Breeding Strategies to Improve Productivity and End Use Quality Under Moisture Deficit and Higher Temperature

A basic breeding scheme for a self-pollinated crop such as wheat is outlined in Fig. 9.1. This scheme represents either a modified bulk or selected bulk selection strategy. In a modified bulk strategy individual plants identified in the early generations are grown as individual plots in the following generation. These progeny are usually derived from three-way or top crosses (involving three parents) or simple crosses between two parents, and individual plants are usually selected from the

P1 x P2

Characterization of parents using molecular markers & phenotypic data

Top cross F1 - individual plants Only th°s1esmDa,rkers segregating in selected using phenotype & MAS I P1 & P2 are applied

F2 plots derived from individual top F2 llbuk f!ar-nos is I I I I I I I I I II I I I I I cross plants (or selected bulks planted of marker positive plants)

All markers applied in F4 (for high priority permitting

All markers applied in F7

selected to form F3

F4 - a bulk of leaves or m m m m m m m m individual p|ants are HHHHHHHH high priority "crosses) budget tested using MAS

F7- is derived from single spike selections from F6

Bioassays of selected plant to confirm resistance

Multi-locational testing over several years with increasing sites and decreasing numbers of lines with time

Fig. 9.1 An example of a conventional breeding scheme using either a modified bulk or selected bulk strategy. The time from cross to homozygous line identification is 4-7 years and a further 4-5 years of yield and quality evaluation and seed multiplication are required before the selected genotype is released to farmers first segregating generation (either top cross F1 or F2). A bulk of selected spikes or plants is then used to advance the population to the next generation. This continues until de-bulking in the F5 or F6 generations to produce homozygous inbred lines.

A selected bulk differs in that selected F2 plants are bulked to form a single F3 bulk per cross. The populations are advanced in the same way until de-bulking in the later generations. The scheme in Fig. 9.1 assumes the use of molecular marker assisted selection (MAS) and drought and/or heat screening during the segregating phase (F2-F6). There are many variations on these schemes and the process described is one of among many possible strategies. The crossing, selection and evaluation strategies outlined in the following sections will be discussed in the context of the strategy in Fig. 9.1. Crossing

Once genetic variability for adaptation to the prevailing stresses in the TPE has been assembled, the challenge for the plant breeder is to combine this variability in a crossing program that also encompasses the key biotic and market constraints. Genotyping technology has improved significantly in recent years and the breeder should have DNA fingerprints of all the key progenitors in the breeding program. Historically, breeders used coefficients of parentage that assumed no selection when determining relatedness among materials. The breeder will use this information to better design crosses. The degree of relatedness among parents selected for crossing will reflect the breeding objective, the complexity of the target trait and the available resources.

DNA profiles generated using microsatellites or DArT (Diversity array technology) (Mace et al. 2008) provide good genome coverage and offer more realistic estimates of diversity. Assuming that much of the variation for drought tolerance is additive (Trethowan and Mujeeb-Kazi 2008), the breeder can identify the least related lines from among the best performing materials under stress to combine in crossing. Association genetics studies may also be useful in identifying genomic regions linked to improved yield performance. Crossa et al. (2007) genotyped MET entries at the International Maize and Wheat Improvement Center (CIMMYT) spanning a 25-year period and related these profiles to wheat cultivar performance. Their analysis identified genomic regions unrelated to genes controlling phenology, morphology and disease resistance that were associated with superior yield in many environments globally. Combining these regions in crosses may provide the additive variance for stress response needed to improve the broad adaptability of crop germplasm. Characterization of parental materials is not just confined to MET data and molecular analysis. Determination of the physiological responses of progenitors within the breeding program to abiotic stress will allow the breeder to combine physiological mechanisms in crosses (Reynolds and Trethowan 2007). Progeny developed in this way at CIMMYT have shown superior performance in global MET experiments (Yann Manes, 2008 personal communication).

When introducing variability from primary synthetic wheat, a landrace or a translocation stock cultivar, it is in most instances sensible to make at least one backcross or top cross to an elite parent before proceeding to F2. This is because most breeding programs cannot manage the extremely large populations required to exploit a simple cross between an adapted cultivar and a primary synthetic. In contrast, the backcross F2 will produce a higher frequency of agronomically acceptable progeny. This principle was used at CIMMYT to produce synthetic derivatives with drought and heat tolerance, broad adaptation and high yield (Trethowan and Mujeeb-Kazi 2008).

