Theor Appl Genet DOI 10.1007/s00122-016-2828-6
ORIGINAL ARTICLE
Characterization of a common wheat (Triticum aestivum L.) high‑tillering dwarf mutant Tao Xu1 · Nengfei Bian1 · Mingxing Wen1 · Jin Xiao1 · Chunxia Yuan1 · Aizhong Cao1 · Shouzhong Zhang1 · Xiue Wang1 · Haiyan Wang1
Received: 4 July 2016 / Accepted: 12 November 2016 © Springer-Verlag Berlin Heidelberg 2016
Abstract Key message A novel high-tillering dwarf mutant in common wheat Wangshuibai was characterized and mapped to facilitate breeding for plant height and tiller and the future cloning of the causal gene. Abstract Tiller number and plant height are two major agronomic traits in cereal crops affecting plant architecture and grain yield. NAUH167, a mutant of common wheat landrace Wangshuibai induced by ethylmethyl sulfide (EMS) treatment, exhibits higher tiller number and reduced plant height. Microscope observation showed that the dwarf phenotype was attributed to the decrease in the number of cells and their length. The same as the wild type, the mutant was sensitive to exogenous gibberellins. Genetic analysis showed that the high-tillering number and dwarf phenotype were related and controlled by a partial recessive gene. Using a RIL2:6 population derived from the cross NAUH167/Sumai3, a molecular marker-based genetic map was constructed. The map consisted of 283 loci, spanning a
Communicated by J. Dubcovsky. Electronic supplementary material The online version of this article (doi:10.1007/s00122-016-2828-6) contains supplementary material, which is available to authorized users. * Xiue Wang
[email protected] * Haiyan Wang
[email protected] 1
State Key Laboratory of Crop Genetics and Germplasm Enhancement, Cytogenetics Institute, Nanjing Agricultural University/JCIC-MCP, Nanjing 210095, Jiangsu, China
total length of 1007.98 cM with an average markers interval of 3.56 cM. By composite interval mapping, a stable major QTL designated QHt.nau-2D controlling both traits, was mapped to the short arm of chromosome 2D flanked by markers Xcfd11 and Xgpw361. To further map the QHt. nau-2D loci, another population consisted of 180 F2 progeny from a cross 2011I-78/NAUH167 was constructed. Finally, QHt.nau-2D was located within a genetic region of 0.8 cM between markers QHT239 and QHT187 covering a predicted physical distance of 6.77 Mb. This research laid the foundation for map-based cloning of QHt.nau-2D and would facilitate the characterization of plant height and tiller number in wheat.
Introduction Bread or common wheat (Triticum aestivum L.) is a major staple food crop worldwide. Improvement of yield potential of wheat is of great significance to ensure sufficient of grain production. To meet the need of future food supplies brought by increasing population and reduction of farmland, new strategy must be undertaken. Since harvest index has been almost maximized, many wheat researchers suggest that a net increase in biomass would be necessary to sustain an increase in grain yield (White and Wilson 2006). A new plant architecture characterized by an increase of grain numbers per panicle and a reduction of unproductive tillers has been proposed (Khush 2001). Tillering capacity is an important trait of plant architecture for grain yields because tiller number per plant determines spike number and affects grain production directly (Naruoka et al. 2011). Several lines of evidences indicate that there is a highly negative correlation between tiller number and plant height in rice and wheat (Lanning et al. 2012; Lu et al. 2015;
13
McIntosh et al. 2013; Yan et al. 1998; Zhang et al. 2011). To obtain ideal plant architecture, understanding the regulatory mechanisms of these two important traits is an urgent challenge. The characterization of high-tillering, dwarf mutants with massive tillers and short culms may provide valuable information. To date, more than 20 reduced height (Rht) genes have been identified in wheat (McIntosh et al. 2013). However, the genes most widely used in wheat breeding are Rht-B1b, Rht-D1b, Rht-D1c and Rht8 (Korzun et al. 1998; Li et al. 2012; Peng et al. 1999; Worland et al. 1998). Rht-B1b, RhtD1b can increase yield by 6.1 and 14.1%, respectively; Rht-B1b produces significantly more effective tillers per plant, leading to a higher yield per plant (Sial et al. 2002). Under Mediterranean climatic conditions, Rht8 results in a slight increase in spikelet