Generation mean analysis of some physiological traits in the hybrid maize cv. SC704 under different water regimes

Introduction

Various abiotic stresses such as drought, reduce the growth and yield of plants below the optimum level (Cramer et al. 2011). Maize (Zea mays L.) as one of the most important crops in the world, is drought sensitive, and its yield significantly decreases under water stress conditions (Shiferaw et al. 2011; Todaka et al. 2015). Drought causes degradation of photosynthetic pigments and thus decreases the light energy absorption on the photochemical apparatus and damages thylakoid membranes and centers of photosystem reaction (Batra et al. 2014). The chlorophylls and PSII absorb the light energy

which can be used to drive photosynthesis processes. When this energy is lost from PSII, it appears as heat or emitted chlorophyll fluorescence (Kalaji et al. 2012). Chlorophyll fluorescence has been extensively used to study the variation in the efficiency of light processing and stress tolerance in plants (Lichtenthaler et al. 2005; Stirbet et al. 2018). Relative leaf water content (RWC) has also been utilized for the evaluation of drought tolerance in crops (Dhanda and Sethi 2002). Decreasing this trait causes stomata closure and CO2 assimilation declines under drought stress (Gindaba et al. 2004).

Studies about the inheritance of chlorophyll fluorescence components are very limited in maize. Hola et al. (2010) examined the response of several photosynthetic parameters of maize plants to drought stress in five inbred lines and their F1 hybrids. They reported that the nonadditive genetic effects were often more important than the additive effects for the photosynthetic characteristics. However, the genetic changed when the plants were exposed to drought conditions. Also, it was not possible to predict the response of F1 hybrids to drought based on the photosynthetic performance of their parents. Simic et al. (2014) conducted a quantitative trait loci (QTL) analysis on nine chlorophyll fluorescence parameters in maize at four environments that differed in weather conditions. The chlorophyll fluorescence parameters were analyzed among 205 recombinant inbred lines derived from the intermated B73 × Mo17 maize population. The results revealed 10 significant QTLs for seven chlorophyll fluorescence parameters, of which five were co-localized after combining over the four environments. These results indicated the role of polygenic inheritance in governing the chlorophyll fluorescence parameters under study. They also detected one pleiotropic locus on chromosome 7, coinciding with the gene gst23 that possibly related to efficient photosynthesis in different field conditions.

Generation mean analysis is a biometrical method for estimating the genetic effects including additive, dominance and especially epistatic interactions by using means of different generations (Mather and Jinks 1982; Viana 2000). Information about the nature and magnitude of gene actions involved in the tolerance to water deficit stress can be useful for breeding drought-resistant varieties in maize. Thus, a generation mean analysis was designed to study the inheritance of physiological characteristics in maize under different water regimes.

Materials and Methods

Plant material and experiment

Seven generations including B73 (the maternal line), MO17 ( the paternal line), SC704 (F₁), F₂, BC₁, BC₂ and F₃ were grown in PVC pipes with 20 cm diameter and the height of one meter, under normal irrigation and two water regimes (55% and 75% of available water depletion) in the research station of the University of Tabriz. With consideration of a proper distance between the seeds, two maize seeds were sown per pipe to secure at least one plant appearance in each pipe. Then, thinning was done at the tillering stage and only one plant was left in a pipe. Eventually, 20 plants for each generation were evaluated. Plants were grown in similar conditions in terms of soil type, light, temperature and pipes size. Also, the experimental site was protected from the rainfall by a plastic cover throughout the cropping period.

Chlorophyll fluorescence parameters

Chlorophyll fluorescence was measured on a fully developed leaf using a hand-held chlorophyll fluorometer (Opti-Sciences, OS-30p). The leaves were adapted in dark for 15 min at first and then fluorescence was recorded at 650 μmol m^-2 s^-1. Fluorescence values recorded included: F₀: minimum fluorescence, F_m: maximum fluorescence, F_V: variable fluorescence = (F_m - F₀), F_V/F_m: maximum quantum yield of photosystem II (PS II) and F_V/F₀: maximum primary yield of the photochemistry of PS II.

