Zichng Zhng,Jingjing Co,To Gu,Xi Yng,Qiong Peng,Linyng Bi,*,Yongfeng Li,*

a Institute of Plant Protection,Jiangsu Academy of Agricultural Sciences,Key Laboratory of Food Quality and Safety of Jiangsu Province-State Key Laboratory Breeding Base,Nanjing 210014,Jiangsu,China

b Biotechnology Research Center,Hunan Academy of Agricultural Sciences,Changsha 410125,Hunan,China

Keywords:Rice Barnyardgrass Grain yield Root traits Shoot traits

ABSTRACT Barnyardgrass (Echinochloa spp.) is the most common noxious weed in rice paddies as it inhibits rice growth and reduces grain yield.To date,little information is available on above-and belowgroundgrowth changes in rice due to neighboring barnyardgrass.This study aimed to investigate the changes in root traits and shoot growth of rice when it is grown with different kinds of barnyardgrass. Japonica rice plants (var.Nanjing 9108) were co-cultured with two varieties of Echinochloa crusgalli (L.) Beauv.(EP,var.mitis(pursh) Petern;EH,var.zelayensis(H.B.K.)Hitchc),and E.colonum(L.)Link (EL)in the field in 2017 and 2018.Four treatments included control(i.e.,weed free rice plants)and co-cultures with each of three barnyardgrasses(EP,EH,and EL).The results revealed that EP,EH,and EL treatments significantly reduced rice grain yields by 30.6%–36.2%,42.5%–46.5%,and 10.6%–14.3%,respectively.Shoot growth including shoot dry weight,leaf photosynthetic rate,zeatin (Z) and zeatin riboside (ZR) in grains,and activities of key enzymes involved in sucrose-to-starch conversion in grains and root traits,such as length density,root dry weight,total absorbing surface area,active absorption surface area,oxidation activity,and Z+ZR contents in roots were dramatically reduced during post-heading stages of rice when grown with the three kinds of barnyardgrass.Moreover,above-mentioned rice shoot growth indices were strongly and positively correlated with root traits.These results suggested the decrease in rice shoot growth and root traits during post-heading stages contributes to the reduction in the rice yield when it grows with barnyardgrass neighbors.

1.Introduction

Rice(Oryza sativaL.)is one of the most important food crops in the world,providing 35%–60% of the total daily caloric intake for approximately three billion people.Rapid population growth and economic development have resulted in growing pressure to increase food production.It is estimated that by the year 2035,it will be necessary to produce about 60% more rice to meet food needs,corresponding to annual increases in yield of at least 1.2%[1–3].However,the average annual increase in rice yield worldwide is less than 1%,and the outputs of many Asian countries have not increased in the past 20 years [4,5].

During its growth process,rice is inevitably affected by various unfavorable factors that may cause yield losses.Weeds are one of the biotic factors that limit high,stable rice yields.It is estimated that crop yield losses caused by weeds in China now exceed 3 million tons every year [6].Barnyardgrass in particular is a problematic weed in rice paddies;it is also an example of crop mimicry,as it closely resembles the rice crop at the seedling stage [7] and is only easily distinguishable in the late growth stage,rice yield loss is unavoidable [8].Barnyardgrass consists of different species and varieties.The most noxious barnyardgrasses areEchinochloa crusgalli(L.) Beauv.var.mitis(pursh) Petern andE.crusgalli(L.)Beauv var.zelayensis(H.B.K.) Hitchc in the lower reaches of the Yangtze River.The speciesE.colonum(L.) Link occurs less frequently but has spread rapidly from upland areas to paddy fields in recent years [9].Research on barnyardgrass has shown that the duration and timing of weed emergence,weed density,weed species,and crop sowing methods can substantially affect the magnitude of competition and yield losses and may cause 21%–79%loss in rice yield [10–16].Our recent work showed a significant decrease in the grain quality of rice when neighbored by barnyardgrass [17].

