1 Division of Agricultural &Environmental Sciences, University of Nottingham, Suton Bonington, Campus, Loughborough, Leicesterhsire, LE12 5RD, UK. Email John.Foulkes@nottingham.ac.uk
2 John Innes Centre, Crop Genetics Department, Colney Lane, Norwich, NR4 7UH, UK
3 ADAS Boxworth, Cambridge CB3 8NN, UK
The association of three target traits for drought resistance (early flowering, high accumulation of stem water soluble carbohydrate (WSC) reserves and high green flag-leaf area (GFLA) persistence) with yield performance under drought was analyzed utilizing a doubled-haploid (DH) population derived from a cross between Beaver and Soissons. The aims were to: (i) quantify associations between target traits and yield response to drought and (ii) search for quantitative trait loci (QTLs) for drought-resistance traits. Flowering had neutral effects on drought resistance, suggesting there may be a trade-off between water-saving behaviour in the shorter pre-flowering period with early flowering and a reduced capacity to access water associated with the production of a smaller rooting system. The value of large stem soluble carbohydrate reserves for drought resistance could not be confirmed in the UK environment. Stem WSC was positively associated with grain yield under both irrigation and drought. An effect of major significance for this trait was related to the 1R arm of chromosome 1BL/1RS. The genetic trait which showed the clearest correlation with yield performance under drought was green flag-leaf area persistence. The coincidence of QTL for senescence on chromosomes 2D and 2B under drought-stressed and optimal environments, respectively, indicate a complex genetic mechanism controlling expression of this trait involving the remobilization of resources from the source to the sink during senescence.
Genotype, water, anthesis, stay-green, synteny, breeding.
Winter wheat is the most extensive arable crop in the UK, grown on about 2 million ha per annum. Current national average yield is in the region of 8 t/ha. About 30% of the wheat area is grown on drought-prone soils (Foulkes et al. 2001) and the annual yield loss to drought, which typically occurs post-anthesis, is in the region of 15% (Austin 1978; Foulkes et al. 2002). With predicted climate change and more frequent summer droughts these losses will be exacerbated. New varieties may be one way of combating these drought effects. In the present study the potential usefulness to plant breeders of three target traits for drought-resistance: early flowering, high accumulation of stem WSC reserves and high green flag-leaf area (GFLA) persistence, were considered. When selecting for a given trait in UK wheat germplasm to improve drought resistance it is important to consider its effects on yield-forming processes in the absence of drought in case these act detrimentally. This is because drought cannot be predicted with certainty in the UK’s variable climate. Understanding the genetics of traits is important for any plant breeder, and molecular markers improve the efficiency of breeding by allowing manipulation of the genome through marker-assisted selection. So the identification of QTLs associated with target traits was sought in the present work.
The genetic control of drought-resistance traits was analyzed utilizing a DH population derived from a cross between Beaver (ppd-D1, late flowering; moderate-to-high stem WSC accumulation; poor GFLA persistence) and Soissons (Ppd-D1, early flowering; low-to-moderate stem WSC accumulation; good GFLA persistence). The aims were to: (i) quantify associations between target traits and yield response to drought and (ii) search for QTLs for drought-resistance traits.
Two experiments examined a DH mapping population derived from Beaver x Soissons located at the drought-prone site of ADAS Gleadthorpe, Nottinghamshire (loamy sand, to 35 cm over medium sand) conducted in 2000/01 and 2001/02. In 2000/01, 34 DH lines were examined; and in 2001/02 46 DH lines (i.e. an additional 12 lines) were examined. Experiments were sown as first wheat crops on 17 December 2000 and 14 October 2001 and treated with a full programme of fungicides, pesticides and herbicides to minimise effects of disease, pests and weeds. In each experiment, 220 kg/ha nitrogen fertilizer as ammonium nitrate granules was applied in a four-split programme. Applications of fungicides were given at GS31, GS39 and GS59 to keep diseases to very low levels. No plant growth regulators were applied.
