*Article* **Climate Change, Crop Yields, and Grain Quality of C<sup>3</sup> Cereals: A Meta-Analysis of [CO2], Temperature, and Drought Effects**

**Sinda Ben Mariem <sup>1</sup> , David Soba <sup>1</sup> , Bangwei Zhou <sup>2</sup> , Irakli Loladze <sup>3</sup> , Fermín Morales <sup>1</sup> and Iker Aranjuelo 1,\***


**Abstract:** Cereal yield and grain quality may be impaired by environmental factors associated with climate change. Major factors, including elevated CO<sup>2</sup> concentration ([CO<sup>2</sup> ]), elevated temperature, and drought stress, have been identified as affecting C<sup>3</sup> crop production and quality. A meta-analysis of existing literature was performed to study the impact of these three environmental factors on the yield and nutritional traits of C<sup>3</sup> cereals. Elevated [CO<sup>2</sup> ] stimulates grain production (through larger grain numbers) and starch accumulation but negatively affects nutritional traits such as protein and mineral content. In contrast to [CO<sup>2</sup> ], increased temperature and drought cause significant grain yield loss, with stronger effects observed from the latter. Elevated temperature decreases grain yield by decreasing the thousand grain weight (TGW). Nutritional quality is also negatively influenced by the changing climate, which will impact human health. Similar to drought, heat stress decreases starch content but increases grain protein and mineral concentrations. Despite the positive effect of elevated [CO<sup>2</sup> ], increases to grain yield seem to be counterbalanced by heat and drought stress. Regarding grain nutritional value and within the three environmental factors, the increase in [CO<sup>2</sup> ] is possibly the more detrimental to face because it will affect cereal quality independently of the region.

**Keywords:** cereals; yield and quality; high [CO<sup>2</sup> ]; predicted future climate; high temperature; grain quality traits; drought stress

#### **1. Introduction**

Food security is threatened by the impacts of climate change on agriculture and by increasing the world population [1,2]. Actually, climate change has already slowed global agricultural productivity growth, and in a recent study, Ortiz-Bobea et al. [3] found that anthropogenic climate change (ACC) has reduced global agricultural total factor productivity since 1961 by about 21%, with a greater impact for warm regions such as Africa (−34%) than for cooler regions such as Europe and Central Asia (−7.1%). Over the next few decades, climate change is expected to affect more the world's supply of cereal grains, impacting their quantity and quality due to the complex effects of elevated atmospheric [CO2] and changing temperature and rainfall patterns on crops [4]. Cereals contribute to a substantial part of the world's plant-derived food production and comprise a majority of the crops harvested. In fact, FAO statistics show that in 2016, sugar cane had the highest production globally, followed by corn, wheat, and rice [5]. Adding to that, according to the Foreign Agricultural Service/USDA, preliminary world production in 2018 of maize, wheat, and rice was estimated at around 1076, 763, and 495 million tons, respectively [6]. Further, their nutritional quality has a significant impact on human well-being and health, especially in the developing world [7]. Thus, one of the major

**Citation:** Ben Mariem, S.; Soba, D.; Zhou, B.; Loladze, I.; Morales, F.; Aranjuelo, I. Climate Change, Crop Yields, and Grain Quality of C<sup>3</sup> Cereals: A Meta-Analysis of [CO2], Temperature, and Drought Effects. *Plants* **2021**, *10*, 1052. https:// doi.org/10.3390/plants10061052

Academic Editor: Othmane Merah

Received: 31 March 2021 Accepted: 21 May 2021 Published: 24 May 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

challenges that plant breeders are facing currently is to increase cereal grain production while taking into consideration an adequate grain nutrient content.

Numerous effects of elevated atmospheric [CO2] on plants have been documented through a photosynthesis-mediated CO2fertilization effect, including increased carbon (C) assimilation, growth, yield, and C content [8,9]. Thus, elevated [CO2] could enhance the concentration of photosynthesis-derived carbohydrates in grains, starch being the major component [10–12]. Since grains are predominantly composed of carbohydrates (mostly in the form of starch), it has been suggested that increases in starch concentrations can cause a dilution effect onother nutrients, including proteins, lipids, vitamins, and minerals. In addition, adjustments in the photosynthetic apparatus and later on the redistribution from senescing leaves to grains must be considered as the key mechanisms. Due to the different biochemistry of C<sup>3</sup> and C<sup>4</sup> photosynthesis, the positive effect of elevated [CO2] on photosynthesis is more pronounced in C<sup>3</sup> crops such as wheat and rice but less notable in C<sup>4</sup> crops such as maize [13]. Rising [CO2] is likely to lead to "globally imbalanced plant stoichiometry (relative to pre-industrial times)" [14], which in turn would "intensify the already acute problem of micronutrient malnutrition" [14], particularly regarding minerals such asFe, Zn, and I, as well asprotein (or N) [12,15,16]. Elevated [CO2] has been reported to decrease mineral concentrations in barley grains (−6.9%), rice grains (−7.2%), and wheat grains (−7.6%), and increasing the ratio of non-structural carbohydrates (TNC) to protein by 6–47% in grains and tubers [9]. For the grain crops barley, rice, and wheat, the reduction in protein mediated by elevated [CO2] was reported to be 15, 10, and 10%, respectively [15]. In their meta-analysis of the impact of elevated [CO2] on wheat grains, Broberg et al. [12] found a significant reduction in the concentration of the majority of minerals (Ca, Cd, Cu, Fe, Mg, Mn, P, S, and Zn), while B and Na were not significantly affected, and K was significantly increased (<2%). These meta-analytic results are in line with those from individual wheat FACE experiments [17–21]. Two minerals, Fe and Zn, are already deficient in the diets of hundreds of millions of people, and CO2-induced reductions in Fe and Zn have been reported in the edible parts of major crops [9,22,23] and are projected to have negative effects on human nutrition [24,25]. Furthermore, emerging evidence points to elevated [CO2] affecting nutrients beyond protein and minerals that are essential to humans, such as vitamins and carotenoids [26,27]. The decrease in mineral concentrations is notable in C<sup>3</sup> plants but less so in C<sup>4</sup> plants [9,23] and is consistent with differences in physiology; the simulation of carbohydrate production by elevated [CO2] is stronger in C<sup>3</sup> plants, while reduced transpiration is present in both C<sup>3</sup> and C<sup>4</sup> plants.

