+Bioline International Official Site (site up-dated regularly)

Identification of Quantitative Trait Loci for Anthesis-Silking Interval and Yield Components Under Drought Stress in Maize

(Key Laboratory of Crop Genetics and Breeding, Ministry of Agriculture, Institute of Crop Breeding and Cultivation, Chinese Academy of Agricultural Sciences, Beijing 100081, China)
* Author for correspondence. Tel: +86 (0)10 68918596; Fax: +86 (0)10 68975212; E-mail: .

A genetic linkage map with 89 SSR marker loci was constructed based on a maize (Zea mays L.) population consisting of 184 F₂ individuals from the cross, Huangzao 4 × Ye 107. The 184 F₃ families were evaluated in the field under well-watered and drought-stressed regimes in Shanxi Province of China. The objectives of the study were to identify genetic segments responsible for the expression of anthesis-silkinginterval(ASI), ear setting and grain yield, and to examineif the quantitative trait loci (QTLs) for ASI or yield components can be used in marker-assisted selection (MAS) to improve grain yield under drought conditions. Results showed that under well-watered and drought-stressed regimes, three and two QTLs involved in the expression of ASI were detected on chromosomes 1, 2 and 3, and 2 and 5, respectively. Under well-watered regime, two QTLs for ear setting were detected on chromosomes 3 and 6, explaining about 19.9% of the phenotypic variance, and displayed additive and partial dominant effects, respectively. Under drought-stressed condition, four QTLs for ear setting were detected on chromosomes 3, 7 and 10, which were responsible for interpreting 60.4% of the phenotypic variance, and showed dominant or partial dominant effects. Under well-watered condition, four QTLs controlling grain yield were identified on chromosomes 3, 6 and 7, while five QTLs were identified under drought stress on chromosomes 1, 2, 4 and 8. The gene action was of additive or partial dominant effects, and each QTL could explain 7.3% to 22.0% of the phenotypic variance, respectively. Under drought conditions, ASI and ear setting percentage were highly correlated with grain yield, which can be used as secondary traits for grain yield selection. Based on linked markers detected and gene action analyzed, an MAS strategy for yield improvement under drought condition couldbe established, which consists of QTLs contributing to decreased ASI and to increased ear setting and grain yield, respectively.

Key words: Zea mays ; drought stress; anthesis-silking interval; ear setting; grain yield; quantitative trait loci (QTLs)

In maize, when drought stress occurs just before and during flowering time, water stress results in increasing the length of the anthesis-silking interval (ASI), and yield losses are more severe (Grant et al, 1989). Selection for grain yield under drought condition has often been considered inefficient because of the low heritability especially as yield decreases (Blum, 1988). Under these conditions, selection for secondary traits, which are correlated with grain yield and have relatively high heritabilities, may increase selection efficiency.

ASI has been considered as a precise secondary trait, which has high heritability and is correlated with grain yield under drought stress. Selection for short ASI can contribute to a high ratio of ear setting to grain yield (Bolanos et al, 1993; 1996; Ribaut et al, 1997; Edmeades et al, 1999). However, conventional selection for ASI requires carefully managed drought conditions, which limits its use in maize breeding programs. With the development of molecular markers, it is possible to identify major quantitative trait loci (QTLs) regulating specific drought responses, and it will provide an efficient way to improve drought tolerance in maize germp lasm (Stuber et al, 1987; Zehr et al, 1992; Schon et al, 1994; Veldboom et al, 1994; 1996). Therefore, identification of genomic segments responsible for the expression of ASI and yield components, with the aim in marker assisted selection (MAS), should be very useful. To date, no QTL mapping work for drought resistance in maize and subsequent MAS has been carried out in Chinese germplasm. In this study, the objectives were to identify QTL conferring the expression of ASI and yield components, and to examine if the QTLs for ASI or yield components can be used in MAS to improve grain yield under drought conditions.

The maize (Zea mays L.) inbred lines, Huangzao 4 and Ye 107, were used as parents. Huangzao 4 yields well under drought and expresses a short ASI, while Ye 107 is low yielding under drought and has a long ASI. An F₂population derived from t he cross, Huangzao 4 × Ye 107, was used to construct the genetic linkage map.

The 184 corresponding F₃families were evaluated for drought response in Linfen City, Shanxi Province of China, in 2001, where there is less rainfall during the growing season (from May to August). Therefore, timing and intensity of water deficits can be easily managed. The experiment was arranged in a complete block design with two replicates under two water regimes, well-water (WW) and severe stress (SS). The plot consisted of one four-meter-row with 0.76 m spacing between rows. Plots were over-planted and later thinned to 17 plants for each plot. Water was applied by furrow irrigation. All the treatments received the first two irrigations (1 and 20 d after planting). After this period, irrigation was applied every 20 d to the WW regime. Drought stress was obtained at 65 d. After flowering period (about 90 d after planting), both two treatments were irrigated to encourage adequate development of the kernels that had been set. In fact, there was less than 50 mm rainfall before July 20 of 2001 when maize had reached the grain-filling stage, so the intensity of water was well managed.

