Effects of planting dates and shading on carbohydrate content,yield, and fiber quality in cotton with respect to fruiting positions

ZHAO Wen-qing, WU You, Zahoor Rizwan, WANG You-hua, MA Yi-na, CHEN Bing-lin, MENG Ya-li,ZHOU Zhi-guo

Key Laboratory of Crop Physiology Ecology and Production Management, Ministry of Agriculture/Jiangsu Collaborative Innovation Center for Modern Crop Production (JCIC-MCP), Nanjing Agricultural University, Nanjing 210095, P.R.China

1. lntroduction

The indeterminate growth habit of cotton causes formation of bolls at different fruiting positions (FP) to grow at different times under various environmental conditions (Bondada and Oosterhuis 2001). Many studies have documented that the FP1 contributed more to the total cotton yield and produced better fiber than bolls at any other positions of the same sympodial branch (Jenkinset al. 1990b; Pettigrew 1995; Heitholt 1997; Davidoniset al. 2004). However,the contribution of FPs differs between cultivars and under different environmental conditions. Under optimal conditions, the distal FPs (FP3 and above) could sustain a 58.8% retention rate of cotton bolls in high cotton yield fields (7 657 kg ha–1, 24 240 plants ha–1) (Guet al. 2010).In contrast, the distal FP (FP3) produced less cotton bolls with weaker fiber under normal conditions, but the rate of reduction was less in low temperature-tolerant cultivars (Maet al. 2014). Thus, cotton yield and fiber quality on distal FPs might be more plastic under various environmental conditions, which could minimize reduced yield and fiber quality in adverse conditions. Studies about the underlying physiological changes in distal FPs and their relationship with inner FPs could be used to identify genotypes that are resistant to adverse conditions or to develop new varieties for better productivity under environmental stress.

The above-mentioned changes are related to source-sink(leaf-boll) interactions. In cotton, the development of a boll and its subtending leaf are closely associated (Wullschleger and Oosterhuis 1990). In a mature cotton boll, most of the carbon originate from the subtending leaf (Ashley 1972;Grindlay 1997). Efficient production of carbohydrates(mainly sucrose and starch) in leaves and translocation of carbohydrates towards bolls is essential for maintaining cotton yield and fiber quality (Jianget al. 2006; Ahmadiet al. 2009). In mature fiber cells, cellulose constitutes more than 90% of the dry weight (Delmer and Amor 1995), which means that the process of cotton fiber formation is primarily a process of cellulose synthesis. Sucrose, transported from the subtending leaf, is the initial carbon source for cellulose synthesis and supplies UDP-glucose as the immediate substrate for cellulose polymerization (Delmer and Haigler 2002; Williamsonet al. 2002). Sucrose content in cotton fiber and subtending leaf are highly correlated with the cellulose accumulation and final fiber quality (Wanget al.2009; Gaoet al. 2012), and all of these factors are influenced by genetic and environmental factors.

Both low temperature and light de ficiency are vital environmental constraints for cotton production (Yeateset al. 2010; Lvet al. 2013). These two important constraints can occur singly or combined in cotton-growing areas of the Yangtze River Valley and the Yellow River Valley in China under multiple cropping systems due to late harvesting of preceding full-season winter crops (Chenet al. 2014b).Previous studies have shown that low temperatures as a result of late planting significantly increased fiber sucrose content, and also decreased sucrose transformation rate and cellulose content, which consequently reduced lint yield and decreased fiber quality due to carbohydrate deficiency(Donget al. 2006, 2010; Shuet al. 2009; Caoet al. 2011; Luet al. 2017). In some studies, low light decreased lint yield(SassenrathColeet al. 1996; Dusserreet al. 2002; Lvet al.2013; Echer and Rosolem 2015), increased fiber length, and lowered strength of fiber and micronaire (Pettigrew 1995,2001). In other studies, results suggested that low light decreased or did not significantly affect fiber length (Zhao and Oosterhuis 2000; Wanget al. 2005). Moreover, recent research showed that both low temperature and low light had severe effects on carbohydrate content in both cotton leaf and fiber, which decreased cotton yield and fiber quality(Liuet al. 2015b; Huet al. 2016).

Many studies have described low temperature and light effects on cotton fiber and/or the subtending leaf (Chenet al.2014a, b; Liuet al. 2015b; Huet al. 2016). However, it is still not fully understood how low temperature, reduced light,and their interaction affect cotton source-sink relationships at different FPs. Therefore, an experiment with a variety of planting dates and shading treatments was designed to determine the effects of low temperature and light on cotton yield, fiber quality, and carbohydrate content in leaves and fibers of two different FPs in low temperaturetolerant and low temperature-sensitive cultivars. The goals of this experiment were to elucidate differences between physiological mechanisms of cotton fiber development in different FPs in response to temperature-light stress.

2. Materials and methods

2.1. Experimental design

A two-year field experiment was conducted at Pailou Experimental Station, Nanjing Agricultural University(118°50′E, 32°02′N), Jiangsu Province, China, in 2010 and 2011. Two cotton cultivars were used: Kemian 1 (low temperature-tolerant) and Sumian 15 (low temperaturesensitive), which are both widely grown in the Yangtze River Valley (Shuet al. 2009; Liuet al. 2013). Soil was clay,mixed, thermic, typic Al fisols (udalfs; FAO Luvisol) based on a soil pro file depth of 0–20 cm. Before planting cotton,nutrient contents of soil were: 17.5 and 18.5 g kg?1organic matter, 1.1 and 1.0 g kg?1total N, 62.3 and 80.5 mg kg?1available N, 17.6 and 18.8 mg kg?1available P, and 98.3 and 110.5 mg kg?1available K in 2010 and 2011, respectively.