During the last two decades, the air temperature has increased by 0.85 ◦C [28]. In fact, annual average minimum temperatures in Spain have increased over the last century by 1.5 ◦C, and by 0.6 ◦C during the last 25 years [29]. The most probable outcome of climate ensemble model projections foresees increasesof1.8 to 4.0 ◦C by the end of the 21st century (2090–2099) relative to the period 1980–1999. These numbers originate from the best estimate of greenhouse gas time series deduced from the six marker scenarios alone [30]. Heat stress is a major constraint to sustainable cereal production, with reductions in grain yield being associated with high temperatures during the reproductive or grain-filling stages in wheat [31,32] and rice [33–35]. High-temperature impacts on grain filling can vary enormously, depending on timing (days after anthesis) and duration. Both chronic moderately high temperatures (25–35 ◦C) and heat shocks (>35 ◦C) during the grainfilling phase are frequently associated with an increase in grain protein concentration in wheat [31,36,37] and rice [35,38]. Indeed, high temperature primarily impacts the accumulation of starch in wheat grain, with accumulation beginning earlier than under cooler temperatures, the duration of its accumulation also being reduced, and the result is a greater concentration of protein in the grain. Further, the duration of protein accumulation is reduced, while the rate of protein accumulation is substantially increased. In addition, leaves senesce before the heads mature, suggesting that high temperatures might enhance N remobilization from leaves and stems [39,40]. Moreover, the timing and duration of heat stress during grain filling have been shown to be important sources of variation

in dough properties in wheat [41]. Grain protein and mineral composition are quality characteristics that can change due to high temperature, and they respond to changes in enzymes involved in starch and protein synthesis. Yang et al. [42] observed that the activity of glutamate synthase was enhanced by heat stress, while sucrose phosphate synthase, sucrose synthase, and soluble starch synthase were significantly decreased during grain filling. However, Monjardino et al. [43] found that protein concentration was negatively affected by heat stress during the early stage of endosperm development. They found that among the protein fractions, zeins are the most affected by heat stress. In fact, zein accumulation was repressed under high temperature rather than being degradedin the early developmental stages. In rice, elevated temperature also alters grain protein and mineral nutrient composition [35,44]. Ferreira et al. [36] showed that the total quantity of N per grain in wheat is generally little affected by the growing temperature but, due to the above-mentioned lower grain yield, the percentage of N on a dry weight basis rises under higher temperatures. Similar increases in the percentage of dry weight have been reported for wheat [45,46].

Increasing greenhouse gas emissions may also lead to rainfall reductions in the coming decades, which will increase the frequency and intensity of drought in the Mediterranean basin [47–49]. Climate change projections for the Mediterranean region indicate a precipitation decrease of 25–30% for the last decades of the 21st century [50]. Adding to that, the seasonality of rainfall is much more important. In fact, the expected shortage in Mediterranean rainfall should impact summer precipitation much more than winter precipitation. Mediterranean crop growth, however, is mainly driven by winter rain. Moreover, drought is considered one of the most important factors limiting crop yields around the world. Wheat crop responses to water scarcity depend on several factors, including plant development status, duration, and intensity of the stress and genetic variables [51]. Although rainfall during winter has been traditionally abundant and coincides with the lowest evapotranspiration rates, the occurrence of drought in winter during the early stages of the crop cycle has been recently reported [52].

This can further constrain wheat growth and thus final grain yield, mostly through a decrease in ear density and the number of kernels per unit crop area [53,54]. Grain yield reductions mediated by drought have been widely reported in wheat [55,56], and depending on the genotype, the reductions may reach up to 50%. The TGW is also reduced significantly, above 30% in droughted wheat [51,56,57]. Drought stress leads to reduced photosynthetic area and acceleration of leaf senescence during late grain filling in cereals, resulting in a shorter grain-filling period. In wheat, this smaller photosynthetic area and accelerated leaf senescence limit the amount of assimilates translocated to the grain, which implies reductions in grain yield [51]. Grain composition is also affected. Drought stress affects starch accumulation [51,58] more severely than N accumulation during grain filling, putatively influencing the conversion of sucrose into starch [59]. This tends to increase the grain protein concentration (expressed as % protein) in wheat [35,55,57] and rice [60]. In some cases, the opposite effect has been observed in wheat [61,62], possibly related to differences in stress levels and plant development status [63]. Knowledge of the effects of drought stress on grain mineral composition is scarce [7,64]. Crusciol et al. [60] explained the increase in rice grain N, Ca, Mg, Fe, and Zn concentration under rainfed conditions being due to a dilution effect because productivity was higher in irrigated than rainfed systems.

All the above-mentioned changes in grain composition linked to the changing environmental conditions are expected to have important implications for the nutritional quality of foods. During the last decade, different meta-analyses have characterized elevated [CO2] effects on crop yield and quality traits. However, comparatively little attention has been given to how other target environmental parameters such as temperature and drought will affect crop yield, and especially grain nutritional characteristics. Considering the economic and social importance of cereal crops and the impact of climate change not only on grain production but also on the nutritional value, this meta-analysis aims to provide an overview

of the effects and interactions of multiple climate stressors, specifically high [CO2], drought and elevated temperatures, on the productivity and grain quality of C<sup>3</sup> cereals.

#### **2. Results**

#### *2.1. [CO2], Temperature and Drought Stress Effects on Grain Yield Components*

The overall effect of elevated [CO2] on C<sup>3</sup> crops resulted insignificant increases in grain yield and thousand grain weight (TGW) of30.10% and 7.41%, respectively (Figure 1). Nevertheless, a contrasting drastic loss in grain yield and TGW was observed under high temperatures and drought stress. Results presented in Figure 1 indicate that the heat and drought stress effects were similarfor TGW and recorded −20.17% and −20.29% reductions, respectively, but the negative effect of drought on cereal grain yield was larger than the effect of elevated temperatures (−70.53% vs. −24.85%). − − − −

**Figure 1.** Change (%) in the mean of yield and TGW of plants grown under elevated [CO<sup>2</sup> ], high temperature, and drought stress relative to the control. Data within parentheses indicate the number of observations. Error bars indicate the standard error of the mean. \* indicates a statistically significant difference at *p* < 0.05.

#### *2.2. [CO2], Temperature and Drought Stress Effects on Grain Quality* 2.2.1. Starch

− In cereals grown under high [CO2], there was a significant increase in grain starch concentration (5.65%), whereas there was a significant decrease (−9.91%) under elevated temperature (Figure 2). Regarding water availability, there was no significant change as responses were sprayed in a broad range of both positive and negative changes (Figure 2).