In some instances the breeder may attempt to combine variability in double crosses or four-way crosses (crosses between F1 progeny), although these usually result in F2 progeny that are generally too variable to manage. However, should a reasonable degree of relatedness exist among two or more of the progenitors then such crosses may make sense. In reality the plant breeder gradually improves the frequency of favorable alleles in the breeding program over the span of a career and often several cycles of crossing and selection are required to pyramid genes for traits of economic importance. Selection

Once the desired crosses have been made the selection of the segregating materials becomes vital. If molecular markers for known genes are available, they can be tracked in the segregating phase. Allele enrichment in the top-cross F1, backcross F1 and F2 using markers for known genes will greatly increase the frequency of lines carrying the target genes in the subsequent fixed line progeny and can be useful for accumulating genes governing root health (William et al. 2007; Bonnett et al. 2005). If quantitative trait loci (QTL) of significant effect relevant to the TPE are available they can be introduced into elite germplasm using a MAS scheme similar to Fig. 9.1. However, significant QTL x environment interaction and genotype specificity tend to limit this approach. The breeder is often faced with multiple QTLs of relatively minor effect that are genotype and environment dependent. In this instance one possibility is to use a recurrent selection scheme, combined with molecular markers and empirical selection under stress, to provide a mechanism whereby these minor QTLs can be combined. In such a scheme parents would be genotyped using markers and QTLs combined in crosses and tracked using markers. The progeny would be genotyped and screened in multi-environment trials under stress at F4 and the progeny selected on the basis of yield and genotype. These progeny would then be randomly inter-mated to continue the process of allele accumulation. This approach favorably skews gene frequency towards better adaptation under stress.

However, in the absence of QTLs for abiotic stress tolerance, favorably skewing gene frequency to greater levels of water or temperature stress tolerance will require one of two approaches. The segregating materials can either be selected in the TPE under all the prevailing stresses within any given year and site, or in managed selection environments that mimic the TPE. Effective selection in the TPE is dependent upon the occurrence of the TPE in the year of selection. The heritability of selection is extremely low for most water limited environments and year effects are almost always the largest component of variance (Ribaut et al. 1996; Ahmad and Bajelan 2008).

On the other hand, managed environments can increase the heritability of selection but their effectiveness is dependent upon correlation with the desired TPE. An analysis of global MET data of wheat lines developed for semi arid environments at the International Maize and Wheat Improvement Centre (CIMMYT) showed that the germplasm did not adapt to certain dry environments (Trethowan et al. 2001b). All the materials in this study were developed using simulated post-anthesis drought stress in the field in northwestern Mexico. However, the patterns of adaptation improved when a broader range of managed stresses were employed that better correlated with the target environment (Trethowan et al. 2005). The relevance of the managed selection environments at CIMMYT was confirmed in a retrospective study of genotypes previously tested in global METs in managed stress treatments. It was clear that specific stress patterns correlated with specific environments (Trethowan et al. 2005). However, given the vagaries of the target environment there will never be sufficient data to draw water-tight conclusions. This calibration of managed stress environments must be continual and an integral part of the breeding strategy.

While physiological characters, such as total soluble stem carbohydrates and transpiration efficiency are useful in differentiating parental materials and improving the efficiency of crossing for stress tolerance, their determination is generally not manageable in segregating generations where large numbers of lines have to be assessed. However, easy to measure physiological traits such as canopy temperature depression (CTD) do correlate with plant performance under both heat and drought stress (Reynolds et al. 2007a). If a modified bulk or selected bulk scheme is used, it is possible to measure CTD quickly and efficiently on large numbers of F3 or F4 plots (Reynolds and Trethowan 2007) (Fig. 9.2). The breeder's eye is the best physiological tool available, however CTD measured on breeder selected plots can show a significant range in temperature responses. When these materials were carried forward and fixed lines derived from them it is interesting to note that none of the



20 21 22 23 24 25 26 Vegetative canopy temperature (°C)

Fig. 9.2 The canopy temperature of F4 bulks under drought stress before and after flowering



23.0 23.5 24.0 24.5 25.0 25.5 26.0 26.5 Average Two CT readings F4 pops Y03-04

Fig. 9.3 The relationship between canopy temperature (CT) of F4 bulks in 2004 and the yield of fixed lines derived from them in 2006 (Yann Manes, unpublished data)

23.0 23.5 24.0 24.5 25.0 25.5 26.0 26.5 Average Two CT readings F4 pops Y03-04


Fig. 9.3 The relationship between canopy temperature (CT) of F4 bulks in 2004 and the yield of fixed lines derived from them in 2006 (Yann Manes, unpublished data)

high-yielding lines had warmer canopies in the F4 generation (Fig. 9.3). This illustrates that CTD can be a reliable tool for identifying stress tolerant lines.

A similar approach can be used to select for end-use quality. A combination of molecular markers and high throughput small scale tests that correlate with end product quality can be used in the early generations to favorably skew gene frequency. This is possible where pedigree or modified bulk breeding strategies are used and seed from small plots are available. For example, these tests may be conducted on F2 derived F4 plots tested at more than one location.

As mentioned earlier, the deployment of more water-use-efficient and heat tolerant materials in conservation farming systems will improve overall productivity while conserving moisture and resources and reducing costs. Capturing the genetic response to these systems will require selection under zero-tillage with crop residue cover equivalent to that in the TPE. Alternatively, as crop emergence and establishment are important components of adaptation to conservation agriculture, planting segregating bulks deeper than normal combined with selection for short-statured plants from among those that emerge is an effective way of favorably skewing gene frequency (Trethowan et al. 2005). Trethowan et al. (2009) reported the results of a selection study in which segregating materials from the same cross were selected either always under zero-tillage or always under complete tillage. In general, the materials selected under zero-tillage performed better in both tillage systems.

