Paul Kellam - Blocking the flu

Paul Kellamhas been investigating whether screening for variations in a gene called IFITM3 could form a test for flu risk.
09 December 2015

Interview with 

Paul Kellam, Wellcome Trust Sanger Institute


A small group of influenza viruses or virons


Kat - We're heading into the flu season, and people at risk such as pregnant women, the elderly and those with certain health conditions are being advised to get a flu jab to help them stay free of the virus this winter. But could doctors soon be adding genetic risk into the mix when deciding who might benefit most from a jab? Paul Kellam, from the Wellcome Trust Sanger Institute outside Cambridge, has been investigating whether screening for variations in a gene called IFITM3 could form the basis of just such a test.

Paul - A particular gene, and it's known as interferon inducible transmembrane 3 or IFITM3 for short, has a variant in it and that variant is associated with more severe influenza disease than if you don't have that gene variant.

Kat - Is that a straight one to one relationship? If you've got that gene variation, you are going to end up in the hospital when you catch flu?

Paul - So genetics really doesn't work like that. There are modifiers. What you're looking at is an increased risk so an increased relative risk of developing disease given your genetics. This gives you a four or so fold increase risk of developing disease. So, these aren't absolutes. These are modifiers. Not everybody that ends up with severe disease, whether it's influenza or other viruses, will have this variant. So, it's not a variant that is universally everywhere, causing everything that we see, but nonetheless, in the subset of people, this has a very clear association.

Kat - It certainly seems to be important as you say, but what's it actually doing. If genes are the instructions that make proteins in cells, what's going on at this cellular level?

Paul - So, the gene variant unfortunately, we do not know its mechanism of action and that's true of a lot of human genetics. What we do know, what this protein does and therefore, by implication, we can start to infer what the gene variant does. So, the protein is expressed in the cell. It decorates what are known as endosomes and these are internal vesicles that viruses use to get from the outside of the cell to the inside. So, the virus gets taken up at the cell surface, ends up in an endosome, the endosomes lower their pH as they get deeper and deeper into the cell, and that low pH causes the virus to trigger and to breakout.

Kat - So, they're becoming more acidic.

Paul - They're becoming more acidic, that's right. The endosomes acidify as they go deeper into the cell. What this protein seems to do is make it much more difficult for the virus to break through the endosome membrane. Whether it makes it more sticky, less fluid is not really clear, but they seem to create an extra hurdle, an extra barrier for the virus to get out of the endosome. This is enough to really, really change the disease course.

Kat - We've talked about flu and about swine flu. Is that the only kind of virus that this protein is working against?

Paul - Well, no and that's one of the interesting things about this gene and this class of genes. Because they are blocking common features of what viruses need to get into cells or use to get into cells, then they have a broader activity spectrum. So this particular gene will work against viruses such as dengue virus, other influenza viruses, yellow fever virus, Ebola virus, to name but a few. So, we're starting to look at broad spectrum antiviral molecules.

Kat - A lot of drugs that we have target molecules in the cell and molecules in pathogens and stop them from working, they block them. I'm thinking of some of the cancer drugs that block the molecules that tell cells to grow. But what you're trying to do here is boost the level of this protein. How can you do that then, because that's kind of hard?

Paul - It is kind of hard. Fortunately, biology helps us in this way. Proteins turn over in cells. some proteins turnover very quickly, some turnover very slowly, and have much longer half-lives.

Kat - So, that's kind of creating it, destroying it, creating it, destroying it?

Paul - That's right. Some recent work that has come from a group in the US has identified the mechanism for turning over this protein IFITM3. What they found is that there's a particular pathway in the cell that helps to destroy IFITM3. Now, we know the proteins that interact together then the theory is if you block that interaction specifically, you'll leave more IFITM3 on the endosomes and that will increase your antiviral effect.

Kat - And that sounds potentially like, "Wow! This is an antiviral drug that could work for lots of people against all kinds of viruses!" So, when are we going to get it?

Paul - Well, it's always a difficult question. This is very much at the early stages. This is understanding the basic biology and mechanism. And then you've got to start on that long and hard road of drug development. In the end, you've got to be able to find a molecule that really works, a small drug, and that that drug has a good therapeutic effect. That means the beneficial side is much, much bigger than the detrimental side of any toxicity associated with the drug. As yet, we don't have any evidence that you can do that very easily in these sort of systems.

Kat - Turning more to the genetics angle, are there any ways that we could use this genetic information that people with certain gene variants respond in different ways to the flu? Could that be useful as well?

Paul - Well, I think that's a very interesting way of thinking how you can get quick wins from this sort of information. One of the things that people can relate to very easily is how infectious diseases feel. People have gastroenteritis, they have respiratory tract infections. As you say, you've had flu. You know what it feels like. Therefore, if it's possible to understand the genetics that changes how you respond to a pathogen and identify people that are at higher risk of severe disease then you have a way of stratifying them if you like, for treatment. Now for something like influenza, that treatment is very straightforward. It's identify those individuals and encourage them to have a vaccination because the vaccine tends to protect against the strain of flu that's coming. So, that's a very easy way of thinking how genetics can go through actual information that people can really relate to. The other way of thinking about it is in severe influenza seasons when people are coming in to accident and emergency wards and the clinicians are faced with who is going to have flu but get better as opposed to those that are going to have flu and get worse and really need to worry about them as a clinician. At that point, stratifying based on genetics starts to give you that extra information that hopefully becomes clinically useful.

Kat - That was Paul Kellam from the Wellcome Trust Sanger Institute.


Add a comment