In defence of Embryonic Stem Cells as a new model system for Developmental Biology

Over the last year a trend has emerged, highlighted by several articles, to defend model systems, specifically in Developmental Biology. The reason for this is a perceived (and real) threat in the funding of research in yeast, C. elegans, Drosophila, sea urchins and other systems e.g ascidians or arabidopsis (the issue of the animalcentric view of developmental biology is an interesting one) despite their research potential supported by many and obvious substantial contributions to our understanding of biological systems. One of the latest in this series of statement was published recently (1: http://dmm.biologists.org/content/10/12/1381). The authors do an excellent exposition of the contributions of several model systems to basic Biology and the impact that they have had on biomedical research, as a way to justify the need to continue the support of their research base. Surprisingly, the authors don’t include vertebrate models and Xenopus workers might query this omission. However, in addition to defend the aforementioned systems, they take issue with emerging research in Stem Cells, stating at the outset that “those who do not know the history of yeast, flies and other non-vertebrate organisms and their contributions to biomedicine may consider such studies as unworthy of participation or of funding. Given the growing excitement about human stem-cellbased cell culture systems, particularly organoids, some may argue that organisms that do not recapitulate all of the complexity of humans will cease to be useful in the near future. This line of thought is not only incorrect but could hamper scientific progress” (1). In a rather barbed comment towards the end of the essay, the authors state “We would go further to argue that organoids will not and should not replace non-vertebrate model organisms as discovery tools” (1) Here he authors refer directly to an article in the special issue of Development on organoids (2) in which it was argued exactly what is the theme of this post, that there is no way to separate the two fields and that putting them together promises much for the future.

As someone who is aware of that history and also works with Embryonic Stem Cells (ESCs) (DOI) I say that these statements are wrong and that to pitch research with Stem Cells and Organoids in contrast to that with model systems is not helpful. The issue is not about Stem Cells/Organoids versus model systems but about good and bad Science; and there is a lot of both in model systems and in Stem Cells, though I would say that from a strictly Science based perspective, right now, there is a lot of very good Science with model systems (DB) and a lot of bad science with Stem Cells at the moment. Issues of significance are a different matter. When making these arguments we should bear in mind that in comparison with model organisms, Stem Cells/Organoids are barely at the start of their historical life. In some ways Stem Cells, as an experimental system, are having the same disruptive effect that, say, Drosophila had when it erupted into a field dominated by marine invertebrates at the beginning of the XX century (3) and yet, history proofs that Drosophila, once it found its bearings, made very significant contributions and insights (3, 4). Difficult to compare historical periods but we should wait for time to make its mark in new fields of research. The discussion is not only about the quality of the Science that these areas of research can produce, it is also a matter of questions and systems  – will expand below. Notwithstanding these considerations, I want ti insist that it is not helpful to create confrontations, particularly when research in Stem Cells has always been an integral part of Developmental Biology and therefore, to try to present it as something new and competitive is to deny an essential part of the field and systems that the authors try to defend: Developmental Biology. Our understanding of cell lineages in many organisms e.g leeches, and systems e.g the nervous system in all metazoan, of gametogenesis, homeostasis and regeneration is founded on the biology of Stem Cells. We should celebrate and welcome this. It is true that over the last ten years the field of Stem Cells has undergone a renaissance (as did Drosophila in the 1980s after a hiatus of almost 40 years) and that some people want to present it as something new, but it does not work. Research with Stem Cells is Developmental Biology and what has changed is that in their renaissance, Stem Cells and, in particular Embryonic Stem Cells (ESCs), have become a new model system for Developmental Biology.

Nobody will deny that the model systems highlighted by the authors have played a key role in modern Biology. In particular they have been instrumental in revealing the elements needed for biological systems to build themselves and to function. The genetic and molecular screens of the 1980s and 90s uncovered a set of agents that, although we did not know at the time, turned out to be elements of a universal tool kit for the construction of living systems, and many have been shown to be central targets of disease (5, 6). This work has led us to appreciate the critical role of transcription in cell fate assignments, that of cell biology in morphogenesis and how the combination of the two contributes to the building of tissues and organs. In learning about the universality of many of these elements we are uncovering that they form conserved modules whose functioning we still do not understand; continuing work on these systems is making progress to understand these modules. This is the basis for much of the arguments as those presented here (1) and it is right. However, Science, even today is an issue of questions and systems i.e. one should not work on a system simply because it is experimentally sound, because it yields results, because many experiments can be done or because it has a glorious past. One should work on a system because it is suited to ask, and maybe answer, specific questions and one should match question and experimental system. In this regard, invertebrates have many features in common with vertebrates but there are features of vertebrates that do not figure in the Biology of invertebrates. For example, the large scale cellular rearrangements that characterize most vertebrate embryos do not find a correlate in Drosophila or C. elegans, not to mention issues of homeostasis which only arise in the life spans of vertebrates. The issue of Dlp and Hippo that the authors raise in their essay (1) echoes the actions of hormones and chelons, but it is clear that the time and length scales characteristic of vertebrates require studying them in vertebrates. Interestingly, the issue of the regulation and function of Hippo –very much a flavor of the times- might not be the best example since it appears that Hippo can be linked to just about anything and everything and that one would always have to make such studies in a specific context.

At a most significant level, the fundamentally genetic work with model organisms, does not address the question of how cells build tissues and organs and this is because a gene-centric approach can’t do that; I hasten to add that the fact that there is a gene involved in a specific process and that we might be able to interpret gain and loss of function phenotypes does not count in my book. To understand how cells (genes is more difficult) build tissues and organs requires a very different approach from the one we have been using until now. You don’t understand how a car works simply by removing pieces and running it –sometimes it will crash, others it will not start or run slower- you learn how it works by building it and testing what you have built or are building the necessity and sufficiency of every piece (see the famed ‘Can a biologist fix a radio’ ref 7)-. The case of yeast and the cell cycle is an interesting one in this regard. Some of the most salient insights in this field have not come only from mutagenesis and epistasis but from  a combination of modelling and engineering in which the order of the system is challenged and reorganized (8). This has led to discovering minimal systems, the functional meaning of redundancies and of modules with functional sense i.e going beyond genes. Behind these findings lurk principles and designs (see the interesting discussion presented in 9 and 10). Reengineering or engineering a system from the bottom up tells us much that Genetics cannot do. This kind of work is very difficult to do with model systems because they are not suited to this. In the case of some invertebrate models systems because they are too hardwired and one cannot create them from scratch. In the case of vertebrates because of their complexity challenges this type of approaches. Enter Stem Cells and, in particular ESCs as a new model system in Developmental Biology one in which we can ask how cells build tissues, organs, embryos.

ESCs are a derivative from early mammalian embryos with a huge potential to enlighten our understanding of how cells build embryos, tissues and organs. It is true that it is early days, that there is a lot of poor work and a great deal of salesmanship of mediocre science at the moment, but this should not hide the fact of what they offer if they are used correctly, with high standards and a critical eye. And organoids, mentioned in (1) are an excellent example of what these cells (and also Adult Stem Cells (ASCs)) can do and how much we can learn from them about details of Biology that are not accessible in any other way. But, we need to use them judiciously and critically. Some of the arguments for this have been made before (https://amapress.gen.cam.ac.uk/?p=2024 and https://amapress.gen.cam.ac.uk/?p=996 )  and I shall not repeat them here, but will emphasize that Duronio et al are right that Stem Cells and organoids not a substitute for Model Organisms, they are a new experimental system with possibilities that are just beginning to be explored and whose full potential remains to be explored. They are not going to replace vertebrate model systems because the kind of questions that they ask are different and complementary but they sure will change the way we view vertebrate and, in particular mammalian development. Look at the work of a small group of people and their collaborators -J. Briscoe, S. Lowell, V. Wilson, J. Brickman, P. Zandstra, A. Brivanlou and E Siggia, A. Warmfklash, Fu, M. Lutolf, A. Grapin-Botton amongst others- that are exploring this avenue. There are also people who are using developmental biology to pioneer the development of regenerative medicine with ESCs, the work of J. Wells and J. Spence comes to mind. Here it is this dialogue between Developmental and Stem Cell Biology that matters. And in the area of adult organs, optic cups, minibrains and intestinal organoids have been leading a revolution in our understanding of precisely this question of how cells build organs and how we can harness this potential for biomedical research (11, 12). If one looks at this work, one will see that there is nothing to fear and much to be gained by promoting research with Stem Cells IN the context of Developmental Biology. To paraphrase a well known quotation: Stem Cells without model systems are lame, Model systems without Stem Cells are blind.

And as long as we don’t hype the issue and provide perspective, we should not shy away from the statement that understanding the make up and development of human beings is an important aim of Developmental biology. The reasons for this could be sheer curiosity or the thought of applications to disease or regenerative medicine. However, there is no ethically sound model system to study human development. The last few years have shown that while mice are a good proxy for much that is human, the only way to study humans is with humans: the work with minibrains and intestinal organoids, highlights this. And it is here that, hopefully on sound grounds, ESCs and ASCs will become a most important tool in this endeavour because what we are learning from the work with mouse cells is that if properly used, they teach is things. Not everything that happens in an embryo happens in the dish but all that happens in the dish can happen, and often happens in the embryo.

It seems to me that what Duronio et al grapple with is not really Stem Cells and Organoids but something more mundane but perhaps also more fundamental: the squeeze in funding, the large number of researchers and the average high quality of research that makes striving for that funding a challenging task. From this perspective, we should not lose sight that Stem Cells face the same problems than Drosophila. The reason is that funders often want to hear that one is going to cure a disease and not to find something fundamental; this applies to model systems as well as to Stem Cells. Now what is important is that we shall not convince funders by denying what is obvious (the possibilities opened by Stem Cells and Organoids) but by highlighting how Stem Cells and Developmental Biology are not only very related but are, as discussed above, complementary. We should also remember that the discoveries Duronio et al highlight were made in a very different environment than the one we worked on today when, perhaps (I believe certainly) it was easier to discern what was important from what was anecdotal or simply interesting. This is difficult today when every week we are confronted by ‘great’ discoveries which strive to be seen in a rain of technical prowess.

Stem Cells as a model system are here to stay and should be seen as a complement and not a competitor in our current arsenal to understand Nature. Is is interesting to reflect on the observation that one of the great achievements of Development journal has been to create a highly successful section on Stem Cells! What we need to do, though, is raise our game, find good, difficult questions and ways to address them. We should not be complacent and think of experiments simply because we can do them or of systems because we can use them. We should be bold and look for new avenues an systems. Also, we need to admit that much has changed and is changing in the environment in which we do Science and that what worked in the past, does not work today. We need to change and adapt. We need to accept that we work in a very different environment with very little room for the scientist/craftsperson of the XIX and XX century (https://amapress.gen.cam.ac.uk/?p=1537). What we need to do is to find ways to use much that is good in this environment and invent new ways of doing Science. In the context of the essay that triggers this post (1), we need to avoid justifying classical model organisms on what they have done in the past and find ways to make sure that they contribute to the future. Biological systems are such that there will always be a new gene, a new organelle a new relationship to be found, like the collections of the XVIII and XIX centuries were putting together animals and plants. What we need to make sure is that we find the ‘finches’ and have the eye to recognise what they tell us.

 

References

  1. Duronio, RJ., O’Farell, P., Sluder, G., Su, TT. (2017) Sophisticated lessons from simple organisms: appreciating the value of curiosity-driven research. Dis. Model Mech. 10, 1381-1389.
  2. Huch, M., Knoblich, J., Lutolf, M. and Martinez Arias, A. (2017) The hype and hpe of organoids. Development 144, 938-941
  3. Kohler,RE. (1994) Lords of the fly. The University of Chicago Press.
  4. EA Carlson “Mendel’s legacy’ Cold Spring Harbor 2004 .
  5. Anderson, KV and Ingham, P. The transformation of the model organism: a decade of developmental genetics. Nature Genetics 33, 285-293.
  6. Ugur, B., Chen, K. and Bellen, H. (2016) Drosohila tools and assays for the study of human disease. Dis, Model and Mech. 9, 235-244.
  7. Lazebnik, Y. (2002) Can a biologist fix a radio? Or what I learnt whie studying apoptosis. Cancer Cell 2, 179-182.
  8. Coudreuse, D. and Nurse, P. (2010) Driving the cell cycle with a minimal CDK control network Nature 468, 1074-1079
  9. Atay, O and Skotheim, J. (2014) Modularity and predictability in cell signalling and decision making. Mol. Cell Biol 25, 3445-3450
  10. Mellis, IA and Raj, A. (2015) Half dozen of one, six billion of the other: What can small- and large-scale molecular systems biology learn from one another? Genome Res. 25, 1466-1472.
  11. Sasai, Y. (2013) Next-generation regenerative medicine: organogenesis from stem cells in 3D culture. Cell Stem Cell 12, 520-530.
  12. Bredenoord, AL., Clevers, H. and Knoblich, J. (2017) Human tissues in a dish: the research and ethical implications of organoid technology. Science 355, DOI: 10.1126/science.aaf9414

Developmental Biology: the cauldron of the biological sciences

Disclaimer: I am not a historian of Science and these views are, simply, sketches from the fringes of my interest in the subject as a practitioner of Developmental Biology.

Sometimes it is possible to draw some sort of a straight line in the history of Science. For example, one can build a sensible narrative across an axis Galileo-Copernicus-Kepler-Newton….or, sort of, as most of the time a scientific discipline is a pidgin of varying questions, thought traditions, data and, importantly, visions of different scientists and surprising experimental results. Any attempts to trace a scientific discipline to a specific root will be difficult. Visions and technical advances will always collide in unpredictable manners to answer old questions and find new ones. It would be easy to say that the combination of hindu-arabic numbers and modern Calculus is at the root (the stem cell?) of Physics and yet, the babilonians and the greeks did very well without either. The history of Science is, first and foremost, a history of progress through change, a blend of experiments, ideas and reasoning. Furthermore, it is usually when different languages and visions come together that something interesting happens. One can trace paths but there are no unique roots.

