Comprehensive Assessment of Drought Resistance of Proso Millet Germplasm Resources in Shanxi

Haigang WANG, Ling CHEN, Junjie WANG, Xiaoning CAO, Junli DONG, Lun WANG,Tianyu YANG, Zhijun QIAO*

1. Institute of Crop Germplasm Resources of Shanxi Academy of Agricultural Sciences/Key Laboratory of Crop Gene Resources and Germplasm Enhancement on Loess Plateau, Ministry of Agriculture/Shanxi Key Laboratory of Genetic Resources and Genetic Improvement of Minor Crops, Taiyuan 030031, China;

2. Crop Research Institute, Gansu Province Academy of Agricultural Science, Lanzhou 730070, China

Panicum miliaceum, with many common names including proso millet, common millet and white millet, is a grass species used as a crop. Proso millet is well adapted to many soil and climate conditions. It has a short growing season,and needs little water. The water requirement of proso millet is probably the lowest of any major cereal. It is an excellent crop for dryland and no-till farming. Proso millet is an annual grass whose plants reach an average height of 100 cm.Proso millet includes japonica rice and glutinous rice. Proso millet in Shanxi, as main local crop, is often planted in the hilly area of northwest Shanxi. Drought Resistance of Proso millet is identified to learn about the capacity of different species to drought, which is of great significance to the choice of drought-resistant species.

Shan Lun et al.[1]studied on the drought-resistance of proso millet and crops at different growth periods, and they found that the seedling stage is the most sensitive period for crops to grow. Wang Lun et al.[2]evaluated the drought-resistance of proso millet based on the repeated-drought approach,and found there were large differences in the drought-resistance of proso millet of different species. Jia Gengliang et al.[3]applied PEG simulation moisture to study the influence of proso millet and found that there were distinct differences in the electrolyte leakage, MDA content in the leaves, proline content and solution ofdifferent proso millet.Zhang Panpan et al.[4]sifted the drought-resistance indicators of millet,and found that the relative seedling rate,relative root length,relative coleoptile length, relative seedling length, relative seedling weight and relative root weight can be considered as indicators to determine the drought resistance of millet during the germination period. Feng Xiaoming et al.[5]studied the influences of different water treatments to the performance of seedlings and optical parameters, and found that the plant height, panicle length, stem diameter,panicle weight, grain weight per spike,straw weight per plant, node numbers of main stem, 1 000-grain weight and effective panicles declined, but the species which are drought-resistant declined less significantly to species which are less drought-resistant.Zhang Meijun et al.[6]applied PEG simulation to determine the substantial differences in the drought resistance of different millet species. Drought restrains the germination activity of proso millet. Germination drought index, relative seedling rate, relative seedling tendency, relative seedling weight and relative root weight are indicators for the drought-resistance of proso millet at the germination period.Liu Tianpeng et al.[7]applied PEG threatening treatment to identify the drought-resistance of seedlings of 56 millet species. Results prove that through comparisons of PEG processed species and the control there are distinct differences in the droughtresistance of diverse species.

Experts used to explore the drought-resistance of proso millet by studying the biochemical indicators based on moisture threatening. This paper aimed to clarify the drought resistance of proso millet applied in field production.

Materials and Methods

Test materials

The test materials include 11 bred varieties that are widely applied in Shanxi and 9 landraces with good performance in the production(Table 1).

Test design

This paper identified the drought resistance of proso millet in Gansu during the entire growth period from 2012 to 2013. The experiment includes two treatments, drought stress and normal water supply. In the drought stress approach, the millet was only watered for once during the entire growth period, while in the normal water supply, the millet was watered for three times during the seedling, earing and grouting period.We measured nine traits, including plant height, panicle length, stem diameter, panicle weight, grain weight per spike,straw weight per plant,node numbers of main stem, 1000-grain weight and effective panicles, under different treatments.

