Annex I Status of Breeding for Tolerance of Abiotic Stresses and Prospects for Use of Molecular Techniques

CONSULTATIVE GROUP ON INTERNATIONAL AGRICULTURAL RESEARCH
TECHNICAL ADVISORY COMMITTEE

TAC SECRETARIAT

FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS

March 2001

John Bennett,
Plant Breeding, Genetics and Biochemistry Division,
International Rice Research Institute,
DAPO 7777, Metro Manila, Philippines

Importance of abiotic stresses for the CGIAR mandate

Abiotic stresses arise from extremes of climate such as drought, flood and cold, from soil toxicities of such elements as Na, Al and Fe, and from soil deficiencies of elements like P and Zn. In places where climatic extremes are of regular occurrence and predictable, agricultural activity is usually very limited and dependent populations are small, but in many other places where these stresses occur in an unpredictable manner the agricultural activity may be intense and the dependent populations large. It is in these latter regions that abiotic stresses are major contributors to food insecurity and poverty for hundreds of millions of the rural poor. Drought alone affects more than 70 million hectares of rice-growing land world-wide. Soil toxicities and deficiencies on the other hand render more than one hundred million hectares of agricultural land marginal for agriculture, again limiting production and creating poverty for millions. Farmers in these environments adopt a risk-aversion strategy of low inputs, resulting in low outputs, poor human nutrition and reduced educational and employment opportunities, especially for girls. The rural poor are particularly badly affected because of lack of access to alternative sources of employment or food.

Plant breeding for abiotic stress tolerance - a viable option

Most crop species show considerable genetic variation in tolerance to the major climatic and chemical stresses. Plant breeding is therefore a viable option for improving productivity, reducing farmers’ risks and cultivating marginal land. The CGIAR Centers have a comparative advantage in these breeding activities because of their germplasm collections, their new capacity for genetic dissection of complex traits, and their ability to conduct multidisciplinary plant improvement programs in target environments. The Centers have released several high-yielding cultivars with enhanced tolerance of abiotic stresses, including Al-tolerant rice from CIAT, cold-, salt- and submergence-tolerant rice from IRRI, and drought-tolerant maize from CIMMYT. However, these new cultivars represent incremental gains that could certainly be exceeded if more investment were made in the needed multidisciplinary research (Table1).

Partnerships for multidisciplinary research

It is now a particularly auspicious time for the CGIAR to dedicate itself to a concerted effort to improve tolerance of abiotic stresses in the mandated crops. Advanced research organizations have developed the necessary analytical tools for understanding the mechanisms of stress tolerance. Most of this work is conducted in the public sector, often in basic studies on Arabidopsis thaliana. The leading laboratories are eager to be partners in the enterprise. The private sector, with its traditional focus on protecting plants from biotic stresses, at present have little interest in abiotic stresses and may share their new genomic resources with the CGIAR Centers for the benefit of the poor. The linkages between the Centers and their NARES partners are particularly important in defining the research agenda, conducting the breeding programs in realistic environments and ensuring impact.

Table 1 summarizes the steps in enhancing abiotic stress tolerance in CGIAR mandated crops. Also listed are the principal disciplines and partners required at each step. An important early step is identification of target environments and their associated stresses. This information will be used in designing the selection screens for conserved germplasm and breeding materials and planning the evaluation trials. The early identification of recipient cultivars can greatly accelerate the breeding program and define the baseline performance against which genetic improvement will be judged. The identification of beneficiaries helps in determining the relative contributions of genetic enhancement and crop management to overcoming abiotic stress. Poor farmers will rely more on genetics than management, but expectations should be realistic: it is likely that innovative research on crop and natural resource management in relation to abiotic stresses will reveal cost-effective ways in which poor farmers can increase their productivity and income. Just as the Green Revolution varieties of wheat and rice encouraged massive government infrastructure schemes for irrigated environments, we can expect that new germplasm for the fragile rainfed environments will encourage local innovation in management by and for poor farmers.

Table 1: Partners in developing cultivars with tolerance of multiple abiotic stresses

*Includes linkage to public research on abiotic stresses by the Arabidopsis community

Understanding the mechanisms of abiotic stress tolerance

The screening of germplasm collections for different mechanisms of tolerance to a particular stress must be based on sound physiological principles and take into account the target environment and the timing of stress relative to the growth cycle. Yield under stress is often used as a preliminary criterion that can be applied to thousands of accessions, with more discriminating tests being applied subsequently to identify accessions with different mechanisms of tolerance. If the CGIAR Centers do not invest adequately in the characterization of the germplasm that they hold in trust, it is unlikely that anyone else will do so and a valuable resource will remain unexploited.

