Nitrogen Fertiliser Effects on Grain Anthocyanin and γ-Oryzanol Biosynthesis in Black Rice

Abstract

Accumulation of phytochemicals in vegetative tissue under nitrogen (N) stress as an adaptive strategy has been investigated in various crops, but the effect of applied N on grain phytochemicals is poorly understood. This study investigated the effect of applied N on the biosynthesis and accumulation of rice (Oryza sativa L.) grain anthocyanin and γ-oryzanol under different ultraviolet-B (UV-B) conditions in a controlled pot trial using two distinct black rice genotypes. The response of grain anthocyanin and γ-oryzanol content to applied N was genotype-dependent but was not altered by UV-B conditions. Applied N increased grain anthocyanin and decreased γ-oryzanol content in genotype SCU212 but had no significant effect in genotype SCU254. The expression of the OsKala3 regulatory gene was significantly upregulated in response to applied N in SCU212, while the expressions of OsKala4 and OsTTG1 were unchanged. The expression of all three regulatory genes was not significantly affected in SCU254 with applied N. Key anthocyanin biosynthesis genes were upregulated in grain by N application, which indicates that the common increase in anthocyanin in vegetative tissues under N deprivation does not hold true for reproductive tissues. Hence, any future approach to target higher content of these key phytochemicals in grains should be genotype-focused.

1. Introduction

Pigmented black grains are receiving increased attention in the food industry due to the demonstrated health benefits associated with their levels of anthocyanins and other antioxidant compounds [1]. Pigmented black rice (Oryza sativa L.) has higher concentrations of anthocyanin than other cereals [2], with typical concentrations in the grain of around 3 mg/g [3]. Anthocyanins have been reported to prevent sclerosis, improve eyesight, reduce cancer cell proliferation, lower the incidence of obesity and diabetes, and reduce inflammation [4,5,6]. Rice is the most widely consumed natural source of γ-oryzanol, and black rice has comparatively higher concentrations of γ-oryzanols than nonpigmented rice [7]. γ-Oryzanols are reported to lower cholesterol and have anti-inflammatory, antioxidant, antidiabetic, neuroprotective, and anticarcinogenic properties [8,9].

In rice grain, the concentrations of anthocyanin and γ-oryzanol are affected by genetic and environmental factors, including UV radiation, salinity, water availability, temperature, and carbon dioxide (CO₂) levels [10]. It is well documented that nitrogen (N) fertiliser application affects rice yield [11] and quality [12]. Nitrogen is an integral component of chlorophyll, amino acids, ATP, plant hormones, coenzymes, and N-containing secondary metabolites [13]. On the one hand, it may be expected that N deprivation would increase anthocyanin and γ-oryzanol concentrations in pigmented rice grains since the accumulation of these defence-related metabolites is a typical response to low N stress in leaves [14,15]. Competition for nutrients between primary and secondary metabolite pathways under severe N deprivation may limit anthocyanin and γ-oryzanol biosynthesis, even though neither are N-containing compounds. Studies in rice have found that N application can increase [16,17] or have no effect [16,18] on anthocyanin concentrations in the pericarp of the grain. The inconsistent results across studies may be due to different environmental conditions (e.g., temperature or UV-B levels) or genotypes investigated, but it highlights that knowledge of N impacts on anthocyanin accumulation in rice grains is incomplete. To the best of our knowledge, only one study has examined the effect of N fertiliser on γ-oryzanol in rice, with the authors indicating that the highest γ-oryzanol concentrations in the grain (161 mg/kg DW) occurred under N starvation, while concentrations under a range of N fertiliser rates varied from ~93 to 115 mg/kg DW [19].

UV-B radiation has been reported to affect anthocyanin concentration in leaves and fruit of a range of plant species [20,21,22,23]. This includes the leaves of rice, where exposure to UV-B radiation increased anthocyanin concentration in young plants [24,25]. Further, levels of phenylalanine lyase (PAL), an enzyme precursor for anthocyanin biosynthesis, increased four-fold in the leaf upon 12 h of exposure to UV-B radiation (0.6 W m²/s) in 6-day-old rice seedlings [26]. The link between anthocyanin concentrations in vegetative tissues and UV-B radiation is unsurprising, given that anthocyanins play a vital role in photoprotection by filtering UV-B and protecting cells from photodamage [25]. However, while UV-B is widely reported to impact anthocyanins in vegetative tissues, little is known about the effect of UV-B radiation on anthocyanin concentrations in rice grains. Similarly, we are unaware of any studies that have investigated the impact of UV-B radiation on γ-oryzanol concentrations in rice grains, despite evidence that γ-oryzanol is effective at absorbing both UV-A and UV-B when used in cosmetics [27].