In each experiment, irrigated and unirrigated treatments were randomised on main-plots and DH lines randomised on sub-plots in a split-plot design in three replicates. Sub-plot size was 6 x 1.4 m. The two irrigation treatments were: (i) fully irrigated to maintain soil moisture deficit (SMD) < 60 mm from onset of stem extension (GS31) to harvest and (ii) unirrigated. Irrigation was applied using a linear overhead irrigator. Date of flowering (GS61; Tottman 1987) was assessed in all sub-plots of each experiment. Combine grain yield was assessed in all sub-plots of each experiment in a 1.4 x 5 m area of the sub-plot, and values adjusted to 15% moisture. Percentage WSC of stems and attached leaf sheaths was assessed in ten randomly selected fertile shoots per sub-plot at GS 61+75oCd (base temperature 0oC) as described by Foulkes et al. (2002), in 2002 only. The percentage of total leaf lamina area remaining green was assessed for the flag leaf from a visual in situ assessment of all fertile shoots per sub-plot at GS61 + 35 days. For stem WSC and %GFLA assessments, measurements were undertaken on different calendar dates according to differences in reaching the specified developmental stage.
2001:- Visible onset of drought in the unirrigated treatment coincided with ear emergence, and in the irrigated treatment 155 mm of irrigation water was applied. Averaging across DH lines, drought decreased grain yield from 9.07 to 6.24 t/ha (P< 0.001). The yield loss in response to restricted water availability for individual lines ranged from 1.87 to 3.56 t/ha (P< 0.10).
2002:- Visible onset of drought coincided with flag leaf emergence, and in the irrigated treatment 170 mm of irrigation water was applied. Averaging across DH lines, drought decreased yield from 7.56 to 5.91 t/ha (P< 0.05). The yield loss in response to restricted water availability for individual lines ranged from 0.46 to 2.74 t/ha (P< 0.01).
Drought had only a minimal effect on flowering, advancing GS61 on average by 1 day in both seasons; genotypes responded similarly to drought. Averaging across irrigation treatments, flowering ranged from 14 to 26 June in 2001 and from 28 May to 6 June in 2002. Correlations of flowering date on grain yield under irrigated and unirrigated conditions are shown in Table 1. Generally there was a trend for later flowering to be associated with greater yields under both irrigation and drought. It is concluded that early flowering was not associated with drought resistance in the present study. It has been suggested that the benefit of water-saving behaviour in the pre-flowering period with early flowering may be cancelled out by the development of a smaller, less extensive root system (Foulkes et al. 2004), and the reported results tended to support this.
Table 1. Pearson’s correlation coefficient between the DH line means for flowering date (GS61) and grain DM yield (t/ha 85% DM) in the irrigated and unirrigated treatments in 2001 and 2002.
0.12 NS (df = 33)
0.29* (df =45)
0.32 NS (df = 33)
0.31* (df = 45)
Stem WSC was only measured in 2002. There was a positive linear relationship between stem WSC and grain yield under irrigated conditions (R2 = 0.19, P< 0.01; Fig. 1). Previous work has suggested that reduction in current assimilation during grain filling induces greater stem reserves mobilisation to, and ultisation by, the grain (see review by Blum (1998)), and therefore it was expected that lines accumulating most reserves under irrigated conditions would show smaller absolute yield losses under drought. However, in the present study there was no statistically significant relationship between stem reserves under irrigation and ability to maintain yield under drought (Fig. 1). This was because the positive slope of the regression of stem reserves on yield was similar under irrigation and drought (Fig. 1; statistical sig. of regression under drought: R2 = 0.36, P< 0.001). Similar effects were observed in the 2000/01 season in a different mapping population (Rialto x Spark) at the same site (Foulkes, unpublished data). For stem WSC accumulation in the irrigated treatment and unirrigated treatment in 2002, an effect of major significance was associated with the 1R arm of chr 1BL/1RS (Snape, unpublished data).
Figure 1. Linear regression of stem water soluble carbohydrate (t/ha 100% DM) at GS61+75oCd on grain DM yield (t/ha 85% DM) for 46 lines of Beaver x Soissons DH mapping population in 2002 at Gleadthorpe, in irrigated (Irr) and unirrigated (Unirr) treatments.
Drought had a large effect on %GFLA, decreasing it on average from 51 to 7% in 2001 and from 40 to 11% in 2002 (P< 0.001); genotypes responded differently to drought in both years (P< 0.001). Averaging across irrigation treatments, %GFLA ranged from 13 to 43% amongst lines in 2001 and from 6 to 45% in 2002. The regression of %GFLA on yield in the irrigated and unirrigated treatments in 2002 is shown in Fig 2. The relationship was not statistically significant under irrigation but there was a significant positive linear relationship under drought (P< 0.05). It is concluded that %GFLA is a trait with significant genetic x water availability interaction, which contributed to yield maintenance under drought stress in the present study. For %GFLA in the irrigated treatment, a QTL of major significance was found on chromosome 2B and in the unirrigated treatment on chromosome 2D (Verma et al., 2004).