I am saying this while reflecting on an article published in PLoS Biology by Scott Gilbert, claiming that Developmental Biology is the ‘stem cell of biological disciplines’ (http://journals.plos.org/plosbiology/article/authors?id=10.1371/journal.pbio.2003691 ). The reason for this musing is the feeling that Developmental Biology, specifically Developmental Biology, is under threat as a discipline (a widespread thought in the field these days) and that this kind of revelation should lead us to place the discipline in its rightful place and make funders and other scientist give it a credit which many people think is overdue. I do not see any of this to be the case. Developmental Biology is no more under threat than other branches/areas of Biology (talk to prokaryotic biologists or virologists!). What is under threat (more change than threat) is a certain way of doing Science –but I shall leave this to another time, as here I want to focus on some specific aspects of the piece by Gilbert. All I will say for now about this is that we need to do is to adapt and to change. The point I would like to make here is that, even if the drama underpinning Gilbert’s statement were true, one does not help the cause by rewriting history and shoehorning facts. The author states at the beginning that it is good to create myths to defend a realm and while this could well be true, mythologies are not history and one should be careful with the consequences that go with creating myths which can mislead current generations of students, and pay lip service to the cause it pretends to help.

Let me go to the two specific claims I take issue with: that developmental biology is the birth place of Genetics and Neurobiology (I would add here immunology and cell biology but then this would be too long). Let me start with Genetics, something I have worled with, and make things clear from the beginning: the questions that led to the emergence of Genetics had nothing to do with Embryology (Developmental Biology did not exist at the beginning of the XX century), absolutely nothing. The fact that embryos of different species might have been used to answer some questions posed by the gene based theory of inheritance does not mean that Embryology was at the roots of Genetics. The questions that concerned Mendel, and later de Vries, Tschermak and Correns, had to do with hybrids, plant hybrids to be more specific, and the laws that regulate the inheritance of some of their phenotypic characteristics. If one goes back in history, one will find that the forerunners of Mendel, plant breeders in Germany like Kolreuter, were not interested in questions of Embryology. This bit is easy to demonstrate and I suggest that anybody with some interest in the matter reads the little book by Robert Olby ‘Origins of medelism’ (Schoken books New York 1966) .

The intervention of TH Morgan in the story lends itself to misconstruction as he was, first and foremost, an embryologist. However, at the time, it was not rare –as it is the case today- that people were interested in Evolution and, in particular, in matters of the inheritance of characters (not the genetic basis of development). Morgan had an interest in these matters which were of paramount interest to biologists at the time, with a particular focus in the nature of mutation. It is this that led him to start working with Drosophila and, in an inadverted manner, to lay down the foundations of Genetics in Columbia University with a small group of superb students led by Sturtevant, Bridges and Muller. All this without a single thought for embryology. It is true that Morgan did refer, particularly in his Nobel lecture, to the connections between Genetics and Embryology, but he did not see how to do it and, actually, in his latter years in Caltech he returned to embryology without even attempting to bridge the gap (see the excellent biography of Morgan by G. Allen, Princeton Univ. Press 1978).

With his work, Morgan starts what Sydney Brenner has called the ‘great deviation”, a proces in which Genetics and Cell Biology need to get on good footing before Embryology can answer the questions it had at the start of the XX century. In many ways and with the exception of the work of Spemann and his collaborators, Embryology enters into suspended animation until the emergence of developmental biology when the consolidation of a number of disciplines –in particular Genetics and Biochemistry- allows a new vista of old questions. Thus Morgan pioneers the Genetics branch of this path and does so in a somewhat unconscious manner. In this regard, Boveri’s work is interesting and important but is a thin (though important) strand in the cell biology stream which also develops –for the most part- answering specifc questions (the residence of the particles of inheritance). Let me repeat, the work of Sturtevant, Muller and Bridges under Morgan –and on their own- has nothing to do with and owes nothing to Embryology. Like classical physics lies in the axis mentioned above (Galileo-Copernicus-Kepler-Newton) one can draw an axis for Genetics: Kolreuter-Mendel-Bateson-Morgan-Avery-Watson and Crick-Nirenberg; will led the reader to decipher (of degoogleized) some of those names. There is not a whiff of Embryology here which at the time was concerned with other questions, many of which were difficult to tackle.

Once Developmental Biology emerges as a discipline, an offshoot of the Entwincklungsmechanick and Embryology, it has something to do with Genetics, particularly in the latter part of the XX century, but Embryology –as such- has very little to do with Genetics. An interesting observation about this lack of a relationship between Genetics and Embryology in most of the first half of the XX century can be seen in what came to me called “the modern synthesis which tried to reconcile Genetics and Evilution. There are two disciplines that are left out of this endeavour: Embryology/Developmental Biology and Paleontology. The reasons are different but in the case of what is the object of this discussion, the reason was that at the time the connection between Genetics and Developmental Biology is, for the most part, non existent. Later on, in the 80s there is much of a relationship, but it is not the one expressed by Gilbert, it generates Evo-Devo.

The application of Genetics to questions associated with the development of plants and animals, changes the way we practice the discipline and, most significantly, provides insights that we are still digesting. Thus, while Genetics does not owe much, if anything, to Developmental Biology -and much less to Embryology, Developmental Biology owes a huge amount to Genetics. The reasons for this are many and are made very clear every day but the most important one is that IF there is one discipline that is stem of biological sciences, this is Genetics. However, Genetics is not the stem cell of Biology (I think this is the wrong way of looking at history of Science) but the unifying language, the unifying discipline and the one that ever since its emergence provides a clarity of thought and argument to old biological questions and this includes those related to development.

I shall be brief on the second issue: that of the ‘historical lineage’ relationship between Developmental Biology and Neurobiology. The claim is not very difficult to challenge as Neurobiology is not just the study of the development of the nervous system but, more significantly, its function. The origin of Neurobiology lies in the discovery of electricity. Volta and Galvani, Hodgkin and Huxley had no interest in Embryology nor developmental biology but it is them and their work that creates the roots and the basis of Neuobiology. There is little one can add to this well established fact. Ross Harrison, referred to in the Gilbert article,  is a pioneer of Developmental Neurobiology and, in particular, of tissue culture and ex vivo approaches to biological questions. He should be celebrated (as he has) for that. To place him at some root of such an advanced, complex and insightful branch of Biology as Neurobiology, is to reduce Neurobiology to Developmental Neurobiology, a somewhat narrow view of he field. Even on the example given by Gilbert, it is true that certain aspects of Neurobiology have their roots in embryos –but NOT EMBRYOLOGY- for it was the vision of Ramon y Cajal to use embryos to look at the structure of the nervous system, where he thought that a simpler system would yield answers to the disputes with Golgi raging at the time about the individuality of neurons. Thus, I think this is a better example to relate Developmental Biology to Neurobiology but, again, Ramon y Cajal was not interested in developmental questions but in very specific issues of cell biology associated with the nervous system.

In Biology, as in Science, one needs to use specific systems to find about the world through experiments but this does not mean that certain disciplines associated with those systems are at the root of further developments. Furthermore, I don’t believe that there are stems to whole branches of Science. Particular branches: quantum mechanics, astronomy, organic chemistry, Cell Biology and, of course, Genetics are pidgins of many ideas and languages and in the blends that they represent lies their greatness and their value.

I appreciate the enthusiasm of the piece in PLoS Biology but we should be careful not to distort history for the younger researchers. In the context of the theme of the essay we should not create weak phylogenies but rather thrive in the diversity that gives so much to Science and encourage interactions between disciplines as it is here that discoveries lie.  Developmental Biology is, today. a pidgin of  many disciplines: Genetics, Biochemistry, Cell Biology, Evolution, where the  recent application of Physics and computations methods is stimulating the development of these fields and providing, as it should, new vistas of old problems in Developmental Biology. Rather than a stem cell, I encourage you to see Developmental Biology as an alchemic cauldron where the blend of disciplines produces exciting and interesting findings which, slowly, will find their way into practical applications. Cell Biology and genomics are good examples where the application of these disciplines to developmental systems has revealed issues that cannot be addressed, because they don’t exist, in different contexts e.g transcriptional states during cell fate transitions or the emergence of tissue level behaviours from individual cells. More recently, the application of Physics and Maths is changing our view of developmental systems, providing intriguing new insights and also transforming the fields (Physics and Maths) in surprising and interesting ways

As for the real problem that the article identified and tried to address, it is not the lack of recognition of Developmental Biology . Furthermore, it is not specific to Developmental Biology. The issue is deeper and it has more to do with the way we do Science today as opposed to a few years ago, the fabric and form of Science. I have mentioned some of the issues before. By creating a mythology of the past we do not solve this. What we should be doing is identifying the exact challenges that we face and rising to them by looking at the future in search of creative solutions.

 

Notes

  1. I shall not discuss the uses of the words ‘evolution’ and ‘development’ in the late XIX century which are raised in the article. This would require more time, space and study that I am able to provide now. The issue is typical of nomenclature and the use of words in Science which acquires large dimensions in Biology where definitions are loose. It is very similar to what is happening with ‘epigenesis’ nowadays.
  2. There are many excellent texts about the history of Genetics but I like, specially EA Carlson “Mendel’s legacy’ Cold Spring Harbor 2004 . There are less single books on the history of Developmental Biology –though there are many collections. I suggest two: J. Oppenheimer “Essays on the history of embryology and biology’ MIT press 1967 and “A history of embryology’ edited by T. Horder, J. Witknowski and C. Wiley Cambridge University Press 1985.

 

On the Value of Imitating Nature (imperfectly)

COI: our group has an interest in these matters and the post reflects this and contains references to our work.

The idea of making a human being from natural or unnatural parts has been more than a curiosity for centuries. The story is told in a little known book by Philip Ball (“Unnatural: the heretic act of making people” Vintage books, London 2012) and has many a fascinating angle, though the bit I really enjoy is the development of in vitro fertilization by Patrick Steptoe and Robert Edwards. So much Biology behind what today is a routine clinical procedure! This is an important point: whatever technology there is around, it is underpinned by basic Science. No application can be developed without the support and inspiration of a scientific question and this applies to making cars or humans. But, why would one want to make a human being in a lab?

If you want to make a human all you have to do is to combine an egg with sperm and let it run. The entry of the sperm into the egg triggers a process that generates an embryo which, in turn, lays down the templates for tissues and organs and their relative organization in space. Leaving aside parthenogenesis (the activation of the process without fertilization), this is the way you make any animal and, in many ways, this is what Steptoe and Edwards did for the human by recreating the event in laboratory conditions. However, the process of activating development does not tell you much more than what you knew already since, once the process is triggered you are back to just watch how form unfolds over time i.e. Life, if you want to be dramatic, requires Life and all its pieces working together harmoniously. Can we do better? The advent of Embryonic Stem Cells (ESCs) has given the age old challenge of creating organisms in the laboratory a new lease of life, though very few people would look at it from the perspective of the actual making of a human.

ESCs are clonal derivatives from mammalian blastocysts that can be grown in culture and differentiated in a controlled manner into most cell types of an organism.  Recently it has been found that these cells have the ability to organize themselves into tissues and organs and that we have the capacity to steer (but not to control) these processes. Thus disembodied brains, eye cups, intestines and muscles emerge from culture dishes under so called ‘defined conditions’. These structures are imperfect, often functionally wanting and, for the most part, just happen (rather than are created) and promise much more than they can deliver at the moment. Importantly, amplified by the media, they create dangerous hype and false hope. Intriguingly, they also raise questions about what is an organism that we have not thought much about yet.

A different approach to the problem is to go beyond the organs themselves and ask the question of whether ESCs can be used to build an organism or, for starters, an embryo. There are two reasons for this. Here I shall only state them and hope to find some time to expand on them in the near future. The first one is the sheer scientific value of what one can learn from such an experiment about how organisms are made. The second one, probably more practical, suggests that if you want to build tissues and organs, you better try to copy the embryo, to imitate Nature. Perhaps by starting with embryos one can endow cells with the properties of the real objects, improve the yield of organs and tissues and get them to a functional state. An interesting consideration arises from these musings: is it not what Steptoe and Edwards did an imitation of Nature? Did they not get an embryo out of a dish? Sure, in the end the embryo, the mammalian embryo, will need the mother but they did create a human embryo in the lab. But what do we learn from such imitations about the emergence of form? If one wants to learn how cells make embryos and we rely on the properties of the system to create embryos, what have we learnt? By counterfeiting a painting you don’t learn about the process that created it. Science teaches us something from decomposing, from asking about minimal conditions. A good example can be found in the roots of Developmental Biology (the branch of Biology that studies the emergence of an organism from a fertilized egg). For many decades in the XIX century biologists looked at and described the embryonic development of different organisms and thus developed a good roadmap of what happens after fertilization. It was the agenda laid down by Wilhem Roux and its dramatic execution by Hans Driesch with his embryo splitting experiments, that teaches us something, namely the regulative power of embryos and, in doing so, it lays down the agenda for Experimental Embryology and later Developmental Biology in the XX century ( see A New Sort of Engineering I and II).

The initial question behind the experiments of Driesch and what followed was to find out the minimum number of elements that can give rise to an organism but, from there and because of the answers that emerged, this approach has been rich in contributions to our understanding of biological systems. The reductionist/deconstructionist approach that underlies Experimental Embryology generated a very successful research program that was further underpinned by Genetics: instead of using scalpels and needles, use mutations to understand the system. Genetics has shown not only the processes but has identified the parts. Its combination with molecular biology has proven most effective in the dissection of the entrails of the System and has revealed that the making up of an organism is not what it seems. A classic favourite of mine is the organization of the early Drosophila embryo into stripes, beautifully choreographed in their emergence and spatial organization but which experimentation shows are individually controlled elements of an evolutionarily fitted system. The visible coordination, the obvious symmetry is an illusion created by Natural Selection; the system is built piecemeal with the whole being a collection of parts, seamlessly woven into each other for, probably, function, over millions of years of tinkering. This should not come as a surprise. After all, how are Diptera (flies) made but from disconnected parts, called imaginal discs, which develop autonomously and independently from each other inside the larva (a feeding machine that provides the raw energy and matter for the growth of the parts of the adult). Although the imaginal discs do not raise eyebrows, particularly for those that work with them, it is a strange thing that legs, eyes, genitalia and wings develop independently from each other only to come together in an assembly line during metamorphosis in the ultimate example of self-assembly.

