Society for Endocrinology - a world-leading authority on hormones


Issue 144 Summer 2022

Endocrinologist > Summer 2022 > Features


AN INTERVIEW WITH...CYNTHIA ANDONIADOU

| Features



Cynthia Andoniadou

Cynthia Andoniadou

Cynthia Andoniadou is Reader in Stem Cell Biology and Associate Dean for Postgraduate Research at King’s College London, where she established her research group in 2013. She is also the recipient of the Society for Endocrinology’s 2022 Starling Medal, which honours an emerging outstanding scientist whose work has contributed to exceptional advances in endocrinology. Here, she talks about her life in endocrinology to Kim Jonas.

Kim: Can you describe your central area of research?

Cynthia: Yes, my main interest is endocrine stem cells. We have been focusing on pituitary stem cells and recently expanding to the adrenal medulla. The central questions that underlie the research aim to understand what these cells do throughout life, and how do they do it. We want to know how they contribute to and regulate endocrine function. Are they there just as a supply of new cells, in case of the requirement for organ plasticity (for example if there is physiological challenge), or do they have additional roles?

So, we have started the work of identifying and characterising the stem cells in an organ, to try to understand their behaviour, particularly during homeostasis, physiological challenge and disease. In terms of their cell biology, the part that I’m most interested in is signalling: in particular, paracrine signalling emerging from the stem cells and influencing the other cells in the environment. The research that we have published in the last few years has revealed that stem cells secrete ligands into their environment, which modify the behaviour of other cells.

An example in the pituitary gland is WNT signalling, and how WNT ligands promote the proliferation of more committed progenitors. This is essential for growth of the gland; without this paracrine stem cell action, the gland doesn’t expand, which, in mice, leads to hypopituitarism.

There are multiple aspects of the contribution of stem cells that haven’t been explored, for example, during the adaptation of the gland or during disease. Our unpublished research is hinting that stem cells can influence more than just proliferation, so they might be involved in multiple aspects of pituitary gland function.

K: How did you begin to explore the role of stem cells through endocrine research? What led you to that path of stem cell discovery?

C: My research went full circle. I did my undergraduate degree in genetics and microbiology. Very soon into the course, I realised I hated microbiology, so I focused more on genetics. I did a PhD at the National Institute for Medical Research, London, in the Developmental Genetics Division, working on stem cells of the central nervous system and, specifically, the role of SOX2, which marks many epithelial stem cells.

My work was mostly based in vitro, so, for my postdoc, I wanted to go back into the organism and focus more on developmental biology. I joined Juan Pedro Martinez-Barbera’s lab at the UCL Institute of Child Health, London, and started my postdoc research on anterior forebrain development. It was a fantastic training opportunity for me. A large part of it involved manipulating genes and signalling pathways using genetically altered mouse models.

I was using a specific Cre-driver to activate the WNT signalling pathway in the forebrain, and the Cre recombinase was expressed not only in the forebrain but also the pituitary gland. I was disappointed when there was no phenotype in the forebrain. However, when we looked at the pituitary gland, we realised the mutation led to pituitary tumours. Looking at that phenotype in more detail revealed that there was involvement of stem cells, which got me back into studying stem cells, this time in the pituitary gland. So that was exciting for me, because I really do love stem cells.

K: Can you tell us about the multi-omics approaches that you’ve been using recently, why you’ve taken that direction and how it’s benefited your research?

Cynthia and her team, clockwise from top: Val Yianni, Thea Willis, Emily Lodge, Alice Santambrogio, Cynthia, Yasmine Kemkem (not present: Carlos Abascal Sherwell Sanchez).

Cynthia and her team, clockwise from top: Val Yianni, Thea Willis, Emily Lodge, Alice Santambrogio, Cynthia, Yasmine Kemkem (not present: Carlos Abascal Sherwell Sanchez).

C: We started off doing bulk RNA sequencing and purifying populations in the lab to compare stem cells with non-stem cells, and realised that this approach can give very misleading information. It is fine if you only have pure cell populations to compare, but all endocrine cells were present in one sample, and the resolution of the technique was not sufficient to draw certain conclusions.

When single cell technologies started emerging, it presented a fantastic opportunity to see what the relationships are between cells. Specifically, we wanted to study the communication between cells and to know exactly which genes stem cells express and, therefore, the proteins they potentially secrete, which cells might be perceiving as signals. This is all key to much of the research that we are doing.

We knew from immunofluorescent staining and cell culture experiments that not all stem cells behave in the same way or express identical markers. Therefore, we wanted to know what the degree of heterogeneity was amongst this population, and had the genetic tools to purify stem cells and characterise them better. We then started using single cell sequencing techniques to analyse stem cells of different genetically modified mutants, for example, ones with impaired secretion.

