Fiona Watt talks with StemBook about the epidermis, the Human Induced Pluripotent Stem Cell Initiative, and women in science

Discussion |

Fiona Watt talks with StemBook about the epidermis, the Human Induced Pluripotent Stem Cell Initiative, and women in science

Lisa Girard


Fiona Watt, DPhil, is Director of the Centre for Stem Cells and Regenerative Medicine at the King's College London Centre for Stem Cells and Regenerative Medicine. Additionally, she is one of the founding members of the Human Induced Pluripotent Stem Cell Initiative, and Deputy Editor of eLife. Her lab is known for its work on mammalian skin cells and uses epidermal stem cells as a model to understand stem cell-niche interactions in cell fate decisions. Fiona spoke with StemBook editor, Lisa Girard, recently and below is an edited transcript of that conversation.


Fiona thank you so much for talking to StemBook, I really appreciate your time. One of the first things I wanted to let StemBook readers know a little bit about is what kind of projects are currently going on in your lab?

 My lab has a longstanding interest in the epidermis and how stem cells maintain it in adult life. We approach this from two different angles. One involves studies with cultured human epidermal cells and the other is making and analyzing mouse models. The main interest we have with cultured human cells is looking at how they interact with specific components of the stem cell niche and measuring their responses in terms of differentiation or proliferation, which is a really stripped down approach to the epidermis. Our work on mouse models has made us particularly interested in how changes in the epidermis cause responses in the underlying dermis. We have a big project on at the moment where we are trying to do lineage tracing and fate mapping in fibroblasts. I call this our “descent into the dermis” in a rather dismissive way, but it is actually humbling in that there are phenotypes that we have had in the lab for years without paying attention to what’s happening in the dermis, so that’s quite fun. But it’s very different from a reductionist approach where you are looking at single human epidermal stem cells in very specific environments.


Right, right. It’s really more in context.

 Yes. One thing I always do and really enjoy is going back to what the in vivo human situation is like. I have really great colleagues at King’s College that are fantastic dermatologists. It is quite easy for us to say, “We have a nice phenotype, we don’t understand it, what do you think this might mean in a human context?” So that anchors both the mouse modeling and the in vitro human work to what’s happening in human skin and I really enjoy that.


 And I think you’ve also found some important signals from the extracellular matrix that validates your in vivo models?

 Yes, the extracellular matrix is actually one of my first loves because, quite simply, if you deprive epidermal cells of contact with the extracellular matrix they differentiate. It is such a simple and clean experiment it’s been great fun exploring it over the years in different ways. It is a theme that runs through our work and at the moment we are looking at very high resolution at how engagement of extracellular matrix receptors affects the decision to differentiate or not.


 Complementary to looking at these signals, you have also done some work using single cell profiling studying the heterogeneity in the undifferentiated epidermal cells. I was wondering if you could talk a little bit about how this has informed how you think about stem cell fate?

 In 2006, we published a paper in which we did single cell gene expression profiling because we were looking for new stem cell markers-and that was very successful. But in the back of my mind, as the years went by, was that when we had looked at markers that were co-expressed on a population basis they clearly weren’t always co-expressed at a single cell level. In that first paper, we just didn’t have the sensitivity to pick up differences in gene expression that were less than about seven-fold. So, when we had the opportunity, with advancing technology, to use a more sensitive approach we went back and profiled a certain number of cells then looked at whether or not the gene expression profiles clustered in any way. We obtained evidence from that analysis to suggest that, in fact, underlying what seemed like a uniform population we could distinguish two distinct populations of cells and it would appear that the differences between them are not stochastic.

 I was rather disappointed by the way this work was reviewed because what the reviewers were after was another new stem cell marker. We were able to deliver that, but to me, the real interest is in trying to understand the relationship between the two populations of cells. It would appear to me that they probably don’t flip from one state to the other because these two types of cells are predicted to have very different relationships with their niche. At the moment, the technical challenge is to marry the transcriptional analysis with analysis of protein levels. The next stage is to capture a cell, label it with an antibody to a protein, and then have a look at the genes that are expressed. I think that’s really going to help us understand what the heterogeneity means. So far, we have completely failed to map the heterogeneity we see in our cultured cells to what’s happening in human skin but that’s partly a technical challenge. There is much more to do and maybe we just have to be patient. We waited a few years for the sensitivity of the technology to increase and then we went back to the problem and now, with equipment that is coming on the market to look at single cells in different ways, we’ll be able to gain new insights into the functional significance of the heterogeneity we have reported. I always say that if an experiment is really good you should revisit it every few years to get more out of it. I’d really like to understand what the relationship is between those two populations of cells we have defined and what the relevance is to cells in situ in the skin rather than isolated in a dish.


