Selection methods in plant breeding based on mode of reproduction

This article discusses Selection methods in plant breeding based on mode of reproduction. Some plants reproduce by (more or less strict) self-fertilization where pollen from a plant will fertilise reproductive cells or ovules of the same plant. Other plants only (mainly) allow cross-pollination where pollen from one plant can only fertilize a different plant. Asexual propagation (vegetative propagation)can also occur in plants (e.g. runners from strawberry plants) which gives a new plant which is genetically identical to its parent plant. All these differences change the way plant breeders work. Apomixis is the phenomenon that seed are produced, but in an essentially asexual way, so that parent and offspring belong to one clone just as in case of 'normal' asecual' propagation.

Importance of mode of reproduction

The mode of reproduction of a crop determines its genetic composition, which, in turn, is the deciding factor to develop suitable breeding and selection methods (what is meant is for instance: a self-ferilizing crop like wheat is bred by basically different methods than a cross-fertilizing plant like rye grass; this is not a discussion on e.g. biotechnology vs. organic work, this is all within classical, traditional breeding). Knowledge of mode of reproduction is also essential for its artificial manipulation to breed improved types. Only those breeding and selection methods are suitable for a crop which does not interfere with its natural state or ensure the maintenance of such a state. It is due to such reasons that imposition of self-fertilization on cross-pollinating crops leads to drastic reduction in their performance. Likewise, it is practically impossible to maintain permanent heterozygosity in self-fertilizing crops rendering the development of hybrids an unattractive breeding approach. Asexual propagation is another type of reproduction where any plant or part of it can be used for multiplication without even a slight genetic change from generation to generation. The methods of breeding and multiplication for such crop are thus entirely different than those of sexually reproducing crops.

For teaching purpose, plant breeding is presented as four categories: Line breeding (autogamous crops), population breeding (allogamous crops), hybrid breeding (mostly allogamous crops, some autogamous crops), clone breeding (vegetatively propagated crops). Within these categories, there are several breeting methods possible. Moreover, there is a wide spectrum of techniques and technologies to make several steps in these methods easier, faster, more precise, or allow them at all, microspore culture, embryo rescue, mutagenesis, Agrobacterium-mediated gene transfer, fingerprinting, tilling, QTL analysis etc.

Self fertilizing crops (autogamous crops)

Certain restrictions caused the mechanisms for self fertilization (partial and full self fertilization) to develop in a number of plant species.

Some of the reasons why a self fertilizing method of reproduction is so effective are the efficacy of reproduction, as well as decreasing genetic variation and thus the fixation of highly adapted genotypes. Most of the loci get fixed at a high rate; this can be ascribed to the fact that with each generation of self fertilization the rate of heterozygotes decreases by 50%. Homozygosity will thus be obtained in 5-8 generations. The 3rd reason for the efficacy of self fertilization is that in mixed stands of self and cross pollinating crops, the self fertilizing plants can donate pollen to both plant types, where the cross fertilizing plants are restricted concerning the contribution it can make to the population with regard to pollen donation.

Almost no inbreeding depression occurs in self fertilizing plants because the mode of reproduction allows natural selection to take place in wild populations of such plants. Thus, the genetically non-superior or unstable plants are removed from the population at an early evolutionary stage. Populations derived from self pollination are sometimes not as evolutionary adaptable as with other reproductive methods, but are known to utilize specific ecological niches more effectively.

Critical steps in the improvement of self fertilizing crops are the choice of parents and the identification of the best plants in segregating generations. The breeder should also have definite goals with the choice of parents. Self fertilizing cultivars are easier to maintain, but this could lead to misuse of seed.

Some of the agronomically important, self fertilizing crops include wheat, rice, barley, dry beans, soy beans, peanuts, tomatoes, etc. New new genetic variation is released in a given germplasm pool by crossing within this pool or with foreign genotypes, followed by segregation; moreover, mutagenesis and introduction for alien DNA (transgenes) are further options to create new variation.

