Effect of molybdenum levels on photosynthetic characteristics, yield and seed quality of two oilseed rape (Brassica napus L.) cultivars

ABSTRACT

Molybdenum (Mo) is an essential element for higher plants. The objective of this study is to evaluate the effect of different Mo levels on the photosynthetic characteristics, yield and seed quality of two oilseed rape cultivars (ZS11 and L0917) using a pot experiment at four Mo levels [0 (control), 0.15, 0.3 and 1.0 mg kg⁻¹]. Results showed that the dry matter significantly increased with Mo concentration increasing at the stem elongation stage. The net photosynthesis rate (P_n), stomatal conductance (G_s) and transpiration rate (T_r) were the highest at the 0.15 mg kg⁻¹ level but reduced at 0.3 and 1.0 mg kg⁻¹ levels. The yield and harvest index were highest at 0.15 and 0.3 mg kg⁻¹ for ZS11 and L0917, respectively. Gray correlation model analysis showed that the synthetic characteristics of seed quality (expressed as gray relation degree) followed the sequence of Mo1.0 > Mo 0 > Mo 0.3 > Mo 0.15 mg kg⁻¹ for L0917 and Mo 1.0 > Mo 0.15 > Mo 0 > Mo 0.3 mg kg⁻¹ for ZS11. Our results demonstrate that the yield of oilseed rape reached a maximum at 0.15 and 0.30 mg kg⁻¹ for ZS11 and L0917, respectively, while the integrated seed quality is optimal at 1.0 mg kg⁻¹ for both cultivars in Mo-deficient soil.

KEY WORDS:

• molybdenum levels
• oilseed rape
• photosynthesis parameters
• seed yield and quality
• gray correlation analysis

1. Introduction

Molybdenum (Mo), a micronutrient required by higher plants, plays important roles in growth and metabolism in the form of Mo-enzymes, including nitrate reductase, sulfite oxidase, xanthine dehydrogenase, aldehyde oxidase, and the mitochondrial amidoxime reductase (Havemeyer et al. 2006; Mendel 2011; Mendel and Kruse 2012). In these enzymes, Mo has both structural and catalytic functions as well as direct involvement in redox reactions involved in nitrate assimilation, sulfite detoxification, purine degradation and synthesis of abscisic acid (ABA) (Mendel 2011). However, plant Mo accumulation depends on the availability of molybdate in soil strictly (Reddy et al. 1997), and oxyanion molybdate (MoO₄²⁻) is the predominant form in solution at pH higher than 4.2 for plant use (Lindsay 1979). When grown under Mo deficiency, plants exhibit a series of visible physiological phenotypes for hindering plant growth, including less biomass, low photosynthesis and nitrogen metabolism (Kaiser et al. 2005; Sun et al. 2009; Kovács et al. 2015).

Oilseed rape (Brassica napus L.), one of the three major oil crops in the world, provides edible oil and raw materials for bio-energy production (Karp and Richter 2011; Girondé et al. 2014). As one of the most important commercial and agricultural crops, oilseed rape is sensitive to Mo deficiency in soil, which affects its growth, production and seed quality. Liu et al. (2010a) reported that the dry matter, chlorophyll concentration and net photosynthesis ratio (P_n) of seedling leaves were increased with supplying Mo in oilseed rape, but affected by different phosphorus (P) levels. Additionally, the yield of oilseed rape was obviously increased only in the presence of P fertilizer (Liu et al. 2010b). Yang et al. (2009) also reported that the increase of seed yield with Mo addition was achieved under the addition of zinc (Zn) and boron (B) fertilizers. On the other hand, the seed quality, including the oil content and oleic and linolec acid contents, were increased significantly under the application of Mo and B fertilizer, while the contents of stearic acid, sulfuric glucosid and erucic acid were decreased (Chen et al. 2004). Similar results were also demonstrated under application of Mo and P fertilizers (Liu et al. 2012).

However, few studies have been performed so far on the effects of only Mo supply on the oilseed rape yield and synthetic characteristics of seed quality. In the present study, for a better understanding of the relationship of Mo levels with yield and quality of oilseed rape, the yield and photosynthetic characteristics were investigated under four Mo levels with two cultivars. Additionally, we also studied the quality indicators such as oil content, fatty acid composition, erucic acid, glucosinolate, mineral nutrients and protein of both cultivars under four Mo levels using the gray correlation model.

