Breeding Techniques
To some people, plant breeding may simply mean crossing two flowers in the hope of producing a new cultivar. In some cases, this may in fact give the desired result. Genetic improvement of plants is a key activity of many research institutes and a priority for most of our fruit industries. This is the second in a series of three articles which aim to give growers a better understanding of what is involved in plant breeding programmes.

Introduction
The previous article (1) explained what is involved in developing and maintaining a germplasm collection. The germplasm collections provide a crucial foundation for plant improvement programmes undertaken by HortResearch. This article describes how breeders gain information on the inheritance of characters of interest, and then use this information to assess alternative breeding strategies and to select parents of high breeding value.

The simplest method of plant breeding is what humans have been doing for centuries: selecting the best plants from natural and local germplasm and growing them in preference to other cultivars (cultivated varieties). The next level up is making controlled crosses between individual cultivars that are exceptional in one or more characters. This method is very effective and relatively simple where qualitative traits are involved. That is, where development of the trait or characteristic is controlled by the action of a single gene, such as in colour and form of flowers.

The majority of desirable traits, such as plant height, number of fruit, yield etc are controlled by several genes, each contributing to the formation and expression of the trait. The action of these genes is usually additive: that is, the trait expressed represents the sum of positive and negative effects of the individual genes. These are referred to as quantitative traits, and because of the additive effect of the genes and the influence of environmental factors on them, a full range of trait characteristics may be observed. For example, plants don't just come in two sizes, short and tall, they cover every size in between.

Population Improvement
Quantitative genetics involves understanding the inheritance of fruit and tree characteristics to determine appropriate breeding strategies for population improvement. Traditionally, in fruit breeding, contemporary cultivars have been crossed and new selections made from the families of seedlings that have either one parent or both parents in common. This strategy has been successful, particularly when very large numbers of seedlings have been raised (e.g., 40,000 seedlings of one particular cross). Plant species contain thousands of genes, and large numbers of crosses are usually required to ensure that every possible combination of genes has a chance of being expressed. However, this method gives very little genetic information to feed back into the programme. There is often no long term strategy and usually few of these seedlings contribute to further breeding generations.

Strategies of population improvement, using `recurrent selection', aim at increasing the frequency of favourable genes in a population progressively over successive generations, while maintaining reasonable levels of genetic variability. With an appropriate crossing design, useful genetic information can be obtained which can be used to select the best combinations of parents for future crosses and to choose the selection strategy that will give the greatest genetic gain for each character or trait.

Genetic diversity is very important as use of only a small number of cultivars as parents eventually leads to high levels of inbreeding and consequent loss of genetic variability and potential for further improvement. Inbreeding depression reduces vigour and increases undesirable traits. This narrow genetic gain can also give wide susceptibility to environmental factors or pests and diseases.

Two thirds of the apple cultivars released in the world over the last 30 years have originated from just five parents. This means that apple breeders around the world are working with a very narrow genetic base. Their breeding objectives are confined to the development of improved existing cultivars and not to the creation of entirely new types of apples. Over the past five years, HortResearch has developed the first true apple breeding population in the world by gathering 500 genetically unrelated families. The aim is to broaden the genetic base used in apple breeding and maintain high-quality breeding lines for future development of new commercial apples.

This is achieved by increasing the frequencies of desirable genes in the population, step-by-step for a range of important characters. Thousands of seedlings are planted out from many different families, and replicated at several sites. Seedlings are then evaluated for several agronomic traits. The best individuals (one per family) are selected and crossed to produce the second cycle of the breeding population. At each cycle, improved breeding stocks are extracted from the breeding population and pair crossed to produce families from which commercial cultivars can be selected.

Similar, but smaller, breeding populations have been, or are being, established for kiwifruit, apricots, berryfruit and hops.

Increasing Ploidy
Breeders are also able to make use of a range of alternative breeding techniques or strategies. One of these is the manipulation of the numbers of chromosome sets (or ploidy). Colchicine is an alkaloid obtained from the root of the autumn crocus. It interferes with cell division in plants, so that all the chromosomes remain in one cell (giving double the number of chromosomes). Chromosome doubling in plants can be achieved by treating the vegetative apex of plants with colchicine (and some other chemicals). Doubling of an entire chromosome complement may result in an increase of cell volume and, consequently in an increase of plant parts (especially vegetative ones). This can be a useful tool in breeding and selecting for larger fruit size.

Organisms in which each complete genome (chromosome set) is present three times, are called triploids (3x). They are formed when an ovule with double the chromosome number is fertilised by pollen with a normal reduced chromosome number. When tetraploids (4x) are available, triploids are made by crossing tetraploids and diploids (2x). Triploid organisms may develop more vigorously than do their diploid equivalents. In some species, they are better than the tetraploids and are commercially used. Because of the odd number of chromosomes however, triploids are completely sterile.