Drosophila has taught us much but there is one feature which is very revealing and is always overlooked as trivial, namely this coming together of the adult organism from parts, that the whole that we see flying around is an illusion created by the convergence of chance and necessity, of teleology and Natural Selection working together for reasons and benefits that we do not understand.  What the development of Drosophila tells us is that Biology is not intuitive and that behind a seamless whole there is much patchwork. But what does this have to do with making humans?  Allow me to rephrase the question, rather than making humans -which I would argue has already been achieved- what we want is ways to learn how humans make themselves, how cells make humans. At this, perhaps we should aim lower first and start with a system we can work with and which is sufficiently similar to teach us the ropes. How about mice and, at even at a humbler level, mouse embryos? If we can make the seed, for that is what an embryo is, maybe we can learn about the organism. As usual, there are many ways to ask questions and we and others are trying 1 but while some aim to counterfeit the embryo, some of us have decided to see what cells can do if we set them free or if we constrain them. For example, arranging ESCs in constrained spatial arrangements and asking them to differentiate is teaching us about the connection of specific genetic circuits and the spatial self-organization of differentiating ESCs 2, 3. On the other hand, the development of 3D gastruloids has revealed many surprises as it looks as if, against the established wisdom of the field, the development of the axes of a mammalian embryo does not require the extraembryonic tissues 4-6. Furthermore, this work (still in its early days) reveals what could be interpreted as the equivalent of the imaginal discs of the mouse: that the head and the trunk can emerge independently of each other and that depending on the signals the cells receive at a particular time, they will assemble into specific and different parts of the embryo 7. I can hear the embryologists raising their voices and claiming that this is an artefact of the experimental system and this echoes much of what was in the air when Drosophila started to reveal the modular nature of its inner engineering. It may be, but I suspect that they are telling us something and that these contraptions are real.

What this says is that we are moving into an era in Developmental Biology in which Engineering is going to play a central role and will do so in two ways. One, in the manner that we usually understand engineering i.e. the development of efficient and reproducible ways towards a practical end. This has already had an impact in Biology and Biomedical research: genetic engineering, as well as artificial hearts, bones and limbs bear witness to this. Now we are starting to apply this to the activities of ESCs, though we should be careful to control the hype that can surround this research. In addition, we should look at engineering as a discovery tool. In some ways and inadvertently, we have already since when Driesch and his followers start chopping the embryo into pieces, they are using some engineering principles. They are doing it in a crude way (can a biologist fix a radio?) but they are. The analysis of the stripes in the early Drosophila embryo is another more sophisticated example of this and one that has revealed some profound principles of the link between the molecular logic of a cell and how it is used to build the blueprint of an organism.

I do not see a reason to build an embryo of a mouse or of a human. This is a relatively interesting headline-grabbing challenge.  On the other hand, I do see many reasons to ask questions about how cells build organisms or how much of an organism can arise from the autonomous behaviour of cells. In doing so we are learning interesting facts. For example that there are length scales in the organization of the basic cell populations (germ layers) that make up an embryo and, significantly, as pointed out above that mammalian embryos are, like Drosophila, piecemeal entities put together by evolution. In some ways, the emergence of organs in a dish hints at this but the reliability and reproducibility of this event might need the generation of the components of an embryo. I suspect that in understanding this there are prizes that will help us in the long term goal of engineering tissues and organs. Rather than imitating Nature we should aim to break the mirage that Nature and our minds have created in this wonderful balanced and continuous unit that we call an organism.

Note. The figure at the top is “the human condition” from R. Magritte.

  1. Simunovic, M. & Brivanlou, A.H. Embryoids, organoids and gastruloids: new approaches to understanding embryogenesis. Development 144, 976-985 (2017).
  2. Warmflash, A., Sorre, B., Etoc, F., Siggia, E.D. & Brivanlou, A.H. A method to recapitulate early embryonic spatial patterning in human embryonic stem cells. Nat Methods 11, 847-854 (2014).
  3. Etoc, F., Metzger, J., Ruzo, A., Kirst, C., Yoney, A., Ozair, M.Z., Brivanlou, A.H. & Siggia, E.D. A Balance between Secreted Inhibitors and Edge Sensing Controls Gastruloid Self-Organization. Dev Cell 39, 302-315 (2016).
  4. Turner, D., Glodowski, C., Alonso-Crisostomo, L., Baillie-Johnson, P., Hayward, P., Collignon, J., Gustavsen, M.W., Serup, P., Schroter, C. & Martinez Arias, A. Interactions between Nodal and Wnt signalling Drive Robust Symmetry-Breaking and Axial Organisation in Gastruloids (Embryonic Organoids). bioRxiv doi.org/10.1101/051722. (2016).
  5. Turner, D., Alonso-Crisostomo, L., Girgin, M., Baillie-Johnson, P., Glodowski, C., Hayward, P., Collignon, J., Gustavsen, M.W., Serup, P., Steventon, B., Lutolf, M. & Martinez Arias, A. Gastruloids develop the three body axes in the absence of extraembryonic tissues and spatially localised signalling. bioRxiv doi.org/10.1101/104539 (2017).
  6. van den Brink, S.C., Baillie-Johnson, P., Balayo, T., Hadjantonakis, A.K., Nowotschin, S., Turner, D.A. & Martinez Arias, A. Symmetry breaking, germ layer specification and axial organisation in aggregates of mouse embryonic stem cells. Development 141, 4231-4242 (2014).
  7. Baillie-Johnson, P., van den Brink, S., Balayo, T., Turner, D.A. & Martinez Arias, A. Generation of aggregates of mouse ES cells that show symmetry breaking, polarisation and emergent collective behaviour. JOVE doi: 10.3791/53252 (2014).

A Short Tale about Brachyury

Once upon time….somebody said that Genetics is Ariadne’s thread of Biology, the only way to guide us out of the labyrinth that is a biological problem. Nowhere has this been more true than in the analysis of development, the processes underlying the emergence of an organism. In this spirit, it is the systematic application of genetic analysis to Drosophila melanogaster and Caenorhabditis elegans spearheaded a deluge of information that started modern ‘Developmental Biology’ (the study of the dynamics of pattern and form in embryos, as opposed to ‘Embryology’, the detailed description of the different stages associated with embryos). The success of these two invertebrates took its time to permeate the mammalian embryo. There are good reasons for this. Two possibly important ones come to mind: the challenge that is to study an embryo that develops inside the mother and, not unrelated, the difficulty of applying Genetics to such an object. In retrospect we can see that most of the mutations that have given us insights into developmental events are embryonic lethal and it has been the pathological, or if you don’t want to sound so dramatic say phenotypic, analysis of those mutations that has yielded a view of the process. Notwithstanding husbandry and breeding details, the mammalian embryo, small and hidden in the confines of the uterus, is not an ideal target of systematic screens (the mammalian genetics papers) but with patience and focus, Genetics was applied and the usual combination of serendipity and method have yielded their fruits.

It could be said that there are many kinds of Ariadne’s threads. For example, mammals have an interesting aspect of their make up that is not a bad one into the labyrinth and that, in my view, has not been exploited as much as it might: haploinsufficiencies. Dominant phenotypes caused by the loss of one allele that are related to the function of a gene and it is with a haploinsufficiency that mammalian Developmental Biology started with the discovery of the mutation that lends its name to the gene Brachyury. In some ways the story (or the path if you want to stick to the mythology) of Brachyury is a story which highlights the highs and the lows of the connection between Genetics and Development, that reveals how easy it is to be distracted by the underlying complexity of biological processes, how molecular biology can simplify things but also how we should not be complacent and forget that genes are not causes but simply tools and elements to understand a process. It is also an example, as it unfolded, of the power of the blend of Genetics and Embryology that forms the core of XX century Developmental Biology. Also, perhaps surprisingly to some, Brachyury paved the way for the genetic analysis of development as it was the first mutation (genes are related to mutations but are not the same) associated with a developmental defect, many years before Poulson’s work with Drosophila Notch (1).

170213 T CrossWe do know today that Brachyury, also known as T or T/Bra, is a gene that encodes a transcription factor of a family called the T-family because of their structural relatedness, that play crucial roles in early development (2). We also know that it has a central role in the processes of gastrulation and axial extension in chordates and that its existence extends to invertebrates where it has a variety of roles in early patterning of embryos (3).

The origin of Brachyury lies in the 1920s Paris where Nadine Dobrovolskaia-Zavadskaia, a Russian emigrée from the revolution, was working at the Institute for Radium (an interaction between the Curie Laboratory and the Institut Pasteur) looking for the ability of X-rays to induce mutations in mammals (4). In the course of a screen involving 3000 crosses of mice across three or four generations she found two mutations that bred true, one of them she called T. The mutation had a dominant phenotype reflected in the length of the tail, hence the name T for taillessness (the capital for the dominant phenotype) but was, also, embryonic lethal (5).

An obsession with T starts and, in parallel with the discovery of recessive alleles, t, and interacting mutations (in laboratory and wild mice), descriptions of the phenotypes of different genetic combinations suggest an involvement of T, whatever it was, with the development of spinal cord and of the vertebral column (6-8 and then, if you want to follow on, read the more general accounts in refs 9 and 10). One thing is to describe a mutant phenotype –a forensic exercise- and a different one to wonder whether it tells something general about the connection between genes and developmental processes; I am, not sure that we have yet resolved this problem. It was, probably, Salome Gluecksohn-Schoenheimer who began to do the second through her analysis of T and t mutants (8, 10). There had been descriptions of the defects of the mutants (6-8) but she went further. Having worked in Spemann’s laboratory (11) the defects in the development of the notochord, neural tube and somites associated with T mutants, made her think of the effects of the organizer and of a possible relationship between T and its activity. These thoughts were no doubt encouraged by her discussions with, particularly C. Waddington, and the parallels she drew with mutations affecting the patterning of the Drosophila wing, suggested that T was saying something of how genes related to developmental processes (10, 11). But these were early days to see, let alone study, this relationship. Ariadne’s thread was in a badly tangled ball and the genetics of T did not help. Mutants with T like phenotype, some allelic, some not but which acted as modifiers were isolated in wild and laboratory strains, and their relationship to the original T mutant proved genetically complex (see brief discussion of these matters in refs 9, 10 and, more extensively, 12). The notion that these genetic interactions were linked to the function of T, led to a labyrinthine situation.

The problem was, and still is, that the genetic analysis of a biological process has its limitations; it is a bit like palm reading. It requires interpretation and, in the case of complex processes –like Development- it is not the way to unravel Mechanism. To quote a very prescient thought from Gluecksohn-Schoenheimer  “A mutation that causes a certain malformation as the result of a developmental disturbance carries out an “experiment” in the embryo by interfering with the normal development at a certain point. By studying the details of the disturbed development it may be possible to learn something about the results of the “experiment” carried out by the gene. However to discover anything about the nature of the action of the gene is a much harder task. It is necessary for this purpose to be able to trace back all the results of the action to certain original causes” (8). Furthermore, Genetics for its own sake (the analysis of the complexity of interactions, alleles, pseudoalleles, complex complementation associated with the breeding of a trait) can intervene, capture our imagination and lead us astray. A bit like maths without Physics; elegant and fun but lacking bite.  However, Genetics was what there was and the genetics of T proved a bit of a tangle (12), particularly when it came to the question of the link of the mutants (and at the bottom the wild type gene) to the generation of the embryo.

The question of what was the relationship between genes and development was an important issue in the 60s and the 70s. It would emerge from the molecular and cellular analysis of T, and in parallel of the bithorax complex (BX-C) in Drosophila by Ed Lewis. How genes controlled development did not look a simple affair at the time. Being interested in the topic as a graduate student in the late 70s and the early 80s I remember reading the seminal –but somewhat arcane- 1978 paper by Lewis on the BX-C (13) and a profoundly puzzling one on T (14). I must admit at having been simultaneously fascinated and befuddled by both but particularly by the mouse one. The complexity and the challenge to unravel them was part of the excitement to solve the puzzle. The Drosophila case seemed easier to unpack, perhaps; and it was. The reason probably lies in the rather Cartesian organization of the embryo –which was rather well laid out at the time- and the linear manner in which events unfold that allowed a rather rapid connection between genes and specific developmental events. It has interested me recently, how much hard work was going on to outline the battleground to apply Genetics to mammalian embryos when the buzz was about the genes of Drosophila. But T was not forgotten as it held a key to how mammalian embryos were built and therefore needed to be addressed.

170213 T ExpressionFrom here, the story gathers pace and the riddles find their solutions. And so it was that the era of molecular cloning clarified matters. In the 80s the systematic application of DNA cloning and analysis to the harvest of mutants screens from Drosophila and C. elegans, paved the way for similar work in other organisms, particularly mammals, and started to lift the fog that Genetics has laid on T. The BX-C turned out to encode three transcription factors and a complex regulatory region (15,16), while T/Bra encoded one transcription factor (17,18). The nature of the devilish genetics of T/t still lingered in the background but the brutal simplicity of molecular analysis delivered its verdict: T was, IS, a transcription factor expressed in the notochord as well as the progenitors of the spinal cord and the mesoderm (see Figure). The question then was, not how this related to the complex genetic analysis revealed by the multiple alleles and crosses –this is, apparently, still work in progress- but how this related to the activity of the protein, encoded by the gene and revealed by its loss of function. And this was just the beginning (in developmental biology, knowing what kind of protein is encoded by a gene –that we have come to know from a mutant phenotype- often opens more questions than it answers). The question was now the original one and again Gluecksohn-Schoenheimer, presciently posed it;  in 1938 thinking about how Spemann’s group approached the problems associated with embryonic development she mused about how in an ideal world to approach mammalian development: ‘The events that take place in the development of the mammalian embryo have not been subjected to an extensive causal analysis so far. The reasons for this are to be found mainly in the lack of suitable methods. It is not possible yet to use transplantation, isolation or vital staining techniques on mammalian embryos as they have been used on amphibian embryos. In the course of time it probably will be possible to analyze the mammalian embryo by transplantation and isolation just as thoroughly as has been done with the amphibian. For the present, however, the experimenter is not able “to take an active part in the course of events that take place during the embryogeny of the mammalian embryo,” nor “to alter the course of events at a chosen point in a chosen manner and draw conclusions on their relations from the resulting changes.” (Spemann 1936.)’ (8). An amazing paragraph for the time. She was more right than perhaps she thought and the molecular biology of T was going to pay high dividends that would make her musings true. The 80s gold rush of gene cloning revealed that T/Bra was conserved across phyla and its discovery in Xenopus led to a fruitful link with mesoderm induction that Jim Smith and his colleagues pursued in an enlightening manner for many years, establishing how T/Bra worked at the molecular level and established a connection between T/Bra and gastrulation (see e.g 19, 20). In parallel, Rosa Beddington began to apply the techniques that were being developed to study mouse development to the molecular insights and reagents thus unravelling the connections between T/Bra and mammalian development that lurked behind the genetics for so long (21-23). This work was soon picked up by one of her collaborators, Valerie Wilson, who has been pursuing the intricacies of the relationship between T/Bra and the mammalian body plan for the last many years answering many of the questions that had been posed by the early Genetics of the mutant. There is still much to do because, thought we have learnt much, the question of the relationship between gene and effect, mutant and phenotype, remains. But now we have the tools and the framework to try to answer it. We just need not be distracted by the ease to gather facts and remember the questions.