Figure 2. Linear regression of percentage green flag leaf area at GS61+ 35 days on grain DM yield (t/ha 85% DM) for 46 lines of Beaver x Soissons DH mapping population in 2002 at Gleadthorpe, in irrigated (Irr) and unirrigated (Unirr) treatments.
The present study indicated that early flowering has neutral effects on drought resistance in the UK’s temperate climate. This is supported by recent findings on the effects of the Ppd-D1 allele on grain yield responses to drought (Foulkes et al. 2004). This allele confers photoperiod insensitivity, hence early flowering. There may be a trade-off between water-saving behaviour in the shorter pre-flowering period with early flowering and a reduced capacity to access water associated with the production of a smaller rooting system.
The value of larger stem soluble carbohydrate reserves for drought resistance in the UK environment could not be confirmed from the present results, as stem WSC was positively associated with grain yield under irrigation and under drought. Evidence for significant deposition of stem WSC reserves in grains in the absence of post-anthesis stress in wheat was provided by Gebbing et al. (1999) and our results bear out the potential importance of stem WSC for grain yield potential even in the absence of post-anthesis stress. The slope of the regression of stem WSC on yield, although statistically significant under drought, was no greater than that under irrigation in this investigation. Interestingly the 1BL.1RS translocation was associated with greater accumulation of reserves, consistent with previous reports of 1BL/1RS conferring greater biomasses at harvest in modern adapted UK genotypes (Foulkes et al. 2001).
The genetic trait which showed the clearest correlation with yield performance under drought was green flag-leaf area persistence. This trait showed a significant positive correlation with yield under late-season drought. The identification of QTLs for senescence on chromosomes 2D and 2B under drought-stressed and optimal environments, respectively, indicates a complex mechanism controlling expression of this trait. Importantly, chromosome group 2 of wheat are the homologues of chromosome F of sorghum and of chromosome 10 of maize (Gale and Devos 1998) where QTLs for stay green have been reported by Tuinstra et al. (1997) and Bertin and Gallais (2001), respectively. This evidence provided by comparative mapping in other species supports the present identification of candidate genes on group 2 chromosomes in wheat associated directly with stay-green effects. These QTLs may represent novel sources of drought resistance for breeders to exploit. This information will be of value to breeders for further fine-mapping to identify markers tightly linked to genes controlling the trait, and for gene discovery.
Austin, R.B., 1978. Actual and potential yields of wheat in the UK. ADAS Quarterly Review, 29, 277-294.
Bertin, P. and Gallais, A., 2001. Genetic variation for nitrogen use efficiency in a set of recombinant inbred lines. II - QTL detection and coincidences. Maydic , 46, 53-68.
Foulkes, M.J., Sylvester-Bradley, R. and Scott, R.K., 2001. The ability of wheat cultivars to withstand UK drought: resource capture. Journal of Agricultural Science, Cambridge, 137, 1-16.
Foulkes, M.J., Sylvester-Bradley, R. and Scott, R.K, 2002. The ability of wheat cultivars to withstand UK drought: formation of grain yield. Journal of Agricultural Science, Cambridge, 138, 153-169.
Foulkes, M.J., Sylvester-Bradley, R., Worland, A and Snape, J., 2004. Effects of a photoperiod sensitivity gene Ppd-D1 on yield potential and drought resistance in UK winter wheat. Euphytica, 135, 63-74.
Gale, M.D. and Devos, K.M. 1998. Plant Comparative Genetics after 10 Years. Science, 656-658.
Gebbing, T., H. Schnyder and W. Kubach. 1999. The utlization of pre-anthesis reserves in grain filling in wheat. Assessment by steady-state 13CO2/12CO2 labelling. Plant Cell and Environment, 22, 851-858.
Totman, D.R. 1987. The decimal code for growth stages of cereals. Annals of Applied Biology, 110, 441-454.
Tuinstra, M.R., Grote, E.M., Goldsgrough, R.B. and Ejeta, G., 1997. Genetic analysis of drought tolerance and components of grain development in Sorghum bicolor (L.). Molecular Breeding, 3, 439-448.
Verma, V., Foulkes, M.J., Caligari, P., Sylvester-Bradley, R. and Snape, J. 2004. Mapping QTLs for flag leaf senescence as a yield determinant in winter wheat under optimal and droughted environments. Euphytica, 135, 255-263.