According to previous research, 25 April and 10 June are the optimal and late planting dates, respectively, in the Yangtze River Valley (Jianget al. 2006; Liuet al. 2015b).Thus, two planting dates, 25 April and 10 June, and two shading treatments, CRLR100% (control, with no shading)and CRLR60% (photosynthetically active radiation was reduced to about 60% of control by covering plants with a white polyethylene net that was 12 m long, 7 m wide,and 2 m high, supported by an iron stand), were used in both years. Shading started when 50% of white flowers at FP1 of the 7th fruiting branches (FB7) were blooming and continued until the bolls opened in both 2010 and 2011. The microclimate data at 17 days post anthesis (DPA) showed that the average air temperature and relative humidity were only slightly affected by shading. Only the photosynthetically active radiation (PAR) was significantly affected by shading(Huet al. 2016). Three replicates of each treatment were set up randomly in the field. Each plot was 6 m wide and 10.5 m long. Row spacing of cotton plants was 80 cm and interplant spacing was 25 cm. Cotton seeds were sown in a nursery bed and then transplanted to the field at the seedling stage with three true leaves. Furrow-irrigation was applied as needed and conventional weed and insect control measures were applied.

2.2. Sampling and processing

White flowers on FP1 and FP3 of FB7 were tagged on the same day in the same planting date with small plastic tags noting the flowering date. Labeled bolls and leaves subtending to the bolls were collected at 10, 17, 24, 31,38, and 45 DPA in OPD, and at 10, 17, 24, 31, 38, 45, 52,and 59 DPA in LPD. On each sample day, collections were conducted at 9:00–10:00 a.m., and an ice box was used to transport samples to the lab. The fibers were excised from the bolls with a scalpel and the leaves were washed with distilled water and then dried for carbohydrate analysis.

At the maturation stage, tagged bolls on FP1 and FP3 of FB7 in each treatment were harvested and ginned individually. The quality of ginned fiber was analyzed by the Cotton Quality Supervision, Inspection, and Testing Center of the Ministry of Agriculture of the People’s Republic of China. Meanwhile, cotton bolls on FP1 and FP3 from the central two rows of each plot were hand-harvested. Bolls were dried at 70°C and weighed. After ginning, fiber biomass was measured for calculating lint percentage. To calculate lint yield, we used the formula: Lint yield per ha=Boll number per ha×Weight per boll×Lint percentage

2.3. Measurements of sucrose, starch, and cellulose contents

Sucrose was extracted from fiber and leaf samples and assayed (Pettigrew 2001). About 0.1 g dried leaf tissues or 0.3 g fiber samples were extracted with 5 mL of 80%ethanol. The ethanol samples were incubated in an 80°C water bath for 30 min, and then centrifuged at 4 000 r min–1for 5 min. The pellets were extracted twice more using 80%ethanol and three aliquots of supernatant were collected together and diluted to 25 mL with 80% ethanol for sucrose determination. The sucrose assay was conducted according to the method described by Hendrix (1993).

Leaf starch was extracted from the ethanol-insoluble residue. Ethanol from the centrifuge tube was removed through evaporation, and 2 mL of distilled water was added into the tube. Starch in the residue was released in a boiling bath for 15 min and cooled to room temperature. Then,2 mL of 9.2 mol L?1HClO4was added to the mixture and hydrolyzed for 15 min. A total of 4 mL of distilled water was added into the tube and the samples were centrifuged at 4 000 r min–1for 10 min. The supernatant was collected and the residue was extracted one more time using 4.6 mol L–1HClO4. Supernatants were then mixed and diluted to 25 mL with distilled water for starch determination. The starch concentration was measured spectrophotometrically at 620 nm using an anthrone reagent with glucose as the standard.

The fiber cellulose content was extracted and assayed according to the method described by Updegraff (1969).About 0.5 g of each fiber sample was digested by an aceticnitric reagent (3 mL) in a centrifuge tube. Tubes with a lid (to reduce the possibility of evaporation) were placed in a boiling water bath for 30 min. The water level of the bath was kept at the same level as the liquid in the tubes. Samples were centrifuged at 4 000 r min–1for 5 min, then decanted and the supernatant was discarded. The residue was washed twice using distilled water and then dried. H2SO4(67%, v/v)was added to the tube and held at room temperature for 1 h. The solution was diluted from 1 to 100 mL with distilled water and 1 mL of this dilution was placed in a culture tube for cellulose determination. The cellulose content was measured spectrophotometrically at 620 nm using an anthrone reagent with pure cellulose as the standard.

2.4. Statistical analysis

Data were subjected to an analysis of variance (ANOVA)using SPSS statistics package version 17.0 (SPSS Inc.).Differences at theP＜0.05 level were considered statistically significant using the least significant difference (LSD) test.The coefficient of variation (CV, %) was calculated as the ratio of the standard deviation to the mean. The Δ (%),where, Δ (%)=(Indicator value on FP1–Indicator value on FP3)/Indicator value on FP1×100, was calculated to determine differences in FP indices under different planting and shading treatments. R, where, R=Indicator value on FP1/Indicator value on FP3, was calculated to determine the carbohydrate distribution ratio between FPs under different planting and shading treatments.

