3.1 Exploiting Stomatal Closure: Deficit Irrigation and Partial Rootzone Drying
It has been known for some time that chemical compounds synthesised in drying roots can act as long-distance signals, which induce stomatal closure in the leaf or restrict leaf growth via arrest of meristematic development. As a result, in some cases, stomatal closure can occur without significant changes in shoot water status. This occurs where plant water potential is buffered by controlling stomatal aperture via feed-forward mechanisms.
Plants that show this response are said to be "isohy-dric". Even where the leaf water potential is similar, plants exposed to a water deficit will have a lower xylem water potential - and it is this that controls leaf growth - on account of a reduced gradient of water potential from the roots through the leaves due to reduced water flux (Tardieu et al. 2010); this difference between xylem and leaf water potential may result in reduced leaf growth in the plants experiencing a water deficit. Careful manipulation of soil water availability to induce a mild water deficit allows minimisation of the impact of the deficit on shoot water status (Davies et al. 2000) . This has been exploited in "Deficit Irrigation" strategies, where less than 100% of crop evapotranspiration is replaced by irrigation, and in variations on Deficit Irrigation, one of which is known as Partial Rootzone Drying. In Partial Rootzone Drying, water is applied only to one side of the roots, so that the other side is exposed to drying soil, with the side being irrigated switched at intervals. Such techniques have allowed water savings without reducing yield (e.g., dos Santos et al. 2003, Grant et al. 2004).
The potential of exploiting plant responses to dry soil/substrate is not confined to food production. In landscaping under semi-arid conditions, transplantation is more likely to result in successful growth if the plants have been pre-conditioned to dry conditions. Thus deficit irrigation is increasingly being used in the production of ornamental nursery stock with reduced shoot height and/or leaf area, increased root-collar diameter, root growth potential, and root:shoot ratio, increased osmotic adjustment and water use efficiency, and low stomatal conductance, leaf water and turgor potentials, and relative water content (Franco et al. 2006). Variations on the idea of deficit irrigation can be applied to this end - for example Banon et al. (2006) exposed Nerium oleander seedlings to both deficit irrigation and low air humidity on the nursery, prior to transplant, with the result that mortality after transplant was reduced from 92 to 32% compared to control plants; Franco et al. (2006) highlight that microclimate management during the nursery phase can be an effective means of producing high-quality seedlings capable of withstanding transplant shock and capable of rapid establishment in arid landscapes. Even where water is not limiting, deficit irrigation can be used to control the size and quality of hardy ornamental nursery stock (Cameron et al. 2006) , particularly where the application of deficit irrigation is combined with an efficient method of sensing plant water requirements that is suited to application on the nursery (Grant et al. 2011).
In addition to the impact of long-distance signalling, mild water deficits also exert direct or indirect impacts on yield and the quality of harvested products: for example, reduced leaf growth may improve the light environment around fruit while the fruit are developing.
3.2 Exploiting Genotypic Variation in Breeding Programmes
Long-term sustainability requires not only more resource-efficient production systems but the development of crops with improved resource use efficiency, and crops which are tolerant to specific stresses. Clearly there is substantial variation between cultivars in tolerance of water deficits, which could be exploited in breeding programmes. This has been clear for many years in the major food crops (e.g., Fischer and Wood 1979, Quarrie 1983, Farquhar and Richards 1984), but such variation is also now being revealed in a wide range of other crops - for example, fruit crops (Klamkowski and Treder 2008, Grant et al. 2010) , energy crops (Zub and Brancourt-Hulmel 2010), and forest trees (Brendel et al. 2008) .
Despite such genetic variation, there has been limited progress to date in developing drought tolerance (Tuberosa and Salvi 2006). This is partly due to a shortage of research exploring the physiological impact of altering crop genetics. The genetic resources for investigating drought tolerance have increased dramatically in recent years, but phenotypic research has not kept pace (Miflin 2000, Verslues et al. 2006). This in turn is largely a consequence of the difficulty of measuring plant physiological processes with the required spatial and temporal resolution to match molecular investigations.
