




What determines whether a limb grows into an arm or a leg? Research into insect development suggests that a relatively small number of “master genes” control human development, too
THE MYTH that babies are found under gooseberry bushes was a convenient way for Victorian parents to avoid the embarrassing subject of the facts of life. But babies, like most other living things, are not created whole: they grow from a single fertilised egg cell, and continue to grow from birth to adulthood.
Advertisement
This process is controlled by genes. Most genes carry information that directs cells to make proteins. In most cases, a single gene carries the information needed to make one type of protein. When a gene is active, the cell “reads” information from that gene and uses it to make a particular type of protein. The gene is said to code for that protein. When a gene is switched off, its information is not used to make the protein.
Some genes tell the cells in an embryo where they are and whether they should grow into an arm or a leg, or whatever. We are now beginning to understand more about how these genes work.
Almost all animals begin life as a single cell called a zygote. This forms when a sperm and an egg cell (the gametes) fuse together. Each gamete contributes half of the zygote’s complement of genes. These genes control development as the zygote grows into an animal consisting of many billions of cells. For the animal to live, each cell must develop correctly according to its position in the embryo.
In humans, the zygote divides in two after about 30 hours. At this point the two cells are identical. We know this because each cell has the capacity to grow into an individual by itself. Identical twins arise when a two-celled embryo splits. But as the cells continue to divide, those in different parts of the embryo become committed to different tasks. Some are delegated the task of making skin, others that of making muscles, and so on. Our bodies contain more than 200 clearly distinct types of cells. The real number of different types is probably much higher.
When a cell becomes specialised, it loses the ability to give rise to a whole person. This process of specialisation is called differentiation. If we could understand how differentiation is controlled, we would be close to understanding development.
Differentiation of a cell involves changes in the activity of its genes. Probably tens of thousands of genes are needed to make a person. Almost every type of cell in the body contains a complete set of genes, but in most cells only some of these are switched on.
The cell’s function determines which genes are active. So, in hair cells, the gene for keratin – the fibrous protein in hair – is active; in the cells of various glands, genes that direct the cells to make hormones are switched on; and so on. Differentiation is essentially the control of gene activity.
Fortunately, understanding the process is not quite as difficult as it sounds. A lot of genes can be ignored, including the so-called housekeeping genes. These are active in almost all cells: they run the basic processes that cells need to stay alive. Housekeeping genes rarely affect differentiation.
A relatively small number of master genes regulate directly or indirectly the remaining genes. The evidence that the master genes exist comes from research into the development of fruit flies (Drosophila). èƵs have identified about 100 genes that a fruit fly needs to develop correctly.
The picture is less clear in the case of our own development. We are fairly certain that it is controlled by master genes but no one knows how many there are. èƵs are studying how master genes work and hope to find out what controls them.
Odd insects
Useful clues
ONE GROUP of master genes seems to operate in both insects and vertebrates. This observation is very exciting because these two types of animal are so different: any genes that operate in both of them are probably of fundamental importance.
By the standards of molecular biology, this group of genes has had a long history. The genes were discovered because scientists decided to study some very strange-looking insects. The insects developed more or less correctly, but some of their parts grew in the wrong place. For example, a leg might grow out of the insect’s head, in the place where an antenna would normally be.
Nearly a century ago, an English geneticist called William Bateson studied these transformations, as scientists called them. Because one part of the body appeared to change into a likeness of another he called the phenomenon homoeosis from a Greek word meaning “likeness”.
Biologists realised that these weird-looking insects might be able to tell them a lot about animal development. After all, if a leg grows in place of an antenna, the controlling mechanisms have somehow gone wrong.
At that time, DNA, the blueprint of life, was yet to be discovered. But we now know that almost all genes are part of the DNA contained in the chromosomes in the nucleus of the cell. DNA is in the form of two strands wound together in a double helix. The backbone of the double helix has a series of molecules called bases attached to it.
The information needed to make a protein is stored in the sequence of the bases on one of the strands of DNA. If this sequence changes, the structure and properties of the protein for which the gene codes may change too. A gene which has a different sequence to normal is said to be mutated.
Although for much of this century biologists did not know exactly what a mutation was, they could see what effect a mutation had. By looking at how fruit flies inherit developmental abnormalities, they deduced the existence of the genes that make each part of the fly develop in the right place.
Their success followed pioneering work by the American biologist Edward Lewis, some 50 years after Bateson. Without the benefit of modern technology, they found mutated genes that each gave rise to a transformation in a different part of the fruit fly. For example, a mutation in one gene affected the development of part of the head; in another, mutations affected a section of abdomen. èƵs have coined the name homoeotic genes for these genes, from Bateson’s term “homoeosis”.
When a homoeotic gene is mutated, the development of part of the insect becomes shunted onto the wrong track. A single mutation in a homoeotic gene can cause a leg to develop in place of an antenna. Yet building a structure such as an antenna or a leg requires the concerted action of hundreds of genes. So homoeotic genes must control the activity of the hundreds of other genes needed to make each part of the body.