The story of T/Bra is a good example of the way in which the blending of Genetics. Embryology and Molecular Biology have enlightened the relationship between genes and development. History can be anecdotal but it is also informative. The recent history of developmental biology is very focused on Drosophila and C. elegans and it may come as a surprise to many that T/Bra, as a question and as a reality, was there before Bicoid and Wnt and Notch, highlighting the questions that needed an answer. History, in this case, also highlights the perils of the purely Genetic analysis of a biological process and the need to remember that Genetics is a language, a formal language, to ask questions. In the analysis of Development, it leads us to the elements of the system but might not be the element of choice to see how they come together to make an organism. The position of Genetics to Biology is a bit like Mathematics to Physics: a language that allows one to formulate a question in formal terms which then provides a machinery to work towards the answer but the output of this operation needs to be interpreted.

The history of Brachyury also has an interesting element in that it highlights the important contribution of women to the field; most of the important breakthroughs and insights in the story come from women: Nadine Dobrovolskaia-Zavadskaia, Salome Gluecksohn-Schoenheimer, Virginia Papaioannou, Rosa Beddington, Val Wilson. An influence worth emphasizing as we celebrate Women in Science week.

In the end, the tale continues. Rather than the English and “they lived happily ever after’, we could quote the French “ils vécurent heureux et eurent beaucoup d’enfants” (they lived ever happily after and had a lot of kids), the kids being the myriad of questions that have been raised by the great discoveries about T over the last twenty years. Discoveries which open the door to answer to the questions that T/Bra leads us into: about Development, Genetics, Evolution, about Stem Cell biology and, in the near future, of the engineering of living systems.

 

NB One appreciates that this piece just skims through the surface of the story and its implications. Still, one hopes that this will inform at some level and, also, encourage thinking about the connections between Genetics and Developmental processes. I am grateful to Peter Baillie-Johnson for the suggestion of the title. The image on the genetics of T is from: http://www.biologydiscussion.com/gene/genes-types/genes-types-top-6-types-of-genes-genetics/67413

 

References

1. Poulson, D. 1940. The effects of certain X-chromosome deficiencies on the embryonic development of Drosophila melanogaster. J Exp Zool 83: 271–325.

2. Papaioannou VE. (2014) The T-box family: emerging roles in development, stem cells and cancer. Development 141, 3819-3833.

3. Technau, U. Brachyury, the blastopore and the evolution of the mesoderm. BioEssays 23, 788-794

4. Korzh, V. and Grunwald, D. (2001) Nadine Dobrovolskia-Zavadskaia and the dawn of developmental genetics.  Bioessays 23, 365-371.

5. Dobrovolskia-Zavadskaia, N., 1927 Sur la mortification spontane´e de la queue che la souris nouveau-ne´e et sur l’existence d’un caracte`re (facteur) he´re´ditaire “non viable.” C. R. Seances Soc. Biol. Fil. 97: 114–116.

6. Dobrovolskia-Zavadskaia, N, Kobozieff N, and Veretennikoff S. Etude morphologique et genetique de la brachyourie chez les descendants de souris a testicules irradies. Arch de Zool Exp 1934;76:249±358.

7. Chesley P. (1935) Development of the short-tailed mutant in the house mouse. J Exp Zool 1935;70:429±459.

8. Gluecksohn-Schoenheimer, S. (1938) The development of two tailless mutants in the house mouse. Genetics 23: 573–584.

9. Pappaioannou, V. (1999)The ascendency of developmental genetics, or how the T complex educated a generation of developmental biologists. Genetics.151. 421-425.

10. Gluecksohn-Schoenheimer S. (1989) In praise of complexity Genetics 122, 721-725.

11. Gluecksohn-Schoenheimer S. (1992) The causal analysis of development in the past half century: a personal. Development 1992 Supplement

12. Silver L.M. (1985) Mouse t haplotypes. Annu. Rev. Genet. 19, 179-208.

13. Lewis, EB (1978) A gene complex controlling segmentation in Drosophila. Nature 276, 565-570.

14. Artzt, K., McCormick, P. and Bennett, D. (1982) Gene mapping within the T/t complex of the mouse. I. t-lethal genes are nonallelic. Cell 28: 463–470.

15. Bender W, Akam M, Karch F, Beachy PA, Peifer M, Spierer P, Lewis EB, Hogness DS. (1983) Science 221, 23-29.

16. Gehring, WJ (1992) The homeobox in perspective Trends in Biochem. Sci. 17, 277-280.

17. Hermann, B. G., S. Labiet, A. Poustka, T. King and H. Lehrach,  1990 Cloning of the T gene required in mesoderm formation in the mouse. Nature 343: 617–622.

18. Kispert A, Koschorz B, Herrmann BG. (1995) The T- protein encoded by Brachyury is a tissue specific transcription factor. EMBO J. 14, 4763-4772.

19. Smith JC, Price BM, Green JB, Weigel D, Herrmann BG. (1991) Expression of a Xenopus homolog of Brachyury (T) is an immediate-early response to mesoderm induction. Cell 67, 79-87.

20. Saka Y, Tada M, Smith JC. (2000) A screen for targets of the Xenopus T-box gene X-bra. Mech Dev. 93, 27-39.

21. Beddington RS, Rashbass P, Wilson V. (1992) Brachyury, a gene affecting mouse gastrulation and early organogenesis. Dev Suppl. 1992:157-65.

22. Wilson V, Rashbass P, Beddington RS. (1993) Chimeric analysis of T (Brachyury) gene function. Development. 1993 Apr;117(4):1321-31.

23. Rashbass P, Cooke LA, Herrmann BG, Beddington RS. (1991) A cell autonomous function of Brachyury in T/T embryonic stem cell chimeras. Nature 353, 348-351.

Engineering the Future of Developmental and Stem Cell Biology

Final poster.001A meeting was held recently at the Pasteur Institute on the topic “Engineering the embryo: beyond Systems Biology”. The event brought to my mind a question I pose to the final year undergraduate class: how should we approach a biological problem? like physicists or like engineers?

The relationship between Physics and Biology has a long and very distinguished history, strewn with technical contributions that have often changed the direction and pace of biological research. Microscopy and X-ray crystallography would not have happened without the intervention of the physicists. To see this you don’t need to go further than the 2014 Nobel prize to E. Betzig, S. Hell and W. Moerner for the deep developments in superresolution techniques that are having a most dramatic impact in cell and developmental biology. However, there are also profound and long lasting conceptual contributions. Neurobiology has benefitted enormously from the input of physicists and molecular biology was shaped by Schrodinger’s “What is life?” and Max Delbruck’s Phage School. More recently, over the last twenty years, a cadre of young physicists have led a renaissance of the relationship between Physics and Biology. Single molecule and single cell techniques have opened up our eyes to the statistical processes underlying molecular and cellular Biology and taught us the beginning of how to deal with them. For those of us who have been lucky to be part of this feast, it has been both fun and insightful. If you have missed it, I suggest you catch up.

The fact that a biological system integrates multiple variables was never in doubt but it has been our ability to access those variables that has changed the game. Faced with a deluge of information of specific processes, we have learnt to measure and to use models to understand those measurements and, in turn, perform precise experiments. In many ways we are at the beginning of this game but the input of the physical sciences forged over the last ten years has already left an important imprint permeating much of forefront biological research. There is much to do and we now know how. Why then mulling over engineering?

The word ‘engineering’ has many meanings but, in general, evokes images of machines and blueprints. If you belong to the last century, like some of us do, you might think of steam engines, bridges, chemical plants, of belts and braces. If you belong to this century you might think of jets, computer chips, speed trains and clean technology, of electronics. Engineering is, in many ways, applied Physics. It is about toiling with materials to do something useful in a reproducible manner. Where a physicist uses an equation to understand/explain a process, an engineer uses it to explore a practical solution to a problem. Where a physicist uses a phase space to explore the behaviour of a system, an engineer will look at it seeking the domain that might work for a specific process in the real world. A phase space may often be the end game of the physicists work, for the engineer it is the starting point. While much of engineering deals with the physical world, there have been many inroads into Biology and, as in the physical world, there are many kinds of engineering in Biology: fermenters, biochemical reactors, bacterial circuits, mechanical organs are examples that come easily to mind. A good example of bioengineering you can relate to is beer making (yeast engineering in disguise). However, over the last few years, two fields -developmental and stem cell biology- have been coming together unknowingly with a common nexus through Engineering, a new kind of Engineering that was the motivation of the Pasteur meeting. Let me explain.

It is about five years that the tragically late Yoshiki Sasai surprised everybody by showing that Embryonic Stem (ES) cells could be coaxed into forming eye cups and forebrains1,2. His work coincided with the observation that intestinal stem cells would build crypts on their own3 and was followed by a plethora of other reports showing that (but not how) embryonic and adult stem cells can give rise to tissues, organs and structures. These observations are often hailed as the dawn of a new field, organoid biology, though one wonders if this is not, really, Developmental Biology by another name and with one significant difference: embryos, the realm of Developmental Biology, are reproducible, organoids more often than not, aren’t; this has consequences and sets a target.

Sasai was, at heart, a developmental biologist and, having worked with Eddy de Robertis (but probably from before) he was aware of classical experiments in which when animal caps from frogs are left to their own devices, they will make eye cups and forebrains (see Hurtado and de Robertis for a practical review 4). Furthermore, as shown in an insightful review in Development 5 he clearly knew the early development of the eye cup and the lens and thus understood the necessity to relate whatever happens in an ES cell culture to what happens in embryos. Thus he had a good sense for the fact that the autonomous potential of ES cells to organize into tissues autonomously was neither new nor unusual, it reflected what those cells are meant to do. The autonomous gastrulation of frog tissue (exogastrulae), the ability of cells from limb buds to to self organize into digits and the remarkable but much forgotten experiment 6 in which when frog animal cap cells are jumbled up, reagregated and exposed to signals they organize in space as the normal animal caps do, are some examples of this. If you allow me,  ‘organoid biology’ is a rebranding of ‘tissue and organ morphogenesis’ with, perhaps, the added spice that it is making use of our ability to differentiate stem cells rather than the material provided by embryos. It also carries a lot of hype.

The sight of disembodied organ-like structures in culture dishes easily captures the imagination. Probably it is for this reason that the report that a human brain had been grown in a dish 7sparked a huge amount of attention and interest amidst researchers and the public. The press fuelled this interest with headlines such as “Mini brains allow scientists to study brain disorders” or ‘lab grown minibrains aid Alzheimer disease researchers’ that created expectations. What had been achieved did not live up to the billing. What had been observed was not that different from what had been achieved in the mouse system but the fact that it was human, as these things tend to be, and that it had been achieved from iPS cells (genetically engineered ES cells) was an irresistible combination. There was also a sideline about microcephaly and the possibility of modelling brain diseases but one wonders how much of this had to do with journal headline and how much with real science. We certainly look forward to further reports on this front.

There is little question that this was an achievement but what the media forgot to say is that the events that led to those structures are the creations of cells over which we have little control. Furthermore, that in the v1.0 there were no brains as such but rather whimsical structures with a mixture of elements from a brain (forebrain, midbrain and eye tissue), that the system was hardly reproducible, that we had little understanding over what had happened. Feynman famously left on his final blackboard the statement “that which I cannot build, I do not understand’. If in ignorance we extrapolate this to the organoids that grow in culture dishes we would have to say: that which we can build we do not understand. The reason is, of course, that we have not built anything ourselves, that it is the cells -with rules that at the moment we cannot fathom- that have done it (as they are programmed to do), that we are at the mercy of the cells, privileged spectators of their productions. Can we change this? What is missing? Enter the engineers.

Figs for Paris.002If the system was engineered it would be reproducible and this would allow us to learn. Reproducibility is a first and most important target of the organoid game at this stage and it is where the interaction with engineers is key. Embryos are reproducible systems and it is because they are reproducible that we can use them to learn, that we can detect small changes in patterns to obtain clues of how genes relate to the building of tissues and organs. As F. Jacob pointed out, organisms are the result of evolutionary tinkering, bricolage, a more rudimentary form of engineering and so, perhaps, a way to understand them is to look at them from the point of view of engineering. The goal is not to GUESS how an organism is made, BUT to KNOW how it is made and for this only if we obtain reproducibility shall we understand. If we want to use organoids, we need to strive to make them reproducible and then, not only we shall be able to use them but, they will also teach us about the processes and interests that they represent. Recently there has been some progress in some of the organoid cases and the possibilities are clear. Thus an engineering of the minibrain system has allowed insights into the working of the Zika virus and promises more 8,9, and an engineering of the already robust intestinal organoids, has created the conditions for a sophisticated degree of reproducibility that will contribute to ongoing studies10. These are examples to follow because for the most part the field (if we admit that it is a field) remains hostage to the vagaries of the interactions between cells and culture.