3. Results

3.1. Weather data

Weather data were collected from the National Meteorological Information Center (Nanjing Weather Station, China).Parameters related to temperature and solar radiation during fiber development for 2010 and 2011 are in Table 1. The cumulative photo-thermal index (PTI) was also calculated(calculation equations are in Zhaoet al. (2012) and given in Table 1. Based on these data, the development period of cotton bolls located in FP3 was delayed by 2 d (in OPD)or 8 d (in LPD) compared to FP1. This delay was related to the lowered MDT (mean daily temperature), MDTmax(mean daily maximum temperature), MDTmin(mean daily minimum temperature), MDSR (mean daily solar radiation), and PTI in FP3. The coefficients of variance (CV) of PTI between FPs was higher than those of MDT, MDTmax, MDTmin, and MDSR in both 2010 and 2011. In addition, CV of PTI under LPD tripled compared to the CV under OPD, which was consistent with the delay in cotton boll development period. Thus, the difference in environmental conditions for different FPs and among treatments was primarily in PTI.

3.2. Effects of planting dates and shading on cotton yield distribution and fiber-quality indices at different FPs

Lint yield distributionLint yield was decreased due to late planting dates and shading treatments (T), which reduced boll number and boll weight. The extent of reduced lint yield showed an increasing trend from FP1 to FP3. Analysis of variance showed that T, FP, and T×FP significantly affected cotton boll number and lint yield, while T and FP significantly impacted boll weight. Lint percentage was only significantly influenced by T. In both years, boll number and lint yield of cotton bolls located at FP3 were significantly lower than FP1 in all planting dates and shading treatments. Boll weight of cotton bolls located at FP3 was also significantly lower than FP1 except under OPD-CRLR100%. Δ between FP1 and FP3 for boll number and lint yield was close and far above Δ for boll weight and lint percentage (Table 2).

?

Table 2 Cotton yield and yield components at FP1 and FP3 and analysis of variance for combinations of different planting dates and shading treatments

Averaged over 2 years, the lint yield of Kemian 1 at FP1 was reduced by 24, 35, and 57% under OPD-CRLR60%,LPD-CRLR100%, and LPD-CRLR60%, respectively, while lint yield at FP3 was reduced by 31, 51, and 72%, respectively,compared with OPD-CRLR100%. The lint yield of Sumian 15 at FP1 and FP3 had a similar trend in Kemian 1: the decreases under OPD-CRLR60%, LPD-CRLR100%, and LPD-CRLR 60% were 27, 45, and 66%, respectively at FP1, and 35, 63, and 81%, respectively, at FP3, compared with OPD-CRLR100%. In addition, CV of lint yield and yield components on FP1 was lower than those on FP3. Meanwhile, the CV for cultivar Kemian 1 was lower than that for Sumian 15 at both FP1 and FP3 (data not shown). The results above illustrated that (1) yield and yield components were more variable at FP3 than at FP1 under different treatments; (2) bolls on FP3 contributed less to yield when planting date was delayed and shading was increased; (3) cotton yield and yield components of Kemian 1 on both FP1 and FP3 were more stable than Sumian 15 under adverse temperature and light conditions.

Cotton fiber qualityFiber upper-half mean length(UHML) and fiber strength were significantly affected by T and FP, and micronaire was significantly affected by T, FP, and T×FP (Table 3). For cultivar Kemian 1,UHML, fiber strength, and micronaire were lower at FP3 than those at FP1 under all the planting dates and shading treatments except OPD-CRLR100%,while for cultivar Sumian 15, they were lower at FP3 under all planting dates and shading treatments. Fiber strength and micronaire at both FP1 and FP3, and UHML at FP3 had optimal characteristics under OPD-CRLR100%, followed by OPD-CRLR60%, LPD-CRLR100%, and LPD-CRLR60%.However, UHML at FP1 was the longest under OPD-CRLR 60%, followed by OPD-CRLR100%, LPD-CRLR100%, and LPD-CRLR60%. Thus, Δ between FPs for fiber strength and micronaire increased as PTI decreased, while the largest Δ between FPs for UHML was seen under OPD-CRLR60%.This indicated that 40% shading had little effect on fiber quality at FP1 but decreased the quality at FP3 in OPD. However,40% shading significantly decreased fiber quality on both FPs in LPD, and showed an increasing trend from FP1 to FP3. The phenomena above, along with higher CV between treatments for FP3 compared to FP1 (data not shown),showed that fiber quality was more variable at FP3 than that at FP1 under different planting dates and shading treatments.

3.3. Effects of planting dates and shading on carbohydrate contents in leaves subtending to cotton boll

Sucrose content in subtending leavesFor both cultivars,as DPA progressed, sucrose content in subtending leaves inOPD decreased, while the content in LPD showed an inverted“V” shape and peaked at 24 DPA (Fig. 1). Before 38 DPA,sucrose content in the subtending leaf at FP1 was higher than that at FP3 on each measurement day. After 38 DPA,no significant differences in sucrose content of subtending leaves were observed between FPs. Compared with OPDCRLR100%, sucrose contents in subtending leaf to cotton boll were decreased under OPD-CRLR60% in both FPs,but increased under LPD-CRLR100% and LPD-CRLR60%.Compared with LPD-CRLR100%, sucrose content also decreased under LPD-CRLR60%. Thus, shading decreased the leaf sucrose content, whereas LPD increased it in both FP1 and FP3.