In each part of an insect embryo, a different set of homoeotic genes is active. One useful though oversimplified way to understand the function of these sets of genes is to think of them as providing the combinations to unlock different developmental programmes, thus activating the genes needed to make a wing, a leg or whatever.
Homoeobox genes
Controlling development
IN THE past 10 years, the techniques available for studying genes have become extremely sophisticated. èƵs have been able to uncover much more about how homoeotic genes work, by determining and manipulating the sequence of bases in a gene, for example.
Researchers have found that all the homoeotic genes in fruit flies contain a short segment of DNA in which the sequence of bases is nearly identical. Biologists often call small sections of DNA associated with genes “boxes”. Sticking to this tradition, they called this segment the homoeobox.
In each of the genes, the homoeobox codes for a small section of protein that can bind to DNA. Proteins coded for by homoeotic genes can therefore bind to specific DNA sequences near other genes. This binding makes it possible to switch other genes on or off, by intefering with the mechanisms which “read” the information from genes. This explains how the proteins coded for by homoeotic genes can directly control the activity of other genes.
Once they had found the homoeobox, scientists looked for it in other genes and in other animals. The result was a flood of homoeobox genes (genes containing a homoeobox). In fruit flies alone, the search has yielded nearly 50 homoeobox genes. These genes also turned up in other animals, including ourselves and other vertebrates.
Some of the homoeobox genes in vertebrates turned out to be astonishingly similar to the original homoeotic genes in fruit flies. To prevent confusion with all the other homoeobox genes, these vertebrate genes are called Hox genes. The word “Hox” is a shortened version of “homoeobox”.
Although many homoeobox genes were found, not all play a major part in development. Many control only a few genes needed in particular cells in certain circumstances. Others ensure that the correct housekeeping genes are switched on.
Despite this, many scientists believe that the Hox genes of vertebrates play a similar role in development to that of homoeotic genes in insects. But they are still wary of calling Hox genes “homoeotic” genes. This is partly because no one has shown conclusively that Hox genes help to control our development. Nevertheless, there are several remarkable similarities between Hox genes and the homoeotic genes of insects.
One of the most striking similarities is the arrangement of these genes on chromosomes. In fruit flies, homoeotic genes are grouped in two clusters. One cluster contains homoeotic genes that affect the head end of the body, the other one contains homoeotic genes that affect the tail end. These two clusters were once part of a single giant cluster, in which the genes controlled the fate of every part of the body. èƵs know this because in some insects, such as red flour beetles, this giant cluster is still intact.
Common clusters
No coincidence
HUMANS have several clusters of Hox genes. To scientists’ amazement, the order of Hox genes in these clusters is very similar to that of the homoeotic genes in the giant cluster of red flour beetles. This similarity is far too unlikely to be a coincidence. It means that Hox genes and homoeotic genes must be related to each other. This implies that our Hox genes have a similar role to that of insect homoeotic genes.
Further research has strengthened this belief. In insects, each homoeotic gene is active in a particular region of the embryo. These regions, which overlap, run along the body from head to tail. The order of the genes in a cluster and the regions of the embryo in which they are active match each other exactly. So homoeotic genes next to each other in a cluster help to control adjacent regions of the embryo.
The combination of active homoeotic genes is an accurate guide to the position of any point along the body. This information makes it possible for the homoeotic genes to select the correct developmental programme for each part of the developing insect.
The pattern of activity of Hox genes in vertebrate embryos is more complex. Nevertheless, their activity in the developing brain and spinal cord is similar to that of the homoeotic genes in insect embryos. Most of the Hox genes exert their control at the base of the spinal cord. But as you move along a cluster of Hox genes, the region of activity of each gene extends a little farther along the spinal cord towards the brain. So, in both insects and vertebrates, the region of activity of a homoeotic, or Hox, gene depends on its position in the cluster.
These similarities in the arrangement and activity of homoeotic genes and Hox genes lead many scientists to believe that Hox genes help to control human development. But scientists must now find the genes that are controlled by homoeotic or Hox genes. There must be hundreds of these: they are the genes involved in building each part of the body. So far, none has been isolated. If this search is unsuccessful, the theory about how homoeotic genes work may have to be revised.
Heads and tails
Put in position
HOMOEOTIC genes are just one piece of the developmental jigsaw. They are important but they are not everything. We know this because, even when a homoeotic gene has a mutation and operates incorrectly, a fruit fly will still have all its segments. So, creating the pattern of segments in insects must be the work of other genes; biologists usually call these the segmentation genes.
In fact, the bodies of insects are more complicated than just a series of segments. They are built up from a series of compartments. Each segment of the insect contains two compartments, a front one and a back one. The boundaries between these compartments are well defined. During normal development, cells will not cross from one compartment to another. (This is in sharp contrast to the situation in vertebrates, where cells are much more mobile during development.)
Insects evolved from simpler animals, such as centipedes, where most of the compartments and segments are very similar. In modern insects, however, every compartment is specialised – not only in its appearance but also in its function. During development, the segmentation genes divide the embryo into compartments; then the homoeotic genes give each compartment its identity.
To do this, the segmentation genes must provide information that allows one part of the embryo to be labelled as different from all the other parts. èƵs call this positional information.
