Keywords

1 Introduction

The human body needs micronutrients for proper growth and development and to maintain good health (Maret 2017; Palanog et al. 2019). However, deficiencies in these elements and associated health risks are commonly reported in all age groups, especially in preschool children, women, and elderly people in the developing world (Caulfield et al. 2006). An estimated one-third of the global population suffers from micronutrient malnutrition, mainly because of the large dependence on cereal staples for daily nutritional needs without access to a diversified diet and supplementation (Ritchie et al. 2018). The urgent need to address micronutrient malnutrition has been widely recognized globally; hence, decreasing childhood mortality and maternal death by eradicating malnutrition is an important Sustainable Development Goal (Hanieh et al. 2020).

Among the micronutrients, zinc (Zn) is most essential for vital organs, enzymatic activity, tissue growth and development, cognitive function, immunity, etc. There is therefore a need for a regular daily supply of Zn in the required quantity to have healthy and productive populations (Prasad et al. 2014; Chasapis et al. 2020). However, an estimated two billion people suffer from Zn deficiency-related health consequences and most of them are resource-poor urban and rural dwellers (Rampa et al. 2020). The disability-adjusted life years (DALYs) due to Zn malnutrition strongly impact annual GDP growth, and hamper economic development in the developing world (Gödecke et al. 2018). Multiple interventions such as fortification of foods, micronutrient supplementation, and food diversification have been employed to mitigate Zn malnutrition; however, recurring costs and poor accessibility and awareness among the rural masses have resulted in limited success (Bouis 2017). Increasing the mineral density in the edible part of the major staple crops, which is also popularly called “biofortification ,” has been proven to be effective in alleviating malnutrition without much additional cost. This complementary food-based approach is the safest and cheapest way to deliver nutrients on a larger scale to the target populations (Bouis and Saltzman 2017).

Rice is among the target staple food crops for Zn biofortification in different countries of South Asia, Southeast Asia, and Africa (Siwela et al. 2020). The Biofortification Priority Index prepared by HarvestPlus clearly identified several countries in Asia with an immediate need for Zn biofortification (HarvestPlus 2020). The International Rice Research Institute (IRRI) and its national partners in target countries are making efforts to develop Zn-biofortified rice varieties. The first set of high-Zn rice varieties has been released for commercial cultivation in Bangladesh, India, the Philippines, and Indonesia (Inabangan-Asilo et al. 2019). Efforts are in place to mainstream the breeding of high-Zn rice by applying advanced breeding techniques and genomic tools to make sure Zn will be an essential component of all future varietal releases from the main breeding pipelines of IRRI (CGIAR 2018).

Over the past decade, great progress has been made in our understanding of Zn homeostasis in rice from a biofortification perspective and in the development of high-Zn rice. In this chapter, we would like to provide some insights into the recent advances in developing Zn-biofortified rice for the target countries.

2 Zn Is Critical for Human Health

Zinc plays an important role in the catalytic function of most of the enzymes needed for the structural stability and functioning of more than 3000 proteins, helps to maintain membrane stability, and protects tissues and cells from oxidative damage (Cakmak 2000; Broadley et al. 2007; Andreini et al. 2009; Maret and Li 2009).

Zinc deficiency is one of the major causes of child mortality worldwide (Black et al. 2008), which has been estimated to affect more than 178 countries (WHO 2003). Zn-deficient children are highly prone to diarrhea, respiratory ailments, poor cognitive function, and stunting (Brooks et al. 2004; Sazawal et al. 2007; Tielsch et al. 2007; Young et al. 2014). Zn deficiency during the first 1000 days for children after birth causes irreversible damage leading to less chance of survival, poor immune system and cognitive ability, and stunting (UNICEF 2013). Hence, a regular daily supply of Zn is highly essential, but this is rarely achieved by most resource-poor people. Thus, adequate Zn nutrition is essential for good health, especially for children and pregnant women for growth and development (IZiNCG 2009). The daily Zn requirement of individuals varies from 9 to 11 ppm depending on age, gender, and health conditions, but preschool children and pregnant and lactating women need more Zn (IOM 2001; Welch and Graham 2004; Iqbal et al. 2020; Alqabbani and AlBadr 2020).