Table 2 Changes of millet under normal water supply and drought resistance stress

Data analysis

Excel 2003 and DPS8.05 are applied for data statistics analysis.

Evaluation of the drought resistance of millet in the field Function analysis The function values of tested materials are shown as(1)[8]

In the formula,μ (xi)was the subordinative function value of the ith main component.xiwas the score of a certain main component, while ximax and ximineach represented the maximum and minimum values of all tested materials.

The average weight of principal component

where, wirefers to the importance of the ithprinciple component, pirefers to the contribution rate of the ith principal component of each material.

Comprehensive drought resistance D value[9]

Results and Analysis

Changes and correlation analysis of plants under different water supply treatments

Table 2 shows the investigation of nine traits under different treatments,including plant height, panicle length,stem diameter, panicle weight, grain weight per spike, straw weight per plant, node numbers of main stem,1 000-grain weight and effective pani-cles. Each indicator under drought stress was slightly less than that of normal water supply.

Table 3 indicates the correlation analysis of nine traits under drought stress,and the result suggests positive correlation of strain height, ear length,node number and strain weight. The ear length was related to the node number and the grass weight. The stalk diameter and node number was of significant relation with ear weight and strain. The number of stalk node was of significant relation with ear weight and strain height. The relation between each trait would spoil the analysis of each feature.In order to eliminate such influence, we analyzed the drought resistance of millet based on principal analysis method.

Principal component analysis

The principal component analysis of nine traits(Table 4)suggest that the contribution rate of five principal components include 45.388%, 20.153%,10.964%, 10.435% and 6.918%, and its accumulated contribution rate was 93.858% , while other contribution rates can be neglected.Therefore, the five principal components represented 93.858%. The contribution rate of the first principal component was the largest, 77.343%. The coefficient of strain height,the number of stalk node and strain weight was larger than the coefficient of other features, which suggested that the first principal was composed of strain height,node number and strain weight. The contribution rate of the second principal component was 20.153%, and the coefficient of earring length, earring weight and particle weight was obviously larger than that of other traits. The contribution rate of the third principal component was 10.964%,and the coefficient of effective earring coefficient was obviously larger than that of other traits.The contribution rate of the forth-principal component was 10.435%, and the coefficient of 1 000-grain weight was obviously larger than that of other traits. The contribution rate of the fifth principal component was 6.918%, and the coefficient of stalk thickness coefficient was larger than that of other traits.

Table 3 The relevance of nine features

Table 4 Characteristic vector of the first five main components, principal component feature value, contribution rate and accumulation contribution rate plant height, panicle length, stem diameter, panicle weight, grain weight per spike, straw weight per plant, node numbers of main stem,1 000-grain weight and effective panicles

Comprehensive evaluation of drought resistance and selection of drought resistance indicators

According to the relevance of drought resistance D and the functional value (Fig.1) (r=0.519, P＜0.05), the output of different species can explain the intensity of drought resistance.

The five principal components and the eigenvectors are applied to calculate the comprehensive drought resistance D value of each material(Table 5). As can be seen in Table 5,the drought resistance D value of“Yellow Millet” was the largest one,which suggested that the drought resistance of such species was the strongest,while that of “Jingshu No.6”was the weakest. The average drought resistance value was 0.46,which was lower than the comprehensive drought resistance of homemade species,at 0.53.

Based on the obtained compre-hensive value of drought resistance and the regression equation of drought resistance during the entire growth period, the drought resistance indicators of the entire growth period were sifted.Taking the comprehensive value of drought resistance D value and each trait and variables,we obtained the optimal regression equation based on regression analysis:

Y=-0.508 + 0.053 9X7+0.0227X9,in which X7and 0.022 7X9each represented the earring weight and strain weight, and the direct coefficient was 0.561 3 and 0.570 3. The relevant coefficient r and the determinant coefficient R2was 0.949 and 0.901 respectively,and F was 77.086.The equation was extremely significant (P ＜0.01).Among the nine traits, except that the effective earring, 1 000 particle weight and comprehensive drought resistance were not related, other traits showed significant relevance to other forms(Table 6).

