论文部分内容阅读
Abstract Waterlogging is one of the most factors limiting wheat production in the middle and lower reaches of the Yangtze River Plain of China， especially in the middle and late stages of wheat. Wheat varieties ‘Jingmai102’ （JM102） and ‘Yangmai158’ （YM158） were planted to study the dynamic changes of photosynthetic characteristics in flag leaf and the influence of waterlogging at anthesis on the yield and components and dry matter accumulation and remobilization of winter wheat in above ground. The results showed that the SPAD values slightly increased at 1 day after anthesis （d）， and then kept decreasing with the increase of waterlogging time. The decrease in SPAD value was more remarkably in YM158 than that in JM102. As for the chlorophyll fluorescence parameters， the photochemical efficiency （Fv/Fm）， potential activity （Fv/Fo） of photosystem II， and electronic transmission （Fm/Fo） on photosystem II increased first and then decreased with the increase of waterlogging days after anthesis. The quantum ratio of heat dissipation （Fo/Fm） had a tendency opposite to that of Fv/Fm， and the change range of JM102 was lower than that of YM158. For the grain yield and components， waterlogging at anthesis decreased the dry weight of single stem， grain yield， 1 000kernel weight， spikelet per panicle， and harvest index， and the reduction of JM102 was smaller than that of YM158. As for the accumulation and remobilization of dry matter， the accumulation of dry matter after anthesis decreased significantly under waterlogging condition （WL）， and the reduction of JM102 was smaller than that of YM158. In conclusion， waterlogging at anthesis significantly affected the photosynthetic characteristics， yield and components in both varieties， but different varieties exhibited different tolerances to waterlogging stress and YM158 was more sensitive to water stress than JM102.
Key words Chlorophyll fluorescence parameters; Waterlogging; Grain filling rate; SPAD; Wheat yield and component; Dry matter accumulation and remobilization
Wheat （Triticum aestivum）， providing 20% of calories and proteins humans[1]， is an important agricultural crop in the middle and lower reaches of the Yangtze River Plain of China， which is prone to heavy rainfall and soil compaction and has poor drainage systems. Waterlogging is a signi cant constraint on crop production worldwide. Many agricultural soils for cultivating wheat are frequently exposed to waterlogging[2]， which affects crop yield and causes economic losses. The precipitation indices in China showed an upward trends in the past 40 years[3]. Moreover， it is expected in the near future that there will be more risk of waterlogging due to an increase of occurrence of more intense precipitations and extreme events of high rainfall around the world， as a result of climate change[4]. Waterlogging， as one of the abiotic stresses， causes a multitude of changes in metabolism of cotton， impairing crop growth and leading to a decline in yield[5]. Waterlogging also reduces oxygen supply to the roots， which induces a change in the intensity of dark respiration， an increase in the use of carbohydrates， and the synthesis of antioxidants[6-7]. Waterlogging causes a reduction in leaf area， and dry matter accumulation[8]， vegetative development， physiological traits and yield and components[9]. The combined effects of soil waterlogging and compaction by Tang[10]indicated that waterlogging at different crop stages reduces the grain yield， yield components， grain filling traits， and leaf greenness， and waterlogging has a considerably harmful effect on root and shoot growth in the tillering stage. Waterlogging at the jointing stage decreases leaf greenness， stomatal conductance， or leaf water content， ultimately causing premature the senescence of leaves， and grain yield and harvest index decrease when 1 000kernel weight is reduced instead of the number of spikes[11]. Moreover， waterlogging at elongation affects the grain yield and remobilization of dry matter and nitrogen in wheat[12]. Waterlogging at booting stage advances the grain filling period of wheat， decreases the filling rate， shortens the grouting time， and reduces the 1 000kernel weight[13].
It has been reported that the kinetic parameters of chlorophyll fluorescence are sensitive to abiotic stresses， such as waterlogging， salinity， heat， and drought stress， which significantly affect the photosynthesis and chlorophyll fluorescence of flag leaves in wheat[8， 13-16]. For example， drought stress gradually decreases the PSII electron transport[17-18]. Leaf maximum chlorophyll fluorescence （Fm） and maximum fluorescence at steady state （Fms） increase， whereas maximum quantum yield （Fv/Fm）， quantum yield of electron transport （Qp）， electron transport rate （ETR）， and nonphotochemical quenching （Qn） decrease due to the imposition of salt stress[19]. Experiments on the effects of waterlogging at different growth stages on chlorophyll fluorescence and grain filling characteristics of winter wheat flag leaf indicated that waterlogging at the tillering stage and the filling stage slightly affected the photosynthesis ability and grain filling process of the flag leaves， whereas waterlogging at the jointing stage and the booting stage showed considerable influence on the photosynthesis ability and the filling process of the flag leaves[20]. Waterlogging accelerated leaf senescence， especially under compacted conditions， which significantly decreased photosynthetic capacity， resulting in a lower maximal PSII photochemical efficiency （Fv/Fm）， apparent electron transport rate （ETR）， effective quantum yield of photosystem II （ΦPSII） and photochemical quenching （qP）. It has been reported that among the different wheat stages， anthesis was most sensitive to waterlogging， and the effects of waterlogging at this stage were nearly irreversible[10].
Currently， no consensus on the effects of waterlogging stress at anthesis on wheat yield and chlorophyll fluorescence has been achieved.Further research in this area is clearly needed. In this study， we tested wheat for waterlogging during flowering. The fluorescence induction curves of wheat were measured at 0， 4， 7 and 10 d， respectively， to determine the dynamic response curve of chlorophyll fluorescence parameters under waterlogging conditions.
Materials and Methods
Experimental conditions and plant material
The experiment was conducted at Jingzhou experimental station of Yangtze University， Jingzhou， Hubei， China （30°21′N; 112°8′E） during the wheat growing season of 2016-2017， and repeated in 2017-2018. Two winter wheat （Triticum aestivum L.）， YM158 and JM102 were grown in plastic pots （22 cm in height and 25 cm in diameter） filled with 7.5 kg of clay soil. And YM158 is less tolerant to water， which was selected by previous waterlogging experiments. JM102 is the local variety. To meet nutrient requirements of the plants， we used 3 g of nitrogen fertilization （pure N）， 1 g of which was applied when seeding as compound fertilizer， and 2 g was applied per pot at the wintering period and stem elongation stage. Five seeds were sown in each pot， and three seedlings remained after thinning at the threeleaf stage.
