- Top of page
- Materials and Methods
Root growth in anaerobic waterlogged soils depends on the O2 available via diffusion from above-ground parts through root aerenchyma (Armstrong, 1979; Colmer, 2003a). Aerenchyma provide a low-resistance internal pathway for O2 diffusion in roots (Armstrong, 1979; Justin & Armstrong, 1987). Wetland species, such as rice and Hordeum marinum, possess constitutive aerenchyma, and the volume is further enhanced when O2 deficiency occurs in the root zone (rice –Colmer, 2003b; H. marinum–Garthwaite et al., 2003). By contrast, dryland species, such as wheat, form aerenchyma following waterlogging, but the volume is typically lower than that in most wetland species (wheat –Thomson et al., 1992; Huang et al., 1997; wetland species –Justin & Armstrong, 1987). Moreover, the movement of O2 to the apex is further enhanced in the roots of many wetland species by the formation of a barrier to radial O2 loss (ROL) in the basal root zones (Armstrong, 1979; Colmer, 2003a), a feature lacking in the roots of dryland species (Jackson & Drew, 1984).
The introduction of new genes from undomesticated species into crops has been suggested as one approach to improve waterlogging tolerance (wheat –Taeb et al., 1993; maize –Mano & Omori, 2007). A degree of tolerance to waterlogging has been demonstrated in Lophopyrum elongatum, a wild relative of wheat (Taeb et al., 1993; McDonald et al., 2001). However, waterlogging tolerance reported for a disomic chromosome addition line of L. elongatum in wheat (Taeb et al., 1993) could not be confirmed (McDonald et al., 2001). Moreover, L. elongatum and the L. elongatum–wheat amphiploid both lack a barrier to ROL in roots (McDonald et al., 2001). The identification of H. marinum as a waterlogging-tolerant (McDonald et al., 2001; Garthwaite et al., 2003; Malik et al., 2009) and salt-tolerant (Garthwaite et al., 2005; Malik et al., 2009) species in the Triticeae, and successful hybridization of wheat with H. marinum (Islam et al., 2007), might provide a new source of genes for abiotic stress tolerance in wheat (Colmer et al., 2006a). An H. marinum–wheat (T. aestivum cv Chinese Spring) amphiploid demonstrated enhanced salt tolerance compared with the wheat parent (Islam et al., 2007), but waterlogging tolerance of the amphiploid has not been evaluated previously.
Waterlogging tolerance in H. marinum is associated with superior root aeration traits, when compared with its dryland relatives (McDonald et al., 2001; Garthwaite et al., 2003). Hordeum marinum develops some aerenchyma even in well-aerated conditions and, in O2-deficient medium, develops moderately higher porosity in adventitious roots than do wheat (McDonald et al., 2001) and barley (Garthwaite et al., 2003). In addition, putative suberin deposits in the walls of the hypodermis form a barrier to ROL in the roots of H. marinum when grown in stagnant solution (Garthwaite et al., 2008). The experiments reported here evaluated root aeration traits (adventitious root aerenchyma, total porosity and formation of a barrier to ROL) and tolerance to growth in stagnant solution in four H. marinum–wheat amphiploids: the earlier reported amphiploid from accession H21, evaluated for salt tolerance by Islam et al. (2007), plus three new amphiploids. In addition to the solution culture experiments, tolerance to soil waterlogging and root ROL barrier expression were also evaluated for H. marinum (H21), wheat and their amphiploid in a soil pot experiment.
Materials and Methods
- Top of page
- Materials and Methods
Amphiploids (genome AABBDDXX) of bread wheat (Triticum aestivum L. cv Chinese Spring, genome AABBDD) with four accessions (H21, H87, H109 and H155) of Hordeum marinum Huds. (genome XX), produced by AKMR Islam, were used in this study. The H numbers identify accessions from the Nordic Gene Bank, kindly provided by R. von Bothmer (Swedish Agricultural University, Alnarp, Sweden). Wheat was used as the male parent in crosses to H. marinum (female parent) owing to the small anther size of H. marinum, which restricts pollen yields and thus impedes the use of H. marinum as the male parent. The production of the H. marinum (H21)–wheat (cv Chinese Spring, abbreviated as CS) amphiploid has been described in Islam et al. (2007); the same approach, but with hybridization of CS onto three different H. marinum accessions, was used to produce the three new amphiploids reported here for the first time.