In rice grains, anthocyanins are concentrated in the pericarp layer [28], and corresponding biosynthetic (structural) genes involved in anthocyanin biosynthesis have been well studied. Early biosynthetic genes involved in anthocyanin biosynthesis include chalcone synthase (CHS) and flavanone 3-hydroxylase (F3H), while core late anthocyanin biosynthetic genes are dihydroflavonol 4-reductase (DFR), anthocyanidin synthase (ANS), and UDP-glucose: flavonoid 3-O-glucosyltransferase (UFGT) [29]. These genes encode enzymes directly involved in the formation of anthocyanin, and their transcription is controlled by the MBW (MYB-bHLH-WD40) complex consisting of the MYB-type TFs, basic helix-loop-helix (bHLH) transcription factors (TFs) and WD-repeat proteins (WD40) [30,31]. Although MBW genes are highly conserved in their protein sequence across different plant species [32], different MBW complex genes are responsible for anthocyanin biosynthesis in leaves, pericarp, husks, culms, and awns in rice [30]. In rice, OsKala3 and OsC1 are the pericarp and leaf-sheath specific MYB TFs, respectively [33]. Likewise, OsKala4 and OsRb are the bHLH TFs for pericarp and leaf sheath, respectively [34]. Although the effect of N on regulatory genes for anthocyanin biosynthesis has been studied in berries and leaves [35,36], N effects on pericarp-specific regulatory genes for anthocyanin biosynthesis in rice have not been explored.

We hypothesised that application of N may have a significant negative effect on anthocyanin and γ-oryzanol biosynthesis in rice grain, and the effect may be genotype-dependent and further affected by UV-B radiation. To test this hypothesis, a controlled environment pot trial was set up to investigate grain anthocyanin content and the expression of regulatory and structural genes for anthocyanin biosynthesis in response to N in distinct black rice genotypes in the presence and absence of UV-B radiation. The effect of N on grain γ-oryzanol concentrations under different UV-B conditions was also quantified.

2. Materials and Methods

Two japonica black rice varieties, SCU212 and SCU254, were selected to test the effects of N on grain anthocyanin and γ-oryzanol biosynthesis under UV-B and non-UV-B conditions. These two genotypes were chosen from a black rice collection obtained from the International Rice Research Institute (IRRI) Los Baños, the Philippines, based on different levels of key secondary metabolites. SCU212 (Accession ID 125541) is a North Korean variety with comparatively higher anthocyanin and γ-oryzanol content than the Chinese variety, SCU254 (Accession ID 82240) [37]. A controlled-environment pot trial was conducted in a greenhouse at Southern Cross University, Lismore, NSW, Australia. The roof of the greenhouse was made out of polycarbonate sheets, which block 100% of external UV-B radiation.

2.1. UV-B Light Setup

Polycarbonate sheets were used to divide the glasshouse into two independent compartments: UV-B and non-UV-B. Two Phillips UV-B broadband lamps (TL 40 W/12 RS SLV 25, 102 V, 40 W) (GMT Lighting, Hawthorn, VIC, Australia) were fitted above two benches on the UV-B side of the glasshouse, maintaining a distance of 50 cm from the top of the plants. A Solarmeter Model 6.2 Sensitive UV-B meter (Solar Light Company, Inc., Glenside, PA, USA) was used to measure irradiation on the top of the rice plants, and UV-B radiation was maintained in the range of 150 to 250 µW/cm² throughout the growing period by adjusting the height of the UV-B lamps.

2.2. Soil Collection and Pot Set Up

Soil from the 0–15 cm layer was collected from the Brookside field site of Southern Cross University, Lismore campus. Further details of the field site and chemical characterisation of the soil can be found in [38]. The soil was sieved to 4 mm to remove large particles and foreign material. Nondraining 4 L pots were filled with 2.5 kg of soil, leaving a head space of 10 cm. A day before transplanting, 0.5 g triple super phosphate and 0.3 g of sulphate of potash were added to all pots. Two N treatments, +N (1 g urea applied to appropriate pots prior to transplanting) and −N (no urea), were implemented with five experimental replicates per treatment.