**Figure 2.** Change (%) in the mean concentration of grain starch of plants grown under elevated [CO<sup>2</sup> ], high temperature, and drought stress relative to the control. Data within parentheses indicate the number of observations. Error bars indicate 95% CI. \* indicates a statistically significant difference at *p* < 0.05. −

#### 2.2.2. Total Protein

− − − Grain total protein concentration was negatively affected (−8.90%) by high [CO2]. However, it was significantly increased by temperature and drought (10.40% and 12.44%, respectively, as shown in Figure 3. Among the proteins that were studied, the gluten, gliadin, and glutenin concentrations were analyzed under elevated [CO2]. The grain gluten, gliadin, and glutenin concentrations presented in Figure 4 reveal a significant decrease in the gluten and gliadin concentrations (−11.54% and −7.41%, respectively). In contrast, rising [CO2] decreased the glutenin concentration, but it was not significant. Regarding the effects of drought and heat stress on these proteins, it was not possible to generate statistically powerful results due to the low amount of data (less than three repetitions). − −

**Figure 3.** Change (%) in the mean concentration of grain total protein of plants grown under elevated [CO<sup>2</sup> ], high temperature, and drought stress relative to the control. Data within parentheses indicate the number of observations. Error bars indicate 95% CI. \* indicates a statistically significant difference at *p* < 0.05.

**Figure 4.** Change (%) in the mean concentration of grain gliadins, gluten, and glutenins of plants grown under elevated [CO<sup>2</sup> ] relative to ambient level. Data within parentheses indicate the number of observations. Error bars indicate 95% CI. \* indicates a statistically significant difference at *p* < 0.05.

#### 2.2.3. Mineral Composition

− − The results presented in Figure 5 show an overall decrease in micro-macronutrients in C<sup>3</sup> grains under elevated [CO2]. Across all the data, the mean change ranged between −4.70% (recorded for P) and −39.41% (recorded for Mo). The changes in B and Se were not significant. Among all the measured elements, only Na concentration increased significantly (52.05%) under high [CO2]. Heat stress had no significant effect on any of the grain mineral concentrations, and this could be due to data scarcity and small sample sizes leading to high data variability. Slight increases in Mg and N of1.91% and 6.31%, respectively, were recorded, whereas the Ca, Fe, Mn, and Zn concentrations were reduced. Regarding water scarcity, the data analysis showed distinct effects between minerals (Figure 5). In fact, drought stress induced an accumulation of Ca and N in grains and recorded a significant increase by 19.92% and 9.56%, respectively. However, no significant increase was obtained regarding Fe, Mg, P, and Zn concentrations. Under low water availability, S and K concentrations declined, but not significantly, by −10.43% and −7.59%, respectively. − − − −

**Figure 5.** Change (%) in the mean concentration of grain minerals of plants grown under (**A**) elevated [CO<sup>2</sup> ]; (**B**) high temperature; (**C**) drought stress relative to the control. Data within parenthesis indicate the number of observations. Error bars indicate 95% CI. \* indicates statistically significant difference at *p* < 0.05.

#### **3. Discussion**

#### *3.1. [CO2], Temperature and Drought Stress Effects on Grain Yield Components*

Current scientific knowledge indicates that grain yield and quality will face serious challenges under the projected future climate. In line with previous papers [65–67], our meta-analysis shows that the predicted elevated [CO2] will increase crop grain production [8,17,68]. However, as noted by studies conducted over recent decades, it is essential to consider that the [CO2]-derived "fertilization" effect might decline or be eliminated when combined with stressful growth conditions, such as drought and temperature stress [69–71]. Moreover, in cereals such as wheat, increased grain yields have been associated with increases in the numbers of tillers and grains per spike rather than spike number or grain size [72,73]. The results of the current study have also revealed an association with an increase in the number of grains rather than their weight (larger increase in grain yield than TGW).

Both high temperature and drought negatively affected crop yield. Data analysis showed that yield was more markedly affected under drought than under heat stress conditions. Lower yields in stressed plants can be associated with (i) a shortened duration of the grain-filling period and/or (ii) a lowered photosynthetic rate during grain-filling. Dixit et al. [67] applied a crop simulation model to assess the impact of climate change on wheat production and found a loss of 15% in wheat grain yield in stressed plants, which was associated with a reduction in the number of days to reach grain maturity. Indeed, Mitchell et al. [74] attributed the direct negative effect of rising temperature on wheat yield to the temperature-dependent shortening of the phenological stages. Such decreases in the duration of grain filling would imply a shorter time available for accumulating resources for grain formation [31,46]. The time and duration of heat stress could cause different physiological responses in the plant, therefore, affect crop production. Many studies have reported that heat stress applied prior to anthesis negatively affects the grain yield of wheat due to many reasons [31,75]. High temperature accelerates leaf senescence and reduces post heading duration [75]. Adding to that, heat stress significantly reduces seed germination and negatively affects microspores and pollen cells, leading to non-functional florets or abortion of fertile florets and resulting in male sterility [76]. In fact, the decline in grain yields under high day temperatures was primarily caused by a reduction in the seed set percentage. Meanwhile, under high night temperature, the combination of decreased spikelet number per panicle, grain weight, and biomass production in addition to decreased seed set percentage contributed to the grain yield loss [77]. Altenbach et al. [51] reported that high temperature during anthesis promoted both grain shrinkage and a decrease in weight. Additionally, under heat stress conditions, plants tend to have a shorter grainfilling period, which reduces grain size and thousand kernel weight, while under drought conditions, plants tend to produce fewer grains per spikelet (and/or fewer tillers) [78]. This finding matches the TGW analyses stated above. In fact, heat stress and water scarcity showed similar effects on TGW in the current data analysis, suggesting that under drought conditions, the drastic decline inC3cereal yields is instead linked to a decrease in grain number produced per plant.

#### *3.2. Effects of [CO2], Temperature and Drought Stress on Grain Quality*

Another major consideration is the effect of climate change on grain quality. While crop breeding is already much more focused on yield traits, comparatively little attention has been given to grain quality traits. This is a matter of great concern because, as described in more detail below, environmental stress will affect the relative abundance of starch, protein, and minerals [9,79,80].

#### 3.2.1. Starch

Starch is the most abundant end-product of cereal growth and development, representing around 70% of the dry weight (*w*/*w*) of grains [81]. Rising [CO2] increases photosynthetic rates in C<sup>3</sup> plants; increased carbohydrate translocation from the source (leaves and

stems) to the sink (grains) is expected to increase the starch content in grains [82]. Indeed, the current data analysis has shown that growth under elevated [CO2] has a significant positive effect on the grain starch concentration, which contrasts with the non-significant results reported by Högy and Fangmeier [83] and Broberg et al. [12]. Fangmeier et al. [84] reported that elevated [CO2] significantly increased starch only for plants under high levels of N fertilizer.