Solution culture Expt 1
Experimental design The experiment was as follows: two treatments (aerated or stagnant) × six genotypes × three replicates, in a completely randomized design, in a naturally lit phytotron (20 : 15°C day : night) located in Perth, Western Australia (experiment conducted during April and May). Growth and adventitious root aerenchyma were evaluated. The six genotypes were as follows: four H. marinum–wheat amphiploids (H21–CS, H87–CS, H109–CS and H155–CS), H. marinum (H21) and wheat (T. aestivum cv CS). Space limitations in the phytotron prevented the inclusion of all four H. marinum parents in this first experiment, but all were included in Expt 2; data are also available on the growth of H21, H87, H109 and H155 in stagnant solution from an earlier screening of a diverse set of H. marinum accessions (Malik et al., 2009).
Plant culture Hordeum marinum was germinated 3 d before wheat and the amphiploids to ensure that plants were at a similar developmental stage (Haun, 1973) at the start of treatment. Seeds were surface sterilized with 0.4% commercial bleach, rinsed with tap water, imbibed in aerated 0.5 mM CaSO4 for 3 h and then transferred to plastic mesh floating on aerated 0.1-strength nutrient solution in 4.5-l pots. The pots were covered with aluminum foil so that the seeds were in darkness.
Seedlings of H. marinum were exposed to light after 4 d, and wheat and amphiploid seedlings were exposed to light after 3 d. At the time of exposure to light, the concentration of the nutrient solution was increased to 0.25-strength. Seven-d-old H. marinum and 4-d-old wheat and amphiploids were transplanted into 4.5-l pots containing full-strength nutrient solution (McDonald et al., 2001). The composition of the nutrient solution at full strength was as follows (mol m−3): K+, 5.80; Ca2+, 1.50; Mg2+, 0.40; NH4+, 0.625; NO3−, 4.375; SO42−, 1.90; H2PO4−, 0.20; Na+, 0.20; H4SiO4−, 0.10; and micronutrients (mmol m−3): Cl, 50; B, 25; Mn, 2; Zn, 2; Ni, 1; Cu, 0.5; Mo, 0.5; Fe-EDTA, 50. The nutrient solution contained 2.5 mol m−3 2-N-morpholinoethanesulfonic acid (MES), and the pH was adjusted to pH 6.5 using KOH (to give the final K+ concentration as listed above). The solution in all pots was renewed every 7 d. Six pots of each genotype were established and each pot contained three plants.
Fourteen-day-old H. marinum and 11-d-old wheat and amphiploids were exposed to a hypoxic pretreatment (three pots of each genotype) by flushing the nutrient solution with N2 until the O2 concentration in the solution reached c. 0.03 mol m−3; these pots were left stagnant overnight. The next day, the nutrient solution was replaced with deoxygenated stagnant nutrient solution containing 0.1% (w/v) agar. The dilute agar prevents convective movements in the solution and so impedes the re-entry of O2 and promotes the accumulation of ethylene (Wiengweera et al., 1997). The remaining three pots of each genotype also had the nutrient solution (without agar) refreshed and continued to be well aerated.
Growth measurements Harvests were taken at the time treatments were imposed (initial) and after 21 d of treatments. Plants were divided into leaves, stems, seminal roots and adventitious roots. The numbers of stems (main stem plus tillers) and of adventitious roots were recorded. The relative growth rate (RGR) was calculated using the formula:
[dry mass1, dry mass (g) at time 1 (initial harvest); dry mass2, dry mass (g) at time 2 (final harvest); t1 and t2, time 1 and time 2 in days; loge, natural logarithm].
Aerenchyma in adventitious roots The percentage of aerenchyma was measured in transverse sections taken at 50 and 100 mm behind the tip of adventitious roots. Sections of fresh roots were taken using a hand-held razor blade and examined under a light microscope equipped with a digital camera (Nikon coolpix45000, Osaka, Japan). The percentage of aerenchyma (expressed per unit cross-sectional area) and stele size were analyzed in each digital picture using an image analysis program (ImageJ, Version 1.24o, Millersville, PA, USA).
