THE first signs appear in infancy. Parents may notice that their child has problems focusing. As they grow up, more clues appear. The child may be happy and always smiling, but their movements are jerky, they struggle to talk and learning difficulties become apparent.
This is Angelman syndrome. Some symptoms resemble autism, but Angelman is due to a specific mutation shutting down a key gene in the brain. The thing is, there’s nothing wrong with the gene itself – it’s just turned off. If we could switch it on again in children with Angelman syndrome, they might grow up with far fewer problems.
Advertisement
This February, biologist reported that his team had managed to do exactly this in mice with the Angelman mutation. This is not just exciting news for those with the syndrome – it’s exciting for us all. Many diseases are caused by cells turning genes on when they should be off, or off when they should be on. If this approach works in people – Segal is in very early discussions about human trials – it could be used to treat everything from schizophrenia to obesity and drug addiction.
In the past couple of years, gene-editing has been making headlines, thanks to a revolutionary tool called CRISPR that has made it cheap and easy to alter the DNA inside cells. But Segal’s team at the University of California, Davis, is one of many that is adapting gene-editing tools so they can be used to control genes rather than change them. Put another way, they are using it to edit the epigenome rather than the genome.
It is early days, but the field is advancing at a breathtaking pace. Epigenome editing might turn out to be even more important and revolutionary than gene editing. “Ultimately the epigenetic approach may be better for therapy,” says Segal, “but this all remains to be seen.”
The basics first. The genome is basically a set of recipes – genes – for making proteins. The vast differences between, say, a muscle cell and a neuron are all down to which proteins get made when. So controlling the activity of genes is absolutely crucial, and cells do it in several ways.
The main one is by proteins called transcription factors that bind to specific DNA sequences and turn nearby genes on or off. Another, longer-lasting way cells can control the activity of genes is by adding chemical marks to DNA or the proteins that package it up. These “epigenetic” marks work by blocking the binding of transcription factors, for instance. The pattern of epigenetic marks on a cell’s DNA are referred to as its epigenome.
In recent years, our understanding of the role of the epigenome in health and disease has deepened. “About 10 or 15 years ago we thought it was all about the genome. Now we realise that a lot of it is about the epigenome,” says of Duke University in Durham, North Carolina. But until recently there was no effective way to tinker with the epigenome, he says. “We couldn’t understand what it was doing, or how we could reprogram it to treat disease. CRISPR epigenome-editing is transforming our ability to do this.”
“Until now we didn’t have the tools to reprogram the epigenome to treat disease“
The trick is to turn gene-editing tools like CRISPR into epigenome-editing tools. This is possible because all gene-editing tools have two main parts: a guide part that targets a specific spot in the genome, and a scissors part that cuts DNA at this site. Remove the scissors and replace them with other gadgets, and you can do all kinds of tricks. It’s like one of those kitchen tools that can be turned into a mixer or mincer or juicer by swapping attachments – although it can take a good deal of tinkering to get the attachments to work.
Two main things can be done this way. The first is to create artificial transcription factors that turn on or off specific genes or sets of genes.
In 2013, we discovered that the Cas9 protein, the key part of the CRISPR machinery, could be tweaked so it instead of editing them. Cas9 can be turned into a basic transcription factor simply by blunting the blades of the “scissors” part of the machinery. This “dead Cas9” binds to DNA and stops other transcription factors binding.
Replacing the scissors with other attachments can make artificial transcription factors even more powerful – capable of turning genes on as well as off, for instance.
The second thing that can be done is to swap the scissors for enzymes that add or remove epigenetic marks.
The advantage of changing epigenetic marks is that the effect of a single treatment could last for years, perhaps even for life. Artificial transcription factors, by contrast, will often only work for the days or weeks they remain in a cell (although for some purposes this may be desirable).
It is not just about CRISPR. Earlier gene-editing tools, such as the DNA-binding proteins known as zinc fingers, were first adapted for epigenome editing more than a decade ago. Segal used a zinc finger protein modified to act as an artificial transcription factor in the mouse study.
“Our grandparents’ lifestyle choices could have lasting effects on our health“
But progress has been slow with zinc fingers, because you have to design and test a new protein for each DNA sequence you want to target, a costly and laborious process. With CRISPR, the protein stays the same: to change the target all you have to do is add a different bit of guide RNA. That’s easy, quick and cheap. “It’s a revolution, there is no doubt about this,” says epigeneticist of the University of Stuttgart in Germany, author of .
CRISPR is not the only new kid on the block. There is much excitement about another method, , that could turn out to be even simpler to use.
All these new tools are making it easier to tackle the many unanswered questions about epigenetics. For instance, there has long been debate about cause and effect; some biologists have suggested that genes have epigenetic marks added because they are already inactive, rather than it being the addition of marks that inactivates them. Now we can test this.
Gersbach is one of the pioneers. Last year, his team developed versions of Cas9 that could remove acetyl marks – a type of epigenetic tag – from the proteins around which DNA wraps. They showed this was enough to in cells growing in dishes. They also showed that adding acetyl marks to these histone proteins .
Other teams are working on CRISPR-based tools to add or remove another type of epigenetic mark called a methyl group, on histones and DNA, and it seems that changing methyl marks also alters gene activity. In April, for instance, a team in Japan used CRISPR to cut out DNA sequences in embryonic kidney cells that had been tagged with methyl marks and replace them with non-methylated sequences. This alone was enough to .
