Iron and zinc biofortification in polished rice and accumulation in rice plant (Oryza sativa L.) as affected by nitrogen fertilization

Abstract

With focus on maximizing grain yield in rice (Oryza sativa L.) production, especially in China, information available in the literature on how nitrogen (N) fertilization of rice crops affects biofortification of iron (Fe) and (Zn) in grains is limited. The objective of the experiment was to investigate to what degree application of N fertilizer attained the optimum Fe and Zn concentration in rice grains as well as grain yield under pot conditions. Two rice cultivars of the indica ‘Zhenong 952’ and the japonica ‘Bing 98110’, grown widely in the area of the Yangtse River Delta in southern China, and fertilized with four rates of urea (0, 0.50, 1.00 and 1.50 g N pot⁻¹), were investigated. The results showed that, in the pot trails, the optimum application of N alone on rice crops could increase the concentration of Fe in the polished rice. By considering both health and commercial reasons, when N application reached 1.00 g pot⁻¹, the optimal Fe and Zn concentrations were attained as well as grain yield for ‘Zhenong 952’, and for ‘Bing 98110’ the optimum N application was 1.50 g pot⁻¹. Fe appeared not to be so easily mobilized as Zn in the plant. The ratio of Zn deposited in the brown rice was about 40% of total Zn in the plant, irrespective of N application. However, deposited Fe was only about 3% of total Fe. Fe concentration in brown rice was only about ½ that in rice husk, 1/5 that in peduncles, 1/10 that in leaves, and only a little more than 1% of that in roots. These results suggested if we wanted to increase the amount of Fe in grains the translocation mechanism of Fe in rice plant must be clearly understood first.

Keywords:

• Biofortification
• Grain yield
• Human health
• Iron
• Nitrogen
• Zinc

Introduction

Iron (Fe) and zinc (Zn) deficiencies that weaken immune function and may impair growth and development (Welch, 2002) are widespread in the world, especially in developing countries, where it is estimated that 40–45% of school-age children are anaemic and that about 50% of this anaemia results from Fe deficiency (WHO, 1998). In China, more than 20% of the people are influenced by Fe deficiency (Yu, 2005). A major etiological factor is the low concentration of Fe and Zn from diets based on staple cereals.

Rice (Oryza sativa L.) is the dominant staple food for more than half of the world's population (Zimmernann & Hurrell, 2002; Wang et al., 2005). It provides 23%, more than that provided by wheat and corn, of all the calories consumed by the world's population, and even provides 50–80% of the energy intake of the people in developing countries (IRRI, 2006). Rice, however, is a poor source of many essential mineral nutrients, especially Fe and Zn, for human nutrition. The IRRI (2006) report stated that polished rice contains an average of only 2 mg kg⁻¹ Fe and 12 mg kg⁻¹ Zn whereas the recommended dietary intake of Fe for people is 10–15 mg, and that of Zn is 12–15 mg (Welch & Graham, 2004). Currently, malnutrition of Fe and Zn afflicts more than 50% of the world's population (Tucker, 2003; Welch, 2005). Heavy and monotonous consumption of rice with low concentration of Fe and Zn has been considered a major reason for this (Welch & Graham, 1999; Graham et al., 2001).

There are several potential approaches to increase the concentration of Fe and Zn in staple foods, including food nutrient fortification and supplementation programs (Michaelsen & Friis, 1998; Hurrell, 2002; Davidsson, 2003), and conventional breeding and genetic engineering (Graham et al., 1999; Bouis, 2000; Lucca et al., 2001; Zimmernann & Hurrell, 2002; Holm et al., 2002). Although fortification or supplementation in rice has proven to be effective for certain nutrients (Poletti et al., 2004), it is so expensive and has such unpleasant side-effects that people can either hardly afford it or show poor compliance with it (Frossard et al., 2000). Biotechnology and plant breeding currently show low efficiency, high cost and limited success (Schachtman & Barker, 1999) such that they cannot settle the present severe problem of Fe and/or Zn deficiencies in human nutrition. Agriculture is the primary source of all nutrients required for crops and then for human health, and fertilization is the key point of nutrient-integrated management in agronomic approaches to enhance crop quality and yield, so that fertilization could be one of the sustainable and low-cost strategies to improve Fe and Zn density in edible portions of staple food crops (Rengel et al., 1999).

