Higher Temperatures during Grain Filling Affect Grain Chalkiness and Rice Nutrient Contents

: High temperature effects attributable to climate change can affect rice quality. The chalky area of rice grains is often used to evaluate of rice grain starch quality, but the overall effect of high temperatures on grain chalkiness and overall nutrient quality has not been fully clariﬁed. Thus, in this study, we assessed high temperature effects on grain weight, chalkiness, and nutrient contents. Rice grains were classiﬁed into four groups on the basis of the chalky area in scanned grain images: P (0%), S (0–15%), M (15–40%), and L ( ≥ 40%). Then, the amylose, protein and mineral nutrient concentrations were assessed in each chalkiness classiﬁcation. High temperatures during grain ﬁlling markedly decreased the grain weight and the amylose content of milled rice but increased the chalky area of the grains as well as protein content and the concentrations of most minerals. There were signiﬁcant negative correlations between mineral contents and both grain weights and amylose contents of milled rice. These results indicate that increases in grain chalky areas due to high temperatures during grain ﬁlling also increase grain mineral contents.


Introduction
Rice is the staple food for more than half the global population, supplying them with important calories and nutrients [1]. It is the most important crop in Asia, where more than 90% of rice is produced, but it is also an important food source in Madagascar and West Africa [2,3].
However, climate change is expected to affect rice yield and quality. Various studies have reported that high temperatures above 26-27 • C during the grain filling stage reduces yield [4][5][6] and quality due to poor accumulation of storage materials such as starch [7]. One study reported that high temperature could reduce the starch content by up to 40% [8]. Moreover, high temperatures in the period after heading cause the emergence of chalky grains, immature thin grains, and cracked grains, leading to decreased quality of rice in external appearance and taste [7,9,10].
Some studies have found that higher CO 2 conditions can lead to increased rice yields [11][12][13], but other studies have reported reductions in nutrient grain contents under high CO 2 conditions. For example, it has been reported that high CO 2 conditions lead to decreases in proteins; vitamins; and minerals such as zinc, iron, calcium, magnesium, copper, and manganese [14][15][16], and that these nutrient reductions could lead to undernutrition in people [17]. For example, Smith and Myers (2018) [18] indicated that protein, iron, and zinc contents in food crops could be reduced by 3-17% at 550 ppm CO 2 compared with current conditions. Such decreases in nutrients are likely to increase the risk of under-nutrition in the world, especially in the poorer rice-dependent countries [2,15]. Currently, there are about 2 billion people in the world that have vitamin and mineral deficiency [2]. Deficiencies in minerals such as iron and zinc are a major health problem in pregnant women, infants, and children [19]. They increase the risk of death in pregnant women as well as suppressing growth and causing immunodeficiency in children [20,21]. For example, zinc deficiency accounts for 4.4% of deaths in children under five years of age [20]. These under-nutrition problems could be increased by the combination of population growth in these countries together with the risk of reduced rice grain yields and nutrient contents due to climate change.
In addition to reductions in grain yield and nutrient contents, some reports have also reported increases in the occurrence of chalky grains under higher temperatures. Chalky grains are turbid in appearance because of light scattering caused by a coarse structure that contains gaps between starch granules and is the result of poor starch accumulation [22]. It is thought that this is due to high-temperature conditions decreasing starch synthesis as well as reducing activation of α-amylase, which is a starch-degrading enzyme [23]. Moreover, chalky grains are smaller than perfect rice grains in length, width, and grain thickness [24,25]. Several types of chalkiness have been reported depending on the position of the chalkiness in the grain, including white-back, basal-white, and milky-white [26]. In addition, the size of the chalky area has been reported to increase with temperature. In Japan, the grain chalky area starts to increase when the mean temperature during 20 days after heading is over 23-24 • C and exceeds 20% of the grain surface area when the temperature is more than 27 • C [27]. Furthermore, it has been reported that the type of grain chalkiness varies depending on the temperature conditions during the grain filling period and the rice cultivar [24]. Some studies have also assessed the relationship between grain chalkiness and the grain quality and starch structure [28,29]. However, no reports have investigated how differences in grain chalkiness resulting from higher temperatures would affect other grain nutrients such as protein and minerals. Since rice is such an important staple food, it is important to understand how climate change could affect the nutrient quality of rice, as well as the relationship between grain chalkiness and grain nutrient concentrations.
Therefore, the purpose of this study was to investigate how different air temperatures during the grain filling period affected grain chalkiness and nutrient concentrations, including amylose, protein, and mineral concentrations.