3 Rice Biofortification with High Grain Zn

Rice is the single most important source of energy and nutrition for more than half of the world’s population (Gross and Zhao 2014). It is a major staple crop in more than 40 countries and supplies at least 20% of the daily caloric intake of more than 3.5 billion people (FAO 2014). Asia, with 60% of the global population, consumes more than 90% of the total rice produced annually (Milovanovic and Smutka 2017). Annual per capita rice consumption exceeds 100 kg in some Asian countries (FAO 2016). However, milled rice is less nutritious; thus, most of the poor people who largely depend on rice without access to a mineral-rich diverse diet suffer from hidden hunger, including Zn deficiency.

Food-based solutions were found to be safe and effective in controlling and preventing micronutrient deficiencies, especially when multiple deficiencies occur (Torheim et al. 2010; Szymlek-Gay et al. 2009). Several studies reported that the consumption of a diverse diet and crops enriched with mineral elements provides more nutrition (Brown et al. 2002; WHO 1998). Recently, biofortification of staple crops has become a popular method for tackling malnutrition. It is the process of increasing the density of readily bioavailable mineral elements by breeding or biotechnological approaches (Garg et al. 2018) for staple food crops such as rice, which has been obtaining increased attention by breeders and policymakers in recent times. Biofortification has the lowest per capita costs vis-à-vis other interventions, and it is especially easily accessible and affordable for rural populations (Ma et al. 2008). Therefore, increasing grain-Zn content would create a significant impact on human health. One estimate suggested that an additional 8 μg/g of Zn in raw milled rice over the baseline Zn (16 μg/g) in cultivated varieties could help to reach the amount equivalent to 30% of the Estimated Average Requirement per day (HarvestPlus 2012).

4 Trait Development for High Grain Zn

4.1 High-Zn Donor Identification

Rice is endowed with abundant genetic diversity and thereby provides needed genetic variability for rice breeding programs (Rana and Bhat 2004). More than 230,000 rice accessions are maintained in global gene banks, which include landraces, cultivars, varieties, and aromatic and wild rice (Li et al. 2014). Among the different species or subgroups, wild rice, landraces, and aus accessions were found to be a rich source of micronutrients; they have several-fold higher nutrients than cultivated rice (Cheng et al. 2005; Banerjee et al. 2010; Descalsota-Empleo et al. 2019a, b). Aus accessions are genetically closer to popularly grown indica rice varieties, so they can be readily used by breeding programs to improve the Zn content of modern rice varieties. Some aus accessions such as Kaliboro, Jamir, UCP122, DZ193, and Khao ToT Long 227 have higher content of grain Zn (Norton et al. 2014; Descalsota et al. 2018). We are efficiently using aus germplasm in our breeding programs at IRRI and have also widely shared these donor lines with our partners for use in their breeding programs. The accessions of the 3K Genome Project, Multi-parent Advanced Generation Inter-Cross (MAGIC)-derived lines and wild rice introgression lines, were also characterized to identify valuable donors for grain Zn and used in genetic dissection studies (Bandillo et al. 2013; Swamy et al. 2018a; Descalsota et al. 2018; Zaw et al. 2019). Moreover, large scope exists for revisiting gene banks to screen for high grain Zn and other beneficial elements using advanced high-throughput phenotyping technologies. Similarly, a systematic effort to collect and characterize heirloom rice for nutritional value in partner countries will help in breeding for improved nutrition.

4.2 Association Between Yield and Zn

The development of high-yielding Zn-biofortified rice with a combination of desirable agronomic traits and tolerance of pests and diseases is a must for their successful adoption and consumption. Both yield and grain Zn are genetically complex traits and are hugely influenced by external environmental factors (Zaw et al. 2019; Descalsota-Empleo et al. 2019a). In most cases, a negative association was reported between grain-Zn content and yield, and in a few specific germplasm accessions and populations a nonsignificant negative relationship or no relationship was reported (Gregorio 2002; Norton et al. 2010; Morete et al. 2011; Anandan et al. 2011; Nha 2019). Under different soil Zn conditions and in a set of different aromatic accessions and landraces, a positive relationship between grain Zn and yield was reported (Wissuwa et al. 2008; Gangashetty et al. 2013; Sathisha 2013). Therefore, for the identification of stable high-Zn donor lines with higher or more acceptable yield, the use of appropriate breeding methods and selection strategies is needed to successfully combine yield and grain Zn.

