Leaf photosynthetic function duration during yield formation of large-spike wheat in rainfed cropping systems

Plant materials

In our study, eight new high-yield and large-spike wheat lines (2005, 2013, 2026, 2036, 2037, 2038, 2039 and 2040) were used, which had been tested in the Shaanxi provincial wheat variety trial test in 2009 and the Shaanxi provincial wheat variety regional test in 2010, and these lines were bred through many generations over many years. The winter wheat cultivar of Xi’nong 979 (Triticum aestivum cv. Xi’nong 979) has been planted across large areas of the Huang-Huai-Hai production region of winter wheat (Wang & Shangguan, 2015), the detailed characteristics of these materials were provided in Table 1 and would be analyzed in the part of discussion. The seeds were also sown in Sufang Town, Wugong City, Shaanxi Province, northwestern China (Wang, Chen & Shangguan, 2016).

Culture conditions and experiment design

The field experiment was conducted in the Institute of Soil and Water Conservation, Chinese Academy of Sciences and Ministry of Water Resources, Yangling, Shaanxi (34°16′N, 108°04′E) during the winter wheat growing seasons (October–June) in 2010–2011 and 2011–2012. The experiment was located in the sub-humid warm temperate continental monsoon climate zone and generally had flat topography. The soil of the experiment was a well aerated Eum-orthic Anthrosol soil characterized by favorable permeability, strong water and nutrient-retaining capacities, and wide crop adaptation. There were similar precipitation (231.1 and 229.8 mm) trends and other climate conditions during the two wheat growing seasons, so we chose to analyze the data from 2011–2012.

The field experiment was a randomized block design involving nine treatments, and each plot size was 2 m × 2 m with 10 rows (20 cm spacing) of wheat sown at 110 seeds/row. Surrounding the experimental field, there were five guard rows of wheat. Plants were sampled from the central rows in each plot. Each year, winter wheat was planted on 10 October and the same quantity of chemical fertilizer (360 kg hm⁻² N and 70 kg hm⁻²P₂O₅) was applied to the top 20 cm of soil before planting. During the growing period, no irrigation and fertilizer were applied; the site was plowed to bury weeds and pests before sowing, and weeds were hand-hoed several times during the growing period.

Population tiller measurement

In each plot, six rows of wheat were chosen and tagged at sowing stage, and the number of tillers were counted at seedling and tillering stages before winter. After wintering, wheat tiller number data were collected at intervals of 7 d from returning green stage to heading stage. The last survey date of the number of spikes was at mature stage.

Relative value of chlorophyll measurements for intact expanded and detached leaves

A portable chlorophyll meter (SPAD 502; Minolta Camera Co., Osaka, Japan) was used to take SPAD values from the three uppermost fully expanded leaves of leaf 5, leaf 7, and the flag leaf on each plant at 10-d intervals (Fig. 1). A total of six plants were measured per plot, and three SPAD values per leaf, including one value around the midpoint of the leaf blade and two values at 3 cm away from the midpoint were averaged as the mean SPAD value of the leaf (Peng et al., 1993).

From a total of six repeat plants, leaf 5, leaf 7, and the flag leaf were cut when fully expanded and immediately placed in Petri dishes containing distilled water, and cultured in the plant ZPW-280B (DongTuo, Inc., Heilongjiang, China) incubator in vitro. During growth, the photoperiod was 12 h (day, 100−120 µmol m⁻² s⁻¹)/12 h (night), temperature was 25 ± 2/18 ± 2 °C (day/night), and relative humidity was 60%–70% (day/night). Measurement methods and measured parts of detached leaves were the same as for intact expanded leaves.

Photosynthetic rate measurement

For in vitro P_n measurement, the leaves were separated from plants and preserved in liquid nitrogen immediately, the interval times of measure were 30 min (jointing and flowering stages) and 10 min (grain-filling stage) during 09:00−11:00 AM on clear days with a wind speed below 1 m s⁻¹.

The P_n of fully expanded top-down penultimate and flag leaves for all lines and cultivars was measured with a LI-6400XT Portable Photosynthesis System (LI-6400, Li-Cor, Inc., Lincoln, Nebraska, USA) equipped with 2 cm × 3 cm leaf chamber and integrated light source. For the measurement, leaves were intact with no sign of insect or disease attack and the upper one-third of leaves was used. While being measured, the settings were leaf chamber temperature of 25 °C, mean CO₂ concentration of 382.6 ±  2.5 µL L⁻¹, PAR generated by the LED (mixed red and blue) system of 1,300 µmol m⁻² s⁻¹, humidity of 53%–56%, and the gas flow rate of 5 mL min⁻¹. The P_n was measured for 3 min until P_n and transpiration rates stabilized; for each plot, three randomly selected and fully expanded leaves were measured.