We know more about the role and characteristics of the positional information in fruit fly embryos than in any other animal. Such information is present even in the unfertilised egg cell. Studies in which researchers injected cytoplasm from one part of an egg cell into another part, and watched its subsequent development, showed that in the egg there is already a “head” end and a “tail” end.
The amount of positional information builds up during the development of the embryo. Eventually the positional information is so complex that very small groups of cells, destined to become a particular part of the body, can be distinguished.
In fruit flies, the segmentation genes act in a cascade. The first segmentation genes take their cue from the positional information that existed in the egg cell. They divide the embryo into broad regions. The next set of segmentation genes takes these broad regions and converts them into 15 distinct bands. At this stage, each band is just four cells wide. A final set of genes distinguishes the front end of each band from the back end, allowing each band to be divided into two compartments.
The segmentation genes help to control the activity of the homoeotic genes. In a complex set of interactions, they ensure that each homoeotic gene is active only in the correct part of the insect embryo. So homoeotic genes are part of the cascade which starts with positional information in the egg cell and finishes with the genes that build each part of an insect.
èƵs think that they have now isolated most of the segmentation genes in insects. But no one has found equivalent genes in vertebrates.
Long pedigree
Past performance
INSECTS and vertebrates must have inherited their homoeotic and Hox genes from a common ancestor. This ancestor probably looked something like a flatworm and would have lived around 600 million years ago. So homoeotic genes have been around for a long time.
Finally, what about plants? Plant development is very different from animal development but the basic need for positional information and control of gene expression is still there. Homoeosis has even been found in plants. It can be particularly obvious in flowers, with one floral organ replacing another. For example, petals might develop in place of stamens. Several of the many- petalled garden flowers, such as some double petunias, are the result of this sort of transformation.
An American research team has recently isolated a homoeotic gene in plants. Although this gene does not have a homoeobox, its sequence shows that it probably codes for a protein which can control the activity of other genes.
Metamorphosis
FLIES, moths and butterflies all undergo complete metamorphosis. This means that they change from an egg into a larva, then into a pupa, and finally into an adult (or imago). Other insects, such as grasshoppers, undergo incomplete metamorphosis. These insects do not have a pupal stage and the young look much more like adult insects. Given that flies rearrange their bodies so completely during metamorphosis, how is the development of their adult bodies controlled?
In flies, the cells which will eventually make the adult body are kept separate from the cells which make the larva. As the embryo develops, cells which will form the adult fly grow into a series of discs. These are called imaginal discs and they are strung along the inside of the embryo. When the larva turns into a pupa for metamorphosis to begin, cells in the imaginal discs proliferate and each disc grows into a different part of the adult.
Each imaginal disc forms in a different part of the embryo, corresponding to the section of the adult which that disc will eventually make. The pattern of activity of the homoeotic genes in the imaginal discs is the same as that in the surrounding section of embryo. This is why homoeotic mutations appear in adult flies even though the pattern of activity of the genes is established in the embryo.
The cells that grow, during metamorphosis, into the adult insect are present in the larva as a series of “imaginal discs” which remain dormant until the larva begins to pupate. The pattern of activity of the homoeotic gene is established in the embryo, but because the activity of the homoeotic genes in the imaginal discs corresponds to that in the surrounding section of the embryo, homoeotic mutations can still appear in adult flies. So a mutation that affects the development of an antenna will not be seen until the adult insect emerges from its pupa.
Probing for known sequences
One of the key techniques behind research into the development of embryos has been the ability of geneticists to probe for specific sequences of genetic material, such as DNA. This technology allows them to search for genes containing homoeoboxes and to find out where Hox genes are active in vertebrate embryos.
The DNA molecule consists of two strands twisted around each other – the double helix. Each strand is made of four different kinds of molecules called bases: adenine, guanine, thymine and cytosine – or A, G, T, and C. In the double helix, an A on one strand always binds to a T on the opposite strand, and a G always binds to a C. So if the sequence of bases on one strand is AAAGGG it will be TTTCCC on the other. The two strands are complementary to each other.
A single strand of DNA can be used as a probe to detect strands with complementary sequences. So scientists were able to “fish” for homoeoboxes in animals other than fruit flies. They used a single strand of DNA from a known homoeobox gene to probe for complementary strands in the DNA of other animals.
The same sort of technology allows scientists to discover where genes are active. When a gene is switched on, the cell copies one of its strands to make complementary strands of RNA (a substance similar to DNA). These RNA strands carry the information in the gene to the sites of protein synthesis. By probing a section of tissue with a sequence of DNA complementary to the RNA sequence, researchers can see where the RNA is being made – and, therefore, in which cells the gene is active.
Further reading
For a clear account of modern concepts ahout development and of how the development of fruit flies is controlled see Bruce Alberts et al. Molecular Biology of the Cell (2nd edition, Garland Publishing Inc, New York and London). For a more detailed description of Hox genes and their activity in vertebrates see “Homeobox genes and the vertebrate body plan” by Eddy De Robertis et al, in Scientific American, July 1990.