Horticultural Plant Breeding: Past Accomplishments, Future Directions

Genetic improvement of agronomic crops through breeding, which typically are marketed as commodities, are grower directed. Breeding objectives principally involve increasing yield, often based on resistance to biotic and non-biotic stress. For example improvement in hybrid maize yields have relied on increasing yield stability under high populations. In horticultural crops, breeding objectives must be consumer directed because consumers make individual decisions on consumption, and make choices between different cultivars and alternate crop species. Indeed, many horticultural industries are based on a very few cultivars with unique qualities [‘Clementine’ mandarin, ‘Kerman’ pistachio, ‘Williams’ (‘Bartlett’) pear, ‘Dwarf Cavendish’ banana]. Unique quality rather than yield per se must be the overriding breeding objective. Successful examples of quality improvement through breeding include the creation of supersweet maize, based on the incorporation of the shrunken gene; sugary podded snap peas; seedless triploid watermelons; and the yellow-fleshed, aromatic, pineapple. These breeding innovations have essentially created new Industries and represent the future direction for horticultural breeding. A way must be found to incorporate consumer satisfaction in the selection process. Marketing attempts to obscure cultivar identification and change horticultural crops into commodities need to be resisted by the industry. HISTORICAL Plant breeding in the modern sense can be defined as purposeful genetic improvement. Yet genetic improvement of crop plants has an ancient tradition and in fact the greatest feat of plant breeding dates to Neolithic Revolution when our crop plants were domesticated starting about 10,000 years ago. Progress was achieved by selection of elite clones or lines of useful plants, mass growing of elite clones, followed by selection from naturally occurring seedlings. We recognize this technique as mass recurrent selection. A key step to progress in this technique, especially for tree crops, was developing methods of vegetative propagation based on offshoots, cuttings, and finally grafting to fix genotypes. The concept of lineage and prepotent parents had long been recognized in animal breeding and is found embedded in literature from Virgil to Shakespeare. The Greeks and Romans were long familiar with special cultivars of fruits and extolled their virtue. Pliny the Elder, 1 century CE, could write of the specific attributes of a number of apple, quince, and medlar clones. Plant breeding in the modern sense is largely a 19 century discipline (Goldman, 2000; Janick and Goldman, 2003) when experimental studies began to confront the problem of inheritance. Its experimental beginnings date to the 18 century; between 1760 and 1766, when Joseph Gottlieb Koelreuter carried out the first series of systematic experiments in plant hybridization using tobacco. Thomas Andrew Knight demonstrated segregation for seed characters of the garden pea but offered no explanation. Later with the help of his daughter he initiated hybridization within various fruit species and introduced a number of cultivars initiating the practical science of fruit breeding. Charles Darwin was the first to demonstrate and explain a mechanism (natural selection) that could account for the highly branched lineages that nature represents (what we now refer Proc. IS on Hort. in Asian-Pacific Region Ed. R. Drew Acta Hort. 694, ISHS 2005 62 to as evolution) but did not come up with a satisfactory theory of inheritance. Jean Baptiste Van Mons, a Belgian horticulturist practiced long term mass selection in pear from open pollinations and introduced a number of new cultivars. It remained for Gregor Mendel to demonstrate that morphological characters were controlled by factors (later termed genes) that interact to form a phenotype and segregated unaltered from one generation to the next. His famous paper on inheritance in the garden pea published in 1865 included what he called species conversion incorporating backcrossing. Between 1840 and 1880, the seedsman, Louis de Vilmorin carried out practical breeding improvement in vegetables using a method he called genealogical selection (progeny testing) by assessing an individual’s capacity to transmit characters based on lineage incorporating statistical analysis. The genetic revolution that followed the rediscovery of Mendel’s paper in 1905 had a revolutionary impact on plant improvement. Although breeders had unconsciously been using many appropriate procedures via crossing and selection in the 19 century, the emerging science of genetics, especially the fusion of Mendelian and quantitative genetics, put plant breeding on a firm theoretical basis in the 20 century. The relationship between genetics and post-Mendelian plant breeding is exemplified by a number of now routine breeding protocols. One is the extraction and recombination of inbreds combined with selection to produce heterozygous but homogeneous populations (hybrid breeding); backcross breeding in which individual genes can be extracted and inserted with precision and predictability into new genetic backgrounds; and disease resistance breeding based on an understanding that genetics can control disease reaction in plants and that host plant resistance can be an object of selection. What we now call “Conventional” plant breeding was based on a system of extensive germplasm exploration, sexual recombination followed by a series of selection strategies involving statistical techniques to separate genetic and environmental affects. The technique proved to be powerful and dynamic because plant breeding was shown to be evolutionary as improved plants from the breeders art became the parents of subsequent generations. Dramatic advances in biology in the second half of the 20 century introduced a third revolution involving biotechnology, a catch-all term that includes both cell and DNA manipulation. The conventional baseline was 1953, the date of the brilliant Watson and Crick paper on the structure of DNA. One pathway in biotechnology developed from a series of investigations into gene function and structure and another from the culture and physiology of cells using microbial techniques. Molecular biology introduced two new techniques that were to have a profound effect on plant breeding. One was the establishment of molecular markers which made possible genotypic selection and the other was the transfer of genes between non-related organisms via recombinant DNA (transgene technology). By the end of the 1990s herbicide and insect resistance had been transferred into cotton, maize, and soybeans. These advances were quickly adopted, and engendered great expectations for agriculture but the technique has come up against a rising chorus of opposition based on little more than fear of the unknown. The future course of this technology, at least for the short run, cannot be predicted with certainty. Plant breeding in the 20 century has accomplished a number of powerful achievements in agriculture. Perhaps the most important have been the increase in yield and adaptability of our major crop plants. At least half of the tremendous increases in yields of our major crops (as much as 6 to 7 fold) in the 20 century are attributed to genetic improvement. The development of photoperiod insensitive, short stemmed, fertilizer responsive rice and wheat—the Green Revolution of the late 1960s—is the high point of 20 century plant breeding. HORTICULTURAL CROP BREEDING The goal of this paper is to address the future of horticultural plant breeding. In this respect it may be enlightening to distinguish between horticultural and agronomic crops (Table 1) because the case can be made that there is a fundamental difference between these crops that may affect the future course of genetic improvement. The