Table 3 Effect of different planting dates and shading treatments on cotton fiber quality at FP1 and FP3

Maximum sucrose content and sucrose transformation rate were significantly affected by T, FP, and T×FP; the minimum sucrose content was significantly affected by T(Table 4). The maximum sucrose content in the subtending leaf at FP3 was significantly lower than that at FP1 under all the planting dates and shading. The minimum sucrose content in the subtending leaf at FP3 was lower than that at FP1 in OPD but higher than that at FP1 in LPD. Sucrose transformation rate in the subtending leaf at FP3 was lower than that at FP1 (significantly in LPD). Δ and R between FPs for leaf maximum sucrose content increased as PTI decreased, which showed that decreased temperature and reduced light increased the carbohydrate distribution of the subtending leaf at FP1 but decreased it at FP3. As PTI decreased, Δ for the minimum sucrose content had no clear trend, whereas R decreased, leading to a greater increase of Δ and R for sucrose transformation rate. A larger CV of the subtending leaf minimum sucrose content/sucrose transformation rate between treatments for FP3 compared to FP1 was also observed (data not shown), indicating that the minimum sucrose content/sucrose transformation rate were more variable on FP3 than on FP1 under different planting dates and shading treatments.

Fig. 1 Effects of planting dates and shading treatments on sucrose content in the subtending leaf at fruiting position 1 (FP1) and FP3 of two cotton cultivars, Kemian 1 and Sumian 15, in 2010 and 2011. OPD, optimal planting date, 25 April; LPD, late planting date, 10 June; CRLR, crop relative light rate. Vertical bars represent standard errors.

Table 4 Differences in maximum/minimum sucrose content, sucrose transformation rate, and the maximum starch content in subtending leaf (Δ/R) between fruiting position 1 (FP1) and FP3 under different planting dates and shading treatments

Fig. 2 Effects of planting dates and shading treatments on starch content in the subtending leaf at fruiting position 1 (FP1) and FP3 of two cotton cultivars, Kemian 1 and Sumian 15, in 2010 and 2011. OPD, optimal planting date, 25 April; LPD, late planting date, 10 June; CRLR, crop relative light rate. Vertical bars represent standard errors.

Starch content in subtending leavesStarch content in subtending leaves shows a single peak as DPA increased;the peak value was delayed under LPD (Fig. 2). In the late development period of cotton boll (38–59 DPA), starch content clearly increased. On each measurement day,starch content in the subtending leaf at FP1 was higher than that at FP3. Changes in starch content in the subtending leaf at FP1 between treatments were smaller than that at FP3, indicating that FP3 was more susceptible to variable planting dates and shading. Compared with OPD-CRLR 100%, starch content in the subtending leaf to cotton boll both decreased under OPD-CRLR60% at both FPs,but increased under LPD-CRLR100% and LPD-CRLR 60%. Compared with LPD-CRLR100%, starch content also decreased under LPD-CRLR60%. Thus, shading decreased the leaf starch content at both FPs, whereas late planting dates could increase starch content at both FPs. The maximum starch content was further analyzed,and data showed that the maximum starch content was significantly influenced by T and FP (Table 4). Δ for the maximum starch content did not have a clear trend, but R values were all nearly 1.10 under various treatments,indicating that decreased PTI would not influence the leaf starch content distribution rate between FP1 and FP3.

3.4. Effects of planting dates and shading on carbohydrate contents in cotton fiber

Sucrose content in cotton fiber As DPA progressed,sucrose content in cotton fiber under OPD decreased, while the content under LPD slightly increased before 17 or 24 DPA and sharply decreased thereafter (Fig. 3). Sucrose content at FP1 was higher than that at FP3 before 24 DPA in OPD or before 31 DPA in LPD, but it was lower at FP1 than that at FP3 after 24 or 31 DPA for OPD and LPD, respectively.Compared with OPD-CRLR100%, sucrose contents in cotton fiber were decreased under OPD-CRLR60% at both FPs, but increased under LPD-CRLR100% and LPDCRLR60%. Compared with LPD-CRLR100%, sucrose content decreased under LPD-CRLR60%. Thus, similar to leaf sucrose content, shading also decreased cotton fiber sucrose content at both FPs, whereas LPD increased it at both FPs.

Cotton fiber maximum/minimum sucrose content and sucrose transformation rate were significantly affected by T, FP, and T×FP (Table 5). Fiber maximum sucrose content at FP3 was significantly lower than that at FP1 under all planting dates and shading treatments, while the minimum sucrose content at FP3 was higher (significantly in LPD) than that at FP1. Thus, a lower sucrose transformation rate of fiber at FP3 compared to that at FP1 was observed. Δ for fiber maximum sucrose content and sucrose transformation rate were positive, while Δ for fiber minimum sucrose content was negative. Both Δ and R for fiber maximum sucrose content and sucrose transformation rate increased as PTI decreased, while Δ and R for fiber minimum sucrose content had an unclear trend. These results demonstrated that fiber sucrose content distribution and sucrose transformation rate at FP1 improved compared to FP3 as PTI decreased due to late planting and shading. A smaller CV of fiber maximum/minimum sucrose content between treatments at FP3 compared to FP1 showed that the fiber sucrose

Fig. 3 Effects of planting dates and shading treatments on sucrose content in cotton fiber at fruiting position 1 (FP1) and FP3 of two cotton cultivars, Kemian 1 and Sumian 15, in 2010 and 2011. OPD, optimal planting date, 25 April; LPD, late planting date,10 June; CRLR, crop relative light rate. Vertical bars represent standard errors.