Nitrogen (N) application is the main segment in rice fertilization, and Gregorio et al. (2000) have reported that N level is an important favourable factor determining grain mineral content. However, the effects of N on Fe and Zn concentration in rice grains are not unambiguous, though N fertilization on increasing nutrient concentration in other crops had been proved to be effective (Zebarth et al., 2002; Chenard et al., 2005; Wangstrand et al., 2006). And in China, N application on rice crops, ranging from 152 kg ha⁻¹ to 274 kg ha⁻¹ (Jin et al., 2002), is extremely focused on maximizing grain yield. How N fertilization affects Fe and Zn concentration has not been of concern. Furthermore, Fe and Zn accumulation in different organs of the rice plant has not been studied in detail up to now. So the aim of the present study is to evaluate the effects of N on the accumulation of Fe and Zn in the grains and in the other parts of the rice plant and to determine the optimum N rate at which the optimal Fe and Zn density in the polished rice could be attained, as well as grain yield. The authors hope that their findings may be a forward-looking guideline for N fertilization on harvesting high-Fe and -Zn rice that could be beneficial to human health.

Materials and methods

Plant materials and pre-culture

Rice varieties used in the experiment were the indica ‘Zhenong 952’ and the japonica ‘Bing 98110’. Both varieties were widely grown in the area of the Yangtse River Delta in southern China.

Seeds of rice were surface-sterilized in 1.5% (v/v) sodium hypochlorite for 10 min, rinsed thoroughly in deionized water and soaked in a dish containing a shallow layer of deionized water at room temperature overnight, and germinated in moist quartz sand. Seedlings were grown in a growth chamber with a light/dark regime of 14/10 h, temperature of 30/25 °C (day/night), a light intensity of 12 000 lux and a relative humidity of 70–80%. Twenty-day-old uniform seedlings were used in the further experiment.

Experimental sites and design

The pot experiment was conducted in the experimental ground of the Huajiachi campus of Zhejiang University in 2006 and was laid out in a completely randomized design with three replicates per treatment. Every pot in which three 20-day-old uniform rice seedlings were transplanted was a replicate. Black plastic pots used in the experiment were 30.0 cm in height, and 25.5 cm in diameter. The powdery loam soil was air dried and passed a 2 mm sieve, homogenized, and each pot contained 7.5 kg of soil. Some properties of the experimental soil are presented in Table I. Seedlings of ‘Zhenong 952’ were transplanted on April 18 and harvested on July 22, and seedlings of ‘Bing 98110’ were transplanted on July 10 and harvested on October 22.

Table I. Some selected properties of the soil in the experiment before fertilization.

	Organic	Total		Available
	C	N	P	N	P	K	Fe	Zn
pH	g kg^−1		mg kg^−1
5.70	27.30	0.96	0.99	51.77	22.87	86.06	26.26	1.91

Four application doses of N fertilizer urea (0, 0.50, 1.00, 1.50 g N pot⁻¹) were studied. 2/3 of urea-N was applied as the basal fertilizer and 1/3 at the tiller stage. All pots received 0.67 g P₂O₅ pot⁻¹ as triple superphosphate and 1.00 g K₂O pot⁻¹ as potassium chloride before seedlings were transplanted into the pots. All basal fertilizers were homogenized with the soil.

Plant sampling and analysis

Grain samples of ‘Zhenong 952’ and the whole plant samples of ‘Bing 98110’ were collected for analysis. Straw of ‘Bing 98110’ was separated into the peduncle, the flag leaf, the leaf excluding the flag leaf, the leaf sheath, the stem and the root. After being rinsed thoroughly with 0.01 M HCl (AR) and then with deionized water in order to clean the samples, completely all parts of the straw were dried at 70 °C to constant weight.

Rice grains were air dried. The blighted grains were separated before rice seeds were dehusked in an electrical dehusker (model JLGJ-45, China), and then the rice husk was collected. A part of the brown rice was polished with a sample polisher (model JB-20, China) before being ground.