Figs for Paris.003If we agree that the solution is to engineer the process, we need to use the Physics of the system, which, like in any engineering process, underpins the events. This means that in our attempt to engage cells into building tissues and organs we need to engage developmental biologist. If stem cells are the materials, developmental biology is to organoids what Physics is to Engineering, and we need to use it like that. And by Developmental Biology I do not simply mean the garden variety that dabbles in genes and cells but the more quantitative systems one that strives to integrate the rapidly emerging data into models that identifies parameters and tells us how the different variables interact. Just like Evolution tinkers, we shall tinker. At this, as Peter Zandstra exhorted us at the meeting, we need to use what we know in a realistic manner. Build models that contain real dimensions and time scales and use them, like engineers do, to improve, to build, to understand.

Organoids, in its many varieties (embryonic and adult stem cells, micropatterned cell arrays, scaffolding of cells from different tissues) promise much but we need to accept that the process is more important than the end, that understanding their beginnings and how they unfold will allow us to understand and improve the end product. One intriguing feature emerging from the current studies and that we shall have to address, concerns the differences between the events that we see in the dish and what we observe in an embryo (see refs 5 and 11 for discussions). It is too early to say whether these differences are important but, it might be that as in engineering there are many ways to make a bridge or a jet or a computer, the same might apply to a tissue and an organ; to put it clearly, there are many molecular solutions to an organ or a tissue and the embryo uses one but we might be able to use others that are simpler. Much interesting and unknown lies ahead but one of the most exciting prospects is the possibility to explore human developmental biology and, in the end, to create a proper organ and tissue engineering as a first step for a regenerative medicine. But we should not cut corners.

Much of this was discussed on and off stage at the Pasteur meeting which had a sense of first encounter between engineers, developmental and stem cell biologists and of expectation of what can be achieved by working together. There will be a report of the meeting in the Development journal, but importantly interesting things will happen. Stay tuned.

1.         Eiraku, M. et al. Self-organized formation of polarized cortical tissues from ESCs and its active manipulation by extrinsic signals. Cell Stem Cell 3, 519-32 (2008).

2.         Eiraku, M. et al. Self-organizing optic-cup morphogenesis in three-dimensional culture. Nature 472, 51-6 (2011).

3.         Sato, T. et al. Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. Nature 459, 262-5 (2009).

4.         Hurtado, C. & De Robertis, E.M. Neural induction in the absence of organizer in salamanders is mediated by MAPK. Dev Biol 307, 282-9 (2007).

5.         Sasai, Y., Eiraku, M. & Suga, H. In vitro organogenesis in three dimensions: self-organising stem cells. Development 139, 4111-21 (2012).

6.         Green, J.B., Dominguez, I. & Davidson, L.A. Self-organization of vertebrate mesoderm based on simple boundary conditions. Dev Dyn 231, 576-81 (2004).

7.         Lancaster, M.A. & Knoblich, J.A. Generation of cerebral organoids from human pluripotent stem cells. Nat Protoc 9, 2329-40 (2014).

8.         Qian, X. et al. Brain-Region-Specific Organoids Using Mini-bioreactors for Modeling ZIKV Exposure. Cell 165, 1238-54 (2016).

9.         Xu, M. et al. Identification of small-molecule inhibitors of Zika virus infection and induced neural cell death via a drug repurposing screen. Nat Med 22, 1101-1107 (2016).

10.       Gjorevski, N. et al. Designer matrices for intestinal stem cell and organoid culture. Nature 539, 560-564 (2016).

11.       Turner, D.A., Baillie-Johnson, P. & Martinez Arias, A. Organoids and the genetically encoded self-assembly of embryonic stem cells. Bioessays 38, 181-91 (2016).

The Launch of the Human Cell Atlas Project: A Catalogue of Trillions of Cells Without a Question?

A catalogue is not a map and if you want to transform a catalogue into a map, you probably need to understand what the map is for.

The launch of the Human Cell Atlas Project (https://www.broadinstitute.org/research-highlights-human-cell-atlas) has caught the eyes and the ears of newspapers, the mind of some scientists and, obviously, the imagination of the leaders of the project. If you see the headlines, it might tempt you. After all, the prospect and challenge of learning the secrets of 35 trillion cells (3.5 x 10^12) is an impressive undertaking. The number reminds us of Avogadro’s number and, of course, there is the challenge and the excitement for the technologists and the computer scientists associated with such large numbers. There is a lot of money and human power at stake and, in so far as I can tell, very little forethought. Yes, there are the challenges of the logistics behind such projects but, one can think of many projects that aim to collect and organize data from biological systems but do not make it to the headlines. This one, I suspect, builds on an ongoing trend for genomic data gathering, organization and analysis inspired by our technological developments and also on the fact that, when it comes to data, Biology rules because biological data are about the generation, control, selection and maintenance of variability at different scales from molecular to organismal. This does not mean that all in it is meaningful and this, I think, is the main oversight of a project which has more cloth than substance. One has often the feeling that biologists are searching for their own Manhattan project. First it was the human genome (then any genome and many genomes), the connectome (or connectomes), then in a more quiet manner there is the human proteome (it is more complicated) and when we are getting tired of –omes, we begin to talk about Atlases.

Don’t get me wrong, there is something useful here but, are there not better investments of talent and money in the same or related fields at this point in time? I have followed the single cell field from its emergence and have recently commented on its current status (http://amapress.gen.cam.ac.uk/?p=1765) so will not repeat myself since much of what I said there applies to this project. However, let me point out the surprising naïvité behind this project and the lack of understanding of what maps, Atlases if you want to give them a name, mean in Biology. The idea of this project is to obtain information about the transcriptome of individual cells and use this as a reference base for……what? What is the map for? We are told that “Without maps of different cell types, where they are located in the body, and the genes they express, we cannot describe all cellular activities and understand the biological networks that direct them” and, with the usual references to health and medicine, that “A cell atlas has the potential to transform our approach to biomedicine. It would help identify markers and signatures for different diseases, uncover new targets for therapeutic intervention, and provide a direct view of human biology in vivo, removing the distorting aspects of cell culture”. At first sight nothing wrong with this, and a lot of promise, except that we already have experience of similar projects at smaller scale and, when you look at them, these projects pose many questions which we have not answered yet. And many of those questions have a common denominator: what kind of questions can one ask to such data beyond cataloguing, organizing? Forget the hype, what is this for? What is this about? If there are no clear answers to these questions, why going bigger blindly?

A significant problem with this idea of cataloguing (a catalogue is not a map) trillions of cells into classes based on their identities lies in what we already have learnt from similar smaller scale experiments: populations of cells in a similar state exhibit large heterogeneities in gene expression which bedevil their identity (however we define it), that will vary between individuals and certainly with age. We use these heterogeneities to classify but we don’t understand their meaning. So, which age shall we look at? Which individuals? DNA is constant, RNA is not. Furthermore, at a technical level we are still not sure how deep we have to go into the sequence analysis of the cells to get meaningful information, we still do not know how to read this information. Most of the projects already out there seem to be more about sequencing more cells, faster, deeper, about new algorithms to organize the data, about data visualization than about Biology. The Biology tends to be left out, as if the structure of the data will yield the Biology for free. Sure there are biologists in most projects. In several years of maps at a smaller scale than what is proposed here, I have not seen much of ‘sense’ in this field and the questions that it poses are left unanswered or, worst, unuttered. What is more worrying, in a field like the hematopoietic system, which appears to be a test bed for this field, there are two or three papers a month with little cross referencing and few general messages beyond slight reshuffles of the maps that years of classical Biology had produced and, occasionally, a few new markers for cell types. And yet, I am sure that there is much that can be learnt from a proper look at the data which, in turn, would determine how we gather the data. There is something to ponder here, in particular on question of what do we want these catalogues for?

By the mid 1800s physicists knew that the macroscopic variables of a physical system depended on the atoms and molecules that made it up. So, for example, the Temperature and the Pressure of a gas were/are a consequence of the velocities and relative positions of the constituent molecules. The question was how to relate the two. If they had had much of today’s technology, they (some) might have opted for measuring the momentum and position of every molecule and through lengthy calculations work out the answer. There are about 10 trillion molecules in a nMole of a gas so, you can see the headline: Physicists engage into a Molecular Atlas that will transform physics and engineering: Trillions of molecules to be mapped in terms of their velocities and relative positions. Furthermore, we would be told that this will be done for a couple of the fundamental molecules in the Universe thus ushering in a new era of technology. This, of course, does not make much sense and it was the prescience of JC Maxwell and, in particular, L Boltzmann that showed the way to deal with the problem. In this manner they created Statistical Physics which in addition to producing very exciting Science did transform Physics and Engineering. Now, Biology of course is not Physics, and the identity of a cell is not the state of a molecule but there are analogies that can be used in the analysis of transcriptomes and maybe we should engage deeper with those analogies rather than going for the trivial. It is possible that what we want to know is not the ‘identity’ of every cell (though I would argue that our view of this is, still, primitive, superficial and inaccurate) but what those individual identities average to, what is it that is being read at the higher level of organization by the cells. After all, experiments (experiments, not cataloguing exercises) tell us that the heterogeneities that are observed in the analysis of single cells in populations are dynamic, though we do know or understand what the meaning of those heterogeneities are. This observation, alone, suggests that if you want to transform a catalogue into a map, you probably need to understand the meaning of these dynamic heterogeneities. Moreover, for all we know (or ignore) we still do not understand what the macro-variables that cells measure are, what is being represented in those heterogeneities. Doing a map of the trillions of cells of an organism (this being human cells only to reinforce an anthropocentric view which while justifiable is narrow minded) without knowing what these are related to is not that useful or clever. And if you want to compare this to the Human Genome Project, just remember one thing, the genome is the same in every cell and, for the most part in every species, it is this conservation that make SNPs and polymorphisms useful. However, with transcriptomes there will be surprises and we need to think what we want to measure before we go too far.

James Briscoe has written about the semi-comic statement of Microsoft to ‘solve’ cancer within ten years and pointed out how biological systems have this habit of ‘fighting against anything we try to do” (https://briscoelab.org/2016/10/02/the-three-billion-dollar-question/). Nowhere is this best exemplified than in the single cell transcriptomics field, where the intrinsic tendency to generate heterogeneities, can fox the best experimental designs. We still have much to learn.

One suspects that, if and when this project gets under way it will become one of those ‘too big too fail’ exercises that are becoming frequent in Biology. Projects which take funding away from hypothesis-driven or hypothesis-seeking projects which might provide a good focus for these accounting exercises. I would argue that, so far, the field of single cell analysis of transcriptomes has yielded some information but little insight, it has revealed the presence of heterogeneities in expression which are dynamic and pose some questions which are not often discussed but which need to be addressed before going too far, perhaps, in the wrong direction. We may take a page from the history of Physics and think that, perhaps, we should understand Pressure, Volume, Temperature and their relationships, before we attempt to deduce them from the molecules which, probably, underlie them. Because, as I have said before, a catalogue is not a map and their purposes are very different: a map helps you navigate, a catalogue puts order into a collection of items but does not in principle, have a defined purpose.

In the end the project will generate data and the brief of the Broad Institute (https://www.broadinstitute.org/news/international-human-cell-atlas-initiative-gets-underway) contains some interesting statements: “By making the Atlas freely available to scientists all over the world, scientists hope to transform research into our understanding of human development and the progression of diseases such as asthma, Alzheimer’s disease and cancer. In the future, the reference map could also point the way to new diagnostic tools and treatments”. We have heard that before, haven’t we?

I still miss the Biology in the project. If you want a big project, might it not be better to think and apply the current technology to a system or a problem which can teach us how to approach, one day, a human cell Atlas?

Of codes and machines in Biology I; elements for a discussion

Picture1

NB. As I was writing this post, a couple of comments came up on Twitter on whether the machine metaphor was a useful one for biological systems. The discussion did not change the view that is expressed here: a machine is a good metaphor and one that, to a large extent, remains untested.

a

Picture2

I am no fan of science fiction but there is a novel which I have always liked –for the philosophical background more than the plot- and that has come to my mind when recently thinking about genes and cells: Carl Sagan’s ‘Contact’. The key element of the story is the reception of an extraterrestrial message by scientists which, after some toiling, is revealed to contain instructions for building a machine. The instructions do not contain any hint of what the machine is for and only vague ideas of how its detailed shape will look like; both emerge as it is built. The story has many twists and turns, deals with some of the preoccupations of Sagan and, in some ways, the recent ‘Interstellar’ touches in a more saccharine manner, with some of the same issues. But the bit that has grabbed my attention over the last few weeks has been the notion of having a set of instructions without a clear picture of what they are for (self assemble furniture instructions feel like this sometimes) that, when followed, generate a machine and only when the machine emerges, one can think of what it is for.

The reason I was thinking about ‘Contact’ has to do with a preoccupation to figure out what it is that we want to know. I am well aware of things that we need to know, but given that there is always limited time, one needs to decide what it is that one would like to know; understanding is ambitious and, probably, out of reach. Modern cell and molecular biology have placed in front of us a formidable technical arsenal which, in principle, allows us to explore any question we may have. As I have hinted at before, it is unfortunate that a collusion of editorial and career interest are giving precedence to classifications and listings, sometimes one at a time, over real questions but…….. it is also true that questions are difficult to find and, more importantly, to answer. And it is in trying to understand what are important tractable questions, that something caught my eye, something which I am sure some of you either know or have thought about in a different manner. The problem of Developmental Biology is how the information in the DNA is decoded and transformed into the tissues and organs that configure an organism. Most seminars and reviews on the subject start like this only to then proceed with genetic or molecular screens. Nonetheless, as a consequence of this work, classical and popular genetics have created the mirage that there are genes for this and that i.e. that the DNA harbours instructions for digits, eyes, hair colour or height and, in the worst twist often promoted by newspapers, that there are genes for diseases. This is the basis of our current understanding of how the information in the DNA is interpreted and turn into an organism but……. if one thinks about it and thinks from first principles, genes are ONLY instructions to build machines (M in the figure below): ribosomes, transcription and replication enabling machines, membranes, cytoskeletal devices, etc….their remit does not go beyond this. These machines, once built (through the central dogma), become assembled, much like a 747 or a transatlantic liner, into a larger product, a device with the capability to process information, react to it, do work, sense the environment. This device is what we call A CELL. In multicellular animals, subtle variations in the composition and performance of the component machines, lead to different cells that can be further assembled into tissues and organs and these into an organism. Thus, don’t forget that the instruction in the DNA do not code for much that is 3D, and certainly lack any information about the function of the machines or the devices they code for. Like the extraterrestrial message in Sagan’s novel, the instructions are for a machine whose purpose only becomes clear once it is built.