1.2 Trait measurements

For all the trials, male and female flowering was measured on an individual plant basis. ASI was calculated as the difference of family means between the male and female flowering. The number of ears was recorded per plot and included all ears having five kernels or more. Harvested ears were air-dried until the moisture of all the samples reached 12%. Grain yield was expressed at 12% of moisture. The mean value of each trait was used for QTL analysis.

1.3 SSR analysis

Genomic DNA was extracted by using a modified CTAB procedure (Saghai-Maroof et al, 1984). The PCR reactions were performed using a PTC-200 Thermal Cycler (MJ Research, Watertown, MA). PCR reaction volume was 20 µL containing 10 mmol/L Tris-HCl, 50 mmol/L KCl, 0.001% gelatin, 1 U Taq DNA enzyme, 2.5 mmol/L MgCl₂, 0.16 mmol/L each of 4 dNTP, 10% glycerol, 0.3 µmol/L SSR primer, and 50 ng DNA template. The reaction mix was overlaid with 28 µL of mineral oil. After amplification, 3-4 µL 5 × SGB was added to each tube. The amplification products were separated by electrophoresis in a Model 16 cm × 20 cm × 0.1 cm ATTO AE-6220 vertical gel system using 1 × TBE on a 12% undenatured polyacrylamide gel with 28 lanes. The gels ran at 250 consistent voltage for 2.5 or 3 h. After electrophoresis, the gels were silver-stainedby rinsing with 10% acetic acid for 30 min, quick rinsing three times in water, staining with 0.1% silver nitrate for 30 min, rinsing briefly with water, developing with 2.5% Na₂CO₃, and the reaction was stopped by rising the gels briefly with 3% Na₂-EDTA (or 10% acetic acid), respectively.

1.4 Data analysis

In the F₂ population, the fragments amplified from Huangzao 4 and Ye 107 by SSR primers were scored as 2 and 0, respectively, and heterozygosity or missing fragments were scored as “1” and “-”, respectively. Segregation at each marker locus was checked by the chi-square goodness of fit test for the expected Mendelian s egregation ratio. Estimates of the proportion of parental genome for each individual were obtained according to Paterson et al (1988). Linkage analysis of SSR markers was conducted by multipoint analysis using the computer program Mapmaker version 3.0 (LOD = 3.0, r = 0.4). Mapping QTLs and estimation of their effects were performed using composite interval mapping (Zeng, 1994). An LOD threshold of 2.06 was chosen for declaring a putative QTL significance.A QTL position was determined at the local maximum of the LOD plot curve in the region under consideration. The proportion of phenotypic variance explained by a single QTL was calculated as the square of the partial correlation coefficient. Each QTL was represented by a 20-cM interval with the local LOD maximum as center.

2 RESULTS

2.1 SSR linkage map

The 184 F₂ individuals of Huangzao 4 × Ye 107 were genotyped by 102 polymorphic SSR marker loci between two parental lines. Allele frequencies did not deviate significantly from the segregation ratio (1:1) at any marker locus. At 12 of the 102 marker loci, the observed genotype frequencies deviated significantly (P < 0.01) from the expectation ratio (1:2:1). The proportion of the Huangzao 4 and Ye 107 genome among the 184 F₃ families was 50.9% and 49.1%, respectively, which followed a normal segregation distribution (1:1).

The genetic linkage maps were constructed, which contained 89 SSR marker loci, and covered 1 543 cM on 10 chromosomes with an average interval length of 17.3 cM (Fig.1). The arraying order of marker loci in the linkage map was in good agreement with that in SSR bin map (Maize DB 2002), except for three loci, bnlg 292, umc1366 and bnlg1621, which mapped to different positions.

2.2 Phenotype analysis of ASI, ear setting and grain yield

The ASI was longer, while ear setting and grain yield were significantly lower, under the drought stress regime, in comparison with the well-watered regime (Table 1), indicating that long ASI and low ear setting contributed to yield losses under drought condition. Large phenotypic variation was detected for ASI, ear setting and grain yield among the 184 F₃ families. Both ASI and ear setting were found to be normally distributed, while grain yield was not. Nevertheless, all the three traits could be adopted for QTL analysis.

Both ASI and ear setting were significantly correlated with grain yield under drought-stressed condition (P<0.05, P<0.01) (Table 2), indicating that both traits could be used as secondary traits for germplasm improvement for drought tolerance.