All samples were then ground with a sample grinder (model Retsch MM301, Germany), digested in 2.0 ml HNO₃ (GR) and 0.5 ml H₂O₂, and digestion solutions were allowed to cool to room temperature (∼25 °C) and adjusted to a final volume of 25 ml with doubly deionized water. Fe and Zn concentration was determined by an inductively coupled plasma mass spectrometer (ICP-MS, model Agilent 7500a, USA). Reference material (the powder of the polished rice, GBW (E) 080684) and blanks were included in each digestion and Fe and Zn determination.

All statistical analyses were performed using STATISTICA (v. 5.5). Each value represented the average of three replicates. Data were subjected to analysis of variance (ANOVA) and significant differences in mean values were separated using Duncan's Multiple Range Test (p <  0.05).

Results

Grain yield as affected by N fertilization

For the two rice varieties, the grain yield increased significantly with N fertilization ranging from 0 to 1.50 g pot⁻¹, and more grain yield of ‘Bing 98110’ was harvested than ‘Zhenong 952’ at one N level (Table II). The results in Table II also suggested that application of more than 1.50 g N pot⁻¹ could improve grain yield of the two rice varieties, but yield response of ‘Zhenong 952’ to more N application would not be significant compared with that of ‘Bing 98110’.

Table II. Grain yield and Fe and Zn concentration in the polished rice of the indica ‘Zhenong 952’ and the japonica ‘Bing 98110’ as affected by N levels.

	Zhenong 952			Bing 98110
	Yield	Fe	Zn	Yield	Fe	Zn
N levels	g plant^-1	mg kg^−1 DW ^c		g plant^−1	mg kg^−1 DW
N0 ^b	11.32 c^a	1.845 b	13.969 a	19.17 d	1.848 b	20.914 a
N0.5	19.48 b	1.838 b	13.941 a	27.76 c	1.871 b	14.806 c
N1.0	33.16 a	2.085 a	12.958 b	33.89 b	2.171 a	15.623 bc
N1.5	35.48 a	1.967 ab	13.176 ab	43.27 a	1.980 ab	17.399 b
^a Means within a column with the same letter are not significantly different by Duncan's Multiple Range Test at p < 0.05 level (r = 3).
^b N0, N0.5, N1.0, N1.5 mean 0, 0.50, 1.00, 1.50 g N pot^−1 application, respectively.
^c DW = Dry weight

Iron and zinc concentration in polished rice as affected by N fertilization

There was the highest concentration of Fe in the polished rice of ‘Zhenong 952’ with the application of N at 1.00 g pot⁻¹, and compared with no N supply it was increased significantly by 13.01%. With N application rate increased to 1.50 g pot⁻¹, no significant increase was observed (Table II). Similarly, the highest concentration of Fe was attained in the polished rice of ‘Bing 98110’ with the application of 1.00 g N pot⁻¹, and this value was also increased significantly by 17.48% compared with no N supply. In addition, no significant increase occurred when the application of N was increased to 1.50 g pot⁻¹ (Table II). The results we found were similar to those of Gregorio et al. (2000), which showed that Fe increased significantly by an average of 15% with the addition of N at applications between 0 and 135 kg ha⁻¹.

In contrast, when the application rate of N reached 1.00 g pot⁻¹, the lowest concentration of Zn in the polished rice of ‘Zhenong 952’ was attained and it was decreased significantly by 7.24% compared with no N supply. And though there was a slight increase in Zn concentration in the polished rice when N supply increased to 1.50 g pot⁻¹, no significant difference was observed (Table II). The lowest concentration of Zn in the polished rice of ‘Bing 98110’ was attained at 0.50 g N pot⁻¹ application and it was decreased significantly by 29.21% compared with no N supply; in addition, there was a marked 17.51% increase with N application from 0.50 to 1.50 g pot⁻¹ (Table II). However, Gregorio et al. (2000) found that there was no significant difference in Zn concentration with the addition of N at applications between 0 and 135 kg ha^{− 1}. Though at its lowest in ‘Zhenong 952’ with application of N at 1.00 g pot⁻¹ and in ‘Bing 98110’ at 0.50 g N pot⁻¹, Zn concentration of the two cultivars was still higher than the average value of 12 mg kg⁻¹ Zn in the polished rice reported by IRRI (2006).