fertility, which contributes to increased grain production (Rebetzke and Richards 2000). But in other environments, Rht8 has negative effects (Kowalski et al. 2016). In addition, Rht-D1c (Rht10) has been successfully used in wheat breeding (Duan et al. 2012). Nonetheless, the genetic basis of dwarf gene resources is still quite narrow and it is necessary to explore the alternative of dwarfing genes for wheat breeding. Tiller number is inherited quantitatively in most cases and is affected by soil fertility and environmental factors, especially day length and temperature. In wheat, although four single genes (tin1, tin2, tin3, ftin) responsible for tiller inhibition were mapped on chromosome arm 1AS (Spielmeyer and Richards 2004; Zhang et al. 2013), 2A (Peng et al. 1998) and 3A (Kuraparthy et al. 2007), respectively, most of the underlying variation for tillering was found to be controlled by QTL. To date, wheat tillering QTL have been identified in most wheat chromosomes (Cui et al. 2011; Huang et al. 2003; Kato et al. 2000; Kuraparthy et al. 2007; Law 1967; Li et al. 2002; Spielmeyer and Richards 2004; Wang et al. 2016; Xie et al. 2006; Yan et al. 2011; Zhou et al. 2006). Identification and molecular characterization of genes involved in tillering is an essential prerequisite to engineer tillering for cereal crop improvement. Gibberellic acid (GA) is essential for many developmental processes in plants, including plant height and tillering. Depending on their reaction to exogenous gibberellic acid, the Rht genes in wheat are normally classified into two groups: GA insensitive and GA sensitive (Zhang et al. 2006). Dwarf mutant alleles of homoeologous genes RhtB1 and Rht-D1 are GA insensitive and located on chromosome arms 4BS and 4DS in wheat, respectively, while Rht4 on 2BL, Rht5 on 3BS, Rht7 on 2AS, Rht8 on 2DS, Rht12 on 5AL, Rht23 on 5DL, and Rht9 and Rht13 on 7BS are GA-sensitive (Borner et al. 1997; Chen et al. 2013, 2015; Daoura et al. 2014; Ellis et al. 2004; Gale et al. 1985; Korzun et al. 1998).
13
Theor Appl Genet
Wangshuibai is a common wheat landrace from Jiangsu province, China. It is an important genetic resource for resistance to wheat scab (Fusarium head blight, FHB). However, the adult plants are prone to lodging due to relatively longer stems. In this study, a dwarf and high-tillering mutant (NAUH167) was obtained by treatment of Wangshuibai seeds with ethyl methanesulfonate (EMS). The objectives of this study were to characterize and map this mutant locus for high-tillering and reduced height.
Materials and methods Plant materials NAUH167, a high-tillering and dwarf mutant, was obtained by 0.35% (V/V) EMS mutagenesis of 2000 dry seeds of Wangshuibai. The M0 seeds were planted in a greenhouse. M1 generation plants were investigated for different morphological characters, and one mutant with reduced plant height and higher tillers was identified. The mutant was self-fertilized for three generations, the dwarf and high-tillering characters were found to be stably inherited. The M5 mutant was named as NAUH167. The F1 plants and F2 populations derived from reciprocal crosses between NAUH167 and its wild type Wangshuibai were established for genetic analysis, NAUH167/Wangshuibai and Wangshuibai/NAUH167 F2 populations consisted of 340 and 371 individuals, respectively. A population of 93 F2-derived recombinant inbred lines (RIL, F2:6) derived from the cross NAUH167/Sumai3 was constructed by single-seed descent and was used for marker analysis and mapping of the mutant traits (Table S1). The other population, 2011I-78/NAUH167, was composed of 180 F2 progenies and was used for further mapping (Table S2). The 2011I-78 stock is a Chinese Spring (CS) substitution line where CS chromosome 2D is substituted by chromosome 2D from Aegilops tauschii [Chinese Spring (Ae. tauschii 2D) or CS(T2D)]. The 2011I-78 stock was kindly provided by Dr. J. Dvorak (University of California, Davis). The F1 plants, F2 populations, and derived RILS families were grown in the Jiangpu Experimental Station of Nanjing Agricultural University, Nanjing, China (32°01′44.9″N 118°37′41.6″E). Evaluation of the traits The plants of NAUH167, Wangshuibai, Sumai3 and their derived F1, F2, F2:6 (designated as E1, in year 2012), subsequent F2:7 (designated as E2, in year 2013) and subsequent F2:8 (designated as E3, in year 2015) populations were grown in the length of 1.5 m and the spacing of 25 cm per row with 30 seeds. For the F2 population of 2011I-78/
