Carbohydrates accumulation, oil quality and yield of rapeseed genotypes at different nitrogen rates

ABSTRACT

A field experiment was conducted in 2018–2020 to investigate the effect of two nitrogen application rates (N₀ = 0 kg ha⁻¹(control), N₉₀ = 90 kg ha⁻¹, and N₁₈₀ = 180 kg ha⁻¹) on morphological and physiological behaviours of two new rapeseed varieties; V₁ = JYJS01 (semi-dwarf variety) and V₂ = CY36 (tall variety). Our results indicated that N application enhanced the plant growth and seed protein content in comparison to control. Contrastingly, compared with control, sugar contents, i.e. sucrose and reducing sugars, decreased significantly in response to an increase in N. Seed oil content also decreased by 8.5% and 5.5%, 6.7% and 3.9% against N₁₈₀ in V₁ and V₂ during 2019 and 2020, respectively. Furthermore, higher nitrogen application rates decreased seed fatty acid proportions by decreasing the sugar availability for fatty acid biosynthesis. Our results demonstrated the highest seed yield (9.4 g plant⁻¹ in 2018/19 and 9.5 g plant⁻¹ in 2019/20) in V₂ against N₁₈₀, and the lowest seed yield (3.2 g plant⁻¹ in 2018/19 and 3.4 g plant⁻¹ in 2019/20) in V₁ at control. These findings imply that a high nitrogen application rate had increased the yield. At the same time, the carbohydrates translocation from stems to seed for fatty acid biosynthesis decreased, which played their significant role in seed physiology for fatty acid metabolisms, ultimately affected the seed quality. However, there is still a dire need to evaluate how nitrogen application affect the carbohydrate availability for fatty acid metabolism in the seed of new winter rapeseed varieties.

Graphical abstract

KEYWORDS:

• Rapeseed
• nitrogen
• carbohydrates
• proteins
• oil contents

1. Introduction

The current human population is 7.5 billion, which is expected to rise by 9 billion by 2050, which will cause a significant threat to global food security (Mahajan & Tuteja, 2005). Therefore, high-yielding cereals and oilseed crops are s a significant challenge for the farming community to meet the increasing global food demand. Rapeseed (Brassica napus L) is one of the major oilseed crops consumed widely globally as its oil production per hectare is double compared to soybean (Shahidi, 1990). China is the second-largest producer of rapeseed after Canada (Bonjean et al., 2016). Rapeseed is the fourth largest cultivated crop in China after rice, maize and wheat, with ‘The Yangtze River Basin’ being the largest rapeseed producing belt where farmers use an intensive crop production system to obtain higher yields (X. Li et al., 2018). The annual rape output accounts for more than 80% of the total national output. Its oil serves as a vital source of edible oil in the human diet. Moreover, rapeseed oil demand has also increased as a renewable energy source in recent years (A. Ahmad et al., 2011).

Fertilizers are commonly used to boost crop production. Depending on the soil fertility and the type of crop, fertilizers are mainly used to provide the necessary amount of plant macronutrients, namely nitrogen, phosphorus, and potassium (Queirós et al., 2015). However, its efficiency depends upon the rate and timings of fertilizer application and the cultivar under consideration. Nitrogen (N) is one of the essential macronutrients as it plays a vital role in crop growth and development and yield production (Bouchet, Stahl et al., 2016b; Leghari et al., 2016). It is an essential structural component of chlorophyll and is involved in various physiological processes (Bouchet et al., 2016b). N is a highly demanding nutrient for rapeseed growth and development as its production is highly affected by available nitrogen (Öztürk, 2010). Also, in comparison to cereals, it has a higher critical N demand; therefore, applying the correct rate of N fertilizer at the appropriate time is an essential aspect of successful rapeseed production.

Seed oil content, protein and fatty acids in rapeseed are considered essential indexes of the seed quality. Earlier research has demonstrated that carbohydrates in sugars play a crucial role in rapeseed oil quality (Ni et al., 2019). The sugars (sucrose and reducing sugars) can be translocated from stem to seed, where they are the chief source of carbon (sucrose) for fatty acid metabolisms as they are converted into fatty acid during lipid biosynthesis in oilseed, thus affecting the seed quality of rapeseed ultimately (Khan et al., 2018a). Higher rates of N application are known to result in decreased translocation of carbohydrates from stems to seeds, which reduces the carbon source for fatty acid metabolism in seeds and change the fatty acid composition, i.e., enhanced palmitic and stearic acid content and decrease linoleic and linolenic acid content (Khan et al., 2018a). However, how varying rates of N application affect the carbohydrate availability for fatty acid metabolism in winter rapeseed varieties (both tall and semi-dwarf) still needs further investigation.

The N has been documented as an essential nutrient for obtaining the maximum potential of rapeseed yield (Tian et al., 2021). Although yield responses of rapeseed to increasing doses of N fertilizer vary with various environmental variables such as weather, soil type, residual fertility (especially nitrate), soil water content, as well as the cultivar; however, many studies have reported enhanced growth and yield of rapeseed in response to high doses of applied N (G. Ahmad et al., 2007; Hocking et al., 1997; Khan et al., 2018b; Rathke et al., 2005). It has been reported that N enhances the rapeseed yield by influencing several growth parameters, i.e. number of branches and pods per plant, seeds per pod and thousand seed weight (Öztürk, 2010). When applied in excess, N results in considerably reduced seed yield (Khan et al., 2018b). It is known that an optimum dose of N nutrient increases the protein content, but when applied in excess, it enhances N content in the seed and also decreases the oil content and thus the commercial value of rapeseed (G. Ahmad et al., 2007). Furthermore, excessive amounts of N application are also heavy on farmer’s pockets and pollute the environment. Since there are only a few detailed studies on the carbohydrates metabolism, oil quality and yield of rapeseed varieties in response to different N application levels; therefore, we have researched intending to assess the changes in carbohydrates accumulation, oil quality and yield of two newly introduced winter rapeseed varieties JYJS01 (semi-dwarf variety) and CY36 (tall variety) in response to different N application levels, i.e. N₀ = 0 kg N ha⁻¹(control), N₉₀ = 90 kg N ha⁻¹and N₁₈₀ = 180 kg N ha⁻¹.