2.3. Plant Growth

Seeds of SCU212 and SCU254 were germinated on a floating mesh above deionised water. After a week, deionised water was replaced by a full-strength Yoshida solution [39], which was changed weekly. Three-week-old, evenly-sized rice seedlings were transplanted into the soil with one seedling per pot. Rice was grown under flooded (anaerobic) conditions for the duration of the experiment.

For UV-B acclimatisation, UV-B lights were set up to “turn on” for 4 h of UV-B radiation with 15 min “off time” in between each hour (10 am to 3 pm) at 1 week after transplanting. When plants were 1 month old (around the tillering stage), UV-B lights were turned on continuously for 4 h from 10 am to 2 pm when UV-B intensity was highest outside the glasshouse.

Within each compartment (UV-B and non-UV-B), pots from each N treatment were arranged in a completely randomised design for each genotype, and pots were rerandomised every second day. Genotypes were not mixed because of differences in plant height so as to maintain the same distance between plants and the UV-B lamps throughout the growing period.

A temperature data logger was placed on each bench in the four compartments (two UV-B treatments and two genotypes) to monitor temperature throughout the experiment. Temperatures in the greenhouse ranged from 20 °C to 32 °C throughout the experiment. Photosynthetically active radiation (PAR) was measured from 11 am to 2 pm each day using a PAR meter (Apogee, MQ500) (Apogee Instruments, Inc. Logan, UT, USA) to ensure both UV and non-UV blocks received similar PAR, which ranged from 900 to 1100 µmol/m²/s on sunny days.

2.4. Tagging and Harvesting of Grain Samples

An earlier study by [37] reported that maximum anthocyanin concentrations/content in both SCU212 and SCU254 occurred around 15 to 20 days after flowering (DAF), whereas γ-oryzanol concentrations continued to increase throughout grain development. Grain anthocyanin and γ-oryzanol concentrations were therefore quantified at 20 DAF (dough stage) in both genotypes in the present study. The expression of regulatory and structural genes involved in anthocyanin biosynthesis was examined at 10 DAF.

All rice panicles were tagged when at least 50% of florets had visible anthers. At 20 DAF, tagged panicles were cut, and grains were quickly separated from panicles and stored at −20 °C for future anthocyanin, γ-oryzanol, carbon (C), and N quantification. Grain samples were also collected at 10 DAF, snap-frozen in liquid N and stored at −80 °C to look at regulatory and structural gene expression.

Morphological data, including number of tillers, grain yield per pot, straw biomass, seed size, and plant height, were recorded at final harvest (maturity) at 30 to 35 DAF.

2.5. Sample Preparation and Extraction of Anthocyanin and γ-Oryzanol

Samples stored at −20 °C were freeze-dried to remove moisture without compromising seed anthocyanin and γ-oryzanol concentrations. Grains were dehusked and ground to powder before extracting anthocyanin and γ-oryzanol, as described in [37].

Briefly, a subsample of ~50 mg finely ground grain was extracted in a 15 mL falcon tube with 10 mL of methanol acidified with 1 N HCl (85:15, v/v) for anthocyanin extraction. For the γ-oryzanol extraction, a subsample of ~50 mg finely ground grain was extracted with 1 mL absolute methanol in a microcentrifuge tube. The mixture was sonicated, and the crude extract was centrifuged at ~12,000× g for 10 min at room temperature. The resultant supernatant (~800 µL) was pipetted into 2 mL Agilent HPLC vials (Agilent, Santa Clara, CA, USA) and stored at −20 °C until analysis using high-performance liquid chromatography (HPLC).