4.3 Molecular Dissection of Grain Zn

4.3.1 QTLs and Meta-QTLs Associated with Grain Zn

Zn uptake, transport, and accumulation in the grain are governed by a complex network of quantitative trait loci (QTLs) and genes. A comprehensive review of QTLs identified for grain Zn was carried out and detailed discussion presented by Swamy et al. (2016). Several QTLs with moderate to high phenotypic variance were reported for grain Zn on all 12 chromosomes of rice. At IRRI, our group has also carried out several QTL mapping studies using biparental and multiparental populations and germplasm collections (Table 1). Swamy et al. (2018a) reported eight QTLs for grain-Zn content. All of these QTLs were distributed across the rice genome, having the lowest frequency (one QTL) on chromosomes 1, 9, and 11 and the highest frequency (seven QTLs) on chromosome 12. Chromosome 7 had the second highest number (six) of QTLs. However, the QTLs on chromosomes 7 and 12 were consistent over different backgrounds and environments. The QTLs detected on chromosome 7 contributed 5.3–35.0% of the phenotypic variance for grain-Zn content in different backgrounds, while the QTLs on chromosome 12 contributed 9–36% (Swamy et al. 2016, 2018a). In another study, Swamy et al. (2018b) detected nine QTLs responsible for Zn on chromosomes 2, 3, 6, 8, 11, and 12 through two doubled-haploid (DH) populations derived from crosses of PSBRc82 × Joryeongbyeo and PSBRc82 × IR69428. Recently, association mapping experiments using diversity panels for grain Zn led to the identification of seven QTLs on chromosomes 1, 2, 4, 6, 7, and 12 by Descalsota et al. (2018) and three QTLs on chromosomes 1, 5, and 7 by Zaw et al. (2019). All of these findings show that numerous QTLs for Zn highlight the genetic complexity of this trait.

Table 1 QTLs identified for grain-Zn content in rice
Full size table

Meta-QTL analysis provides consolidated, precise, and smaller confidence intervals for multiple QTLs reported for a trait (Goffinet and Gerber 2000; Arcade et al. 2004; Swamy et al. 2011). Jin et al. (2015) identified 22 meta-QTLs on ten different chromosomes for grain-Zn content (rMQTLs). Similarly, Raza et al. (2019) carried out meta-QTL analysis of grain-Zn QTLs reported from 24 mapping populations and three diverse germplasm sets and identified 46 MQTLs. Seven meta-QTLs (rMQTL2.1, rMQTL4.4, rMQTL6.4, rMQTL8.2, rMQTL8.3, rMQTL8.4, and rMQTL12.4) were found to be common between two studies (Jin et al. 2015; Raza et al. 2019). In another study, 208 QTLs for grain Zn from 26 studies were projected on the consensus map and eventually 45 meta-QTLs were identified (Soe 2020). Overall, the confidence intervals of all the MQTLs were narrower vis-à-vis the mean values of the original QTLs. Several consistent QTLs and associated markers were identified , which are useful for efficient marker-assisted selection (MAS) programs. In addition, precise meta-QTL regions provide an opportunity to shortlist candidate genes for further functional validation.

4.3.2 Network of Metal Homeostasis Genes

Mapping of major-effect QTLs/genes for grain Zn and understanding their molecular basis can fast-track the development of Zn-biofortified rice through MAS. The genomic regions of important QTLs associated with grain Zn identified in numerous studies contained multiple hypotheticals and functionally annotated genes that function as metal chelators and ion transporters. A list of important genes associated with Zn homeostasis in rice is summarized in Swamy et al. (2016). Rice roots produce chemicals that free up mineral elements from the soil complex and promote their root uptake from the soil (Widodo et al. 2010; Nozoye et al. 2011). Several genes/gene families are involved in biosynthesis of phytosiderophores, mineral uptake, transport, and loading such as OsDMAS, OsSAMS, OsNAS, OsTOM1, and OsNAAT (Inoue et al. 2003, 2008; Bashir et al. 2006; Johnson et al. 2011). Zinc finger transcription factors such as OsZIP1, OsZIP3, OsZIP4, OsZIP5, and OsZIP9 are major Zn transporters within the rice plant (Ramesh et al. 2003; Ishimaru et al. 2005; Lee et al. 2010a, b). In separate studies conducted using connected populations, ZIP family genes such as OsZIP5 and OsZIP9 were identified along with another 140 candidate genes (Nha 2019). In a study using DH populations, OsZIP6 was identified as a primary candidate gene associated with grain Zn (Calayugan et al. 2020). Similarly, OsVIT and OsYSL family genes are involved in Zn transport across the tonoplast and phloem, respectively (Sasaki et al. 2011; Kakei et al. 2012; Zhang et al. 2012; Lan et al. 2013).