The selection of some economically important traits such as elevated micronutrient concentration is hampered by the high cost of analysis. In these instances, the identification of linked molecular markers would greatly reduce the cost of selection. Molecular markers would ideally be used in the early segregating generations to skew gene frequency with subsequent ICP-MS (inductively coupled mass spec-trometry) analysis of the relatively smaller number of fixed line progeny remaining at the end of the selection process. Markers have the advantage of being phenology independent, as differences in maturity within a population can confuse traditional selection approaches for traits such as Fe and Zn concentration. If the breeder has access to differential stresses (usually generated using limited irrigation or different planting dates) it would be informative to test the stability of expression of micro-nutrient concentration across two environmental extremes, typical of the current and predicted TPE.

In crops such as wheat and barley, double haploids provide an option for the rapid production of genetically stable homozygous lines. It is generally advisable to make the double haploids on F2 or F3 progeny once screening for simply inherited but economically important traits has been completed, thus greatly increasing the frequency of useable materials among the resultant double haploids. However, double haploids are unlikely to be particularly useful for the improvement of complex traits such as tolerance to water and temperature stress. Without selection under these stresses to improve allele frequency for plant response the probability of finding a double haploid with all the desired alleles is very small. Evaluation

Once fixed lines have been identified, usually derived from single plants in the F5 generation or greater, the efficacy of the stress response must be confirmed. In the CIMMYT wheat program these progeny are tested first in a series of managed stresses. These are a combination of managed pre-anthesis, post-anthesis and/or continuous stresses generated using limited irrigation in an arid environment (Trethowan et al. 2005) and late planting is used to generate a consistent heat stress from anthesis through the grain-filling period. Selected lines are then tested globally in METs covering the target wheat growing areas of the developing world. The lines selected using these crossing, selection and evaluation principles have performed well globally. Lage and Trethowan (2008) analyzed the performance of synthetic derivatives deployed in the Semi-Arid Wheat Yield Trial distributed by CIMMYT and found that some synthetic derivatives showed superior yield response across a wide range of environments when compared to the best locally adapted cultivars. Synthetic derivatives selected in managed stress environments in Mexico also performed well when tested across variable Australian environments (Ogbonnaya et al. 2007). These authors reported that derivatives yielded 8-30% more than the best locally adapted cultivars, clearly demonstrating a significant genetic correlation between Mexican managed stress environments and sites in Australia.

In a reverse study, Gororo et al. (2002) developed synthetic hexaploid derivatives from locally produced primary synthetics in Australia and tested these materials in both Australia and Mexico. The synthetic derivatives were higher yielding than their elite recurrent parent in 38 of 42 comparisons across environments in both countries, indicating a significant degree of transferability of drought stress response.

There is less information on the response of materials selected under terminal heat stress and tested in high temperature TPEs. In one of the few available studies, Lillemo et al. (2005) analyzed the yield performance of lines distributed globally in CIMMYT's High Temperature Wheat Yield Trial. These lines were developed by late sowing segregating materials in northwestern Mexico with subsequent selection for plant phenotype and grain weight and eventually yield once fixed lines were identified. The selection environment in Mexico clearly correlated with many heat stressed environments globally and materials with stable and superior performance in the heat stress TPE were identified.

In Australia, fixed lines are tested widely in the TPE over a number of years as large genotype x year interactions obscure genetic potential (Chapman et al. 2000). In contrast, in many developing countries fixed lines are tested in METs on research stations, largely for logistical and economic reasons (Ceccarelli and Grando 2007). The materials are only grown on farm once they have been released to farmers. Clearly, the efficiency of cultivar selection from among homozygous lines derived through selection will be dependent on how well the chosen yield testing environments correlate with the TPE.

Nutritional, processing and product quality are generally assessed using grain samples collected from selected genotypes from METs sown across the TPE. The extent of the analysis will reflect the importance of quality within the local, regional and global market place. The analysis of quality is expensive and generally limited to less costly indirect tests in initial assessments, with more detailed analysis of dough properties and end-product quality in subsequent trialing of selected entries.

Was this article helpful?

0 0
Renewable Energy 101

Renewable Energy 101

Renewable energy is energy that is generated from sunlight, rain, tides, geothermal heat and wind. These sources are naturally and constantly replenished, which is why they are deemed as renewable. The usage of renewable energy sources is very important when considering the sustainability of the existing energy usage of the world. While there is currently an abundance of non-renewable energy sources, such as nuclear fuels, these energy sources are depleting. In addition to being a non-renewable supply, the non-renewable energy sources release emissions into the air, which has an adverse effect on the environment.

Get My Free Ebook

Post a comment