2.6. Analysis of Anthocyanin and γ-Oryzanol Concentration

Anthocyanin and γ-oryzanol concentrations were measured using an Agilent 1260 Infinity II-HPLC instrument (Agilent Technologies, Palo Alto, CA, USA) equipped with an auto sample injector (G7129C), vacuum degasser, quaternary UHPLC 1260 flexible pump (G7104C), and diode array detector (DAD, 1260) as [37]. An Agilent 1.8 µm ZORBAX Extend-C18 reverse phase column (Agilent, Santa Clara, CA, USA) was used for both assays. The mobile phases used for anthocyanin quantification were 100% MilliQ water with 0.05% trifluoroacetic acid (TFA) (Solvent A) and 100% acetonitrile with 0.05% TFA (Solvent B). For γ-oryzanol, 60% acetonitrile + 40% water in 10 mM ammonium formate and 0.1% formic acid (as solvent A), and 10% acetonitrile + 90% isopropanol in 10 mM ammonium formate and 0.1% formic acid (as solvent B) were used. The same mobile phase gradients were used as described in [37] for anthocyanin and γ-oryzanol quantification at a flow rate of 0.3 mL/min. The injection volume of the sample was 3 µL, and a column temperature of 40 °C was set for both analyses. The UV absorbance was detected at 520 nm for the major anthocyanin cyanidin 3-O-glucoside (C3G) and 325 nm for γ-oryzanol. The standard for C3G (as Cl salt) was purchased from Phytolab GmbH & Co.KG, Vestenbergsgreuth, Germany and the standard for total γ-oryzanols was obtained from Selleck Chemicals, Houston, TX, USA.

2.7. Carbon and Nitrogen Determination

Concentrations of C and N in grain and straw samples were measured using an LECO TruMAC CNS analyser (LECO Corporation, St. Joseph, MI, USA) at the Environmental Analysis Laboratories (EAL), Southern Cross University, Lismore, Australia, using methods from Rayment, G. and D. Lyons [40].

2.8. RNA Extraction and qPCR

Total RNA was extracted from seed samples harvested before peak anthocyanin stage (10 DAA) for both genotypes as previously described in [37], using Fruit-mate solution for RNA purification (Takara, Otsu, Japan, #9192), the TRIZol^TM Reagent for RNA extraction (Invitrogen, Carlsbad, CA, USA, #15596018) and the RNA purified using Direct-zol^TM RNA Miniprep kit (Zymo Research, Irvine, CA, USA, #R2052). The QuantiTect Reverse Transcription kit (Qiagen, Hilden, Germany, #205313) was used to synthesise first strand cDNA from 1 µg of total RNA, and qPCRs were performed using QuantiNova SYBR Green PCR kit (Qiagen, Hilden, Germany, #208057) and a QIAquant^TM real-time PCR cycler (Qiagen, Hilden, Germany), according to manufacturer instructions. The PCR cycling parameters included 2 min at 95 °C, 40 cycles of 5 s each at 95 °C, and 10 s at 60 °C. All reactions were carried out in a qPCR 96-well skirted plate (Qiagen, Hilden, Germany, #209002).

Primers for the three regulatory genes (OsKala3, OsKala4, and OsTTG1) controlling the expression of anthocyanin biosynthesis genes and five biosynthetic genes (OsCHS, OsF3H, OsDFR, OsANS, and OsUFGT) in rice as used by [33] were obtained from Sigma Aldrich, St. Louis, MO, USA, along with primers for the housekeeping gene ubiquitin (OsUBI). Three distinct biological replicates and three technical replicates were performed for each sample. The relative expression of genes in the +N treatment was normalised to the cycle threshold (Ct) value of housekeeping gene (OsUBI) and Ct of the same gene of interest in the −N treatment as a control using the 2^−ΔΔCt method [41].

2.9. Statistical Analysis

R studio was used to calculate means of all data and to perform one-way analysis of variance (ANOVA) for each genotype under plus and minus UV-B conditions, and to test for significant differences between +N and −N treatment means.

3. Results

3.1. Effect of Nitrogen on Yield Components and Grain Traits under Different UV-B Conditions

Nitrogen treatment had a significant effect on almost all of the measured morphological traits at grain maturity (35 DAF) for both genotypes regardless of UV-B conditions (

Table 1. Plant height, tiller number, grain yield, shoot biomass, grain moisture % and grain size for SCU212 and SCU254 black rice genotypes under +N and −N fertiliser treatments with and without UV-B radiation.