To develop rice cultivars with traits that enable the crop to suppress weed growth,better understand the mechanism of competition between rice and weeds is a must.Roots are an integral part of plants and are involved in the acquisition of water and nutrients,plant hormone synthesis,and anchorage of plants [18].Previous studies demonstrated that some root traits of rice,including root length,root dry weight,root absorption area,and root oxidative activity,are significantly positively correlated with rice yield[19–21].Cytokinins are an important class of phytohormones that regulate cell division,chloroplast biogenesis,bud and root differentiation,stress tolerance,and organ senescence [22–24].It has been reported thatjaponica/indicahybrid rice combinations exhibit a much higher zeatin content in their root exudates than their restorer lines,in line with a stronger root activity of the hybrids;therefore,root cytokinin content could be an important trait of root physiology [25].Previous studies also showed that when rice was grown with barnyardgrass,the rice root mass predominated vertically and laterally within the soil profile of plots [26] and the root oxidation activity,root-bleeding sap,and zeatin+zeatin riboside contents decreased as the density of barnyardgrass increased[27].Shoot traits such as shoot biomass,leaf photosynthetic rate,and the activities of sucrose synthase in grains play important role in the rice yield formation [28,29].However,little empirical information is available on the changes that occur in the root and shoot traits of rice when grown alongside barnyardgrass,especially for different kinds of neighboring barnyardgrass.

The study’s objective was to investigate the changes in root traits and shoot growth of rice when it is neighbored by three kinds of barnyardgrass.The relationships between rice root morphological and physiological traits and shoot traits were examined.Root dry weight,root length density,root diameter,root total absorbing surface area,root active absorbing surface area,root oxidation activity,Z+ZR in roots and grains,the leaf photosynthetic rate,the activities of sucrose synthase (SuSase),adenosine diphosphate glucose pyrophosphorylase (AGPase) and starch synthase (StSase)in grains,and shoot dry weight of rice were all determined.Characteristic biological traits of barnyardgrass were also measured,such as plant height,and shoot and root dry weight.The findings of this study revealed the rice response mechanism to cooccurring barnyardgrass plants and provides a theoretical basis and technical guidance for high-yield rice cultivation and weed control.

2.Materials and methods

2.1.Field site and plants

The experiment was conducted at a farm located at the Jiangsu Academy of Agricultural Sciences,Jiangsu Province,China(32°18′N,118°52′E) during the rice growing season (May-October)of 2017 and 2018.The soil here is sandy loam(Typic Fluvaquents,Etisols,USDA classification) and consists of 19.2 g kg-1organic matter,106.8 mg kg-1total N,35.6 mg kg-1Olsen-P and 82.6 mg kg-1exchangeable K.The soil field capacity moisture content (0.181 g g-1) was gravimetrically measured after constant drainage;its bulk density was 1.33 g cm-3.Weather records,including average air temperature,sunshine hours,and precipitation during the rice-growing period,over both years,were obtained from a weather station located at the experimental site and are shown in Table S1.

A high-yield rice variety,Nanjing 9108(ajaponicavariety),was selected for study and is currently extensive used in local production.We chose two barnyardgrass varieties,Echinochloa crusgalli(L.) Beauv.var.mitis(pursh) Petern (abbreviated here as EP) andE.crusgalli(L.) Beauv.var.zelayensis(H.B.K.) Hitchc (abbreviated here as EH),both of which occur in almost all rice fields as well as a barnyardgrass species,E.colonum(L.) Link (abbreviated here as EL),that has gradually spread from upland areas to paddy fields in recent years as weeds.Barnyardgrass seeds were collected from multiple rice fields around Nanjing in 2016.Rice and barnyardgrass were grown in the paddy field from 2017 to 2018.Rice seedlings were raised in a seedbed (sown on May 10–11) and were transplanted on June 9–10 at a hill spacing of 20 cm × 20 cm with two seedlings.Nanjing 9108 was headed (50% plants) on August 20–24 and then harvested on October 15–16 in both study years.

2.2.Treatments

The experiment was comprised of four treatments in a complete randomized block design with four replicates.Each plot size was 5 m × 6 m,and the plots (n=16) were separated by an alley that was 0.5 m wide with a plastic film barrier inserted into the soil at a depth of 50 cm.The experimental treatments were as follows,control (weed free),rice grown with EP,rice grown with EH,and rice grown with EL.

