Dear readers, today I’m happy to share my Q&A with Madeline Lancaster on organoids. Dr. Lancaster is a Group Leader at the MRC Laboratory of Molecular Biology, University of Cambridge (UK). She became well known as brain organoids expert after publication of the study from Jürgen Knoblich’s lab 2 years ago.
Dear Dr. Lancaster, for introduction, can you please tell us how did you get to stem cell research and why brain organoids?
Actually, it was something of an accident, which I suppose is how much of science starts out. During my PhD in the lab of Joseph Gleeson at UCSD, I saw a number of fellow researchers making use of 2D in vitro neural differentiation in the form of neural rosettes. I was very impressed with the capabilities of neural rosettes to more faithfully recapitulate early neurodevelopmental processes and so when I started my post-doc in the lab of Juergen Knoblich, we planned out a project making use of mouse neural rosettes. However, because I was so new to it, I hit a few stumbling blocks and had trouble getting the neuroepithelial cells to settle down in 2D and make nice rosettes. Instead, I ended up with 3D balls of neurepithelium that looked very interesting, so I developed them further by embedding in Matrigel. That was the birth of the brain organoid method essentially, and the natural step was then to develop it further with human stem cells.
Let’s talk about biology of tissue self-assembly. Is there any difference between cells self-assembly and self-organization? What are the major mechanisms behind of cells self-assembly into tissue? Do cells have some kind of “spatial memory” or ability to “sniff chemical signals” in suspension? How is cells self-assembly universal among different types of stem- and mature cells?
There is indeed a difference between self-assembly and self-organization, and the difference can be more subtle depending on the system, but generally it comes down to the energy needed. Self-assembly does not usually require added energy and happens spontaneously, as in the case of protein folding for example. However, higher intrinsic order that requires input, such as energy input, would be termed self-organisation. In the context of organoids, it is likely that both of these processes are at play. Although we have not specifically looked in organoids, other similar systems, such as reaggregates of embryonic tissues, have shown that cells can sort out according to differences in expression of cell surface adhesion proteins. This type of organisation is an example of self-assembly as it represents orderly arrangement of cells to a state of lowest energy simply due to their differential adhesion properties (the so-called Differential Adhesion Hypothesis). However, the generation of tissue architecture through a more concerted effort of stem cells to generate specific fates with different properties and their active positioning through processes such as spindle orientation, would, in my view, represent a process of self-organisation. This process is also at play in organoids, for example in the cerebral cortex where radial glial progenitors undergo oriented cell divisions to position more differentiated daughter cells basally.
Although organoids are in vitro and in suspension, within the tissue it is reasonable to assume that there are certain chemical signals, such as growth factor gradients, since the tissue is quite dense and would very likely maintain such a gradient despite being in suspension. However, this remains to be determined and I think organoids would provide a very interesting system in which to study the contribution of such growth factor gradients, versus more stochastic cell sorting out events, to tissue morphogenesis.
I wonder how suspension of cells should be “treated” to trigger and maintain self-assembly? Is addition of extracellular matrix (ECM) sufficient or something else should be done?
It really depends on the organoid in question. It seems that some tissues require the addition of ECM in order to promote organisation, while others do not. However, I don’t think the ECM itself triggers self-assembly, as much of the early organisation that occurs in organoids, for example during the embryoid body phase, occurs before the addition of ECM. I think ECM is required later to support and trigger certain morphogenetic events, particularly epithelial budding in gut and in brain organoids. Timing really seems to be key in all steps though. It seems that if we can simply provide the right environment at the right time, the tissue is capable of intrinsic organisation and self-assembly. And my guess is that this is actually not far off from what is happening in an embryo. If you think about it, an embryo also develops due to intrinsic organisational cues.
There are many articles in the media about success and advances in generation of organoids from stem cells, but I’d like to focus on technical challenges. Can you please talk about methodological issues in organoid techniques in general and in brain organoids in particular?
In general, organoids develop quite stochastically since there is a lack of external body axes and other organisational cues to help pattern the tissue. Therefore, this leads to quite significant variability between organoids and between preparations of organoids. I think this applies to essentially all current organoid methodologies and certainly to a great extent to brain organoids. However, when we focus on specific brain regions, then you find a stereotypic organisation and developmental program that is quite reliable from organoid to organoid. Nonetheless, I think any technical readout making use of organoids, requires a large n in order to be confident in the findings.
What is a progress in generation of bigger vascularized organoids? How long organoids could be maitain in culture with adequate nutrients supply and vasculature?