Relative water content measurement

The RWC was measured following the methodology of Barr & Weatherley (1962). First, six disc samples were cut from the youngest fully expanded leaf at jointing, flowering, and grain-filling stages, and immediately weighed to obtain their fresh weight (FW). Second, the same discs were placed on distilled water for 24 h in Petri dishes in darkness and, after gentle blotting to remove excess water, they were weighed to obtain turgid weight (TW). Finally, discs were weighed after drying at 60 °C for 48 h until constant dry weight (DW) was reached. Using these three parameters, RWC was calculated using the following Eq. (1):

(1)${\displaystyle \mathrm{RWC}\left(\text{\%}\right)=\left(\mathrm{FW}-\mathrm{DW}\right)∕\left(\mathrm{TW}-\mathrm{DW}\right)\times 100.}$

The RWC was measured at 30-min intervals at jointing and flowering stages, and 10-min intervals at grain-filling stage.

Root activity measurement

In the field, the root system was excavated to a size of 20 cm (length) × 20 cm (width) × 30 cm (depth) at the sampling periods of jointing, flowering, and grain-filling stages. The wheat roots were washed with deionized water and excised at five cm from the root tips. Then, root activity was determined by the TTC method (Lindström & Nyström, 1987).

Photosynthetic pigment measurements

Immediately after the leaf had fully expanded, six leaves were collected from each plot, and each leaf was soaked with 80% acetone and the concentrations of carotenoids (Car), chlorophyll (Chl) a, and Chl b were determined by measuring absorbance at 663, 646, and 470 nm, respectively, using an UV-2300 spectrophotometer (Techcomp, Inc., Shanghai, China). The Chl concentrations and contents were calculated using the following formulae (Arnon, 1949):

(2)${\displaystyle {\mathrm{C}}_{\mathrm{a}}=\left(12.21\times {\mathrm{A}}_{663}\right)-\left(2.81\times {\mathrm{A}}_{646}\right)}$ (3)${\displaystyle {\mathrm{C}}_{\mathrm{b}}=\left(20.13\times {\mathrm{A}}_{646}\right)-\left(5.03\times {\mathrm{A}}_{663}\right)}$ (4)${\displaystyle {\mathrm{C}}_{\mathrm{x}•\mathrm{c}}=\left(1000\times {\mathrm{A}}_{470}-3.27\times {\mathrm{C}}_{\mathrm{a}}-104\times {\mathrm{C}}_{\mathrm{b}}\right)∕229}$ (5)${\displaystyle \mathrm{Chl contents}\left({\mathrm{mg g}}^{-1}\right)=\left(\mathrm{C}\times {\mathrm{V}}_{\mathrm{T}}\times n\right)∕\mathrm{FW}\times 1000}$ where C_a, C_b, and C_x•c are the concentrations of Chl a, Chl b, and Car, respectively; C is the Chl concentration (mg L⁻¹); V_T is the volume of extracting solution (mL); FW is the fresh weight of the leaves, and n is the dilution multiple.

Dry weight measurement

In each plot, 100 plants that had headed and flowered on the same days were chosen and tagged, and flowering date was defined as the time when 50% of the plants had flowered (Wang & Shangguan, 2015). Ten plants were collected as samples approximately every 5 d from the first day after the onset of flowering. Every plant was divided into single spike and leaf + stem + sheath parts. Then the two sample parts were used for DW measurement immediately after sampling by deactivating at 105 °C and then drying at 80 °C to a constant weight.

Yield trait measures

At maturity, 20 plants were randomly selected in each plot for measurement of spike length, number of grains per spike, kernel weight per spike, number of spikes, and 1,000-grain weight. At the same time, six rows (one m length) were selected for estimation of grain yield, excluding borders in each plot.

Statistical analysis

The data in the tables are the average value of three replicates in the form of means ± SE (standard error). The significant differences (P < 0.05) were tested by SPSS ver. 14.0 (SPSS Inc., Chicago, IL, USA), and the differences among the treatments were tested by Duncan’s multiple range test.

Dynamic changes in tiller death rate

During the tillering period of seeding–heading and mature stage, there were fewer tillers for large-spike lines than for Xi’nong 979 (Fig. 2). There were different tillering peaks for different wheat materials, the peak for lines 2005, 2039, 2040 and CK was 2 weeks after returning green, and that for line 2013 was 3 weeks after returning green. Although there were two tillering peaks during the growth periods for lines 2036 and 2037, they were 7 d and 3 weeks after returning green. At mature stage, there was twice the number of tillers for Xi’nong 979 compared with large-spike lines, showing their obvious differences in tiller death rates.

Time courses of declination of chlorophyll relative values

The dynamic changes of SPAD were measured in intact expanded and detached leaves at different leaf positions for wheat lines. For the in vivo situation, the SPAD peaks were at different positions for wheat lines, the peaks for leaf 5 were around 0 and 6 d after expansion, and the highest values of leaf 7 and the flag leaf were around 24 d. In intact expanded leaves, the degenerative processes of SPAD values for leaf 7 and flag leaf included a relative steady phase and a rapid declining phase, and these were observed in the different wheat lines. The SPAD of leaf 5 showed a uniform decline for the different lines; the average SPAD of leaf 5, leaf 7, and the flag leaf of large-spike lines was 4.29, 6.53, and 6.04 higher than CK, respectively (Figs. 3A–3C).