2. Material and methods

2.1. Plant materials

The two important varieties, semi-dwarf; JYJS01 (V₁) and tall; CY36 (V₂) are selected because of their higher seed yield in China’s Sichuan Province. The variety JYJS01 (V₁) was supplied by the Kele rapeseed research and extension limited company of Sichuan Province, and the variety CY36 (V₂) was supplied by the crop research institute of Sichuan Academy of Agricultural Sciences, China.

2.2. Experimental site and design

The two-year field experiment was conducted at the site of Chengdu Plain of the upper reaches of Yangtze River Basin in Sichuan province of China (102°54 E, 104°53 E, 30°05 N, 31°26 N). It is a subtropical region with an annual total rainfall of 1775 mm, an average temperature of 16.2°C and an annual total sunshine duration of 1050 h. During the winter season, the day-to-day average rainfall is generally less than 150 mm, and the temperature is above 0°C from December to February. The growing season climate, including the monthly precipitation and mean temperature, is illustrated (Figure 1). The experimental site soil with clay loam texture 54% sand, 28% silt and 17% clay (hydrometer method), with total nitrogen 0.9 g kg⁻¹ by Kjeldahl acid digestion method (Keeney, 1982; Sparks et al., 2020), total phosphorous 0.6 g kg⁻¹ (Sparks et al., 2020), total potassium 6.3 g kg⁻¹ (Huoyan et al., 2016), and available nitrogen 63.5 mg kg⁻¹ automatic discontinuous analyzer (Clever chem200, Germany), available phosphorous 40.6 mg kg⁻¹, accessible potassium 96.4 mg kg⁻¹ (Jackson, 2005), and organic matter 30.3 g kg⁻¹ in the top 0–25 cm soil layer and pH 6.6 (Pietsch & Mabit, 2012).

Figure 1. Monthly rainfall and the mean temperature during the growing seasons 2018–2020 from September- May.

The experiment was carried out in two successive years, 2018–2020. The field experiment was designed as a split-plot with two N application rates as the main plots with three replications and two cultivars as the subplots with three replications. The area of the subplot was measured as 9 × 2 m. The total area of plots was fertilized with 90 kg P₂O₅ ha⁻¹, and the source of P₂O₅ is CaH₆O₉P₂ and 90 kg K₂O ha⁻¹and, the source of K2O, is K₂SO₄ before sowing. Seeds were sown to the sub-plot with the density of 3 kg seeds ha⁻¹. The R × R (row to row) distance was 30 cm, hill spacing was 20 cm, and two plants were reserved in each hole. The sowing was done on 29 September 2018. The plant’s initial flowering stage and ending flowering stage was on March 1 and March 31, 2019, respectively, and harvested on April 30 of 2019. The sowing date in 2019 was on October 11 and, the plant initial flowering stage and ending flowering stage were on February 24 and March 24, 2020, respectively, and were harvested on April 24 of 2020. The three Urea fertilizer treatments were given as N₀ = 0 kg N ha⁻¹(control); N₉₀ = 90 kg N ha⁻¹; N₁₈₀ = 180 kg N ha⁻¹. The total amount of fertilizer was split into two stages, one was applied as basal fertilization before sowing, and the other was top-dressed in late November at the 6-leaf stage.

2.3. Sampling and measurements

2.3.1. Data collections

The plants were harvested in four separate sets at vegetative (January 4^th), flowering (March 1^st), pod filling (March 31^st) and maturity stages (April 30). The plant height was determined (at three phenological stages vegetative, flowering and pod filling) by measuring the distance from cotyledon nodes to the top of the plant. Harvesting of the plants was carried out when two-thirds of the seeds were brown. Six plant samples were randomly selected from each plot and steadily rooted to examine the morpho-physiological components for further experiments. Afterwards, the yield components of these selected plants were measured at maturity, including the seed number per pod, pod number per plant, and the seed yield per plant. Primary branch height was examined (at maturity) by measuring the distance from cotyledon nodes up to the first branch, and the main inflorescence length was determined (at maturity) by measuring the distance from the uppermost branches up to the top of the plant.

2.3.2. Pod area

The pod area of rapeseed plants of 20 plants in each plot was determined at the pod filling stage by following the previously published study (Zhou & Xi, 1993).

The formula for measuring the pod area was:

${\mathrm{S}}_{\mathrm{a}}\left(\mathrm{m}{\mathrm{m}}^2\right)=\pi \mathrm{d}{\mathrm{h}}_{1}1/3\pi \mathrm{d}{\mathrm{h}}_2$

(Where h1 = 0.8 H, h2 = 0.2 H, H is pod length and d is pod width).

2.3.3. Seed quality and oil contents

The seed quality was evaluated at S₃₀ (30 days of post-flowering). Near-infrared reflectance spectroscopy (NIRS; Foss NIR Systems Inc., U.S.A.) was used to evaluate the seed quality by following the standard procedure (Mika et al., 2014).

2.3.4. Total chlorophyll contents of leaves and pod wall

Total leaf and pod wall chlorophyll contents were extracted by grinding samples in 10 mL of 80% acetone and centrifuged at 1000 rpm for 3 min at 4°C. Then, these samples were placed in 10 mL of 80% aqueous acetone solution in the darkness for 24 h at room temperature. The supernatant was then separated and analyzed with a spectrophotometer (SpectraMax i3x from Austria) at wavelengths of 470, 645, and 663 nm.

The formula for measuring the total leaf and pod wall chlorophyll contents was:

Chlorophyll a (mg/g) = 12.7 (D 663) −2.69 (D 645) V/ (1000 × W)

Chlorophyll b (mg/g) = 22.9 (D 645) −4.68 (D 663) V/ (1000 × W

Total chlorophyll content = (mg/g) D652 × V/ (34.5 × W)/Chla + Chlb

Where the D, W, and, V represent the absorbance value, fresh weight of sample and volume of acetone respectively. The total chlorophyll contents were determined at vegetative, flowering and pod filling stage from all experimental plots in each replication.

2.3.5. Total nitrogen

The fresh samples were oven-dried at 105°C to stop the enzymatic activities for one hour, and then it was kept at 80°C to obtain the constant dry weight. Afterwards, the samples were ground, and the Kjeldahl method was used to determine the nitrogen content of the plant (Keeney, 1982; Sparks et al., 2020).