2. Materials and methods

2.1. Plant material and growth description

Two sequenced oilseed rape cultivars (ZS11 and L0917) were used for the experiment. The experiment was established with alfisols soil at Huazhong Agricultural University (30°28ʹ26ʹʹN, 114°20ʹ51ʹʹE, 30 m above sea level) in Wuhan, Hubei, China, from 31 October 2013 to 5 May 2014. The soil was air-dried, ground and sieved through a 2-mm sieve, followed by placing 6 kg soil in each pot, which was 55 cm in height and 23 cm in diameter. The physico-chemical characteristics of the experimental soil were: pH 4.8, organic matter 17.46 g kg⁻¹, alkaline hydrolysis nitrogen (N) 117.3 mg kg⁻¹, available P 37.70 mg kg⁻¹, available potassium (K) 76.70 mg kg⁻¹ and Tamm reagent-extractable Mo 0.101 mg kg⁻¹ soil [the threshold of Mo deficiency in soil is 0.15, referencing Bao (2000)]. Basal fertilizers applied were 0.20 g N, 0.15 g phosphorus pentoxide (P₂O₅) and 0.2 g potassium oxide (K₂O) per kg soil from ammonium sulfate ((NH₄)₂SO₄), dipotassium phosphate (KH₂PO₄), and potassium chloride (KCl), respectively. Microelements were supplemented to each pot, including boric acid (H₃BO₃) 2.86 mg, manganese chloride (MnCl₂ · 4H₂O) 1.81 mg, zinc sulfate (ZnSO₄ · 7H₂O) 0.22 mg, copper sulfate (CuSO₄ · 5H₂O) 0.08 mg and ethylenediaminetetraacetic acid-ferric sodium salt (FeNaEDTA) 36.71 mg per kg soil before sowing. Additionally, 0.6 g urea was added to each pot at stem elongation and flower stages, separately.

Four molybdate levels, 0 (Mo 0), 0.15 (Mo 0.15), 0.3 (Mo 0.3) and 1.0 (Mo 1.0) mg kg⁻¹, were applied as sodium molybdate ((NH₄)₆Mo₇O₂₄ · 4H₂O, AR). Each treatment was replicated 9 times, in which three replicates were used to determine the dry matter at the seedling stage, three replicates were used to determine the dry matter and photosynthetic characteristics of leaves at the stem elongation stage and the other three replicates were grown until grain maturity, with two plants for each pot. The plants were separately harvested at the seedling, stem elongation and grain maturity stages, then air-dried and weighed. Samples were oven-dried at 65°C for dry weight determination.

2.2. Elemental analysis

Oven-dried seed samples were ground and sieved through a 1-mm nylon sieve for the elemental analysis. Approximately 300 mg of each sample was transferred into test tubes and digested with 6 mL of nitric acid: perchloric acid (4:1, volume/volume). The temperature was held at 170°C for 2 h, then increased to 200°C and kept steady for 6 h until the liquid was colorless and transparent. The digest solution samples were used to analyze the elements of Mo (graphite furnace analysis), iron (Fe), manganese (Mn), copper (Cu) and zinc (Zn) (flame atomization analysis) by atomic absorption spectrometry (Z-2000; Hitachi, Japan).

2.3. Photosynthetic parameters

The net photosynthesis rate (P_n), stomatal conductance (G_s), transpiration rate (T_r) and intercellular CO₂ concentration (C_i) were measured on the intact topmost fully expanded leaf from the main stem at the stem elongation stage using an Li-6400 Portable Photosynthesis Analysis System (Li-COR, Lincoln, Nebraska, USA) as described by Liu et al. (2010b), with the measuring condition set as an open-circuit gas channel system, 500 μmol s⁻¹ air velocity and 1300 μmol m⁻² s⁻¹ photo-radiation intensity.