Picture3

Fig1. Top, A general principle whereby functional interactions between machines built by instructions from a blueprint, configure a device that works according to the laws of Physics and Chemistry. Bottom, a translation of the principle to a biological system.

Something interesting happens once the machines and their suprastructures, the cells, are assembled: space arises (the information in the DNA does not contain, convey or encode space, maybe time, but not space). As a consequence of the generation of space (surfaces and volumes), mechanics makes an appearance and it does so at two levels: the molecular one (the machines) and the cellular one (the device).  Thus, the emergence of space leads to the emergence of mechanics which feeds-back on the processes of decoding and assembly and, in more than one way, tissues and organs and ultimately organisms, are the outcome of these feedbacks and the interactions they create. The decider of what the output of these interactions should be is function, as dictated in a blind manner by Natural Selection.

Picture4

Fig 2. Tissues and organs do  not arise from instructions but result from the feedbacks that the emergent properties of the performance of the machines assembled in cells. More explicitly from the feedbacks that generate new activities and behaviours in those machines and cells.

Much has been made of the selfish gene hypothesis (and I hasten to add that I am not an evolutionary biologist and know little about evolution) which suggests that the machines and the devices have the aim of reproducing the genes which encode them. I find this naïve, deterministic and anthropocentric. There might me no purpose in the assembly of those machines, maybe all is a game of molecular flaunting and selection is not simply looking at the replication of the genes but at the performance of the machines (I realize that I shall need to expand this here but, like Fermat, for that I need more space than the one I can afford here). Perhaps it is a widespread unconscious focus on the genes that for a long time has placed our emphasis on the decoding, which misses many of the important points and problems the need to be addressed. In this regard, it is interesting that while cell biologists have focused on the machines and, sometimes, how they are connected, developmental biologist have focused on the decoding in a kind of naïve manner: a gene for each season. I would surmise that the important problems in developmental biology require, at least for now, thinking about the machines, the devices they configure and their functioning rather than in their component parts. Only in this manner we shall be able to go beyond genes/instructions and start to look at tissues and organs from the same perspective as Nature does: the ability of the cell to integrate and process information. Work on the decoding of gradients in developmental systems (James Briscoe, Marcos Gonzalez Gaitan, Thomas Gregor, Johannes Jaegger, and Arthur Lander) or of oscillators during somitogenesis (Alexander Auhlehla and Andrew Oates) do just that and, to my mind, are some good examples of the way forward, of how to integrate quantitative cell biology, hypothesis testing and modelling. These efforts gauge the information processing ability of cells rather than in the micromanaged organization that most molecular biologist do. Perhaps phage and E. coli supervise closely every step of their biology but, while not impossible, this is not probably the way multicellular developmental systems operate. They are likely to use coarse graining, space and time averaging of molecular events, system level strategies that we can easily miss by focusing on the details of the molecular events.

Much of what we have done so far to understand higher order organization of biological systems (and there is work to do) deals with temporal aspects of the system: simple gene regulatory networks that try to capture sequences of gene expression linking them to specific cellular events. The challenge comes when we face the nature and details of the feedback that space creates on the decoding, the assembly and performance of the machines and their coming together into the devices that we call cells. It is the feedbacks involved in these processes that we need to understand and represent. Turing when thinking about the problem of development in his 1952 paper, was well aware of the challenge when he wrote:

“……one proceeds as with a physical theory and defines an entity called ‘the state of the system’. One then describes how that state is to be determined from the state at a moment very shortly before. With either model the description of the state consists of two parts, the mechanical and the chemical. The mechanical part of the state describes the positions, masses, velocities and elastic properties of the cells, and the forces between them.(….) The chemical part of the state is given (in the cell form of theory) as the chemical composition of each separate cell; the diffusibility of each substance between each two adjacent cells rnust also be given….(…). The interdependence of the chemical and mechanical data adds enormously to the difficulty (of understanting the state of the system) and attention will therefore be confined, so far as is possible, to cases where these can be separated”

This is a pre-molecular way of looking at the problem but remember that Turing was trying to think about Development and doing this in a very prescient manner. The interdepence he refers to is the essence of the feedbacks mentioned above which turn cells into tissues and organs. Along similar lines, a few years ago F. Julicher pointed out something to me which highlights the essence of the problem and, at the same time, the difference between Physics and Biology. It can be stated simply: whereas in a physical system, Chemistry generates mechanics and the reverse is not true, in biological systems both work (Physics generates Chemistry and Chemistry generates Physics) and it is in this feedback and interactions that, probably, lies the essence of biological systems.

The current focus and craze on screens and single cell transcriptomics (proteomics is coming) should not make us forget that these endeavours only address the parts. Moreover, as I have discussed before (http://amapress.gen.cam.ac.uk/?p=1094) the system we are trying to understand has evolved to respond to selection (and remember that this is what screens are about, a highly selective selection of the ability of cells to respond to a stimulus) and the response is based on dynamic heterogeneities that we still do not understand but which, one suspects, have something to do with the feedbacks I have outlined. The exciting thing (to me) in ‘Contact’ is the machine, what it does and how both its structure and purpose unfold as the system is built. The parts that configure the system could have been used to build anything but the instructions turn them into a space-time machine. In Biology, right now there is too much emphasis on the parts. This would not be a bad thing were it not because we endow parts with functions that correspond to the wholes they are part of. How do we avoid this? The real aim of Systems Biology is to avoid falling into this trap. It is therefore unfortunate that in most realms, Systems Biology has become a proxy for data analysis. Cell and Developmental biologists should embrace a proper version of Systems Biology because right now it is the only way to get out of the describe-the-parts loop and move towards understanding on the way to re-engineer cells and tissues.

Acknowledgement: I want to thank F. Julicher and S. Grill for very enlightening and inspiring discussions on the subject of this post. Also, note that there is a I in the title. There will be a II which aims to deal with ‘questions in Biology’.

Epilogue: The Twitter feed I mentioned above led to many interesting suggestions. They all described different kinds of machines. Even the notion of City, much liked in the thread, can be construed as a machine. What this reflects is that machines are ways to describe (and then engineer) assemblies of parts with some functional aim. It is difficult not to see this in the fabric of a cell and, in many ways, it is a useful, working notion. If properly used it could be helpful as in its day was the vision of the central dogma, as information transfer and decoding.

Expensive or Insightful Biology?: Single Cell Analysis as a Symptom

160727_800px-Musei_Wormiani_HistoriaLists, catalogues and classifications have always been the business of the biological sciences. The nature cabinets of the XVII and XVIII centuries, the collections that occupied much of the XIX century and which fuelled the work of Darwin are good examples of this. Beetles, butterflies, fish, pigeons, plants occupied (and occupy) the time of individuals, often amateurs, interested in Nature. The nature of this enterprise is captured in Umberto Eco’s book “The Infinity of Lists”

When we don’t know the boundaries of what we want to portray, when we don’t know how many things we are talking about (….) when we cannot provide a definition by essence of something and so, to be able to talk about it, to make it comprehensible or in some way perceivable, we list its properties (…………….). We call this representative mode the list, or the catalogue

Indeed: to make something whose limits or meaning we ignore, we make lists, if they are organized according to some criterion (and since Linnaeus but even before, they are) they have the potential to reveal something of the essence of that which is being classified. Physicists and chemists know well how this works: stars, spectra and the elements come to mind. But the level and intricacy of what the biological world offers to the catalogue aficionado is different, probably, boundless. To go back to Eco, it is unclear where the limits of the biological world lie. No wonder E. Rutherford said that one could reduce the sciences to Physics and stamp collecting; he may have had the biological (then natural) sciences in mind and this perhaps is why S. Brenner famously retorted that what Rutherford did not know is that there are some stamps that are worth collecting.

In the History of Science lists have the potential to highlight generalities which allow precise questions to be asked and answered. Physics and Chemistry have been good at reaping the benefits of this activity. In Biology a most famous outcome of this cataloguing is, of course, Darwin’s great work which revealed a principle running through the continuum of transformations that stares from large collections, ordered collections (the word ordered and in what manner the order comes about being important here), of plants and animals. In a different way, the work of Mendel is a culmination of less structured but no less significant collection of lists of the output of many lists; after all, it is seeing patterns in the outcome of crosses of plants that leads to genetics. In all cases the assumption is that if the lists are arranged according to the right criterion, they will reveal an order and, behind that order, some mechanism -in the sense of a causal explanation for a set of observations and not as the usual Figure 7 characteristic of modern biology papers- that will provide an insight into a system. In the end, sometimes, the insight can lead to the manipulation of the system for the benefit of the observer: lists lead to science that leads to engineering which leads to progress.

There is a danger in these lists and it is that they might become an end of themselves. That the scientist becomes a collectionist, forgets Brenner and gives credence to Rutherford. Surely the lists are valuable resources for those that want to ask questions, but the truth is that as we turn into list makers, we can forget that there are questions behind the observations and habit turns us lazy and content in our collecting. Sometimes one feels that this is happening in the biological sciences, that biologists are becoming professional collectionists. There might be a reason for this:  the essence of biological systems is the generation, selection and competitive propagation of novelty and variation. As a result, every species, every genome, every cell in every genome, every organelles in the cell, every protein in the organelle, is subject to this continuous generation of variation, to the exploration of a large space of form and functions. If one assumes that every cell type in an organism is different and that these differences are species specific, one can do a simple calculation: the range of different cell types varies between 3 in a plachozoan to about 1014 in a human and if, as it is currently assumed, there are on the order of 8.5 x 106 organisms on the earth, one could say that there might be on the order of 1020 different cell types to explain (NB this is assuming that all individuals within a species are similar and forfeiting the development of an organism during which large numbers of transient cell types are generated that differ from their final types). This number, 1020 , is already a large number relative to the approximate number of stars, 1012 . It may be small relative to number of atoms, 1080 in the Universe –and one has to remember that atoms need to be proportionately distributed into 117 elements which is where the differences appear i.e all atoms of an element are essentially the same and thus, the 1080 number needs to be tempered by its being bundled up in the abundance of each element. It is here, in this notion of similarity of all the atoms of an element, that the main difference between the biological and physical systems appear. The stars are very similar to each other in composition, and this is why we can study them from a distance by using the spectra. On the other hand, every organism, every cell type in every organism is different, unique. In fact you and I are very similar but our cells in similar places in similar organs are likely to be different. Enter DNA, which is the way to explain uniqueness in Biology: if we accept, as we must, that every cell type responds to a ‘transcriptional code’ of sorts, and we focus just in humans with our approximate 20,000 genes (I am not interested in philosophical discussions of what is a gene and hope that you and I will agree that this is a lower bound), simple calculations allow for 220000 combinations, to account for those 1014 different cells (and don’t forget those developmental intermediates). If you throw this number into your calculator, it will be confused as it will approach infinity. Of course, the toilings of Natural Selection ensure that only part of that repertoire is used but still, the number is large and dwarfs anything the inorganic world can produce. Surely we are stardust, like the moons of Jupiter, but DNA and RNA have found a way to turn that dust into a creative material device.

Where am I going with this? Over the last few years technical developments have allowed us to peer into single cells at the level of their transcriptional complement and, with increasing accuracy, at the level of their genomes. The observation is that even within what histologically is a (one) cell type, there is a great deal of heterogeneity. It is difficult to silence the genome, and we are learning that cells –particularly in development- are exploring their transcriptional space in a dynamic manner. The result is that within an organism much of that space of 220000 combinations is likely to be explored and much of it represented. The technical developments are allowing increasing volume and accuracy in the observation of this process (gene expression at the level of single cells) and of the delivery of these results. In consequence this creates interesting challenges for classifying, for making lists, which are taken on by groups of computational biologists whose interests lie in dealing with complexity rather than in understanding its meaning. Meetings are held on the subject of gene expression at the level of single cells and while at the moment the possibilities lie in honing our ability to describe the expression patterns of single cells and of characterizing the genomes of cells in tumours, the holy grail on the horizon is the analysis of epigenetic marks at the level of single cells and the ambition of getting the genome, epigenome, transcriptome and proteome in single cells. Our infatuation with these techniques, what it reveals and the possibilities associated with it are powerful and thus reviewers and editors lurk in the background to ask you for a single cell analysis of your favourite system, if everything else has failed to hamper the publication of your work. But, at the moment, it is also expensive and begs the question of where does it lead to? What is the meaning of this work? Are we paying lip service to Rutherford?

160727_Untitled.001The analysis of single cell gene expression can have -and sometimes had- an impact in three areas of Biological research: Cancer Biology, Immunology and my area of interest, Developmental Biology, which aims to understand how an organism builds itself. In all cases, single cell analysis allows the identification of ‘rare cells’ which sometimes have a function and sometimes, they don’t. The issue is that more often than not and in the best tradition of Biology, these studies reveal the temptation of collecting data under the banner of its ‘importance’ without realizing that we have fallen to a fad, that cataloguing has taken precedence over understanding. The description of a biological process demands a link between a cellular and a genetic description of the process and there is little doubt that the arrival of single cell transcriptomics and associated techniques, particularly single cell lineage tracing, has revolutionized the field. However we should be careful not to be swayed by the collectionist syndrome and remember that behind the data there are questions and that if we cannot see them, we should acknowledge that. We should not confuse cataloguing and collecting with Science. In some ways there is no great difference between beetles and genes, and we might be developing a XXI cabinet of genes and cells. It might require more challenging techniques than those collections of the past but there is no difference between collecting one or the other. Already papers in journals tend to be divided into two: either analysis of gene X in tissue Y in organism Z, or increasingly, single cell analysis of process W in organism Z. And in the best tradition of classical Systems Biology, one hopes that in the analysis of the data, the question and the answer will emerge at the same time as one stares in hope at the data.

