African Crop Science Journal African Crop Science Society ISSN: 1021-9730 EISSN: 2072-6589 Vol. 3, Num. 2, 1995, pp. 223-229

Biotechnology and sorghum improvement for drought and temperature stress tolerance

N. SEETHARAMA International Crops Research Institute for the Semi-Arid Tropics (ICRISAT) Patancheru. 502 324 Andhra Pradesh. India

Code Number:CS95030
Size of Files:
   Text: 29K
   No associated graphics files

ABSTRACT

The limited progress made so far in breeding for drought, heat and coId resistance in most crops is achieved mainly by empirical approach. In spite of the great surge in physiological research on drought resistance and the contributing plant traits, its application in breeding is only modest. The problems of establishing internal consistency in the correlations between presence of traits and the intervening processes are rarely proven beyond doubt. Therefore, in spite of several advantages offered by the analytical approach, impact will be limited until we understand the physiological and biochemical components of critical traits.

However, the time is ripe to test the usefulness of simpler traits like specific metabolite or stress-induced protein accumulation by such techniques as genetic transformation. The complex traits related to overall crop development or performance under stress may be studied with molecular markers. More basic research is needed on further evaluation of the plant traits in a practical breeding programme than in plant/water relations per se or wider search for traits in the sorghum germplasm or from alien species. We can expect more rapid progress in improvement for heat and cold resistance than for drought in the near future.

Key Words: Genetic transformation, molecular markers, resistance, stress-induced proteins

RESUME

Des progres limites realises dans l'amelioration pour la resistance a la secheresse, la temperature et le froid dans la pluspart Des cultures ont ete obtenus par une approche empirique. Malgre son grand interet en recherche physiologique sur la resistance a la secheresse et la contribution Des caracteristiques Des plantes, son application en amelioration n'est que modeste. La consistance dans les correlations, entre les traits et les mecanismes qui interviennent, est rarement prouvee au dela Des doutes. Ainsi, Malgre les nombreux avantages offerts par l'approche analytique, l'impact restera limite jusqu'a ce qu' on ait compris les composants physiologiques et biochimiques qui determinent les traits importants. Cependant, le moment est venu de tester l'utilite de traits simples tels que le metabolite specifique ou l'effet de stress dans le processus d 'accumulation de proteine a l'aide de la technique telle que la transformation genetique. Quant aux traits complexes lies au developpement complet de la culture ou a la performance en cas de stress, ils peuvent etre etudies a l'aide des marqueurs moleculaires. La recherche de base est necessaire pour une evaluation ulterieure des caracteristiques des plantes dans un programme pratique d'amelioration, ce qui n'est pas le cas pour les rapports entre l'eau et lesplantes ou la recherche pour les traits dans le germoplasme du sorgho ou dans les especes etrangeres. Il semble que l'on puisse avoir rapidement de progres dans l'amelioration pour la resistance a la chaleur et au froid que pour la resistance a la secheresse dans un proche avenir.

Mots Cles: Transformation genetique, marqueurs mnleculaires. facteur stressant de l'induction des proteines

INTRODUCTION

Sorghum [Sorghum bicolor(L.) Moench] is widely distributed from 35'S to 35'N, 0 - 2250 m above sea level, and in regions receiving 300 - 1200 mm rainfall and soil pH 5.0-10.0 (Seetharama et al., 1990). In spite of this, sorghum yields are most affected by abiotic stresses, especially under the low-input management in the tropics. While yield loss due to drought (water deficit) stress is regarded as the most important, loss due to temperature stress (high and low) is underestimated, especially if one considers the lost opportunity for higher production (e.g. due to decrease in radiation use efficiency). Sorghum often experiences all these types of stresses at different stages of growth and development within a season. For example, sorghums grown in Botswana and in India during the post-rainy season suffer from low temperature stress at the beginning of the season, and from drought and heat stress during later stages of development. The interaction between the above stresses is quite complex and may be further compounded by nutrient and biotic stresses.