2.4. Seed quality analysis

Harvested seeds were air-dried. Grain quality was measured using near-infrared reflectance spectroscopy (NIRSystems 3700, FOOS, Hillerød, Denmark). The gray correlation analysis was performed as described by Jia et al. (2010) and Huang and Huang (1996). Briefly, (1) establish the reference and comparison sequences. According to the different characteristics of each quality indicator, we selected the maximum of protein, oil content, unsaturated fatty acid and mineral element, and the minimum of saturated fatty, erucic acid and glucosinolate as comparison sequences, denoted X₀. The initial values of seed quality were non-dimensionalized by dividing X₀ with them. (2) Calculate the relational coefficients (ξ). The gray relational coefficients (ξ) of each comparison sequence to the reference sequence are represented by Equation (1). (3) Calculate the relational degree. After the gray relational coefficients are obtained, the weights of different quality parameters in the oil content are determined by Equation (2). The product of the coefficient and weight is usually defined as the gray relational degree. (4) Order the relational degree. The product values of the coefficients and weights are arranged in the order of maximum to minimum.

(1)

(2)

2.5. Statistical analysis

All data were statistically analyzed with a two-way analysis of variance (ANOVA) procedure using SPSS 20.0 software (IBM, SPSS, Inc., Chicago, IL, USA), and the mean values of each treatment underwent multiple comparisons using the least significant difference (LSD) test at P < 0.05 level.

3. Results

3.1. Effects of Mo levels on dry weight of oilseed rape

There was a rising trend in the dry matter of oilseed rape with the increase of Mo level at seedling and stem elongation stages for both cultivars, but no significant difference was observed at the harvest stage, suggesting that the oilseed rape is more sensitive to Mo deficiency at seedling and stem elongation stages (Fig. 1).

3.2. Effects of Mo levels on photosynthesis parameters at the stem elongation period

The basic photosynthesis parameters including the net photosynthesis rate (P_n), stomatal conductance (G_s), transpiration rate (T_r), and intercellular carbon dioxide (CO₂) concentration (C_i) were measured at the stem elongation stage. The P_n, G_s and T_r reached a maximum at Mo level 0.15, followed by a decrease with the increase of Mo level (Fig. 2A, B, D), and the C_i also decreased with the increase Mo level (Fig. 2C).

Figure 1. The dry weight at (A) seedling, (B) stem elongation and (C) harvest stages of oilseed rape (Brassica napus L.) grown in pot soils treated with different Mo levels. Bars indicate standard error (n = 3). Different letters indicate significant differences at P < 0.05.

Figure 2. The photosynthesis parameters of (A) net photosynthesis rate (P_n), (B) stomatal conductance (G_s), (C) intercellular CO₂ concentration (C_i) and (D) transpiration rate (T_r) at the stem elongation stage of oilseed rape (Brassica napus L.) grown in pot soils treated with different Mo levels (mean ± standard error [SE]; n = 3). Different lowercase letters indicate a significant difference at P < 0.05.

3.3. Effects of Mo levels on grain yield and yield components of oilseed rape

Grain yields reached the maximum at Mo 0.15 and Mo 0.3 for ZS11 and L0917, respectively. The number of pods per plant reached the maximum at Mo 0.15 for both cultivars (Table 1). The harvest index showed the same tendency as the yield, which reached the maximum at Mo 0.15 and Mo 0.3 for ZS11 and L0917, respectively (Table 1).

Table 1. The grain yield, yield components and harvest index of oilseed rape (Brassica napus L.) grown in pot soils treated with different molybdenum (Mo) levels.

Cultivars	Treatments	Yield (g plant^−1)	Thousand seed weight (g)	Number of seeds per pod	Number of pods per plant	Harvest index (%)
L0917	Mo 0	5.92c	3.39b	14b	151ab	1.91ab
	Mo 0.15	7.63bc	3.82b	15b	184a	2.28ab
	Mo 0.3	8.72ab	3.39b	14b	174a	2.50a
	Mo 1.0	7.42bc	3.26b	15b	149ab	2.02ab
ZS11	Mo	8.59ab	4.71a	23a	87c	1.77b
	Mo 0.15	10.88a	4.70a	22a	108bc	2.32ab
	Mo 0.3	9.57ab	4.55a	21a	89c	2.13ab
	Mo 1.0	9.20ab	4.43a	23a	93c	2.06ab
T value	Cultivar	14.72**	68.69**	86.10**	44.57**	0.45
	Treatment	2.77	1.64	0.62	1.40	1.93
	Cultivar × Treatment	0.90	0.42	0.27	0.38	0.35

Note: Different letters indicate significant differences at P < 0.05; ** and * indicate F-test significant at P < 0.01 and P < 0.05, respectively.