Single cell analysis of expression is the epitome of this strange hypothesis-free science that is often hailed in reviews and social media. We are in the midst of it. Slowly we fool ourselves that large data and cataloguing will lead us to the essence of a process, that it will allow us to talk about something that we cannot define. And while it is true that Biology has a habit of revealing principles from lists I cannot help but thinking that with this trend of hoarding data, we are losing perspective of the processes that still need addressing. It would be good if, as R. Feynman said, we don’t confuse naming something with understanding something. Developmental Biology in particular, is losing itself in this naming game and single cell analysis will –unless checked- provide the ultimate distraction from questions that are there but we are too…..may I say ‘lazy’? to ask. We should not forget that there are things to explain, that cataloguing is a way to answer, and sometimes to unlock, those ‘things’ but also that we need to make an effort to search for them.

The allure of the information that can be gathered in one of those experiments is enormous but one needs to remember that in addition to being expensive and data rich, it needs to be insightful. The difference between a collectionist and a scientist should lie not in the ability to make observations but in the ability of the second to use the observations to answer specific questions about Nature. Biological systems have a boundless ability to generate (constrained) variability and it seems to me that the challenge is to understand the nature of the machine –for it is a machine- that generates, processes and uses that variability, written in that tape that is the DNA, interpreted by the transcriptional machinery and supervised by Natural Selection. It is the process, not its output, that needs to be explained. Questions are cheaper than data gathering but good questions are hard to come by.

Epilogue

One of the most disturbing aspects of the current trend in the single cell field is the lack of cross reference or discussion of the data. Often the same system is surveyed in more than one paper without any reference to the other, related, pieces of work or even, on occasion, to the general problem. While this is in keeping with the current trend in the biological sciences in which the publication rather than the finding is what matters, it is no less disturbing. If we do not get hold of the boundless nature of that data by using questions to clean it up and thus reveal what is good and bad data, we shall do a disservice to the system that puts up the money for that research and, more importantly, to Science itself. Surely, there is meta-analysis, Darwin’s great work can be construed as a meta-analysis- but nowadays, often this is done not so much with a question in mind but with the idea of multiplying the data-analytical power. The boast tends to be not in what has been learnt but in how large the data set is. And in the end, the danger is that, increasingly, what we do is expensive collecting; XXI century cabinets of genomic data, without a good reason, without a good question –which exists. We seem to have relinquished our ability to interpret what we observe and lost our interest in asking questions because, I agree, it is easier to order and catalogue this diversity that we call Biology. Still, the issue buzzes in the back of my mind: there are questions, important questions, to be asked and….all that data!

Boltzmann, Darwin and THE current challenge of the life sciences

Boltzmann2

Ludwig Boltzmann 1844-1906 (http://en.wikipedia.org/wiki/Ludwig_Boltzmann)

The XIX century will be called the century of Darwin (L. Boltzmann)

While most people have heard of Einstein and Newton and Feynman, Boltzmann is not a household name when thinking about famous physicists. Ludwig Boltzmann was a theoretical physicist extraordinaire who at the end of the XIX century, in that Vienna that was going to give so much to the world in the ensuing years, taught us a most interesting way of thinking in material terms about the structure of matter and abstract concepts like heat and energy. Spurred by his philosophical inclinations, in his latter years he wanted to transcend what he had done and thought, by looking at Evolution from the physical perspective. In this process he clearly absorbed much of Darwin at a time that darwinism was not as popular as it would become later: “… If you would ask me about my heartfelt conviction, whether the nineteenth century will be called one day the iron century or the century of the steam engine or the century of the electricity, I answered without any doubt it will be called the century of the mechanistic conception of nature, the century of Darwin…”. There is little doubt from this statement that Boltzmann understood Darwin but there is also an inkling, if you know something about the work of each of these individuals, that he might have had a deeper insight than he let us know in his writings.

Physics and Biology share one challenge: the mechanistic understanding of the relationship between events that happen at the limit of our visual detection –the microscopic world- and what we can observe and sense i.e. measure (any act of perception is a more or less conscious measurement) at the macroscopic level. The way we do this is nicely put in a statement attributed to the physicist Jean Perrin, which suggests that one of the cornerstones of Science is the craft of revealing the invisible through the visible. In some respects this is what we do in Biology when we draw those diagrams that are meant to represent events supposed to happen inside cells. While some of them are probably accurate (and for accuracy on the basis of our current understanding of our molecular structural knowledge, see D. Goodsell visions of the cell: http://mgl.scripps.edu/people/goodsell/) others do not capture, yet, what they want to represent. And so, there is a two way road from the macroscopic to the microscopic. A topic of many talks in Biology is, we are told, that what we want to know is the relationship between the genotype and the phenotype, between the genes and the cell. However, behind this statement there is the dream of some sort of a linear relationship between both which has not and will not be found because 1) it does not exist and 2) this might not be the right question to ask. If you are an evolutionary biologist you spend a great deal of time relating genes to the structure of populations and therefore you know about the problems of simple linear models and of the slippery nature of quantifiable variables which are sometimes needed to deal with biological systems. However, it is precisely in the challenge of relating genes to, for the sake of argument let us say phenotypes, that the connection between Boltzmann and Darwin emerges and might provide some inspiration for today’s challenges.

Figs for Blog.002A

Figure 1. One of the big challenges in Biology is how to relate the events that are described by molecular networks with the organs and tissues that characterize the make up of an organism.  It is obvious that cells and their lineages are the vehicles for this transformation.

The breakthrough of Boltzmann stemmed from his belief in the reality of atoms and their fundamental role in the understanding of physical systems. A belief it had to be since at that time it was impossible to penetrate the structure of a cell, let alone that of a molecule or an atom. Taking this view as a starting point, he developed a theory which provided a mechanistic explanation (watch it, not in the sense of the modern biologists i.e. figure 7 of your NSC paper, but rather, to quote my colleague Ben Simons, as a causal explanation for an observation) for observables like Pressure, Temperature or Energy. He showed how if one accepted the existence of atoms, one could derive these properties from the spatially constrained interactions between them. Since the number of molecules in a macroscopic observable is enormous (remember Avogradro’s number is 6.02 X10 23 molecules in a mole), even those who were interested in the subject, found it very difficult to comprehend how could one devise a mechanistic and mechanical way to connect these large numbers to the observables. If you were a committed newtonian you would have to calculate the trajectories and energies of every atom and its interactions with all the other atoms and then find a way to compute the total sum (or product) of the resulting numbers! The way forward, as Boltzmann saw, was assuming the reality of the atomic structure of matter, to perform a proper statistical analysis of the behaviour of ensembles of molecules in different conditions. He reckoned that with such large numbers, the connection between the elements and the properties of the system was through statistics –in its infancy at the time- and that under the simple conditions of an ideal gas, a statistical treatment of the kinetic relationships between individuals in populations of molecules (microscopic) would yield the macroscopic measurable (Pressure, Temperature, Kinetic Energy…); a proper treatment of the problem shows how the observables result from the constrained averaging of the individual variables. It was a deep insight that what mattered were the statistical properties of the population rather than the details of the individual behaviours which became averaged at the higher level. This work provided a solid foundation for the work of the Scottish physicist JC Maxwell who had calculated the distributions of velocities of an ideal gas on similar terms, thus laying a significant foundation for the kinetic theory of gases -this is why today we talk about the distributions of velocities and energies in physical systems as the Maxwell-Boltzmann distribution. But Boltzmann took the basic ideas of a statistical analysis of the structure of matter further and provided a material basis for that most elusive notion: Entropy (which in thermodynamic terms can be defined as the amount of energy, thermal energy, which is not available to do mechanical work). With apologies to the physicists (if any reads this) for the simplification, he envisioned matter as a problem in combinatorials of its constituents: a particular structure being one, and only one, of a huge number of configurations of its constituent elements. If that structure disappears, or changes, it means the system has acquired a new configuration and will search for the original one in the large space of all the other configurations. Not surprisingly it will find many ‘disorderered’ ones before finding the original one. Entropy, Boltzmann saw, is a measure of that number of non-structured configurations. He extrapolates this to the Universe and suggests Life as the chance result of a fluctuation in a small space of a large heat bath. It is these thoughts about the Evolution of physical systems that probably led him to consider darwininan concepts: “… The struggle for existence of the living beings is not a fight for basic materials—these materials are available in air, water and soil in sufficient quantities for all organisms—it is also not a fight for energy that is available in the form of inconvertible heat in every body but it is a fight for [negative] entropy, which becomes available by the transition of energy from the hot Sun to the cold Earth. In order to exploit this transition as much as possible, the plants spread out the incredibly large surface of the leaves and force the energy of the Sun before it falls down to the temperature of the Earth in a not yet understood way to perform synthetic chemical reactions that are still completely unknown in our laboratories. ..”. Much food for thought here and I shall leave it for another time. Suffice to say that the deep gauntlet that lies in here was taken later by E Schrodinger who in his famous book “What is Life” discussed at length some of these notions and introduced the eye catching but misleading notion of negative entropy, free energy really (Gibbs or Helmholtz); he might have been influenced by his youth in Vienna studying Physics under the aura or the great Boltzmann.

Figs for Blog.002B 

Figure 2. Boltzmann’s insights that allowed him to use statistics of the mechanical properties of the particles under several constrains to deduce the macroscopic properties of the system. In the process he provided a physical description of Entropy (S) in terms of the configurations of the system (W).

What does this have to do with where we are at the moment? What is the point of all this to modern Biology? The current challenge, as some of us perceive it, is not to see how genes generate a phenotype but to link the molecular and the cellular realms. To explain cellular activities (motility, change of fate, higher order structure and dynamics of cell populations, etc) in terms of their molecular underpinning. In all this and what has become a game changer is our ability to measure or, if you will, to see and then to measure, and to be able to do this at the level of individual cells. What we are getting out of this process is large amounts of data, information, that we are accumulating in databases that are more or less centralized and organized. What we are lacking is not just methods to process this information, but questions, conceptual frameworks to interpret what the analysis of the data (which is more data) yields. The question then can be reduced to how the myriads of genes, proteins and their interactions at one level, generate behaviours at a different scale. How do the macromolecular complexes that underpin cell movement  and shape, the structure of a tissue or the dynamics of a tissue in homeostasis, generate those observables?. In this work, there are two connected relations: from the molecules to the cell and then from the cell(s) to the tissue. This statement contains the implicit statement that THE CELL is a vehicle to link molecules to tissues and organs. The numbers of the game are very large (genes, transcripts, cells) and become larger if we consider single cells, which is becoming routine. It is here that the work of Boltzmann becomes an inspiration. The secret will be the averaging and the way biological systems do what physicists call coarse graining, will provide the understanding; but first we need to define the variables that need to be averaged and the calculations that need to be made. Progress is being made but it is slow because, unfortunately, the emphasis is still in mindless data collection and on the naïve belief that describing it is understanding.

It was probably this deep insight into the population averaging of the properties of very large number of components of a system that led Boltzmann to have an intuitive understanding of Darwin. After all, the importance of large numbers and their dynamics is implicit in Darwin’s theory of natural selection and becomes explicit in the postdarwininan interpretation as in the work is R. Fisher, S. Wright and others, genes play the role of the atoms, and statistics is not just central, but develops around these ideas. Qualities, phenotypes, arise from the multivariate statistics of the effects of multiple genes. It is interesting, as has been discussed by J Gunawardena that much Genetics was developed without an understanding of the molecular structure of the gene and that for many years, the gene was a mathematical entity  (Biology is more theoretical than Physics, Mol Biol Cell. 2013 Jun;24(12):1827-9).

Figs for Blog.001

Figure 3. Outline for a statistical mechanics inspired solution to the problem (for further thoughts see references at the end). At the more microscopic level there are Gene Regulatory Networks (GRN) which generate dynamic (revolving arrows) patterns of activity at the level of single cells (intrinsic component). An interaction between these patterns and external signals (extrinsic components) generate patterns of fates at the cellular level that result in distributions of cell populations which are the result of distributions of gene expression in those populations. In turn these interactions across scales result in macroscopic structures. At the moment we do not know what these significant variables are nor what are their relationships but there are glimpses of this in the literature (see references at the end).

We need to look at Physics for inspiration and the current impasse needs, quickly, some new paradigms to move from description to understanding. The single cell analysis of developmental processes and, in particular stem cell populations has raised the possibility that statistical mechanics can offer a useful paradigm. What you have read for the last few minutes is a statement in support of such programme. But what we shall need is to define the macroscopic and the microscopic variables in a precise and meaningful manner. Then, progress will follow. Perhaps Boltzmann was right and the XIX century belongs to Darwin, as much as the XX belongs, at least in Biology, to the gene. In this series, the XXI should be the century of the cell and I hope that it does not take us 100 years to realize that to name and count genes and proteins is as futile a task as that which Boltzmann circumvented: to calculate the position and momentum of every particle of a gas. In many ways Biology is the unwritten chapter of statistical mechanics, the chapter that beckons at the end of any text book in the matter.

Darwin gave Biology a way to use the information that has been collated by naturalists in their collecting frenzy (which Darwin practiced in his early days). Today, instead of beetles and plants, we collect sequences and genomic landscapes and this is important and useful. However, the wonder of these objects and the useful information they contain should not deter our attention from the real task in hand which is to formulate the questions that will allow us to link genes (and epigenes) to cells and cell populations and through these to tissues and organs.

A brief list of related references (to build a field: the statistical mechanics of biological processes)

Karsenti E. Self-organization in cell biology: a brief history. Nat Rev Mol Cell Biol. 2008 Mar;9(3):255-62. doi: 10.1038/nrm2357 (E. Karsenti is a pioneer of the attempts to understand biological systems bridging the microscopic and macroscopic realms. He has done most of his work trying to understand how molecular ensembles generate cells which is a first step towards higher levels of understanding. His work is very influenced by I Prigogine).

Lander AD. Making sense in biology: an appreciation of Julian Lewis. BMC Biol. 2014 Aug 2;12(1):57. With Julian Lewis in mind, an insightful meditation of models in Biology.

Gunawardena J. Beware the tail that wags the dog: informal and formal models in biology. Mol Biol Cell. 2014 Nov 5;25(22):3441-4. doi: 10.1091/mbc.E14-02-0717. Models? What kind of models?