Despite no significant effect due to drought, we revealed an overall decrease in grain starch concentration under drought stress. Worch et al. [85] observed that changes in endosperm starch content positively correlated with grain yield and concluded that grain starch content is one of the leading causes of reduced yield in crops subjected to drought conditions. This can be due to water deficit compromising both production of photoassimilates (source of carbon skeletons for the synthesis of starch) and the activity of enzymes involved in starch biosynthesis in the endosperm. Thus, the lower starch content observed in grains of genotypes subjected to water deficit could be correlated with the availability of reducing sugars [86].

Elevated temperature also negatively affected the starch concentration in grains. It has been reported that the reduction in starch concentration under high-temperature conditions is due to two factors; (i) shortening of the grain-filling period, which may reduce the duration of starch accumulation [51], and (ii) impairment of starch metabolism. While data for grains of plants exposed to high temperatures are scarce, Hawker and Jenner [87] and Keeling et al. [88] reported the inhibition of starch metabolism by high temperature (generally around 30 ◦C), possibly due to thermal denaturation negatively affecting the activity of starch synthase.

#### 3.2.2. Total Protein

Elevated [CO2] has been documented to reduce grain protein (or N) content in edible parts of crops [14–16]. In line with these earlier studies, the current meta-analysis showed that elevated [CO2] significantly decreased grain protein concentrations. This reduction has been associated with increased photosynthesis and accumulation of grain carbohydrates, leading to reductions in the amount of grain protein (due to a dilution effect) [17,35]. However, Goufo et al. [89] reported decreases in protein without associated increases in starch in grains of rice exposed to elevated [CO2]. Decreased protein concentrations in cereal grains under elevated [CO2] might be a consequence of reduced leaf protein concentrations in photosynthetic tissues, leading to decreased seed protein [84,90]. The suppression of nitrate assimilation by elevated [CO2] could be another contributor [91]. Our study also showed that there was a change in protein composition in grains of plants grown at elevated [CO2]. In line with the results of Wieser et al. [92] and Högy et al. [17], gluten, gliadins, and glutenin concentrations decreased under increasing [CO2]. Differences in the amounts and proportions of gluten protein fractions and types have significant effects on dough mixing and rheological characteristics. One of the most important characteristics for baking quality is bread volume, which has been strongly correlated with crude protein, total gluten proteins, and glutenin macropolymers [4,93]. Consequently, a reduction in bread quality can be expected due to the higher sensitivity of gluten fractions to elevated [CO2].

Grain protein content is sensitive to environmental conditions and controlled by a number of factors, particularly the duration and rate of grain filling and the availability of assimilates, which are negatively affected in crops subjected to stressful growth conditions [94,95]. In contrast to elevated [CO2], we found that high temperatures increased the grain protein concentration by 10.4%, which could be attributed to greater remobilization of shoot-derived protein. The grain protein concentration is expressed as a percentage of grain dry mass, which alongside the lower size and weight of the affected grains (also detected in our meta-analysis), would contribute to them having lower carbohydrate levels and consequently higher grain protein [96]. We note that the increase in grain protein concentration (10.4%) is almost the same as the decrease in grain starch concentration (−9.9%), suggesting that starch depletion increases the relative content of total protein.

Drought affects plant phenology and physiology. Water scarcity has been previously described as reducing photosynthetic rates, shortening the grain-filling period [11,97], and accelerating leaf senescence after anthesis. We detected significant increases in grain total protein associated with low water availability. Bhullar and Jenner [98] reported that during the grain-filling period, drought stress hinders the conversion of sucrose into starch but has a milder effect on protein biosynthesis. Our findings did not corroborate Bhullar and co-workers' conclusions. As mentioned before, the fact that the grain starch concentration was not significantly affected by drought would discard the lower carbohydrate level as a factor that induces increased grain protein content. Singh et al. [7] observed that together with lower rates of carbohydrate accumulation in the grain of plants subjected to drought, the increase in flour protein was mainly due to higher rates of grain N accumulation. The present meta-analysis supports this assertion because grain N concentration was affected by drought. Adding to that, the increased grain protein concentration under drought could be explained by the shortened maturation time common to stress conditions, which tends to favor protein over starch accumulation in cereal grains [99]. Drought, among other stresses, accelerates the translocation of senescence-inducing resources (including amino acids) from leaves to seeds during grain filling. Several studies have demonstrated that the contribution of reserve mobilization to the final grain yield is higher under stressful conditions than relatively well-irrigated conditions [100–102].

#### 3.2.3. Mineral Composition

The present study showed that elevated [CO2] leads to an impoverishment of macro/ microelements in grains. Moreover, there is a variation among minerals in the magnitude of the reductions, and this supports previous results [17,18,20,26,103]. In fact, only the Na concentration was significantly increased, with surprisingly few studies having investigated this element in relation to the effect of [CO2], and so there is little background information to explain this trend. Basically, most studies have focused on the main minerals that affect human health, such as Fe, Zn, P, K, and Ca, and have underlined a common decline in these minerals under rising [CO2]. With respect to our results, the concentrations of Zn, Fe, S, Ca, Mg, P, Mn, K, and Mo were significantly decreased. Such reductions have been associated with increased production of spikes and grains that translates into a grain nutrient-dilution effect, diminishing the nutritional value. Furthermore, by reducing transpiration (linked to stomatal closure due to long-term exposure to elevated [CO2]), high [CO2] can reduce the mass flow in the soil toward roots, which diminishes the availability of mobile minerals in the rhizosphere [14]. While carbohydrate dilution should lower all other nutrients in plant tissues evenly [104], other effects of elevated [CO2] on plant physiology are not evenly distributed among the minerals. For example, reduction in transpiration and elevated biosynthesis affect some minerals more than others. This means that the stoichiometry of plants exposed to elevated [CO2] should "differ not only in C:(other elements) ratios but also in the ratios among other elements (e.g., C:N, N:P, and P:Zn should be different)" [14]. Indeed, Loladze's meta-analysis of over 7500 pairs of observations from studies of elevated [CO2] published over 30 years (1984–2014) showed a significant reduction in foliar Mg (and N, P, K, Ca, S, Fe, Zn, and Cu) but not the Mn content in C<sup>3</sup> plants, and underlying biochemical mechanisms responsible for the increased Mn:Mg ratio have been proposed [105].

