Drosophila: eye genetics and mosaic analysis

Drosophila is arguably the most versatile and one of the most powerful eukaryotic genetic model systems. Flies are easy to maintain and breed in large numbers in the lab and have a generation time of 10 days (at 25 C). The genome is relatively compact (1.7 x 108 bp and has been completely sequenced - the annotated version has been freely available on the web since March 2000.

In order to illustrate general principles of Drosophila research techniques in these lectures I will focus on one particular aspect of the fly's biology, the development and function of the compound eye,. . Drosophila has a typical insect compound eye. Each eye is composed of several hundred simple units called ommatidia arranged in an extremely regular array. In the fly each ommatidium consists of a core of 8 photoreceptor cells (R1-R8) surrounded by 4 cone cells (equivalent to the vertebrate lens in function) pigment cells and a sensory bristle. The number of cells, their identities and functions within each ommatidium is invariant.

The eye develops from the eye-antennal imaginal disc. Not all ommatidia form at the same time. Differentiation occurs in posterior to anterior sequence: the first ommatidia to differentiate do so at the posterior pole of the eye disc, the last at the anterior pole. The morphogenetic furrow marks the boundary between dividing, non-differentiating cells and differentiating ommatidia. The further posterior ommatidia are from the furrow the closer they are to having all cell types (R1-R8 cone cells and pigment cells) differentiated. Cells anterior from the furrow divide asynchronously, while cells in the furrow are non-dividing. Immediately after leaving the furrow cells go through two last synchronous divisions. This strict control of cell division makes the eye an ideal system for analysing the developmental effects of cell cycle genes. Programmed cell death is another important component of the developmental process - once all ommatidia have differentiated any cells which have not been incorporated into ommatidia die.

The cellular composition of each ommatidium is as invariant as that of an adult C.elegans and even simpler as it consists of even fewer cells. However unlike C.elegans the lineage of ommatidial cells varies both between ommatidia and between individuals in a way that suggests that lineage plays no part in determining ommatidial cell fate.

The eye is an effective model for processes including cell signalling, neuronal connectivity, control of cell proliferation and vesicular transport. Many different types of mutant have been isolated affecting for example, the size, the pigmentation and the morphology of the eye. The second half of the lecture focuses on mutations that affect the development of one specific photoreceptor the R7 cell. The R7 photoreceptor is the only one which is sensitive to UV light. This make it possible to screen for R7 defects using simple behavioural screen for flies which fail to show a phototactic response to UV. Two mutants isolated in this way proved to lack the R7 photoreceptor -it never develops. The relationship between these two mutants, sevenless (sev) and bride of sevenless (boss), was initially worked out using the classical genetic technique of mosaic analysis.

 

Generating genetically mosaic eyes by X-radiation induced mitotic recombination

Recombination between homologous chromosomes is well known to occur during meiosis. It can also occur during mitosis, although usually at a very low frequency. In flies it is possible to increase the rate of mitotic recombination by several orders of magnitude through exposing flies to a short pulse of X-irradiation. This can be used to induce clones of cells which are homozygous for one or more mutations in flys in which most cells are heterozygous for these mutations. Assume we have larvae which are heterozygous mutant at a fully recessive marker gene locus (a good example of such a locus is the eye pigment gene white). Usually when cell of these larvae divide they give rise to two daughter cells which are heterozygous like their parents - both daughters can be said to show the parental combination of alleles. X-ray induced recombination between the chromosome carrying the wild type and the chromosome carrying the mutant allele of white can however produce one recombinant daughter cell which is homozygous for the wild type allele and one recombinant daughter cell which is homozygous for the mutant allele. Both recombinant cells will go on to produce "clones" of progeny with the same homozygous genotype in the adult fly. As the marker gene is recessive, cells of the homozygous wild type clone will be indistinguishable from the heterozygous cells around it. But cells of the homozygous mutant clone will be visibly different from their neighbours making it possible to see what combination of cell types a single progenitor cell can give rise to.

Induced mitotic recombination is an important technique because it can not only be used to produce genetically marked clones of otherwise normal cells, but also genetically marked clones of mutant cells. This is because recombination involves not just single genes but entire chromosomal segments distal to the recombination breakpoint. Somatic recombination can be induced in larvae or embryos that are heterozygous mutant both for a marker gene and developmentally interesting genes which map distal to the marker gene. In adults subjected to this treatment marked clones will be homozygous mutant for the marker gene and for the developmental gene(s)and the phenotype of the clone can be analysed for defects (such as never including an R7 cell). Mosaic analysis can be used to:-

 

  1. Analyse the effect of recessive embryonic lethal mutations on adult structures like the eye. To do this clones are induced after embryogenesis is complete.

 

  1. Show which cells need to express a gene for the product of that gene to be effective. For example, in developmental processes that depend on cell-cell signaling, genes in the pathway that generates the signal need to be active (wild type) in the cells that secrete the signal, while genes involved in transduction of the signal (e.g. the signal receptor) need to be active in the cells that respond to the signal.

 

Mosaic analysis of sev and boss shows sev is required in the R7 cell and boss in the R8 cell. R7 is normally the last photoreceptor to develop and R8 the first. R7 and R8 are adjacent in developing ommatidia. The results of the mosaic analysis imply that the R8 cell is needed for R7 to form. A simple theory, which has been proved correct, is that R8 produces a signal (coded for by boss) that activates a receptor on a neighbouring cell (coded for by sev) instructing that cell to become the R7 cell.

 

References

http://www.sdbonline.org/fly/aimorph/eye.htm

 

Thomas BJ and Wassarman DA (1999)

A fly's eye view of biology

Trends in Genetics vol 15 pp184-190