To my knowledge, no one has successfully created fully in vitro vascularised organoids. This is a huge hurdle and one that the field of tissue engineering has been working on for many years. I am not an expert in engineering approaches to generate such in vitro vascularised constructs, but my feeling is that this hurdle will require a coming together of these fields to successful achieve in vitro vascularised organoids. Without vasculature, of course organoids are limited in their development. This can be overcome by, for example, breaking apart the organoids and “splitting” them, as in the case of gut organoids. Alternatively, organoids can be ectopically transplanted into a host organism for vascularisation and blood supply, as has been done for liver buds, for example. But in general, the full complexity of an organ cannot be recreated in a dish without achieving vascularisation and so I think this will be a major focus of future methods.
Is it possible to create desired brain region by changing spatial configuration/ polarity or chemical signals? Would it be possible to make from one “common” organoid or source cells/ organoids should be pre-selected from the mix?
Generating specific individual brain regions is certainly possible and has been pioneered by the late Yoshiki Sasai, who successfully generated the first brain regions in 3D. These included cortex, cerebellum, pituitary, and perhaps most well known are optic cup organoids. These different brain regions are possible to generate in isolation by providing certain combinations of growth factors at specific time points. For example, the addition of FGF19 gives rise to cerebellar tissue in 3D culture.
What is your take on potential ethical issues, related to creation of “mature” complex (with connectivity) and sized human brain organoids?
I don’t claim to be an expert in bioethics, but there is a solid foundation of research into network formation in model organisms that indicates that neural networks cannot reach maturity without both input, such as sensory input, and output. Thus, without both sensory input and output, such as the ability to interact with the surroundings, organoids will not form mature “thinking” networks. I think we are a very long way off from being able to create such a construct for organoids that would allow both input and output, so my feeling is that this will not be an issue in the immediate future. However, it may be possible in the more distant future and then it will be necessary to revisit the issue at that time.
We are all curious, of course, about possibility of translation organoids into medical applications. Can you please talk about such “translational landmarks” as efficiency, reproducibility and scalability of organoids? One particular question is about Matrigel – is there any alternative industry-grade defined ECM for organoids?
Organoids from a variety of organs have enormous potential for therapeutic applications. However, the reproducibility is still perhaps the largest hurdle. We can attempt to overcome this simply by using a large n, and this seems to work quite well for detecting phenotypes. Therefore, I hope it would also be successful in detecting drug effects such as in drug screening assays, but that remains to be shown. With regards to Matrigel, there are alternatives, but since it’s not completely clear what Matrigel is entirely made of, it has been difficult to generate a wholly synthetic alternative. I imagine that once animal studies show therapeutic potential of transplanted organoids, there will very likely be a number of scientists working on this problem.
Speaking of using organoids for Pharma R&D, what do you think is advantage of organoids over other sources of 3D tissues, such as bioprinted, defined iPS/ES-derived lines, and organ-on-a-chip? Our 2014 poll showed that organoids are not yet ready for prime time (see – http://stemcellassays.com/2014/04/cells-on-drugs-results/).
In general, the organoid field is in need of some proof of principle work showing that organoids, of any sort, can provide a platform for therapeutic applications such as drug discovery. My guess is that a number of companies are beginning to dip their toes into the organoid field, but once one takes the plunge and shows that a drug has actually been successfully developed with the help of organoids, then I think the pharmaceutical industry as a whole will jump on this technology.
What are the perspectives for using of organoids for cell- or organ replacement therapy? What indications and organoid types, in your opinion, are more promising and closer to clinical trials?
Gut organoids are certainly the closest to cell therapy approaches, since intestinal organoids have already been shown to successfully engraft when transplanted into the mouse intestine. I could imagine the most immediate need would be for liver, kidney and retinal organoids. Brain organoids are not as likely to be useful for whole organ transplants for obvious reasons, but they may provide a good way for directed differentiation of particular neural types that could then be transplanted. For example, spinal cord might be useful for victims of spinal cord injury.
You just started your own lab. I believe it’s very exciting time for you! Any research plans would you share with our readers? What advice would you give to young folks who are entering the field and are trying to make right career decision?
Yes, it’s truly an amazing time in my career. I think I’ve never had so much fun than I am right now with setting up the lab! It is a bit hectic though and I feel like I never have enough time or hands to do everything I would like to do. But that’s alright and I think probably one piece of advice I would give to another young soon-to-be group leader would be to not try to do too much and to instead focus on one or two key questions. For me, that’s brain size regulation. My lab is interested in using organoids to study developmental processes that regulate brain size determination, both in evolutionary terms but also in relation to neurodevelopmental disorders that affect brain size. I hope that by going down this path, we can learn something about what makes our brains unique and what goes wrong in the case of certain neurodevelopmental disorders like microcephaly.
Dr. Lancaster, I’d like to thank you for your time and for opportunity to talk! Good luck with your new lab!