Towards Context Aware Cell Line Development: Interview with Duygu Dikicioglu, UCL

Hello and welcome to this interview for Oxford Global.

Today, I'm delighted to be joined by Duygu Dikicioglu, an Associate Professor in Digital Bioprocessing and Biochemical Engineering at UCLs Department of Biochemical Engineering.

She'll be joining us at Cell UK 2025, and we're very interested to hear about her presentation, which will be titled ‘Variant Interpretation in HEK293, Leveraging Genomic Heterogeneity for Context Aware Cell Line Development.’

Very interested to get going with this interview, Duygu, but thank you for joining me today.

Thank you very much for giving me the opportunity to actually talk a bit more about this and hopefully gather some more interest for the event.

Fantastic.

So to start, could you just basically introduce yourself and your research focus here at UCL?

Yes, of course.

So thank you very much.

You've already done the major introductions.

So I work at UCLA Department of Biochemical Engineering, and my focus is on how we can make better use of digital tools in bioprocess engineering in a very brief snapshot.

So my group focuses a lot on the use of automation, modelling and different types of data management strategies, how we can use data analytics and statistics, particularly in the context of systems bioprocess engineering and systems biology.

So we developed tools and pipelines to address specific challenges around digitalisation and data that are highly relevant and particularly of concern to bioprocessing issues.

Fantastic, thank you.

And as we said at the top, your presentation is on the HEK293 cell line. Why this particular cell line and why is it so central to bioprocessing and biomedical research?

Yes, so I mean, of course, biomedical research and bioprocessing are kind of complementary, but at the same time they have different needs and different utility.

So the cell line that we will be focusing on is particularly attractive in both aspects.

So from a bioscience or perhaps a biomedical research point of view, these cell lines are derived from human tissue, and they are biologically relevant, much more relevant than say a CHO cell line, for example, if you want to study human specific cellular pathways, protein expression, post translational modifications.

Now if we move on to more on the bioprocessing angle of things, it's derivatives.

The derivatives of the HEK293 cell line are extensively used for producing adenoviral and lentiviral vectors.

So these are critical tools for selling gene therapies and vaccine development.

We have seen an example of it as we well known in the COVID-19 periods in the vaccine.

So the cell line itself has some advantages that actually make it possible for us.

It has an ease of relative ease of culturing and transfectability.

It is a versatile platform for genetic manipulation, and it allows us scalability for manufacturing purposes because they can survive in suspension cultures and in serum filled media, which makes it particularly interesting for our purposes for bioprocessing as well as biomedical research.

Of course, this comes with some challenges because despite this widespread use and attractiveness, they're genomically not as well characterised as we would hope them to be, which introduces a number of assumptions that and risks that have been carried out ever since they haven't started actively to be employed in research.

Great.

And I wanted to get more into that. So, you mentioned that there are some major gaps in how we understand the genomic structure of HEK293 and HEK293T.

What kinds of assumptions are typically made and why are they so problematic?

This is a very actually comprehensive question.

So I mean we will get much more in depth into it in the talk obviously.

But so one of the most widely used assumptions that we have is that, I mean, we inherently make that assumption that when we talk about this HEK293 cell line or HEK293T cell line, we actually subconsciously treat them as genetically uniform cell lines that remain stable over time and also between labs.

So when we make comparisons, when we interpret them, of course, this can be problematic at times.

Sometimes we assume that they are highly unemployed, and we actually are in our research show that it's not necessarily the case all the time.

There are some frequent chromosome rearrangements, deletions, mutations in their genome, and this varies from lab to lab.

So when we rely on information that emerges from another lab of course, or even when we monitor the same culture over the course of time.

This variability is a factor that we need to take into account more effectively.

Now when we think about the production cell lines HEK293T cell lines, they are a derivative of the original cell line that expresses an SV40 large T antigen.

However, these HEK293T derivatives are not only the difference from the original cell line with this expression.

This process of expression introduces additional genomic complexity that we often overlook.

We know that studies, previous studies have shown on from our own research.

We also know that there are additional genomic changes, insertions, rearranges that actually cause susceptibility, more susceptibility to DNA damage that emerges from this variation.

So all the other variants from the HEK293 cell lines, including HEK293T are a kind of at the risk of sometimes being treated as interchangeable.

This is also another common misconception, especially with regards to their transcriptomic, proteomic profiles.

And this creates of course assumptions that perhaps allow us to overlook certain key points in relation to their production capabilities.

Now one final thing that I would like to talk about in this domain is the fact that of course HEK293 cells are derived from embryonic kidneys, but they are nevertheless A derived cell line that underwent like extensive transformation and immortalization.

And this actually is the emergent factor from many of even oncogenic pathways being altered.

So they are far away from being like a generic human cell model.