Changes in the elemental composition in grains are also detected under heat and drought stress. Previous studies suggested that both stress factors tend to increase mineral concentrations (including Fe, N, S, Zn, K, and P). However, the low number of reports means that there is relatively large uncertainty about the magnitude of the increase. The observed increase in grain protein and N concentrations (and the concomitant decrease in starch) under elevated temperature means that there is more N per unit of starch [106]. In addition, Fe and Zn tend to increase under drought. Although water plays a significant role in mineral uptake and later mobilization within the plant, with these processes decreasing during water stress, our meta-analysis agrees with Ge et al. [107] reporting that soil drought stress improved transport mechanisms and/or routes for some minerals, such as Fe and Zn, leading to increased grain concentrations of these elements. Moreover, according to other studies [107,108], the increase in the levels of Fe and Zn may be related to the more efficient remobilization of these nutrients from leaves to grains. However, according to other authors [109], the increase in Fe and Zn concentrations is linked to sink strength at the single grain level. More specifically, Miller et al. [109] observed in maize how the mineral content in drought-sensitive genotypes (which produced lower numbers of grains than the tolerant ones) was higher than in fully watered plants. According to this explanation, the increase in nutrients in the grains may be related to the number of grains formed, with each grain being a specific sink [86]. Furthermore, as we mentioned above, heat and drought cause a decrease in the number and size of cereal grains, which suggests that there might be a concentration effect due to the smaller grains [110].

#### **4. Materials and Methods**

#### *4.1. Data Search and Selection Criteria*

To find relevant studies related to the issue of the current meta-analysis, literature searches of primary research in published peer-reviewed journal sources were conducted from Google, Web of Science, and Scopus in June 2017. To search the literature, the following keywords were used: grain yield, cereal, high [CO2], elevated temperature, drought stress, climate change, and C<sup>3</sup> grain quality. More than 150 papers were found, but 78 articles were selected according to the following criteria: (i) the article studies the effect of at least one climate parameter, including [CO2], temperature, and drought, (ii) the article contains at least one response variable from the following list: grain yield, thousand grain weight (TGW), starch, total protein, gluten, glutenins, gliadins, and a set of minerals (Al, N, B, Ca, Cd, Co, Cr, Cu, Fe, K, Mo, Mg, Mn, Na, Ni, P, Pb, S, Se, Si, and Zn). The most abundant C<sup>3</sup> species that are reported in the literature are wheat, rice, and barley. All papers included in this meta-analysis were published between 1990 and 2019 (Table A1). The study is based on comparing plants grown at elevated [CO2] (550–900 ppm) using Open Top Chamber (OTC) facilities or in the field using Free-Air-CO2-Enrichment (FACE) systems with those grown at ambient [CO2] (currently at ca. 400 ppm). Studies comparing different ranges of temperature, from ambient (10–25 ◦C) to elevated temperature (28–37 ◦C), and two levels of irrigation (limited irrigation or well-watered) are also included in the current report. Response means of plants grown under the different environmental conditions stated previously were taken from tables. The time of occurrence of stress during the crop cycle and the duration of stress applied differ among the studies, as indicated in Table S1. Most studies reported that treatments were maintained until the end of the experiments, when the plants reached maturity.

#### *4.2. Data Analysis*

All the data described above were organized in an Excel datasheet pairwise (control and experimental value) for each experimental factor ([CO2], temperature, water) (Table S1). The datasheet was loaded into and analyzed in RStudio v1.1.456 [111]. For the effect size metric, we used the natural log of the response ratio, lnR = ln(HF/LF), where LF and HF are reported mean nutrient concentrations at low and high treatment, respectively, with the treatment being any of the three climate factors considered in this study (CO2, temperature, or water). The log response ratio eliminates asymmetry between percentage decreases limited to 100% and unlimited percentage increases; it is a standard approach for analyzing elevated [CO2] and other ecological studies [112]. After performing statistical analyses, all the results were back-transformed to regular percentage changes using the formula: (exp(lnR) − 1)\*100%.For estimating the 95% confidence intervals for the mean effect size, a non-parametric test, namely bootstrapping with 999 replacements, was used for sample sizes of seven or more (i.e., when seven or more independent studies reported any given nutrient concentration at low and high treatments) [27]. The advantage of this approach is that it does not require the distribution of effect sizes to be normal. However,

for the confidence intervals to be accurate, they can be applied only for sufficiently large sample sizes (>7). For sample sizes <7, we had a choice of discarding the data completely, which would result in the loss of potentially valuable information, or making a normality assumption and applying a parametric method. We chose the latter for sample sizes of 3 to 7. No confidence intervals were derived for sample sizes of two or less. In all cases, unweighted methods were used, with each study having equal weight.

#### **5. Conclusions and Perspectives**

This study highlights that while current and near-future environmental conditions will severely affect cereal yield, the nutritional value of cereal grain will also be affected.

It seems that within the three factors related to climate change investigated, the rise in atmospheric [CO2] is possibly the one more detrimental and difficult to face because elevated [CO2] will impact grain quality traits all over the world while the impacts of the increase in temperature and the decrease in water availability will be localized or easy to counterbalance. In fact, although the increase in [CO2] might promote yield enhancement and starch accumulation through higher rates of photosynthesis, the grains of these plants will have lower concentrations of total proteins and minerals, leading to reduced baking quality and deficient nutritional value. On the other hand, even if both high temperature and drought severely decrease crop yields, the available data shows that grain quality will be differentially affected. Heat stress will negatively affect grain starch concentration due to depleted starch biosynthesis metabolism and shortening of the grain-filling period, but it might increase total proteins and N concentration. Regarding water availability effects, grain yield could be conditioned by the final starch concentration of affected plants. Adding to the increase in the Fe and Zn concentrations, we found that total protein concentrations are significantly increased, which is probably due to a dilution effect on starch and the accelerated reserve remobilization from source to sink to compensate for the nutrient uptake deficit that results from low soil water content. According to numerous climatic models, precipitation patterns are expected to change in the future with more frequent drought events in semiarid and arid regions but, it is also predicted that in other regions, precipitation will likely increase. Therefore, while drought and elevated temperature can be potentially mitigated (by increasing irrigation, planting crops at higher altitudes within a given latitude, or displaced to cooler and wet latitudes within a country), the effect of rising [CO2] is present at all latitudes and will act independently of where crops will be established. Hence, [CO2]-induced reductions in grain quality would be much more challenging to mitigate.

Our study highlights the fact that within the context of the present and near-future environments, it is crucial to increase crop yield through the development of stress-adapted cultivars. While the current breeding programs and agricultural incentives are almost exclusively yield-based, breeding for improved cereal quality can meaningfully improve the nutritional status of humanity. For this purpose, a better understanding of how environmental growth conditions (such as elevated temperature, drought, etc.) affect grain yield and nutritional parameters of cereals will help developing more nutrient-dense crops. Adding to that, exploring genetic diversity and variability of major crops is needed to discover genotypes more resilient to ongoing climate change.

**Supplementary Materials:** The following is available online at https://www.mdpi.com/article/10 .3390/plants10061052/s1, Table S1: The collected data and information used in the meta-analysis.