Waterlogging of wheat for 7 d at anthesis with pots putted into the horizontal pool with 27 cm water level， which meant keeping 2 cmwater above the pot. The index was measured at days 0， 1， 3， 5， 7， and 10 after anthesis （d）， and the 10thd was the day that waterlogging was stopped for 3 d. The dry matter sampling time is at days 0， 7， 14， 21， and 28.
Plant and climatic measurements
The flag leaf chlorophyll content was measured at 9：00 Am with a chlorophyll meter （model SPAD 502， Minolta， Japan）， and each treatment measured 15 flag leaves growing evenly. The average was determined. The chlorophyll fluorescence parameters were measured by Ultraportable modulation chlorophyll fluorometer MINIPAMII. For each treatment， three flag leaves with uniform growth and the same direction of light were measured and averaged. The chlorophyll fluorescence parameters were measured at 8 pmwhen the flag leaves exhibited a dark adaptation period for 30 min. The minimal fluorescence level （Fo） was measured by a weak red light， which was sufficiently low （< 0.1 μmol/（m2·s）） and thus did not induce any significant variable in fluorescence. The maximum fluorescence of darkadapted leaves （Fm） was measured by applying a 0.8 s saturating pulse （8 000 μmol/（m2·s））. All Fo measurements were performed with the measuring beam set to a frequency of 6 000 Hz， whereas all measurements of Fm were performed with a measuring beam automatically switching to 20 kHz during the saturating flash. Other parameters were also measured in the same range as mentioned above. Statistical analysis
Experiments were designed as randomized complete blocks， with each replicate representing a separate block. Treatment effects in the experiment were analyzed using ANOVA in SPSS software， version 14.0. Treatment means were separated by least significant difference （LSD） test at P≤0.05 unless otherwise specified.
Results and Analysis
Dynamic change of flag leaf SPAD value under waterlogging treatment
As showed in Fig. 1， the change of SPAD in control （CK） increased from booting stage to anthesis and decreased with the change of grain filling time. A significant （P≤0.05） difference in SPAD value between each period of the two varieties was observed， and the average value of SPAD in YM158 was significantly higher than that of JM102. The change of SPAD values in waterlogging treatment （WL） continued to decline with the increase of waterlogging time but increase slightly at 1 d. The decline observed in YM158 was larger than that in JM102. Significant effects （P≤0.01） in the WL on chlorophyll content of both varieties were observed at 10 d. In conclusion， waterlogging significantly decreased the SPAD values， and different wheat varieties exhibit different resistance to water stress. The resistance of JM102 was better than that of YM158.
Effects of waterlogging at anthesis on chlorophyll fluorescence parameters of wheat
The photochemical efficiency Fv/Fm is proportional to the quantum yield of PS II photochemistry and is highly correlated with the quantum yield of net photosynthesis. Fv/Fm changes only slightly in nonstressful environments and is almost free from the influence of difference species and growing environments. However， Fv/Fm markedly declines under stress， making it useful as a probe and indicator. As shown in Fig. 2 A， from 1 to 7 d， the Fv/Fm values changed in CK， and the WL showed the same trend. The change in the trend of Fv/Fm WL was greater than that in CK and that in YM158 was greater than that in JM102 under WL. From the 7thto 10thd， the Fv/Fm value in JM102 of CK increased 2.81% and decreased 4.66% in CK and WL， respectively. And the Fv/Fm values in YM158 increased 3.37% and decreased 5.24% in CK and WL， respectively.
Mingmei WEI et al. Differences in Chlorophyll Fluorescence Parameters， Yield and Its Components Between Different Genotypes of Wheat Under Waterlogging Conditions at Anthesis
The variation trend of other fluorescence parameters such as Fv/Fo， Fm/Fo and Fo/Fm are shown in Fig. 2 B， C and D， and Fv/Fo， Fm/Fo and Fo/Fm reflected the potential PS II activity， the electronic transmission on PSII and the quantum ratio of heat dissipation， respectively. And the Fv/Fo and Fm/Fo had the same tendency as the photochemical efficiency （Fv/Fm）， but the Fo/Fm had a tendency opposite to Fv/Fm. Furthermore， from 7 to 10 d， the Fo/Fm value of JM102 in CK decreased 17.80%， while the WL of which increased 16.29%， and the Fo/Fm value in CK and WL of YM158 decreased and increased by 19.32% and 19.31%， respectively. The Fv/Fo and Fm/Fo values of JM102 increased by 15.64% and 13.52% in CK but decreased by 29.70%and 22.99% in WL， respectively. The Fv/Fo and Fm/Fovalues of YM158 increased by 17.48% and 14.88% in CK but decreased in WL by 30.52% and 23.94%， respectively. In summary， waterlogging at anthesis poorly affected the photosynthesis of wheat， and the degree of influence between the two varieties was different. YM158 was more sensitive to water stress than JM102. Waterlogging decreased the photochemical efficiency， potential PS II activity， electronic transmission on PSII， and quantum ratio of heat dissipation， and the magnitude of the decrease in YM158 was more obvious.
Chlorophyll fluorescence is useful for quantifying the changes in the function of the photosynthetic apparatus. Intricate relationships between fluorescence kinetics and photosynthesis help our understanding of photosynthetic biophysical processes.
