Why flies are relevant to medical research
This essay was written by Jean-Paul Vincent and Iris Salecker and was first published in the 2004 Mill Hill Essays.
Sequencing the human genome was a milestone for humankind. However, the inflated media reaction following completion of the first draft obscured the fact that genome sequence alone provides little information about what a gene does within an organism. This realisation has spawned a new buzzword – functional genomics – to describe research to find the role of individual genes. In humans this is largely done by identifying genetic defects that are associated with a particular disease or trait. This is laborious and is limited to a few traits, such as a disease caused by mutation of a single gene. So far, out of the thirty thousand human genes, fewer than one thousand have been found to be associated with a disease. Clearly, a more comprehensive and sensitive approach is needed to uncover gene function. Because of many practical and ethical issues, much of this research must be done with model organisms such as worms, flies, fish, chicks or mice.
All forms of life are worthy of scientific research but for an animal species to serve as a model of human biology it must fulfil two key criteria. It must be relevant, that is similar enough to humans, and it must be tractable, that is amenable to a broad set of experimental approaches. In the year 2000, one such model organism made headlines: the sequence of the entire genome of the fruit fly Drosophila melanogaster had been completed. This was only the second genome from a multicellular organism to be sequenced and it prepared the scientific community for tackling the tenfold larger human genome. The fruit fly has been at the forefront of genetic research for the past one hundred years, remaining tightly intertwined with progress in modern biology. In this essay, we examine why Drosophila is set to stay an essential tool in the postgenomic era as it will enable the investigation of complex traits involving networks of interacting genes.
It is arguable that functional genomics was invented around 1910 at Columbia University in New York by Thomas Hunt Morgan. He brought Drosophila into his laboratory to investigate the basis of inheritance, since it is easy to breed, does not require much space, and has a short life cycle of about ten days. He and his students collected flies with many inheritable traits, such as altered eye colours or misshapen sensory hairs, defects that we now know are associated with mutations in specific genes. The first identified mutations were spontaneous, caused by such random events as cosmic rays. Thus, mutant flies were only found by looking endlessly at millions of flies bred in T. H. Morgan’s famous fly room. The first recorded Drosophila mutant had white eyes instead of the normal red colour. The gene was called white, thus initiating a tradition of naming genes according to the defects seen in the corresponding mutant. This tradition has since been upheld in a sometimes colourful manner by generations of Ph.D. students and post-doctoral fellows dreaming of identifying a new mutant so that they could give it a ‘cool’ name. This led to the name shavenbaby to describe a gene required for the formation of specialized hairs on the surface of larvae, and to the name he-is-not-interested for a gene involved in male courtship. A student of T. H. Morgan, Hermann J. Muller, showed that X-rays are mutagenic, that is they cause mutations. We now know that it is because they damage the DNA of eggs and sperm. When flies are exposed to X-rays or other mutagens such as the chemical ethyl methanesulphonate (EMS) defects in the genome are induced at a high rate. Using X-rays to induce mutations allowed the Morgan group greatly to speed up the production and identification of new mutant offspring with abnormal eye pigmentation. This method of identifying many genes involved in a given biochemical process is now called “forward genetics”.
One great success story of forward genetics is the identification of many genes required for embryo development. Janni Nusslein-Volhard and Eric Wieschaus, working in Heidelberg during the late seventies, generated a large collection of Drosophila mutants and kept those which had a disrupted arrangement of specialized hairs on the surface of larvae, as this reflects alteration in the pattern of cell types. Although the identification of so many genes required for embryonic development was a momentous achievement in itself, its impact was greatly enhanced by parallel progress in molecular biology, which allowed the identification of the mutated genes. Until then, genes were inheritable entities that could be mapped only approximately to a position on a chromosome. Recombinant DNA technology provided the tools to assign a specific DNA sequence to genes. Drosophila has proved to be exceedingly tractable for forward genetics, allowing the discovery of gene function at an unprecedented pace. Many Drosophila genes were subsequently found to have homologues (genes with similar sequences) in higher organisms. As a result, several vertebrate genes have been named after Drosophila genes. For example, chick or mouse counterparts of the Drosophila gene called hedgehog were named Sonic Hedgehog and Indian Hedgehog. What better measure of the impact of Morgan’s pioneering work than the spreading of quirky Drosophila gene names to wide areas of biology?