Table 5 The performance of the first five main components and the comprehensive drought resistance D value of nine traits

Table 6 Relevant coefficient of nine traits and the comprehensive drought resistance D value

Conclusions and Discussions

Comprehensive evaluation of drought-resistance of Proso millet during the entire growth

Proso millet is an essential crop in the arid and semi-arid area, and the drought resistance is one of the main features of the product. The drought resistance is the result of the comprehensive effects of many factors,and the effect of a single element is subtle[10-12]. At present, studies on the drought resistance of proso millet focus on the germination and seedling stages[4,6-7]. In order to clarify drought resistance of proso millet applied in field production in Shanxi,we identified its drought resistance during the whole growth period and measured nine traits, including plant height, panicle length, stem diameter,panicle weight, grain weight per spike,straw weight per plant, node numbers of main stem, 1 000-grain weight and effective panicles, under different treatments.Principal component analysis and stepwise regression analysis were used for comprehensive evaluation of drought resistance during the whole growth period and screening the drought index.

Selection of comprehensive drought resistance of proso millet

Drought resistance is a kind ofself-regulation of crops to unfavorable environment, which is not only related to the species types, biochemical features, but also has direct relation with the environment.Therefore,identification of drought resistance of millet was complicate. In order to sift available drought resistance indicators, this paper identified the features of crops during the entire growth period. Zhang Panpan et al.[4]sifted the drought-resistance indicators of millet, and found that the relative seedling rate, relative root length, relative coleoptile length,relative seedling length, and relative seedling weight. Zhang Meijun et al.[6]applied PEG simulation to determine the substantial differences in the drought resistance of different millet species. Drought restrains the germination activity of proso millet. Germination drought index, relative seedling rate, relative seedling tendency, relative seedling weight and relative root weight are indicators for the droughtresistance of proso millet at the germination period.Liu Tianpeng et al.[7]applied PEG threatening treatment to identify the drought-resistance of seedlings of 56 millet species.

The drought-resistance indicators vary from its identification method[13-14]and identification period[15-16].Based on principal component analysis and stepwise regression analysis, the study on Proso millet as drought resistance crops is of vital significance,which provides theoretical basis for the study on drought resistance.

[1]L SHAN (山仑), LK GUO (郭礼坤).Drought resistance of crops during seedling period and its demand for water(春播谷类作物成苗期间的抗旱性及其需水条件)[J].Acta Agronomica Sinica(作物学报),1984,10(4):257-263.

[2]L WANG(王纶),QF WANG(温琪汾),LP CAO (曹厉萍),et al.Studies on the selection of drought resistance of species and drought resistance mechanism (黍稷抗旱种质筛选及抗旱机理研究)[J].Shanxi Agricultural Science (山西农业科学),2007,35(4):31-34.

[3]JIA GL(贾根良),DAI HP(代惠萍),FENG BL (冯佰利), et al. Influences of PEG simulation threatening the features of millet seedlings (PEG 模拟干旱胁迫对糜子幼苗生理特性的影响)[J].Acta Botanica Boreali-Occidentalia Sinica(西北植物学报), 2008, 28 (10): 2073-2079.

[4]ZHANG PP(张盼盼),FENG BL(冯佰利),WANG PK (王鹏科),et al.Study on the identification and utilization of drought resistance indexes of Broomcorn Millet at Germinating stage (糜子芽期抗旱性指标鉴选与利用研究)[J]. Journal of Hebei Agricultural Sciences (河北农业科学),2010,14(11):22-27.

[5]FENG XM (冯晓敏),ZHANG YQ (张永清). Effects of water stress on seedling growth and photosynthetic characteristics in broomcorn(水分胁迫对糜子植株苗期生长和光合特性的影响)[J]. Acta Agronomica Sinica (作物学报), 2012,38(8):1513-1521.

