Image credit: Wellcome Sanger Institute
This week, the latest high-impact paper from the Wellcome Sanger Institute’s Human Cell Atlas team was published. The researchers describe, for the first time; new cell types in the heart, the characteristics of the cells that make the heart beat, a molecule that could be an early warning sign of heart failure, a newly-discovered way heart cells communicate with each other, and a new resource that could help repurpose drugs.
The study brought together histologists, cardiologists, immunologists, cell biologists, software developers, bioinformaticians and specialist technicians – primarily over Zoom as much of the work was done during the pandemic. The team is one of the first in the world to use techniques that enable them to see individual cells in unprecedented detail. Building on previous work in their group in spatial genomics, they were also able to locate each individual cell to its precise location in the heart.
The lead authors of the paper are James Cranley, PhD student and trainee cardiologist, and Kazumasa Kanemaru, Postdoctoral Fellow. Both work in Sarah Teichmann’s group at the Sanger Institute. Here, they talk about some of the serendipitous collaborations that enabled this huge achievement, and share their excitement about the latest, cutting-edge techniques in spatial genomics and single-cell sequencing.
James and Kazumasa started at the Sanger Institute in 2020 and 2021. James had paused his cardiology training to undertake a PhD, while Kazumasa joined from the University of Tsukuba, Japan, as a Postdoctoral researcher.
James was drawn to the project after the first version of the Heart Cell Atlas was published in 2020. At the time, he was treating patients with cardiac arrhythmias and other types of heart rhythm disturbances.
“When I was looking for PhD projects I could see the power of single-cell transcriptomics to unpick biology in an unbiased fashion. At that time there was no published heart cell atlas, so, seeing that Sarah was a leader in the field, I approached her to discuss atlasing the heart. It transpired her team had already been working on a first version, which was nearing publication. This work was ground-breaking but some important cells were not described. This included cells from the cardiac conduction system – the cells which initiate and coordinate the heartbeat. Making a version two atlas that included the conduction system cells and using spatial transcriptomics became the core of my PhD fellowship application.”
James Cranley working in the lab
James Cranley
Kazumasa Kanemaru at work in the lab
Kazumasa Kanemaru
Kazumasa trained as an immunologist in Japan, though is also a qualified medical doctor. Before joining the Sanger Institute, he was researching the function of single molecules in mouse dermatitis models.
“Although I enjoyed and understand the importance of detailed analyses focusing on certain molecules, I felt that my work wasn’t comprehensive, like what we're doing now. Then single-cell genomics and spatial transcriptomics arrived – there was a leap in the technology which sheds light on human biology. It seemed to me that with this technology we could really discover something from humans. That’s when I approached Sarah to see if I could join her group.”
Both reflect on the complementary nature of each other’s skill sets.
“It was one of the lucky things, I think. James brought and integrated his knowledge from cardiology – a whole bunch of knowledge about the cardiac system that we just didn’t have as molecular biologists,” says Kazumasa.
“Well, the immunology side was vital, too!” laughs James. “One of the stories in the paper - not to do with the conduction system – was that we used spatial technologies to find niches, or groups, of cells that dwell together and function and interact with each other. And one of those niches was an immune niche, which hasn't been described before.”
The cells they describe form a defensive structure around the surface of the heart, which guards against invading pathogens. It would protect the heart from an infection in the lungs, for example. “Kazumasa’s knowledge of the immune system and excellent annotation of the cell types allowed us to make that discovery,” adds James.
New skill sets
As well as James and Kazumasa, there are bioinformaticians, histologists, technical experts and software developers all working together on cell atlas projects like this one. However, to start with, one key skill set was missing - they didn’t know an anatomist specialised for the cardiac conduction system.
So James looked up the author of his medical school heart text book, and found Prof. S. Yen Ho based at Imperial College London. Just one email later, and she was on board. “We needed an expert in the conduction system – it was amazing to get her involved. She’s been fantastic,” says James.