Germplasm accessions with high tolerance of a particular abiotic stress are usually not directly useful for agriculture. The genes conferring stress tolerance must be introgressed into improved backgrounds, a task often rendered difficult by the genetic complexity of the trait and our poor understanding of it at the molecular level. Another limitation is the difficulty of applying a uniform level of stress over a field. Great skill and considerable expense are involved in exposing a population of a thousand breeding lines (or a thousand genebank accessions) to uniform stress from drought, salinity, iron toxicity or zinc deficiency. And lack of control of soil type and texture and general climatic conditions can lead to genotype x environment interactions that confound even the most carefully planned experiments. Finally, the target environment is unlikely to feature a single abiotic stress: submergence, drought, iron toxicity and Zn deficiency may be encountered in a single season at a single location. New cultivars with multiple tolerances of abiotic stress are essential.

CGIAR Centers have responded to these challenges by developing interdisciplinary teams focused on specific, high-priority stresses, such as drought, salinity and aluminum toxicity. The powerful new tools of biotechnology have been allied with skills in physiology to design informative experiments. Some blind alleys have been entered, but progress overall has been encouraging. Breakthroughs in one crop are frequently relevant to other crops because of the common background of plant development and metabolism. Studies on one stress may help studies on other stresses because of common principles operating in diverse stress-response pathways.

A major theme in stress biology is the relationship between the evolutionary history of a crop and its whole-plant response to stress. Wild relatives of crop plants generally adopt a fail-safe strategy that, in times of stress, allocates scarce resources to just a few seeds to ensure their vigor as seedlings in the next generation rather than attempting to fill all seeds. In adapting a crop to productive agriculture, it may be necessary to supply just enough crop management or genetic improvement to prevent plants from responding to stress by unnecessarily adopting the same fail-safe strategy. Common principles that apply over species and over stresses are emerging from current studies, raising the possibility of multiple payoffs within the CGIAR system from investments made in any particular crop or stress.

Molecular tools

Completion of the Arabidopsis and rice genome sequences was announced in 2000 and 2001, respectively. The rate of development of new molecular tools will increase dramatically. For example:

• Microsatellite-based simple sequence repeats (SSR), widely used for marker-aided selection, were until recently quite laborious to find and map. Now they will be available in vast numbers, and few genes of interest will be far from a marker.

• IRRI’s activities in proteomic analysis of drought and salt responsiveness in rice led frequently to the detection and partial sequencing of interesting proteins not present in any sequence database. Now the genes corresponding to these proteins reside in databases and wait to be mined.

• Until recently we were unaware of the full complement of plant genes and were ignorant of the genes responsive to any given stress. Now Rice GeneChips carry 24,000 genes and can be regarded as a “mother array” from which trait- or stress-specific “baby arrays” can be readily derived. The “baby arrays” will be cheaper to make and use and easier to interpret.

Other useful molecular tools are also becoming available, such as insertional mutants in rice and maize, and deletional mutants in rice. These resources will allow a direct connection to be made between a gene and a phenotype. IRRI is developing a public platform in functional genomics to which many different institutes will contribute (IRRI itsef, its NARES partners, other CGIAR centers, AROs from developed and developing countries, and the private sector. The position of rice as the model cereal means that this public platform will be useful for the functional genomics activities of CIAT, CIMMYT, ICARDA, ICRISAT and WARDA as well as IRRI.

Table 2 summarizes the current prospects for isolation of stress tolerance genes from CGIAR mandated crop. The analysis is based on information about mapping of major genes and QTLs for abiotic stress tolerance in rice and other crops and about the Arabidopsis genomic initiatives. It assumes also that the public platform for rice genomics will begin operating as planned in the next 12 months. The probability of success is given a higher rating if major genes or QTLs for tolerance of the indicated stresses have been closely mapped in a cereal or if homologous Arabidopsis genes have been identified, sequenced and annotated. The difficulty in applying each stress uniformly over a large mapping population was also taken in account in judging whether QTLs might be isolated; traits that are difficult to screen for, and hence to map accurately, were given a low rating.

Table 2: Probability that four molecular approaches will lead to the discovery of genes able to enhance abiotic stress tolerance in the field

Drought research - a case study

Drought is the most important and most intractable of the abiotic stresses. As the water crisis deepens, the emphasis is on water saving through irrigation systems that are water-efficient. This means developing plants that are high-yielding even when grown under recurrent mild water deficit. At IRRI we use the term “aerobic rice” to refer to both water-efficient irrigated rice and rainfed rice made much more productive through limited irrigation. It is likely that research on water-efficient irrigation will benefit from studies on drought tolerance under rainfed conditions. Table 3 summarizes these and other opportunities to understand drought tolerance and apply the knowledge in breeding programs.