This paper emphasizes drought stress more than the less complex heat and temperature stresses. It is easier to measure and manipulate temperature stress than drought stress. Therefore, they are studied more frequently than drought by physiologists and biotechnologists. Discussion of either model systems or any specific traits will be avoided, as they always have a tendency to make things appear simple. An attempt will be made to project a holistic view, and then to specifically discuss how biotechnology can help to improve sorghums for drought and temperature stress tolerance.

TERMINOLOGY AND CONCEPTS

Drought and drought resistant cultivars. Crops experience drought stress when the water requirements of the plants at one or more growth stages exceed the available water in the root zone because of inadequate precipitation. In sorghum, it can occur at one or more of the following stages: early season drought (from germination or emergence to floral initiation, or approximately until the time of full ground cover), mid-season drought (reproductive development), and terminal drought (seed-set and maturation). Other soil problems such as excessive concentration of aluminum, sub-soil compact layer, etc., can also cause drought stress, but these are generally not considered separately. However, some patterns of stresses are closely linked to drought itself(e.g. lodging: Jordan et al., 1984). Generally, drought resistance is the ability of a genotype to yield satisfactorily in areas subjected to periodic water deficits (Turner, 1979). The definition of phenotype is very important in the context of drought resistance, as genotypes are selected on the basis of phenotypic expression.

Drought resistance trait or mechanism. In the absence of established correlations between field performance under drought and plant traits or mechanisms, plant performance cannot be directly used to characterize genotypes as drought resistant. However, it may be useful in evaluation of germplasm and breeding, and is critical for the application of biotechnological tools.

BREEDING EFFORTS

Plant breeders have developed cultivars tolerant to environmental stress without being aware of the selective effect of the testing environments or the underlying mechanisms. In most cases, the elite cultivars adapted to the region are described as drought resistant, based mainly on multilocation or multi-year performance. To date, the progress in breeding for drought resistance is slow and sometimes even doubted. Among the specific reasons listed for slow progress, the multiplicity of drought patterns and plant responses are the foremost (Seetharama et al., 1983). However, by organizing our thoughts and actions around a few fairly distinct patterns, the problem can be reduced to a manageable size.

Approaches to screening genotypes for drought resistance. In spite of the apparent advantages the physiological approach of screening for specific constitutive or inducible (under stress) plant traits offers, this approach lacks wide support. The reason is that there is rarely a unique and consistently quantifiable relationship between the measured trait response and the final yield. On the other hand, empirical or agronomic screening can only be of local interest and has to rely on the consistency of stress patterns in the normal rainy season over many years. In the long run, agronomic and physiological approaches are complementary. Emphasis on either of them should be based on the need and opportunity that exists for the individual researcher (Blum, 1987) and the stage of development of a crop improvement programme.

A challenge to physiologists is to include integration of component traits into processes that determine yield. Seetharama et al. (1983) advocated wider use of 'operational traits' such as leaf area or yield components, which are intermediate in complexity, and integrative in effect. However, such traits are also very plastic and covary with increase in one compensating for another. Therefore, they have to be used only as an adjunct to other criteria such as phenology and rooting pattern. Finally, regardless of the philosophy under which a cultivar is bred, only an empirical field test can provide satisfactory proof of a cultivars' drought resistance. The field tests need not be confined to measurements of final yield only~ biomass, lodging, quality, or growth during stress period can also be used.

Empirical field screening can be more efficient if 'hot spots' and rain-out shelters are available. In the tropics, screening for physiological traits can be done during the normal season or during the rain-free dry seasons, when one can manipulate drought stress, intensity and pattern by manipulating irrigation regimes. While yield under stress can be used to identify genotypes. indices based on relative yield loss under stress is more useful when corrections are made for differences in variables such as time to flower. Flexibility with the measurement and interpretation of physiological traits and integration of values over the specified periods of stress can be useful.

Sources of drought resistance. Some scientists believe that access to diverse germplasm and knowledge of environmental characteristics of the germplasm collection sites could guide in selecting lines most apt to have desired characteristics in a target region. However, there is no evidence for the usefulness of 'pointed' collections for specific traits. Use of wild species for improving crops for specific environmental stresses, unlike for breeding disease and pest resistance, are rare (Downton, 1984). Genetic engineering may change this situation.