The well-characterized Zn metal homeostasis genes can be manipulated through genetic engineering to improve grain-Zn content in rice (Trijatmiko et al. 2016). The advanced genome editing techniques using zinc-finger nucleases (ZFNs), transcription activator-like effector nucleases (TALENs) , and clustered regularly interspaced short palindromic repeats (CRISPR)/Cas systems can be used to induce modifications at specific genomic loci (Kim et al. 1996; Christian et al. 2010; Jinek et al. 2012; Chen and Gao 2013; Gao 2015). The Zn homeostasis genes can be major target sites for genome editing to improve grain-Zn content in rice.

4.3.3 Bayesian Network Analysis of Grain Yield and Zn

A Bayesian genomic prediction network (BN) provides valuable information on interactions between multiple traits and SNP markers and helps to establish relationships among them. It clearly depicts the strength and direction of associations among traits and SNP markers. In a way, it helps to validate the QTLs, genes, or trait associations identified by genome-wide association analysis (Zaw et al. 2019). In a MAGIC Plus population, a BN was used among Zn- and yield-related traits. The results clearly showed a complex relationship among traits (Fig. 1). Among the agronomic traits studied for their relationship with grain Fe and Zn, only panicle length had a direct effect on Fe and Zn content in rice (Descalsota et al. 2018). Zaw et al. (2019) conducted BN analysis in a global MAGIC population using 8110 SNP markers and 16 traits, including grain Zn. At a BN strength of more than 0.5, strong direct associations were reported among traits such as yield → zinc, zinc → filled grains, iron → zinc, and iron → grain length. Zn was associated with eight markers for each of the traits. In general, Fe and Zn content have strong positive correlations, thus providing huge opportunities to improve both minerals together. It is interesting that in both BN studies there was no direct effect of yield on Zn, indicating that combining high yield potential and high grain-Zn content is possible in order to develop successful Zn-biofortified rice varieties. We emphasize that there is a need to thoroughly dissect the influence of panicle length on grain Zn. It is commonly observed that increased yield dilutes Zn content, which results in negative correlations between these traits. There is therefore a need to make adjustments for grain-Zn mapping studies (McDonald et al. 2008).

Fig. 1
figure 1

Bayesian network analysis of grain Zn and agronomic traits. Note: Zn zinc, Fe iron, HT plant height, DF days to flowering, NT number of tillers, PT productive tillers, GL grain length, GW grain width, TGW thousand-grain weight, YLD grain yield

Full size image

4.4 Multi-Trait Genomic Selection for Zn Biofortification

QTL mapping and GWAS methods are routinely used for molecular dissection of complex traits; however, they have limited power in detecting minor-effect loci (Bernardo 2008; Collard and Mackill 2008; Ben-Ari and Lavi 2012). In contrast, genomic selection (GS) considers genome-wide effects, including both major and minor loci, and thereby assesses the genomic estimated breeding values (GEBVs) of breeding lines (Meuwissen et al. 2001). With the recent advances in statistics, deep machine learning models are helpful in accurately estimating GEBVs and their cross-validation in training or reference sets (Montesinos-López et al. 2019). Since GS captures total genetic variance, it addresses the existing limitations of GWAS and QTL mapping to improve traits (de los Campos et al. 2009). In addition, it speeds up selection cycles, which enhances annual genetic gain and saves cost significantly (Shamshad and Sharma 2018). Therefore, great opportunity exists for employing GS-related strategies that capture both major- and minor-effect alleles to increase the genetic gain for grain yield and grain-Zn content in rice.

Selection for higher yield and other desirable agronomic traits along with high grain-Zn content is an integral part of Zn biofortification; however, both yield and Zn are genetically complex and difficult to manipulate or simultaneously improve (Garcia-Oliveira et al. 2018; Zaw et al. 2019). The rate of genetic gain for grain yield becomes stagnant at ~1% yearly. This is not sufficient to meet future demand for rice, not to mention the strong impacts of complex genetic architecture and genotype–environment interactions (Peng et al. 2000, 2004; Wassmann et al. 2009). The combined genetic gain for yield and Zn will be relatively inferior when compared with that for individual traits. Therefore, implementing multi-trait-based population improvement through genomic selection is an efficient approach.

