initiated by degradation of the basement membrane to which the epidermal progenitor cells are attached and is terminated by rapid exit from the cell cycle, rather than a gradual slowdown, as occurs in some other tissues. Different developing tissues can therefore achieve their final size using distinct growth termination strategies.

ECM degradation in the Drosophila abdominal epidermis initiates tissue growth that ceases with rapid cell-cycle exit

Published in Current Biology

Published

Ciccarelli banner

Uncovering cancer-immune system interactions could inform how patients respond to immunotherapy

Researchers at the Francis Crick Institute and King’s College London have revealed the complex interactions between cancer and the immune cells that surround a tumour, with the potential to inform how patients will respond to immunotherapy. The researchers analysed thousands of samples across 32 types of cancer to examine the way that cancer dynamically interacts with the tumour immune microenvironment (TIME), allowing the disease to flourish.

Focusing on a class of genes called cancer drivers, they identified 477 of these cancer drivers that interact with multiple features of the TIME, suggesting that they drive the formation of cancer by disrupting biological processes within the cell as well as interfering with the immune system.

Mechanistic insights into the interactions between cancer drivers and the tumour immune microenvironment

Published in Genome Medicine

Published

Polarity in dividing cells

Coupling cell division and polarity to keep cells organised

The vast majority of cells exhibit ‘cell polarity’ – they typically must distinguish their tops from their bottoms and their front from their backs. In complex organisms like animals or humans, cell polarity must be coordinated between cells to generate functional tissues and organs. Such coordination poses a challenge during embryonic development or in regenerating tissues as cells are continuously growing and dividing. To ensure cells are oriented correctly with respect to one another, cell polarity and cell division must be coupled.

When a cell divides, it generates a flow of material towards its centre, which aids the process of cell division and contributes to the forming boundary between what will become the two new daughter cells. Here the researchers show that this flow of material also transports a key molecule, PAR-3, into this forming boundary. The local flow-dependent accumulation of PAR-3 breaks the internal symmetry of the daughter cells and ensures that the polarity of each daughter cell is oriented properly with respect to its sister. This simple physical mechanism for coupling cell division and polarity may be a general method for keeping cells organised in actively dividing tissues.

Cleavage furrow-directed cortical flows bias PAR polarization pathways to link cell polarity to cell division

Published in Current Biology

Published

Synthetic sugars

Molecular decision making in glycosaminoglycan synthesis

Cell-surface and secreted proteins play critical roles in human development, growth factor signalling, and cell adhesion. Proteoglycans are an important subset of these proteins and are modified with long chains of sugar molecules called Glycosaminoglycans (GAGs) such as heparan sulphate (HS) or chondroitin sulphate (CS), but they all start with the same four sugars – only after the addition of the fifth sugar is the fate of the growing chain sealed.

While protein and DNA synthesis are template-driven, from DNA or RNA, synthesis of the proteoglycan GAG chains are not. In a collaboration between the Crick and Imperial, the researchers devised a synthesis system to allow precise control of eight of the enzymes in the biosynthesis pathway. They discovered that chrondroitin sulphate is the “default” modification, and that the enzyme responsible for priming chrondroitin sulphate synthesis modifies all sites equally. They also found that the enzyme responsible for priming heparan sulphate synthesis (EXTL3) has a positively charged patch that interacts with negatively charged amino acids near the attachment site and will only modify certain substrates. This will help to predict how mutations surrounding the glycosaminoglycan attachment sites could be implicated in diseases like cancer or developmental conditions.

Molecular mechanism of decision-making in glycosaminoglycan biosynthesis

Published in Nature Communications

Published

Zebrafish embryo stages

Accelerating developmental biology research with deep learning

Zebrafish are often used in biological research due to their transparent embryos and rapid development, allowing scientists to easily observe and study their growth processes. This makes them particularly valuable for understanding human biology and diseases. Sometimes, these fish can experience developmental delays due to genetic issues or experimental treatments. Traditionally, scientists have manually checked the fish's growth against standard charts, a slow and not always precise method.