ADAPTIVE MECHANISMS, STRESS LEVELS, AND BREEDING

Where is breeding for drought resistance most profitable? Prolonged and catastrophic droughts leave little scope for exploiting additional drought avoidance or tolerance in crops, as in sub-Saharan Africa. But drought resistance traits can be important in stabilizing crop production in large areas of rainfed agriculture that frequently suffer from moderate to serious intermittent drought (McWilliams, 1986). When the stress levels are mild, infrequent, and unpredictable, empirical performance at optimal conditions may still be the best guide, as breeding exclusively for drought tolerance may not be cost-effective in such cases.

Priorities for breeding. These are: (1) to reduce drought risk and to achieve maximum productivity by combining choice of sowing date and cultivar maturity for the target reg~on (drought escape); (2) to maximize water use during the period of water availability by incorporating such traits as rapid leaf area development and deep rooting (drought avoidance); and (3) to incorporate plant characteristics useful in increasing yield stability and resistance to catastrophic events during stress periods (e.g. heat stress, photo-inhibition, etc.) drought tolerance.

Some may argue that the third component described above should be the first to be tackled. However, experience to date has shown that selection for drought escape following avoidance has been the most effective component of improved cultivars in many crops, and this trend is likely to continue as it favours higher yields under stress than tolerance mechanisms. If plants can be genetically engineered for specific tolerance mechanisms useful for the target region, the above conclusion may be reversed.

Breeding strategy. Under mild stress, the rankings of sorghum genotypes for potential and drought yields are similar. Under such situations mere testing for drought resistance may be adequate. The consistency, rather than the magnitude of genotype differences is critical for selecting test environments. While the materials excelling in specific traits can be easily exchanged, final selections must be tailored to meet the local drought profile and intensity.

Range of materials, and testing. Because contrasting adaptive strategies are present among sorghum genotypes, there is need for generating a wide range of cultivars by breeders working for large and diverse regions. There is a constant need for 'fine-tuning' at local levels, and the final choice ofcultivar lot the region should be left to the (well-informed) growers. The selections for yield and wide adaptation. determined on the basis of multi-location testing, may or may not be useful in selecting for drought resistance. The apparent dichotomy in the approaches of CIMMYT. Mexico (Osmanzai at al., 1987) and ICARDA, Syria (Srivastava et at., 1987). with respect to their wheat improvement programmes, arises from the difference in the extent of wheat growing areas and severity of drought in the two respective regions. Crop improvement scientists working for large regions (e.g. those in International Centres) need a range of cultivars to take advantage of several sets of local conditions.

TRAIT-BASED CROP IMPROVEMENT APPROACH

Plant breeders rarely have the single objective of breeding lot drought resistance; instead, drought resistance is one of the traits selected. It is now widely appreciated that by moving from use of an undefined environment to a logistically efficient testing scheme, breeders can select for enhanced drought resistance. The key is to direct research towards a specific pattern of the drought in a given zone and gradually attempt to combine for drought resistance at other growth stages.

Choosing traits. For empirical selection, consideration of phenological and morphological traits should precede over measurement of physiological and biochemical traits, as the former group of traits can be readily transferred and easily verified elsewhere. However, in advanced breeding materials, the range in variability for these traits is very narrow. At that point. it is imperative that one should search for other traits. Interpretation of data on expression of metabolic traits is not straight forward and is crop specific. For example, high abscisic acid (ABA) accumulation represents drought tolerance in sorghum and maize, but causes drought susceptibility in wheat (Jones, 1983). When the metabolic traits or phenomena are directly related to crop phenology or morphology (e.g. phytochrome) they may be more useful.