In the situation in vitro, SPAD values showed continuous declines, and the average values of leaf 7 and flag leaf of large-spike lines were 5.64 and 4.62 higher than CK, respectively (Figs. 3D–3F).

Time courses of RWC and P_n decline

At the jointing stage, the leaf RWC of large-spike lines and CK showed declining trends; the rate of decline of leaf RWC for lines 2037, 2038 and 2040 was 1.74%, 2.32%, and 0.10% lower than for CK after 70 min in vitro (Fig. 4A). The P_n of lines 2036, 2037, and 2038 was higher than for CK, and values were 0.65, 0.55 and 1.18 µmol CO₂ m⁻²s⁻¹ after 130 min in vitro, respectively, compared to a negative P_n of CK for the same times. The average duration times of P_n of lines 2037, 2038, and 2040 were longer than for the other lines and CK (Fig. 5A), indicating that the three lines had strong drought resistance.

At the flowering stage, the leaf RWC in all wheat lines slowly declined, and the values for lines 2013 and 2026 were higher than CK after 130 min in vitro; the decrease amplitudes of leaf RWC for lines 2005, 2013, 2026, and 2037 were smaller than for CK after 10–130 min in vitro, showing that they maintained a good water balance (Fig.  4B). The declining trend of P_n in eight lines and CK was stronger at flowering than at jointing stage; the P_n of lines 2036, 2038, 2039, and 2040 were higher than CK after 130 min in vitro; the decrease in amplitudes of P_n in lines 2005, 2013, 2026, 2039, and 2040 were smaller than CK after 10–130 min in vitro (Fig. 5B), showing that they had a long photosynthetic duration.

At the grain-filling stage, the leaf RWC of all lines was within 78–90%; the values of lines 2036, 2038, and 2040 were higher than CK after 100 min in vitro, and decreases in amplitudes of the RWC of these lines were smaller than CK after 10–100 min in vitro (Fig. 4C). The variation tendencies of P_n for wheat materials differed, except for line 2036, the values of the large-spike lines were higher than for CK after 100 min in vitro, and the decrease amplitudes of P_n in all large-spike lines were lower than CK after 10–100 min in vitro (Fig. 5C), showing that large-spike lines had strong drought resistance and good water retaining capacity.

Relationship between P_n and TTC

The pooled analysis showed a strong linear and positive correlation between P_n and TTC of large-spike lines at jointing, flowering, and grain-filling stages in this rainfed environment (P < 0.001; Fig. 6). The results indicate that TTC might be a good tool for indirect assessment of P_n.

Photosynthetic pigment contents

The differences in Chl a content between large-spike lines and CK were not significant (Table 2). At the jointing stage, the Chl a contents of lines 2036, 2037, 2039, and 2040 were higher than CK; the Chl b contents of large-spike lines were all lower than CK; and the Car contents of lines 2026, 2036, 2037, 2039, and 2040 were higher than CK. At heading stage, the Chl a and b contents of lines 2026, 2036, 2037, 2039, and 2040 were higher than CK; and the Car contents of lines 2026, 2036, 2037, and 2040 were higher than CK. At the flowering stage, the Chl a, Chl b, and Car contents of lines 2026 and 2040 were higher than CK. At the grain-filling stage, the Chl a, Chl b, and Car contents of line 2040 were higher than CK.

DW of single spikes and other parts

The DW of leaf + stem + sheath for large-spike lines and CK all increased after the flowering stage; with average maximum DM reached at 15 d after flowering being 3.57 and 2.21 g plant⁻¹, respectively, and then gradually declining. The DW of leaf + stem + sheath for large-spike lines and CK all reached their minimum values at 44 d after flowering: 2.11 and 1.21 g plant⁻¹, respectively. The DW of leaf + stem + sheath of large-spike lines and CK began to decrease at 20 d after flowering and the photosynthetic assimilation substances were gradually transported to grain, and were reduced to 59.17 and 54.78% of the maximum DW by 44 d after flowering, respectively (Fig. 7).

The DW of wheat grain showed an “S” growth pattern from flowering to maturity; the average DMW of single spikes for large-spike lines and CK all reached maximum values 44 d after flowering: 3.34 and 2.38 g plant⁻¹, respectively. The average aboveground biomass of large-spike lines at 5 d after flowering and maturity stage were 4.13 and 5.45 g plant⁻¹, respectively, which were 77.03 and 51.62% higher than for CK (Fig. 7).

Yield and primary components

The average spike length, number of grains per spike, kernel weight per spike, and 1,000-grain weight of the eight large-spike lines were significantly higher than those of CK (Xi’nong 979) (P < 0.05), and the number of spikes per hectare was significantly lower for large-spike lines than CK (Table 3). The yields of lines 2005, 2026, 2037, 2038, 2039, and 2040 were 3.00, 8.97, 8.33, 5.18, 5.24, and 7.03% higher than those of CK, respectively. These results indicated future positive effects for improving grain yield of large-spike lines by coordinating the relationship among agronomic traits.