Iron and zinc accumulation in rice plant under different levels of N application

For the whole plant, regardless of N application, Fe in the plant was mainly deposited in the root, in which about half of the Fe accumulated. In the aerial part, about 40% of Fe deposited in the straw (including the stem, the leaf sheath and the leaf), and only about 3% of total Fe was accumulated in the brown rice (Figure 1). From Table III, we could see that the concentration of Fe in brown rice was less than 1.5% of that in the root, about 1/10 of that in the leaves, 1/5 of that in the peduncles and even only about a half of that in the rice husk. With the addition of N application from 0 to 1.0 g pot⁻¹, the ratio of Fe accumulated in the brown rice to the whole plant was increased by 8.23%, and when N increased to 1.5 g pot⁻¹, the ratio decreased again to 2.79% (Figure 1).

Figure 1.  Ratio of Fe and Zn in organs to the whole plant of the japonica ‘Bing 98110’ as affected by N levels.

Table III. Concentrations of Fe and Zn in different organs in the japonica ‘Bing 98110’ as affected by N levels.

	N0		N0.5		N1.0		N1.5
	Fe	Zn	Fe	Zn	Fe	Zn	Fe	Zn
Organs	mg kg^−1 DW
Brown rice	9.398 g	41.875 b	8.904 g	36.124 c	11.348 f	33.695 c	10.150 g	35.859 b
Rice husk	24.170 f	8.879 g	22.031 f	8.421 f	22.013 e	10.421 f	21.020 f	10.180 f
Blighted grain	24.394 f	23.251 d	22.626 f	20.894 d	21.306 e	22.461 d	24.003 f	20.320 d
Peduncle	52.295 e	27.668 c	44.217 e	20.431 d	50.934 d	21.440 d	47.720 e	18.724 d
Flag leaf	72.234 d	19.139 e	64.617 d	14.778 e	68.294 c	10.828 f	72.912 d	12.549 e
Leaf excluding flag leaf	107.990 b	16.322 f	128.444 b	15.272 e	85.845 b	16.606 e	94.606 c	11.569 ef
Leaf sheath	98.923 bc	23.122 d	57.581 d	13.203 e	64.551 c	14.501 e	52.835 e	12.624 e
Stem	95.662 c	48.943 a	88.108 c	50.756 b	87.901 b	66.904 a	105.358 b	45.848 a
Root	641.151 a	29.273 c	722.308 a	54.158 a	991.523 a	59.112 b	760.415 a	32.621 c
The whole plant	125.135	26.497	128.760	26.004	155.969	28.441	132.113	22.255

However, Zn appeared to be easily mobilized from the root to the grains (Table III). Irrespective of N application, for the whole plant, the ratio of Zn deposited in the brown rice to total Zn in plant was about 40%; the next highest amount was in the stem, whereas in the root the value was only about 10% (Figure 1). With the addition of N application from 0 to 1.0 g pot⁻¹, the ratio of Zn accumulated in the brown rice to the whole plant decreased by 28.01%, and when N increased to 1.5 g pot⁻¹, the ratio increased again to 39.29% (Figure 1).

The ratio of Fe/Zn was about 5 in the whole plant regardless of the treatment N levels (Table III), in accord with the findings of Bdase's (He & Meng, 1987) which showed that the ratio Fe/Zn in the plant was 3∼15, and that the optimal value was 5.