Proving the usefulness of traits. Examples of control of drought resistance by single genes arc rare. Proving the usefulness of such traits is easy with isogenic tines. Most breeders arc unwilling to invest the necessary resources into breeding for physiological response to stress. Although it may be impractical to prove the benefits of such traits in directly improving yields under drought, the effect of the trait on specific components of yield and survival can be easily assessed. But the readily observable trait is associated with drought resistance. then such characters will be widely used e.g. the reduction in time interval between artthesis and silking in maize (CIMMYT, 1987)1. Cost-effectiveness of screening for physiological traits is an important consideration. Kanangara et al. (1983) found that the size of correlation coefficients between ABA levels and sorghum grain yield was similar to that between the amount of irrigation water applied and yield. This showed no advantage in measuring ABA levels, which was not practical in the early eighties. Now, however, with the availability of monoclonal antibodies, this technique is more practical.

Contribution of individual traits must be compatible with other traits useful in productivity. The number of characters contributing to yield stability under drought is large (e.g. >20 in sorghum; Seetharama et al., 1982), but one can choose a few (2-3) either together or during different generations.

We must aim at optimal, rather than maximum expression of drought resistance traits. An example will follow. In West Africa, the length of the growing season is reduced by 3-4 weeks per degree as one moves away from the equator. In India, on the other hand, the reduction is far less (0.9 week 'N^-1). This would mean that one should select for low photoperiod response in India and high photoperiod response in West Africa (F.R. Bidinger, ICRISAT Centre, personal communication). Changing root/shoot ratio is another example.

BIOTECHNOLOGY AND BREEDING FOR DROUGHT RESISTANCE

The lack of significant impact on sorghum drought resistance by traditional breeding has raised expectations from biotechnology. The greatest difficulty in the application of biotechnology to improve stress resistance in plants is the inadequate quantitative understanding of (or even in some cases the conception of steps involved) plant responses to drought. Therefore, the possibility of explaining all aspects of drought resistance at a molecular level may be unreasonable. Biotechnologists and physiologists need to establish a more secure base for their experimentation by improving communication with others. A trait in a breeder's or crop physiologist's sense is some response that can be quantified, or its presence or absence merely noted. If this trait represents a reasonably small number of genes, molecular geneticists may be able to assist with plant improvement for drought. Therefore, greater application of molecular approaches to drought resistance will have impact only when we are able to dissect each drought response ('operational trait') in terms of the 'gene' in the molecular biological sense. The recent approaches using gametes, somatic cells, and tissues as units for inducing mutation and selection as means of enhancing variability within the accepted range of varieties may be fruitful. Distinction must be made between cellular and whole plant responses to water stress. Resistance to low water potential developed in the presence of osmoticum in culture medium can be lost in the absence of it (Bressan et al., 1981).

Genetic transformation may prove useful for efficient transfer of traits governed by a single or a few genes from unrelated species to the target crop. Such attempts, however, are not yet successful even in a model system (Barrels et al., 1992). Nevertheless, more studies are needed to transform plants to express stress-induced proteins and other metabolites under a variety of conditions to better understand their roles. Most of the stress resistance sources are restorer lines; therefore, to begin with, it may be worthwhile to target B-lines for genetic transformation. The anti-sense technology is also useful in studying the effect of specific genes. With the advancement made in the construction of the sorghum RFLP map, it is timely to start utilizing the maps and DNA probes for selecting complex resistance traits or mechanisms. Unfortunately the cytogenetics of sorghum (diploid) is not well developed to take advantage of proper genetic stocks (e.g. reciprocal chromosomal substitution between lines to locate genes on chromosomes for complex characters like lodging and drought in wheat).

Biotechnology may offer new diagnostic tools to investigate stress metabolism (e.g. cheaper and sensitive methods to quantify plant hormones, stress metabolites or protectants, etc.). One specific use of the new tools should be to optimize the productivity and drought resistance traits, as many of them are negatively associated in the field (Seetharama et al., 1983) and in in vitro systems (Singh et al., 1989). The potential of mutagenesis to provide genetic material for elucidating biological mechanisms of drought resistance (Axtell et al., 1992) needs to be fully exploited. Much more emphasis needs to be placed on physiological genetics in the future.