Such studies show that in at least some cases, editing the epigenome does alter gene activity. Already, biologists are starting to think of the medical applications. “There is very great potential,” says Jörg Tost of France’s CEA Genomics Institute, who the prospects of epigenome engineering.
For a start, we could turn one kind of cell into another, to generate cells for treating diseases. Stem cell biologists have developed various methods for controlling a cell’s destiny, such as using growth factors, but epigenome editing should be even more powerful and precise. Gersbach has already used it to turn fibroblasts – the cells that make collagen – into neurons, and other teams are doing similar things.
Even more exciting is the possibility that we could use epigenome-editing tools to flick genes on or off in our bodies, as Segal did in mice. “When a gene is off when it is supposed to be on, or on when it is supposed to be off, you can simply go in and fix that problem and treat the disease in that way,” says Gersbach.
This could apply to a huge range of disorders, from cancer to neurodegenerative diseases. “In principle, it could cure many diseases rather than just treating the symptoms,” says Jeltsch.
Epigenome-editing tools injected into people’s bodies could be designed to enter only specific cell types, or to work only once they are inside certain cells, or both. They could be used to turn whole sets of genes on or off, not just individual ones. And it should even be possible to that change their behaviour depending on whether a particular cell is diseased or healthy.
How far away are such treatments? “I cannot give a timeline. But it’s not 30 years, it will be much earlier than that,” says of Imperial College London, who studies epigenetic changes in tumours. Tost shares this optimism; he expects the results of the first animal trials of CRISPR-based epigenome editing to appear within three or four years, and the first human trials to begin in as little as a decade.
But before then, there are some major obstacles to overcome, the biggest of which is delivering the epigenome-editing machinery to cells in the human body. “This is for sure the most critical bottleneck,” says Jeltsch.
The delivery issue is partly why treatments based on RNA interference (RNAi) – a way of turning off genes by triggering the destruction of RNA – have been slow to materialise. CRISPR should be far more powerful, not least because it can turn genes on as well as off. But the CRISPR machinery is larger and even harder to deliver.
There is reason to be optimistic; biologists have spent decades developing ways of delivering cargoes to cells, from packaging them in viruses to encapsulating them in fatty particles, some of which should work with CRISPR. And as with gene editing, some epigenome-editing treatments could involve removing, say, immune cells from the body, reprogramming them to attack cancer or halt an auto-immune attack, then implanting them back into the body. This would bypass the delivery issue.
“In principle, it could cure many diseases rather than just treating the symptoms“
Another big issue is safety. A few gene therapies have proven safe enough to get approval, and in principle epigenome editing should be even safer because it does not alter the sequence of DNA. But we’ll have to wait for the results of human trials to find out for sure. Until the risks become clearer, doctors will focus on serious diseases that so far have no effective treatments, such as ±áłÜ˛ÔłŮľ±˛Ô˛µłŮ´Ç˛Ô’s.
Even so, epigenetics is still a relatively young field, with a lot of basic questions that need to be answered. What determines how long epigenetic changes last? What proportion of cells in a tissue will need to be altered to treat various diseases? “We need to learn a lot about epigenetics, to work out which screw to turn to change things in a predictable fashion,” says Jeltsch.
But at last we have the tools we need to answer these questions. That by itself is a massive advance, and this is just the beginning. “Epigenome editing is in its infancy. It’s going to be very exciting to see what happens in the next two to five years,” says Segal. “And I hope I’m part of it.”
Epigenome editing for healthier kids?
Are you suffering because of the sins of your grandparents? There is now strong evidence that it’s not just genes that get passed on – lifestyle choices such as what your parents and grandparents ate or whether they smoked affects your risk of getting obesity, diabetes, cancer and other conditions. What’s less clear is how, but the mechanism added to DNA to control gene activity.
These epigenetic marks are supposed to be erased in embryos, but some biologists think a number of them may persist – or that they are re-established by RNA present in sperm.
If proved right, this means we inherit epigenetic changes that have lasting effects on our health. And if, for instance, they affect the architecture of the brain, it might be too late to do much about it by the time the effects become apparent. There is one solution, though: editing the epigenome of embryos to reduce our children’s chances of suffering from conditions like diabetes. And we will soon have all the tools needed to do that (see main story).
The question is whether it will be safe and desirable. The issue will be looked at as part of a US National Academy of Sciences investigation into genome editing of human embryos.
In principle, though, the case for editing the epigenome of embryos could be more compelling than that for editing the genome. It should be safer and less controversial because the child’s DNA sequence remains unaltered. And while virtually every genetic disease can be prevented by various forms of screening – meaning there is no need to edit genomes to prevent them being inherited – there is as yet no way to screen for epigenetic abnormalities. But it may be possible to identify parents-to-be whose lifestyle choices place their potential children at risk.
One potential issue would be ensuring that any epigenomic edits are not promptly erased. Another is how many edits will be required. “Even if epigenetic changes are shown to be involved, there may be thousands of them, too many to change,” says Romain Barrès of the University of Copenhagen in Denmark, who has found that the sperm of obese men contain thousands of epigenetic changes.
Still, it is possible to make multiple edits with CRISPR, points out Jerome Jullien of the University of Cambridge, who showed earlier this year that in frogs, epigenetic abnormalities in sperm can lead to developmental problems. “I agree that the more sites to modify the more difficult. But the number of sites is still unknown.”
This article appeared in print under the headline “Fine-tuning the genome”
