An introduction to programmes for development

The cytology laboratory in the University of Leicester has a long tradition in studies of the lampbrush chromosomes that are found in the growing oocytes of most animals as well as in certain stages of the life cycle of at least one simple plant, the giant unicellular alga Acetabularia mediterranea (Callan, 1982). At some stage in very early diplotene of oogenesis of an amphibian, the oocyte nucleus begins to enlarge and many regions of the nuclear chromatin begin to transcribe RNA. RNA polymerase molecules attach to hundreds of sites along the chromosomes and move progressively along the DNA strand, synthesizing long and complex molecules of RNA as they go. The process intensifies as more polymerase becomes available, and the consequence is an unfolding of the chromosomal DNA to form thousands of loops or transcription units, and ultimately the appearance of the lampbrush form that has been so extensively studied in late previtellogenic and early vitellogenic oocytes. Exactly how this unfolding takes place is not yet known. Because of technical difficulties, the earliest stages of lampbrush formation have never been described, and they probably differ in detail from one organism to another. However, it does seem to be a process that happens quite quickly, leads to a most interesting and problematical form, and is perhaps best thought of as an activation of widespread transcription followed by an unfolding of hitherto inactive chromatin.

The initial plan of the Symposium unfolded quite quickly following discussions with a number of British and American developmental biologists with active research programmes based on a wide range of developmental phenomena. Right from the outset it was accepted that developmental biology was a field that covered all types of organism, and the Symposium should therefore provide participants with opportunities to learn about and think about systems that stretched from microbes to man, emphasizing on the one hand the generality of most developmental phenomena, but at the same time highlighting certain important differences between simple and complex systems and between plants and animals.

First and foremost was the need to focus on genes and chromosomes, since they are the primary source of encoded information for development. Then followed a tendency to examine ways in which genes were subject to various kinds of programme that had evolved to direct development along certain defined and well-regulated pathways. To produce a symposium on such topics would be easy and its content would not be hard to predict in the context of the times and trends in molecular and developmental biology. So it was decided to take a small step away from biotechnological analysis on the principle that genes and chromosomes operate within cells and organisms that are subject, like everything else, to physical and mathematical constraints. The modern approach to ‘epigenetics’ is through computers, and in every sense this Symposium seemed to offer the right opportunity to examine some of the ways in which modern computing techniques were being applied to analyse and rationalize certain aspects of the behaviour of cells and the morphogenesis of more or less complex systems. To attempt to bring together and generate useful interaction between hard-core experimental biologists - few of whom have progressed far beyond the use of word processors or the construction of programmes in simple ‘BASIC’ - and a small but distinguished group of pioneers in the application of computing to developmental analysis was risky. The language barrier was likely to be for-midable, and scope for mutual respect was unpredictable. Nevertheless it was worth a try, and it seems to have met with some success.

The first part of the Symposium deals with chromosomes as sources of genetic information and as surprisingly changeable structures (Bostock). The surprise comes for persons belonging to the generation that was vigorously indoctrinated in the notion of the constancy of the genome and the dogged dependability of its DNA. There is a treatment of some of the more dramatic ways in which chromosomes can differentiate their expression (Bostock and Bird), some of the special ways in which they can influence the final form of the organism (Müller), the interdependence of chromosomes, and the importance of their size and position in the cell nucleus (Heslop-Harrison and Bennett).

The second part of the Symposium concentrates on definable programmes of gene expression: switching on and off of genes on cue to direct the production of a simple but highly evolved organic form (Brammar and Hadfield); programmes evolved for the production of specific but quite complex cellular components (Schloss); the orchestration of sets of genes to bring about or maintain particular phases of development (Hodgkin and Baulcombe) ; the molecular biology of gene regulation and control of expression (Kalfayan and Lindquist); the interplay between external environmental factors and the transcription of nuclear genes (Ellis and Trewavas); and the programming of genes to produce specific developmental decisions that commit cells along new pathways of development (Johnson).

The impression is most likely to be that the genome is in overall and quite highly disciplined control, but that it has evolved to be acutely sensitive to a rather special set of factors operating from both inside and outside the cell. The whole scenario will be one of orderliness and purpose: genes switching on and off for the timely production of appropriate and specific materials that constitute the basis of cellular differentiation.

For persons who are concerned with what happens before embryonic development ever starts it is a slightly perplexing scenario. Returning to the astonishingly widespread transcription that continues on lampbrush chromosomes for weeks, months or even years before egg maturation, we ask: Is there a programme here? Why are so many segments of the nuclear DNA, over 10000 of them in some organisms, simultaneously transcribing at maximum rate to judge from the close packing of polymerases on the axes of transcription units? What does all this represent? Is it all part of a highly complex and sophisticated programme of conditioning for an egg that must later cope with maturation, fertilization and the cleavages of early development? Is it perhaps a rather indiscriminate event that has persisted to ensure the correct levels and proportions of expression of relatively few genes that really are absolutely essential for the initiation and maintenance of development? Or does it signify something else that we do not yet understand? Whatever the answer, and we do not know it, the lampbrush phase of oogenesis remains a challenge, and if it contains a programme of gene expression then there must indeed be much more to an egg than is evident from its outward rounded simplicity.

The frontier of this Symposium lies between the biologists and the computer scientists, between biology and biochemistry and mathematics and modelling, and the final part of the Symposium consists of five contributions by computer scientists who have strong interests in the events of development. One of the purposes of this Introduction is to offer some comments that may help to define the frontier and make it more negotiable in both directions.