What traits can be studied? The following traits deserve further attention by physiologists and biotechnologists: ( 1 ) maturity and developmental plasticity; (2) seed and seedling vigour; (3) seedset and growth under stress; (4) dry matter distribution under normal and stress conditions; (5) non-senescence and resistance to lodging; (6) root system: components, plasticity and deep rooting; (7) photosynthesis and transpiration efficiency; (8) plant hormone levels, and stress-induced proteins and (9) molecular diversity of core collections to identify truly diverse germplasm.

Some attempts are already in progress to study non-senescence and the fundamental processes like plant development. Most of them are in the USA where they largely rely on file molecular marker technology. These studies may serve global sorghum research better if the proper linkages exist between sorghum workers in different regions.

TEMPERATURE STRESS

Sorghum crops often experience heat stress more commonly than cold stress. Heat stress is usually associated with drought. especially during germination and grain filling stages. The correlation between heat stress and production of heat-shock proteins (Peacock et al., 1993) is not followed by any suitable validation experiments to prove their practical significance.

Cold stress can affect sorghum at germination, seedling growth, microsporogenesis, and seed setting stages (Seetharamaetal et al., 1994). Although there are several local landraces identified as resistant at each stage, development of high yielding new cultivars with significantly higher levels of cold tolerance has not been developed. Since the temperature stress related genes (hsps: heat-shock proteins, and or cold-regulated proteins) are expressed in the pollen, gametophytic selection for improvement of these traits is worth exploring (Ottaviano et al., 1991; Tarchini et at., 1994). Effects of sub-lethal temperatures on overall plant development, respiration rate, and source-sink relationships (Wardlaw. 1993) may be useful in understanding plant adaptations and ways of increasing productivity under normal, drought and temperature stress conditions.

REFERENCES

Axtell, J., Pelters, P. and Zehr, U. 1992. Mechanism of drought resistance in sorghum. Pages 157163. In:Proceedings of the 47th Annual Corn & Sorghum Research Conference.

Bartels, D., Velasco, R., Schneider, K., Forlani, F., Furini, A. and Salamini, F. 1992. Resurrection plants as model systems to study desiccation tolerance in higher plants. In: Biotechnology for Arid Land Plants. Marby, T.J., Nguyen. H.T., Dixon, R.A. and Bonnes. M.S. (Eds.), pp. 47-58. IC^2 Institute, University of Texas, Austin, TX, USA.

Blum, A. 1987. Breeding crop varieties for stress environments. CRC Critical Review of Plant Sciences 2(3):199-238.

Bressan, R.A., Hasegawa, P.M. and Handa. A.K. 1981. Resistance of cultured higher plant cells to polyethylene glycol induced water stress. Plant Science Letters 21:23-30.

CIMMT (International Maize and Wheat Improvement Center. 1987.1983 Report on Wheat Improvement. CIMMYT. El Batan, Mexico. p.18.

Downton, W.J.S. 1984. Salt tolerance of food crops. 1984. CRC Critical Review of Plant Sciences 1:183-201.

Jones. H.G. 1983. Plant and Microclimate. Cambridge University Press. 271 pp.

Jordan, WR., Clark, R.B. and Seetharama, N. 1984. The role of edaphic factors in disease development. In: Sorghum Root and Stalk Rots:A Critical Review. Mughogho, L.K. and Rosenburg. G. (Eds.), pp. 81-97. ICRISAT, Patancheru 502324. A.P., India.

Kanangara, T., Seetharama, N., Durley, R.C. and Simpson, G.M. 1983. Drought resistance of Sorghum bicolor. Changes in endogenous growth regulators of plants grown across an irrigation gradient. Canadian Journal of Plant Sciences 63:147-55.

McWilliams, J.R. 1986. The national and international importance of drought and salinity effects on agricultural production. Australian Journal of Plant Physiology 13:1-13.