Researchers at the Crick developed KimmelNet, an artificial intelligence tool, to make this process quicker, less subjective and more reproducible. KimmelNet analyzes standard microscope images of zebrafish embryos to determine their growth stage and can reliably spot developmental delays, requiring just a small number of images to do so confidently. Furthermore, the tool adapts well to new data, and its performance can be even further enhanced with some additional fine-tuning.

This innovation could significantly speed up research involving zebrafish, making studies more efficient and reliable. Plus, the approach has the potential to be applied to other organisms as well, broadening its utility in the field of biological research.

Automated staging of zebrafish embryos with deep learning

Published in Life Science Alliance

Published

New insights into HIV infection

A study from the Bishop lab has looked into HIV-1 uncoating, the process by which the viral core breaks down during infection. Their work suggests that uncoating or remodelling of the HIV-1 capsid lattice occurs at the nuclear pore, and that this step is essential for a productive infection.

HIV-1 requires capsid remodelling at the nuclear pore for nuclear entry and integration

Published in PLOS Pathogens

Published

Gamma delta T cells monitor tissue health

Although distress signals from microbes and tissue damage have long been appreciated as instigators of immunity, how surface tissue (epithelial) health is monitored remains poorly understood. Work from the Hayday lab has identified how gamma delta T cells, a population of specialised immune cells, sense the body’s status quo, enabling them to assess the health of surface tissues and protect against UVR-induced DNA damage and inflammation, two cancer-disposing factors.

Normality sensing licenses local T cells for innate-like tissue surveillance

Published in Nature Immunology

Published

Structural insights into influenza infection

Hemagglutinin (HA), the receptor binding and membrane fusion glycoprotein of influenza virus, is synthesised as a precursor (HA0) that requires cleavage and exposure to low pH for fusion activity during virus entry. Researchers in the Rosenthal and Gamblin labs have used cryo-EM to image an extensive conformational change in the HA0 protein at low pH that may mimic an intermediate in the structural transitions by which HA mediates membrane fusion. Unlike the case with HA, however, the change is reversible. The results provide insight into the role of cleavage in membrane fusion activation and have implications for the action of antiviral drug candidates and cross-reactive anti-HA antibodies that can block influenza infectivity.

Reversible structural changes in the influenza hemagglutinin precursor at membrane fusion pH

Published in Proceedings of the National Academy of Sciences of USA

Published

Give and take in the exometabolome

Microbial communities are composed of cells of varying metabolic capacity, and regularly include auxotrophs that lack essential metabolic pathways, whose usefulness to the community is therefore puzzling. A study from the Ralser lab revealed that metabolic changes in auxotrophs enrich the microbial community exometabolome—the mixture of secreted extracellular metabolites—and increase drug resistance. The findings could help the development of more effective antimicrobial treatments.

Microbial communities form rich extracellular metabolomes that foster metabolic interactions and promote drug tolerance

Published in Nature Microbiology

Published

Dopaminergic neurons generated from human induced pluripotent stem cells. Blue stain for the nuclei and yellow stain for tyrosine hydroxylase, a dopaminergic neuron marker.

Scientists harness the power of AI to shed light on different types of Parkinson’s disease

Researchers at the Francis Crick Institute and UCL Queen Square Institute of Neurology, working with technology company Faculty AI, have shown that machine learning can accurately predict subtypes of Parkinson’s disease using images of patient-derived stem cells. They generated stem cells from patients’ own cells and chemically created four different subtypes of Parkinson’s disease. They then imaged the disease models in microscopic detail and labelled key cell components including lysosomes, and ‘trained’ a computer program to recognise each subtype, which was then able to predict the subtype when presented with images it hadn’t seen before. Their work has shown that computer models can accurately classify four subtypes of Parkinson’s disease, with one reaching an accuracy of 95%. This could pave the way for personalised medicine and targeted drug discovery.