Solution culture Expt 2
Experimental design and plant culture This experiment was conducted in order to measure the root porosity and profiles of ROL along adventitious roots in an O2-free medium. Plants were grown in aerated or stagnant treatments before the measurements. Four H. marinum accessions (H21, H87, H109 and H155), the four respective amphiploids with CS wheat (i.e. H21–CS, H87–CS, H109–CS and H155–CS) and CS wheat were evaluated. Plants were grown using the same procedures and in the same 20 : 15°C (day : night) phytotron (during April–June in a second year) as described above for solution culture Expt 1. The experimental design involved growing plants in two batches, staggered with time, owing to the time-consuming nature of ROL measurements. The first batch contained all four H. marinum parents, and the second contained the four amphiploids and CS wheat. Plants were exposed to aerated or stagnant treatments for 21–25 d before ROL measurements (described below). Pots within batches were completely randomized. Thus, the overall data set consisted of two treatments × nine genotypes × three replicates.
Adventitious root porosity The porosity (% gas volume per unit root volume) of adventitious roots was measured using the method described by Raskin (1983) and the equations as modified by Thomson et al. (1990). Measurements were taken for adventitious roots of 100–150 mm in length.
ROL from adventitious roots ROL was measured from intact adventitious roots that had not formed lateral roots. Plants were exposed to aerated or stagnant treatments for 18–21 d before ROL measurements. Intact plants were transferred to a 20°C controlled-temperature room. The shoot of each plant was held just above the root–shoot junction in a rubber lid so that the roots were immersed in a clear Perspex chamber (100 mm × 100 mm × 180 mm) containing deoxygenated stagnant solution with the following composition: agar (0.1% w/v), K+, Cl− (5.0 mol m−3) and Ca2+, SO42− (0.5 mol m−3). ROL from adventitious roots was measured at various positions, using root-sleeving O2 electrodes (i.d., 2.25 mm; height, 5.0 mm) fitted with guides to keep each root near the centre of each electrode (Armstrong & Wright, 1975; Armstrong et al., 1994).
Solution culture Expt 3
This third solution culture experiment was run simultaneously with the soil pot experiment described below (April–June, in a third year), enabling the comparison of ROL from the roots of plants grown in stagnant agar solution with that of plants grown in waterlogged soil. Again, the purpose was to compare ROL patterns along adventitious roots of H. marinum (H21), CS wheat and the H21–CS amphiploid. Thus, the experimental design was three genotypes × three replicates, grown in stagnant deoxygenated 0.1% agar nutrient solution. Plants were grown in the same 20 : 15°C (day : night) phytotron as used for solution culture Expt 1, and ROL measurements (described above) were taken after 26–27 d of treatments.
Experimental design The soil growth experiment was two treatments (drained or waterlogged) × three genotypes × three replicates. Hordeum marinum (H21), CS wheat and the H21–CS amphiploid were used. In addition to the pots for the growth experiment, three additional waterlogged pots for each genotype were also raised to provide plants for ROL measurements. The experiment was conducted in a completely randomized design in the same 20 : 15°C day : night phytotron as used for solution culture Expt 3.
Plant culture and measurements Hordeum marinum was germinated 3 d before wheat and the amphiploid to obtain plants at a similar developmental stage (Haun, 1973) at the start of treatments. Seeds were surface sterilized with 0.4% commercial bleach, rinsed with tap water and placed on moist filter paper in Petri dishes in the dark at 20°C. After 2 d, two germinated seeds were sown at a depth of 10 mm in each pot (12 pots of each genotype). The pots (height, 450 mm; diameter, 105 mm) were made from polyvinyl chloride (PVC) tubes fitted with a PVC base. A hole, 5 mm in diameter, was drilled in the side, 15 mm above the bottom, and connected to an open-end hose that was clamped for waterlogged pots and left open for drained pots. The bottom 50 mm of each pot contained gravel, above which was 4 kg of soil [sandy surface horizons of a waterlogging-prone duplex soil from south-western Australia; pH (1 : 5 w/v soil : 0.01 M CaCl2), 4.6; EC (1 : 5 w/v soil : water) = 0.19 dS m−1]. Nutrients, premixed through the soil before planting, were as follows (μg g−1 soil): NH4NO3, 357; KH2PO4, 60; K2SO4, 160; CaCl2.2H2O, 178; MgSO4.7H2O, 22; ZnSO4.7H2O, 11; MnSO4.H2O, 11; CuSO4.5H2O, 6; H3BO3, 0.8; Na2MoO4.2H2O, 0.2 (Malik et al., 2001). After 7 d, plants were thinned to one per pot. Pots were watered daily with deionized (DI) water to maintain the soil at 90% of field (i.e. pot) capacity.