Several selection methods are applied in self fertilizing crops, such as mass selection, single plant-, pedigree-, bulk population-, back cross-, recurrent-, as well as single seed descend (SSD) selection and the use of DH(doubled haploid) lines. In most breeding programs a combination of these methods are applied. The different selection methods can be summarized as follows:

Mass selection

This method of selection depends mainly on selection of plants according to their phenotype and performance. The seed from selected plants are bulked for the next generation. This method is used to improve the overall population by positive or negative mass selection. Mass selection is only applied to a limited degree in self fertilizing plants and is an effective method for the improvement of land races. This method of selection will only be effective for highly heritable traits. one shortage of mass selection are the large influence that the environment has on the development, phenotype and performance of single plants. It is often unclear whether the phenotypically superior plants are also genotypically superior and strong environmental differences may lead to low selection efficacy (heritability).

ingle plant selection (pure line selection)

A variety developed by this method will be more uniform than those developed by mass selection because all of the plants in such a variety will have the same genotype. The seed from selected plants are not added together but are kept apart and used to perform offspring tests. This is done to study the breeding behaviour of the selected plants. The high uniformity in stand and performance has been stressed in the past, but the risk of highly specialized pathogens evolving is very high. More genetic variability could buffer the crop against such pathogens as well as stability of production under varied environmental conditions.

Selection Methods for the Development of Pure Breeding Cultivars from Crosses

Crosses between varieties, germplasm introduction and breeding lines are made to create new gene combinations. In the generations to follow, superior genotypes (presumably having superior genes and gene combinations) are selected and fixed in the homozygous state by means of self fertilization and selection. These selections are tested extensively with the goal of releasing the few best of them for cultivation.

Pedigree selection

Parental lines are crossed and selection of plants with new gene combinations already takes place in the F2 generation (the generation of plants formed from (natural) selfing of the F1 hybrids). The offspring of selected populations in the generations to follow are repeatedly subjected to selection and always single plants are taken as source of new offpspring lines, until genetic uniformity is reached. Only then, seed yield of several plants per line is combined and mixed to have more seed for testing. Records are kept of the origin of the selected individuals or lines. The amount of generations of single plant and line selections, as well as selection intensities, can be varied in practice according to the crop and availability of facilities.

It is usually traits with high heritability that are quick and easy to measure that are concentrated on as long as single-plant-selection is carried out (in the early phase. In the later phase, when plots are the units of assessment, traits like yield with lower heritability are assessed and taken as basis for selection. One of the main objections against this method is that the genetic variation, available for selection of quantitative traits, are drastically decreased in later generations due to the single-plant-selection carried out in early generations (like from F2 to F5). Seed purification and multiplication is usually incorporated in one of the final generations of pedigree selection. This method is very labour intensive.

Breeding efficiency is one of the goals with early generation testing. This is done by early identification of superior heterogeneous populations. The early elimination of inferior populations and subsequent concentration of selection efforts within superior populations is assumed to result in increased efficiency. Accurate evaluation of heterogeneous populations is essential to the success of this method and assumes that transgressive segregants from inferior populations will not exceed selections from superior populations in performance.

Bulk population selection

With this method of selection the offspring from a crossing are planted at planting densities equal to commercial planting densities. During this period, which may include a number of generations, the level of homozygosity in the bulk population increases.

This method is simple and cheap and involves less work than pedigree selection in the earlier generations. It is necessary to plant large populations to ensure that the best segregates are selected when selection starts. Segregating generations are subjected to another single plant selection step. Fewer records are kept during earlier generations than with pedigree selection. This type of selection is especially carried out with crops which are usually planted at high planting densities, e.g. small grain crops.

ingle seed selection (SSD)

This method was introduced as a means to transform the entire F2-generation as fast as possible into a generation of homozygous plants, albeit heterogeneous population (thus prevent the loss of variation and have the plants homozygous, i.e., tru-to-type breeding). This method is hence used to decrease the time that is passing with genotypes not yet being homozygous. The decrease is enabled by the fact that with only one seed desired per plant, plants can be pushed to have two or three, sometimes even more seasons per calendar year. This can cause the process from the starting a breeding cycle intil release of cultivar to decrease by 1-3 years.

This method does not eliminate weak plants early such as in other methods, there is no provision for selection of superior plants in the F2 generation. Modification of this method is possible and record keeping is not necessary in early generations.

Doubled haploid method

Haploid plants may be produced by chromosome elimination in wide crosses, ovule culture or by anther and microspore culture. Anther and microspore culture, however, are mostly used because of its ability to produce haploid plants with much larger quantity compare with the other two methods. Stresses are usually necessary to alter the development pathways of microspores from producing pollen to forming haploid plants. Presently, the so-called inducer method is very much used in maize breeding (hybrid maize)

The chromosome numbers of the haploid plants are doubled with the use of colchicine. Spontaneously doubled haploid plants, however, can also be produced directly from the three methods. Embryo rescue methods can be used to ensure that seed from these wide crosses or stigma culture plants survive and doesn’t get aborted. This method has the potential of shortening genetic improvement cycles in comparison to pedigree or bulk methods. Like the single seed selection method, early generations are not subject to selection, but most of the lines are eliminated during the field evaluation trials.