Study Site and Plant Material
A culture pot study was performed during May-October in 2020 in an open-air steel frame building, covered in wire netting to protect the plants from birds, at the School of Agriculture, Meiji University, Kawasaki, Kanagawa, Japan (35 • 61 N, 139 • 54 E). The test cultivar used was IR64, which is widely cultivated in tropical Asia [30].

Experiment Setup
Three treatment groups were established with different transplanting dates to change the air temperature during the grain filling stage. The T1 group was transplanted on 20 May, the T2 group on 3 June, and the T3 group on 17 June. In all the treatment groups, 1/5000-a Wagner pots were filled with rice paddy soil that had been treated to remove the plant residue and ensure uniform soil particle size. Three seedlings per pot were transplanted into each pot on Day 21 after seeding. The applied fertilizers were N/P/K = 3:3:3 g pot −1 as basal dressing, followed by N (Urea) = 0.286 g pot −1 as topdressing on Day 50 after transplantation. These pots were installed in a concrete frame filled with water (3 m length × 3 m width × 0.3 m depth) and left flooded throughout the cultivation period. There were 15 pots in each treatment group. Five replicate pots were analyzed for each treatment. Air temperature in the study building was recorded with a thermometer in a thermistor (TR-52i; T&D Corp., Matsumoto, Japan) in a forced ventilation duct.

Sample Preparation of Milled Rice
The rice was harvested at the maturity in all treatment groups. It was air-dried in a glass greenhouse for 5 days, followed by threshing, husking (Otake FC2K Dehusker; Ohtake Seisaku-sho, Aichi, Japan), and screening (1.6 mm aperture). The resulting brown rice from each pot was measured three times with a grain scanner (RSQI 10A; Satake Co. Ltd., Hiroshima, Japan) for grain length and grain width, and three times with a rice analyzer (RGQ1 100B; Satake Co. Ltd., Hiroshima, Japan) for grain thickness. Then, the rice from each pot was polished with a grain polisher to a polishing ratio of 90-91% (Pearlest; Kett, Tokyo, Japan).

Classification of Grain Chalkiness
About 10g of milled head rice (grains retaining 75% or more of their length before milling [31]) from each pot was classified for chalkiness on the basis of the proportion of chalky area in an image of each grain taken with grain scanner. There were four chalkiness classifications: Perfect (0% = P; no chalkiness), Small (S; 0-15%), Medium (M; 15-40%), and Large (L; over 40%). Additionally, the weights of milled rice were ascertained for each classification group and were converted into the weight of milled rice per grain. Classified milled rice from each pot was then crushed using a grain sample crusher (TQ-100; Kett Tokyo, Japan) to be analyzed for protein, amylose, and minerals contents, as explained below.

Protein Contents
The protein contents of milled rice were determined by measuring the nitrogen concentration using a CHNS/O elementary analyzer (vario MICRO cube; Elementar, Frankfurt, Germany) and multiplying the protein conversion factor of 5.95.

Amylose Content
The amylose content was obtained using the iodine colorimetry method [32]: 1 mL of 95% ethanol and 9 mL of 1N NaOH were added to 0.1 g of powdered milled rice, which was gelatinized at room temperature [33]. The gelatinized solution was diluted and neutralized with acetic acid, and then potassium iodide solution was added. Finally, the mixture was incubated in a thermostat bath at 27 • C for 20 min to develop color, and absorbance was measured at 620 nm (according to Juliano's method). Since the amylose content partly reflects the accumulation of carbohydrate in grains [34], in this study, the amylose content was used to estimate the starch content [16].

Mineral Nutrient Concentrations
About 0.1 g of classified milled rice was put in a 25 mL test tube. It was degraded by adding about 3 mL of dilute nitric acid (HNO 3 ), after which it was kept at room temperature for 3 days. The digested solution was then transferred to a 10 mL volumetric flask and filled up to volume with dilute nitric acid. The resulting solution was then filtered with a membrane filter (DISMIC filter 25SS045RS, polyether sulphone, 0.45 µm pore size; Advantec, Tokyo, Japan) to prepare for measurement. Macronutrients (K, Mg Na, Ca) and micronutrients (Zn, Cu, Mn, Fe, Ni, Mo) were analyzed using inductively coupled plasma optical emission spectrometry (ICP-OES, Optima 8300; PerkinElmer Co., Ltd., Massachusetts, USA) or inductively coupled plasma mass spectrometry (ICP-MS, NexlON 300D; PerkinElmer Co., Ltd., Massachusetts, USA), respectively.

