Order from chaos – making sense of bowel cancer’s scrambled DNA

Last year, researchers at our London Research Institute published what became – after the discovery of the Higgs boson – the second most-referenced science paper of 2012

Their study looked at how tumours ‘evolve’ during treatment, and showed that, genetically speaking, different parts of a patient’s kidney tumour were extremely diverse. No two regions they analysed were identical.

Although not the first study to demonstrate this diversity – known as ‘intratumoral heterogeneity’ –   this paper kick-started a wider discussion of its causes and implications. Understanding how diversity develops in a tumour is important – because this is how cancers develop resistance to chemotherapy, and ultimately what makes cancer such a killer.

Today, the same group has published new research in the journal Nature that starts to reveal diversity’s origins.

They’ve been studying a phenomenon called ‘chromosomal instability’ in bowel cancer cells, where cells’ DNA becomes more and more disordered as they grow and divide, causing ever-greater genetic chaos. Patients with more unstable tumours tend to do worse – so understanding how it develops is important.

In a series of meticulous and detailed experiments, the researchers have found compelling evidence that chromosomal instability is caused by the malfunctioning of a particular (and unexpected) step in cell division, and identified three genes involved.

This gives us a better understanding of instability’s causes, which hopefully will galvanise future work to exploit it, and ultimately to improve things for patients. Let’s look in detail at what they did.

Chromosomal chaos

Each cell in our bodies contains 46 chromosomes – long DNA molecules that bear our genetic material.

Before a cell divides in two, these need to be precisely and exactly copied, so that each ‘daughter’ cell has exactly the right number of chromosomes – a process called DNA replication.

Once the DNA has been copied, a second step, called chromosome segregation occurs – the chromosomes are pulled apart as the cell pinches in two (we’ve discussed this in detail, with videos).

But in rapidly dividing cancer cells, things aren’t nearly so neat and tidy.

As they repeatedly divide, the DNA in some types of cancer cell becomes extraordinarily messed-up. Instead of the 46 chromosomes, cells frequently end up with random numbers.

And as well as changes in number, chunks of one chromosome can be fused to another, or copied, or deleted.  This is known as ‘chromosomal instability’, or CIN for short. Certain cancer ‘cell lines’ (lab-grown cancer cells) have highly unstable chromosomes (referred to as ‘CIN⁺’) whereas others are much more stable (CIN^–).

Professor Charles Swanton’s team have been studying the difference between them, and trying to work out in detail where the instability stems from.

The team’s new study focused in particular on bowel cancer cells, and on why and how their chromosome number and structure changes over time.

Down the microscope

The lead researchers in Professor Swanton’s team – Dr Rebecca Burrell and Dr Sarah McClelland – began with a tried and tested technique: they looked down a microscope at hundreds of dividing CIN⁺ and CIN^– bowel cancer cells, to see if there were any obvious differences between them.

In particular, they were looking for tell-tale signs of different types of DNA error that would give clues as to the cause of the instability.

As the diagram below shows (click to enlarge), errors occurring during each of the two phases of cell division leave types of damage.  Errors during DNA copying lead to so-called ‘acentric’ and ‘bridged’ chromosomes, whereas errors during segregation lead to ‘lagging’ chromosomes.

When they studied the cells, they saw that those with unstable chromosomes often left chunks of DNA behind as they divided (see picture). A closer look at these ‘remainder’ chromosomes revealed that they were usually broken chunks of chromosomes, or two chromosomes stuck together – in other words, acentric and bridged chromosomes. ‘Lagging’ chromosomes were much less common.

In other words, this was clear-cut evidence that chromosomal instability was being driven – in large part – by errors occurring before the chromosomes segregated, rather than during division itself.

This was unexpected and surprising – researchers had long suspected that chromosomal instability was caused by defects during segregation.

Further experiments shed more light on instability’s cause: unstable cells were copying their DNA extremely slowly, suggesting that they were ‘stressed’ during DNA replication.

‘Replication stress’ is a well-known phenomenon to cell biologists. If we imagine DNA replication to be like the construction of a new building, then replication stress is like the builders running out of bricks, or starting building before the foundations are properly laid – the building is much more likely to collapse.