The purpose is the same as for SSD: transform the genetic material from F1 as fast as possible into a bulk of homozytous plants. This method is very labour intensive and the most expensive of the procedures that increases the amount of generations per year. For this method to be successful, the plants must be genetically stable.

Backcrossing: a selection method for the upgrading of genotypes

This is a type of repeated crossing and selection where an lacking, specific (or very few) gene(S) (honestly speaking an allele) can be incorporated into otherwise already elite, superior cultivars. The so-called recurrent parent(al variety) is highly productive and commercially successful but lacks this specific gene (e.g. disease resistance). This trait is e.g. present in a wild or exotic (donor) parent. After the initial cross there is the backcross with the recurrent parent. After each backcross, thos hybrid plants that 'still' harbour the desired gene are identified and selected and are backcrossed again with the recurrent parent.

This technique is easy when traits are to be added which are easily inherited, dominant, and easily identified in hybrid plants. If unwanted genes are closely linked to the desired gene, the unwanted genes are probably transferred together with the desired gene and the offspring may suffer from this. One advantage of the back crossing method is that extensive testing (for other than the desired trait) is not necessary. This method is used to e.g. transfer male sterility into pure lines, which is a prerequisite for many hybrid breeding programs.

Marker assisted backcrossing is routinely applied in breeding programs for gene introgression.

Traditional selection methods

A relatively small gene pool is created with traditional selection methods, there is not a lot of opportunity for gene recombination and few opportunities for the breakage of linkage blocks. Breeders are looking to overcome these shortcomings by increasing the amount of crosses made, as well as to start testing for performance in earlier generations and to introduce systems for improvement of populations (e.g. recurrent selection).

Other breeding strategies include breeding of multiline cultivars (varieties; a composite of genetically nearly identical lines which have different genes for one (few) trait(s), e.g. stem rust resistance). Lines that are basically genetically identical, except for a single gene, are called isogenic lines. Mixed varieties (composites) are also used and are less uniform than pure varieties. They should be revised in regular intervals to prevent shifts in the proportion of components. Recurrent selection is most often used in cross fertilizing crops, but can as well successfully be used in self fertilizers - use of a male sterility gene may be necessary in this case.

Hybrid breeding is a very important category for cross fertilizing crops and has the following advantages compared to the natural alternative which is population breeding:

* More Heterosis (due to the inter-pool cross situation)
* Maximal selection intensity possible (one - the best - genotype makes the cultivar)
* Easier to create varieties with multiple resistance genes
* Attractive method for private breeders because of built-in mechanisms for protection of varieties.

Raising hybrid seed has been one of the major goals of horticultural and agricultural practice, because hybrid plants are more productive (due to hybrid vigour) and more uniform than plants derived from random pollination. To raise hybrid seed, self-pollination and sib-pollination (pollination by a plant of the same line) of seed mother must be circumvented. One method is hand emasculation of the line used as female parent, which is then naturally cross-pollinated by pollen from the line serving as male parent and planted in an adjacent row. However, this process is very labour intensive and invariably expensive. If the crop plants can be made self-incompatible by the introduction of the genes controlling self-incompatibility, then all seeds produced will be hybrids resulting from cross-pollination between two different lines. This would facilitate the production and increase the yield of hybrid seed and, at the same time, reduce the labour costs. Yet, the necessary step of forming homozygous parents by self-fertlilization is not easy with self-incompatible material. Therefore, the most prominent means to circumvent the selfing of the seed mother is to use a genetically pollen-free seed mother (cms, cytoplasmic-genic system of pollen sterility)

Selection of cross-pollinated crops

The natural state of self-fertilizing crops is homozygosity. If under selection, this results in homogeneity, if not or not strictly so, it results in heterogeneous stand of plants which are geneticall different but nevertheless all homozygous (like a landrac). Cross-fertilizing popuations of crops are characterized by a high degree of heterozygosity and heterogeneity. Plant species where normal mode of seed set is through a high degree of cross-pollination have characteristic reproductive features and population structure.Existence of self-sterility, self-incompatibility, imperfect flowers, and mechanical obstructions make the plant dependent upon foreign pollen for normal seed set. Each plant receives a blend of pollen from a large number of individuals each having different genotypes. Such populations are characterized by a high degree of heterozygosity with tremendous free and potential genetic variation, which is maintained in a steady state by free gene flow among individuals within the populations. It is inappropriate, and could be rather hazardous, to take one or a few individuals to investigate or improve these populations. The enhanced fitness of heterozygotes over homozygotes of cross-pollinated crops has been manipulated in the form of two different breeding approaches namely, population improvement and hybrid breeding in such crops.In the development of hybrid varieties, the aim is to identify the most productive heterozygote from the population, which then is produced with the exclusion of other members of the population. In contrast, the population improvement envisages a stepwise elimination of deleterious and less productive alleles through repeated cycles of selective mating of genotypes that are more productive. Population improvement is slow, steady and a long-term program, whereas the production of hybrids is aimed to maximize the genetic gains in much less time. Both of these breeding approaches are complementary rather than mutually exclusive and are based on sound genetic theory. The different selection methods can be summarized as follows:

Mass selection

It is the simplest, easiest and oldest method of selection where individual plants are selected based on their phenotypic performance, and bulk seed is used to produce the next generation. Mass selection proved to be quite effective in maize improvement at the initial stages but its efficacy especially for improvement of yield, soon came under severe criticism that culminated in the refinement of the method of mass selection. The selection after pollination does not provide any control over the pollen parent as result of which effective selection is limited only to female parents.The heritability estimates are reduced by half, since only parents are used to harvest seed whereas the pollen source is not known after the cross pollination has taken place.

Recurrent selection

This type of selection is a refined version of the mass selection procedure and differs as follows:

* Visually selected individuals out of the base population undergo progeny testing
* Individuals selected on basis of the progeny test data are crossed with each other in every possible way to produce seed to form the new base population.

Half-sib selection with progeny testing

Selections are made based on progeny test performance instead of phenotypic appearance of the parental plants. Seed from selected half-sibs, which have been pollinated by random pollen from the population, is grown in unreplicated progeny rows for the purpose of selection. A part of the seed is planted to determine the yielding ability, or breeding value, for any character of each plant. The seed from the most productive rows or remnant seed from the outstanding half-sibs is bulked to complete one cycle of selection.

Full-sib selection with progeny testing

A number of full-sib families, each produced by making crosses between the two plants from the base population are evaluated in replicated trials. A part of each full-sib family is saved for recombination. Based on evaluation the remnant seed of selected full-sib families is used to recombine the best families.

elections with test cross performance

The purpose of this type of selection is a slight deviation from the concept of intra-population improvement in the sense that the population is improved, not only for performance, but also with respect to combining ability with a specific reference population. It involves genetic modifications of the population with an aim on its better use for the exploitation of heterosis. It involves three steps:

# Self-pollination and test crossing of individuals
# Evaluation of test crosses in replicated trials
# Recombination from selfed remnant seed of selected plants

elfed family selection

The plants in the original base population are selfed to produce S1 progenies, which are evaluated in the next season in replicated multi-environmental trials to identify promising S1 families. The remnant S1 seed of such selected families is then recombined in the third season as a result of which one cycle is completed in three seasons. Hence, the units of selection and recombination are S1 progenies.

Breeding of Asexually Propagated Crops

Asexual reproduction covers all those modes of multiplication of plants where normal gamete formation and fertilization does not take place making these distinctly different from normal seed production crops. In the absence of sexual reproduction, the genetic composition of plant material being multiplied remains essentially the same as its source plant.

Clones of mother plants can be made with the exact genetic composition of the mother plant. Superior plants are selected and propagated vegetatively; the vegetative propagated offspring are used to develop stable varieties without any deterioration due to segregation of gene combinations. This unique characteristic of asexual reproduction helped to develop a number of cultivars of fruits and vegetables including grapes, apples, pears and peaches.

Improving asexual plant material through selection

The selection in these crops is restricted to the material introduced from other sources, such as field plantations. The promising selections are tested in large scale trials which, if successful, can be multiplied and released for commercial cultivation.The improvement of asexually propagated plants through induced mutations has distinct advantages and limitations. Any vegetative propagule can be treated with mutagens and even a single desirable mutant or a part of a mutated propagule (chimera) can be multiplied as an improved type of the original variety.

election of asexual plants

Selection, in the case of asexual plants, can be defined as the selection of the best performing plant and the vegetative propagation thereof. Because plants are not totally genetically stable, it can be expected that deviations would occur through the years. Selection is thus an ongoing process where deviants are selected or removed from the selection program. The main purpose of selection is to better the quality and yield of forthcoming plantations. Such as any breeder, the breeder has to have a good knowledge of the characteristics of the cultivar under consideration. Different approaches can be followed in the selection process of asexual plants, such as mass selection and clone selection from clone blocks.