Effect of waterlogging on dynamic change of single stem dry matter
Waterlogging also affects the accumulation of dry matter in wheat at anthesis. The dynamic changes of dry matter in single stem under WL are shown in the Fig. 3. The dry matter of single stem increased rapidly from anthesis to mid grain filling stage， and the increase in WL and CK of JM102 and YM158 was 39.16%， 56.10%， 43.82% and 43.82%， respectively. The increase in WL was smaller than that in CK， and the decrease from CK to WL in JM102 was greater than that in YM158. From the mid grain filling stage to mature stage， the increase of dry matter under WL was not significant， but the CK increased significantly， and the increase in WL and CK treatment of JM102 and YM158 was 3.64%， 18.43%， 7.48%， and 29.88%， respectively. The effect of waterlogging from anthesis to mid grain filling stage on JM102 was greater than that on YM158， but from mid grain filling stage to mature stage， the effect of waterlogging on YM158 was greater than that on JM102. These results could indicate that the waterlogging at anthesis decreased the accumulation of dry matter significantly， and JM102 may have a recovery mechanism to waterlogging.
Effects of waterlogging on accumulation and remobilization of dry matter
Waterlogging at anthesis significantly affected dry matter accumulation after anthesis， thereby reducing the contribution to grain. As showed in Table 1， waterlogging significantly decreased the dry matter of single stem at maturity， and the decrease in JM102 was 21.1% which was less than that in YM158 that was 22.9%. And waterlogging at anthesis decreased the dry matter accumulation in vegetative at maturity， and the decrease in JM102 and YM158 was 15.9% and 13.7%， respectively. But the contribution of postanthesis dry matter remobilization to yield （CPDMA） of JM102 was 74.21% and 24.58% in CK and WL， and the CPDMA of YM158 was 77.27% and 20.74% in CK and WL， which mean that the reducing of CPDMA in JM102 was less than that of YM158. Effect of waterlogging on grain filling rate
As showed in Fig. 4， the grain filling rate of two varieties in CK showed the same trend. The changing trend of the grain filling rate was an initial increase and the maximum value at day 14 to 21， followed by a gradually decrease. The grain filling rate of YM158 was generally higher than that of JM102. The grain filling rate in WL from day 0 to 7 and from day 7 to 14 were greater than that of the CK from day 0 to 7 and from day 7 to 14. However， from day 14 to 28， the filling speed of CK was higher than that of WL. And from the date at the 7thd， the single grain weight of YM158 under WL was not only better than that of JM102 but also better than that of YM158， which underwent no treatment， which responded that the YM158 responded first to water stress and showed greater effect to waterlogging than JM102.
In conclusion， waterlogging at anthesis affected the grain filling rate， which was a result of promoting grain filling in the early stage of grain filling. However， in the late stage of grain filling， grain filling rate significantly decreased in both varieties. YM158 was more sensitive to water stress than JM102.
Effects of waterlogging on grain yield and its components
Waterlogging at anthesis showed significant effects （P<0.05） on grain yield and its components as showed in Table 2， and the damage of waterlogging on YM158 was greater than that of on JM102. The results indicated that YM158 was more sensitive to water stress. As showed in Table 2， the 1 000kernel weight （TKW） of YM158 was greater than that in JM102 under controlled conditions. However， the TKW of JM102 was greater than that of YM158 under waterlogging， and the reductions of TKW in YM158 and JM102 were 39.9% and 19.9%， respectively. The same variation tendency in single stem grain yield of the two wheat varieties was observed， the reductions of grain yield in YM158 and JM102 were 38.9% and 31.3%， respectively， which indicated that the JM102 showed greater performance than YM158 under water stress. But as for the spikelet per panicle， JM102 was greater than that of YM158 whenever at CK or WL， while the reduction in YM158 was 25.8% lower than that in JM102， which was 27.0%.
A significant correlation in the harvest index per pot at 5% probability level was reflected between the two varieties under WL. The same trend was observed as the grain yield， and the harvest index of YM158 was 0.361 greater than that of JM102， which was 0.337 under CK， but JM102 was 0.294 greater than that of YM158， which was 0.287 during WL at anthesis. In summary， waterlogging at anthesis significantly affected wheat grain filling rate， harvest index， yield and its components， and the accumulation and remobilization of dry matter （P<0.05）. When the wheat was waterlogged for 7 d at anthesis， YM158 reflected sensitively on the grain filling. The grain filling rate of YM158 increased faster than JM102， which was less sensitive under the water stress at early filling. At the later filling， JM102 performed greater than YM158， which might be because JM102 was the tolerant variety and exhibited stronger recovery ability when exposed to water stress. Waterlogging at anthesis reduced the grain yield and its components of wheat. However， the decline is different for different varieties， and different indicators of the same species also show different rates of decrease.
Discussion
Effects of waterlogging on SPAD values and chlorophyll fluorescence in flag leaves
A significant positive correlation exists between SPAD value with chlorophyll a， chlorophyll b， and total chlorophyll content in wheat leaves， and the absolute content of chlorophyll in wheat leaves can be predicted by their SPAD value[21]. This indicates that the changes in chlorophyll content can be reflected by SPAD values. As shown in Fig.1， the SPAD value reached the maximum during anthesis and then constantly decreased until maturity， indicating that the chlorophyll content reached its maximum at anthesis. Waterlogging significantly decreased the SPAD value in both genotypes， but the decline in different genotypes was different. Such decreases could be explained by the fact that waterlogging induced some metabolic disorders， thus limiting the development and the functionality of root system， such as the water and mineral uptake， from previously leached soil， especially nitrogen[22]. The chlorophyll loss was linked with variations in the photosynthetic characteristics of the leaves. A slight increase at 1 d from the WL was also observed， which might be explained by the defense mechanism in plants. Plants exhibit a capacity of "stress memory"， that is， the process of priming or hardening that involves prior exposure to a biotic or an abiotic stress factor， enhancing plant resistance to subsequent stress exposure[23]. Li et al.[20]showed that preanthesis waterlogging priming can effectively alleviate yield loss under postanthesis waterlogging in wheat. Waterlogging decreased the SPAD values， and different wheat varieties exhibited different resistance to water stress. The resistance of JM102 was better than that of YM158. Chlorophyll fluorescence is useful for quantifying the changes in the function of the photosynthetic apparatus. Intricate relationships between fluorescence kinetics and photosynthesis help our understanding of photosynthetic biophysical processes. As expected， the chlorophyll fluorescence parameters of winter wheat in PSII showed strong responses to environmental stress. In our study， the chlorophyll fluorescence parameters， such as photochemical efficiency （Fv/Fm）， potential PS II activity （Fv/Fo）， the electronic transmission on PSII （Fm/Fo）， and the quantum ratio of heat dissipation （Fo/Fm） were affected to some extent by water stress. These factors reliably produced differences in PSII photochemistry. Our results indicated that water stress reduced Fv/Fm， Fv/Fo， and Fm/Fo but finally increased Fo/Fm.