		SCU212			SCU254
	Traits	−N	+N	p-Value	−N	+N	p-Value
With UV	Plant height (cm)	69	74.6	*	50.6	58.2	**
	No. of tillers per plant	3.6	7.8	**	2.5	12	**
	Grain yield (g/plant)	3.51	10.17	**	2.15	22.81	**
	Straw biomass (g/plant)	6.67	16.8	**	7	31.21	**
	Grain moisture (%)	21.3	23	ns	22.1	29.1	**
	Grain size (g/100 grains)	1.81	1.91	*	2.24	2.2	ns
Without UV	Plant height (cm)	78	87.4	**	51.25	62.4	**
	No. of tillers per plant	3.2	8.2	**	2.6	11.2	**
	Grain yield (g/plant)	3.04	13.02	**	2.45	22.64	**
	Straw biomass (g/plant)	7.25	22.1	**	4.77	24.14	**
	Grain moisture (%)	22.1	26.3	*	24.5	30	**
	Grain size (g/100 grains)	1.63	1.73	**	2.35	2.18	*

Note: ns, nonsignificant difference (p > 0.05); * and ** indicate significant difference at p < 0.05 and p < 0.01, respectively.

). In SCU212, N application increased tiller number by 2-to-2.5-fold, grain yield by 3-to-4-fold, and shoot biomass by 2.5-to-3-fold, irrespective of UV-B conditions. A significant increase in plant height and grain size due to N fertiliser was also observed under both UV conditions. In SCU254, N application resulted in an increase in tiller number by 4.5-to-5-fold, grain yield by 9-to-10.5-fold, and shoot biomass by 4.5-to-5-fold (Table 1). Plant height and grain moisture were also significantly higher in +N as compared to −N.

3.2. Anthocyanin and γ-Oryzanol Response to Nitrogen Fertiliser under Different UV-B Conditions in Rice Grains at the Dough Stage

In SCU212, the application of N resulted in a significant increase in cyandin-3-O-glucoside (C3G) concentration by 20% and 23%, in the presence or absence of UV-B radiation, respectively. A similar trend was observed for grain C3G content. In contrast, in SCU254, the application of N had no significant effect on grain C3G concentration or content, regardless of UV-B treatment conditions (

Table 2. C3G concentration and content, γ-oryzanol concentration and content for SCU212 and SCU254 black rice genotypes under +N and −N fertiliser treatments with and without UV-B radiation.

		SCU212			SCU254
		−N	+N	p-Value	−N	+N	p-Value
With UV	C3G concentration (mg/100 g)	758.34	913.77	*	575.03	593.31	ns
	γ-Oryzanol concentration (mg/100 g)	58.08	52.87	*	40.63	39.55	ns
	C3G content (mg/100 grains)	12.25	15.97	*	11.64	12.27	ns
	γ-Oryzanol content (mg/100 grains)	0.96	0.89	*	0.82	0.8	ns
Without UV	C3G concentration (mg/100 g)	666.61	819.25	**	592.85	614.51	ns
	γ-oryzanol concentration (mg/100 g)	55.19	51.78	**	48.97	48.54	ns
	C3G content (mg/100 grains)	10.9	12.76	*	12.34	12.48	ns
	γ-Oryzanol content (mg/100 grains)	0.89	0.81	**	0.77	0.76	ns

Note: ns, nonsignificant difference (p > 0.05); * and ** indicate significant difference at p < 0.05 and p < 0.01, respectively.

).

Unlike C3G, grain γ-oryzanol concentration significantly decreased with the application of N in SCU212 under both UV-B treatment regimens: from 58.1 to 52.9 mg γ-oryzanol/100 g DW with UV-B radiation and from 55.2 to 51.8 mg γ-oryzanol/100 g DW in the absence of UV-B. Grain γ-oryzanol content followed a similar trend to γ-oryzanol concentration in SCU212 in response to N application. In SCU254, there was no significant effect of N on grain γ-oryzanol concentration or content, irrespective of UV-B conditions. However, γ-oryzanol concentrations were around 20% higher in the absence of UV-B radiation (around 48 mg/100 g DW across N treatments) compared to grain collected from plants that received UV-B radiation (around 40 mg/100 g DW across N treatments). γ-Oryzanol content in the grain differed by less than 10% under the different UV-B and N treatments (Table 2).