In mass selection there are some factors that must be considered when selecting plants in a mother block, e.g. vineyard. Time of selection is a big factor, because you have to select when most of the characteristics of the plant are clearly showing. With asexual perennials the best time is just before harvest. For the best results the selected plant must be evaluated during the next season, when growth-abnormalities, leave disfigurations and virus symptoms are best visualized. Mass selection is done annually on the same plant for a minimum of three years. A plant that does not conform to the requirements in any given year of the selection cycle is discarded from the program.

Older plantations which were exposed to harsh growth conditions are seen as a preferred selection sources. The plants that grew under these circumstances and performed well are seen to have good genetic properties. In these older plantations natural selection took care of most poor performing plants.

New clone development

The development and registration of new clones take place by means of local clone selection in old plantations, as well as the importation of high quality clones from abroad, for local evaluation.

A clone is the vegetative offspring of one specific mother plant; it does not show any genetic, morphologic or physiologic deviations from the mother plant. Evaluation takes place with the different selected clones after selection. The dissimilar clones are compared to each other to determine their quality and resistance capabilities. Breeding is not involved in clone selection; the clone cannot be bred for resistance of certain types of viruses, emphasis must be put on making sure that the clone material leaving the nursery are virus free. Techniques are developed to test the clones for any harmful viruses. Harmful viruses sometimes do not show in the preliminary evaluations. Phytosanitary development (virus detection and virus eradication) is thus performed in laboratories and greenhouses, parallel with field- and quality evaluation in field clone trials.

When a clone complies with the minimum quality and phytosanitary standards prescribed by the Plant Improvement Association (PIA), it is officially registered for certification and commercialization. The PIA is an association that complies for all plant improvement, including grapes, apple, pear and peaches.

Clones are made of cuttings from a field-grown mother plant. Due to bad management and infection from neighbouring plantations there is only a few virus free mother plants in selecting plantations. The clone developers had to incorporate techniques such as tissue cultures and in vitro propagation to develop virus free clones from the limited mother material. The apical meristem is free of any harmful viruses. By using the apical meristem for tissue culture a virus free clone can be developed.

Multiplication

During the first phase of multiplication, a nucleus cutting from each candidate and/or registered clone is kept in a PIA approved insect free nucleus block green house. From here all future multiplication and evaluation will be done.During the second phase of multiplication, rootstocks and grafted cuttings are established in foundation blocks in insect free facilities and open field isolated areas, from where further evaluation are done.

The scion and rootstock material from the second phase source are grafted and callused. The grafted plants are then planted in isolated areas for the establishment of mother blocks. These blocks are used for multiplication purposes. The third phase is the establishment of scion mother blocks from the above mentioned source on farms of contracted collaborator producers in pre-selected virgin soil of which about 4 km² are maintained.

References

*CHAHAL, G.S & GOSAL, S.S., 2002. Principles and procedures of Plant Breeding, Alpha Science International, United Kingdom.
*Falconer, D.S., 1989. Introduction to Quantitative Genetics. 3rd Ed. Longman. Burnt Mill.
*FRISCH, M. & MELCHINGER, A.E., 2005. Selection Theory for Marker-assisted Backcrossing. Genetics: Published Articles Ahead of Print, published on March 31, 2005 as 10.1534/genetics.104.035451 [http://www.genetics.org/cgi/rapidpdf/genetics.104.035451v1]
*GOUSSARD, P.G. (2004) Improving grape vine material through selection. Department of Viticulture. Stellenbosch University. Lecture notes.
*HOLSINGER, K.E., 2000. Reproductive systems and evolution in vascular plants. PNAS. June 20, vol. 97, no. 13: 7037-7042.
*KENNEDY, B.K., 2004. Genetic barrier to self-pollination identified. EurekAlert! 19 May. [http://www.eurekalert.org/pub_releases/2004-05/ps-gbt051704.php]
*KOHLI, M.M. & FRANCIS, M., 2000. Application of biotechnologies to wheat breeding. Proceedings of a conference at La Estanzuela, Uruguay, November 19-20, 1998. Montevideo, Uruguay: CIMMYT. [http://ecoport.org/ep?SearchType=earticleView&earticleId=162&page=2055]
*KWV, South Africa. (2005). Setting new global standards for vine plant improvement. Vititec ©
*MARAIS, G.F. 2005. Seleksiemetodes vir kruisbestuiwende gewasse. Department of Genetics. University of Stellenbosch. Lecture notes.
*TAKEBAYASHI, N. & MORELL, P.L., 2001. Is self-fertilization an evolutionary dead end? Revisiting an old hypothesis with genetic theories and a macroevolutionary approach. American Journal of Botany. 88:1143-1150.
*VIVIER, M.A. (2004). Improving grape cultivars. Department of Viticulture. University of Stellenbosch. Lecture notes.