It would seem that there are likely to be three main ways in which gene expression in higher organisms may be finely regulated; three categories, that is, that may be defined in the context of the Symposium. The first is through the influence of repressor and activator molecules: proteins that specifically block or facilitate the synthesis of messenger RNAs by binding at or near to transcription initiation sites. There are, of course, many examples of repressors in bacterial systems (see for example, Ptashne et al., 1980), and equally satisfying examples of activators and repressors are to be found in eukaryotes (Nevins, 1983). Es-sentially, there are intracellular protein complexes that act directly on a gene to inhibit or facilitate its expression.

The second category is one in which the expression of a gene is altered by changing the DNA sequences in its neighbourhood. The cellular oncogenes are now the best examples of this kind of phenomenon (Bishop, 1983; Land, Parada & Weinberg, 1983), and in a sense it is these oncogenes that provide us with what may be the first track across the frontier of the Symposium. The point is that at least some oncogenes seem to code for proteins that are associated with the cell surface. The relationship between the expression of a specific gene or group of genes and a change in the cell surface, in the outward direction of gene action leading to cell surface change, has not yet been clearly demonstrated, but the notion is there and it is gaining credibility day by day.

The third category is presently the least well understood but in many ways it is the most exciting. It is summed up in the notion that the shape of a cell plays an important role in determining the pattern in which its genes are expressed. In experimental terms, place a certain cell on a certain substrate and it will adopt a certain shape and express a certain group of genes. Change its substrate and it will change its shape and proceed to express a different set of genes. Such influence by the immediate physical environment on the differentiation of cells has been observed in studies of mouse mammary gland epithelium (Gordon & Bernfield, 1980), and chondrocytes (Nathanson & Hay, 1980; Belsky, Vasan & Lash, 1980), and there is every reason to expect that more examples will be found. In addition, once a particular differentiative pathway has been followed, the substrate probably plays an important role in maintaining the stable differentiated state of many cell types (Toole, 1981).

Could it be that through things like oncogenes and studies of cell surface/ substrate interactions we are coming nearer to understanding the significance of interplay between growth, in the sense of genome-directed intracellular events, and form, in the sense of cell shape and interaction with substrates and other cells?

The cells of higher organisms do not normally change or differentiate unless they are in contact with something. Contact involves the cell surface. Cellular differentiation that accompanies development means change of shape, movement, repositioning and new patterns of cell behaviour. All these involve the cytoskeleton and a quite well-defined range of structures and substances that are concerned with shape and motility. To what extent does size, shape and movement provide the driving force for the creation of new patterns of interaction, and to what extent are these new patterns of interaction initiated and sustained by changes in programmes of gene expression? This would seem to be the question that we should ask ourselves in the framework of the whole Symposium.

Plant developmental biologists have made a major contribution to the Symposium, and since this is unusual for a BSDB production it seems appropriate to offer some comment on the extent to which plant and animal systems relate to one another in terms of general developmental biology. There are just two special circumstances that we should take note of in plants. The first is the cell wall, and the second is the fact that most plant cells are in direct cytoplasmic continuity with one another. What then, we may ask, can the cell membrane possibly have to do with differentiation and development other than to regulate the passage of molecules into and out of the cell? Perhaps very little, but the cytoskeleton and the forces and programmes that alter cell shape are certainly highly important in almost precisely the same manner as in animals. In the simplest terms one might say that apart from differences in the role of the cell membrane in differentiation, the only thing that plants have done in contrast to animals is to evolve a tool, auxin, that plasticizes the cell wall. Once this has happened, the cytoskeleton takes a hand, the cell changes shape, and then it recasts the wall around its new form. The question arises once again: to what extent does the form of the cell influence the programme of expression of its genes, and vice versa?

Of course, the idea of linking growth and form is not a new one. It was foremost in the minds of most 19th century embryologists, and it was widely explored by D’Arcy Wentworth Thompson in the two-volume work that he published in 1917. D’Arcy Thompson spent much of his life and thought trying to import into what we now cell developmental biology the concepts of physics and mathematics and trying to introduce along with them the method of experiment. Regrettably he did not have a computer! Instead, he and his predecessors used soap bubbles or balls of dough with varying amount of yeast, so creating models of cellular aggregates in which individual cells grew at different rates. Above all, D’Arcy Thompson dedicated himself to showing that in living organisms, what we can learn from the simplest cases in terms of mathematics and physics, includes the principles that determine the most complex. The following quotation from ‘On Growth and Form’ makes his point. In the case of the growing embryo we know from the beginning that surface tension is only one of the physical forces at work; and that other forces, including those displayed within the interior of the cell, play their part in the determination of the system. But we have no evidence whatsoever that at this point, or at that point, or at any, the dominion of physical forces over the material system gives place to a new condition where agencies as yet unknown to the physicist impose themselves on living matter, and become responsible for the conformation of its material fabric’.

Well, now we have some evidence. We know a great deal about genes and chromosomes and about programmes of gene expression, and we are learning fast about interplay between genes and the cellular components that are responsible for generating and accommodating physical force; and now we have computers. The crucial question again, put in another way: to what extent is there an interplay between the biological and biochemical forces that operate outwards from the genome, and the physical and mathematical inevitabilities of organic form?

It is to be hoped that persons reading this book will regard it truly as a symposium volume and not a collection of papers, and will make the effort to examine each of the contributions and evaluate them individually within the framework of the whole. The Symposium was an experiment. Its success depends on open mindedness and eagerness to understand the fundamental principles that govern growth and form in all living organisms.