3.4. Effects of Mo levels on seed quality of oilseed rape at harvest

Seed quality, next only to yield, is the second most important breeding objective in oilseed rape, including oil content, fatty acid (four unsaturated fatty acids and two saturated fatty acids) composition, erucic acid, glucosinolate, mineral nutrient and protein, etc. The quality indicators varied in their response to Mo levels, especially for different cultivars. For L0917, the oil content, linolenic acid and stearic acid reached the maximum at Mo level 1.0, while the contents of linoleic acid and erucic acid reached peak at Mo level 0.15. However, for ZS11, the oil content and Fe concentration reached the maximum at Mo level 1.0 and the content of oleic acid, linoleic acid, palmitic acid and glucosinolate reached a peak at Mo level 0.15 (Table 2). In order to analyze the effect of different Mo levels on the comprehensive quality of oilseed rape, the gray correlation model was used to evaluate the synthetic characteristics of seed quality by calculating the gray relational coefficients according to the reference and comparison sequences (Table 3). The mean of the gray relational coefficients of all the quality indicators was adopted as the gray relational degree, which represents the synthetic characteristics of seed quality. Based on the values of gray relational degree for the four Mo levels, the comprehensive quality for cultivar L0917 and cultivar ZS11 followed the order Mo 1.0 > Mo 0 > Mo 0.3 > Mo 0.15 and Mo 1.0 > Mo 0.15 > Mo 0 > Mo 0.3, respectively (Table 4), which means the integrated seed quality was optimal at Mo level 1.0 for both cultivars in Mo-deficient soil.

Table 2. Effects of different molybdenum (Mo) level on the protein, oil concentration and oil components in seed of two cultivars.

				Unsaturated fatty acid				Saturated fatty acid				Microelement concentration (mg kg^−1)
Cultivar	Treatment	Protein (%)	Oil content (%)	Oleic acid (%)	Linoleic acid (%)	Linolenic acid (%)	Eicosapentaenoic acid (%)	Palmitic acid (%)	Stearic acid (%)	Erucic acid (%)	Glucosinolate (µmol g^−1)	molybenum (Mo)	iron (Fe)	manganese (Mn)	copper (Cu)	zinc (Zn)
L0917	Mo 0	23.76c	40.29e	67.45bc	13.72b	5.55b	2.84a	4.23abc	0.48ab	7.94c	27.9bcd	0.159c	261.5a	902.0b	6.53a	56.23ab
	Mo 0.15	23.66c	41.12de	68.29b	15.06a	5.71b	3.29a	4.35ab	0.56a	11.43a	32.12b	0.220bc	173.6b	1226.5a	4.90abc	57.90a
	Mo 0.3	23.31c	42.74bc	68.09b	13.63b	5.81b	2.96a	4.21bc	0.30c	10.7ab	27.16bcd	0.282b	193.4b	1180.9a	3.21bc	53.74ab
	Mo 1.0	23.35c	42.18 cd	63.28c	13.87b	7.13a	3.58a	4.13 cd	0.37bc	8.53bc	28.38bc	0.294b	256.4a	1118.8a	5.33abc	48.00bc
ZS11	Mo 0	25.29b	42.88bc	66.88bc	13.77b	5.62b	1.44b	4.11 cd	0.36bc	9.28bcd	25.6bcd	0.182bc	197.6b	506.3c	5.44ab	40.95 cd
	Mo 0.15	26.74a	41.10de	74.44a	16.05a	6.03b	1.09b	4.42a	0.29c	9.87abc	54.75a	0.216bc	185.5b	584.5c	5.47a	52.27ab
	Mo 0.3	25.93ab	43.57b	70.15b	13.4b	5.71b	1.90b	4.15bcd	0.49ab	9.68abc	22.4 cd	0.259bc	194.7b	578.6c	3.22bc	36.24d
	Mo 1.0	25.83ab	45.22a	70.34ab	11.68c	6.06b	1.68b	4.00d	0.22c	9.55abc	19.59d	0.474a	260.6a	511.4c	3.13c	34.25d
F-value	Cultivar	70.45**	27.14**	10.65**	1.78	1.79	70.45**	1.91	5.24*	0.01	0.65	2.678	1.247	190.6**	1.635	41.82**
	Treatment	1.35	16.02**	3.46*	21.02**	9.56**	1.35	8.96**	2.74	3.15	19.25**	11.65**	11.67**	4.862*	5.107*	8.661**
	Cultivar × Treatment	1.71	5.47**	2.51	6.75**	4.46*	1.71	1.00	6.37**	1.80	11.47**	2.98	2.844	1.90	1.335	1.647

Note: Different letters indicate significant differences at P < 0.05; ** and * indicate F-test significant at P < 0.01 and P < 0.05, respectively.