Dates of Growth Stages and Environmental Conditions during Growth of Each Treatmet Group
The number of days before heading differed between the treatment groups, with a difference of about 1 week between the shortest T2 group and the longest T1 group ( Table 1). The growth period from heading to maturing was in a range of 41-45 days. The mean air temperature during the whole growing season of the three groups differed by about 1 degree Celsius. However, during the 20 days after heading (20 DAH), there was a larger difference in mean air temperature, ranging from 29.1 • C for the T1 group to 28.2 • C and 23.5 • C for the T2 and the T3 groups, respectively. Therefore, grain filling in the T1 and T2 treatments was under high temperature conditions, while that of the T3 treatment was under more typical non-high temperature conditions.

Grain Size and Changes in Grain Weight and Amylose and Protein Contents
The brown rice of the T3 group had significantly greater grain length than in the other two groups (Table 2). However, the grain width and grain thickness were similar or were significantly higher in the T1 and T2 groups than in the T3 group. After classifying the grain chalkiness of the milled rice, we found that the grain weight of the P classified grains without any chalkiness was highest in all the treatment groups ( Table 3). As the chalkiness increased, the grain weight in the T1 and T2 groups decreased, although there was no such reduction in grain weight in the T3 group.  There was no significant difference in amylose content between chalky ratio classifications, although it was slightly higher in P (Table 3). In contrast, the protein content increased along with an increase in the chalky ratio, being significantly lower in P or S classifications in all treatment groups compared with the higher chalky ratio M and L classifications. Then, the relationships between the grain weight of milled rice, amylose, and protein contents were investigated according to the chalky ratio in all the treatment groups ( Figure 1). There was a positive correlation between grain weight and amylose (p < 0.001), whereas there were negative correlations between the grain weight of milled rice and protein (p < 0.05), and between protein and amylose (p < 0.01).

Mineral Concentrations by the Chalky Ratio in Treatment Groups
Mineral contents varied between chalkiness ratio classifications and growing season treatment groups ( Table 4). The relationships also differed between mineral type. In the T1 and T2 groups, many of the minerals had lower concentrations in more chalky grains (M or L classification) than in the less chalky grains (P or S classification). This was true for K, Mg, Ca, Zn, Mn, Fe, and Mo in the T1 group and for K, Mg, Zn, Fe, and Mo in the T2 group. However, in the T3 group, there was an opposite relationship between the concentrations of many minerals and chalky ratio, with higher concentrations of K, Mg, Zn, Cu, Fe, and Ni in the least chalky category P. In general Cu, Ni, and Mo contents were higher in grains with lower chalky ratios in all treatment groups. Mo concentrations tended to be higher in L and M chalkiness categories than in S or P, with differences between L or M and P ranging from 4 to 39%. Moreover, Ni was lower in L than in P by 17-42%.

Mutual Correlation between Grain Weights of Milled Rice, Amylose, Protein, and Mineral Contents
Mutual correlation between grain weights of milled rice, amylose, protein, and mineral contents was investigated using the average of all chalky ratios for each treatment group (Table 5). Significant and negative correlations were found between grain weight and K, Ca, Zn, Fe, and Mo contents (r = −0.607 to −0.778), and between amylose and K, Ca, Ni, and Mo contents (r = −0.621 to −0.745). However, no significant correlation was found between protein and any of the minerals. Furthermore, no significant correlation was found between Mg, Na, Cu, Mn, or Ni contents and either grain weight or amylose. There were significant correlations between mineral contents, with higher Fe contents associated with higher K, Ca, Cu, and Mo concentration (p < 0.001). *, **, and *** respectively denote significant correlation at p < 0.05, p < 0.01, and p < 0.001.