The effect of waterlogging on photosynthesis of plants is multifaceted， which not only directly causes abnormalities of photosynthetic apparatus but also affects photosynthetic electron transport[8]. Chlorophyll fluorescence kinetics plays a unique role in the light systems absorption， transmission， dissipation， and distribution of light energy during leaf photosynthesis. The maximum photochemical efficiency Fv/Fm is an important indicator to measure the degree of photoinhibition[13]， which can be used to characterize the conversion efficiency of PSII primary light energy. Under nonstressful conditions， change in Fv/Fm is small. The great reduction in Fv/Fm indicates that the plant is photoinhibited[24]， and the reaction center is irreversibly destroyed or reversibly inactivated[25]. Fv/Fo indicates the potential photochemical activity of PSII， and under adverse conditions， Fv/Fm and Fv/Fo values are significantly reduced， PSII potential photosynthesis activity is inhibited， and photosynthetic electron transport and photosynthetic membrane energization are inhibited[26]. In agreement with many previous studies， our results showed that water stress affected the chlorophyll fluorescence and limited the photosynthesis capacity.
As shown in Fig. 2， Fv/Fm， Fv/Fo， and Fm/Fo values slightly increased at the 3rdd both in CK and WL， whereas the quantum ratio of heat dissipation （Fo/Fm） slightly decreased， and the WL showed greater magnitude of change. These results indicate that shortterm water stress may improve the photosynthetic capacity of plants， and the defense mechanism in plants may play a role， which was consistent with the findings that waterlogging priming improves tolerance to waterlogging stress in wheat[27]， from the 3rdto 7thd， the Fv/Fm， Fv/Fo， and Fm/Fo values decreased in both CK and WL， the Fo/Fm increased in both CK and WL， and the WL showed a great magnitude of change. At the 7thd， the waterlogging treatment was stopped. However， given that the water was not discharged from the potting soil in time， the roots of the plants were still in the state of water accumulation. From the 7thto 10thd， the Fv/Fm， Fv/Fo， and Fm/Fo values sharply declined in WL but increased in the control. The Fo/Fm value showed a tendency that is opposite to that of Fv/Fm. From the WL， the variation of the value of chlorophyll fluorescence parameters of JM102 was smaller than that of YM158， and the Fv/Fm， Fv/Fo， and Fm/Fo values were greater than that of YM158. These results indicate that different wheat varieties show different resistance to water stress， and the resistance of JM102 was better than that of YM158. The decrease in Fv/Fm， Fv/Fo， and Fm/Fo values was more pronounced for the YM158 genotypes than the JM102 genotypes， which showed lower tolerance level for waterlogging stress. These decreases in the Fv/Fm ratio can be considered as an indicator of electron transport chain deterioration in the PSII[26， 28-29]. Thus， for the YM158 genotypes， the photochemical process was more sensitive to waterlogging stress compared to the JM102 genotypes， which could maintain relatively normal values of Fv/Fm ratio under this constraint.
A difference in temperature existed between the two treatments due to the difference in moisture in the two treatments， and the value of the chlorophyll fluorescence parameter of CK changed in such a short period of time， which might be dominated by temperature. Temperature affects chlorophyll fluorescence parameters. At low temperatures， Fv/Fm increases with temperature， but when the temperature is increased to a certain level， high temperature inhibits the chlorophyll fluorescence parameters[30-31].
Effect of waterlogging on dry matter accumulation and yield in wheat
Wheat grown in western Japan accumulated dry matter in the grains during the 14day period before maturity， which accounted for nearly 40% of the final grain weight[32]. This indicated that the dry matter accumulating during this stage was required. The results for waterlogging at anthesis showed a highly significant effect on plant development， grain yield， and components of winter wheat. Waterlogging at anthesis affected the grain filling rate. In the early stage of grain filling， waterlogging promoted grain filling， but the grain filling rate was significantly decreased in the late grain filling stage and ultimately reduced the TKW and grain yield of wheat. The grain filling period is an extremely important growth stage in the growth and development of wheat. The yield of wheat depends mainly on the grain weight after flowering， and the grain weight depends mainly on the influence of the filling rate and the duration of grain filling[33]. Waterlogging at anthesis induced leaf senescence and affected grain filling rate. Consistent with what Araki et al. reported[11]， waterlogging at jointing and after anthesis in wheat induced early leaf senescence， impaired grain filling， and ultimately reduced the grain yield. However， the effective number of grains per ear significantly decreased in WL as showed in Fig. 3 and Fig. 4， which was opposite to the reported by Hideki Araki. Filling rate is mainly controlled by genetics， whereas the duration of grain filling is greatly affected by environmental factors[34].
Grain yield of the sensitive cultivar （YM158） showed a significant reduction only when waterlogging was prolonged to only 7 dat anthesis. The results of research with waterlogging imposed at anthesis displayed great differences in wheat yield losses related to waterlogging duration. Marti et al.[35]found a linear reduction in grain yield with increasing waterlogging durations from 4 to 24 d imposed before anthesis and a loss of 50% with the longest treatment.
Conclusions
Waterlogging at anthesis greatly influenced the chlorophyll fluorescence and grain filling process of winter wheat. From the perspective of photosynthetic capacity， waterlogging during flowering can effectively reduce chlorophyll content， decrease Fv/Fm， Fv/Fo， and Fm/Fo， and increase Fv/Fo. Waterlogging at anthesis accelerated plant senescence， increased the filling rate， shortened the duration of grouting， and decreased grain weight. Waterlogging significantly reduced the grain yield of single stem， spikelets per panicle， 1 000kernel weight， dry weight of single stem， harvest index per pot. Waterlogging at anthesis reduced the photosynthetic capacity and accumulation of dry matter after anthesis in wheat， eventually reducing the CPDMA. All of the above indicators were affected by water stress， but the degree of influence of waterlogging varied between the varieties. JM102 was less sensitive to water stress than YM158.