Table 3. The gray relational coefficient between the reference and comparison sequences of two cultivars and the weights of different parameters in oil content.

				Unsaturated fatty acid				Saturated fatty acid				Microelement concentration
Cultivar	Treatment	Protein	Oil content	Oleic acid	Linoleic acid	Linolenic acid	Eicosapentaenoic acid	Palmitic acid	Stearic acid	Erucic acid	Glucosinlate	molybdenum (Mo)	iron (Fe)	manganese (Mn)	copper (Cu)	zinc (Zn)
L0917	Mo 0	1.0000	0.8824	0.9722	0.8286	0.6599	0.6754	0.9467	0.4175	1.0000	0.9404	0.4839	1.0000	0.6191	1.0000	0.9371
	Mo 0.15	0.9903	0.9190	1.0000	1.0000	0.6835	0.8415	0.8898	0.3316	0.4945	0.7019	0.6305	0.5612	1.0000	0.6327	1.0000
	Mo 0.3	0.9578	1.0000	0.9932	0.8191	0.699	0.7129	0.9569	1.0000	0.553	1.0000	0.9127	0.6227	0.9205	0.4582	0.8568
	Mo 1.0	0.9614	0.9704	0.8543	0.8448	1.0000	1.0000	1.0000	0.6482	0.8527	0.9054	1.0000	0.9565	0.8304	0.7006	0.7155
ZS11	Mo 0	0.9430	0.9455	0.8983	0.8633	0.9252	0.7876	0.9703	0.5851	1.0000	0.7453	0.5931	0.7879	0.8704	0.9939	0.8056
	Mo 0.15	1.0000	0.9078	1.0000	1.0000	0.9945	0.6780	0.8953	0.7383	0.9338	0.3334	0.622	0.757	1.0000	1.0000	1.0000
	Mo 0.3	0.9674	0.9609	0.9397	0.8446	0.9395	1.0000	0.9599	0.4224	0.9542	0.8622	0.664	0.7802	0.9889	0.6857	0.7453
	Mo 1.0	0.9635	1.0000	0.9422	0.7672	1.0000	0.8857	1.0000	1.0000	0.9686	1.0000	1.0000	1.0000	0.8777	0.6772	0.7225
Weight		0.0762	0.0794	0.0756	0.0713	0.0698	0.0612	0.0780	0.0489	0.0664	0.0570	0.0559	0.0678	0.0617	0.0628	0.0680

Table 4. The synthetic characteristics of seed quality (expressed as gray relational degree) of different molybdenum (Mo) levels.

	L0917		ZS11
Treatment	Relational degree	Sequence	Relational degree	Sequence
Mo 0	0.8242	2	0.8476	3
Mo 0.15	0.7784	4	0.8573	2
Mo 0.3	0.8309	3	0.8477	4
Mo 1.0	0.8827	1	0.9203	1

4. Discussion

In this study, we investigated the effects of different Mo levels on photosynthetic characteristics at the stem elongation stage, including P_n, G_s, T_r and C_i. Generally, G_s is often used to denote the extent of opening of stomata, T_r to reflect the intensity of transpiration in plants and C_i to indicate the assimilation ability of mesophyll cells for CO₂ in plants, which are important indices of the photosynthesis of plants and have close relationships with P_n (Farquhar and Sharkey 1982). The P_n, G_s and T_r reached maximum at Mo 0.15 mg kg⁻¹ for both cultivars and then declined with increasing Mo level, which means that a higher Mo level might cause a decrease in photosynthesis and transpiration. Liu et al. (2010b) observed that P_n in oilseed rape cultivar Zhongyouza No. 2 reached the highest value at 0.30 mg kg⁻¹ Mo level. In our previous research, the seed yield was found to be increased with the application of 0.75 kg hexaammonium heptamolybdate tetrahydrate ((NH₄)₆Mo₇O₂₄ · 4H₂O) per hectare in a field experiment for 30 oilseed rape cultivars (Qin et al. 2015). In this experiment, we also observed that grain yields reached a maximum at the Mo 0.15 level and Mo 0.30 level for ZS11 and L0917, respectively. These results suggested that the photosynthesis and yield were both increased with the application of Mo, and the proper Mo application level usually ranged from 0.15 to 0.30 mg kg⁻¹ soil in pot experiments with Mo-deficient soil.

