MPNforum Magazine

CALR — the Tony Green perspective

The article that follows contains an interview with Professor Tony Green, a true giant in the field of molecular biology and the myeloproliferative neoplasms,

Professor Claire Harrison Guy’s and St. Thomas’

whose career has spanned   Oxford, Cambridge, Australia and most of the critical scientific insights into our field. This magnitude of success might call to mind an image of a ruthless highly ambitious or perhaps eccentric academic. Tony, whom I have had the privilege to work alongside for the past 13 years, is none of these. So when Zhen gave me the opportunity to introduce Tony I readily agreed.  

You can easily research Tony’s scientific pedigree and achievements but unless you know him it would be difficult to appreciate that he is a great family man, father of 3, and taking great joy in recently becoming grandfather to 2. He plays bridge, enjoys long walks and holidays with his wife Sarah. He also causes great excitement in my house as my teenager confuses him with Professor Green the popstar at the end of a telephone.

But it is the time, intellect, rigor and energy that he puts into his work that will have benefitted you the reader as someone touched by MPN. These benefits have been substantial. Here he describes the discovery of the CALR mutations; but he has also led the description of JAK2 V617F and JAK2 exon 12 mutations and broadened our understanding of their mechanisms, genotype/phenotype correlations of these as well as leading the PT-1 study which is the largest ever prospective study of ET patients with over 1200 enrolled to name but a few of his many achievements.

MPNforum:  Thank you for sitting down with us Professor Green.  Your recent paper in the New England Journal of Medicine has set off a wave of excitement in the MPN scientific and patient communities.  We would like to more fully introduce you to the American MPN patient and caregiver community and then clarify some CALR issues.

What drew you to hematopoiesis and MPNs as a field of specialization?

I trained as a hematologist initially  I did a  PhD in early days of molecular biology back in the 1980’s  Got interested in malignancy then when oncogenes were just being described for the first time. After, I did a post doc in Australia with Don Metcalf in Melbourne Australia   ( Walter and Eliza Hall Institute of Medical Research.   He was the father of hematopoiesis in many ways; describing and isolating haematopoietic cytokines and cytokine receptors.

Then I returned to this country as a consultant-level clinician scientist to set up my own group here in Cambridge and that was back in 1991. And I’ve been here ever since.   It’s a difficult place to leave actually.

So I was a Wellcome Trust senior fellow at that point.  The Wellcome Trust is sort of like the Howard Hughes organization.  I became professor of medicine here in 2000 and head of the department at the same time.  So I’ve been head of the university department of hematology here in Cambridge and chair of Haematology in Addenenbrookes hospital for the past 14 years.

Personal  interests  outside of hematology?

Three children, two grandchildren   granddaughters.  I guess I’ve always enjoyed things like walking, up mountains preferably, the  Himalayas, Kilimanjaro in the past…. I spent six weeks on the mountain,  part of the group making a film so we camped up on Kilimanjaro.  Nowadays  it’s more around Scotland and places in the UK and Europe. So walking, scuba diving but time with family and friends is  the most important

How did you get into CALR fishing expedition.  The amount of data necessary to crunch. Did you come across it through directed research or did it come about through raw data crunching?

Let me very briefly take you back a long way and then get up to date.  When I got back to this country in 1991 I decided to work in two areas. One was fairly basic science, gene regulation in blood stem cells, the other was myeloproliferative neoplasms.. I decided back then that the latter  was an important area to research and  biologically very interesting.

Did various things since 1991 along the way.  We pursued a number of approaches to understanding the pathogenesis of the MPNs . For example we mapped 20q deletions and very recently finally got to the bottom of the molecular mechanism. I established the MPN clinical serice in Addenbrookes and also the PT-1 suite of trials with Tom Pearson back in 1997 . Claire (Harrison) and Peter Campbell joined me after Tom (Pearson) retired. We reported the high risk study in NEJM in 2005  and we are just winding the Intermediate and low risk trials up.