Co-culturing began 5 days after rice plants were transplanted and was maintained through the rice maturity period.Seeds of three kinds of barnyardgrass were sown separately in seeding trays on the day the experimental field was irrigated,harrowed,and leveled.Each kind of barnyardgrass was transplanted 5 days after the rice seedlings were transplanted.Our survey data showed barnyardgrass density was ca.six plants per m2at some lager farms(>600 ha-1)in rice–wheat rotation areas in Jiangsu Province.Some of factors,such as a shortage of labor,a short time gap between rice and wheat rotation,a large rice-planting area,and untimely weed control,could be important reasons for this phenomenon.To better characterize and compare the response of root and shoot traits of rice to the three kinds of barnyardgrass,the specific site of each barnyardgrass seedling was designed with rice located in the four corners of the plot and one barnyardgrass seeding transplanted in the intersection of the diagonal lines (rice,barnyardgrass=4,1,about six barnyardgrass plants per m2).Naturally occurring background weeds were removed by hand throughout the experiment.Except for drainage during the rice mid-season,the field was continuously flooded with 2–3 cm water until 1 week before rice was harvested in all treatments.

In each year,N (60 kg ha-1urea),P (30 kg ha-1single superphosphate),and K(40 kg ha-1KCl)were applied and incorporated before transplanting.N was applied at the early tillering stage 8 days after transplanting (DAT) (40 kg ha-1),panicle initiation(45 DAT,50 kg ha-1),and initiation of spikelet differentiation (61 DAT,50 kg ha-1).The total N application was 200 kg ha-1,which was within the recommended range.

2.3.Sampling and measurements

Rice root length density,diameter,and dry weight were determined at the mid-tillering(20–22 DAT,MT),panicle initiation(remaining leaf age of rice was 3.5,PI),and heading(50%rice heading,HD) stages,and at 10,20,and 30 days after rice heading.Samples of the roots of the four rice plants around the barnyardgrass plant were removed from each plot with a spade.The soil cube remove from around each rice plant’s roots was 20 cm × 20 cm × 20 cm(length × width × depth).Roots were carefully rinsed and detached from their nodal bases discarding the broken roots belonging to the barnyardgrass.In order to measure the changes in rice root length density and diameter when neighbored by barnyardgrass,roots were placed in a glass tray with a thin layer of water,scanned with an Epson Expression 1680 scanner (Seiko Epson Corp.,Tokyo,Japan),and then analyzed with the WinRhizo root analysis system (Regent Instruments Inc.,Quebec,Canada).Root length density was calculated from the root length and the volume of the soil core(V),i.e.,root length density(cm cm-3)=root length(cm)/volume(cm3).The roots and shoots of rice were ovendried at 70 °C to a constant weight,and then the dry weight was measured.The crop growth rate (CGR) (g m-2day-1) was calculated using the following formulas,CGR=(DW2–DW1)/(t2–t1),in which DW1and DW2are the first and the second measurement of dry matter weight(g m-2),respectively,andt1andt2represent the first and second time (day),respectively of measurement.At the same time,two barnyardgrass plants were also sampled in each plot with the same method to determine the root dry weight,shoot dry weight,and leaf area of barnyardgrass.In addition,the height of the barnyardgrass was measured from the base of the plant to the tip of its uppermost leaf.

On the aforementioned dates,same as the above sampling method,two rice samples (four rice plants around the barnyardgrass plant as a sample) in each plot were sampled for measurements of root total absorption surface area,active absorption surface area,oxidation activity,and Z+ZR contents.The root total absorption surface area and active absorption surface area of fresh root samples were determined by the methylene blue dyeing method [30].Root oxidation activity (μg α-NAg-1DW h-1) was measured according to the method of Ramasamy et al.[31].The remaining rice roots were used for Z +ZR measurements;the root Z+ZR extraction and purification methods are described by Bollmark et al.[32] and He [33].Z+ZR contents (nmol g-1DW) were analyzed by an enzyme-linked immunosorbent assay (ELISA) as previously described[34,35].The recoveries of the root Z+ZR contents were 78.9%–82.1%.

The number of rice tillers was also recorded by hand-counting the total number of tillers in a pre-marked 1 m2area every 7 days until 77 days after transplanting at each plot.The photosynthetic rate of the top fully-expanded rice leaves was determined at mid-tillering,panicle initiation,heading,early grain filling,mid grain filling,and late grain filling.A gas exchange analyzer (Li-Cor 6400 portable photosynthesis measurement system,Li-Cor Lincoln,NE,USA) was used for these measurements from 9:00 to 11:00 AM when photosynthetic active radiation above the canopy was 1000–1100 μmol m-2s-1.Six leaves were used for each treatment.