References
[1] BRAUN HJ， ATLIN G， PAYNE T. Multilocation testing as a tool to identifyplant response to global climate change[J]. Climate Change and Crop Production， 2010， 7： 115-138.
[2] SHAW RE， MEYER WS， MCNEILL A， et al. Waterlogging in Australian agricultural landscapes： a review of plant responses and crop models[J]. Crop and Pasture Science， 2013， 64（6）： 549-562.
[3] TIAN JY， LIU J， WANG JH， et al. Trend analysis of temperature and precipitation extremes in major grain producing area of China[J]. International Journal of Climatology， 2016， 37： 672-687.
Key words Chlorophyll fluorescence parameters; Waterlogging; Grain filling rate; SPAD; Wheat yield and component; Dry matter accumulation and remobilization
Wheat （Triticum aestivum）， providing 20% of calories and proteins humans[1]， is an important agricultural crop in the middle and lower reaches of the Yangtze River Plain of China， which is prone to heavy rainfall and soil compaction and has poor drainage systems. Waterlogging is a signi cant constraint on crop production worldwide. Many agricultural soils for cultivating wheat are frequently exposed to waterlogging[2]， which affects crop yield and causes economic losses. The precipitation indices in China showed an upward trends in the past 40 years[3]. Moreover， it is expected in the near future that there will be more risk of waterlogging due to an increase of occurrence of more intense precipitations and extreme events of high rainfall around the world， as a result of climate change[4]. Waterlogging， as one of the abiotic stresses， causes a multitude of changes in metabolism of cotton， impairing crop growth and leading to a decline in yield[5]. Waterlogging also reduces oxygen supply to the roots， which induces a change in the intensity of dark respiration， an increase in the use of carbohydrates， and the synthesis of antioxidants[6-7]. Waterlogging causes a reduction in leaf area， and dry matter accumulation[8]， vegetative development， physiological traits and yield and components[9]. The combined effects of soil waterlogging and compaction by Tang[10]indicated that waterlogging at different crop stages reduces the grain yield， yield components， grain filling traits， and leaf greenness， and waterlogging has a considerably harmful effect on root and shoot growth in the tillering stage. Waterlogging at the jointing stage decreases leaf greenness， stomatal conductance， or leaf water content， ultimately causing premature the senescence of leaves， and grain yield and harvest index decrease when 1 000kernel weight is reduced instead of the number of spikes[11]. Moreover， waterlogging at elongation affects the grain yield and remobilization of dry matter and nitrogen in wheat[12]. Waterlogging at booting stage advances the grain filling period of wheat， decreases the filling rate， shortens the grouting time， and reduces the 1 000kernel weight[13].
It has been reported that the kinetic parameters of chlorophyll fluorescence are sensitive to abiotic stresses， such as waterlogging， salinity， heat， and drought stress， which significantly affect the photosynthesis and chlorophyll fluorescence of flag leaves in wheat[8， 13-16]. For example， drought stress gradually decreases the PSII electron transport[17-18]. Leaf maximum chlorophyll fluorescence （Fm） and maximum fluorescence at steady state （Fms） increase， whereas maximum quantum yield （Fv/Fm）， quantum yield of electron transport （Qp）， electron transport rate （ETR）， and nonphotochemical quenching （Qn） decrease due to the imposition of salt stress[19]. Experiments on the effects of waterlogging at different growth stages on chlorophyll fluorescence and grain filling characteristics of winter wheat flag leaf indicated that waterlogging at the tillering stage and the filling stage slightly affected the photosynthesis ability and grain filling process of the flag leaves， whereas waterlogging at the jointing stage and the booting stage showed considerable influence on the photosynthesis ability and the filling process of the flag leaves[20]. Waterlogging accelerated leaf senescence， especially under compacted conditions， which significantly decreased photosynthetic capacity， resulting in a lower maximal PSII photochemical efficiency （Fv/Fm）， apparent electron transport rate （ETR）， effective quantum yield of photosystem II （ΦPSII） and photochemical quenching （qP）. It has been reported that among the different wheat stages， anthesis was most sensitive to waterlogging， and the effects of waterlogging at this stage were nearly irreversible[10].
Currently， no consensus on the effects of waterlogging stress at anthesis on wheat yield and chlorophyll fluorescence has been achieved.Further research in this area is clearly needed. In this study， we tested wheat for waterlogging during flowering. The fluorescence induction curves of wheat were measured at 0， 4， 7 and 10 d， respectively， to determine the dynamic response curve of chlorophyll fluorescence parameters under waterlogging conditions.
Materials and Methods
Experimental conditions and plant material
The experiment was conducted at Jingzhou experimental station of Yangtze University， Jingzhou， Hubei， China （30°21′N; 112°8′E） during the wheat growing season of 2016-2017， and repeated in 2017-2018. Two winter wheat （Triticum aestivum L.）， YM158 and JM102 were grown in plastic pots （22 cm in height and 25 cm in diameter） filled with 7.5 kg of clay soil. And YM158 is less tolerant to water， which was selected by previous waterlogging experiments. JM102 is the local variety. To meet nutrient requirements of the plants， we used 3 g of nitrogen fertilization （pure N）， 1 g of which was applied when seeding as compound fertilizer， and 2 g was applied per pot at the wintering period and stem elongation stage. Five seeds were sown in each pot， and three seedlings remained after thinning at the threeleaf stage.
Waterlogging of wheat for 7 d at anthesis with pots putted into the horizontal pool with 27 cm water level， which meant keeping 2 cmwater above the pot. The index was measured at days 0， 1， 3， 5， 7， and 10 after anthesis （d）， and the 10thd was the day that waterlogging was stopped for 3 d. The dry matter sampling time is at days 0， 7， 14， 21， and 28.