2.3.6. Extraction for sugars determination

The fresh stems and pod wall samples were taken and oven-dried at 105°C for 30 minutes and kept overnight at 80°C until a constant weight was gained to determine the sugar contents. Then, 0.1 g of ground powder of the dried sample was added in 6 mL of 80% ethanol. After that, the samples were boiled in the water bath at 80°C for 40 min and centrifuged at 5000 rpm for 5 min. Then the supernatant was poured into 50 mL tubes of the main solution. This method was repeated twice. After that, 80% ethanol was added into the main solution tube to the volume up to 50 mL. Later on, 0.1 g activated carbon (charcoal) was added to each 50 mL test tubes and kept overnight until the green color disappeared. This solution was filtered and used for the following experiments (Asghar et al., 2020).

2.3.6.1. Sucrose determination

For sucrose content measurements, a 0.9 mL extracted sample was added to the test tube and then added 0.1 mL of 2 M NaOH and heated for about 10 min in the water bath. After heating, samples were kept at room temperature for 15 minutes to cool down. Then, 1 mL of 0.1% resorcinol and 3 mL of 10 M HCL were added to the solution, and the samples were heated at 80°C for 10 min. In the end, the supernatant was taken to read the absorbance at 480 nm by using a spectrophotometer (SpectraMax i3x from Austria).

2.3.6.2. Reducing sugars

Reducing sugars were evaluated by mixing 1.5 mL of the extracted solution, 0.5 mL dd H₂O and 1.5 mL DNS solution in 10 mL test tubes. These tubes were then kept in the water bath at 80°C for 10 min. Later on, the supernatant was taken to read the absorbance at 520 nm using a spectrophotometer (SpectraMax i3x from Austria).

2.4. Statistical analyses

All data were collected and analyzed by the analysis of variance (ANOVA) in conjunction with Duncan’s multiple range test to identify significant differences between treatment levels and combinations of treatments at p < 0.05. Analyses were carried out using the IBM SPSS Statistics 21.0 and Microsoft Excel 2016, and Origin 8.5 were used for graphical presentation.

3. Results

3.1. Morphological and agronomic parameters

In the current findings, the effect of different nitrogen application rates on the morphological and agronomic parameters of two rapeseed cultivars was evaluated. As depicted in Table 1, both cultivars’ average primary branch height was enhanced by 51.3% in 2018/19 and 59.7% in 2019/20 at N180 compared to N₀. Furthermore, the main inflorescence length of both cultivars was increased by 28.1% and 31.4% at N₁₈₀ over N₀ in 2018/19 and 2019/2020, respectively. Moreover, the number of pods and seed yield was enhanced with the increasing nitrogen rate. The maximum number of pods (237.3 and 240.9) was noted in V₂ at N₁₈₀ compared to N₀ in 2018/19 and 2019/20, respectively. Similarly, the seed yield was also recorded maximum (9.4 g Plant⁻¹ and 9.5 g Plant⁻¹) in V₂ at N₁₈₀ compared with N₀ in 2018/19 and 2019/20, respectively. The mean stem diameter was also increased by 32.4% and 25.2% at N₁₈₀ compared to N₀ in 2018/19 and 2019/20, respectively. The branch number and seed number per pod were amplified significantly with the increasing nitrogen rate in both years (2018–2020). It was also observed that the V₁ gained more pod surface area, i.e., 704 mm² in 2018/19 and 769.3 mm² in 2019/20 at N_90, as compared to N_0. Our results revealed that at N_180, the V₂ had obtained a higher number of seeds per pod than V_1.

Table 1. Effect of different nitrogen rates (N₀=0kg N ha⁻¹; N₉₀=90kg N ha⁻¹; N₁₈₀=180kg N ha⁻¹) on morphological parameters of two varieties (V₁=JYJS01 (semi-dwarf), and V₂=CY36 (tall height) of rapeseed during the growing periods of 2018/19 and 2019/20

Year		Treatment	Varieties	Primary branch height (cm)	Main inflorescence length (cm)	Pod number Plant^−1	Stem diameter (mm)	Branch number Plant^−1	Seed number Pod^−1	Pod surface area (mm^2)	Seed weight (g/plant)
2018/19	Varietal difference	N0	V1	63.0c	41.0c	50.7c	10.3b	2.3c	18.7b	457.3b	3.2b
			V2	62.7c	37.3a	56.3c	11.4b	2.7b	20.7c	611.0b	4.0c
		N90	V1	73.7b	48.7b	138.0b	12.3ab	4.3b	23.0a	704.0a	4.7b
			V2	73.7b	39.9a	157.0b	13.0ab	5.0a	23.3b	641.0a	5.9b
		N180	V1	84.33a	54.0a	220.7a	13.3a	5.7a	25.0a	666.7a	8.2a
			V2	105.7a	46.3a	237.3a	15.3a	5.7a	26.0a	545.0a	9.4a
		Treatment difference	N0	62.8c	39.2b	53.5c	10.8b	2.50c	19.7c	534.2b	3.6c
			N90	73.7b	44.3ab	147.5b	12.7ab	4.7b	23.2b	672.5a	5.3b
			N180	95.0a	50.2a	229.0a	14.3a	5.7a	25.5a	605.8a	8.8a
			T	***	***	***	**	***	***	***	***
			V	***	**	ns	ns	ns	*	ns	*
2019/20	Varietal difference	N0	V1	62.9c	42.5b	52.4c	11.0b	2.2b	20.1b	429.2b	3.4b
			V2	59.3b	38.8a	60.8b	12.1a	2.3b	21.5b	614.8a	4.1c
		N90	V1	76.9b	50.0b	146.8b	12.9ab	4.9a	21.6b	769.3a	4.7b
			V2	74.8b	46.0a	158.8ab	13.6a	5.1a	23.8ab	660.4a	6.1b
		N180	V1	89.7a	59.2a	229.7a	13.5a	5.7a	25.5a	675.7ab	8.8a
			V2	105.4a	47.8a	240.9a	15.3a	5.8a	25.7a	559.2a	9.5a
		Treatment difference	N0	61.1c	40.7b	56.6c	11.5	2.3b	20.8b	521.4b	3.7c
			N90	75.8b	48.0a	152.8b	13.2ab	5.0a	22.7b	714.9a	5.4b
			N180	97.6a	53.5a	235.3a	14.4a	5.8a	25.6a	617.3ab	9.2a
			T	***	**	***	ns	***	**	ns	***
			V	ns	*	ns	ns	ns	ns	ns	ns

Values in a column followed by the same letter are not significantly different at P<0.05 as determined by the LSD test; n=6.The ns, *, ** and *** represent the non-significant, significant at P<0.05, 0.01, 0.001, respectively.