Our work has led to us teaming up with wonderful collaborators, Stuart Sealfon and Frédérique Ruf-Zamojski, from the Icahn School of Medicine at Mount Sinai (New York, USA), who are experts in multi-omics techniques. Together, we analysed the human pituitary stem cell compartment and found that this is very similar to that of the mouse, which our collaborators had characterised by multi-ome, bringing confidence in the relevance of much of our stem cell work in mice.

Analysing human samples from frozen, post-mortem, pituitaries differs from the single cell RNA-seq approaches that we previously used, as the method employed used isolated single nuclei. As well as RNA-seq, this allows us to perform ATAC-seq, a method used to determine to what degree chromatin is accessible across the genome. This provides a valuable insight into how cells are regulated, gaining the label ‘multi-ome’. During ATAC-seq, sequencing adapters are inserted into the accessible DNA regions, which can then undergo high throughput sequencing. ATAC-seq does not require prior knowledge of genomic elements (such as promoters and enhancers). It can act as a powerful tool to help determine normal mechanisms of gene regulation and identify how these might go wrong, especially when combined with RNA-seq assessing gene expression.

This allows us to interrogate and understand the networks that are controlling cell fate, and capture changes that might happen during ageing or with disease. As this technique can use frozen tissue, it opens up the possibility of using archived tissue samples. It was crucial to first start with a multi-ome reference of normal glands, at different ages and across sexes.

K: How do you think the developments in -omics will open up the fields for us as endocrinologists, with the prospects for future discoveries that they may bring?

C: It is important to mention that the amount of data generated from just a single piece of tissue is enormous. It’s enough to sustain multiple labs with multiple projects until the end of their careers!

At this point, I don’t foresee it being used for diagnosis (famous last words). It’s unlikely to replace current techniques that assess which cells are normal or what mutations might be present, mostly because of the cost and time it takes for data to be analysed. However, as discovery research, it can lead to a much deeper understanding of disease pathogenesis, and eventually may be of great value to personalised medicine.

At these early stages, we might just have to sequence or multi-ome everything, to gain global information on baseline cell states. As a next phase, disease cell states can help us identify and design new drug targets, and select appropriate treatments that target cells selectively (for example, only cells identified as ‘tumourigenic culprits’ in a tumour). Eventually, having extended multi-ome libraries of diseased tissues, with associated clinical data or outcomes (such as response to specific treatments), will be a worthwhile investment. It will be critical to retain the data in open access format for professionals.

K: With every new technique comes limitations. What are the considerations when using these new approaches?

C: Tissue integrity is crucial. This can be difficult if you are working with human samples that are collected via post-mortem or partially processed by pathology labs. With mouse samples, which we primarily work on, we can control the tissue handling and processing more.

I offer a word of caution on data processing. Anyone can pay and outsource the running of a multi-omics experiment if they have access to the samples, but doing the analysis is the particularly tricky part. You need to have access to trained computational biologists who understand the data and have the biological knowledge to ask the right questions in the right way. So, we need to have common language and understanding with computational biologists.

As a computationally challenged biologist, I can’t understand code that I am shown and can only advise how queries are approached, but I am lucky that my team are proficient in both! New pipelines for data analysis algorithms are emerging all the time, and something that we are lucky to have is a worldwide network of collaborators, who support one another during analysis, and exchange code if required. If we have a problem then we can put that forward, discuss it all together, and come up with solutions. Analysis is going to be a little bit slower until the necessary code is developed to extract everything we need, but, once this is all in place, the analyses will be far more user-friendly.

A final limitation is space. These datasets are enormous, and not all institutions have yet caught up with the server space required for the datasets generated by their investigators. This will remain a major consideration in the future, when there will hopefully be concerted efforts to pool data from multiple studies, and curate them all in a user-friendly format, accessible online.

K: I wanted to end by congratulating you for being awarded the 2022 Starling Medal!

C: Thank you! I am so grateful to the Society for Endocrinology and, especially, as a basic scientist, as I feel truly welcomed into the endocrine community. As a lab, and in the stem cell field, I think we are gaining momentum at present, which is a really nice place to be.

The Starling Medal represents recognition of the cumulative work of everybody who has shaped the research. There are many people who have contributed to this, including all the researchers who have been part of my lab, our clinical, translational and basic collaborators, and the scientists who trained me along the way.

When it comes to my lecture, I will try my best to include contributions from as many people as possible, since it feels strange to receive a medal bearing only your name, when a hundred names are also behind the work.

Nominate the next Starling Medal winner!




This Issue:

Summer 2022

Summer 2022

The Endocrinologist

...

Autumn 2022

Autumn 2022