Treatments were imposed 14 d (H. marinum) and 11 d (wheat and H21–CS amphiploid) after germination; plants were at the 2.3–2.7 Haun stage (Haun, 1973). Waterlogging was imposed by filling the pots with DI water from the bottom. Drained controls continued to be watered daily to weight, so that the soil was at 90% field (i.e. pot) capacity. An initial harvest of three pots of each genotype was taken at the time treatments were imposed. Plants in the three extra waterlogged pots of each genotype destined for ROL measurements were carefully washed from the soil after 21–22 d of waterlogging. The roots of intact plants were transferred into O2-free medium in Perspex chambers and ROL measurements were taken in a 20°C controlled-temperature room, as described above. The final harvest of the growth experiment was taken after 42 d of treatments. Roots were washed out of the soil and the length of the longest adventitious root on each plant was recorded (waterlogged only). Shoot and roots were separated, dried at 60°C for 48 h and dry masses were determined. Three pots (one plant per pot) of each genotype × treatment combination provided three replicates.
Data were analyzed by calculation of the means, standard errors and analysis of variance (ANOVA), where appropriate, using GenStat 8.2 (Lawes Agricultural Trust, Rothamsted Experimental Station, Hertfordshire, UK).
- Top of page
- Materials and Methods
Root aeration traits and waterlogging tolerance were evaluated in four H. marinum–wheat amphiploids, and compared with those in four H. marinum accessions and the wheat parent (CS). The results demonstrated the transfer of a barrier to ROL, and improved waterlogging tolerance, from a wild relative (i.e. H marinum) into a wheat genetic background. The presence of a barrier to ROL in the basal zones of adventitious roots of the four amphiploids was evidenced by higher (1.8–6-fold) ROL at 10 mm behind the tip compared with that at the most basal positions measured (Fig. 1). In addition to ROL barrier expression, and like H. marinum (21–25% adventitious root porosity), the amphiploids also showed slightly higher adventitious root porosity (20–24%), above that in wheat (16%), when grown in stagnant agar solution (Table 4). Thus, the present study demonstrates, for the first time, the transfer of root aeration traits from a waterlogging-tolerant wild relative to wheat via wide hybridization of the two species and the production of amphiploids. The wild relative used, H. marinum, inhabits marshes (von Bothmer et al., 1995) and develops a barrier to ROL in adventitious roots (Garthwaite et al., 2003; Malik et al., 2009; present study), a feature that, combined with aerenchyma, promotes O2 movement towards the tips of roots when in waterlogged soil (Armstrong, 1979).
The expression of a barrier to ROL in each of the amphiploids was not as strong as in their H. marinum parents (Fig. 1). Barrier formation in the amphiploids was, however, stronger than that in wheat (Figs 1, 2). The four H. marinum accessions used to produce amphiploids with wheat differed in barrier strength when grown in stagnant solution (Fig. 1); accessions H21 and H109 had stronger barriers than H87 and H155 (ROL values in basal positions were 1 and 18 nmol m−2 s−1 in H21 and H109, compared with 67 and 88 nmol m−2 s−1 in H87 and H155). Moreover, barrier formation occurred closer to the root tip (20–30 mm) in H21, H87 and H109 than in H155 (60 mm behind the root tip). The variation in the strength of the barrier to ROL in the various amphiploids did not, however, follow the differences in barrier strength amongst the H. marinum parents. Therefore, the genome of each H. marinum accession must interact with that of wheat to determine root ROL barrier expression in the various amphiploids.