To check that replication stress led to chromosomal instability, the researchers used a chemical that interfered with copying DNA, managing to flip chromosomally stable cells into a state that looked just like their unstable counterparts.

Genetic faults

But was there order amidst the chaos? Was there a particular genetic defect -common to CIN⁺ cells and absent in CIN^– cells – that could help explain what they were seeing?

The team then analysed the DNA of CIN⁺ bowel cancer cell lines and – crucially – of similarly unstable tumour samples from cancer patients. When they compared this to DNA from CIN^– (stable) cancer cells, they spotted that a part of one particular chromosome (known as chromosome 18q) was usually missing in the unstable tumours.

Loss of this chromosome has been documented in many cancer samples over the years but no-one was really sure what its significance was.

To see whether loss of one or more of the many genes on 18q might cause chromosomal instability, they used a technique called ‘RNA interference’ to painstakingly switch off these genes in stable bowel cancer cells, one-by-one.

After many careful experiments, they showed that switching off each of three particular genes – PIGN, MEX3C and ZNF516 – led to errors in chromosome segregation, and chromosomal instability.

None of these genes has been previously linked to chromosome stability.

To confirm this link, the researchers showed that ‘knocking out’ each of these these genes, in bowel cancer cells with stable chromosomes, caused their chromosomes to become unstable – they had problems during DNA replication, leading to tell-tale mistakes during chromosome segregation.

Healthy supplements?

As a final check, the researchers went back to a study, published in 2011, that found that growing cells in the presence of the building blocks of DNA – chemicals called nucleosides – could help prevent problems in replication – presumably by making sure there were enough ‘bricks’ at the ‘construction site’.

To prove that replication problems were causing the segregation mistakes, the researchers grew chromosomally unstable cells in the presence of extra nucleosides.

Excitingly, Burrell and the team found that these extra nucleosides dramatically slowed the appearance of chromosome errors in the dividing cells.

Putting it all together

This is an important study. Studying chromosomal instability is only possible in cells in a lab, because it’s such dynamic process. But by bringing together a variety of techniques – the analysis of tumour samples from patients, painstaking years of microscope work, the careful tracking of the speed of DNA replication in different cell lines – our researchers have identified three genes that seem to cause chromosomal instability when they’re deleted.

The fact that loss of chromosome 18q is a common occurrence in many types of cancer suggests that this could be a crucial mechanism at work in the cancer genome. And this in turn helps explain why and how tumours can become so genetically diverse, and therefore so adept at evading treatments.

This work has four key implications:

First, it shows that there’s order underpinning cancer’s chaos. If we can understand the forces at work inside cancer’s DNA, we might be able to develop strategies to stop the disease in its tracks.

Second, the fact that giving cancer cells extra nucleosides seems to reduce DNA instability suggests we could try to restrain chromosomal instability as tumours grow – although a lot more work is needed to test this possibility, never mind work out if it’s a sensible way to help patients.

Third, it increases what we know about how bowel cancer develops. Loss of 18q has been linked to the change from what’s known as an adenoma, or polyp, to a fully-fledged cancer (or carcinoma). This helps explain what’s going on at a molecular level, and also explains why patients with ‘unstable’ bowel tumours often do worse than others.

And finally, it further highlights the need to spot bowel cancers early, before events like 18q loss have occurred. The NHS invites people from the age of 60 to take part in bowel screening (or from age 50 if you live in Scotland) – screening that’s been shown to help spot the disease at an early stage and prevent premature bowel cancer deaths.

If you needed an incentive to take up your invitation (and currently, less than 60 per cent of people invited do so), you need look no further than the work of these laboratory scientists in a corner of Lincoln’s Inn Fields, and their work on chromosomal instability in cancer cells.

Henry

• Huge thanks to Dr Rebecca Burrell for her help and support in drafting this post and advising on the diagram. 

Reference

• Burrell R.A., McClelland S.E., Endesfelder D., Groth P., Weller M.C., Shaikh N., Domingo E., Kanu N., Dewhurst S.M. & Gronroos E. & (2013). Replication stress links structural and numerical cancer chromosomal instability, Nature, 494 (7438) 492-496. DOI: 10.1038/nature11935