The next three references deal with the all important issue of time which is not dealt with here but is very important in linking molecular, cell and developmental biology:

Kicheva A, Cohen M, Briscoe J. Developmental pattern formation: insights from physics and biology. Science. 2012 Oct 12;338(6104):210-2. doi: 10.1126/science.1225182.

Kutejova E, Briscoe J, Kicheva A. Temporal dynamics of patterning by morphogen gradients. Curr Opin Genet Dev. 2009 Aug;19(4):315-22. doi: 10.1016/j.gde.2009.05.004.

Nahmad M, Lander AD. Spatiotemporal mechanisms of morphogen gradient interpretation. Curr Opin Genet Dev. 2011 Dec;21(6):726-31. doi: 10.1016/j.gde.2011.10.002.

The next four references discuss in an explicit manner the need for an approach based in statistical mechanics to understand the dynamics of cell populations in development.

Chalancon G, Ravarani CN, Balaji S, Martinez-Arias A, Aravind L, Jothi R, Babu MM. Interplay between gene expression noise and regulatory network architecture. Trends Genet. 2012 May;28(5):221-32. doi: 10.1016/j.tig.2012.01.006.

Garcia-Ojalvo J, Martinez Arias A. Towards a statistical mechanics of cell fate decisions. Curr Opin Genet Dev. 2012 Dec;22(6):619-26. doi: 10.1016/j.gde.2012.10.004

MacArthur BD, Lemischka IR. Statistical mechanics of pluripotency. Cell. 2013 Aug 1;154(3):484-9. doi: 10.1016/j.cell.2013.07.024.

Trott J, Hayashi K, Surani A, Babu MM, Martinez-Arias A. Dissecting ensemble networks in ES cell populations reveals micro-heterogeneity underlying pluripotency. Mol Biosyst. 2012 Mar;8(3):744-52. doi: 10.1039/c1mb05398a.

On the dynamics of cell populations:

Klein AM, Simons BD. Universal patterns of stem cell fate in cycling adult tissues. Development. 2011 Aug;138(15):3103-11. doi: 10.1242/dev.060103. This is an important insight from physics on the dynamics of cell populations.

A new sort of engineering: II. Organizing self organization in Space and Time

Note: This is the second part of the last post and is not its final form. It will be updated and cleaned up in the New Year but wanted to share these thoughts with those of you who cared to read them before the treadmill catches up with me in the New Year.

 

The vis essentialis of Wolff, the Entelechia of Driesch, the new physical laws promised to Delbruck by Bohr, all found echoes in the famous book “What is life” by E. Schroedinger. This book, that meant so much to a few who went on to change Biology, deals with two questions concerning the physical nature of Life. The first one, central at the time, is the structure of the hereditary material. The second, more abstract and less emphasized in discussions of the book, focuses on the need to explore the thermodynamic basis of Living systems and in doing so it raises, again, the possibility that there might be new laws of Physics lurking in biological systems:

“Living matter, while not eluding the ‘laws of physics’ as established up to date, is likely to involve ‘other laws of physics’ hitherto unknown……. from all we have learnt about the structure of living matter, we must be prepared to find it working in a manner that cannot be reduced to the ordinary laws of physics. And that not on the grounds that there is any ‘new force’ or what not, directing the behaviour of the single atoms within a living organism, but because the construction is different from anything we have yet tested in the physical laboratory.

We must therefore not be discouraged by the difficulty of interpreting life by the ordinary laws of physics. For that is just what is to be expected from the knowledge we have gained of the structure of living matter. We must be prepared to find a new type of physical law prevailing in it”

By the time the book was written, the inevitability of Genetics as the key to unlock the chemical underpinning of living systems was widely accepted, and the double helix, the genetic code, the unravelling of the biochemistry of metabolism and the principles of gene regulation that followed, soon became the vindication of this statement. But, as Crick said, this was, and still is, all chemistry. New physical principles elude any questioning; perhaps, in the end, there are none.

A few weeks ago I attended a meeting of The Company of Biologists in Surrey (UK), “From stem cells to human development”. I went with a mixture of scepticism and curiosity. After all, how could one study human development? By studying I mean, not just describing it and comparing the normal with the pathological but rather doing the kind of work that, through the use of model organisms, has brought so much insight into the molecular and cellular mechanisms underlying the development of embryos. If we are just beginning to grasp how genes govern the development of mice and fish thanks to experimental intervention, how are we going to do the same with humans? It is not just that the material for these studies is difficult to obtain, it is that, with all reason, we need to be mindful of the ethics of this work. Such thoughts were in my mind fuelling low expectations. The meeting turned out to be, for me, a great surprise and the answer to many of my questions, I should have known, laid hidden in the title of workshop through stem cells to development”.

The meeting was a series of examples of what is becoming a clear fact to those in the know: cells are the vehicle between the genes and the organisms. Cells transform the instructions that lay dormant in the genome not just into proteins but into shapes and complex multicomponent forms. The structure of cells, and not just their physical organization but also their computational structure, drive their assembly in the macroscopic arrangements of different cells that we call tissues and organs, and this manifests itself, more than anywhere else, in the surprising self organizing activity of stem cells, embryonic and adult. Take cells with the right potential, place them in the appropriate culture conditions, ignite them with a signal and a genetically driven process will be unleashed that will transform a sequence of nucleotides into a multicellular structure. And the meeting showed us how human eyes, neocortexes, intestines, lungs and blood, emerged from stem cells, embryonic stem (ES) cells. Watching these unfold it is difficult not to think of these processes as manifestations of the vis essentialis and see in these cultures the opportunity to tackle its physical, or if you want to be more conservative, physical-chemical nature. And it is in watching these wondrous processes that the possibility of novel physical principles lurks again in the background.

The notion that cells derived from embryos have a self organizing activity had been known, but perhaps not appreciated, for some time. Thus, Holtfreter and Barth (discussed in a modern light by Hurtado and de Robertis 2007 Dev Biol. 307, 282-289) had observed that animal cap cells from Ambystoma maculatum salamanders, will differentiate autonomously into structures which resemble anterior cortex and develop eyes in culture. Furthermore, attempts to understand limb development and patterning contain numerous reports of mesenchymal cells, jumbled up and wrapped in ectodermal coats, generating digit like structures with recognizable identities. In the premolecular era, this type of experiment was the bread and butter of the experimental embryologist but though guiding much developmental biology, at the time there was little chance of understanding them. Over the last twenty years the application of the methods of classical genetics to development and pattern formation have yielded a catalogue of genes associated with particular processes. In this endeavour you remove a gene, look at the consequences for the organism and then try to work out what was the job of the gene in the process that has gone awry. Connections between genes are worked out through the process of epistasis. While this works very well with linear systems and particularly with metabolic routes, things can get out of hand with complex processes involving non linear systems like cellular machines, or processes like most of development and pattern formation. In particular, it is possible to find molecules that will induce nervous system from ectoderm or digits out of mesenchyme but, what about the process itself? But there is more to a process than its outcome; in fact the process is more interesting than the outcome. Is there a way to tackle its dynamics (the vis essentialis), the way cells proportion tissues and organs (entelechia)? Here perhaps, one needs to take a page (or a piece of the blackboard) of Richard Feynman who famously said: “that which I cannot build I do not understand”. This has been taken much at heart by synthetic biologists who ever since the ‘repressilator” (Elowitz, M. and Leibler, S. A synthetic oscillatory network of transcriptional regulators. Nature. 2000 Jan 20;403(6767):335-8) have been testing the information processing capacity of genetic circuits. But, in terms of Development, we are very far from being able to synthesize the very complex circuits and networks that drive the emergence of organs and tissues. Enter the cell.

Over the last five years there have been an increasing number of reports highlighting the ‘self organizing’ activities of cellular ensembles derived from stem cells. In work pioneered with intestinal stem cells, two cells from adult intestines can give rise in culture to structures that resemble very much in terms of composition and organization, the villi that configure the mammalian intestine. The same has now been shown to be true for other tissues, including lung and liver. However, in an extreme version of these experiments, the late Y Sasai (De Robertis EM. Yoshiki Sasai 1962-2014. Cell. 2014 Sep 11;158(6):1233-5) and his colleagues were able to, under specific culture conditions, coax ES cells to develop into retinas and neocortex. What is more, if the ES cells were from mouse, the emerging structures were sized as in a mouse but, if the starting point were human ES cells, the end point has the size of a human retina (Sasai Y, Eiraku M, Suga H. In vitro organogenesis in three dimensions: self-organising stem cells. Development. 2012 Nov;139(22):4111-21. doi: 10.1242/dev.079590). What else do you need to think about a driving force that is tailored to a particular species and which is able to asses in such a precise manner its final size and proportions? The meeting in Surrey had its share of these phenomena: human ES cells proliferating and turning into intestines, lungs and blood, which highlighted the cells as the architects of the organism through the interpretation of genetic programmes. But how do we get at the physical basis of these structures? Are there any new principles or physical laws behind these observations?

I do not think that self organization is the right notion for what is happening in these experiments; in some way what these cells do is, principally and certainly initially, to organize; however, as it is likely that everybody understands the term self organization, I shall stick to it – though in places I would rather remove the self: the ensemble self organizes, a cell organizes. In this context, there is something important, perhaps fundamental, we do not yet grasp –and need to come to terms with- in the fact that it has to be stem cells that do this and that their label of origin determines the final structure. The most straightforward interpretation would be that what stem cells can do and do remains deeply buried in their genetic programmes and that this is what fuels their organizing potential. Experiments in Drosophila shed some light on this. The Eyeless gene is at the top of a genetic hierarchy which controls the development of an eye. Eyeless is conserved in vertebrates, Pax6, in terms of structure and function -though Pax is involved in several functions, it is also involved in several aspects of eye development. Surprisingly, expression of Eyeless in any cell of the developing fruit fly will lead to a compound eye and, in this context, Pax6 will do the same thing. The simplest explanation for this is that Eyeless drives a deterministic programme for eye. Pax6 will do pretty much the same thing i.e. if expressed in Drosophila it will do the same as Eyeless. What this must mean is that Eyeless and Pax6, which are transcription factors, elicit a programme which in Drosophila’s software will produce a compound eye, the eye of Drosophila. There ought to be a homologous programme in vertebrates as Pax6 is required for the development of the eye. This is not self organization, but highlights that once a programme is initiated in a cell or a group of cells, it will be followed to term. It is maybe that some of what is going on in the cultures of stem cells have a component of this: a programme gets activated which in a deterministic manner will lead to the particular structure. The self organizing component comes into the picture the minute that there are different cells in the culture which now will not lose their way and will assemble themselves into specific structures. It is interesting and ill understood that this organization requires a 3D organization. However, it is important to realize that in this self organizing potential and much of what we can do is to steer this potential with alchemic precision. However as Jeremy Gunawardena once pointed out to me, there is hope since Chemistry is Alchemy with numbers!

Many questions emerge from these observations as well as many experimental possibilities but, new physical laws? New physical principles? Unlikely. Nonetheless and without getting too philosophical, there are two issues where Biology and Physics meet in these experiments which might lead to new notions or conceptual frameworks about the nature of space and time, what I would call the nature of biological space and time. Enough space left to just outline these and, paraphrasing Fermat’s margin, state that there is much to discuss on this, but not enough room here (or patience left in the reader) so, I will just make a few statements. The experiments with stem cells provide a system to determine how cells measure space i.e why mouse and human ES cells will each produce structures with sizes appropriate to their genetic blueprints? Where is this encoded? How is it decoded and executed? How do adult stem cells keep homeostasis of the size and shape of tissues? How do cells create such defined forms and shapes with a high degree of reproducibility? In an interesting observation, aggregates of ES cells have a critical mass to develop into specific patterns: above it, chaos; below it, inactivity. How do they know? How do they sense? But in addition to Space, there is also Time. And it is in this notion that new concepts, perhaps principles, will emerge. Time is central to biological systems as their dymamics, at any scale, is a most intrinsic property. Time, as is well known to physicists, is the most subjective of variables (as the physicist Sean Carroll puts it paraphrasing St Augustine : “I know what time is until you ask me for a definition about it, and then I can’t give it to you.”) however, biological time is not related in any simple manner to astronomical or sidereal time and when we do make this correlation we might be making a mistake. Take the process of somitogenesis (somites are the building blocks of vertebrates} which is run by a molecular oscillator whose elements are conserved across species but whose period is different in different embryos. But this difference is in astronomical time, perhaps from the perspective of the ‘system’ it does not matter, they are the same. How should we think about this? We can see these transformations in many processes which are run by the same molecular networks but take different astronomical times in different organisms. Is there a difference between the time of the networks and astronomical time? Time, as measured by the activity of genetic circuits (a network is not a circuit), is important for the correct decoding of ‘morphogen gradients” and is probably encoded and created by the activity of those circuits. If there is a process in which time is central it is gastrulation in birds and mammals. Furthermore, here, the convergence of space and time is the essence of the process. In these embryos, gastrulation is associated with a structure called the Primitive Streak: cells get progressively drawn into a groove through which they transit to generate the primordial of the different organs. There is an order: extraembryonic, blood, endoderm, heart, somatic…..what determines what a cell does is not its position in the embryo but the time and the order at which it enters the streak. There is a temporal programme which must be written in the gene networks and the circuits they produce. As JA Wheeler said, “time is the way to ensure that things do not happen all at once”. Nowhere is this more clearly stated than in many biological processes and most and best of all in the process of gastrulation in birds and mammals. It is in understanding how genetic circuits generate and interpret time that many new insights will come about in biology and how we might find some novel physical notions (and its relationship to space). It is also likely that this is the vis essentialis and Entelechia.

There is a tinge of engineering in the way we are handling ES cells, but of XVIII and XIX century engineering in the sense that we are tinkering with something we do not quite understand and somehow getting it to work. However, unlike mechanical, civil or chemical engineering where humans run the system, in this engineering of cellular organization, the system rules and runs the scientists, using a blueprint that, for the time being remains hidden in the deep cellars of the cells –to emphasize that we really do not know where it lies.

Interesting and exciting times ahead as accessing that blueprint will reveal new principles and mechanisms, if not physical, certainly biological. Not just the notion of how cells generate time, their time, but also the averaging of fluctuations at any level at the higher level of organization e.g molecular to cellular, cellular to tissue. A new kind of engineering indeed.