Discussion and conclusion

In light of the results obtained in the present study, irrespective of N application, the concentration of Zn in the polished rice of the indica ‘Zhenong 952’ and of the japonica ‘Bing 98110’ was higher than the average value of Zn reported by IRRI (2006), and that of Fe was less (Table II). For the indica ‘Zhenong 952’, with N applied at 1.00 g pot⁻¹, the concentration of Fe in polished rice was highest, and the grain yield was increased sharply three-fold compared with no N supply, and no significant increase occurred when N application increased to 1.50 g pot⁻¹. In rice production, no fertilization was impossible, though Zn was highest with no N supply, the optimal N application for attaining high Fe and Zn rice as well as grain yield should be 1.00 g N pot⁻¹ considering Fe concentration and grain yield (Table II). For the japonica ‘Bing 98110’, with N applied at 1.00 g pot⁻¹, the concentration of Fe in polished rice was highest. However, Zn was highest at no N supply. With the addition of N application increased to 1.50 g pot⁻¹, the higher concentration of Zn was attained as well as the highest grain yield; Fe had no significant decrease with N application at 1.50 g pot⁻¹ compared with 1.00 g pot⁻¹, so considering Zn concentration and grain yield, the optimal N application for attaining high-Fe and -Zn rice as well as grain yield should be 1.50 g N pot⁻¹ (Table II). Generally, an increased rate of N fertilization aimed at Fe and Zn biofortification in polished rice tended to elevate nonsynchronously the concentration of Fe and Zn in rice grains under pot conditions. The results suggested that while rice crops tended to maintain nutrient concentration in the grain within predetermined limits, application of fertilizers could alter the balance (Rengel et al., 1999), so understanding the factors that influenced the balance might be important when selecting for accumulation of a specific nutrient, especially for Fe or Zn. As seed developed in the parent plant, concentration of nutrients was dependent on soil type, nutrient availability, crop species and, to a lesser extent, season and cultivars (Ascher et al., 1994), and this might be an explanation as to why our results for Zn were not in coincidence with those of Gregorio et al. (2000).

Irrespective of N application, for the whole plant, Zn appeared to be easily mobilized from root to the aerial part, especially into brown rice to a considerable extent (Figure 1). In contrast, the accumulation of Fe in brown rice was very difficult. Compared with about 40% of Zn deposited in brown rice (Figure 1), the ratio of Fe accumulated in brown rice was only about 3% of total Fe in the whole plant (Figure 1). The concentration of Fe in brown rice was only a very little more than 1% of that in the root, and was 20% of that in the peduncle and only about 50% of that in the rice husk (Table III). Thus it could be seen that there are many barriers to Fe translocation and accumulation into the grains. The key point of solving Fe accumulating by folds in the grains was not on the Fe uptake from roots but on the translocation of Fe into the grains. We had known that the accumulation of Fe in grains was controlled by a number of processes including root-cell uptake, root-shoot transfer, and the ability of leaf tissues to load Fe into the vascular phloem that was responsible for delivering Fe to developing grains via the phloem sap. However, the homeostatic mechanism by which the phloem loading and unloading of Fe was tightly controlled in the plant was poorly understood (Welch, 2002; Hell & Stephan, 2003). So further research should be carried out to understand these processes if we want to increase Fe content in grains.

Rice contains very small amounts of many essential micronutrients, but half of the world's population eats it daily and depends on it for their staple food. Therefore, even a small increase in its nutritive value would be highly profound for the alleviation of malnutrition, e.g. Fe and Zn deficiencies, and would then be of benefit for human health. This suggests that appropriate measures for the integrated nutrient management in agronomy suitable for high-Fe and -Zn rice production should be established. On the other hand, we have to accept that if we want to attain a 100% or more increase of Fe and Zn in polished rice, plant breeding and genetic modulation of high-Fe and/or -Zn rice varieties are absolutely necessary. Only high-Fe and/or -Zn rice varieties, combined with the optimum agronomic practices, could produce high-nutritional quality rice.

The goal of increased Fe and Zn concentration in rice is to better satisfy human demands, so the bioavailability of increased Fe and Zn in polished rice should be of concern. Caco-2 cell mode and piglet mode techniques were recently used to assess bioavailability and biological effects of micronutrients in crops. However, little study has been conducted using these techniques so far, especially in China, to value the bioavailability of the increased Fe and/or Zn in edible parts of crops that is what we want to study in the near future.

Acknowledgements

The study was funded by HarvestPlus-China Program (#8022) and the National Natural Science Foundation of China (No. 30370838). We thank X.F. Yang for her help on the experimental preparation and G.X. Hu for the samples determination.