 Distinct from what is happening in your lab, I know you have been very involved with the Human Induced Pluripotent Stem Cell Initiative. Could you tell us about that?

 That has been just wonderful in the last year or so. From our work with cancer cells I became a little frustrated that people spend a lot of money doing whole genome sequencing of tumors when it is obvious to someone with my background that there is a lot of heterogeneity in every tumor. It got me thinking, with our stripped down environments, maybe you could take cells from different tumours and discover from the way they behave in vitro what the underlying combinations of mutations are that make them abnormal. I started talking to colleagues in different fields and I realized that even though we had very different backgrounds we all had the same interest in whether you could develop cell based assays that would allow you to get a good phenotypic readout of significant genetic variation in healthy individuals. So the birth of this project we call HipSci was really a bunch of us, including Sydney Brenner, getting together at the Wellcome Trust and realizing that although we all speak very different languages when we make the effort to communicate we can come up with a great collaborative project. In Hipsci we are generating and characterising a large collection of human iPS cells to discover how genomic variation impacts on cellular phenotype and identify new disease mechanisms.


Sounds like a lot of really good synergy.

 Yes, it is. We now have a strategic award that is jointly funded by the Wellcome Trust and the Medical Research Council. My co-leader on the grant is a geneticist who has been behind several massive sequencing projects, Richard Durbin.


 Are there particular diseases that you are interested in or are you just looking at the cohorts available?

 The first set of samples we will examine are from cilliopathies. What I love about cilliopathies is that although the genes responsible are known it’s clear that they have really pleiotropic effects on cells behavior. They affect Hedgehog signaling, Wnt signaling, cell shape and the cytoskeleton, so they are absolutely perfect for phenotypic analysis.

 We are also putting out a call for researchers around the country to propose other informative disease cohorts for analysis via HipSci. We are interested in both monogenic disorders and those with a more complex, or unknown, genetic component.

 Of course it’s always essential to have appropriate ethical consent for collecting samples from healthy and diseased individuals. One of the features of HipSci is that the cells and the accompanying genomic information will be widely available and so we have to make sure there is appropriate protection for the individual donors. Researchers all over the world are concerned about getting this right and we’ve been discussing the issues with them.


 That’s really exciting. In what timeframe do you anticipate the initiative being able to provide data?

 Well, we are getting it already. We have the first cell lines coming through the pipeline. Our lab has got the high content imaging and assays set up. The genomic data is coming through as well. I tend to think of projects as being linear, but this is one where the different partners around the country are doing the different bits of work necessary to stitch it all together at the end into something that we hope will be quite beautiful.


One thing I have admired for many years is your advocacy for women in science, and I know that there has been a lot of work done describing the issues facing women in science. So I was wondering if you could go one step further and describe some actionable items that institutions have taken to recruit and retain talented scientists that are women?

 Yes, I can mainly just talk about the UK situation. When I was starting out, I was the only female graduate student in our department. I think equal numbers of women are now doing PhDs and being postdocs. When it comes to being a group leader, there is a falling off in the number of women-which I think is understandable. The issue is whether enough is being done to support those who do want to carry on. Fortunately there are now mechanisms for mentoring, providing advice, and support.

 Importantly, the UK is good in terms of legal protection and maternity leave. It is quite standard now for a woman to take six months of maternity leave and many take a year. Whether or not that is desirable for your career is for the individual to decide. When you are a postdoc or PhD student your career can be quite modular. So if you had a fantastic PhD, went off for five years to have babies, then went back and did a postdoc, it wouldn’t be difficult to find a lab to work in. There are issues around the cost of childcare, but those are not unique to scientists. In the UK there are a number of different childcare options, including workplace nurseries, which are subsidized. One thing that King’s College London has done, which I think is fantastic, is to put aside a pot of money for which women scientists can apply on the basis merit and need – this will pay for their childcare when they first come back to work.

