Abstract
The sequence of the Drosophila melanogaster genome presented in this issue of Science is the latest milestone in nine decades of research on this organism. Genetic and physical mapping, whole-genome mutational screens, and functional alteration of the genome by gene transfer were pioneered in metazoans with the use of this small fruit fly. Here we look at some of the instances in which work on Drosophila has led to major conceptual or technical breakthroughs in our understanding of animal genomes.
In 1910, T. H. Morgan, having chosen Drosophila for his studies of heredity, was rewarded with the first of many mutants, a white-eyed fly. Morgan was soon joined in the famous Fly Room at Columbia University by three principal students, A. H. Sturtevant, C. B. Bridges (see Fig. 1), and H. J. Muller. Within the space of 5 years, they formulated a revolutionary chromosome theory of heredity (1). Their accomplishments, which led to Morgan winning the Nobel Prize in 1933, are all the more remarkable because their sole experimental method was to do controlled crosses with these mutants and count progeny.
(A) Bridges (left) and Sturtevant in 1920. (B) Morgan in 1917. The photo of Morgan, who was camera shy, was taken by Sturtevant using a camera hidden in an incubator and operated remotely by means of a string. The books and microscope in the background were at Sturtevant's desk (1). Both photos courtesy of the Archives, California Institute of Technology.
In 1913, Sturtevant constructed the first genetic map and showed that genes are arranged in a linear order (2). In two papers published in 1914 and 1916, Bridges, exploiting chromosome nondisjunction in XXY females, provided the elegant first proof that chromosomes must contain genes; this ruled out the alternative possibility, assumed by some at the time, that chromosomes and genes were separate hereditary elements (3). In 1918, Muller introduced the use of balancers, chromosomes bearing inversions that allow the stable maintenance of lethal mutations as heterozygotes in a manner that does not require selection (4); this is only now becoming possible in the nematode and is still not possible in mice.
The physical mapping of genes has its roots in the discovery by Heitz and Bauer in 1933 of salivary gland polytene chromosomes in the flyBibio hortulanus (5). Polytene chromosomes could easily be seen in the microscope because, after numerous rounds of replication, the chromosomes remained aligned and were patterned in cytological bands. T. S. Painter at the University of Texas promptly realized their importance and in 1934 published the first drawings of Drosophila melanogaster polytene chromosomes, which included the chromosomal localization of several genes (6). In 1935 and 1938, Bridges published polytene maps of such accuracy that they are still used today (7). Making extensive use of chromosomal rearrangements, Bridges also constructed cytogenetic maps that assigned genes to specific sections and even specific bands (see Fig. 2A). We now know that these maps are often accurate enough to place genes within intervals of less than 100 kb.
(A) Corresponding points in the polytene chromosome map and the linkage map for the tip of the second chromosome [modified from (35)]. The region shown covers about 5 Mb of DNA. (B) In situ hybridization (36) of a cloned segment of Drosophila DNA to polytene chromosomes, demonstrating the first mapping of a cloned gene to its chromosomal location [modified from (13)].
In 1927, Muller showed that ionizing radiation causes genetic damage and that mutations, including chromosomal rearrangements, be induced with x-rays, a finding for which he received a Nobel Prize in 1946 (8). In the late 1930s, two groups demonstrated the feasibility of generating deficiencies and duplications by combining x-ray–induced chromosomal aberrations with closely spaced break points (9). This method was systematically exploited by D. L. Lindsley, L. Sandler, and 14 co-workers in 1970 to generate an ordered set of duplications and deletions spanning the major autosomes in ∼500-kb segments (10). This work initiated the concept of whole-genome scanning in metazoans for phenotypic perturbations; such a resource has never been duplicated for any other metazoan. It is interesting to note that this genome-wide effort occupied a higher percentage of the total Drosophila research community at the time than has the current genome project.