3.2. Plant height and biomass accumulation

Different nitrogen treatments (N₀, N₉₀, and N₁₈₀) significantly affected (P < 0.05) the plant height and biomass accumulation at different growth stages (Figures 2 and 3). The plant height at the pod filling stage was measured in both varieties. The plant height of three replication of both varieties at N₁₈₀ was 180.5 cm and 179.5 cm in 2018/19 − 2019/20, respectively. At N₁₈₀, biomass accumulation of both varieties was observed at the maturity stage, and these were 6699.8 kg ha⁻¹ and 6850.0 kg ha⁻¹ in 2018/19 − 2019/20, respectively. At the maturity stage, the biomass accumulation was higher at N₁₈₀ than N₀.

Figure 2. Effect of different nitrogen application rates, N₀ = 0 kg N ha⁻¹(control); N₉₀ = 90 kg N ha⁻¹; N₁₈₀ = 180 kg N ha⁻¹ at vegetative, flowering, and pod filling stage on plant height of two varieties, (a) V₁ = JYJS01 (semi-dwarf), and (b) V₂ = CY36 (tall height) in the years of 2018/19 and 2019/20. The ns, *, ** and *** represent the non-significant, significant at P < 0.05, 0.01, 0.001, respectively. Values on the bars followed by the same letter are not significantly different at P < 0.05 as determined by the LSD test; n = 3.

Figure 3. Effect of different nitrogen application rates, N₀ = 0 kg N ha⁻¹(control); N₉₀ = 90 kg N ha⁻¹; N₁₈₀ = 180 kg N ha⁻¹ on the biomass accumulation of two varieties, (a) V₁ = JYJS01 (semi-dwarf), and (b) V₂ = CY36 (tall height) at vegetative, flowering, pod filling stage and maturity stage The ns, *, ** and *** represent the non-significant, significant at P < 0.05, 0.01, 0.001, respectively. Values on the bars followed by the same letter are not significantly different at P < 0.05 as determined by the LSD test; n = 3.

Among N₀, N₉₀ and N_180, the highest plant height and biomass accumulation were observed at N₁₈₀ in both years. The rapeseed plant exhibited the plant height and biomass accumulation trend as N₁₈₀ > N₉₀ > N₀ (Figures 2 and 3). In this study, no significant difference was found between the two varieties for plant height but only at the pod filling stage in 2018/19, while the significant varietal difference for biomass accumulation was observed between the two varieties except at the maturity stage in 2019/20 (Figure 3b). In the treatment N₁₈₀, the plant height (171.3 cm and 175.1 cm) and biomass accumulation (6321.1 kg ha⁻¹ and 6589.0 kg ha⁻¹) were obtained relative to N₀ in V₁ at the pod filling stage in 2018/19 and 2019/20, respectively (Figures 2a and 3a). Similarly, plant height (189.7 cm and 184.0 cm) and biomass accumulation (7078.5 kg ha⁻¹ and 7111.0 kg ha⁻¹) were obtained at N₁₈₀ compared to N₀ in V₂ at the pod filling stage in 2018/19 and 2019/20, respectively (Figures 2b and 3b).

3.3. Leaf and pod wall chlorophyll contents

The leaf and pod wall chlorophyll content of N_0, N₉₀ and N₁₈₀ is shown in Figures 4 and 5. At all sampling stages (vegetative, flowering and pod filling), the results for leaf and pod wall chlorophyll content revealed that different nitrogen treatments had significant (P < 0.05) effect on leaf and pod wall chlorophyll content of rapeseed plants during both years of experiments (Figures 4 and 5). The maximum chlorophyll content was always noted in N₁₈₀. Moreover, as depicted in Figures 4 and 5, the leaf chlorophyll content was increased with an increase in nitrogen rate except at the vegetative stage in V₂, while the pod wall chlorophyll content enhanced at both stages (flowering and pod filling).

Figure 4. Effect of different nitrogen treatments, N₀ = 0 kg N ha⁻¹(control); N₉₀ = 90 kg N ha⁻¹; N₁₈₀ = 180 kg N ha⁻¹ at vegetative, flowering and pod filling stage on leaf chlorophyll content of two varieties, (a) V₁ = JYJS01 (semi-dwarf), and (b) V₂ = CY36 (tall height) during the years of 2018/19 and 2019/20. The ns, *, ** and *** represent the non-significant, significant at P < 0.05, 0.01, 0.001, respectively. Values on the bars followed by the same letter are not significantly different at P < 0.05 as determined by the LSD test; n = 3.

Figure 5. Effect of different nitrogen treatments, N₀ = 0 kg N ha⁻¹(control); N₉₀ = 90 kg N ha⁻¹; N₁₈₀ = 180 kg N ha⁻¹ at flowering and pod filling stage on pod wall chlorophyll content of two varieties, (a) V₁ = JYJS01 (semi-dwarf), and (b) V₂ = CY36 (tall height) during the years of 2018/19 and 2019/20. The ns, *, ** and *** represent the non-significant, significant at P < 0.05, 0.01, 0.001, respectively. Values on the bars followed by the same letter are not significantly different at P < 0.05 as determined by the LSD test; n = 3.

At all growth stages, the leaf chlorophyll content was maximum at the vegetative stage as compared to other stages. The leaf chlorophyll content was significantly decreased from vegetative to pod filling stage. The same trend was observed in pod wall chlorophyll contents, which was maximum at the flowering stage then declined at the pod filling stage. The maximum average leaf chlorophyll content of both varieties was recorded in treatment N₁₈₀ (1.7 mg/L and 1.7 mg/L at vegetative growth stage), whereas the minimum leaf chlorophyll content was measured (1.2 mg/L and 1.2 mg/L at pod filling stage) in N₀ in 2018/19 and 2019/20, respectively. Overall, the highest pod wall chlorophyll content value of both varieties was 1.2 mg/L and 1.2 mg/L in N₁₈₀ at the flowering stage, while the lowest value of both varieties was 0.3 mg/L and 0.3 mg/L in N₀ at the pod filling stage in 2018/19 and 2019/20, respectively. The varietal difference in leaf chlorophyll content was only significant at the flowering stage. In contrast, the varietal difference in pod wall chlorophyll content was significant in both years except at the flowering stage in 2018/19.