Plant and climatic measurements
The flag leaf chlorophyll content was measured at 9：00 Am with a chlorophyll meter （model SPAD 502， Minolta， Japan）， and each treatment measured 15 flag leaves growing evenly. The average was determined. The chlorophyll fluorescence parameters were measured by Ultraportable modulation chlorophyll fluorometer MINIPAMII. For each treatment， three flag leaves with uniform growth and the same direction of light were measured and averaged. The chlorophyll fluorescence parameters were measured at 8 pmwhen the flag leaves exhibited a dark adaptation period for 30 min. The minimal fluorescence level （Fo） was measured by a weak red light， which was sufficiently low （< 0.1 μmol/（m2·s）） and thus did not induce any significant variable in fluorescence. The maximum fluorescence of darkadapted leaves （Fm） was measured by applying a 0.8 s saturating pulse （8 000 μmol/（m2·s））. All Fo measurements were performed with the measuring beam set to a frequency of 6 000 Hz， whereas all measurements of Fm were performed with a measuring beam automatically switching to 20 kHz during the saturating flash. Other parameters were also measured in the same range as mentioned above. Statistical analysis
Experiments were designed as randomized complete blocks， with each replicate representing a separate block. Treatment effects in the experiment were analyzed using ANOVA in SPSS software， version 14.0. Treatment means were separated by least significant difference （LSD） test at P≤0.05 unless otherwise specified.
Results and Analysis
Dynamic change of flag leaf SPAD value under waterlogging treatment
As showed in Fig. 1， the change of SPAD in control （CK） increased from booting stage to anthesis and decreased with the change of grain filling time. A significant （P≤0.05） difference in SPAD value between each period of the two varieties was observed， and the average value of SPAD in YM158 was significantly higher than that of JM102. The change of SPAD values in waterlogging treatment （WL） continued to decline with the increase of waterlogging time but increase slightly at 1 d. The decline observed in YM158 was larger than that in JM102. Significant effects （P≤0.01） in the WL on chlorophyll content of both varieties were observed at 10 d. In conclusion， waterlogging significantly decreased the SPAD values， and different wheat varieties exhibit different resistance to water stress. The resistance of JM102 was better than that of YM158.
Effects of waterlogging at anthesis on chlorophyll fluorescence parameters of wheat
The photochemical efficiency Fv/Fm is proportional to the quantum yield of PS II photochemistry and is highly correlated with the quantum yield of net photosynthesis. Fv/Fm changes only slightly in nonstressful environments and is almost free from the influence of difference species and growing environments. However， Fv/Fm markedly declines under stress， making it useful as a probe and indicator. As shown in Fig. 2 A， from 1 to 7 d， the Fv/Fm values changed in CK， and the WL showed the same trend. The change in the trend of Fv/Fm WL was greater than that in CK and that in YM158 was greater than that in JM102 under WL. From the 7thto 10thd， the Fv/Fm value in JM102 of CK increased 2.81% and decreased 4.66% in CK and WL， respectively. And the Fv/Fm values in YM158 increased 3.37% and decreased 5.24% in CK and WL， respectively.
Mingmei WEI et al. Differences in Chlorophyll Fluorescence Parameters， Yield and Its Components Between Different Genotypes of Wheat Under Waterlogging Conditions at Anthesis
The variation trend of other fluorescence parameters such as Fv/Fo， Fm/Fo and Fo/Fm are shown in Fig. 2 B， C and D， and Fv/Fo， Fm/Fo and Fo/Fm reflected the potential PS II activity， the electronic transmission on PSII and the quantum ratio of heat dissipation， respectively. And the Fv/Fo and Fm/Fo had the same tendency as the photochemical efficiency （Fv/Fm）， but the Fo/Fm had a tendency opposite to Fv/Fm. Furthermore， from 7 to 10 d， the Fo/Fm value of JM102 in CK decreased 17.80%， while the WL of which increased 16.29%， and the Fo/Fm value in CK and WL of YM158 decreased and increased by 19.32% and 19.31%， respectively. The Fv/Fo and Fm/Fo values of JM102 increased by 15.64% and 13.52% in CK but decreased by 29.70%and 22.99% in WL， respectively. The Fv/Fo and Fm/Fovalues of YM158 increased by 17.48% and 14.88% in CK but decreased in WL by 30.52% and 23.94%， respectively. In summary， waterlogging at anthesis poorly affected the photosynthesis of wheat， and the degree of influence between the two varieties was different. YM158 was more sensitive to water stress than JM102. Waterlogging decreased the photochemical efficiency， potential PS II activity， electronic transmission on PSII， and quantum ratio of heat dissipation， and the magnitude of the decrease in YM158 was more obvious.
Chlorophyll fluorescence is useful for quantifying the changes in the function of the photosynthetic apparatus. Intricate relationships between fluorescence kinetics and photosynthesis help our understanding of photosynthetic biophysical processes.
Effect of waterlogging on dynamic change of single stem dry matter
Waterlogging also affects the accumulation of dry matter in wheat at anthesis. The dynamic changes of dry matter in single stem under WL are shown in the Fig. 3. The dry matter of single stem increased rapidly from anthesis to mid grain filling stage， and the increase in WL and CK of JM102 and YM158 was 39.16%， 56.10%， 43.82% and 43.82%， respectively. The increase in WL was smaller than that in CK， and the decrease from CK to WL in JM102 was greater than that in YM158. From the mid grain filling stage to mature stage， the increase of dry matter under WL was not significant， but the CK increased significantly， and the increase in WL and CK treatment of JM102 and YM158 was 3.64%， 18.43%， 7.48%， and 29.88%， respectively. The effect of waterlogging from anthesis to mid grain filling stage on JM102 was greater than that on YM158， but from mid grain filling stage to mature stage， the effect of waterlogging on YM158 was greater than that on JM102. These results could indicate that the waterlogging at anthesis decreased the accumulation of dry matter significantly， and JM102 may have a recovery mechanism to waterlogging.