In rice, Spindel et al. (2015) reported prediction accuracies of single-trait genomic selection (ST-GS) models for grain yield at 0.31, while Arbelaez et al. (2019) have shown predictive accuracies for grain yield at 0.36. Meanwhile, multi-trait genomic selection (MT-GS) models have illustrated higher predictive abilities than ST-GS models and the results are obvious, especially when low-heritability traits are paired with a genetically correlated secondary trait with higher heritability (Jia and Jannink 2012; Hayashi and Iwata 2013; Guo et al. 2014; Schulthess et al. 2016). Many findings have used MT-GS approaches in crop breeding, but not yet in rice. Schulthess et al. (2016) have confirmed the predictive ability of MT-GS in outperforming ST-GS pipelines for grain yield and protein content in rye. Lado et al. (2018) have verified combining two, three, and four traits in bread wheat in exploiting the benefits of MT-GS under different cross-validation scenarios. The use of correlated traits in MT-GS models gives the best prediction accuracies in a two-trait scenario . GS in maize showed higher prediction accuracy in DH populations than a GWAS panel using the same set of GBS and rAmpSeq markers, and GS outperformed MAS in predicting the performance of Zn content in maize (Guo et al. 2020). Although most of the available GS methods increased predictive ability, Zn breeders should target multiple independent phenotypes from multi-environments. Thus, multi-trait and multi-environment (MTME) models have been established to employ the information on multiple traits evaluated in multiple environments, which improves predictive ability compared to conventional, pedigree, and independent GS analysis (Montesinos-Lopez et al. 2016).

5 Development of High-Zn Rice

5.1 Phenotyping of Grain Zn

To enhance selection accuracy and to significantly improve a breeding program, reliable phenotyping is crucial. Accurate phenotyping for any trait involves a standard protocol with a set of specific standards. Usually in large biofortification breeding programs like the one at IRRI, we handle a huge number of breeding lines every season, so there is a need for quick turnover of materials with accurate phenotyping for grain Zn. Efficient high-throughput dehulling, milling, and Zn measurement protocols and equipment are needed for successful Zn biofortification of rice (Fig. 2) (Swamy et al. 2016; Guild et al. 2017). Several low-throughput qualitative, semi-quantitative, and quantitative methods are available for the estimation of grain-Zn content in rice and other cereals. Inductively coupled plasma optical emission spectrometry (ICP-OES) is used to assess nutrient density in grains (Zarcinas et al. 1987); this method is more accurate but low-throughput and input-intensive and it requires trained staff. X-ray fluorescence (XRF) is a rapid non-chemical-based method to measure grain-Zn content in milled rice, which has decreased cost per unit and simultaneously increased selection intensity although it still requires grain processing under a contamination-free environment of exogenous Zn sources.

Fig. 2
figure 2

Phenotyping for grain-Zn milled rice

Full size image

5.2 Setting a Zn Target for Rice Biofortification

The development of nutritional targets for crops for biofortification breeding was established by a group of experts taking into account the food habits of the target populations, nutrient losses during food processing, and nutrient bioavailability (Hotz and McClafferty 2007). The breeding target was designed to meet the specific nutrient requirement of the target populations considering the baseline micronutrients existing in popular rice varieties and extra micronutrient content to be added to the crop of interest. Zn-biofortified rice is expected to provide >40% of the Estimated Average Requirement, which is enough to help overcome Zn-deficiency-induced health risks (Bouis and Saltzman 2017). There is a plan to release high-Zn rice varieties in three phases: the first set of varieties will have an additional Zn content of 6–8 ppm, the second wave of varieties will have 8–12 ppm, while the third wave of high-Zn rice varieties will have 12 ppm of additional grain Zn (Fig. 3).