Clearly, Drosophila has allowed rapid progress in basic biology but how relevant is this to human biology? On the surface, a fly is a far cry from a human and yet we are more like flies than most people would suspect. Amazingly, every one of the genes identified by J. Nusslein-Volhard and E. Wieschaus turned out to have a homologue in higher organisms, including humans. Conversely, many human genes have a clear homologue in Drosophila. For instance, over 85% of human genes that have been associated with a disease have a Drosophila counterpart. Humans have about twice as many genes as flies but the extra genes rarely represent truly novel functions. They simply allow for more complex and subtle regulation of core molecular pathways. Of course, this molecular conservation does not imply that genes perform the same overt role in different organisms. The Drosophila gene called wingless, which, as the name implies is required among other things for wing development, has a homologue in mice called wnt-1. Clearly wnt-1 does not specify wings in the mouse but in both mice and flies it encodes a signal that helps cells to communicate with each other during development. Incidentally, many cancers in humans are associated with the misregulation of this signaling pathway.
In the case of wingless and wnt-1, their homology is reflected in the fact that they have the same basic function, cell communication. However, in several cases homology extends further. For example, Tinman is a transcription factor, a protein that binds to part of a gene and helps to control the process by which genes are decoded into proteins. Tinman was identified first in Drosophila as being a master regulator of heart development. It has a homologue in frogs and mice and in these organisms also it is essential for early heart development. Similarly, a set of genes that control the early steps of eye development are conserved from flies to mice. One of them is the gene eyeless, which is required to specify eyes. Amazingly, mis-expression of this gene in many areas of the fly’s body causes the formation of additional little eyes, for example on the legs. The homologous gene Pax6 is found in higher organisms where it is also required for eye development. So in the course of evolution the basic molecular modules used for cell communication and organ specification in the fly have also been adopted to control the development of higher organisms including humans.
Flies have a heart, rudimentary kidneys and cells that act like liver cells. Moreover, perhaps surprisingly, they do have a sophisticated brain with about three hundred thousand nerve cells. This allows them to process complex sensory inputs, have a sense of time, and to learn and remember. Seymour Benzer and his colleagues at the California Institute of Technology in Pasadena pioneered the approach of linking genes to behaviour in flies. Like humans, flies have an inner clock, which makes them observe regular phases of activity and sleep. The first behavioural genetic screen designed by Benzer and his students made it possible to isolate mutant flies with a disrupted sense of time. They identified mutant flies that wake up and sleep abnormally at irregular times. Other mutants display slower or faster patterns of activity. Genetic analysis showed that these behavioral defects were linked to different mutations in a single gene, which they called period. Subsequent studies revealed that this “clock” gene is conserved from flies to higher organisms. Another fundamental question in neurobiology – how we learn and memorize – also proved to be accessible to forward genetic analysis in flies. Flies can be trained to memorize particular odours. In “fly school”, they are repeatedly exposed to two smells, one of which is associated with a negative experience. Even days later, given a choice, flies will walk towards the smell that was not associated with the negative experience. But some mutant flies forget their training session, and will walk to both sources of smell. The first such mutants to be isolated were called dunce, rutabaga and amnesiac. Subsequent studies showed that these genes are all involved in regulating the activity of cyclic AMP, a molecule used for communication between nerve cells. Further work has shown that cyclic AMP metabolism affects memory function in higher species too. Recently, researchers have begun to explore the genetic basis of addiction to drugs and alcohol, using Drosophila as a model system. Flies are naturally attracted to alcohol, because it indicates the presence of rotting fruit, a nice place to lay eggs. Like humans, they show complex behavioural changes when exposed to excess alcohol: initially they become hyperactive, but then lose coordination, and eventually become sedated. Drosophila researchers set out to find genes involved in the response to alcohol by examining the movements of flies exposed to alcohol vapour. They called one of the first identified mutants cheapdate, because the flies were more sensitive to alcohol than normal flies. It turned out that cheapdate is a mutation in the memory gene amnesiac. This helped to establish a link between alcohol sensitivity and the signaling molecule cyclic AMP. These few examples illustrate how flies can be used to assess the role of genes in complex brain functions that are likely to be relevant to human health. Additional work suggests that physiological parameters such as growth and organ function can be investigated in a similar manner. Of course it is important to note that the fly cannot be a universal model. Clearly, flies lack bones, lungs, and an adaptive immune system. Nevertheless the extent of similarities between flies and humans is astounding.