Theor Appl Genet
NAUH167 (in year 2015), 30 seeds per plot were sown in a 2.2 m row and the spacing of 25 cm per row to guarantee enough space between plants. Field management followed local agronomic practices during the growing seasons. At maturity, 10 plants from the middle of each row were harvested and phenotyped for plant height, effective tiller number and total tiller number per plant. To understand the difference of developmental dynamics between Wangshuibai and NAUH167, the number of tillers per plant was recorded every 7 days from the beginning of the jointing stage, including the seedling stage, sixth leaf stage, jointing stage, booting stage, heading stage and late maturity stage in years 2011 and 2012, respectively, in the Pailou Experimental Station of Nanjing Agricultural University, Nanjing, China (32°01′05.4″N 118°51′26.1″E). To compare more traits between Wangshuibai and NAUH167, 10 plants from the middle of several rows were individually harvested for recording the agronomic characters in years 2011 and 2012, respectively. The plant height, internode length, spike length, spikelet number per spike, effective spikelet number per spike and effective tiller number per plant were measured at the late maturity stage. Length and width of the flag leaf were measured at filling stage. All phenotypic data were analyzed by t tests using SPSS 16.0 software (Norusis 2008). Preparation of tissue sections and microscopic observation Internode elongation is mainly caused by cell division in the internodal meristem, followed by cell elongation. Therefore, tissues from peduncles (1 cm above the first node) were dissected and frozen using the Leica CM 1950 Cryostat (Leica, Germany) to compare the intercellular characteristics of the mutant and its wild type. Longisection of the internode were prepared as described previously (Gaude et al. 2012), and observed using a Zeiss Axio Imager D2m microscope (Zeiss, Germany). The adaxial sides of the flag leaf were sampled at the heading stage for cell observation using a Leica DM2500 microscope (Leica, Germany). The length and width of 40 cells from an internode or a flag leaf were measured using ImageJ software (Abràmoff et al. 2004), and the mean values were calculated on the basis of three replicates. All data were analyzed by t tests using SPSS 16.0 software. Markers from public databases and developed in this study For preliminary molecular-marker analysis, a total of 1964 wheat SSR markers were used to test polymorphisms between NAUH167 and Sumai3 and to construct linkage map of the mapping population NAUH167/Sumai3
RIL2:6 (Paillard et al. 2003; Pestsova et al. 2000; Song et al. 2005; Wang et al. 2009; Xue et al. 2008). After a QTL was mapped on 2D, additional molecular markers were developed to saturate the QHt.nau-2D region by the following approaches: (1) According to the published wheat ESTs sequences obtained from GrainGenes 2.0 (http://wheat. pw.usda.gov/GG2/index.shtml), which were cytogenetically mapped to deletion bins 2DS 0.47-1.00, 1173 ESTPCR markers and 286 CAPS markers treated with four different four-base pair enzymes were designed; (2) EST sequences of the flanking markers of the major QTLs were blasted to the Ae. tauschii 2DS genomic database. Based on the sequences of the interval scaffolds, 316 ESTbased markers were designed to detect the polymorphism between 2011I-78 and NAUH167. All the above primer pairs were designed using online software Batchprimer3.0 (http://probes.pw.usda.gov/batchprimer3/) and the primers were synthesized by Shanghai Invitrogen Biotechnology Company limited. Molecular marker analysis Genomic DNA was extracted from 2 g fresh leaves at the third leaf stage according to the cetyl trimethylammonium bromide (CTAB) method (Murray and Thompson 1980). PCR amplifications for the molecular markers analysis were carried out in a 10 μL reaction mixture containing 50–200 ng genomic DNA, 0.3 mmol/L of each primer, 200 mmol/L of each dNTP, 2.5 mmol/L of MgCl2, 0.8 unit of Taq DNA polymerase, and 1 × PCR buffer. Amplifications were conducted using Senso Quest Labcycler Standard Plus (Senso, Germany) with the following program: 94 °C for 3 min, followed by 34 cycles of 94 °C for 30 s, 50–60 °C for 50 s and 72 °C for 70 s, and finally 72 °C for 10 min. PCR products were resolved in 