At heading time,80–100 panicles that headed on the same day were tagged in each plot.Eight tagged panicles from each plot were sampled at 10,20,and 30 days after heading.All grains from each panicle were removed.The sampled grains were used to measure Z+ZR contents and the activities of SuSase,AGPase and StSase.The method for measuring Z+ZR contents in grains is the same as the measurement of Z+ZR in roots.The method for preparation of SuSase,AGPase and StSase were according to those described by Yang et al.[36].

Grain yield was determined for all plants in the 10 m2sites,except for the border plants in each plot and adjusted to a moisture content of 0.14 g H2O g-1fresh weight.Aboveground biomass and yield components,i.e.the number of panicles per square meter,the percentage of filled grains,and grain weight,were determined from 40 rice plants around barnyardgrass plants (excluding the border plants) sampled randomly from each plot.The percentage of filled grains was defined as the filled grains (specific gravity 1.06 g cm-3)as a percentage of total number of spikelets.The number of spikelets per panicle was calculated from the grain yield,grain weight (14% moisture content),and percentage of filled grains,i.e.the number of spikelets per panicle=grain yield per square meter/(number of panicles per m2× 1000 grain weight × percentage of filled grains).

2.4.Statistical analysis

Analysis of variance was performed using SAS/STAT statistical analysis package (v6.12,SAS Institute,Cary,NC,USA).Data from each sampling date were analyzed separately.Means were tested by least significant difference atP=0.05 (LSD0.05).

3.Results

3.1.Barnyardgrass plant growth characteristics

The shoot dry weight,root dry weight,leaf area,and height of barnyardgrass at different rice growth stages are presented in Fig.1A–H.The shoot dry weight of the three kinds of barnyardgrass showed no significant difference at mid-tillering,but a significant difference was observed at other measurement times.EH had the greatest shoot dry weight,followed in descending order by EP and EL.Similar trends were observed for root dry weight among the three kinds of barnyardgrass at all the measurement times except for rice maturity.There was no significant difference in the root dry weight at maturity between EP and EH,but both EP and EH exceeded EL.The leaf area of the three kinds of barnyardgrass showed no significant difference at mid-tillering,while significant changes were detected at the panicle initiation and heading stages of rice.No difference in leaf area was observed between EP and EH,but the leaf areas of both were significantly higher than that of EL.The height of barnyardgrass differed significantly at different measurement times.Although EH was significantly taller than EP and EL at all the measurement times,EP height was similar to EL at mid-tillering and panicle initiation stages yet taller at the heading and maturity stages.

3.2.Rice yield and its components

Table 1 shows the yield and yield components of rice when grown with the three kinds of barnyardgrass in 2017 and 2018.The interactions between year and treatment were not significant for grain yield (Table S2).Compared with the control,the grain yield of rice when grown with barnyardgrass was reduced,but this reduction differed greatly among the barnyardgrass treatments(Table 1).EP,EH,and EL significantly reduced rice grain yield by 30.6%,36.2%,and 14.3%,respectively,in 2017 and by 42.5%,46.5%,and 10.6%,respectively,in 2018.These results indicated that the competitiveness of the three kinds of barnyardgrass was dissimilar,ranked as EH>EP>EL.Based on the yield component analysis,no significant difference in the panicle number was observed among all treatments compared with that of the control.However,the rice grain number per panicle,seed setting rate,and 1000-grain weight decreased significantly under the EP and EH treatments,and the seed setting rate and 1000-grain weight were lower in EL compared with those in the control (Table 1).

Table 1 Changes in grain yield and yield components of the rice variety,Nanjing 9108,when grown with neighboring barnyardgrass in the field in 2017 and 2018.