3.4. Total nitrogen percentages in leaves, stem, pod wall, and seed

In the current study, the total nitrogen percentages in leaves, stem, pod wall and seed were measured and the application of different treatments, N₀, N₉₀ and N₁₈₀, significantly (P < 0.05) affected leaves, stem, pod wall, and seed total nitrogen percentages (Figures 6, 7 and 8)., The application of high nitrogen treatment at all tested plant parts increased the total nitrogen percentages. In both varieties at treatment N₁₈₀, the maximum total nitrogen percentage was 2.2% and 2.8% in leaves at the flowering stage, whereas it was 1.6% and 1.7% in the stem at the vegetative stage as compared to N₀, in 2018/19 and 2019/20, respectively (Figures 6 and 7).

Figure 6. Effect of different nitrogen application rates, N₀ = 0 kg N ha⁻¹(control); N₉₀ = 90 kg N ha⁻¹; N₁₈₀ = 180 kg N ha⁻¹ at vegetative, flowering and pod filling stage on leaves nitrogen percentage of two varieties, (a) V₁ = JYJS01 (semi-dwarf), and (b) V₂ = CY36 (tall height) during the years of 2018/19 and 2019/20. The ns, *, ** and *** represent the non-significant, significant at P < 0.05, 0.01, 0.001, respectively. Values on the bars followed by the same letter are not significantly different at P < 0.05 as determined by the LSD test; n = 3.

Figure 7. Stem nitrogen percentage at different nitrogen treatments, N₀ = 0 kg N ha⁻¹(control); N₉₀ = 90 kg N ha⁻¹; N₁₈₀ = 180 kg N ha⁻¹ at vegetative, flowering, pod filling and maturity stage and on varieties, (a) V₁ = JYJS01 (semi-dwarf), and (b) V₂ = CY36 (tall height) in the years of 2018/19 and 2019/20. The ns, *, ** and *** represent the non-significant, significant at P < 0.05, 0.01, 0.001, respectively. Values on the bars followed by the same letter are not significantly different at P < 0.05 as determined by the LSD test; n = 3.

Figure 8. Effect of different nitrogen application rates, N₀ = 0 kg N ha⁻¹(control); N₉₀ = 90 kg N ha⁻¹; N₁₈₀ = 180 kg N ha⁻¹ on pod wall and seed nitrogen percentage of two varieties, (a) V₁ = JYJS01 (semi-dwarf), and (b) V₂ = CY36 (tall height) at maturity stage in the years of 2018/19 and 2019/20. The ns, *, ** and *** represent the non-significant, significant at P < 0.05, 0.01, 0.001, respectively. Values on the bars followed by the same letter are not significantly different at P < 0.05 as determined by the LSD test; n = 3.

On the other hand, in both varieties, total pod wall nitrogen percentage showed different behaviour. The varietal difference of pod wall nitrogen percentage was not significantly increased in 2018/19 compared to 2019/20. In contrast, it was augmented significantly at N₁₈₀ relative to treatment N₀ (Figure 8a and 8b). The total seed nitrogen percentage of both varieties was 2.6% and 2.4% at N₁₈₀ during 2018/19 (Figure 8a) and 2019/20 (Figure 8b) when compared with N₀, respectively. The varietal difference was highly significant at the flowering stage as compared to other plant developmental stages.

3.5. Seed composition

The differential nitrogen treatments had significantly altered the seed composition in the two tested cultivars (Table 2). According to the findings of this study, the seeds protein was enhanced significantly with the increasing nitrogen application rate. The total seed protein of both varieties increased by 19.2% and 24.3% at N₁₈₀ in 2018/19 and 2019/20 compared to N₀ treatment, respectively. The V₂ obtained the highest and the lowest protein content at N₁₈₀ and V₁ at N₀, respectively, in both years. The highest oil content was obtained by V₂ at N_0, and V₁ obtained at N₁₈₀ in both years. An inverse relation was found between the seed protein and oil content. The mean oil content declined significantly by 7.0% and 5.4% at N₁₈₀ compared to N₀ in 2018/19 and 2019/20, respectively.

Table 2. The consequence of different nitrogen treatments, N₀=0kg N ha⁻¹; N₉₀=90kg N ha⁻¹; N₁₈₀=180kg N ha⁻¹) and varieties, V₁ = JYJS01 (semi-dwarf) and V₂ = CY36 (tall height) in rapeseed on oil content, protein, glucosinolate, erucic acid and fatty acid composition during the growing seasons of 2018/19 and 2019/20

year

Treatment

Variety

Protein content (%)

Oil content (%)

Glucosinolate (%)

Palmitic acid (%)

Stearic acid (%)

Oleic acid (%)

Linoleic acid (%)

Linolenic acid (%)

Arachidionic acid (%)

Erucic acid (%)