Table 5 Differences in maximum/minimum sucrose content, sucrose transformation rate, and final cellulose content of cotton fiber(Δ/R) between fruiting position 1 (FP1) and FP3 under different planting dates and shading treatments

content adjustment scale at FP3 was smaller than that at FP1, while a larger CV of fiber sucrose transformation rate between treatments at FP3 compared to that at FP1 (data not shown) indicated that the fiber sucrose transformation rate was more variable at FP3 than that at FP1 under different planting dates and shading treatments.

Cellulose content in cotton fiber As DPA progressed,cellulose content in cotton fiber increased at both FP1 and FP3. On each measurement day, cotton fiber cellulose content at FP1 was significantly higher than that at FP3(Fig. 4). Final cellulose content was most important to cotton fiber yield and quality. From Table 5, final cellulose content of fiber was significantly affected by T, FP, and T×FP. Both Δ and R for fiber cellulose content increased as PTI decreased.Compared with OPD-CRLR100%, the final cellulose content in fiber at FP1 was decreased by 11.4, 23.2, and 38.9% on average for cultivar Kemian 1, and by 12.0, 26.8, and 43.2%on average for cultivar Sumian 15, under OPD-CRLR60%,LPD-CRLR100%, and LPD-CRLR60%, respectively. Final cellulose content at FP3 was decreased by 32.6, 41.7,and 56.7% on average for cultivar Kemian 1, and by 29.2,44.5, and 60.9% on average for cultivar Sumian 15, under OPD-CRLR60%, LPD-CRLR 100%, and LPD-CRLR60%,respectively. Thus, cotton fiber cellulose was more variable at FP3 than that at FP1.

Fig. 4 Effects of planting dates and shading treatments on cellulose content in cotton fiber at fruiting position 1 (FP1) and FP3 of two cotton cultivars, Kemian 1 and Sumian 15, in 2010 and 2011. OPD, optimal planting date, 25 April; LPD, late planting date,10 June; CRLR, crop relative light rate. Vertical bars represent standard errors.

4. Discussion

4.1. Effects of planting dates and shading on cotton yield distribution and fiber-quality indices at diffe rent FPs

Previous studies have measured the effect of low temperature, light de ficiency, and their interaction on lint yield for the entire cotton plant (Zhao and Oosterhuis 1994;Dusserreet al. 2002; Iqbal and Ahmad 2003; Bozbeket al.2006; Arshadet al. 2007; Liuet al. 2015b; Huet al. 2016).In most studies, the number of bolls was determined for the whole plant, and the effect of low temperature and light deficiency on yield components on different boll locations was averaged. Some of these studies concentrated on different fruiting branches of the main stem, but few studies paid attention to different FPs of the same fruiting branch(Heitholt 1997; Anjumet al. 2002). In this study, we focused on differences in the FPs for cotton yield, fiber quality,carbohydrate content in source (subtending leaf), and sink( fiber) organs under low temperature and light deficiency,conditions that occur frequently during the cotton growing season in the Yellow River Valley and Yangtze River Valley(Donget al. 2006; Shuet al. 2009).

Yield components and fiber quality traits varied significantly within a cotton plant due to the boll period and environmental differences (Jenkinset al. 1990a; Zhaoet al.2012). In this study, lint yield, fiber strength, and micronaire at both FPs were significantly decreased under late planting and shading. The decrease under LPD-CRLR60% was greater than that under OPD-CRLR60%, which indicated that late planting dates and shading intensi fied the yield reduction and the quality decline. The yield reduction under late planting and shading at both FPs was primarily due to decreased boll number and boll weight, which is consistent with previous studies (Zhao and Oosterhuis 1994; Dusserreet al. 2002; Iqbal and Ahmad 2003; Bozbeket al. 2006;Arshadet al. 2007). Further analysis found that cotton yield reduction at FP3 was primarily caused by decreased boll number under OPD-CRLR 100% and decreased boll number and boll weight under other treatments. Fiber lengths at FP1 and FP3 also decreased significantly due to low temperature induced by late planting but had varying responses under shading treatments. For the optimal planting date, shading increased fiber length at FP1 but decreased it at FP3, while for the late planting date, shading decreased fiber length on both FPs. These results indicate that the effect of shading on fiber length was influenced by temperature, which supplements the theory that the effect of shading on cotton fiber depend on cultivar, timing of shading,and temperature (Pettigrew 1995, 2001; Roussopouloset al.1998; Zhao and Oosterhuis 2000). These results might also partly explain the inconsistent result of a previous study where shading decreased fiber length (Pettigrew 1996) and other studies that found that shading increased or did not affect fiber length (Echer and Rosolem 2015). Meanwhile,compared with FP1, lint yield, fiber length, fiber strength,and micronaire were lower at FP3, which is similar to Maet al. (2014). This might be related to the lower PTI and longer shading period of FP3 than those of FP1 under the same planting date. CVs of the indices between planting dates and shading treatments at FP3 were larger than those at FP1, proving more variable characteristic of FP3.Compared with Kemian 1, the decreased amplitude of cotton yield and fiber quality traits under late planting and shading was larger in Sumian 15, which also con firmed our previous study that Sumian 15 is more responsive to planting date and shading (Lvet al. 2013; Chenet al. 2014a; Maet al.2014; Liuet al. 2015a).