Table 3. Opportunities for enhancing drought tolerance

One traditional approach to increasing yield under drought is to avoid the stress through cultivation of short-duration varieties. This approach is most effective in areas with a likelihood of drought early or late in the season, but it is necessary in a good season to accept the yield penalty implicit in a shorter growth cycle. A modern improvement on this approach is to reduce the yield penalty by enhancing early seedling vigor, so that the crop gets off to a faster start; genes for this trait have been identified and can be used in traditional breeding or in genetic engineering. A second approach is to grow normal duration varieties with increased root density at depth, to facilitate extraction of water from a greater soil volume. Quantitative trait loci (QTLs) for root density at depth have been detected and efforts to identify the corresponding genes and exploit them for breeding are under way. This research is greatly assisted by the development of physical maps of rice and other crops, as well as the completely sequencing of the rice and Arabidopsis genomes. However, deep roots are almost invariably associated with poor tillering and low yield

A keenly awaited development is the isolation of genes controlling root elongation under drought. Some of these genes cause the drought-affected roots to enter a quiescent state in which they are less vulnerable, while other genes may stimulate growth of unaffected roots in the same plant. Another group of genes under intense study are those that help rice roots to push through the hard pan ~15 cm below the soil surface to reach moist soil underneath.

The above approaches involve drought escape or drought avoidance. Similar progress is being made in relation to drought tolerance. Research on Arabidopsis has identified a master switch that controls genes involved in tolerance of drought, salt and cold - all stresses that cause a water deficit in cells. When the sensitivity of this switch to stress is increased, the ability of the plants to tolerate all three stresses is greatly enhanced. A rice homologue of this gene has been isolated and similarly modified in the expectation of achieving drought tolerance in rice. Genes for osmotic adjustment to water loss have also been mapped in several cereals. These genes control the cellular accumulation of amino acids, sugars or ions such as potassium; high concentrations of these small, osmotically active chemicals enable cells to retain water and hence their normal structure. Genes encoding aquaporins, water-channel proteins, may help plants to acquire and distribute available water faster, while genes controlling the deposition of hydrophobic barriers between cells and on the surface of leaves help create barriers to the loss of water to soil and atmosphere, respectively.

The list of exciting avenues for drought research includes the study of the genes controlling the long-range signaling between roots and leaves (to close stomata and reduce water-loss) and between roots and developing grain (to ensure that available resources are allocated to a few seeds rather than spread thinly over many).

Should all of these possibilities be pursued? Until recently, each stress biologist tended to focus on one or two avenues that seemed promising. The efforts were fragmented, uncoordinated and impossible to evaluate within a broad perspective. Now, with the advent of genome-wide tools such as microarrays, gene chips and proteomics, these disparate approaches can be integrated into a single approach, with the hope of identifying the key events and intervention points. The fact that the Rice GeneChips produced by Affymetrix contain 24,000 out of the 40-50,000 genes of rice means that the behavior of most of the relevant genes under drought stress can be studied unbiased by personal preferences or blinkered by ignorance of whole pathways. However, the interpretation of this vast outpouring of data will require access to special genetic resources and special knowledge of traits and environments. Here lies the comparative advantage of the CGIAR Centers and their NARES collaborators.

Priorities for the CGIAR

If poor farmers had access to cultivars with enhanced tolerance of abiotic stresses, they would reduce their economic risks, improve the livelihood and nutrition of their families, put their marginal land to work, and protect the environment by providing an alternative to slash-and-burn activities. Many farmers in developed countries would also wish to see such advances made by the CGIAR Centers applied to their own crops that increasingly face similar problems. To achieve these advances, CGIAR and NARES scientists must work together and with others to take advantage of the germplasm resources held in trust by the Centers’ gene banks. Links with scientists in advanced laboratories, including the Arabidopsis community, will be essential to achieve rapid progress. CGIAR Centers can work together on environmental characterization, germplasm evaluation protocols, genomics analysis, molecular breeding strategies, crop management strategies and research on participatory plant breeding. The CGIAR Centers should remain major supporters of public sector genomics initiatives for mandated crops; it is not clear how willing and able agricultural biotechnology companies will be to share their genomics information.

An initial five-point program for CGIAR research on abiotic stresses is summarized below:

• Apply new screening protocols to the CGIAR GeneBanks to discover germplasm with novel stress-tolerance mechanisms;

• Combine diverse mechanisms to enhance yield under stress without jeopardizing yield in the absence of stress;

• Use functional genomics for gene discovery to improve molecular breeding strategies for stress tolerances;

• Use farmer participatory evaluation of new stress tolerance cultivars under diverse conditions;

• Develop CG-wide working groups on drought and salinity to share advances, especially in functional genomics.