As one of the most important oil plants in China, the quality of Brassica napus has aroused wide concern among scientists. The main quality indicators for oilseed usually include the oil content, fatty acid composition, erucic acid and glucosinolate concentration, etc., which play an important role in human nutrition and health (such as cardiovascular diseases) (Thelen and Ohlrogge 2002; Gül and Amar 2006). In contrast, the mineral element in oilseed rape seed is still poorly understood. Mineral deficiency is the source of the most massive ‘hidden hunger’ and malnutrition in the world today. A detailed analysis of nutrient elements in oilseed rape revealed that potassium (K), phosphorus (P), sulphur (S), magnesium (Mg), iron (Fe), manganese (Mn) and zinc (Zn) were mainly stored in the cotyledons, while the seed coat contained most of the Ca, Cu, Mo and B in the seed (Eggert and Von Wirén 2013). Recently, the content of mineral nutrients, as a part of the quality of oilseeds, has also attracted wide mention. For example, Zn fertilizer application increased Zn concentration of seed, thereby producing positive effects on rapeseed oil content and fatty acid composition (Wang et al. 2014). Additionally, the application of microelement fertilizers, such as Cu and B, also increased fat content (Sienkiewicz-Cholewa and Kieloch 2015). Several researchers found that Mo application can increase seed oil content, which is beneficial to seed quality (Laaniste et al. 2004; Chen et al. 2005). Other researchers observed that Mo application can increase the protein content of the seed, without showing any significant influence on fatty acid (Cai and Zhang 2000). On the other hand, it is well known that seed quality indicators like saturated fatty acid, erucic acid and glucosinolate have a negative relationship with seed quality. Meanwhile, we found that parameters varied obviously in their response to Mo level. For instance, the linolenic acid and Mo concentration reached a maximum at Mo level l.0, while the oleic acid and linoleic acid reached peak at Mo level 0.15. In this case, the gray correlation model was used to analyze the key factors affecting the characteristics of the system, as has been done in many previous studies. For example, the relationship of the agronomic traits with the per plant yield of six soybeans (Glycine max L.) was analyzed using the gray correlation model (Shen et al. 2012) and the relationship among agronomical traits, physio-biochemical traits and drought tolerance of wheat cultivars (Triticum aestivum) was analyzed using gray relational grade analysis method (Wang et al. 2007). In this study, the gray correlation model was used to evaluate the effects of Mo levels on the synthetic characteristics of seed quality, and the optimal result for both cultivars was found at Mo level 1.0, indicating that a high Mo level can improve the seed quality. This can be attributed to the altered activity of Mo-containing oxidoreductases; the effect of Mo deficiency on Mo enzyme activity, amino acid, sugars, organic acids and purine metabolites has been confirmed (Ide et al. 2011). Currently, Mo application in Mo-deficient soils cannot achieve a balance between high yield and high quality in oilseed rape, and how to achieve this balance can be explored in future work.

5. Conclusion

In the present study, the overall results show that Mo is a limiting factor not only for grain yield but also for seed quality. The oilseed rape can achieve a higher photosynthesis parameter and yield under low Mo application (0.15–0.3 mg kg⁻¹) compared to Mo-deficient soil, but the synthetic characteristic of seed quality is optimal at high Mo application (1.0 mg kg⁻¹). This finding from this research may facilitate a better understanding of the relationship of Mo fertilizer application with yield and seed quality and provide useful information for improving agricultural production in oilseed rape.

Acknowledgments

We greatly appreciate the critical review and revision of the manuscript by Dr. Ron Mclaren (Emeritus Professor of Environment Soil Science, Lincoln University).

Additional information

Funding

This work was supported by the The Fundamental Research Funds for the Central Universities [2013PY135], the 948 Project from the Ministry of Agriculture of China [2015-Z34 and 2016-X41] and the National Natural Science Foundation of China (Program No. 41171240).