The foundation for modern genome research can be traced to a grant application (11) written in 1972 by D. S. Hogness of Stanford University. Anticipating the first successful cloning of eukaryotic DNA a year later (12), Hogness proposed using large insert clones to construct physical maps of whole chromosomes to facilitate the detailed study of chromosome structure (Fig. 3). The first random clones of any organism were generated in the Hogness laboratory in early 1974, and a cloned DNA segment was mapped to a specific chromosomal location a few months later (13) (see Fig. 2B). By early 1975, clone libraries representing the entire genome had been generated (14) and screened for clones carrying specific sequences (15) with the newly developed method of colony hybridization (16). Overlapping segments of chromosomal DNA cloned in bacteriophage lambda (17) covering more than 200 kb were constructed by “chromosome walking” by the end of 1978 (18, 19). An inversion that linked this region to the Bithorax complex of homeobox genes was used to achieve the first positional cloning of a gene, Ultrabithorax, in early 1979 (18, 20). By late 1980, many mutant alleles had been located on the restriction map of the complex and shown to be the result of chromosomal breakage or transposable element insertion (21) (see Fig. 4).
Diagram taken from D. Hogness's 1972 grant application (11) showing his proposed strategy for making a physical map of a whole chromosome, starting with ordering large insert clones based on the F factor [now known as bacterial artificial chromosomes (BACs)] and then subcloning each of these into bacteriophage lambda or plasmid vectors. “One could then obtain a set of overlapping segments covering all the DNA in the chromosome, and the overlaps between segments could be detected and mapped… . In this way, many of the sophisticated physical techniques can be applied in an ordered manner to specific segments of a Drosophilachromosome” (11).
Poster displaying a partial map of the Bithorax complex displayed at the Stanford University Biochemistry Department retreat at Asilomar, California, in late 1980. Note the molecular mapping of various mutant alleles relative to the scale in kilobases derived from the restriction map of the cloned region.
In 1980, C. Nusslein-Volhard and E. Wieschaus extended to animals the use of a systematic genome-wide mutational screen to attempt to identify all genes involved in a fundamental process (22), a feat that had previously been attempted only in microorganisms. Their work on embryonic development soon led to the discovery of the components of most major signaling pathways, as a new generation of fly workers were eager to isolate and sequence the genes defined in their screen using the techniques of positional cloning and transposon tagging (23). In recognition of this work, Nusslein-Volhard and Wieschaus shared the 1995 Nobel Prize.
An important breakthrough for manipulating the genome was made in late 1981 when methods for making transgenic flies with the use of transposable element vectors were developed and used to achieve the first rescue of a mutant phenotype in an animal by gene transfer (24). The availability of stable, single-copy, integrative transgenesis enabled a range of powerful techniques to be developed in Drosophila, many of which have since been adapted to other metazoans. These methods include the use of enhancer traps to screen for genes based on their pattern of expression, developed in 1987 (25), large-scale insertional mutagenesis with engineered transposable elements, developed in 1988 (26), site-specific recombination for generating chromosomal rearrangements, developed in 1989 (27), and two-component systems for controlling ectopic gene expression, developed in 1993 (28).
Ironically, the success in cloning and studying individual genes dampened enthusiasm for an organized genome project, which was seen as unnecessary. Over 1300 genetically characterized genes—nearly 10% of all the genes in Drosophila—have been cloned and sequenced by individual labs (29). This is over twice the percentage of genes in any other animal for which both the loss-of-function phenotype and sequence have been determined. Nevertheless, for flies (30) as well as other animals (31), less than a third of genes have obvious phenotypes when mutated, emphasizing the critical importance of genome sequencing as a gene discovery method.
The annotated sequence of the Drosophila genome reported in this issue (32) is the product of both publicly and privately funded efforts and is the first application of the whole-genome shotgun approach (33) to the sequencing of an animal genome. It provides a model for the large-scale annotation of a genomic sequence, which was accomplished through the concerted efforts of 40 experimental and computational biologists from 20 institutions in five countries. These sequencing and annotation efforts follow the collaborative tradition of Drosophila research established over 80 years ago; as observed by J. Schultz, “it derives from Morgan, and paradoxically has not so much to do with cooperation as with the paramount importance attached to getting on with the work. I cannot recall any instance of explicit discussion of the value of cooperation; it was always taken for granted, and taught by example” (34).