3.3.Rice root length density,diameter,and dry weight

Rice root length density increased at first,and then decreased,exhibiting different patterns of change during various growth stages of rice when grown with barnyardgrass(Fig.2A–F).Rice root length density showed no significant difference among treatments at the rice mid-tillering or panicle initiation stages,while significant changes were detected at the rice heading and grain filling stages.Compared with the control,rice root length density under EP,EH,and EL treatments decreased markedly at the rice heading and filling stages,with decreases of 17.4%–35.2%,24.0%–43.4%,and 9.5%–15.9%,respectively.Root diameter showed no significant difference among all treatments on the same measurement date when rice was grown with different kinds of barnyardgrass(Fig.2C,D).Similar to root length density,root dry weight was similar among treatments at the rice mid-tillering or panicle initiation stages,but marked decreases were observed at the heading and filling stages.Reductions ranged from 23.1%–26.9%,26.8%–31.2%,and 9.8%–16.0% in EP,EH and EL,respectively (Fig.2E,F).

Fig.1.Changes in shoot dry weight(A,B),root dry weight(C,D),leaf area(E,F),and height(G,H)of barnyardgrass grown with rice(Nanjing 9108)across the latter’s growth stages in the field in 2017 and 2018.EP,Echinochloa crusgalli(L.)Beauv.var.mitis(pursh)Petern;EH,E.crusgalli(L.)Beauv.var.zelayensis (H.B.K.)Hitchc;EL,E.colonum(L.)Link.MT,PI,HD,and MA indicate the mid-tillering stage,panicle initiation stage,heading stage and maturity stage,respectively.Vertical bars represent±standard error of the mean(n=4)where these exceed the size of the symbol.Those with different letters are significantly different(P ≤0.05);those marked with ns are not significantly different(P >0.05) within the same period of measurement.

3.4.The total root absorbing surface area,root active absorbing surface area,root oxidation activity,and contents of Z+ZR in roots of rice

No significant difference was observed among treatments in the root total absorbing surface area at the early growth stage,but significant differences were detected at the heading and grain fillingstages (Fig.3A–D).Compared with the control,the EP,EH,and EL treatments markedly reduced the rice root total absorbing surface areas by 23.3%–36.1%,30.7%–50.2%,and 13.3%–15.4%,respectively,in 2017 and by 21.1%–34.7%,30.4%–52.0%,and 14.9%–22.2%,respectively,in 2018 after heading.The rice root active absorption surface area in EP,EH,and EL treatments exhibited similar changes as the rice root total absorbing surface area and was greatly reduced by 27.7%–33.8%,41.2%–46.4%,and 12.0%–18.3%,respectively,in 2017 and by 23.5%–35.3%,38.8%–48.1%,and 10.5%–19.9%,respectively,in 2018.

Fig.2.Changes in root length density(A,B),root diameter(C,D)and root dry weight(E,F)of the rice variety,Nanjing 9108,with neighboring barnyardgrass in the field in 2017 and 2018.Control,weed free;EP, Echinochloa crusgalli (L.) Beauv.var. mitis (pursh) Petern;EH, E.crusgalli (L.) Beauv.var. zelayensis (H.B.K.) Hitchc;EL, E.colonum (L.)Link.MT,PI,HD,and GF indicate the mid-tillering stage,panicle initiation stage,heading stage and grain-filling stage,respectively.Vertical bars represent standard error of the mean(n=4).Those with different letters are significantly different at P ≤0.05;those marked with ns are not significantly different at P>0.05 within the same period of measurement.

Fig.3.Changes in root total absorbing surface area(A,B)and root activity absorbing surface area(C,D)of the rice variety,Nanjing 9108,with neighboring barnyardgrass in the field in 2017 and 2018.Control,weed free;EP, Echinochloa crusgalli (L.) Beauv.var. mitis (pursh) Petern;EH, E.crusgalli (L.) Beauv.var. zelayensis (H.B.K.) Hitchc;EL, E.colonum (L.) Link.MT,PI,HD,and GF indicate the mid-tillering stage,panicle initiation stage,heading stage and grain filling stage,respectively.Vertical bars represent±standard error of the mean(n=4).Those with different letters are significantly different at P ≤0.05;those marked with ns are not significantly different at P>0.05 within the same period of measurement.