With some crops (such as grapes, melon, citrus) sterility is an advantage which greatly promotes the commercial use of triploids as they have no seed. In citrus, seedless cultivars, especially of easy-peel mandarins, have good export potential to Asian countries such as Japan.

When hops are used to make beer, seeds create impurities in the brewing process, and they can be a real problem. In Germany, laws prevent the growing of male hops so accidental pollination and formation of seeds in the hops is avoided. In New Zealand, HortResearch has pioneered the development of naturally seedless 'triploid' hops. Triploid hops contain three sets of chromosomes instead of the usual two, and are infertile so they do not develop true seeds.

Most commercial fruit species are polyploid, sometimes at high levels (e.g., strawberry 8x, boysenberry 7x). Berryfruit and kiwifruit species cover a range of ploidy levels which can interfere with interspecific hybridisation breeding strategies. The importance of research on the genetics of these fruit species is well illustrated by a recent discovery in kiwifruit. All genotypes of Actinidia chinensis were thought, until recently, to be diploid (2x). However, some unexpected crossing failures led scientists in HortResearch to check the numbers of chromosomes in some of the introductions. Several were found to be tetraploid (4x) and the crossing failures had mainly resulted when we had unknowingly crossed tetraploid with diploid genotypes. This information is critical to our kiwifruit breeding programmes.

Mutation Breeding
Another technique available to breeders in their quest to develop new cultivars is mutation breeding or mutagenesis. A mutation is a change in the structure of a gene. In most cases, this is deleterious as it means that the gene doesn't produce what it should. However, mutations can result in valuable new traits. Spontaneous mutations happen in nature at a relatively frequent rate. In fruit trees, large numbers of spontaneous mutations have been recorded and used either directly as new cultivars or in breeding programmes. As most fruit crops are reproduced vegetatively, mutations may be used successfully as new cultivars (e.g., changes in fruit colour or time of fruit maturity).

Recent decades have witnessed intensive work on the induction of mutations by using irradiation, chemicals and other mutagenic agents. The frequency of induced mutations almost doubles those occurring naturally and they have been looked on as a powerful tool for the development of new cultivars. However, available mutagens cause not only changes in genes but also chromosomal aberrations, many of which are deleterious in their effect on the trait and on the entire organism. Consequently, there has been a limited number of induced mutations directly usable as new cultivars.

Discovery of spontaneous mutants or sports by observant growers has been an important means of cultivar improvement in many fruit crops such as apples and citrus. For example, the apples `Royal Gala' and `Imperial Gala' are natural mutations of `Gala'.

In kiwifruit, it is more difficult to trace the source of fruit variants because of the replacement cane pruning system. Unusual fruit found in the packing shed one season may not reappear on the vine the following season if the mutation occurred on a single cane removed during normal winter pruning. Desired improvements/differences would be difficult to detect on single canes. Nevertheless, natural mutations of `Hayward' kiwifruit have been observed in orchards in New Zealand. Gamma irradiation of dormant budwood has been used in New Zealand in an attempt to induce useful mutations in 'Hayward'.

Interspecific Hybridisation
Being able to cross different cultivars, races or species is important in increasing biological diversity. Numerous crosses between cultivars and species have been made which, due to incompatibility, mechanical barriers, or geographical isolation, would have never developed under natural conditions.

In most of our programmes, hybrids produced by interspecific crosses give us unique opportunities for studying the inheritance of important characteristics. For example, crosses between Asian and European pears give HortResearch scientists the opportunity to study the genetic control of ripening. Some interspecific Actinidia hybrids show segregation for fruit skin and flesh colour. Interspecific hybridisation is used in several of our programmes (e.g., apples and berryfruit) to bring in genes for pest or disease resistance. It is important to understand the genetic basis of these resistances for this strategy to be most effective.

Conclusion
Characteristics such as red or green skin colour or size of fruit are subject to changing consumer fashions, but there will always be a requirement for high fruit quality and long storage, for example. Once high-quality breeding populations with a high level of genetic diversity have been established, and some understanding gained about how various characteristics are inherited, breeders are in a much better position to be able to quickly and accurately develop new cultivars which meet changing consumer requirements. These requirements, and some of the methods and selection criteria used in the evaluation and selection of new cultivars, will be covered in more detail in the next article in this series.

References
Muggleston S. 1995. What is involved in Plant Breeding (Part I). The Orchardist of NZ. 68(9):49.

Source:
The Orchardist, November 1995, Vol: 68, Number: (10):48