HIERARCHICAL CONTROL PROGRAMS IN BIOLOGICAL DEVELOPMENT

The translation of a two-dimensional linear sequence into an active three-dimensional matrix is essentially a mechanism for decoding specific chunks of genetic memory in terms of the immediate environment. The step from transformation to development, from the decoding on a specific protein molecule to the construction of a multi-cellular organism is a step from transformation to programming. One of the most important discoveries in modern cell theory is that in any given organism, regardless of how complex, a complete set of instructions for creating that particular organism is contained in the nucleus of each and every cell, A sequence of nucleic acids will specify a particular protein, the sequence of such protein specifications is a program for the development and maintenance of that specific organism. That program itself is a complex system. James Bonner discussed how such a program operates.

We know for example, that in humans we have two genes for making the two protein chains of hemoglobin, We know that those two genes are turned on for transcription of the messenger RNA's for making hemoglobin in certain body cells, those of the bone marrow, which gives rise to the reticulocytes and thence to the erythrocytes. Also, we know that those same genes do not make the messenger RNA for making hemoglobin in other cells of the adult organism, Clearly the genes for making hemoglobin are turned on in one kind of cell and make their appropriate messenger RNA, but in other kinds of body cells they are turned off and do not make their messenger RNA, We can think of many cases which show that particular elements of the genetic material are turned on only in particular places.

Hierarchical control programs must have the capability of turning on the right genetic specifiers at the right moment, in the right place, and in the proper sequence, By removing chromatin from a specific cell then transcribing it with RNA polymerase, researchers have determined that in any given cell between one and five percent of the DNA is available for transcribing. Ninety-five to ninety-nine percent is turned off. A class of proteins called histones which are found only in association with DNA of complex cells serve to turn off unwanted chromosomes. They are called repressor molecules. Bonner said that there is a limited number of such molecules and that they are essentially independent of the organism.

During the last five years, my colleague Douglas Fambrough, now with the Carnegie Institute in Baltimore, and I have studied the chemistry of the histone molecules of chromosomes and have found that there are a limited number of kinds of them: We distinguish eight kinds. These eight kinds differ from one another in their chemical properties, in their amino acid compositions, and so on. We have found the same eight homologous histone molecules in the chromosomes of creatures as different as humans, cows, rats, tetrahymena (a protozoan), neurospora (a fungus), peas, cowpeas, and frogs--all of the kinds of higher plants and animals that have been investigated, The chemical properties of the several chromosomal proteins are similar.

Hormones are a special kind of messenger molecule that travels to specific cells and instructs those cells to turn on certain genes.

Consider, for example, cortisone, which is made in the adrenal cortex and travels to the cells of the liver, When cortisone, which is a steroid arrives in the liver, it behaves as though it were saying, "Liver, turn on the following series of genes, so that the enzymes which those genes describe can be made." These are enzymes that have to do with particular metabolic responsibilities of the liver, and these genes become turned on.

There are many such molecules. They operúate by turning on appropriate genes that were previouly repressed. However, they do not accomplish this alone, When they arrive in a cell they bind with a protein that is capable of combining with only one kind of molecule and it is this complex that turns on the gene. There are additional mechanisms that must be included in a development program. Consider, for example, that every cell in a multi-cellular organism is to some extent unique and occupies a unique position in the organism. In order for the cell to know just what it is supposed to be doing at any particular point in the lifetime of the organism it must perform tests.

Let us imagine that wúe are a cell, and that we are going to divide into two, and then into four, and subsequently to develop into something. What sort of' control information would we need in order to perform this play? For example, let us imagine that we are the single cell at the growing apex of the bud of a particular class of plants that has only one apical cell. This divides into two and forms two new cells, the bottom of which starts dividing to make the bud get bigger, while the upper goes on being an apical cell. And in this class of plants, the cells in the bud multiply and multiply until the bud achieves the desired diameter. In the next step in the process, cells start turning into specialized cells; the ones on the outside start turning into epidermis, and the ones on the inside turn into wood, The ones that are in between turn into phloem and cambium. Again let us suppose that we are this apical cell. We divide into two--one is on top, and one is on the bottom. If wúe are to go any further along our development pathway, what must we knowú? We must know whether we are apical or not apical. Because if we are not apical, then we must go to that portion of the apical book that tells us how to make buds. And if we are apical, we must return to the apical bud state, divide again in the right plane and make a further sub-apical cell wúhich can turn into more bud and so forth. Clearly, if we are a cell that is going to develop, we must have ways to test the envir:onment about us and find out whether we are apical or not apical.,

Bonner went on to suggest that one way we can perform such a test is by having certain genes produce a volatile substance. Qbviously the cell that is on the outside will have a low concentration of the volatile substance while one on the inside wúould have a high concentrúation, This low concentration might not turn on the genes for making the cell grow into a bud, while in a high concentration it could be the molecule that binds wúith the messenger RNA to turn on those genes that are responsible for the initiation of the bud pathway. As Bonner put it, "This is the concept we refer to as the development test--the hierarchical concept that a growing cell in a developing organism is continuously performing tests of its environment.

Every cell in every organism must have a set of tests that will tell it where it is in the organism and at what stage of development or maturity the organism is. Even in a simple organism it is quite evident that this will require a complex program and thus it will be hierarchical.

In the development of an organism first there must be a determination of which pathway is to be followed by the cell. "We do not know how many pathways there are. "Bonner explained," I sat down and tried to wúrite them out for pea plants. I immediately wrote out about 30 different developmental pathways that must necessarily be involved in the development of a plant, and I presume that in the development of a higher animal such as ourselves, there must be even more pathways. So we turn on the set of genes which generates this pathway. That is hierarchical level number 1, or the highest development control level."

The second level in the hierarchy, Bonner said, is the turning on of genes that are responsible for sensing the concentrations of particular kinds of molecules in the environment, and the third is the turning on of genes that code for the appropriate enzymes according to the information contained in the environment. In other words, in order to have a development program capable of the creation of complex organisms it must be hierarchical.