Historical Introduction
The avian genome has been manipulated for at least 3000 years, and specialized breeds and strains of domestic birds predate the recognition of genetics as a scientific discipline by at least 2000 years. It is possible that breeds of chickens were established by poultry breeders in China as early as 1400 B.C. (Darwin, 1896) and it is clear from the writings of Aristotle, Columella, and Pliny that Greek and Roman poultry breeders had created strains for the production of meat and for their ability to perform in the cockpit (Crawford, 1984). Darwin (1896) produced the first extensive catalog of poultry stocks around the world, and it is abundantly clear that many specialized and distinctive breeds had been developed in the Orient, in Europe, and in the Americas. The breeding and development of exhibition poultry, cage birds, and pigeons continues unabated today by many individuals who have little or no formal training in genetics but who continue to manipulate the avian genome in remarkable ways. For example, the Birmingham roller pigeon was created in the 1940s by Mr. Pensom, a bus driver from Birmingham, England, and a group of equally dedicated hobbyists from the area. These gentlemen selected a group of pigeons from a breed possessing a behavioral modification known as tumbling that was also described by Darwin (1896). Tumblers lose the ability to fly at about 6 months of age because the take-off hop initiates a backward roll in the air. The Birmingham group of amateur geneticists, however, selected birds who could fly and roll, and the descendants of these birds provide aerial acrobatics for thousands of devoted breeders of the performing roller.
When genetic selection was practiced prior to the development of a theoretical basis to guide the pragmatic breeder, progress relied entirely on the fortuitous āsportsā and āfreaksā that appeared from time to time. These animals were carefully nurtured by the breeder, who mated together pheno-types that most closely resembled the ideal animal. The ideal was originally defined in the mind of the breeder; subsequently, breed organizations formed to formalize the definition in writing. The remarkable diversity that ensued is a tribute to the diligence and stockmanship of generations of poultry enthusiasts. The legacy of mutations, strains, and breeds that we have inherited from these collectors is providing the tools to unravel the physiological processes involved in the replication, transcription, and translation of the message encoded in DNA.
Manipulation of the Avian Genome as Genetic Theory and Practice were Developed
The plethora of genetic variants in the poultry population played an important role in the understanding of inheritance from a Mendelian perspective and in the application of these principles in medicine and agriculture. For example, the first proof that Mendelās laws applied to animals as well as plants was presented to the Evolution Committee of the Royal Society by William Bateson, who described the inheritance of the rose- and single-comb genetic variants in the head furnishings of chickens (Bateson, 1902). While the rose comb had been known for some time as an esthetically pleasing genetic variant, it also became one of the first quantitative trait loci (QTL) to be investigated, when it was used in the selection of chickens whose comb resisted freezing in harsh northern climates. This goal was accomplished by applying the information published by Bateson and Punnett (1905) that demonstrated epistasis in animals for the first time. Their work identified the genotypes of single, rose, pea, and cushion combs as rrpp, R-pp, rrP-, and R-P-, respectively. The quantitative goal of increasing winter hardiness was achieved by incorporating the P and R alleles into a new breed, the Chantecler, whose small cushion comb reduced heat loss and resisted freezing (Cole, 1922).
The rose comb phenotype continued to be attractive to a number of poultry breeders and is a characteristic feature of the Wyandotte breed, which contributed significantly to the formation of modern broiler strains. It was again studied as a QTL when it was recognized that fertility in all rose-combed breeds was poor. Eventually, the source of infertility was attributed to a recessive effect of the dominant allele at the rose comb locus (Table 1). This relationship between a single functional unit of inheritance and a quantitative trait was the prototype for the current investigations of the relationships between sequences of DNA that can be identified using the techniques of molecular biology and traits that are of commercial importance to the poultry industry (see Chapter 16).
Table 1
Effect of Comb Genotype on Fertility in Chickens
Genotype of sire | Phenotype of sire | Fertility following natural mating | Fertility following artificial insemination |
RR | Rose | 78 | 44 |
Rr | Rose | 91 | 84 |
rr | Single | 92 | 82 |
Data from Crawford and Smyth (1964).
The antiquarian geneticists who conserved and propagated the rose comb gene, with their keen eye and genetic skill, should be recognized not only for their contribution to the fundamental understanding of inheritance and the relationship between quantitative and qualitative traits, but also for the provision of the first autosomal locus to be placed on the genetic map of the chicken. In 1928, Serebrovsky and Petrov demonstrated autosomal linkage in the fowl for the first time when they reported that the rose comb gene was separated from the creeper gene by 9.1 crossover units. This map has been expanded during the past 63 years, and the most current edition (see Chapter 5) shows that markers identified using the tools of molecular biology, such as the ev loci, are being integrated into the map of morphological traits that was assembled during the first half of this century. At the same time, our knowledge of chromosome structure and function has expanded rapidly, and this body of information, together with recent advances in the field, have been summarized in this volume by Bloom et al. (see Chapter 4).