In 2005 We were one of several groups that reported the JAK2V617F  mutation.  We then also discovered  the JAK2 Exon 12 mutations associated with PV (NEJM 2007),  we showed that was in the nucleus  and directly regulated  chromatin (Nature 2009) and we pursued a number of mechanisms by which JAK2 mutations affect the behaviour of stem and progenitor cells.

So by 2009-2010 we were learning a lot about what JAK2 did but we still had these bubbling questions.  Two questions (1) What’s going in patients who lack the JAK2 mutation and (2) another is  why do patients with the JAK2 mutation get one disease and other patients with the same mutation  get a rather different disease?

And because of those two questions I set out with  a colleague of mine, Peter Campbell who was actually a PhD student of mine but now is running the genomics side of the Sanger Center.   So Peter and I set out we decided we needed to properly look at the genomes of those  patients with myeloproliferative neoplasms.  So we got funding for this in 2009-10 and  we’ve been basically doing this for the past 4 years.

And it’s been quite difficult for a whole series of technical reasons that make the myeloproliferative neoplasms rather more challenging than some other tumors. But we got around those various technical challenges. And we decided we wanted to do it comprehensively and so we did. WE did 151 patient exomes and this is the first comprehensive  description of the genomic landscape of the MPNs (NEJM 2013)

We identified CALR as a gene that was mutated among thousands – it proved to be a very difficult mutation to find and very easily missed.

Why is it so difficult to find since it’s so omnipresent?

It’s very difficult to find for a whole series of technical reasons.   When you use next generation sequencing a lot depends on the coverage you get at a given locus. So let me give you an example. In the patients we looked at the coverage,  the number of sequencing reads you get over a given gene, on average  for JAK2 for example are 150 or something like that.  For each patient we would have on average 150 sequence reads across JAK 2   In CALR, the figure on average was 10.  So way down. So that’s the first problem.

Secondly,  CALR is very repetitive. And when you do next generation sequencing you rely on taking the reads you generate and the aligning them, mapping them, back to a reference genome.  And if you have repetitive DNA that becomes quite difficult for the computer to do accurately because it aligns them with various parts of the genome, so that was the second reason.

The third reason is the changes in CALR  gene are all due to insertions and deletions  Some are quite big, 50 or so base pairs. If your sequencing read is only 100 base pairs from  normal and you’re missing about 50 base pairs of  that, when you try and map that  back to the reference genome the computer just spits it so out and thinks it’s just rubbish.

Those are the sorts of things that caused people to miss this and I think we and that Vienna  group were lucky.

What is it about Exon 9 that distinguishes it so that’s the locus for the mutations in the gene?

I can only speculate…The mutations are very specific changes that result in a new end to the protein.   And the only way to get that is through changes in exon 9.

So let me jump to that C terminal.  The effect of the mutation is to change the ability of  the calreticulin protein  to take on calcium?

Well that’s one predicted effect, yes. I think we and other groups are trying to work out quite why this novel protein does what it does.  It’s too early yet but I would  speculate strongly it’s likely to have something to do with those calcium binding sites.

What’s the effect of that on the cells, the organic effect on the body?

I’m afraid I’m not going to be terribly useful to you here because these cell biology experiments,….where you look for function… take a long time and we simply don’t have any data yet. The Vienna group had some preliminary data suggesting this was influencing cytokine signaling which seems very plausible but is likely to be only part of the picture.

The Italian group, Rumi, et.al., conclude that PV and ET JAK2 positive are different phenotypes of the same disease while CALR-mutated ET is a distinct disease entity. Would you agree with that now?

It becomes  a semantics issue really.   I’ve been saying for some time,  in  fact ever since a NEJM review in 2006, ,  in which I floated the idea that these diseases were part of a  phenotypic continuum. we published the first data suggesting this in Lancet in 2005.  So that concept of phenotypic continuum  is now gaining ground. Some people didn’t like it to begin with but I think a lot of people are buying into it now and maybe that’s what Rumi is referring to…

 … but progression might be different?