[6]ZHANG MJ (张美俊),YANG WD (杨武德),QIAO ZJ(乔治军),et al.Resistance evaluation and response of 16 millet varieties at germination stage to drought stress (不同糜子品种萌发期对干旱胁迫的响应及抗旱性评价)[J].Acta Agrestla Sinica(草地学报),2013,21(2):302-307.

[7]LIU TP(刘天鹏),DONG KJ(董孔军),HE JH (何继红),et al.Identification and evaluation on the drought resistance of broomcorn millet bred cultivars at germinating stage(糜子育成品种芽期抗旱性鉴定与评价研究)[J]. Journal of Plant Genetic Resources (植物遗传资源学报),2014,15(4):746-752.

[8]HU BL(胡标林),YU SW(余守武),WAN Y(万勇),et al.Drought-resistance identification of Dongxiang common wild rice in the whole growth period(东乡野生稻全生育期抗旱性鉴定)[J]. Acta Agronomica Sinica (作物学报),2007,33(3):426-433.

[9]HU BL(胡标林),YANG P(扬平),WAN Y(万勇), et al. Comprehensive assessment of drought resistance of BILs population derived from Dongxiang wild rice at seedling stage and its genetic analysis(东乡野生稻BILs 群体苗期抗旱性综合评价及其遗传分析)[J]. Journal of Plant Genetic Resources(植物遗传资源学报),2013,14(2):249-256.

[10]LU SP(鲁守平),SUN Q(孙群),HONG L (洪露), et al. Drought resistance evaluation of glycyrrhiza uralensis fisch. from different regions in germination stage (不同种源地乌拉尔甘草发芽期抗旱性鉴定)[J].Journal of Plant Genetic Resources (植物遗传资源学报),2007,8(2):189-194.

[11]WU XS(武仙山),CHANG XP(昌小平),JING RL (景蕊莲).Screening indexes for drought resistance of wheat at grain-filling stage (小麦灌浆期抗旱性鉴定指标的综合评价)[J]. Journal of Teriticeae Crops (麦类作物学报),2008,28(4):626-632.

[12]LU GH (路贵和), DAI JR (戴景瑞),ZHANG SK(张书奎),et al.Drought resistance of elite maize inbred lines in different water stress conditions (不同干旱胁迫条件下我国玉米骨干自交系的抗旱性比较研究)[J].Acta Agronomica Sinica (作物学报),2005,31(10):1284-1288.

[13]ZHANG J (张健), CHI BL (池宝亮),HUANG XF (黄雪芳), et al. Primary study on the vigour index of drought resistance as the appraise index in corn germination stage(以活力抗旱指数作为玉米萌芽期抗旱性评价指标的初探)[J]. Acta Agriculturae Borealisinica(华北农学报),2007,22(1):22-25.

[14]LI GQ(李贵全),LI HF(李慧峰),ZHANG HY (张海燕),et al.Comprehensive evaluation of flowering-podding period drought resistance of soybean(大豆花荚期抗旱性的鉴定与综合评价)[J].Chinese Journal of Eco-agriculture(中国生态农业学报), 2007, 15 (6): 96-100.

[15]XIE JK(谢建坤),HU BL(胡标林),WAN Y (万勇), et al. Comparison of the drought resistance characters at seedling stage between Dongxiang common wild rice and cultivars (东乡野生稻与栽培稻苗期抗旱性的比较研究)[J].Acta Ecologica Sinica (生态学报),2010,30(6):1665-1674.

[16]WANG YH(王育红),YAO YQ(姚宇卿),ZHANG CJ (张灿军), et al. Study on drought-resistance identification methods and evaluation index of dry-land rice (早稻抗旱性鉴定方法与指标研究—IV 旱稻苗期抗旱性)[J].Agricultural Research in the Arid Areas (干旱地区农业研究),2005,23(4):134-137.