Rice root oxidation activity gradually decreased with growth and development and exhibited significant differences among treatments in different growth stages (Fig.4A and B).There were few differences in rice root oxidation activity among treatments at the mid-tillering and panicle initiation stages,but activity differed markedly at the heading and grain stages.Compared with the control,the EP,EH and EL treatments greatly reduced rice root oxidation activity by 14.7%–19.8%,22.0%–25.9%,and 7.6%–10.2%,respectively,in 2017 and by 17.2%–20.6%,25.7%–26.0%,and 8.8%–12.5%,respectively,in 2018 after rice heading.Based on a comparison of all treatments,EH treatments resulted in the largest reduction in rice root oxidation activity,followed by EP and EL.

Rice root Z+ZR contents decreased with growth and development of rice when grown with barnyardgrass(Fig.4C,D).The root Z+ZR contents in the control,EP,EH,and EL showed no difference at the mid-tillering and panicle initiation stages,but markedly decreased at the heading and grain filling stages.Compared with the control,the Z+ZR contents of rice roots in the EP,EH,and EL treatments after heading significantly decreased by 21.3%–35.9%,48.9%–50.7%,and 13.4%–19.9%,respectively in 2017 and by 25.4%–33.5%,42.1%–48.5%,and 12.8%–16.3% in 2018,respectively.

3.5.The tiller number,leaf photosynthetic rate,shoot dry weight,and crop growth rate of rice

The tiller number of rice increased with growth,reached a peak,and then decreased.There was no significant difference among the treatments(Fig.S1).The leaf photosynthetic rate of rice at different growth stages is presented in Table 2.The photosynthetic rate of leaves showed no significant difference among the control,EP,EH,and EL treatments at the mid-tillering and panicle initiation stages.It was significantly reduced under EP,EH,and EL treatments at the heading and grain filling stages compared with the control.The degree of reduction was in the order of EH >EP >EL.The CGR of the EL treatment from panicle initiation to heading and from heading to maturity were lower than that of the control,followed by EP.The CGR of the EH treatment was the least.

Fig.4.Changes in root oxidation activity(A,B)and Z+ZR contents(C,D)in roots of the rice variety,Nanjing 9108,with neighboring barnyardgrass in the field in 2017 and 2018.Control,weed free;EP,Echinochloa crusgalli(L.)Beauv.var.mitis(pursh)Petern;EH,E.crusgalli(L.)Beauv.var.zelayensis(H.B.K.)Hitchc;EL,E.colonum(L.)Link.MT,PI,HD,and GF indicate the mid-tillering stage,panicle initiation stage,heading stage and grain filling stage,respectively.Vertical bars represent ± standard error of the mean(n=4).Those with different letters are significantly different at P ≤0.05;those marked with ns are not significantly different at P >0.05 within the same period of measurement.

Table 2 Changes in the leaf photosynthetic rate (μmol m-2 s-1) of the rice variety,Nanjing 9108,when grown with neighboring barnyardgrass in the field in 2017 and 2018.

Similar to the leaf photosynthetic rate,there was no difference in the shoot dry weight of rice among treatments at the midtillering and panicle initiation stages,but a marked decrease occurred at the heading and maturity stages when compared with the control(Table 3).EP,EH and EL significantly reduced the shoot dry weight of rice by 23.9%–26.0%,28.5%–30.4%,and 16.3%–18.3%,respectively,at heading stage and by 23.0%–30.6%,33.4%–38.2%,and 12.2%–15.7%,respectively,at the maturity stage.The magnitude of the decrease in order from greatest to least was EH >EP >EL.

Table 3 Changes in the shoot dry weight of the rice variety,Nanjing 9108,when grown with neighboring barnyardgrass in the field in 2017 and 2018.

Table 4 Changes in the Z+ZR contents(pmol g-1 DW)in grains of the rice variety,Nanjing 9108 at 10,20,and 30 days after rice heading,when grown with neighboring barnyardgrass in the field in 2017 and 2018.

3.6.Z+ZR contents and activities of SuSase,AGPase and StSase in grains of rice

Table 4 shows the Z+ZR contents in grains of rice when grown with the three kinds of barnyardgrass in 2017 and 2018.EP,EH,and EL treatments significantly reduced the contents of Z+ZR in rice grains when compared that of the control.The EH treatment exhibited the largest reduction in rice Z+ZR contents in grains when grown with barnyardgrass varieties,followed by EP and EL(Table 4).

