An ‘accidental genomicist’ 

Over the years I have realised that I am a sort of ‘accidental genomicist’. Accidental because I never set out to do genomics research. I was just a geneticist trying to understand how yeast cells control their reproduction through a process called the cell cycle, but that turned out to require work which also contributed to genomics. 

Genomics approaches emerged from molecular genetics and the emphasis that work put on DNA. The molecular genetic revolution, embracing techniques such as the isolation and amplification of specific genes (cloning) and the manipulation of the DNA molecule in cells (gene editing), occurred a few years after I started research. 

So my scientific career was quickly involved in developing molecular genetics for fission yeast, and subsequently that led to genomics. The reason for this was that early work focused on single genes important for cell cycle control, but these were only part of a wider network of genes and that required genomic approaches to identify them and find out how they worked.

Mapping life’s machinery

It all began for my lab around 1980. For the previous seven years I had used classical genetics – which studies individual genes and the effects they have on living organisms – to identify around 30 cdc (cell division cycle) genes which were required for the cell cycle of the single celled micro-organism fission yeast. 

I had chosen yeast to work with because it is ideal for classical genetics, great for isolating mutants, and a eukaryote, so basically not very different from mammalian cells. This work allowed us to describe the underlying logic of how all these genes interacted to control the cell cycle, but we had no way of working out what they actually did within the cell. 

In the 1980s, Paul chose to work with yeast because it's ideal for classical genetics, great for isolating mutants and a eukaryote, so not too different from mammalian cells. Credit: Paul Nurse

That all changed with the new methodologies of gene cloning, combined with yeast transformation (the ability to introduce exogenous genetic material), and precise gene manipulation (gene editing). I developed these methods for fission yeast in partnership with my colleague David Beach at the University of Sussex, which allowed us to clone the genes and sequence their DNA. 

This led us to discover the cdc2 gene (which encodes a protein now known as CDK; cyclin dependent kinase), as well as other cell cycle controlling genes. 

In 1984 I moved to the Imperial Cancer Research Fund labs in London’s Lincoln’s Inn Fields – one of the Crick’s precursor institutes – where we used our ability to transform and manipulate genes to transfer them between organisms, allowing them to be studied in different genomes. 

This allowed us to show that functionally equivalent genes to cdc2 were present in a second form of yeast – budding yeast – and also, crucially, in human cells (work done by one of my post-docs, Melanie Lee). These approaches demonstrated that CDK was a part of a universal cell cycle control mechanism operative in all eukaryotes, and they also set the stage for us to start work on fission yeast’s genomics.

Sequencing the whole fission yeast genome seemed to me the best way to identify all the cdc-like genes and get a molecular understanding of how the cell cycle worked. 

In the 1990s, sequencing was a major task, quite different to today. It required a grant specific for that purpose, but when I applied for these resources from UK funders, they said fission yeast was not part of their overall genomic sequencing strategy. 

To get round this block I visited my friends Bart Barrell and Fred Sanger at the Sanger Institute near Cambridge. Bart was a very experienced DNA sequencer who had worked with Fred, a double Nobelist and pioneer of early sequencing methods. Bart had a Wellcome Trust grant to sequence budding yeast which was not being fully utilised, so, with Fred’s encouragement, we used part of that grant money to start sequencing the fission yeast genome. Progress was quite fast, as the fission yeast genome is small, and within months about half of the task had been completed. 

Fission yeast. Credit: Paul Nurse.

I applied to the Wellcome Trust for support to finish the sequence with Bart, but they got cross with us for sequencing fission yeast without their permission and would not allow a grant application. The ever-resourceful Bart suggested we apply to the EU and finish the sequence by co-ordinating about ten labs in continental Europe. 

This ‘cottage industry’ approach took much longer, but in the end we finished the fission yeast genome. It was only the third eukaryotic genome to be fully sequenced at high accuracy and consisted of about 5000 genes. The sequence revealed a number of interesting things. My favourite one was a comparison of the three eukaryotic genomes that had been sequenced, budding yeast, a nematode worm and fission yeast, with those from simpler (prokaryotic) bacterial genomes, that had already been sequenced. These comparisons revealed which genes were obviously specific to eukaryotes. Satisfyingly, this short gene list included cdc2!  

Cracking the cycle

But what I really wanted the genome sequence for was to delete all the genes one by one and so identify all the cell cycle genes required to reproduce a eukaryotic cell. Such a complete gene deletion collection would also be a valuable resource for investigating many other cellular activities. 

Credit: Paul Nurse

However, once again this was blocked by funders, this time because fission yeast was not part of the overall genomic gene deletion strategy. No UK or US research funder would support the work, so this time my lab set up a consortium with labs in South Korea funded by a Korean Biotech company. 

Genes were deleted in the South Korean labs and these strains were then shipped to my lab where Jacky Hayles (my first graduate student, who has only recently retired from the Crick) identified all the cdc genes in the genome by visually screening for mutants which were defective for cell cycle progression. 

This was a major task but was successful in identifying around 300 essential cell cycle genes and another 200 which influenced the cell cycle but were not essential. This collection also formed a valuable resource for many other cell and molecular biology investigations, and is still the most comprehensive genetic resource for studying the eukaryotic cell cycle and its control. 

Mission accomplished, but it had taken around 20 years, much of it in the ICRF and Cancer Research UK London Research Institute and subsequently at the Crick. It was very much a programme that embraced genomics although obviously was never a part of the strategic genomics endeavours. 

However, it has been very useful for understanding control of the eukaryotic cell cycle. The work on cell cycle control in the budding and fission yeasts laid the groundwork for understanding how cells in more complex organisms such as ourselves control their cell cycle. These insights continue to underpin biomedical advances demonstrating how simple modern organisms can help unlock universal biological principles.