**Author Contributions:** Conceptualization, I.A.; methodology, I.A.; software, I.L.; validation, I.A., F.M., and B.Z.; investigation, S.B.M. and D.S.; writing—original draft preparation, S.B.M.; writing review and editing, S.B.M., I.L., and I.A. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by European Interest Group (EIG) CONCERT-Japan (IRUEC), the Spanish Ministry of Science and Innovation (Spanish MINECO projects PCIN-2017-007 and PID2019-110445RB-I00). Sinda Ben Mariem had a PhD grant from the Navarra Government.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data is contained within the article or supplementary material.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **Appendix A**

**Table A1.** List of papers and species used for the data analysis of each environmental factor.



**Table A1.** *Cont.*

#### **References**


## *Review* **Effects of Elevated CO<sup>2</sup> and Heat on Wheat Grain Quality**

**Xizi Wang and Fulai Liu \***

Department of Plant and Environmental Sciences, Faculty of Science, University of Copenhagen, Højbakkegård Allé 13, DK-2630 Tåstrup, Denmark; xiwa@plen.ku.dk

**\*** Correspondence: fl@plen.ku.dk; Tel.: +45-3533-3392

**Abstract:** Wheat is one of the most important staple foods in temperate regions and is in increasing demand in urbanizing and industrializing countries such as China. Enhancing yield potential to meet the population explosion around the world and maintaining grain quality in wheat plants under climate change are crucial for food security and human nutrition. Global warming resulting from greenhouse effect has led to more frequent occurrence of extreme climatic events. Elevated atmospheric CO<sup>2</sup> concentration (eCO<sup>2</sup> ) along with rising temperature has a huge impact on ecosystems, agriculture and human health. There are numerous studies investigating the eCO<sup>2</sup> and heatwaves effects on wheat growth and productivity, and the mechanisms behind. This review outlines the state-of-the-art knowledge regarding the effects of eCO<sup>2</sup> and heat stress, individually and combined, on grain yield and grain quality in wheat crop. Strategies to enhance the resilience of wheat to future warmer and CO<sup>2</sup> -enriched environment are discussed.

**Keywords:** elevated CO<sup>2</sup> ; grain quality; heat stress

**Citation:** Wang, X.; Liu, F. Effects of Elevated CO<sup>2</sup> and Heat on Wheat

Academic Editor: James Bunce

Grain Quality. *Plants* **2021**, *10*, 1027. https://doi.org/10.3390/plants10051027

Received: 30 April 2021 Accepted: 18 May 2021 Published: 20 May 2021

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

#### **1. Introduction**

Global atmospheric concentration of carbon dioxide (CO2) is expected to reach levels of 420 ppm (RCP2.6) to 1300 ppm (RCP8.5) by the end of this century (IPCC, 2013) with a concomitant rise in mean global temperature of about 2 ◦C by 2050 (IPCC, 2014). Extreme climatic events (ECEs), such as heat waves and droughts, which often affect plant growth and pose a growing threat to natural and agricultural ecosystems, are predicted to increase in frequency and severity in many cropping areas [1,2]. Wheat is one of the world's major food crops with an average annual global production of over 750 million tons from 2015 to 2019 (http://faostat.fao.org/ (accessed on 14 May 2021)). It is considered an important source of starch and energy. Wheat provides significant amounts of important nutrients, including proteins and mineral elements, as well as other components that are beneficial to human health, such as vitamins (especially vitamin B), phytochemicals and dietary fibers [3]. The effect of elevated atmospheric CO<sup>2</sup> concentration (eCO2) on wheat grain yield and grain quality has been well studied [4–6]. A general increase in grain yield and a reduction of grain quality of plants grown under eCO2, especially the decrease of nitrogen (N) concentration and thus also protein contents, have often been reported, leading to the conclusion that eCO<sup>2</sup> potentially exacerbates the prevalence of "hidden hunger" for human nutrition [7,8].

Another environmental factor, heat waves, limits wheat yields globally [9]. Air temperatures have increased since the beginning of the century and global temperature is predicted to increase 1.5–5.5 ◦C in the next 50 to 75 years [10]. Such increases in the temperature can lead to heat stress, a severe threat to wheat production, particularly when it occurs during reproductive and grain-filling phases [11]. Previous studies showed that exposure to temperatures above the optimum temperature for wheat at anthesis and grain filling stage (12 to 20 ◦C) can significantly reduce grain yield of more than 20% [12,13]. All the physiological processes of wheat plants are sensitive to temperature and can be damaged by heat permanently. Heat stress during anthesis can increase floret abortion [14]

and temperatures over 30 ◦C during floret formation may lead to complete sterility [15]. Heat stress during reproductive stage can lead to pollen sterility, tissue dehydration, lower CO<sup>2</sup> assimilation, increased photorespiration and reduced time to capture resources due to accelerated growth and senescence, consequently reducing the yield [11,16]. Besides the effects on grain yield, heat stress also has great impact on wheat grain quality. It was reported that heat stress could alter the ratio of gliadin to glutenin, which may lead to weaker dough properties and reduce baking quality [17]. However, there was a great diversity of the dough weakening effect under heat stress in other studies [18,19] because wheat quality mainly depends on genotype (G), environment (E) and their interactive effects (G × E) [20]. Several studies carried out at the CIMMYT showed diverse responses of different cultivars in quality traits under drought and heat stressed conditions. Hernandez-Espinosa et al. [21] reported that grain morphology (grain density and size), protein content and flour yield were strongly affected by the environment, while the traits related to gluten quality (gluten strength and extensibility, and bread loaf volume) were mainly determined by genotype, but the environmental and G × E effects were also important, especially for gluten extensibility. Li et al. [22] tested 15 quality parameters and found similar results that environmental factors have large effect on grain yield, while grain hardness and gluten quality-related traits are mainly controlled by genotype. Moreover, they found that drought and heat stress showed contrasting effects on dough rheological properties, where heat stress decreased dough tenacity (increased extensibility), slightly reduced dough strength and increased bread loaf volume while drought stress is the opposite. These results suggest that wheat quality is determined by many inter-related factors.