 If after your postdoctoral training you decide you want to run a lab but you have been out of the workplace for more than two years, there are fellowships that you can apply for to help you get back into the workforce. So, taking a career break is not an insurmountable problem.

 I think that senior women scientists still have a tough time in some UK academic environments. Work place bullying and harassment are illegal in the UK but are still a sad fact of life for some women researchers and one that remains hidden because women are frightened of the consequences of making a complaint. The problem is definitely not confined to junior scientists, as I know from personal experience.

 One of the things that can alleviate the situation is to aim for 20% representation of women on committees, conferences, and funding bodies. It’s not a big ask and in my experience the dynamics of decision making are a lot better when there is more than one woman in the room. I think we should look at financial penalties for failing to achieve different proportions of women at different stages. At the minimum you should report how many women applied for this job, how many women were shortlisted, these kinds of things.

 But things are definitely changing for the better. One of the drivers in the UK is a scheme called the Athena SWAN awards, which recognize and celebrate good practice in recruiting, retaining and promoting women in the fields of Science, Technology, Engineering, Mathematics and Medicine. Institutions sign up to the Athena Swan Charter and can then apply for bronze, silver and gold awards. They have to achieve a number of well-defined goals to show that they have created a women-friendly environment. – for example, scheduling meetings during core working hours rather than 8am when many mothers (and fathers) are dropping their children at school.

 The UK’s National Institute for Health Research, a major funder of clinical research, has stated that it does not expect to short-list any future Biomedical Research Centres or Units unless the academic partner institution has achieved at least the Silver Award. Most of the people leading the Centers are men - but in some institutions applying for and Athena Swan award is seen as women’s work or a “paper exercise”.  This is very disappointing.


 What do you think about the extended tenure clock?

 I think that’s excellent and certainly available in the UK. Some of the most prestigious awards for junior scientists here have a cutoff for applying that is six years postdoctoral - but if you have a baby, the cutoff can be extended. If you have a grant and one of your staff goes on maternity leave the funders will give the lab head money to pay for somebody to cover for that post and will extend the period of the grant. So the group leader shouldn’t suffer from maternity leave and the person taking it shouldn’t suffer. The great thing about maternity leave is that pregnancy lasts nine months, plus or minus a couple of weeks, so you can plan for it. My male postdocs will often fall off their bicycles or get injured playing sports; you can’t predict when that is going to happen or when they are going to get better!

 I think that managing expectations is the key to a successful maternity leave..  If you are a student or postdoc it is also good to have the discussion up front with your PI about how the project is going to proceed in your absence. If a project is at an early stage it can just be mothballed but in my lab if somebody submits a paper before they go on maternity leave we have the discussion about what experiments are likely to be required for the revision, who will carry them out, and whether they will be a middle author or co-first. So the person on maternity leave isn’t faced with difficult decisions about revising a paper while they would rather focus on something else. So far that has worked out really fine, actually. And it means that sometimes a PhD student will get a couple more middle author papers than they wouldn’t have had otherwise so that’s good for them and their CV. The person on maternity has no worries because they know that their work is being published even though they are not in the lab doing the experiments. As a lab head I know that an important piece of work is not going to be unduly delayed because of maternity leave.

 After you have had a baby you can feel under pressure not to come back to work because your peers in the antenatal class aren’t going back to work, which makes you feel like a bad mother. You do need a lot of positive reinforcement when you come back to the lab – and I try to provide that for my staff. Having had three kids myself, I know what it’s like. At first it seems like a miracle that you have even managed to get your clothes on and find the lab, and you’ve not gotten any sleep.


Yes, yes, you’ve had a whole day already by the time you get to work!

 Exactly. So if you are not coming straight back into a massive publication race that is definitely a plus.  

 There are issues about when is the perfect time to have a baby, and it’s different for different people. I think it’s difficult to have a baby and then go straight on the job market. It’s a bit more straightforward once your lab is up and running. But really there is no right or wrong time.

 Fiona, thank you so much for your time.

 Sure, thank you.