Discussion
In this study, we investigated how changes in temperature during the grain filling period of rice affected grain amylose, protein, and mineral concentrations, and how these changes were related to the chalkiness of the harvested grains. Changes in the air temperature during the grain filling period were achieved by altering the seedling transplanting date. By this method, the mean air temperature at 20 days after heading was more than 28 • C in the T1 and T2 groups, which is a high temperature environment for grain filling, compared with a more optimal temperature of 23.5 • C at the same growth stage for the T3 group (Table 1). This difference in air temperature during grain filling affected the brown rice size, with shorter grains in the T1 and T2 groups than in the T3 group (Table 2). Morita [27] reported that the proportion of chalky grains begins to increase when the mean temperature during 20 days after heading is over 23-24 • C and exceeds 20% in many regions in Japan, from Tohoku in the north to Kyushu in the south, when the temperature is higher than 26-27 • C [27]. Chalky grains are smaller than perfect rice in grain length, grain width, and grain thickness [24,25]. In our study, the chalky ratio was higher, and the brown rice size was smaller in the T1 and T2 groups that were grown under higher temperature grain filling conditions. Chalky rice has grains that are turbid in appearance because of light scattering caused by a coarse structure containing gaps between starch granules that results from poor starch accumulation [22]. A decrease in the capacity of starch synthesis, as well as the activation of α-amylase, which is a starch-degrading enzyme under high-temperature conditions, are reportedly factors that cause chalky rice [23]. In addition, chalky rice is classified into several types on the basis of the position of chalkiness in the grain. Major types of chalky rice induced by high temperature include white-back, basal-white, and milky-white [26]. Moreover, earlier reports have described differences in the frequencies of types of chalky rice depending on both the temperature conditions during the grain filling stage and cultivar [24]. These difference in chalky areas have been used to evaluate the quality and starch structure of chalky rice [28,29].
In our study, we classified grain chalky ratio into four levels and analyzed difference in grains and their amylose, protein, and mineral contents. The grain weight of milled rice and the amylose content tended to decrease as the chalky ratio increased, whereas the protein contents tended to increase with increased the chalky ratio (Table 3). These results demonstrated that changes in chalkiness caused by increased temperatures during grain filling also reflect change in the grain weight of milled rice, amylose content, and protein contents. Kondo et al. [35] pointed out that starch accumulation and synthesis are involved in the mechanisms that cause chalky rice and are related to decrease in the grain weight of brown rice. Seo and Chamura [36] and Asaoka et al. [37] also pointed out that the protein content increases and the amylose content decreases under high-temperature conditions. In our study, there was a close positive relationship between the grain weight of milled rice and amylose content, but a negative one between grain weight of milled rice and protein content (Figure 1). These results suggest that differences in the chalky area of rice, which result from high temperatures during the grain filling stage, occur because of the poor accumulation of starch (low amylose content), decreased grain weight, and increased protein content.
High temperatures increased the concentrations of most of minerals (Table 4), which is in agreement with the results of an earlier study on wheat and rice [38]. Some reports have suggested that the reason for increases in minerals is that transpiration from rice leaves increases at high temperatures, promoting the translocation of elements from underground plant tissues to above the ground tissues such as the grains [39,40]. Moreover, high temperatures can promote soil acidification, which might enhance mineral utilization [40]. In our study, the T1 and T2 treatments grown under higher than optimal temperatures for grain filling (T1 and T2 = 29.1 • C and 28.2 • C mean air temperature at 20 DAH, compared with only 23.5 • C for T3) had higher chalky ratios and generally higher mineral contents, suggesting that higher temperatures could increase both grain chalkiness and mineral contents. Lin et al. [41] reported that there was a relationship between formation of grain chalky areas and the accumulation of starch, protein, and minerals. Moreover, in our study, we demonstrated that grain weight and the amylose contents are related to changes in minerals (Table 5). Lin et al. [41] reported that SEM observation indicated there were readily apparent inter-granule spaces in the chalky parts of grains, indicating lower starch concentrations in these parts than in translucent parts. These findings suggest that the grain weight and the amylose content are lower when the chalky area of chalky rice increases under high temperatures during grain filling, resulting in wider gaps between starch granules, providing more space for the accumulation of minerals. However, there were no such close relationships found for several minerals. Therefore, it is necessary to clarify differences in the structure of starch granules in the chalky areas of rice formed both under high-temperature and optimal temperature conditions in order to investigate reasons for the difference results for some minerals. Moreover, further investigation is needed to clarify the relationships between the chalky ratio and the gap ratio between starch granules.

Conclusions
This study demonstrated that high temperatures during the grain filling stage can increase the size of the chalky areas in rice and increase the minerals contained in milled rice. It is therefore possible that higher temperature due to climate change might have some positive effects in term of improving the mineral nutrition of under-nourished populations who depend on rice in their diet. However, higher temperatures also increase losses during milling because of broken rice, and this leads to a decrease in eating quality and adverse effects on yield. Many aspects of the mechanisms of mineral absorption and accumulation remain unknown, not only in rice but also in other cereals. Therefore, further studies must be conducted under various environmental conditions that are predicted to arise in the future.
Author Contributions: R.S. and F.S. designed the research. R.S., M.A. and F.S. conducted the research and analyses, and wrote the paper. All authors have read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.