Might be different but we don’t have hard data on that yet. There’s retrospective  data from several groups that there may be difference in progression  or transformation but there’s no prospective data yet.

Can you step through very briefly how you did this, It appears you looked at  3500 samples and did exome sequencing of 151 patients and many other <2000 other cancers.  How did that take place? What was the process?

It was the culmination of a lot of work over many years. We started 20 years ago  by setting up a myeloproliferative clinic here. We got patients in, we took regular samples, building a database, a large sample bank on all the patients here.  Half the battle was making sure that was all done well.

At Cambridge we’re really lucky. We’ve got three things. We’ve got a very strong research group working in experimental ways on the MPNs.  Secondly we got a very strong clinical base. We got the clinic which is in in the hospital next door and we’ve got the UK network  of MPN clinical trials and the PT-1 trials and very strong access to clinical samples. The third prong is the link to the Sanger Institute which is the largest genome center in Europe  and is just 4 miles down the road from us and my friend and colleague Peter Campbell runs the cancer genome d sequencing side of the Sanger   So that’s the three prongs that make us  quite unique in the world.

Do you have your own statistician team?

The Sanger Institute  has a large team running  the various  sequencing projects including a lot of bioinformatics support.  Peter runs that so our work on MPNs was one project going through the fairly industrial pipeline they have going at the Sanger.

Massive Parallel sequencing… What  does that mean, specifically?

Just a bit of a jargon term. Other people use Next generation sequencing. Some people use  High throughput sequencing.  They all mean the same thing. It’s just this a way of sequencing individual DNA molecules vs. the old Sanger method of doing it bulk and doing it chemically.

Can you speculate on the diagnostic and therapeutic ramifications of your discovery.

I can more than speculate on the diagnostic side because we’ve got this into our routine diagnostic service that supplies this region of the UK.  Now in my clinic when we have a patient coming in with a high platelet count we first of all do a JAK 2 mutation test and   if that’s negative we then to a CALR test and if that’s negative we do an MPL. That’s now our standard diagnostic workup.

Therapeutics, I have to speculate because that’s a long way away.. We have to know a lot more about the mechanism by which CALR is acting Because that may give us therapeutic  opportunities and new targets. One possibility is CALR itself could be a target because mutant CALR has a tumor specific sequence at the C terminus.   It will be many years before anything gets into patients that might arise out of this finding but I there will be a lot of interest in exploring that.

How do you account for the lack of co-existence between the JAK 2 mutation and the CALR mutation?

I think they’re doing the same thing.  For example JAK2  increases cytokine signaling. I would not be surprised if CALR does the same although the  pathways  through which it does it would be different.  But the consequence would be increased cytokine signaling. That would be my guess.

 Why would they not be co-existent? Why could they not be active in the same pathway?

You don’t need them. If you’ve got one you already up-regulated cytokine signaling  then having another mutation that does that same would be superfluous. It wouldn’t add anything.  There’s no reason to believe it would give an additional advantage to the cell.

Would zygosity play a role. In case of a homozygous JAK2 mutation why could they not  both be active? 

I  agree its not clear why homozygosity for mutant JAK2 seems to alter the phenotype whereas a CALR mutation on top of a JAK2 mutation  is never seen. Perhaps another possible explanation is that a JAK2 mutation and a CALR mutation in the same cell would be toxic for the cell.

MPNforum:    That’s interesting.

Again, many thanks for your time and your many contributions to the MPN community. Hope to talk with you again soon.

Take me back to the Contents

© MPNforum.com, 2014. Unauthorized use and/or duplication of this material without express and written permission is prohibited. Excerpts and links may be used, provided that full and clear credit is given to MPNforum.com with appropriate and specific direction to the original content.