2018/19

Varietal difference

N0

V1

17.5b

48.2a

49.8a

3.4a

1.4a

45.4b

15.5a

8.4b

6.6a

1.6a

V2

18.0b

49.5a

41.9a

3.5a

1.7a

62.0a

16.0a

8.5a

4.6a

1.5a

N90

V1

17.6b

48.1a

47.2a

3.8a

1.5a

49.7a

14.8a

8.8ab

7.0a

1.4a

V2

19.0b

48.9a

40.6a

3.6a

1.7a

65.2a

14.6a

8.2a

4.9a

1.4a

N180

V1

21.4a

44.1b

46.7a

3.9a

1.6a

50.5a

13.3b

9.7a

7.4a

1.4a

V2

21.34a

46.8a

38.7a

3.8a

1.9a

65.6a

13.4a

9.0a

5.5a

1.4a

Treatment difference

N0

17.7b

48.8a

45.8a

3.5a

1.5b

53.7a

15.8a

8.5b

5.6a

1.6a

N90

18.3b

48.5a

43.9a

3.7a

1.6ab

57.4a

14.7ab

8.50b

5.9a

1.4a

N180

21.12a

45.4b

42.7a

3.9a

1.75a

58.0a

13.3b

9.3a

6.5a

1.4a

T

**

**

ns

ns

ns

ns

*

*

ns

ns

V

ns

*

***

ns

***

***

ns

ns

*

ns

2019/20

Varietal difference

N0

V1

16.8b

47.6a

49.5a

3.6a

1.4a

44.2b

15.5a

8.2b

6.6a

1.8a

V2

17.8b

49.0b

41.0a

3.6a

1.7a

61.8a

16.0a

8.5a

4.5a

1.5a

N90

V1

17.8b

47.5ab

46.9a

3.6a

1.5a

49.1a

14.4a

8.8b

6.9a

1.4a

V2

19.4b

48.8b

63.0a

2.8a

1.0a

24.7a

9.9a

9.6a

15.3a

2.0a

N180

V1

21.6a

44.4b

44.1a

3.9a

1.6a

50.3a

12.9b

9.7a

7.6a

1.4a

V2

21.4b

47.1a

38.0a

3.8a

1.9a

67.1a

13.2a

9.0a

5.5a

1.4a

Treatment difference

N0

17.3b

48.3a

54.2a

3.6a

1.6a

53.0b

15.8ba

8.4b

5.5a

1.6a

N90

18.5b

47.7a

43.8a

3.6a

1.6a

57.1ab

14.5ab

8.8ab

5.9a

1.4a

N180

21.5a

45.7a

41.1a

3.8a

1.7a

58.7a

13.1b

9.3a

6.5a

1.4a

T

***

ns

ns

ns

ns

ns

ns

**

*

ns

V

ns

ns

*

ns

**

***

ns

ns

ns

ns

Values in a column followed by the same letter are not significantly different at P<0.05 as determined by the LSD test; n=3. The ns, *, ** and *** represent the non-significant, significant at P<0.05, 0.01, 0.001, respectively.

In addition to seed protein and oil content, our experiment also focused on glucosinolate, erucic acid, oleic acid, palmitic acid, stearic acid, linoleic acid, linolenic acid, and arachidionic acids (Table 2). The content of glucosinolate, erucic acid, oleic acid, palmitic acid, stearic acid, and arachidionic acids was not changed significantly with the different nitrogen treatments (Table 2). The opposite trend was observed between fatty acids (linoleic and linolenic) as the linoleic acid of both varieties declined by 15.8 and 17.1% in 2018/19 and 2019/20 at N₁₈₀ compared to N₀, respectively. In contrast, the linolenic acid of both varieties was improved by 9.3% and 10.7% in 2018/19 and 2019/20 at N₁₈₀ compared to N₀ treatment, respectively. Overall, the V₂ performed better than V₁ via the production of high oil and protein contents under various nitrogen levels.

3.6. Stem Sucrose and Reducing Sugar

Different nitrogen treatments affected the sucrose and reducing sugars (Table 3). The stem and pod wall sucrose and reducing sugars contents were reduced significantly (P < 0.05) under different nitrogen treatments in both varieties. At the flowering stage, the highest average contents of both varieties observed for sucrose and reducing sugars were 17.7 mg/g and 0.6 mg/g, respectively, under treatment N₀, whereas the lowest contents (0.7 mg/g and 0.2 mg/g) were observed at pod filling stage under treatment N₁₈₀ during 2018/19. Moreover, during 2019/20, the maximum contents found for sucrose and reducing sugars were 4.4 mg/g and 0.6 mg/g in N₀ at the flowering stage, respectively. Nonetheless, the minimum contents (2.0 mg/g and 0.5 mg/g) of sucrose and reducing sugars were obtained at the pod filling stage under treatment N₁₈₀, respectively). In the current study, the overall varietal difference was not significant (P > 0.05.) for sucrose and reducing sugars in both years (2018/19 and 2019/20) except for sucrose at pod filling stage in 2019/20. Moreover, at the pod filling stage, the reducing sugar showed a non-significant difference (Table 3).

Table 3. Effect of different nitrogen rates (N₀=0kg N ha⁻¹; N₉₀=90kg N ha⁻¹; N₁₈₀=180kg N ha⁻¹) on stem sugars (Sucrose and reducing sugar contents) of two varieties (V₁=JYJS01 (semi-dwarf), and V₂=CY36 (tall height) of rapeseed during the flowering and pod filling periods of 2018/19 and 2019/20

Stage		Treatments	variety	year	Sucrose (mg/g)	Reducing Sugar (mg/g)	year	Sucrose (mg/g)	Reducing Sugar (mg/g)
Flowering	Varietal difference	N0	V1	2018/19	17.0a	0.5a	2019/20	18.2a	0.7a
			V2		18.3a	0.6a		19.1a	0.7a
		N90	V1		15.6a	0.5a		16.2a	0.6a
			V2		16.3ab	0.5b		18.2a	0.5a
		N180	V1		13.4a	0.3a		12.7a	0.3a
			V2		13.9b	0.1c		8.2a	0.1a
		Treatment difference	N0		17.7a	0.6a		18.6a	0.7a
			N90		16.0ab	0.5a		17.2a	0.5ab
			N180		13.6b	0.2b		10.5b	0.2b
			T		*	**		*	ns
			V		ns	ns		ns	ns
Pod Filling	Varietal difference	N0	V1		4.6a	0.7a		5.9a	0.7a
			V2		4.2a	0.5a		10.3a	1.1a
		N90	V1		3.4a	0.5a		2.9a	0.6a
			V2		1.0b	0.3b		5.2b	0.7a
		N180	V1		0.5b	0.3a		0.8a	0.2a
			V2		0.8b	0.1b		3.2b	0.7a
		Treatment difference	N0		4.4a	0.6a		8.1a	0.9a
			N90		2.2b	0.4a		4.0b	0.6a
			N180		0.7c	0.2a		2.0b	0.5a
			T		***	ns		*	ns
			V		ns	ns		*	ns

Values in a column followed by the same letter are not significantly different at P<0.05 as determined by the LSD test; n=3.The ns, *, ** and *** represent the non-significant, significant at P<0.05, 0.01, 0.001, respectively.

3.7. Pod wall sucrose and reducing sugar

The different nitrogen treatments (N₀, N₉₀, and N₁₈₀) significantly affected (P < 0.05) the pod wall sucrose and reducing sugar content at flowering and pod filling stages (Table 4). High nitrogen treatment decreased the pod sucrose and reducing sugar contents as compared to control. In treatment N₁₈₀, the lowest sucrose and reducing sugar contents were 12.8 mg/g and 0.4 mg/g at pod filling and 1.2 mg/g and 1.0 mg/g at maturity stage in 2018/19 and 2019/20, respectively. Moreover, at N_180, the lowest sucrose and reducing sugar contents were 0.4 mg/g and 0.3 mg/g at the pod filling stage and 0.2 mg/g and 0.2 mg/g at the maturity stage in 2018/19 and 2019/20, respectively. The lowest sucrose and reducing contents were observed at N₀ in both years. The varietal difference was not significant in 2018/19, while it was significant for sucrose and reducing sugar at pod filling and maturity stages in 2019/20, respectively.