3.3. Carbon and Nitrogen Status in Grains at the Dough Stage in Response to Nitrogen Fertiliser under Different UV-B Treatments

Application of N significantly increased the N concentration and decreased the C/N ratio in the grains of SCU212 grown in the presence or absence of UV-B (

Table 3. N concentration, total carbon/total nitrogen (C/N), and N content in grains for SCU212 and SCU254 black rice genotypes under +N and −N fertiliser treatments with and without UV-B radiation.

		SCU212			SCU254
	Traits	−N	+N	p-Value	−N	+N	p-Value
With UV	N concentration (g/100 g)	1.42	2.03	**	1.336	1.388	ns
	C/N ratio	29.82	21.08	**	31.7	30.77	ns
	N content (g/100 grains)	0.023	0.035	**	0.027	0.028	ns
Without UV	N concentration (g/100 g)	1.3	1.81	**	1.336	1.394	ns
	C/N ratio	32.76	23.8	**	32.14	30.59	ns
	N content (g/100 grains)	0.02	0.029	**	0.027	0.028	ns

Note: ns, nonsignificant difference (p > 0.05), ** indicate significant difference at p < 0.01.

). The application of N fertiliser resulted in ~40% increase in grain N concentration in SCU212, while the grain C/N ratio decreased by ~27 to 29% under both UV-B treatments. A similar trend was observed for N content in 100 grains for SCU212, with and without UV-B, where total N content per 100 grains increased by 45 to 50%, irrespective of UV-B conditions. In genotype SCU254, N treatment had no significant effect on either the grain N concentration or C/N ratio. Grain N concentration was ~1.3 g/100 g, and C/N ratio was ~30.5 to 32 in SCU254 under both N treatments, regardless of UV-B treatment (Table 3).

3.4. Expression of the Regulatory and Structural Genes for Anthocyanin Biosynthesis in the Grain in Response to Nitrogen Treatments at 10 DAF

In genotype SCU212, regardless of UV-B treatment, the relative expression level of OsKala3, one of the three subunits of the regulatory MBW complex that controls anthocyanin biosynthesis [33], was significantly higher with N application than without N (Figure 1a,b). The relative expression levels of other regulatory genes OsKala4 and OsTTG1 were not significantly affected by N application, regardless of UV-B treatment in this genotype. The expression levels of both early anthocyanin biosynthetic genes (OsCHS and OsF3H) and late anthocyanin biosynthetic genes (OsDFR, OsANS, and OsUGT) were significantly higher under +N than −N in SCU212.

In contrast, N did not have a significant effect on the relative expression of OsKala3, OsKala4, OsTTG1, OsCHS, OsDFR, OsF3H, or OsUFGT in genotype SCU254, regardless of UV-B treatment (Figure 1c,d). However, the expression level of OsANS was significantly higher in the +N treatment for SCU254 under both UV-B treatments.

4. Discussion

Previous studies in rice suggested that, depending on genotype, N application can either increase [16,17] or have no effect [16,18] on grain anthocyanin concentrations. Our results support the finding that the response of grain anthocyanin concentration to N application varies according to genotype. While we observed increased grain anthocyanin concentration in genotype SCU212, selected based on its high grain anthocyanin and γ-oryzanol content, no response to N fertiliser application was found for genotype SCU254. The decline in grain γ-oryzanol concentration in genotype SCU212 following N fertiliser application is in agreement with the results of [19]; however, the lack of any significant effect of N on grain γ-oryzanol concentrations in genotype SCU254 indicates that the grain γ-oryzanol response to N is also genotype-specific.

In general, a negative correlation between applied N and anthocyanin concentration in leaf tissue and fruits of other crop species has been reported in the literature. Application of N resulted in a significant drop in anthocyanin concentration in grapes [35,42] and blueberries [43]. A similar result was reported in the medicinal plant Labisia pumila (Kacip Fatimah), where anthocyanin concentration in leaves, stems, and roots decreased with increasing N rate [44]. In leaves or any photosynthetic tissues, accumulation of anthocyanins during N stress has been proposed as an adaptive strategy by plants, protecting leaves from photodamage and facilitating the recovery of N from senescing leaves [45]. That anthocyanin concentration was not increased in black rice grains under N deprivation in our study suggests that anthocyanins in grains may not be an adaptive response to stress. In rice grain, anthocyanin production in the pericarp is most likely a recent trait, gained after rice domestication due to natural mutation rather than as an adaptive strategy of the plants, as there is no evidence of wild rice ancestors with a black pericarp [46].