Both the EP and EH treatments significantly reduced the activities of SuSase,AGPase and StSase in grains at 10,20 and 30 days after rice heading,and no difference was observed between EP and EH treatments.There was no difference between the control and the EL treatment in the activities of SuSase,AGPase and StSase in grains at 10 days after heading,but all activities significantly decreased at 20 and 30 days after rice heading when compared with the control (Table 5).

Table 5 Changes in the activities of SuSase,AGPase and StSase (pmol g-1 DW) in grains of the rice variety,Nanjing 9108 during rice grain filling stage,when grown with neighboring barnyardgrass in the field in 2017 and 2018.

4.Discussion

Barnyardgrass is a weed that usually coexists with rice.There are many species and varieties of barnyardgrass,with a large occurrence base.Barnyardgrass has a close genetic relationship to rice that could lead to severe yield reductions or prevent rice harvesting when barnyardgrass is present [8].The competitive ability of barnyardgrass and the consequent reduction in rice yield in the presence of different kinds of barnyardgrass also differ.In this study,a significant difference was observed in the grain yield of Nanjing 9108 when it was grown with three kinds of barnyardgrass in EP,EH and EL treatments,among which EH had the greatest effect on rice grain yield(a reduction of 42.5%–46.5%),followed by EP (30.6%–36.2%),and then by EL (10.6%–14.3%),Additionally,all three kinds of barnyardgrass greatly reduced the rice yield,which verified the findings of previous studies [17,37].However,the mechanism underlying has been not fully understood,there are some possible explanations based on our observations.

Previous studies indicated that the increase of grain yield was mainly attributed to magnifying of total sink capacity with the increase of the number of panicles per unit area and the number of spikelets per panicle as the key factor [38,39].However,the number of panicles per unit area was not affected by barnyardgrass at early stage of rice,in which stage,barnyardgrass exhibited little competition with rice due to sufficient basal fertilizer and small weed seedling size[37].Although the results in this study showed that barnyardgrass had no effect on the number of tillers and dry matter accumulation of rice before the rice panicle initiation stage when rice was grown with barnyardgrass,we still proposed to control weeds by applications of pre-and post-emergence herbicides[40–42] and breeding high-yielding,weed-competitive rice cultivars.The weed-competitive cultivars exhibited desirable agronomic traits at early stage,such as early tillering,high leaf area index,early vigor,and large biomass,which support the relationship of rapid early growth with weed-suppressive ability [43,44].In this study,the diminished rice sink capacity when co-planted barnyardgrass was mainly due to a sharp decrease in the number of spikelets per panicle,down by 15.4%–33.1% compared with the control treatment.Herein we found that co-planted barnyardgrass treatments significantly decrease CGR from panicle initiation to heading.Previous studies indicated that a stronger ability of aboveground biomass production from panicle initiation to heading could promote spikelet differentiation,reduce spikelet degeneration,and increase endosperm cell proliferation at the early seed-development stage,thus leading to an increase in sink size capacity [2,5].The result implies that decrease in CGR may contribute to the lower in rice sink capacity when rice neighbored barnyardgrass.

Rice yield can be defined as the product of yield sink capacity and filling efficiency [45].In this study we found that,co-planted barnyardgrass could not only reduce the sink capacity,but also could adversely affect the rice filling efficiency.Filling efficiency is determined by a complex balance of sink and source.Greater sink capacity requires stronger source ability for stable grain filling.It is generally accepted that more than 60% of grain yield is produced by photosynthesis during the ripening period [46,47].Herein we found that rice leaf photosynthetic rate was significantly influenced by barnyardgrass.Barnyardgrass is a fastgrowing C4plant.In our study,we found that the leaf area,height and biomass of barnyardgrass increased rapidly from rice panicle initiation to the heading stage.The average barnyardgrass height at the rice heading stage was 162.4 cm in EH,138.1 cm in EP and 107.4 cm in EL,i.e.,taller than rice (98.8 cm).The taller barnyardgrass made it easier for these plants to intercept light for photosynthesis,which contributed to the dry matter accumulation aboveground and the shoot growth of barnyardgrass over shorter neighboring rice (Fig.1A–H).The greater height and biomass of barnyardgrass could shade rice;such asymmetric competition could have caused greater decreases in the rice leaf photosynthetic rate during grain filling period.