Table 4. Effect of different nitrogen rates (N₀=0kg N ha⁻¹; N₉₀=90kg N ha⁻¹; N₁₈₀=180kg N ha⁻¹) on pod wall sugars (Sucrose and reducing sugar contents) of two varieties (V₁=JYJS01 (semi-dwarf), and V₂=CY36 (tall height) of rapeseed during the flowering and pod filling periods of 2018/19 and 2019/20

Stage		Treatments	variety	year	Sucrose (mg/g)	Reducing Sugar (mg/g)	year	Sucrose (mg/g)	Reducing Sugar (mg/g)
Pod Filling	Varietal difference	N0	V1	2018/19	18.5a	0.8a	2019/20	20.2a	0.8a
			V2		6.4a	0.7a		11.0a	0.8a
		N90	V1		14.8b	0.6b		15.4b	0.5b
			V2		2.4b	0.4b		14.6b	0.5b
		N180	V1		13.3b	0.4c		12.7c	0.4c
			V2		0.6c	0.2c		11.0c	0.3c
		Treatment difference	N0		18.9a	0.8a		19.7a	0.8a
			N90		15.0b	0.6b		15.0b	0.5b
			N180		12.8c	0.4c		11.8c	0.3c
			T		***	***		***	***
			V		ns	ns		*	ns
Maturity	Varietal difference	N0	V1		5.8a	0.6a		5.4a	0.5a
			V2		5.4a	0.6a		6.4a	0.7a
		N90	V1		2.7b	0.4b		2.2b	0.3b
			V2		2.3b	0.2b		2.4b	0.4b
		N180	V1		1.3c	0.2c		1.5c	0.2c
			V2		1.0c	0.2c		0.6c	0.2c
		Treatment difference	N0		5.6a	0.6a		5.9a	0.6a
			N90		2.5b	0.4b		2.3b	0.4b
			N180		1.2c	0.2c		1.0c	0.2c
			T		***	***		***	***
			V		ns	ns		ns	***

Values in a column followed by the same letter are not significantly different at P<0.05 as determined by the LSD test; n=3.The ns, *, ** and *** represent the non-significant, significant at P<0.05, 0.01, 0.001, respectively.

4. Discussion

4.1. Morphological and agronomic parameters

In this study, the maximum number of branches and pods was obtained at a high nitrogen rate (N₁₈₀). According to Tuncturk and Çiftçi (2007), the number of branches has proliferated by increasing the nitrogen rate. Our results had indicated that the yield components were increased with increasing the nitrogen rate. It was also investigated by Inamullah et al. (2013), that a high nitrogen application rate enhanced vegetative growth vigorously. The continuous relationship among plant growth, seed yield and nitrogen application rate could be ascribed to higher primary branch height, main inflorescence length, pod number per plant (Baghdadi et al., 2014; Inamullah et al., 2013), stem diameter, branch number per plant (N. Khan et al., 2002), seed number per pod (El-Nakhlawy, 2009), pod surface area (El Kholy et al.,) and seed yield gram per plant (Cheema et al., 2001; Ghanbari-Malidarreh, 2010; Hao et al., 2004; Imran et al., 2014; Tian et al., 2021). (Table 1). According to Ahmadi and Bahrani (Islam et al., 2018), the highest rapeseed yield was attained at 225 Kg ha⁻¹ of nitrogen. However, the best nitrogen rate for optimum rapeseed yield differs between 180–240 kg ha⁻¹, depending on the experiment site, environment and former crop establishment (Leghari et al., 2016). Following this study, in the present experiment, the maximum seed yield was obtained at N₁₈₀ (Table 1).

4.2. Plant height and biomass accumulation

The effect of nitrogen fertilization was observed in plant height and biomass accumulation throughout the experiment. Increasing nitrogen application resulted in higher height and biomass accumulation values with a significant difference between the nitrogen treatments. Nitrogen is the essential nutrient for the growth and development of plants, enhancing plant height by its stimulative effect on cell divisions and cell enlargement (Leghari et al., 2016; Tiwari et al., 2002). It was studied by Yin et al. (2011), that corn yield was increased significantly as the plant height increased and also demonstrated by Katsvairo et al. (2003), that plant yield is increased with increment in plant height. Many studies on plant height and biomass accumulation revealed that plant height is positively correlated with biomass accumulation (Amanullah et al., 2008; Greef et al., 1999; Hamid & Nasab, 2001). Similarly, in this experiment, an increase in nitrogen rate had significantly increased the plant height (Figure 2) (Ahmadi, 2010; Islam et al., 2018); thus, it had increased the biomass accumulation (Figure 3) (Mahipat & Mukesh, 2014).

4.3. Chlorophyll content

Chlorophyll pigments are crucial for plant photosynthesis and play an essential role in multiple plant physiological functions (Guo et al., 2009). Nitrogen is an essential constituent of chloroplast, and any change in nitrogen treatment can directly affect the chloroplast and photosynthesis (Fang et al., 2018). In our experiment, the total chlorophyll contents were enhanced as the amount of nitrogen fertilization was proliferated (Figures 4 and 5). Hence, the total nitrogen levels in the plant tissues were enhanced; ultimately, more photoassimilates were available towards the reproductive parts (pods). Our results are in agreement with the previously published studies (Bell et al., 2004; Dordas & Sioulas, 2008; Hokmalipour & Darbandi, 2011; Men et al., 2020). It was depicted that the high rate of nitrogen application had increased the leaf and pod wall chlorophyll content (Men et al., 2020; Ni et al., 2019; Singh et al., 2017). During the flowering period, high chlorophyll content improved the photosynthetic efficiency of plants (Liu et al., 2017), which increased the number of pods with more seed weight (Khan, Zhou et al., 2018b).

4.4. Total nitrogen percentages in leaves, stem, pod wall and seed

A high nitrogen application rate enhanced nitrogen availability for plants under suitable growing conditions (Hocking et al., 1997; Kjellstrom, 1997; Merrien et al., 1988). It was reported by Hocking and Stapper (2001) that the accumulation of N in rapeseed shoots was higher between the rosette and flowering stages, thus, contributing to the more available nitrogen absorption during the flowering stage. Similarly, the maximum nitrogen accumulation in leaves and stems was observed in the current experiment at the flowering stage (Figures 6 and 7). The concentration of nitrogen accumulation in plant parts increased from vegetative to flowering then decreased towards maturity (Chamorro et al., 2002; Hocking et al., 1997; Schjoerring et al., 1995) shown in Figures 6 and 7. The accumulation of nitrogen in all organs (stem and leaves) was declined except seeds. Thus, it approved the dilution effect, which states that a higher rate of biomass accumulation decreased the rate of N accumulation (Hocking et al., 1997; Kjellstrom, 1997; Merrien et al., 1988).