Effects of waterlogging on accumulation and remobilization of dry matter
Waterlogging at anthesis significantly affected dry matter accumulation after anthesis， thereby reducing the contribution to grain. As showed in Table 1， waterlogging significantly decreased the dry matter of single stem at maturity， and the decrease in JM102 was 21.1% which was less than that in YM158 that was 22.9%. And waterlogging at anthesis decreased the dry matter accumulation in vegetative at maturity， and the decrease in JM102 and YM158 was 15.9% and 13.7%， respectively. But the contribution of postanthesis dry matter remobilization to yield （CPDMA） of JM102 was 74.21% and 24.58% in CK and WL， and the CPDMA of YM158 was 77.27% and 20.74% in CK and WL， which mean that the reducing of CPDMA in JM102 was less than that of YM158. Effect of waterlogging on grain filling rate
As showed in Fig. 4， the grain filling rate of two varieties in CK showed the same trend. The changing trend of the grain filling rate was an initial increase and the maximum value at day 14 to 21， followed by a gradually decrease. The grain filling rate of YM158 was generally higher than that of JM102. The grain filling rate in WL from day 0 to 7 and from day 7 to 14 were greater than that of the CK from day 0 to 7 and from day 7 to 14. However， from day 14 to 28， the filling speed of CK was higher than that of WL. And from the date at the 7thd， the single grain weight of YM158 under WL was not only better than that of JM102 but also better than that of YM158， which underwent no treatment， which responded that the YM158 responded first to water stress and showed greater effect to waterlogging than JM102.
In conclusion， waterlogging at anthesis affected the grain filling rate， which was a result of promoting grain filling in the early stage of grain filling. However， in the late stage of grain filling， grain filling rate significantly decreased in both varieties. YM158 was more sensitive to water stress than JM102.
Effects of waterlogging on grain yield and its components
Waterlogging at anthesis showed significant effects （P<0.05） on grain yield and its components as showed in Table 2， and the damage of waterlogging on YM158 was greater than that of on JM102. The results indicated that YM158 was more sensitive to water stress. As showed in Table 2， the 1 000kernel weight （TKW） of YM158 was greater than that in JM102 under controlled conditions. However， the TKW of JM102 was greater than that of YM158 under waterlogging， and the reductions of TKW in YM158 and JM102 were 39.9% and 19.9%， respectively. The same variation tendency in single stem grain yield of the two wheat varieties was observed， the reductions of grain yield in YM158 and JM102 were 38.9% and 31.3%， respectively， which indicated that the JM102 showed greater performance than YM158 under water stress. But as for the spikelet per panicle， JM102 was greater than that of YM158 whenever at CK or WL， while the reduction in YM158 was 25.8% lower than that in JM102， which was 27.0%.
A significant correlation in the harvest index per pot at 5% probability level was reflected between the two varieties under WL. The same trend was observed as the grain yield， and the harvest index of YM158 was 0.361 greater than that of JM102， which was 0.337 under CK， but JM102 was 0.294 greater than that of YM158， which was 0.287 during WL at anthesis. In summary， waterlogging at anthesis significantly affected wheat grain filling rate， harvest index， yield and its components， and the accumulation and remobilization of dry matter （P<0.05）. When the wheat was waterlogged for 7 d at anthesis， YM158 reflected sensitively on the grain filling. The grain filling rate of YM158 increased faster than JM102， which was less sensitive under the water stress at early filling. At the later filling， JM102 performed greater than YM158， which might be because JM102 was the tolerant variety and exhibited stronger recovery ability when exposed to water stress. Waterlogging at anthesis reduced the grain yield and its components of wheat. However， the decline is different for different varieties， and different indicators of the same species also show different rates of decrease.
Discussion
Effects of waterlogging on SPAD values and chlorophyll fluorescence in flag leaves
A significant positive correlation exists between SPAD value with chlorophyll a， chlorophyll b， and total chlorophyll content in wheat leaves， and the absolute content of chlorophyll in wheat leaves can be predicted by their SPAD value[21]. This indicates that the changes in chlorophyll content can be reflected by SPAD values. As shown in Fig.1， the SPAD value reached the maximum during anthesis and then constantly decreased until maturity， indicating that the chlorophyll content reached its maximum at anthesis. Waterlogging significantly decreased the SPAD value in both genotypes， but the decline in different genotypes was different. Such decreases could be explained by the fact that waterlogging induced some metabolic disorders， thus limiting the development and the functionality of root system， such as the water and mineral uptake， from previously leached soil， especially nitrogen[22]. The chlorophyll loss was linked with variations in the photosynthetic characteristics of the leaves. A slight increase at 1 d from the WL was also observed， which might be explained by the defense mechanism in plants. Plants exhibit a capacity of "stress memory"， that is， the process of priming or hardening that involves prior exposure to a biotic or an abiotic stress factor， enhancing plant resistance to subsequent stress exposure[23]. Li et al.[20]showed that preanthesis waterlogging priming can effectively alleviate yield loss under postanthesis waterlogging in wheat. Waterlogging decreased the SPAD values， and different wheat varieties exhibited different resistance to water stress. The resistance of JM102 was better than that of YM158. Chlorophyll fluorescence is useful for quantifying the changes in the function of the photosynthetic apparatus. Intricate relationships between fluorescence kinetics and photosynthesis help our understanding of photosynthetic biophysical processes. As expected， the chlorophyll fluorescence parameters of winter wheat in PSII showed strong responses to environmental stress. In our study， the chlorophyll fluorescence parameters， such as photochemical efficiency （Fv/Fm）， potential PS II activity （Fv/Fo）， the electronic transmission on PSII （Fm/Fo）， and the quantum ratio of heat dissipation （Fo/Fm） were affected to some extent by water stress. These factors reliably produced differences in PSII photochemistry. Our results indicated that water stress reduced Fv/Fm， Fv/Fo， and Fm/Fo but finally increased Fo/Fm.