Although the impact of rising CO<sup>2</sup> concentration and elevated temperature on plant growth and development has been well studied, the combined effects of heat stress with eCO<sup>2</sup> on wheat crop performance remain unclear. Stomatal closure is induced by eCO2, which leads to the reduction of transpirational canopy cooling; however, higher temperature results in higher water vapor deficits in the air, causing an increased transpiration rate thereby offsetting the effect of stomatal closure induced by eCO<sup>2</sup> on crop water relations [23]. On the other hand, reduced stomatal conductance under eCO<sup>2</sup> will slow transpiration rate thereby alleviating the impact of water stress, which is often concurrent with heat stress [24]. However, plants' ability to sustain leaf gas exchange is dependent on genotype and stress intensity [25]. More detailed studies are required to assess the balance of these processes and how they acclimate to different environmental factors, because under eCO<sup>2</sup> growth environment plants' response to these abiotic stresses can be much more complicated. Therefore, this review summarizes current knowledge regarding the effects of eCO<sup>2</sup> and heat stress on grain yield and quality in wheat and aims to (i) introduce wheat quality traits that are sensitive to abiotic growth conditions, (ii) discuss the impacts of eCO<sup>2</sup> and heat stress on wheat grain yield and quality and the underlying mechanisms.

#### **2. Wheat Quality and Grain Protein**

Wheat is the most important staple food in the world due to its wide adaptation to diverse growth conditions and its unique property of gluten protein fraction in the grain. Wheat has evolved itself from emmer wheat into the cultivated species today by both nature and anthropogenic processes since primitive times (ca. 3000–4000 BC). There are two major wheat species nowadays utilized for food production. The first one is common bread wheat (*Triticum aestevum* L.), a hexaploid wheat that is mainly processed into baking products and the other one is tetraploid durum wheat (*T. turgidum* L. var. durum), which is used to make coarse flour (semolina) for pasta making. Wheat species can be classified by grain hardness, color (red, white and amber) and growing season (spring or winter wheat) and within each class, wheat grain can be evaluated by different grading factors [26]. Wheat grain quality is a combination of many specific parameters. The processing and end-use quality is determined by the multiple phenotypic characteristics of grain, flour, dough and final products [27]. Grain protein content (GPC) is considered as one of the most important components affecting the baking quality [28]. Besides, concentration of mineral

nutrients, grain hardness (milling properties), grain size, starch content and minor grain constituents such as lipids and soluble proteins also play important roles in determining end-use properties [10,29].

Baking quality mainly depends on the protein composition and concentrations of wheat grains. Wheat grain proteins are divided into three categories according to their solubility in different solvents: (1) water-soluble non-prolamins, albumins and globulins (ALGL), which play important roles in grain metabolism, development and response to environmental factors [30]; (2) gliadins (GLIA), which are soluble in 70% ethanol at room temperature and (3) glutenins (GLUT), which are soluble in dissolving media [31,32]. Gliadins and glutenins are storage proteins, which decide the baking quality, and they are collectively called gluten proteins. Gliadins are mixtures of single polypeptides and are classified into four subgroups: α-, β-, γ- and ω-GLIA, which can be separated by gel electrophoresis at low pH and by reverse phase high performance liquid chromatography (RP-HPLC) [33], whereas glutenins comprise the subunits that are aggregated together by disulphide bonds which are high (HMW-GS) and low molecular weight glutenin subunits (LMW-GS). Gliadins and glutenins are responsible for different biophysical properties where gliadins decide the dough viscosity while glutenins determine the dough elasticity and strength [32,34].

The concentrations and composition of gluten proteins are dependent on both the genotypic (variety) and the environmental factors (climate, fertilization, soil, etc.) [35], especially the availability of N fertilization [3,36]. Wieser and Seilmeier [31] reported that different N fertilization strongly influenced the quantities of gluten proteins where the effect on gliadins was more pronounced than that on glutenins, whereas albumins and globulins were barely affected. Moreover, the proportions of gluten proteins were changed significantly, where the hydrophilic proteins (ω-GLIA, HMW-GS) were increased by higher N supply and hydrophobic proteins (γ-GLIA, LMW-GS) were decreased. Daniel and Triboi [33] showed similar results that the percentage of proteins and gliadins in the flour increased with the increase of N supply and the proportion of ω-GLIA in total gliadin increased with N while the α- and β-GLIA decreased with N fertilization. In durum wheat, heat and/or drought stress during cultivation affect the grain quality attributes. According to Li et al. [37], drought tends to enhance gluten strength through increased lactic acid retention capacity (LARC) and mixograph peak time (MPT), while heat stress tends to decrease these parameters. Guzmán et al. [38] reported that the concentration of micronutrients (Fe and Zn) and flour yellowness (processing and pasta-making quality) in durum wheat were improved by drought but reduced by heat stress except Zn content, which increased under severe heat stress due to "concentration effect" induced by smaller grain. Moreover, heat stress was found to reduce gluten strength (alveograph energy, W) leading to a weakening effect on grain quality in durum wheat. However, in bread wheat, Hernandez-Espinosa et al. [21] did not detect the absolute weakening dough effect under severe heat stressed condition; both drought and heat stress led to higher gluten strength due to the higher protein content level. Joshi et al. [39] found that Fe and Zn concentrations in wheat grains varied across locations and years and are influenced by higher temperature and soil availability of Zn content in 30–60 cm soil depth. However, the effects of growing conditions are only based on quantitative effects, the influences on the structures, quantities and proportions of flour protein groups and the gluten protein types are largely dependent on the variety [31].

#### **3. The Effects of eCO<sup>2</sup> on Wheat Plants and the Mechanisms Behind**

Elevated CO<sup>2</sup> enhances plant growth (productivity and total biomass) through promoting net CO<sup>2</sup> assimilation rate (A) and improving water use efficiency (WUE) due to reduced stomatal conductance (gs) and transpiration in C<sup>3</sup> plants (e.g., rice and wheat) and therefore leading to a higher yield [40,41]. Wheat grain yield is mainly derived from photosynthates of leaf, stem and ear during the grain filling stage. The translocation of storage of

carbohydrates in stem only contributes 5–10% to final grain weight [42], while flag provides major source of photoassimilate to grain during the period of grain development [43].

According to Wang et al. [4], a meta-analysis of the effects of eCO<sup>2</sup> on wheat physiology and yield showed that eCO<sup>2</sup> (450–800 ppm) significantly increased A by 33% and decreased g<sup>s</sup> by 23%, Rubisco total activity by 26% and Rubisco content by 14%, hence increasing grain yield by 24%. However, the yield increase in free-air CO<sup>2</sup> enrichment (FACE) experiments was 44% less than those obtained in non-FACE (i.e., enclosure) facilities. While Broberg et al. [41] reported that wheat grain yield increased by 26% under eCO<sup>2</sup> (605 ppm) on an average, mainly through the increase in grain number. Similar results were found in rice where eCO<sup>2</sup> (627 ppm), on an average, increased rice yields by 23% through increasing grain mass, panicle and grain number, while FACE experiments showed only a 12% increase in rice yield [44]. In addition, Wang et al. [45] reported that eCO<sup>2</sup> enhanced rice yield by 20% on average, however, the yield responses to eCO<sup>2</sup> were smaller under FACE conditions (+16%) compared with other methods including greenhouses (+37%), growth chambers (+24%), and open-top chambers (+20%).