The four amphiploids formed a ‘partial’ barrier to ROL in the basal zones of adventitious roots (Figs 1, 2), and showed higher root porosity (Table 4), but only two of the four amphiploids (H21–CS and H109–CS) maintained higher RGR, when in stagnant solution, than the wheat parent (H87–CS displayed some improvement at P < 0.1, but not at P < 0.05; experiments were restricted to three replicates owing to limited seed availability of these novel amphiploids). In addition to the root aeration traits evaluated in the present study, other traits associated with waterlogging tolerance (listed in reviews by Bailey-Serres & Voesenek, 2008; Colmer & Voesenek, 2009) might also differ amongst the H. marinum accessions, and potentially amongst the amphiploids; therefore, growth in stagnant solution is probably also influenced by traits in addition to those evaluated in the present study.
A barrier to ROL in basal regions is regarded as an important trait in roots of many wetland species (Armstrong, 1964, 1979; Visser et al., 2000; McDonald et al., 2001; Colmer, 2003a). In the comparison of H. marinum and wheat, barrier formation might be particularly important for determining O2 diffusion to the apex, as these two genotypes have similar rates of root respiration (Garthwaite, 2005; Garthwaite et al., 2008), and porosity is only moderately higher in roots of H. marinum (Table 4). The presence of a ROL barrier becomes relatively more important for longitudinal O2 diffusion in roots with ‘intermediate’ levels of porosity, compared with those of very high porosity (Armstrong, 1979). Thus, barrier induction would have contributed to the 3.7-fold higher ROL at 10 mm behind the root tip of H. marinum (H21), when compared with wheat (Fig. 1), especially when one considers that total ROL (aggregate of all positions for plants raised in stagnant solution) was similar for the two species (534 nmol m−2 s−1 for adventitious roots of H. marinum and 484 nmol m−2 s−1 for wheat). In addition to barrier formation, however, other factors might also differ between H. marinum and wheat, such as how close the aerenchyma extends to behind the apex – high resistance to O2 diffusion within the tip region in wheat roots has been suggested as an additional limitation restricting O2 supply to the apex (Wiengweera & Greenway, 2004).
When grown in waterlogged soil, H. marinum (H21) formed a barrier to ROL (Fig. 2b). For plants raised in waterlogged soil, however, although the H21–CS amphiploid showed higher ROL near the root tips than did wheat, barrier induction could not be demonstrated as measurements were restricted to the apical 50 mm of relatively short (mean length, 87 mm; Fig. 2) roots without laterals. Longer roots in soil had all formed laterals, preventing ROL measurements in basal zones. Barrier formation in longer roots (mean length, 123 mm; Fig. 2) from stagnant agar solution was only first clearly evident at 40 mm behind the tip. Even in rice, a barrier to ROL is poorly developed in shorter (c. 70 mm) compared with longer (c. 120 mm) roots (Colmer et al., 2006b). Thus, the presence of laterals on longer roots prevented the assessment of barrier formation in soil-grown plants of the amphiploids.
The constitutive porosity (i.e. when in aerated conditions) was higher in adventitious roots of the amphiploids, when compared with wheat, but the porosity was even higher in the roots of H. marinum (Table 4). Constitutive aerenchyma, leading to high root porosity, are common amongst wetland species (Justin & Armstrong, 1987; Jackson & Armstrong, 1999; Visser et al., 2000). High constitutive porosity would aid root aeration during the initial stages of waterlogging, as aerenchyma formation in adventitious roots of wheat can take up to 48 h (Malik et al., 2003).
In summary, the adventitious roots of H. marinum form a strong barrier to ROL when grown in stagnant solution, whereas, in wheat, the barrier is weak. Hordeum marinum–wheat amphiploids expressed a barrier to ROL, and slightly more porosity in adventitious roots, with two of the four amphiploids also showing improved growth in stagnant agar solution. Thus, the present study demonstrates success in transferring superior root aeration traits from a ‘wild’ relative (i.e. H. marinum) into wheat through wide hybridization and the production of amphiploids. The amphiploids, however, produce less dry mass (Table 1) and yield less grain (Islam et al., 2007) than the wheat parent; therefore, whether the amphiploids might eventually be of use for wheat breeding programs will be dependent on additional cytogenetic work (e.g. to produce chromosome addition lines, substitution lines and recombinant lines), and whether the key trait(s) of interest are expressed in these lines without also having deleterious effects on agronomic traits (the challenges in using alien genetic resources are summarized in Islam & Shepherd, 1991 and considered in the context of salinity tolerance by Colmer et al., 2006a).