4.5. Seed composition

Numerous research on the influence of nitrogen application on the seed protein and oil content have been reported (Aminpanah, 2013; Asare & Scarisbrick, 1995; Sawan et al., 2001). It was examined that the protein concentration was increased while the oil concentration was decreased by nitrogen application (Colnenne et al., 2002; Merrien & Blanchet, 1984; Mokhtassi-Bidgoli et al., 2013). The former studies had confirmed the inverse correlation between oil and protein contents in sunflower under high nitrogen application (Andrianasolo et al., 2016). Similarly, in our findings, the protein content increased, and the oil content decreased under the high nitrogen application rate Figure 9. Ni et al. (2019), stated that nitrogen inhibited sugar production and led to the diluting effect on the oil concentration. With the collapse of the photosynthetic organ, nitrogen was translocated from the photosynthetic organ to seed (Bennett et al., 2011); therefore, seed protein accumulation continued after seed oil accumulation was finished. As a result, seed oil percentage decreased, and protein percentage increased. Sugars are very vital carbohydrates, and they can be translocated from stem and pod walls to seed and cause their significant role in seed physiology (Bellaloui et al., 2020; Hill et al., 2003; Morley-Smith et al., 2008). Sugars are the chief and essential source of carbon for fatty acid metabolisms and ultimately affect the seed quality parameter of winter rapeseed (Bellaloui et al., 2020; Cai et al., 2020; Huppe & Turpin, 1994; N.A. Khan et al., 2008; Noctor & Foyer, 2000). According to Zhang and Flottmann, the photoassimilates, i.e. sucrose, played an essential function in fatty acid formation as it converted into fatty acid during lipid biosynthesis (Zhang et al., 2018), as is shown in Figure 9.

Figure 9. This figure shows that the high nitrogen application rate had caused a reduction of stem and pod wall sugars: sucrose and reducing sugars (1), which had caused their reducibility for glycolysis (2). This reduction is responsible for the happenings of two significant events. The first one is the decreased production of oil contents (3) and fatty acids and the second event is the enhancement in the protein contents (4). The TCA represents the Tricarboxylic acid cycle.

In this experiment, fatty acids contents such as glucosinolate, oil, linoleic and linolenic acid contents declined, while the stearic acid and protein contents enhanced under the high nitrogen rate. Our findings are in line with the results of Khan et al. (Khan et al., 2018a), which revealed that the linoleic and linolenic acid contents were decreased by 3.96 and 2.41%, respectively, by increasing the nitrogen application. Meanwhile, the oleic acid was enhanced, which is also supported by another study (Ahmed, 2019). It was also concluded by Li et al. (2017) that an increased in nitrogen application rate increased the palmitic and oleic acid while decreased the linoleic, linolenic and erucic acid contents in rapeseed seeds. The possible reason for the decrease in seed oil content at a high nitrogen rate (N₁₈₀) might be due to a reduction in stem and pod wall sucrose and reducing sugar content at flowering and pod filling stages.

4.6. Sucrose and Reducing Sugar

Nitrogen plays a crucial role in rapeseed sugars (sucrose and reducing sugars) accumulation (Bellaloui et al., 2020; Khan et al., 2018a). It is reported that a higher nitrogen rate had decreased the carbohydrates availability in rapeseed pod wall (Ni et al., 2019). The previous study reported that N application increased the total nitrogen content but decreased the sugar content in the pod wall (Cui et al., 2016). Similarly, our experiment exhibited that a high nitrogen application rate declined the stem and pod wall sucrose and reducing sugar content of rapeseed plants at the flowering stage and pod filling stages, Figure 9. According to Ni et al. (2019), the application of nitrogen decreased the carbohydrates in the Brassica napus L. pod wall. Sugars are the chief and important source of carbon for fatty acid metabolisms. According to Zhang and Flottmann (2018), the photoassimilates, i.e. sucrose, played a vital function in fatty acid formation as it is converted into fatty acid during lipid biosynthesis. It was reported that C:N ratio in the pod wall was higher when no N was applied, and oil concentration was also higher (Ni et al., 2019). As it is shown in Figure 9 due to the low availability of sucrose, the conversion of carbohydrates into fatty acids was decreased, which ultimately decreased the seed oil content (Khan, et al. 2018a) (Table 2). A similar reduction in carbohydrates was observed Mokhtassi-Bidgoli et al. (2013), and Pavithra et al. (2014), in which increment in nitrogen application rate had significantly reduced the sugar deposition.

5. Conclusion

Our results have demonstrated that a higher nitrogen application rate of 180 kg/ ha resulted in increased yield compared to control. Higher nitrogen application rates amplified the chlorophyll content in leaves and pod walls at flowering and pod filling stages which enhanced the photo-assimilates availability for reproductive parts of the plants and ultimately increased the pod number and seed weight. Fatty acid and carbohydrate proportions were reduced by increasing nitrogen rates. Decreased translocation of sugars from stems and pod walls to seed (for fatty acid biosynthesis) at flowering and pod filling stages had declined the oil, linoleic, linolenic, erucic and palmitic acid contents, whereas increment in protein content was observed. The key findings of the present study provide us with deeper insights for understandings the effects of nitrogen applications on the seed quality, carbohydrates availability for fatty acid metabolism and ultimately, the yield of rapeseed. However, further detailed research is needed to examine how varying nitrogen application rates affect the carbohydrate availability for fatty acid metabolism in the seed of new winter rapeseed varieties at molecular and genetic levels.

Author Contributions

Conceptualization, Y.W., A.G.; Methodology, Y.W., A.R., H.K., A.G.; Formal Analysis, A.R., H.K., A.G.; Investigation, Y.W.; Data Curation, Y.W., H.K., A.G.; Original Draft Preparation, Y.W., A.R., H.K., A.G.; Writing–Review & Editing, Y.W., A.R., H.K., A.G., M. A.A., B. A., H.H.J, H.Y, I.S., M.I.H. All authors have read and agreed to the submit the manuscript.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

This work was funded by the National Key R&D Program of China (2016YFD0300300).