Fig. 3
figure 3

Zn target set for breeding Zn-biofortified rice varieties

Full size image

5.3 Germplasm Enhancement and Pre-breeding for Grain Zn

Exploitable genetic variability for any trait, its systematic characterization, and efficient use are essential for a successful breeding program. Most elite modern rice varieties and their closest elite genetic pool have low grain-Zn content (Gregorio 2002). Oryza nivara, O. rufipogon, O. longistaminata, and O. barthii accessions, landraces, colored rice, and aus and aromatic accessions were found to have rich grain-Zn content (Swamy et al. 2016, 2018a, b; Ishikawa et al. 2017). But these accessions may not be agronomically desirable because of their poor phenotype and lower yield. Therefore, a systematic pre-breeding for grain Zn is essential to develop high-Zn rice varieties.

The advanced backcross method for genetic dissection of wild rice, and for developing high-Zn introgression lines, is an attractive approach for efficient use of wild rice accessions (Balakrishnan et al. 2020). Several wild rice-derived introgression lines with high grain Zn and yield have already been developed by several groups (Ishikawa et al. 2017; Swamy et al. 2018a). Multi-parent-derived populations to select transgressive variants with a combination of desirable traits have yielded many desirable transgressive variants for grain Zn (Gande et al. 2013; Ishikawa et al. 2017; Descalsota-Empleo et al. 2019a, b). Marker-assisted QTL deployment, QTL pyramiding, and marker-assisted recurrent selection are helpful in germplasm enhancement for grain Zn with other traits (Hill et al. 2008; Boyle et al. 2017).

Several studies have characterized germplasm and advanced breeding lines for grain-Zn content (Gregorio et al. 2000; Brar et al. 2011). Garcia-Oliveira et al. (2009) identified 85 introgression lines with the highest quantities of Zn with a mean value of 27.1 ppm. Martínez et al. (2010) phenotyped grain-Zn content in 11,400 rice samples in both brown rice and milled rice and reported corresponding Zn values of 20–25 ppm and 16–17 ppm, respectively. Gande et al. (2013) identified eight transgressive lines for high Zn content (31.2–35.5 ppm). Some of these introgression lines, transgressive segregants, and breeding lines can be used as donor lines for Zn biofortification and even some can be directly tested and released as high-Zn rice varieties for commercial cultivation.

5.4 Mainstreaming of Zn Biofortification

Zinc-biofortified rice varieties have been successfully released for cultivation by farmers in some of the target countries. However, developing, releasing, and disseminating a few varieties may not create sustainable and wide-scale impact on human health. At IRRI, mainstream breeding programs are shifting from a siloed trait-based breeding approach to a modernized product development pipeline that effectively integrates the improvement of all traits necessary for market acceptance into a single variety replacement strategy. This new strategy involves using population improvement as a mechanism to drive genetic gain for complex traits, while simultaneously increasing the frequency of trait-favorable alleles. Essential to this strategy is the data-driven identification of a core set of elite lines that represent the pool of possible parental lines that can be used in the breeding process (Cobb et al. 2019). Through successive cycles of recurrent selection, mainstream breeding efforts are now able to drive genetic gain and improve the average value of all traits from a product profile in the entire elite gene pool simultaneously (Collard et al. 2017). By integrating selection for high grain-Zn content directly into the mainstream breeding effort, the mean value of grain Zn among the elite breeding lines will eventually be at or above the recommended allowance of 28 ppm in milled grains. Once this occurs, all the most recently developed new varieties released from the mainstream breeding program will have acceptable concentrations of Zn in addition to other traits more valued in the marketplace. With minimal effort, maintenance breeding for Zn can be conducted in elite breeding programs once acceptable grain-Zn content is achieved in order to ensure constant delivery of sufficient Zn to the diets of nutrition-vulnerable rice-consuming populations. Incumbent upon this strategy is the need for sufficient variation to drive genetic gain for complex traits. A three-phased approach is suggested: elite germplasm characterization, elite germplasm enhancement and selection, and mainstream breeding.

6 High-Zn Rice Testing and Release

6.1 Genotype × Environment Effects on Grain Zn

Grain-Zn content is a complex trait found to be significantly influenced by external soil and climatic factors (Chandel et al. 2010; Anuradha et al. 2012; Swamy et al. 2016; Naik et al. 2020). Meteorological factors such as temperature, relative humidity, and rainfall; soil factors such as organic matter, pH, and nutrient status; and agronomic practices such as fertilizer application, tilling, cultivation system, and irrigation (White and Broadley 2009; Joshi et al. 2010; Chandel et al. 2010) need to be taken into account. Grain-Zn content in a study conducted by Wissuwa et al. (2008) was found to be greatly influenced by the native Zn in soils, genotype, and Zn fertilizer application. In a separate study by Wang et al. (2014), water management with alternate wetting and drying (AWD) together with ZnSO4 fertilization showed a positive response for higher yield coupled with higher grain-Zn content in rice. Pandian et al. (2011) conducted field experiments across three locations involving 17 genotypes of rice. The results showed that grain-Zn content varied significantly among the genotypes and locations.