A precious resource
Another aspect of the collaboration involved in a study like this are the organ donor teams. The hearts in the study are generously donated for transplant very occasionally can’t be used for that purpose, in that scenario these incredibly precious samples may become available for research. The team at the Sanger Institute work with organ donor teams in Cambridge, Newcastle and London.
“To preserve the cells it’s vital to process samples fresh. So I’d travel from Cambridge to Newcastle at any time of day or night to get there as quickly as possible,” says James.
The team would often travel to London in the middle of the night too. The dissection of a heart would be a team of four or five people, from both Sanger and Imperial College London, working together to collect tissue and cell samples. “The preservation process takes several hours,” says Kazumasa, “preserving tissue using both freezing methods and formalin-fixed paraffin-embedded (FFPE) techniques.”
Single cell, multiple outputs
The technology used by the team produces a combination of different types of data for each individual cell. The kit that enables this has only been commercially available for two years, and the team is one of the first to publish research using it.
The first type of data they can produce is from the transcriptome – that is the RNA in the cell – which shows which genes are switched on and off. The second dataset uses an assay called ATAC-seq which tells researchers which parts of the genome are ‘open’. While the majority of our DNA is tightly wound up in our cells, some regions are open. These might be regions where regulatory elements, such as transcription factors, bind to switch on multiple genes.
In combination these give detailed information on not only on which genes are active (or ‘expressed’), but also how that is regulated.
James explains the importance of this knowledge. “This is a very fundamental thing, because if you want to make cells in a dish resemble human cells, changing expression of single genes is not enough, you need to drive change in a whole raft of genes. For that, understanding the regulation of expression is vital.”
The other technical innovation is in the spatial data. “We use the cell profiles that we’ve generated using these single-cell sequencing techniques. Then, we map those profiles to their spatial coordinates,” says James. This mapping technique has been developed over the last few years by Vitalii Kleshchevnikov and colleagues working in Omer Bayraktar’s group at the Sanger Institute. They created a computational tool, called cell2location, which combines single-cell sequencing data with spatial information, to visualise the relationships between cells and better understand tissue biology.
Speaking in 2022, Omer said, “I’m very excited about the potential for cell2location to change the way we observe life at a molecular level. Now we have a tool that’s better than a microscope and can provide us with more detail than we could have ever imagined.”
Turning points
This computational mapping using cell2location has been critical, enabling the team to determine the location of the cells they studied in the heart. They were able to detect multicellular niches – groups of cells that work together.
To validate their findings, Professor Ho would annotate the microscope slides, labeling the key parts of the conduction system. The team could then see how the structures they determined with the computational methods matched up.
“I remember the morning, very vividly actually, when Kazumasa showed us his finding that the sinoatrial node structure could be deconstructed into two zones, not visible to the naked eye: a core, and a periphery.”
Heart tissue slide with annotated functions
Move the slider to see the heart tissue slide and the functional areas annotated by Prof. S. Yen Ho. The sinoatrial node is in red.
Central node and peripheral node
Move the slider to see the two different zones within the sinoatrial node - central and peripheral - revealed by spatial genomics conducted by Kazumasa Kanemaru.
“That was pretty exciting,” says James.
“Bringing Professor Ho on board was an important part of that finding,” adds Kazumasa.
The future of the Heart Cell Atlas
One area where their data could make a difference is for researchers working on cell therapies for heart diseases. While it is possible to make heart cells in a dish, when implanted they beat excessively, causing harmful arrhythmias. The Heart Cell Atlas reveals which groups of genes could be targeted to control that over-activation. If that could be achieved, it would be a step change for regenerative medicine in cardiology.
“I’d like to discover more about human biology, ” says Kazumasa.
Their work in the heart is part of the global Human Cell Atlas (HCA) project, which aims to map every single cell type in the human body. The initiative includes 2,900 scientists, from over 1,500 institutes and 94 countries around the world. So far, over 121 million individual cells have been analysed, and each cell has its gene activation profile recorded. Computational pipelines and methods have been developed to process the vast amounts of data.
The findings, freely available at https://www.heartcellatlas.org are already transforming our understanding of health and disease.