4.2. Effects of planting dates and shading on sourcesink and their relationship

Source, sink strength, and their relationship are crucial for cotton yield and fiber quality (Hu 2007; Kuaiet al. 2014). In developing cotton fiber, the maximum sucrose content can reflect the amount of available sucrose, while the minimum sucrose content is the residual sucrose content in mature fiber, and sucrose transformation rate indicates sucrose transformation capacity during cotton fiber development(Shuet al. 2009). In this study, sucrose content in leaf and fiber as well as cellulose content in fiber were all significantly lower at FP3 than those at FP1. Consistent with previous studies (Pettigrew 2001; Shuet al. 2009), shading significantly decreased sucrose content in subtending leaf and fiber at both FPs, but planting dates increased sucrose content in subtending leaf and fiber at both FPs.Although the effects of shading and planting date on fiber and subtending leaf sucrose content were not consistent,both decreased sucrose transformation rates in leaf and fiber, and fiber cellulose content. The reasons might be: (1)under shading treatment, decreased sucrose content altered the further metabolism of carbohydrates into endpoint fiber structural units or other compounds, which could further induce decreases in cellulose syntheses (Pettigrew 2001);or (2) increased accumulation of sucrose might be related to plant responses to adverse conditions (Guyet al. 1992;Savitchet al. 1997); low temperature stress caused by late planting decreased sucrose inversion rate and subsequently cellulose content. In addition, CVs of fiber sucrose and cellulose under LPD-CRLR60% were greater than those under LPD-CRLR100% or OPD-CRLR60%, indicating that effect of planting date plus shading has a greater effect than the individual effects of temperature or light, which corroborates the results of Chenet al. (2003).

In the present study, key physiological factors involved in subtending leaf and fiber development changed under different planting dates and shading treatments, consequently changing yield and fiber quality. The comparison between FP1 and FP3 on the same fruiting branch showed that the variation of these factors was similar under late planting and shading treatments, but variation amplitude and variation coef ficient were higher at FP3. Meanwhile, sucrose content, cellulose content, cotton yield, and fiber quality were also lower at FP3. The results might be related to adverse temperature and light condition during the boll development period at FP3 and less carbohydrate synthesis and translocation at FP3. Although the boll primarily received carbohydrates synthesized by the subtending leaf, the main stem leaf also supplies carbohydrate for fruit development.Fruit produced closer to the main stem can receive more carbohydrates from the main stem leaf than fruit produced at more distal positions. Thus, compared with FP1, bolls at FP3 experienced more serious temperature-light stress and received a reduced supply of carbohydrates, which hindered sucrose metabolism and cellulose synthesis. In addition,compared with FP1, correlation indices between sucrose content, cotton yield parameters, and micronaire, as well as correlation indices between cellulose content and fiber strength were higher at FP3, further showing that bolls at FP3 were more sensitive to adverse temperature and light regimes. Both cultivars exhibited variability in carbohydrate contents of leaves and fiber between FP1 and FP3 under the combinations of different planting dates and shading treatments, which might be the reason for higher sensitivity of lint yield distribution and fiber quality of temperature-sensitive cultivar Sumian 15 to late planting and shading compared to temperature-tolerant cultivar Kemian 1. These results highlight the importance of retaining the bolls at FP3 and could provide further data for studying differences in sucrose metabolism between fruiting positions and would be valuable for cotton cultivators to improve cotton yield and fiber quality under low temperature and low light on distal FPs.

5. Conclusion

Cotton yield and yield components, fiber quality, sucrose transformation rate of leaf and fiber, and fiber cellulose content were all decreased at FP3 compared to FP1 under all treatments. OPD-CRLR60%, LPD-CRLR100%, and LPD-CRLR60% led to significant decreases in lint yield at both FPs and for both cultivars, especially on FP3 and in Sumian 15; this effect was mainly caused by a rapid decline in boll number. Except for the increase of UHML on FP1 under OPD-CRLR60%, all fiber quality indices were decreased under late planting and shading, and a greater reduction was observed on FP3 and in Sumian 15. LPD increased, but shading decreased, sucrose content in leaf and fiber as compared to OPD with no shading, which led to a decreased cellulose concentration, indicating that late planting mainly diminished the transfer of leaf sucrose to cotton fiber and shading primarily decreased the sucrose content from the source. Additionally, the changes in carbohydrates at FP3 and in Sumian 15 were more variable than at FP1 and in Kemian 1. These results suggested that higher reduction in cotton yield and fiber quality at FP3 and in Sumian 15 under late planting and shading treatments are mainly associated with significant decreases in boll number and carbohydrate contents.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (31271654, 31401327, 31471444),the Special Fund for Agro-scientific Research in the Public Interest, China (201203096), and the Jiangsu Overseas Research & Training Program for University Prominent Young & Middle-aged Teachers and Presidents, China(2016).