It has been hypothesized that higher enzymatic activity(involving the breakdown of sucrose in the sink) could significantly increase sink activity by lowering the local concentration of sucrose,thereby generating a gradient that allows for the further unloading of sucrose from the phloem [48].SuSase is proposed to be a key enzyme in sucrose–starch metabolic pathway.Its activity is regarded as a biochemical marker of sink strength [49].AGPase and StSase are considered key enzymes involved in starch synthesis.Their activities are closely associated with the rate and quantity of starch synthesis [50].In this study,we also observed that SuSase,AGPase and StSase activities in the grains of rice neighboring barnyardgrass were significantly lower during the rice grain filling period,suggesting that barnyardgrass significantly reduced sink activity in grains of rice.

It is proposed that an interdependent relationship exists between root and shoot,i.e.,active shoots ensure a sufficient supply of carbohydrates for root development and the maintenance of active root functions;the activation of root functions improves shoot growth by supplying enough nutrients,water and phytohormones to shoots,thus ensuring an increase in productivity[29,51,52].In this study,we found that when rice was grown with different kinds of barnyardgrass,no difference was observed the rice root traits,such as root length density,total absorbing surface area,root activity absorption surface area,and root oxidative activity among all the treatments at the mid-tillering and panicle initiation stages,while significant decrease were observed during the rice grain-filling stage among treatments.The root traits are important parameters indicative of overall root function.Correlation analysis revealed that the root indices above mentioned at the heading and filling stages were significantly positively correlated with rice shoot traits and grain yield (Table S3).In addition to shading rice after rice panicle initiation stage,barnyardgrass also has a larger root system that can absorb a lot of nutrients from soil [16] resulting in nutrient deficiency in the soil,thus causing the aforementioned root indices decrease.The decrease in these root indices reduced the ability of roots to absorb and transport nutrients for shoot growth,which further suppressed aboveground plant growth and development;in turn,the weakened aboveground production capacity inhibited root growth.In crops,the imbalance between roots and aboveground parts of rice neighboring barnyardgrass would decrease the spikelets per panicle,seedsetting rate and grain weight,and ultimately result in decreased rice yield.

Cytokinins are believed to play a major role in promoting cell division and delaying senescence [34,53–55].Herein,we observed that the Z+ZR contents in roots and grains of rice decreased during the grain-filling stage when rice neighbored barnyardgrass.A correlation analysis showed that Z+ZR content in roots was significantly positively correlated with other root traits such as root length density and root dry weight.It has been reported that a higher Z+ZR level contributed to a higher indole-3-acetic acid level in roots,leading to the maintenance of root growth under adverse conditions [56].This suggests that the reduction in the Z+ZR contents in roots of rice neighboring barnyardgrass may contribute to a lower indole-3-acetic acid level in roots and affect root growth,leading to lower grain yield.

5.Conclusions

Rice yield was significantly reduced by 30.6%–36.2%,42.5%–46.5%,and 10.6%–14.3% when grown with EP,EH,and EL,respectively.Among these treatments,the most competitive barnyardgrass was EH,followed by EP and then EL.The root dry weight,length density,total absorption area,active absorption surface,oxidative activity,Z+ZR contents in roots and grains,leaf photosynthetic rate,shoot dry weight,and SuSase,AGPase and StSase activities in rice grains at the heading and grain-filling stages were dramatically reduced in the presence of barnyardgrass.Correlations revealed that the diminished aforementioned root and shoot traits are likely key factors reducing the grain yield of rice when it is grown with neighboring barnyardgrass.

CRediT authorship contribution statement

Yongfeng Li and Lianyang Bai:conceived and designed the experiments;Zichang Zhang,Jingjing Cao,Tao Gu,and Xia Yang:performed the experiments;Zichang Zhang and Qiong Peng:analyzed data;Zichang Zhang:wrote and revised the paper.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (31871982,31672042),the National Key Research and Development Program of China (2016YFD0200805),and the Jiangsu Agricultural Science and Technology Innovation Fund (CX(18)3056).www.letpub.com

Appendix A.Supplementary data

Supplementary data for this article can be found online at https://doi.org/10.1016/j.cj.2020.10.011.