Osmanzai, M., Rajaram, S. and Knaff, E.B. 1987. Breeding for moisture stressed areas. In: Drought Tolerance in Winter Cereals. Srivastava, S.P., Porceddu, E., Acevedo, E. and Verma. S. (Eds.), pp. 151-162. John Wiley & Sons, Inc., New York. NY, USA.

Ottaviano, E., Gorla, M.S., Pe, E. and Frova, C. 1991. Molecular markers (RFLPs and HSPs) for the genetic dissection of thermotolerance in maize. Theoretical Applied Genetics 81: 713-719.

Peacock, J.M., Soman, P. and Howrath, C.J. 1993. High temperature effects on seedling survival in tropical cereals. In: Adaptation of Food Crops to Temperature and Water Stress. Kuo, C.G. (Ed.), pp. 106-121. Proceedings of an International Symposium, 13-18 Aug. 1992, Taiwan. Asian Vegetable Research and Development Center, Taipei 10099. Taiwan.

Seetharama, N., Huda, A.K.S., Virmani, S.M. and Monteith, J.L. 1990. Sorghum in the semi-arid tropics: agroclimatology, physiology, and modelling. In: Proceedings of the International Plant Physiology Congress. Sinha, S.K., Sane, P.V., Bhargava, S.C. and Agrawal, P.K. (Eds.), pp. 142-151. Society for Plant Physiology and Biochemistry, IARI/WTC, New Delhi, India.

Seetharama, N., MagiIl, C.W. and Miller, F.R. 1994. Molecular markers for cold tolerance in sorghum. In: Use of Molecular Markers of Sorghum and Pearl Millet Breeding for Developing Countries. Witcombe. J.R. and Duncan, R. (Eds.), pp. 32-34. Overseas Development Administration (ODA), UK.

Seetharama, N., Reddy, B .V.S. Peacock, J.M. and Bidinger, F.R. 1982. Sorghum improvement for drought resistance. Pages 317-338. In: Drought Resistance in Crops: With Emphasis on Rice. International Rice Research Institute. Los Banos, Philippines.

Seetharama, N., Sivakumar, M.V.K., Bidinger. F.R., Sardar Smgh. Maiti, R.K., Reddy. B .V.S., Peacock. J.M., Reddy, S.J., Mahalakshmi. V., Sacban, R.C., Kanangara, T.C., Durley. R.C. and Simpson, G.M. 1983. Physiological basis for increasing and stabilizing yield under drought in sorghum. Proceedings of the Natiottal Academy of Science, USA. B 49:498-523.

Singh. N.K., LaRosa, P.C., Nelson, D., Iraki, N., Carpira, N.C., Hasegawa, P.M. and Bressan, R.A. 1989. Reduced growth rates and changes in cell wall proteins of plant cells adapted to NaCl. In: Environmental Stress in Plants. Cherry, J.H. (Ed.), pp. 173-194. Springer-Verlag, Berlin. Germany.

Srivastava, J.P., Porceddu, E., Acevedo, E. and

Verma. S. 1987. Drought Tolerance in Winter Cereals. John Wiley & Sons. Inc., New York., NY, USA. pp. 65-78.

Tarchini, R., Gorla, S.M. and Pe. M.E. 1994. Male gametophytic selection: Perspectives for sorghum and pearl millet improvement. In: Use of Molecular Markers in Sorghum and Pearl Millet Breeding for Developing Countries. Witcombe, J.R. and Duncan, R. (Eds.), pp. 46-52. Overseas Development

Administration (ODA), UK.

Turner, N.C. 1979. Drought resistance and adaptation to water deficits in crop plants. In: Stress Physiology in Crop Plants. Mussell, H. and Staples, R.C. (Eds.), pp. 343-372. John Wiley & Sons, Inc., New York, NY, USA.

Wardlaw, I.F. 1993. Temperature effects on source-sink relationships: a review. In: Adaptation of Food Crops to Temperature and Water Stress. Kuo. C.G. (Ed.). pp.159. Proceedings of an International Symposium. 13-18 Aug., 1992, Taiwan. Asian Vegetable Research and Development Center, Teipei. Taiwan.

Copyright 1995 African Crop Science Society