The effect of waterlogging on photosynthesis of plants is multifaceted， which not only directly causes abnormalities of photosynthetic apparatus but also affects photosynthetic electron transport[8]. Chlorophyll fluorescence kinetics plays a unique role in the light systems absorption， transmission， dissipation， and distribution of light energy during leaf photosynthesis. The maximum photochemical efficiency Fv/Fm is an important indicator to measure the degree of photoinhibition[13]， which can be used to characterize the conversion efficiency of PSII primary light energy. Under nonstressful conditions， change in Fv/Fm is small. The great reduction in Fv/Fm indicates that the plant is photoinhibited[24]， and the reaction center is irreversibly destroyed or reversibly inactivated[25]. Fv/Fo indicates the potential photochemical activity of PSII， and under adverse conditions， Fv/Fm and Fv/Fo values are significantly reduced， PSII potential photosynthesis activity is inhibited， and photosynthetic electron transport and photosynthetic membrane energization are inhibited[26]. In agreement with many previous studies， our results showed that water stress affected the chlorophyll fluorescence and limited the photosynthesis capacity.
As shown in Fig. 2， Fv/Fm， Fv/Fo， and Fm/Fo values slightly increased at the 3rdd both in CK and WL， whereas the quantum ratio of heat dissipation （Fo/Fm） slightly decreased， and the WL showed greater magnitude of change. These results indicate that shortterm water stress may improve the photosynthetic capacity of plants， and the defense mechanism in plants may play a role， which was consistent with the findings that waterlogging priming improves tolerance to waterlogging stress in wheat[27]， from the 3rdto 7thd， the Fv/Fm， Fv/Fo， and Fm/Fo values decreased in both CK and WL， the Fo/Fm increased in both CK and WL， and the WL showed a great magnitude of change. At the 7thd， the waterlogging treatment was stopped. However， given that the water was not discharged from the potting soil in time， the roots of the plants were still in the state of water accumulation. From the 7thto 10thd， the Fv/Fm， Fv/Fo， and Fm/Fo values sharply declined in WL but increased in the control. The Fo/Fm value showed a tendency that is opposite to that of Fv/Fm. From the WL， the variation of the value of chlorophyll fluorescence parameters of JM102 was smaller than that of YM158， and the Fv/Fm， Fv/Fo， and Fm/Fo values were greater than that of YM158. These results indicate that different wheat varieties show different resistance to water stress， and the resistance of JM102 was better than that of YM158. The decrease in Fv/Fm， Fv/Fo， and Fm/Fo values was more pronounced for the YM158 genotypes than the JM102 genotypes， which showed lower tolerance level for waterlogging stress. These decreases in the Fv/Fm ratio can be considered as an indicator of electron transport chain deterioration in the PSII[26， 28-29]. Thus， for the YM158 genotypes， the photochemical process was more sensitive to waterlogging stress compared to the JM102 genotypes， which could maintain relatively normal values of Fv/Fm ratio under this constraint.
A difference in temperature existed between the two treatments due to the difference in moisture in the two treatments， and the value of the chlorophyll fluorescence parameter of CK changed in such a short period of time， which might be dominated by temperature. Temperature affects chlorophyll fluorescence parameters. At low temperatures， Fv/Fm increases with temperature， but when the temperature is increased to a certain level， high temperature inhibits the chlorophyll fluorescence parameters[30-31].
Effect of waterlogging on dry matter accumulation and yield in wheat
Wheat grown in western Japan accumulated dry matter in the grains during the 14day period before maturity， which accounted for nearly 40% of the final grain weight[32]. This indicated that the dry matter accumulating during this stage was required. The results for waterlogging at anthesis showed a highly significant effect on plant development， grain yield， and components of winter wheat. Waterlogging at anthesis affected the grain filling rate. In the early stage of grain filling， waterlogging promoted grain filling， but the grain filling rate was significantly decreased in the late grain filling stage and ultimately reduced the TKW and grain yield of wheat. The grain filling period is an extremely important growth stage in the growth and development of wheat. The yield of wheat depends mainly on the grain weight after flowering， and the grain weight depends mainly on the influence of the filling rate and the duration of grain filling[33]. Waterlogging at anthesis induced leaf senescence and affected grain filling rate. Consistent with what Araki et al. reported[11]， waterlogging at jointing and after anthesis in wheat induced early leaf senescence， impaired grain filling， and ultimately reduced the grain yield. However， the effective number of grains per ear significantly decreased in WL as showed in Fig. 3 and Fig. 4， which was opposite to the reported by Hideki Araki. Filling rate is mainly controlled by genetics， whereas the duration of grain filling is greatly affected by environmental factors[34].
Grain yield of the sensitive cultivar （YM158） showed a significant reduction only when waterlogging was prolonged to only 7 dat anthesis. The results of research with waterlogging imposed at anthesis displayed great differences in wheat yield losses related to waterlogging duration. Marti et al.[35]found a linear reduction in grain yield with increasing waterlogging durations from 4 to 24 d imposed before anthesis and a loss of 50% with the longest treatment.
Conclusions
Waterlogging at anthesis greatly influenced the chlorophyll fluorescence and grain filling process of winter wheat. From the perspective of photosynthetic capacity， waterlogging during flowering can effectively reduce chlorophyll content， decrease Fv/Fm， Fv/Fo， and Fm/Fo， and increase Fv/Fo. Waterlogging at anthesis accelerated plant senescence， increased the filling rate， shortened the duration of grouting， and decreased grain weight. Waterlogging significantly reduced the grain yield of single stem， spikelets per panicle， 1 000kernel weight， dry weight of single stem， harvest index per pot. Waterlogging at anthesis reduced the photosynthetic capacity and accumulation of dry matter after anthesis in wheat， eventually reducing the CPDMA. All of the above indicators were affected by water stress， but the degree of influence of waterlogging varied between the varieties. JM102 was less sensitive to water stress than YM158.
References
[1] BRAUN HJ， ATLIN G， PAYNE T. Multilocation testing as a tool to identifyplant response to global climate change[J]. Climate Change and Crop Production， 2010， 7： 115-138.
[2] SHAW RE， MEYER WS， MCNEILL A， et al. Waterlogging in Australian agricultural landscapes： a review of plant responses and crop models[J]. Crop and Pasture Science， 2013， 64（6）： 549-562.
[3] TIAN JY， LIU J， WANG JH， et al. Trend analysis of temperature and precipitation extremes in major grain producing area of China[J]. International Journal of Climatology， 2016， 37： 672-687.