2.3 QTL mapping for ASI

Three QTLs conferring expression of ASI were identified on chromosomes 1, 2, and 3 under well-watered regime, linked with markers umc1160, umc1555 and umc1970, respectively (Table 3), which showed over-dominant or additive gene effects, and could totally explain 30.3% of the phenotypic variance. Two QTLs for ASI were detected on chromosomes 2 and 5 linked with markers bnlg1017 and umc1822 under stressed regime, which could account for 6.4% and 7.2% of the phenotypic variance, respectively, and both displayed dominant gene effects.

2.4 QTL mapping for ear setting

Two QTLs were found with significant additive and partial dominant gene effects for ear setting on chromosomes 3 and 6 under well-watered conditions (Table 4), linked with phi099 and umc1143, respectively, which could explain total 19.9% of the phenotypic variation. Four QTLs on chromosomes 3, 7 and 10 were detected under the stressed regime. Two QTLs were flanked by phi104127 and bnlg197 on chromosome 3, which displayed partial dominant gene effects. And two QTLs on chromosomes 7 and 10 were found, linked with markers umc1066 and bnlg594, respectively.

2.5 QTL mapping for grain yield

Four QTLs conferring expression of grain yield were detected on chromosomes 3, 6 and 7 under well-watered conditions (Table 5). Two QTLs linked with phi099 and bnlg197 were located on chromosome 3, which displayed additive and dominant gene effects, respectively, and could explain 18.4% of the phenotype variance. Two QTLs were found on chromosomes 6 and 7, linked with umc1143 and phi114, which showed partial and additive gene effects and explained 26.6% of the phenotypic variation. Five QTLs were identified on chromosomes 1, 2, 4 and 8 under drought conditions, which together explained 71.5% of the phenotypic variance.

3 DISCUSSION

The expression of quantitative traits is significantly affected by the environment, and the objectives of using inheritance information for QTL were questioned by the geneticists and breeders. Stuber et al (1992) reported that the interaction was not significant between the environment and QTL for yield components of Mo17 × B73 over six locations. Veldboom et al (1996) found that about 50% of QTLs for morphological traits and yield components could be consistently detected in two different environments. However, under drought-stressed and well-watered conditions, the current study only detected one QTL on chromosome 2 controlling the expression of ASI, and three QTLs for ear setting on chromosome 3, while the mapping position of the putative QTLs on the chromosome was different. None of the QTL for grain yield was found on the same chromosome under the two water regimes. The authors consider that the types and gene action of QTLs involved in the expression of ASI, ear setting and grain yield might vary under different water regimes.

It has been considered that ASI has high heritability and is correlated with yield, therefore, selection for ASI can indirectly and significantly contribute to an increase in grain yield under drought conditions (Bolanos et al, 1996; Ribaut et al, 1997; Edmeades et al, 1999). The result of this study also confirmed that ASI is negatively correlated, while ear setting is positively correlated with grain yield under both irrigated and drought-stressed conditions. Huangzao 4 and Ye 107 have been tested to be drought tolerant and sensitive inbred lines based on three-year field evaluation, respectively. Even though many differences occurred on the mapping position and gene actions of QTL for ASI, ear setting and grain yield, some QTLs can also be consistently detected over other traits and environments. The QTL mapping for grain yield showed that Huangzao 4 could contribute some alleles increasing yield, but also contribute some QTL alleles for decreasing yield, and similar cases can also be found in the QTL mapping for ASI and ear setting. Therefore, it is reasonable to establish an MAS approach for yield improvement under drought condition consisting of QTLs contributing to the increased ear setting and grain yield and to the decreased ASI. The alleles linked with markers umc1128 and bnlg490 on chromosomes 1 and 4 contributed to increased grain yield, which displayed additive and partial dominant gene effect, and could explain 11.5% and 14.4% of the phenotypic variance, respectively. The alleles linked with bnlg1017 contributed to decreasing ASI, displaying dominance and explaining 6.4% of the phenotypic variance. The alleles linked with phi104127, umc1066 and bnlg594 on chromosomes 3, 7 and 10, contributed to increased ear s etting, which showed dominant and additive gene effects, and explained 49.7% of the phenotypic variation. Huangzao 4 contributed all the above alleles. Based on linked markers detected and gene action analyzed, the MAS strategy for yield imp rovement under drought condition might be established, which uses the above QTLs for decreasing ASI, and increasing ear setting and grain yield.

The authors greatly acknowledged Prof. GAO Gen-Lai and Mr. WANG Xue-Dong for facilitating the drought evaluation experiment.

The following images related to this document are available:

Photo images