RWC

The measurement of RWCwas accomplished by excising 1-cm disks from the fresh leaves. Five disks from each pipe were weighed immediately and the fresh weight (FW) was measured. Then, the leaf disks were placed in the distilled water for 24 h and the turgid weight (TW) was obtained. Finally, the samples were dried in an oven at 72 °C for 24 h and their dry weight (DW) was recorded. RWC was determined as follows (Salisbury and Ross 1998):

RWC = ((FW - DW) / (TW - DW)) × 100

Statistical and genetic analysis

Generation mean analysis was carried out using the weighted least squares method considering the inverse of the variance of the means within each generation as the weight (Mather and Jinks 1982). In this method, the overall average for each trait is shown as follows: Y = m + α [d] + β [h] + α² [i] + 2αβ [j] + β² [l]

Where, Y= the generation mean, m= F∞ metric, d= additive effects, h= dominance effects, i= additive × additive interaction, j= additive × dominance interaction, 1= dominance × dominance interaction and α, 2αβ and β² are coefficients of genetic parameters. To verify the goodness of fit of the model, a joint scaling test was carried out for each trait at each irrigation level (Cavalli 1952). All genetic parameters were tested for significance using a t-test.

Results and Discussion

The results showed that the scaling tests for the three-parameter model were not significant for F_V under the normal and mild stress conditions (55% AWD),F_V/F_m under both stress conditions and F_V/F₀ under the mild stress condition, which indicates the lack of epistasis for these traits (Table 1). The scaling tests were significant for the other studied traits suggesting the inadequacy of the additive-dominance model or the presence of epistasis type of gene effects.

Estimates of genetic effects based on the three- or six-parameter models for the studied traits under different water regimesare given in Table 2. The generation mean analysis suggested the importance of both main effects (dominance and additive effects) and non-allelic interaction for most of the studied traits. However, for almost all characters the dominance effects were higher in magnitude as compared to the additive gene effects under different water regimes. The opposite signs of [h] and [l] revealed duplicate epistasis for all traits in the normal and water deficit conditions except RWC in the severe stress condition. Negative and positive signs of the [h] parameter showed that partial dominance was towards the decreasing F₀, F_m, F_V andRWC and increasing of F_V/F_m and F_V/F₀ in the cross under consideration, respectively.

F₀: The best-fit model for this trait was the six- parameter model [m-d-h-i-j-l]. Additive × additive

Table 1. Scaling tests (±SE) for physiological traits in maize generations at different water regimes.

ns, * and **: non-significant and significant at 0.05 and 0.01 probability levels, respectively; AWD: available water depletion; F_0, F_m, F_v, F_{v ̸} F_m and F_{v ̸} F₀: minimum fluorescence, maximum fluorescence value, variable fluorescence, the maximum quantum yield of photosystem II (PSII) and maximum primary yield of the photochemistry of PSII, respectively; RWC: relative water content.

and dominance × dominance interactions had a significant role in the genetic control of F₀in all three water regimes. Asadi et al. (2015) reported a high additive gene effect for F₀ in wheat under normal conditions. Contrary, the non-additive gene effect was reported in sugarcane for F₀by Zhang et al. (2010).

F_m: In the normal and mild stress conditions the six-parameter model [m-d-h-i-j-l] and in the severe stress conditions the five-parameter model [m-d-h-j-l] were the best-fit models for F_m. Dominance and dominance × dominance effects were significant and governed the genetics of this character. Therefore, exploiting the non-additive gene action can be effective for improving F_m. The same findings were documented by Hola et al. (2010) in five inbred lines of maize. The authors revealed that epistatic and dominance gene effects were important in controlling the chlorophyll fluorescence parameters.  

F_V: The findings revealed that the three-parameter model [m-d-h] was fitted well for F_v under normal and mild stress conditions. Both additive and dominance gene effects were significant for this trait under the two aforementioned conditions. Whereas, in the severe stress condition the five-parameter model [m-d-h-j-l] was determined as the best model for this character. The dominance and dominance × dominance effects were also significant and had an important role in controlling F_v.

F_V/F_m: The five-parameter model [m-d-h-j-l] was best fitted for this trait in the normal conditions. The dominance    gene   effect   and   dominance × dominance    interaction    were    significant    and

Table 2. Estimates of genetic parameters (±SE),  for the joint scaling test for physiological traits in maize at

different water regimes using generation mean analysis.

See Table 1 for abbreviations.