As many genes originally identified by mutations in flies were found in the human and mouse genome, it became important to uncover their function in these higher organisms. For example, what is the role of wnt-1, the homologue of wingless, in the mouse? It has become possible to address this question by ‘knocking out’ individual genes with a technique developed in the 1980s. Although the complex procedure is lengthy, taking about one year to knock out one gene in a mouse, it opened up huge opportunities in functional genomics. Guided by earlier work with Drosophila, mouse researchers were able to pick specific genes and knock them out. In this way they were able to find that wnt- 1 in mice, like wingless in flies, is used in a variety of contexts where cell communication is required. Gene knock-out technology has introduced a new conceptual approach in genetics, that of “reverse genetics”. Reverse genetics starts with a gene predicted from sequencing information and works its way back towards function by assessing the effect of knocking it out. In principle, this could be done for all the predicted genes in the mouse genome. However, with current methods this would be ethically unacceptable and exceedingly costly. More recently, methods of reverse genetics have been developed for other model systems, for example, by the use of RNAi. RNAs are molecules that play key roles in protein synthesis, including acting as intermediate codes between genes and proteins. RNA interference, RNAi, was invented by worm researchers who found that injection of double-stranded RNA with a sequence matching that of a specific gene could drastically reduce the amount of protein encoded by this gene. Similarly in zebrafish and frog embryos, gene expression can routinely be knocked down by injection of special RNA-like molecules called morpholino oligonucleotides, that are designed to prevent protein production from specific genes. This procedure enables a rapid assessment of gene function in a vertebrate. At the same time, RNAi technology has revolutionized cell culture-based investigations. As long as a process can be accurately reproduced in cell culture, for example virus entry into a cell, the requirement of any gene can be tested by simply adding interfering doublestranded RNA. Libraries of interfering RNAs for all the genes in different organisms, including flies, are being assembled. In an optimized setting these can be applied to cells and the effect assessed in about one week, thus identifying genes required for a specific process.
These advances with cell culture and zebrafish embryos are clearly very important. Do they make Drosophila obsolete? We think that continued advances in Drosophila genetics will ensure that it remains an essential partner in basic biomedical research. Indeed, some of these advances help to overcome current shortcomings of the new functional genomic approaches in cultured cells, worms or fish. One problem with cells in culture is that they cannot encapsulate the interactions that take place within a complex structure such as the brain. Therefore animal approaches, especially with tractable systems, will continue to be essential in the foreseeable future. Despite their effectiveness, morpholino oligonucleotide-based knockdown approaches also have limitations. Because they need to be injected in the early embryo, they allow very little control of when and where in the animal gene activity is abolished.
The latest techniques of Drosophila genetics make it possible to turn genes on or off at prescribed developmental times and in specific cells within the whole animal. This is analogous to being able to turn the power on and off in different parts of a factory and using this to assess what the different units do and when they do it. There are many reasons why it is useful to control when and where a gene is active. One of them is that many genes are deployed over and over again at different stages of an animal’s life cycle. For example, the wingless gene is essential for embryo development, but is also used later in the fly’s life to specify wing tissue. The later functions can only be seen either with very special mutations, that only affect the late function, or by engineering a delayed gene knock-out.
Methods for turning genes on in a controlled manner have been in place for many years in Drosophila. The most popular requires two transgenes, fragments of DNA that have been artificially inserted in the genome, that are brought together in the same animal. One is called the “driver” and encodes a transcription factor that recognizes control sequences in the other transgene, the “responder”, thus turning it on. Many driver strains that are active at different times and places have been generated over the years. Thus it is possible, for example, to turn on a gene only in the developing wing after completion of the embryonic stages. If the responder encodes a toxin, this technique can be used to destroy specific cells, a form of genetic surgery. For example, a small group of cells in the brain secrete insulin. It has been possible to remove these cells selectively and assess the effect on growth of the whole organism.
This technique has also been combined with RNAi to switch genes off in a controlled manner. In this case, the responder element encodes RNA that has been designed to act as interfering RNA against a specific gene. Yet another technique widely used for complete local knock-out is known as FLP/FRT and in its simplest form relies on two transgenes. Using this procedure it is possible to create recognizable patches of cells in which both copies of the relevant gene are mutated and lack gene activity. When and where such patches are generated is under the experimenter’s control. For instance, the activity of specific genes can be abolished only in the subregion of the brain involved in memory whereas all other brain parts are allowed to work normally. Importantly, in Drosophila, the FLP/FRT technique lends itself to sophisticated forward genetic screens. Random mutations are induced in flies that carry all the necessary components to generate patches of mutant cells readily. Mutants that display a desired defect are bred and used for subsequent identification of the corresponding gene. This approach has been used, for example, to identify genes that are required by photoreceptors in the eye to find their appropriate target in the brain where visual information is processed. Nerve processes emerging from the eye must be able to know when to advance, turn and stop in order to make the right connections. Using an FLP/FRT screen, scientists have identified molecules that are required by the photoreceptors during this complex navigation feat. Since flies with patches of mutant cells grow up into adults, this approach can also be combined with behavioral tests, for example to determine whether mutant flies can properly detect ultraviolet light.
Because more and more genome sequences are available, it has become easier to transfer knowledge from one animal species to another. More than ever, it is important to choose the best model organism to address a specific question. The short generation time of Drosophila has spearheaded early studies of heredity. This characteristic has also made sophisticated genome engineering possible, thus promoting the development of an ever expanding variety of techniques. Clearly no single organism can be relied upon to reflect all the activities of humans, but because of its distinct features, Drosophila will continue to be an essential partner in the discoveries that will lead to medical advances in the twenty-first century.