Ahmadi A, Joudi M, Janmohammadi M. 2009. Late defoliation and wheat yield: Little evidence of post-anthesis source limitation.Field Crops Research, 113, 90–93.

Anjum R, Soomro A, Bano S, Chang M, Leghari A. 2002.Fruiting position impact on seedcotton yield in American cotton.Asian Journal of Plant Sciences, 1, 153–155.

Arshad M, Wajid A, Maqsood M, Hussain K, Aslam M, Ibrahim M. 2007. Response of growth, yield and quality of different cotton cultivars to sowing dates.Pakistan Journal of Agriculturl Science, 44, 2.

Ashley D A. 1972.14C-Labelled photosynthate translocation and utilization in cotton plants.Crop Science, 12, 69–74.

Bondada B R, Oosterhuis D M. 2001. Canopy photosynthesis,specific leaf weight, and yield components of cotton under varying nitrogen supply.Journal of Plant Nutrition, 24,469–477.

Bozbek T, Sezener V, Unay A. 2006. The effect of sowing date and plant density on cotton yield.Journal of Agronomy, 5,122–125.

Cao T, Oumarou P, Gawrysiak G, Klassou C, Hau B. 2011.Short-season cotton (Gossypium hirsutum) may be a suitable response to late planting in sub-Saharan regions.Field Crops Research, 120, 9–20.

Chen J, Lv F, Liu J, Ma Y, Wang Y, Chen B, Meng Y, Zhou Z.2014a. Effects of different planting dates and low light on cotton fiber length formation.Acta Physiologiae Plantarum,36, 2581–2595.

Chen J, Lv F, Liu J, Ma Y, Wang Y, Chen B, Meng Y, Zhou Z,Oosterhuis D M. 2014b. Effect of late planting and shading on cellulose synthesis during cotton fiber secondary wall development.PLoS ONE, 9, e105088.

Chen Q, Zhang F, Wang Y, Kenji K. 2003. The physiologic reaction of cucumber to low temperature and low light intensity.Agricultural Sciences in China, 2, 200–205.

Davidonis G H, Johnson A S, Landivar J A, Fernandez C J.2004. Cotton fiber quality is related to boll location and planting date.Agronomy Journal, 96, 42–47.

Delmer D P, Amor Y. 1995. Cellulose biosynthesis.The Plant Cell, 7, 987–1000.

Delmer D P, Haigler C H. 2002. The regulation of metabolic flux to cellulose, a major sink for carbon in plants.Metabolic Engineering, 4, 22–28.

Dong H, Li W, Tang W, Li Z, Zhang D, Niu Y. 2006. Yield, quality and leaf senescence of cotton grown at varying planting dates and plant densities in the Yellow River Valley of China.Field Crops Research, 98, 106–115.

Dong H, Xin C, Li W, Tang W, Zhang D. 2010. Late planting of short-season cotton in saline fields of the Yellow River Delta.Crop Science, 50, 292–300.

Dusserre J, Crozat Y, Warembourg F R, Dingkuhn M. 2002.Effects of shading on sink capacity and yield components of cotton in controlled environments.Agronomie, 22, 307–320.

Echer F R, Rosolem C A. 2015. Cotton yield and fiber quality affected by row spacing and shading at different growth stages.European Journal of Agronomy, 65, 18–26.

Gao X, Wang Y, Zhou Z, Oosterhuis D M. 2012. Response of cotton fiber quality to the carbohydrates in the leaf subtending the cotton boll.Journal of Plant Nutrition and Soil Science, 175, 152–160.

Grindlay D J C. 1997. Towards an explanation of crop nitrogen demand based on the optimization of leaf nitrogen per unit leaf area.Journal of Agricultural Science, 128, 377–396.

Gu L, Wang X, Zhou Z, Chen D, Xu L, Zhou R, Liu Y. 2010.Researches of high yield cotton cultivations in Jiangsu province.China Cotton, 37, 14–16.(in Chinese)

Guy C L, Huber J L, Huber S C. 1992. Sucrose phosphate synthase and sucrose accumulation at low temperature.Plant Physiology, 100, 502–508.

Heitholt J. 1997. Floral bud removal from specific fruiting positions in cotton: Yield and fiber quality.Crop Science,37, 826–832.

Hendrix D L. 1993. Rapid extraction and analysis of nonstructural carbohydrates in plant tissues.Crop Science,33, 1306–1311.

Hu H B. 2007. Genotypic differences in the changing of the matters involved in cotton fiber thickening development and its relationship to fiber strength. MSc thesis, Nanjing Agricultural University, China. (in Chinese)

Hu W, Ma Y, Lv F, Liu J, Zhao W, Chen B, Meng Y, Wang Y, Zhou Z. 2016. Effects of late planting and shading on sucrose metabolism in cotton fiber.Environmental and Experimental Botany, 131, 164–172.

Iqbal M, Ahmad S. 2003. Effect of different sowing dates on cotton (Gossypium hirsutumL.) cultivar.Asian Journal of Plant Science, 2, 461–463.

Jiang G, Meng Y, Chen B, Bian H, Zhou Z G. 2006. Effects of low temperature on physiological mechanisms of cotton fiber strength forming process.Journal of Plant Ecology,30, 335–343.(in Chinese)

Jenkins J, McCarty J, Parrott W. 1990a. Effectiveness of fruiting sites in cotton: Yield.Crop Science, 30, 365–369.