8.0% non-denaturing polyacrylamide gels (Acr:Bis = 19:1, 29:1 or 39:1) using a constant voltage of 200 V for 45–90 min, and band patterns were visualized by silver staining (Bassam and Gresshoff 2007). Linkage map construction and QTLs mapping The linkage map was constructed using JoinMap 4.0 software and the threshold LOD was set at 3.0 for declaration of linkage (Van Ooijen 2006), and the genetic distances between linked loci were transformed from recombination frequencies using Kosambi mapping function (Kosambi 1944). Linkage groups were assigned to chromosomes based on the published genetic maps of wheat (http:// wheat.pw.usda.gov/GG2/index.shtml). The QTLs were detected using IciMapping 3.2 software with the threshold of the LOD score 2.5. The proportion of the phenotypic variance explained by each QTL was also
13
calculated. QTL nomenclature followed the Catalogue of Genes Symbols for Wheat (http://wheat.pw.usda.gov/GG2/ Triticum/wgc/2008/). GA sensitivity assays Cell elongation in the culm is often regulated by products of genes involved in the GA or BR pathways (Wang and Li 2008). The induction and qualitative analysis of alpha amylase in embryo-less half-seeds were performed to test for defects in the GA signaling transduction pathway (Lanahan and Ho 1988). To compare α-amylase activity, 50 embryo-less half-seeds of NAUH167 and Wangshuibai were used for detecting evidence of secreted α-amylase activity. The responses of NAUH167 and Wangshuibai to GA were tested at the seedling stage. Healthy and plump seeds were chosen to be immersed in water for 10 h at 26 °C and then cold-treated for 48 h at 4 °C. Subsequently, the seeds were transferred to Petri dishes layered with two filter papers that were wetted with control or an equal volume of 10 mmol/L GA every 2 days. A constant temperature (26 °C) and dark condition were maintained throughout seedling establishment. GA (Genview, USA) was firstly solubilized by absolute ethanol at 1 mmol/L, and then diluted by ddH2O to final concentration; and ddH2O with equal concentration of absolute ethanol was sprayed as the control. Seedling height and coleoptiles length of 10 plants for each replicate were measured 8 days after germination. The data of three repeats were statistically analyzed by t test using SPSS 16.0 software.
Results Comparison of the morphology of Wangshuibai and NAUH167 The plant heights of wild type Wangshuibai and its mutant NAUH167 were 143.15 and 69.79 cm in average, respectively. Their tiller numbers were 23.27 and 60.86 in average, respectively. NAUH167 showed significantly increased tillers and reduced height when compared with Wangshuibai. Both lines have five internodes on the culm, including the peduncle. The short stature of NAUH167 was associated with reduced length of every internode, with the largest reduction percentage at the peduncle (20.13% of the wild type). In addition, NAUH167 has smaller diameter of internode, shorter spike, slimmer spikelet, darker grains, shorter, narrower flag leaf and a 10-day delayed flowering period in comparison with Wangshuibai (Fig. 1; Table 1). To determine the key growth stage that affects the phenotype of NAUH167, the total tiller numbers of NAUH167
13
Theor Appl Genet
and Wangshuibai were recorded at different growth stages in different years (Fig. S1). The results showed no obvious difference in the total tiller number at the early vegetative stage. Beginning at the jointing stage, and especially after flowering stage, the two lines showed significant differences. After the jointing stage, the total tiller number of Wangshuibai reached its peak. However, NAUH167 still produced tillers throughout the jointing stage until the maturity stage. Finally, the mutant showed more tillers, dwarfism and delayed maturity. The grain filling of Wangshuibai was normal, but the mutant grew slowly, ultimately had lower seed setting rate and lighter 1000-grain weight, mainly due to high temperature at the late development stages. Comparison of stem and flag leaf cells of Wangshuibai and NAUH167 Changes in the number and shape of cells usually alter the plants morphology. To determine the relationship between the cells and mutant traits in NAUH167, the stem parenchyma cells were observed under microscopy (Fig. 2). The results showed that the length of parenchyma cells of the stem in Wangshuibai was 1.17 times longer than that of NAUH167 (P