Prediction of mechanistic subtypes of Parkinson’s using patient-derived stem cell models

Published in Nature Machine Intelligence

Published

Illustrations and equations can be used to make sense of biology.

New software to detect and track individual cells growing in colonies

Cell biology techniques are creating more and more images that need to be analysed. In this work we present software, DM3D, that we developed to detect and track individual cells growing within small colonies. Each colony is between 10 and 30 cells that have been imaged with an advanced microscopy technique, which allows us to record movies of growing cells for long periods of time. The resulting 3D timelapse movies need to be analysed. Our software provides an interactive environment to combine two powerful techniques for cell segmentation: active contours (a way to find the outline of an object in an image) and neural networks (a series of computer algorithms used to understand relationships in data).

The software lets a user open a movie and called a 3D mesh – a type of model which defines the surface of an object with triangles – that represents a cell or a nucleus, then the mesh can be deformed by an active contours algorithm to improve the representation. It provides an interactive way for somebody to create 3D representations of their data, which are then used to train a neural network. The neural network is used to analyse large numbers of cells and colonies over long periods of time.

This work is important because it looks at how can we get the most out of neural networks, and merges two techniques to counteract their individual weaknesses. Neural networks are susceptible to a training history and it can be difficult to interact with the output to tune or verify it. Active contours require an initialisation step and need to be explicitly "told" to avoid artifacts – an output which has been generated by the training process. Combining the two techniques leads to fast and accurate results.

Active mesh and neural network pipeline for cell aggregate segmentation

Published in Biophysical Journal

Published

Malaria parasite inside red blood cells

A DNA-binding protein that regulates malaria parasite development and pathogenesis

An international team of researchers led by Arnab Pain (Professor at King Abdullah University of Science and Technology, Saudi Arabia) and Tony Holder (recently retired from the Crick) with additional scientists in Oxford, California, Saudi Arabia and India, have identified a key protein which regulates human malaria parasite development and disease. During the cycle of malaria gene expression within infected red blood cells, this DNA-binding protein (in the APiAP2 transcription factor family) controls parasite differentiation, replication, and release from the host cell. It also allows the parasite to modify the host red blood cell, which helps it to evade the immune system. By knocking out the gene producing the DNA-binding protein, it was shown to be essential at two stages within the 48-hour cell cycle. Novel therapeutic strategies to combat malaria are suggested by these findings.

DNA-binding protein PfAP2-P regulates parasite pathogenesis during malaria parasite blood stages

Published in Nature Microbiology

Published

Core cell cycle control

The organisational principles of the eukaryotic cell cycle have previously been put down to two opposing models of enzyme activity. Researchers in the Nurse Lab have developed proteomics assays that allow them to monitor the levels of enzymes in yeast that control the cell cycle. They found that the cell cycle is controlled through a hybrid of both models, although the contribution of one strongly outweighs the other. It is likely that these findings in yeast reflect core control principles shared by eukaryotes.

Core control principles of the eukaryotic cell cycle

Published in Nature

Published

New insight into how Toxoplasma exits host cells

The processes governing host cell exit by the intracellular parasite Toxoplasma gondii are regulated by several signalling pathways, but how these pathways are connected remains largely unknown. Researchers in the Treeck lab have used a combination of phosphoproteomics, lipidomics, reverse genetics and biochemical assays to demonstrate the presence of a feedback loop linking calcium signalling with the upstream cyclic nucleotide and phospholipid signalling pathways to enhance signalling during Toxoplasma egress. The work improves understanding of how the parasite integrates various signalling inputs to a single phenotypic outcome.

A positive feedback loop mediates crosstalk between calcium, cyclic nucleotide and lipid signalling in calcium-induced Toxoplasma gondii egress

Published in PLOS Pathogens

Published