Stomatal conductance (gs) largely determines transpirational water loss, and also influences photosynthetic assimilation and hence yield. It is therefore useful to screen genotypes for gs in programmes to improve yield or water use efficiency (e.g., Gutierrez-Rodriguez et al. 2000). Cuvette-methods for measuring gs, however, are not suited to large-scale screening. Infrared thermometry (and more particularly infrared thermal imaging) has opened up the possibility of remotely assessing the temperature of individual leaves or whole crop canopies (Jones 2004) and may be a useful tool for early-generation selection of physiologically superior lines in breeding programmes aiming to increase yield or reduce the impact of stressful environments (Blum et al. 1982, Amani et al. 1996, Reynolds et al. 1999, Olivares-Villegas et al. 2007, Lopes and Reynolds 2010). Chlorophyll fluorescence provides information on photochemistry, including photochemical capacity and electron transport rate. Chlorophyll fluorescence imaging in the field is still in its infancy, but the photochemical reflectance index (PRI) derived from images taken at the 530 nm and 570 nm wavelengths may provide a good indication of the photosynthetic functioning of plant leaves (Inoue and Peñuelas 2006), and has also been suggested as a good indicator of water stress (Súarez et al. 2010) . Combined together, thermal and chlorophyll fluorescence images taken in quick succession on the same area of leaves could be used to rapidly determine the ratio of photosynthetic assimilation to stomatal conductance i.e., WUE. (see review by Chaerle et al. 2007), but to date imaging has not been used for this purpose.
From the above, a picture emerges of imaging techniques potentially providing a solution to the problem of needing to screen physiological responses over large numbers of plants and at different time-points during the imposition of stress or during the crop's development (Munns et al. 2010), as they allow real-time continuous assessment of physiological processes across organs, whole plants, or even whole plant populations or crops. More established phenotyping techniques include assessment of stable isotope composition. During photosynthesis, the extent of discrimination against the naturally abundant and heavier isotope of carbon, 1 3 C, is related to the ratio of internal to external partial pressure of CO2 (pi/pa), which is controlled by both stomatal conductance and photosynthetic capacity, and therefore is indirectly related to water use efficiency (Farquhar and Richards 1984), with greater enrichment being associated with greater photosynthetic water use efficiency. The level of enrichment can be measured by determining the ratio of 13C and 12 C and comparing the ratio to that of a standard, to determine the carbon isotope composition, 813C. 813C integrates the ratio of assimilation to transpiration over the duration in which dry matter is assimilated. The measurement is suited to screening in crop breeding programmes, since only small samples of material are required, harvesting plant material is rapid compared to measurement of, for example, photosynthetic or transpiration rates, and once dried the plant material can be stored until used for isotope analysis. Additionally, 813C shows high heritability (Condon and Richards 1992, Richards 1996). Use of 813C to screen for variation in WUEi has been successfully applied in breeding programmes for water-limited environments (Condon et al. 2004). In the case of bread wheat under rainfed conditions, selection for high S13C was more efficient than direct selection for either high biomass or high yield, and has led to increases in grain yield (Rebetzke et al. 2002). Genotypes with higher photosynthetic water use efficiency should be more productive where water availability is limited. However, under irrigation, genotypes with lower photosynthetic water use efficiency (relating to higher transpiration) have been found in some cases to perform better (Araus et al. 2003). Low water use efficiency in such situations is associated with faster growth, and consequently, at least in cereals, greater total biomass and higher yields (Condon et al. 2004). In such cases, higher yield is associated with poorer enrichment in 13C. For irrigated crops, this needs to be borne in mind in prediction of genotypic advantages on the basis of S13C. S13C analysis may need to be combined with other techniques to select for genotypes with both relatively high instantaneous water use efficiency and relatively high transpiration under the desired environmental conditions. Assessment of oxygen isotope composition (S18O, which reflects evaporative enrichment in leaves due to transpiration, and has been shown in some crops to correlate negatively with transpiration rate (Cabrera-Bosquet et al. 2009) in parallel with S13C could indicate the extent to which WUEi is influenced by transpiration. While isotope composition is undoubtedly an effective measure in many situations, isotope composition analysis (particularly with respect to S18O) is relatively costly and preparation of the plant material is labour-intensive.