Jenkins J, McCarty J, Parrott W. 1990b. Fruiting efficiency in cotton: Boll size and boll set percentage.Crop Science,30, 857–860.

Kuai J, Liu Z, Wang Y, Meng Y, Chen B, Zhao W, Zhou Z,Oosterhuis D M. 2014. Waterlogging during flowering and boll forming stages affects sucrose metabolism in the leaves subtending the cotton boll and its relationship with boll weight.Plant Science, 223, 79–98.

Liu J, Ma Y, Lv F, Chen J, Zhou Z, Wang Y, Abudurezike A,Oosterhuis D M. 2013. Changes of sucrose metabolism in leaf subtending to cotton boll under cool temperature due to late planting.Field Crops Research, 144, 200–211.

Liu J, Meng Y, Chen B, Zhou Z, Ma Y, Lv F, Chen J, Wang Y.2015a. Photosynthetic characteristics of the subtending leaf and the relationships with lint yield and fiber quality in the late-planted cotton.Acta Physiologiae Plantarum, 37, 79.

Liu J, Meng Y, Chen J, Lv F, Ma Y, Chen B, Wang Y, Zhou Z,Oosterhuis D M. 2015b. Effect of late planting and shading on cotton yield and fiber quality formation.Field Crops Research, 183, 1–13.

Lu G, Dai J, Li W, Tang W, Zhang D, Eneji A E, Dong H. 2017.Yield and economic bene fits of late planted short-season cotton versus full-season cotton relayed with garlic.Field Crops Research, 200, 80–87.

Lv F, Liu J, Ma Y, Chen J, keyoumu Abudurezikekey A, Wang Y, Chen B, Meng Y, Zhou Z. 2013. Effect of shading on cotton yield and quality on different fruiting branches.Crop Science, 53, 2670–2678.

Ma Y, Wang Y, Liu J, Lv F, Chen J, Zhou Z. 2014. The effects of fruiting positions on cellulose synthesis and sucrose metabolism during cotton (Gossypium hirsutumL.) fiber development.PLoS ONE, 9, e89476.

Pettigrew W. 1995. Source-to-sink manipulation effects on cotton fiber quality.Agronomy Journal, 87, 947–952.

Pettigrew W. 1996. Low light condition compromise the quality of fiber produced. In: Dugger P, Richter D A, eds.,Proceeding of the Beltwide Cotton Conferences.National Cotton Council of America, USA. pp. 1238–1239.

Pettigrew W. 2001. Environmental effects on cotton fiber carbohydrate concentration and quality.Crop Science, 41,1108–1113.

Roussopoulos D, Liakatas A, Whittington W. 1998. Cotton responses to different light-temperature regimes.Journal of Agricultural Science, 131, 277–283.

SassenrathCole G, Lu G, Hodges H, McKinion J. 1996.Photon flux density versus leaf senescence in determining photosynthetic efficiency and capacity ofGossypium hirsutumL. leaves.Environmental and Experimental Botany, 36, 439–446.

Savitch L, Gray G, Huner N. 1997. Feedback-limited photosynthesis and regulation of sucrose-starch accumulation during cold acclimation and low-temperature stress in a spring and winter wheat.Planta, 201, 18–26.

Shu H, Zhou Z, Xu N, Wang Y, Zheng M. 2009. Sucrose metabolism in cotton (Gossypium hirsutumL.) fibre under low temperature during fibre development.European Journal of Agronomy, 31, 61–68.

Updegraff D M. 1969. Semimicro determination of cellulose in biological materials.Analytical Biochemistry, 32, 420–424.

Wang Q, Wang Z, Song X, Li Y, Guo Y, Wang J, Sun X. 2005.Effects of shading at blossoming and boll-forming stages on cotton fiber quality.Chinese Journal of Applied Ecology,16, 1465–1468. (in Chinese)

Wang Y, Shu H, Chen B, McGiffen Jr M, Zhang W, Xu N, Zhou Z. 2009. The rate of cellulose increase is highly related to cotton fiber strength and is significantly determined by its genetic background and boll period temperature.Plant Growth Regulation, 57, 203–209.

Williamson R E, Burn J E, Hocart C H. 2002. Towards the mechanism of cellulose synthesis.Trends in Plant Science,7, 461–467.

Wullschleger S D, Oosterhuis D M. 1990. Photosynthetic carbon production and use by developing cotton leaves and bolls.Crop Science, 30, 1259–1264.

Yeates S J, Constable G A, McCumstie T. 2010. Irrigated cotton in the tropical dry season. III: Impact of temperature, cultivar and sowing date on fibre quality.Field Crops Research,116, 300–307.

Zhao D, Oosterhuis D. 1994. Effects of shading on cotton photosynthesis, yield and yield components. In: Oosterhuis D M, ed.,Proceeding of Arkansas Cotton Research Meeting.Special Report.University of Arkansas, Fayetteville, USA.pp. 131–135.

Zhao D, Oosterhuis D. 2000. Cotton responses to shade at different growth stages: Growth, lint yield and fibre quality.Experimental Agriculture, 36, 27–39.

Zhao W, Wang Y, Shu H, Li J, Zhou Z. 2012. Sowing date and boll position affected boll weight, fiber quality and fiber physiological parameters in two cotton (Gossypium HirsutumL.) cultivars.African Journal of Agricultural Research, 7, 6073–6081.