It has been established that the exact composition of the MBW complex that regulates anthocyanin biosynthesis differs among various rice tissues [33]. In rice, OsC1(MYB) and OsKala4 (bHLH) regulate the anthocyanins in hulls [47], whereas OsC1 and OsRb (bHLH) are responsible for anthocyanin production in leaves [34]. In rice pericarp, OsKala3 (MYB) and OsKala4 (bHLH) regulate anthocyanin biosynthesis [33]. Effects of N on the expression of different tissue-specific regulatory genes of MYB TF and/or bHLH TF and the expression of different components (MYB-bHLH-WD) of the MBW complex itself in rice pericarp is unknown. In this study, the expression of the OsKala3 regulatory gene (MYB TF) in genotype SCU212 was significantly upregulated by N addition, whereas the expression of the other two regulatory genes, OsKala4 and OsTTG1, remained unchanged. Hence, in rice pericarp, applied N may alter the concentration of grain anthocyanin by specifically upregulating the expression OsKala3 of the MBW complex, suggesting that regulation of the MBW complex is more complex than previously known. In genotype SCU254, where N application did not significantly affect anthocyanin content, the relative expression of OsKala3 and core structural genes OsCHS and OsDFR were not significantly affected, although OsANS expression was upregulated. This shows that the differential upregulation of a single gene in a biosynthesis pathway is unlikely to result in a higher anthocyanin content. This finding contradicts earlier studies, where limiting N supply caused upregulation of both structural and regulatory genes for anthocyanin biosynthesis concomitant with an increase in anthocyanin concentration in various crops such as grape berries [35], apple peel [48], and Arabidopsis thaliana leaves [36,49]. This discrepancy in results is likely explained by the fact that rice pericarp is a nonphotosynthetic tissue, unlike leaves or the berry exocarp.

Previous studies looking at the effect of nitrogen on anthocyanin and γ-oryzanol were either conducted in glasshouses/polytunnels conditions or in normal field conditions [16,17,18]. One of the major differences between these studies is the presence of UV-B in field conditions and minimal to nil UV-B in controlled environment studies in glasshouses or polycarbonate houses. Studies have established that the exposure of leaves and fruits to UV-B radiation can increase anthocyanin content in various crops [22,23,25]. This raises the question of whether the findings from the field and controlled environment studies are heavily influenced by differences in UV-B conditions. Hence, this study looks at the N fertiliser effect on key secondary metabolites of rice grains under both “with” and “without” UV-B radiation. The purpose of the study was not to investigate the main effect of UV-B on these key metabolites, and due to the lack of replication of UV-B treatment blocks, it is not possible to investigate the main effect and the interactive effect of UV-B and N on anthocyanin and γ-oryzanol. However, it is worth noting that a 20% increase in grain anthocyanin content in the presence of UV-B was observed in genotype SCU212, whereas in genotype SCU254, grain anthocyanin was not responsive to UV-B. The increase in grain anthocyanin in SCU212 is consistent with earlier findings where UV-B radiation has been reported to increase anthocyanin concentration in leaves and fruit of different plant species [20,21,22,23], including the leaves of rice [24,25]. Grain γ-oryzanol content was not affected by UV-B radiation in either of the genotypes. γ-Oryzanol is widely used in cosmetic products for sun protection due to its UV-A and UV-B absorption capacity [50]. Although γ-oryzanol has been found in various tissues of rice plants, its probable role in UV-B protection in plants has not been well investigated. Future studies to investigate the main and interactive effect of UV-B and N on key secondary metabolites of black rice grain are therefore warranted.

5. Conclusions

This study indicates that effect of N on grain anthocyanin and γ-oryzanol concentrations is genotype-specific. Further, that N fertiliser application increased the anthocyanin concentration in genotype SCU212 by altering the expression of only OsKala3 regulatory genes without significant change in the expression of other members of MBW complex suggests that our understanding of the MBW complex in rice is incomplete. The increase in pericarp anthocyanin in response to N application, which contrasts the anthocyanin response to N in photosynthetic tissues, is consistent with the notion that anthocyanin levels in rice grains are not an adaptive response to stress but, rather, due to a random mutation that was then artificially selected for by farmers due to consumer preference.