Although conventional breeding has been successful to some extent in developing drought-tolerant cultivars, such selection programmes are expensive and slow, and it has proven impossible to incorporate certain traits, for example, osmotic adjustment, via conventional breeding (Zhang et al. 1999). More rapid development of varieties with drought resistance should be possible through exploitation of mapped molecular markers (Price and Courtois 1999). Restriction fragment length polymorphism (RFLP) and other molecular markers allow loci-controlling traits related to drought tolerance to be identified and located in the genome. Mapping of quantitative trait loci (QTL) allows dissection of complex traits into their components, each of which is controlled by one or more QTL. Genetic markers at the identified QTL can then be used in rapid selection of plants with the desired alleles. As explained by Price and Courtois (1999), the process is as follows: A trait with potential value in breeding for drought tolerance is chosen, parental lines displaying extreme phenotypes for this trait are identified, these lines are crossed to produce progenies that segregate for the trait of interest, the trait in question is determined in each of the progeny, the parents are screened for genetic polymorphism in many molecular markers, the genotype of each progeny is determined for the selected markers, a genetic map is constructed from the marker data, and finally markers associated with the trait are identified using analysis of variance or interval mapping. Near-isogenic lines, produced through repeated back-crossing, allow fine-scale mapping of QTL (Yadav et al. 2011). If QTL for traits expected to relate to drought tolerance are developed, then near-isogenic lines can be used to characterise the impact of specific QTL on yield (or other commercially-desirable characteristics) in the presence or absence of drought (Price and Courtois 1999). Alternatively, recombinant inbred lines can be used - Sanchez et al. (2002), for example, identified four QTL associated with the stay-green trait in sorghum - a trait which confers resistance to premature senescence under water stress post-flowering - using a recombinant inbred line population; these QTL accounted for more than 50% of the phenotypic variance in field trials. Another example of the power of QTL mapping is the development of near-isogenic lines from two elite cotton cultivars (Levi et al. 2011); each near-isogenic line expressed advantageous osmotic adjustment in comparison to the parents. While there was a tendency for the near-isogenic lines to show higher concentrations of certain metabolites than the parents, the authors acknowledge that further analysis is necessary in order to establish a direct impact of these metabolite concentrations on drought tolerance, but highlight the value of developing genomics information on different crops for dissecting the causes of enhanced drought tolerance. Work is also underway to determine QTL associated with such traits as root morphological characters (Price and Courtois 1999) .
Although QTL mapping holds great promise, mapping drought tolerance remains complicated. One reason is that naturally occurring drought is often unpredictable and therefore researchers must artificially create drought - but while this allows control of the timing and duration of drought, it may not be representative of the "real" droughts to which the crops are normally exposed (Price and Courtois 1999) . As with conventional breeding, the choice of trial sites and the way in which drought is imposed remain hugely important.
Stress-induced transcript expression profiles have been used to identify candidate genes associated with QTLs (Diab et al. 2004, Gorantla et al. 2005, Street et al. 2006). Where both a genome sequence and microarrays are available for a species, expressed sequence tags (ESTs) spotted on microarrays can be located on the genome sequence; if QTL regions have already been identified, then all ESTs on a microarray and within a QTL region can be examined for differential expression (Street et al. 2006) . The expression levels when quantified within a mapping population are used to map "expression QTL".
3.3 Exploiting Gene Expression During Drought: Genetic Engineering
I t is now appreciated that an understanding of photosynthetic metabolism under drought is essential if plants are to be engineered that maintain high yield under water deficit (Lawlor and Tezara 2009). A couple of examples provide an indication of the potential, but also the complications, involved in genetic engineering for drought tolerance.
As the first example, over-expression of SODs that function in the water-water cycle in chloro-plasts in theory would improve photosynthesis during environmental stress, allowing the maintenance of close to normal levels of PSII (photosystem II) and PSI (photosystem I) activity during stress, decreasing the inhibition of photosynthesis and reducing ROS levels. As discussed earlier, while ROS have the potential to cause oxidative damage to cells during environmental stress, they also appear to play a key role as signal transduction molecules involved in mediating responses to pathogen infection, environmental stress, and different developmental stimuli (Miller et al. 2010). The steady-state level of ROS in cells therefore needs to be carefully controlled - both under normal metabolism and under stress. Elucidating the mechanisms that control ROS signalling in cells during environmental stress could provide a powerful means to enhance the tolerance of crops to such stress.
As a second example, a number of genes encoding osmolyte biosynthesis have been transformed into model plant species (Zhang et al. 1999) . The transformants generally show enhanced drought tolerance, yet these studies have not actually shown that the increase in accumulated solutes in the transformants caused osmotic adjustment. Zhang et al. (1999) suggest that osmolytes may play a more complex role in conferring drought resistance than simply contributing to osmotic adjustment, but the emergence of the final expected result (increased drought tolerance) without the expected intermediate result (enhanced osmotic adjustment) is a salutary example of the complicated interaction of different processes in drought response. This also highlights how much work is still required before genetically engineered drought-tolerant varieties are likely to be available for crop production.
Was this article helpful?
How would you like to save a ton of money and increase the value of your home by as much as thirty percent! If your homes landscape is designed properly it will be a source of enjoyment for your entire family, it will enhance your community and add to the resale value of your property. Landscape design involves much more than placing trees, shrubs and other plants on the property. It is an art which deals with conscious arrangement or organization of outdoor space for human satisfaction and enjoyment.