Thus, G × E testing is needed to evaluate promising germplasm and the stability of mineral accumulation across generations and at multiple test sites (Gregorio 2002; Wissuwa et al. 2008; Impa et al. 2013; Naik et al. 2020). Hence, the stability of Zn-biofortified genotypes for grain-Zn content in addition to grain yield is essential for commercial release as varieties.

6.2 Zn-Biofortified Rice Varieties Released in Different Countries

Breeding efforts to biofortify rice with high grain Zn have resulted in the successful release of several high-Zn rice varieties in several countries of Asia. Five high-Zn rice varieties (BRRI dhan62, BRRI dhan64, BRRI dhan72, BRRI dhan74, and BRRI dhan84) have been released for cultivation in Bangladesh. In India, two high-Zn rice varieties (DRR Dhan45 and Chhattisgarh Zinc Rice-1) are available for farmers and consumers. Similarly, NSIC Rc 460 and Inapari Nutri Zn have been released for farmers’ cultivation in the Philippines and Indonesia, respectively. All these high-Zn rice varieties have higher grain-Zn content along with desirable agronomic traits and tolerance of biotic and abiotic stresses (Swamy et al. 2016; Tsakirpaloglou et al. 2019). Several promising high-Zn lines have been successfully tested in Myanmar and Cambodia and in some African countries. We are also making efforts to develop the next wave of Zn-biofortified rice varieties with higher grain-Zn content.

6.3 High-Zn Rice Traceability and Product Control

Grain-Zn content in rice is an invisible nutritional trait and no morphological indicators differentiate Zn-biofortified rice from market rice. Maintaining the product integrity of high-Zn rice throughout the value chain is an important component of successful Zn biofortification programs. Close monitoring, supervision, and quality control are necessary with proper certification, labeling, branding, and tracing of the product (www.fao.org/tempref/codex/Meetings/CCNFSDU/ccnfsdu36/nf36_11e.pdf). The development of Zn product-specific molecular marker-based fingerprints and rapid qualitative biochemical marker kits will also help in tracing Zn-rich rice. Blockchain technology is being used in the large-scale dissemination of nutritious crops to ensure quality control and to deliver the right products to consumers (Tripoli and Schmidhuber 2018).

7 Next-Generation Multi-Nutrient Rice Varieties

Breeding for rice varieties with multiple beneficial minerals and vitamins is essential to develop them holistically for one biofortified rice product. Efforts to develop rice varieties with high Zn, high Fe, selenium, vitamin A, proteins, amino acids, etc., should be given a priority. It will also be interesting to combine high nutrient content with traits beneficial to health, such as low glycemic index, antioxidants, and resistance starch. Also, there is a need to diminish the amount of harmful elements such as arsenic and cadmium. The increase in demand for rice varieties with improved grain quality and nutrition means that a suite of rice varieties with different combinations of traits targeted to different regions should be developed.

8 Conclusions

Biofortification of rice with improved Zn content is an efficient means to tackle Zn malnutrition in predominantly rice-consuming developing countries. Some success has been achieved in understanding the molecular basis of Zn accumulation and the effects of G × E, and finally in developing and releasing Zn-biofortified rice varieties for the target countries. In all, ten high-Zn rice varieties have been released in four Asian countries. Efforts have begun to mainstream grain Zn to ensure that the Zn trait becomes an integral part of future varieties. Huge scope exists to apply advanced genomics technologies such as genomic selection and genome editing to speed up high-Zn varietal development. An efficient rice value chain for Zn-biofortified varieties, quality control, and promotion are essential for successful adoption and consumption. The development of next-generation high-Zn rice varieties with higher grain-Zn content, stacking of multiple nutrients, along with good grain quality and acceptable agronomic traits has to be fast-tracked. Healthier rice has a large demand from all stakeholders, so we need to keep up the pace of developing nutritious rice to meet the demand and to achieve nutritional security.