Contents
-Home
-Welcome Letter
-Purpose of Workshop
-Program
-Presentations

Appendix
-Participants
-Steering Committee
-Program Committee

Workshop Report
-Preface
-Final Report

Gary H. Thorgaard

School of Biological Sciences and Center for Reproductive Biology

Washington State University

Pullman WA 99164-4236

Email: thorglab @ wsu.edu

ABSTRACT

Progress in the mapping of genes and traits in aquaculture species has lagged behind that in agriculturally important plants and terrestrial farm animals. This may in part reflect a lack of resources applied to aquaculture species but may also be due to a tendency to uncritically adopt approaches that have been applied to farm animals for use with aquaculture species. Most aquaculture species, however, share attributes such as availability of wild populations, tolerance to genetic manipulation and large family sizes with plants. More rapid progress may be possible with aquaculture species by exploiting such attributes. Studies mapping genes and traits in rainbow trout are described as examples of exploiting such attributes.

INTRODUCTION

Methods for detecting high levels of molecular variation in populations of plants and animals have made it possible to develop detailed genetic maps in a variety of agriculturally important species. Such maps allow comparisons in the order of genes on chromosomes among species. Associations of traits with particular chromosome regions can also be performed in crosses within species (QTL analysis). Such studies ultimately should provide associations of traits with specific genes and may lay the groundwork for marker-assisted selection programs.

However, mapping and QTL studies tend to be expensive and laborious to perform. At this time, genetic maps and QTL analysis for aquaculture species are lagging well behind those for important farm animals (cattle, swine, chickens) and agriculturally important plants. Most aquaculture species share certain attributes with agriculturally important plants. By adopting approaches used with plants, it may be possible to increase the rate of mapping progress in aquaculture species.

Considerable progress has been made in applying modern genetic approaches to the breeding of plants. Some of the attributes that have facilitated this progress include: (1) availability of wild, non-domesticated strains with distinctive agricultural attributes, (2) ability to genetically manipulate plants to produce pure-breeding lines which can be propagated as reference lines, (3) ability to genetically manipulate plants to generate distinctive mating designs (e.g., production of doubled haploids) that facilitate mapping and QTL analysis, and (4) ability to produce and rear large numbers of progeny at a low cost, facilitating mapping and QTL analysis.

Progress in work with farm animals has tended to follow the human genetic model more closely. Inbred lines tend to be used only rarely in research with farm animals. As for humans, microsatellites have been the primary type of genetic marker utilized in most farm animals. In comparison to some of the markers used in plants, microsatellites require substantial costs for marker development. In relation to the attributes which have facilitated progress in plant genetic analysis, in farm animals (1) wild strains are often unavailable, or if available, may be considered to be too divergent to be useful for introgression of desirable attributes, (2) homozygous lines cannot be readily produced in most farm animals other than by many generations of brother X sister mating, (3) mating designs such as doubled haploid analysis are not possible due to the inability of mammals or birds to tolerate androgenesis or gynogenesis, and (4) the fecundity and rearing costs of most farm animals limit the ability to rear large numbers of progeny at low cost.

Our laboratory at Washington State University has pursued an approach somewhat similar to that utilized in plants for the genetic analysis of rainbow trout, Oncorhynchus mykiss, an important aquaculture species. We have collected genetic material from hatchery rainbow trout populations as well as populations outside the region, northern California, which was the source of most hatchery rainbow trout. Sperm was used to produce homozygous diploid trout by androgenesis (Parsons and Thorgaard, 1984, 1985). Such homozygous individuals were then used to produce clonal lines of genetically uniform individuals in the next generation by androgenesis and gynogenesis (Scheerer et al., 1986, 1991) whose clonal identity has been confirmed by DNA fingerprinting (Young et al., 1996). Crosses have been made between lines to produce vigorous hybrids (Young et al. 1995) which were then used to generate doubled haploid progeny by androgenesis (Young et al. 1998). Genetic typing of these doubled haploids allowed a detailed genetic map to be developed (Young et al. 1998) which is now being related to a microsatellite map which has been developed for this species (Sakamoto et al. 2000). Comparisons among the five lines currently being reared indicate that there are substantial differences in a variety of traits among the lines, including immune response (Ristow et al. 1995) and karyotype (Ristow et al. 1999). One trait showing significant differences among lines is development rate; lines originating from Alaska and Idaho showed faster development to hatch than did typical hatchery rainbow trout lines (Robison et al. 1999). A QTL analysis using doubled haploid individuals produced by androgenesis from a hybrid between a female line of hatchery origin (OSU) and a male line of Alaskan origin (Swanson) indicated that a major locus for development rate is located on one chromosome in rainbow trout (Robison et al. 2001).

PROSPECTS FOR USE OF MARKER-ASSISTED SELECTION IN AQUACULTURE SPECIES

In contrast to most farm animals, aquaculture species have several plant-like attributes. These include: (1) availability of wild strains of farmed species, (2) tolerance to genetic manipulations allowing generation of homozygous lines, (3) tolerance to the same manipulations allowing doubled haploid analysis, and (4) capacity to rear large numbers of progeny per mating. The implication of availability of wild strains is that a broad range of genetic diversity is available for sampling and study. The greater the magnitude of differences between strains, the greater the chance that the genetic basis of such differences can be mapped. Tolerance to genetic manipulation facilitates the development of defined, uniform lines which facilitate research. Mating designs such as doubled haploid analysis are also facilitated, which allows easier mapping of genes and traits. Finally, the availability of large family sizes allows more precise mapping of genes and traits. Together, all of these advantages suggest that it may be easier to map genes and traits per unit of labor and expenditure in aquaculture species than in terrestrial farm animals. The longer-term goal of using marker-assisted selection to improve aquaculture strains is likely to be achievable, given these advantages.

CONCLUSIONS AND RECOMMENDATIONS

The model that has been successful in breeding and genetic analysis of crop species may be very applicable to many aquaculture species. Wild strains of aquaculture species which have distinctive attributes are likely to be available. Collection of genetic material from such strains and performing crosses to understand the genetic control of these traits should be high priorities. Genetic manipulations such as androgenesis and gynogenesis may facilitate such genetic analyses. The high family sizes of aquaculture species may allow precise localization of loci affecting the traits and, ultimately, may facilitate molecular characterization of changes underlying the phenotypes.

Short-term (1-3 years)

2. Identify critical genetic resources in these species.

3. Collect and bank genetic resources in these species.

Mid-term (4-7 years)

REFERENCES