The shift in emphasis from morphologically defined traits to those that are physiologically defined and, more recently, to those that are defined by their base-pair sequences, has been the consequence of developments in the tools available to study inheritance. A good example of this change in emphasis can be traced in the development and use of the Sebright bantam. In about 1800, Sir John Sebright incorporated the rose comb, lacing, dwarfing, and hen-feathering genes into a single breed that today bears his name. In the eyes of Sir John and the devotees of the breed who have continued to perpetuate the Sebright bantam, the combination of mutations produces a work of art. Following the recognition that hen feathering (i.e., the female plumage structure of Sebright and Campine males) was inherited in a dominant Mendelian fashion, however, these breeds became the object of intense investigation by physiologists seeking an understanding of sexual differentiation (see review by Wilson et al., 1987). These investigations were stimulated by the apparent similarity between idiopathic femininization of genetic males in humans and the predictable femininization of the plumage of hen-feathered Sebright roosters. Many elegant, but indirect, studies conducted from 1920 to 1960 demonstrated that the gonadal secretions of the hen-feathered Sebright are normal and that the gene acts at the feather follicle to femininize the growing structure. Further definition of the aberrant physiological system that produces henfeathering awaited the development of an aromatase assay that was sufficiently sensitive to detect the massive increase in the ability of this enzyme to convert androgens to estrogen in the skin (Wilson et al., 1987). These studies showed that the growing feather was femininized because the activity of the androgen-binding cytochrome P-450 oxidase in males bearing the Hf allele is increased 20-fold in the heterozygote and 40-fold in the homozygote hen-feathered rooster. From the point of view of the physiologist who is concerned with enzyme function, therefore, the gene is co-dominant because aromatase levels in the heterozygote are intermediate between those of the homozygous parents. From the point of view of the breeder who is concerned with phenotype, however, the trait is dominant because the heterozygote produces sufficient aromatase to femininize the plumage of the heterozygote male.
Subsequent investigations have revealed the sequence of the mRNA and corresponding DNA that code for the extraglandular aromatase, and it would appear that the increase in activity of this enzyme in hen-feathered males is due to a retroviral promoter that has inserted into this gene (Matsumine et al., 1991). Furthermore, in situ hybridization has revealed that this gene is located on chromosome 1 (Tereba et al., 1991). This elegant work provides an excellent example of a morphological trait that can be related to a protein product that, in turn, can be related to a specific segment of the genome located on a specific chromosome in the chicken.
Unfortunately, our knowledge of the physiology and biochemistry of most traits of interest is far less comprehensive than our understanding of hen-feathering. In most cases, breeders manipulate the genome to alter physiological function with little, if any, understanding of the physiological systems that have been altered. Growth rate, for example, is easy to manipulate on a phenotypic basis, although the genetic basis of this massive physiological change is only beginning to be understood. The development of our understanding of the physiological systems that are manipulated by selection for the end product will be gained using the tools of molecular biology to explain morphology in terms of the base-pair sequences coding for the enzymes, hormones, and structural proteins that interact in a series of intertwined metabolic pathways. In this volume Dr. Goddard and his colleagues have assembled recent information concerning our knowledge of some of the hormonal mechanisms that regulate growth in poultry.
Our current inability to connect biochemical descriptions of genes with the trait we would like to manipulate is the major impediment to using transgenic systems. It could be of significant economic advantage, for example, to transfer the sex-linked dwarfing gene from the chicken to the turkey, because the dwarf broiler breeder hen produces more chicks using less feed (see review by Decuypere et al., 1991). If an extrapolation of the phenotypic action of this gene from the chicken to the turkey is justifiable, it is realistic to expect a reduction in feed consumption of about 10 ā 15%. During the lifespan of a turkey breeder hen, this would equal 10 ā 15 kg of feed. One of the major factors that prevents the execution of an experiment to test this extrapolation is the lack of information regarding the molecular biology of the gene and the gene product that limits growth in dwarfs. In a perfect world, the geneticist would have a genetic blueprint of the physiological functions governing the desired phenotype. The fields of genetics and physiology have evolved separately, however, and our current knowledge is a fragmented collection of facts drawn from a number of different perspectives. The objective of many individuals during the next decade will be the unification of the fragments into a useful description of the genetic code at the molecular level, the development of reliable predictions of the cellular consequences of altering the code, and, finally, the manipulation of the phenotypic expression of the genetic code in the whole animal. When this molecular understanding of genetics and the physiological consequences of genetic rearrangement is realized, manipulation of the avian genome can be undertaken with unprecedented precision.