Now we might easily think that this is only important for hardcore biological core research, but actually it has much further implications than we initially overthink about this.

Because when we analyse the omics data related to these cell lines, we typically refer to the human reference genome, which actually is the key point in our presentation, which is going to be the key point of our talk when hopefully when we meet at Cell UK conference this year.

And I will try to highlight some of our efforts to how we actually took some action to overcome some of these challenges and bypass these preconceptions so that we have novel techniques and novel standards to analyse these cell lines.

Thank you.

And those widely accepted perceptions about the genomes of these cells that you mentioned, are they reliant on the most up to date research or do we need to reconsider the data?

So of course it's a mix.

Some, not some of our conceptions or perceptions of the information that we have available at hand rely on very old data.

Some are of course emerging from more recent studies and now I don't want to necessarily say that everything that is all data is completely useless.

Of course, like the old, accepted views are very useful for us, especially as a starting point.

But in light of the new research, particularly if the new research is somehow emerging in a way to contradict or conflict some of the well-known preconceptions, then this might mean that in light of the new findings with there are there might be cases that we need to rethink about how we do this.

Now of course there are many assumptions, but the one that I will be particularly focusing on in my talk is particularly going to be related to how we assume the HEK293T cell lines to be extremely genomically unstable at times.

The extent of variability on copy numbers and ploidy the of these cells, the availability of structural variants and how this variation across labs have an impact on how we interpret the data, how we actually use this data to design more useful cell lines for ourselves.

So obviously we cannot completely bypass these, but there are certain steps that we are hoping to propose that will actually counteract some of these challenges, and understanding the genome itself and how to analyse the properly lies at the heart of this endeavour in a way.

And then how might this more context aware understanding of HEK293 variants help biotech or biomanufacturing companies in like practical terms?

Now, if you know how to understand, if you understand these cell lines and if you know that if you can translate that understanding into practical context where we can actually tailor the variants accordingly.

Of course, we have the power to have an impact on all stages of wire processing, starting from improving the transaction efficiency for example, all the way to differing the cell behaviour.

And actually if we are able to modify or modulate how we improve or modify cell behaviour, then this gives us a way to control batch to batch variability of production, maintaining a consistent product quality and reproducibility allows us to develop better control strategies even in GMP environments.

And in general, it will help us enhance our understanding of how we design the upstream bioprocess.

Because if you can understand the response of different variants to different factors that impact upstream process design, such as medium composition on oxygen availability, shear stress better than this will help us to streamline the process and get the most out of these cells that we can.

Now, all of these collectively help better regulatory compliance, obviously, because we can predict things better, we can actually document them better, and we can characterise even the product in a better way to understand its quality attributes and ensure that basically we end up with a functionally effective, safe and consistent product at the end.

Great, thank you.

What are the next steps for your team? Are you looking into expanding into different omics types, for example?

So in my talk, I will focus primarily on our efforts on how to analyse the genome of these cell lines.

But of course we know that it doesn't stop at the genome, and we have the manifestation of these genomic changes occurring at the epigenomic level, transcriptomic level and proteomic level primarily.

So we are currently in the process of moving forward with integrating different types of omics data sets with the information that we have at hand and get to get gather some better insight into how we can design these variants in a more effective way.

And for my final question, what do you hope that the industry takes away from your talk at Cell UK 2025 and where can people learn more about your research or get in touch to collaborate perhaps?

OK, so the second part is quite easy.

The first part is of course, I mean what I would like to relay is a message is probably that we need to understand that just as acquiring high quality experimental data, especially if it's high throughput data, it is equally important to have gold standards established for its analysis and interpretation.

And there are things in light of the recent advances or recent observations and challenges that we observe in what we do, there are advances that we can take this to the next step.

And we will be presenting one such effort that actually improves how we analyse and interpret the data.

So, and this might, I mean, this is one of the ways, many ways that can strip some of the well-established preconceptions that could potentially be misleading in terms of the how we design these cells to work more efficiently and better for us.

And so it can even be important for us to elucidate previously uncovered facts from existing legacy data sets.

So this is, I think what I would like to convey as a like a single message is that we need to maintain a very open horizon.

Because even with existing data, looking at it with a fresh pair of eyes, with a fresh pair of analytics can give us a lot more insight than we would originally expect from that very same data set.

And for further information on this, of course, I mean, my contact details will be, I'm supposing available in the documentation in the relation to the conference.

But I'd also say that like we would be in the process of publishing this soon.

So watch the space or anything that comes out.

And obviously whatever we do will be actually publicly made available.

So it will be a community effort.

Fantastic.

And once again you can see that presentation at Cell UK 2025.

And Duygu, thank you very much for your time and we look forward to seeing that presentation at the conference.

Thank you very much for inviting me and giving me the opportunity to discuss a bit more about my research.

Thank you.

Thanks.