Moreover, the type of cultivars also plays an important role in determining the physiological and yield responses of crops to eCO2. Lv et al. [46] reported that eCO<sup>2</sup> enhanced rice yields by 13.5%, 22.6% and 32.8% for japonica, indica and hybrid cultivars, respectively, in FACE experiments. In wheat, eCO<sup>2</sup> had significantly different effects on A between the two types of wheat cultivars (i.e., spring and winter wheat) with a higher increase (71%) in spring wheat cultivars and only 23% increase in winter wheat cultivars [47]. In addition to grain crops, eCO<sup>2</sup> was also reported to increase the yield of vegetables, ornamentals, non-agricultural herbaceous species and woody species by 33% on average with a doubling atmospheric CO<sup>2</sup> concentration [48]. According to a meta-analysis conducted by Dong et al. [49], eCO<sup>2</sup> (827 ppm) generally enhanced yield of vegetables by 34% mainly through the increasing number of organs and vegetable mass. However, other environmental factors could modulate the eCO<sup>2</sup> effects on yield responses including temperature, light, water availability and nutrient supply. For example, the eCO<sup>2</sup> yield stimulation in wheat plants was stronger in the regions with low agronomic productivity [41] and the enhancement in A of wheat and rice plants was greater under sufficient nutrient compared to that under lower nutrient supply [4,44]. Therefore, optimizing growth environments are required to maximize the eCO2-induced yield benefit in agricultural systems, such as additional light (PAR) and high N availability [50,51].

Although eCO<sup>2</sup> can enhance the productivity and yield of agricultural crops, it on the other hand alters and decreases the plant quality, depressing the concentrations of macronutrients (i.e., carbohydrates, protein, and fat) and micronutrients (i.e., minerals, vitamins and phytonutrients) [7,8,52]. Dong et al. [53] reported that eCO<sup>2</sup> increased the concentrations of fructose, glucose and total soluble sugar in the edible part of vegetables but decreased the concentrations of protein, N, magnesium (Mg), iron (Fe) and zinc (Zn) by 9.5%, 18.0%, 9.2%, 16.0% and 9.4%, respectively. A meta-analysis conducted by Taub et al. [54] on several major food crops showed that eCO<sup>2</sup> (540–958 ppm) reduced cereal grain protein concentrations by 10–15% on an average in wheat, barley and rice and reduced tuber protein concentration in potato by 14%. Moreover, for soybean, there was a smaller but statistically significant decrease of protein concentration of 1.4% compared with that grown under ambient CO<sup>2</sup> concentration (aCO2). In rice, the nutritive value of grains was also negatively affected by eCO<sup>2</sup> under FACE conditions through a decrease in protein by 6% and copper (Cu) content by 20% [55].

It is well known that eCO<sup>2</sup> has direct effect on C and N metabolism in wheat, resulting in changes in the chemical composition in plants [8,56]. In a FACE experiment, Högy et al. [5] reported that eCO<sup>2</sup> leads to an overall reduction in grain protein concentration by 7.4% in spring wheat cultivars, particularly the N- and glutamine-rich gliadin fraction, thus lowering the gluten concentration and reducing baking quality. Along with the lowered protein concentration in grain, the composition of amino acids and their concentrations were also modified under eCO2, and the size distribution was significantly shifted towards smaller grains [5]. Blandino et al. [57] also suggested a 7% decrease of protein content in four winter wheat cultivars, accompanied by a reduction in dough strength of plants grown under FACE conditions. A three-year field trial in Australia demonstrated that eCO<sup>2</sup> consistently decreased baking quality and grain protein content in wheat, and protein composition changed towards a greater glutenin/gliadin ratio in all years [58]. Moreover, eCO<sup>2</sup> could also reduce concentration of minerals in wheat grain. For instance, a significant decrease in concentrations of Zn and Fe in wheat grain has been reported [6]. Similar reduction in grain S [59], Ca, Fe and Zn [60] concentrations was documented for wheat grown under eCO2.

There are many different hypotheses explaining these changes of grain quality traits (summarized in Figure 1) and the most frequently mentioned one is the dilution effect, which includes two aspects, biomass dilution and functional dilution [61]. Biomass dilution mainly results from the accumulation of non-structural carbohydrates (NSC) (i.e., soluble sugars, starch, fructans, etc.) which are initial long-term C-storage products of enhanced photosynthesis induced by eCO<sup>2</sup> and therefore reduces the concentration of other constituents [62]. It means that proteins and minerals are diluted by the increased photosynthetic assimilation of carbon, which poses a great threat to future food security because the calories in the food may be sufficient but undernourishment with essential mineral nutrients. For example, eCO<sup>2</sup> greatly increased the ratio of C to N by increasing the starch and total NSC concentrations and decreasing grain N concentration in wheat [63]. However, Taub et al. [54] found that the magnitude of the negative eCO<sup>2</sup> effect on wheat grains was smaller under high soil N conditions than under low soil N with the decrease of grain protein concentrations by 9.8% and 16.4%, respectively, suggesting that the dilution effect may be compensated by a higher nutrient supply. Besides, a relative increase in the synthesis of C-based secondary compounds that are low in N (e.g., lignins, tannins or other polyphenolics) may lead to dilution as well [64], but there was no consistent response [65–67].

Functional dilution refers to a decrease in dry mass concentration of N (Nm) due to increased shoot specific activity, which means that N concentration declines due to the accumulation of additional photosynthates by shoots [61]. This theory bases on the functional-balance model:

$$\text{Nm} \propto \frac{\text{root mass} \times \text{rate (absorption)}}{\text{leaf mass} \times \text{rate (photosynthesis)}'} \tag{1}$$

where the tissue N concentrations are dependent on the relative activities of roots and shoots and the partitioning of photosynthates is controlled by the relative rates of root absorption of soil nutrients and leaf photosynthesis [68,69]. Functional dilution seems to be pervasive because increased photosynthetic rates for plants grown under eCO<sup>2</sup>

is frequently observed if we assume the shoot specific activity equal to photosynthetic rate [70].

Within either biomass dilution or functional dilution, all other mineral elements except C, H and O that are assimilated through photosynthesis should be diluted to a similar ratio. However, there are many heterogeneous responses of different mineral concentrations for crops under eCO2. For instance, the decrease in Zn concentrations in rice grains under eCO<sup>2</sup> was significantly different from those in Cu, Ca, B and P [6] indicating that dilution is not the only mechanism responsible for decreasing nutrient concentrations in plants under eCO<sup>2</sup> [54,61,71]. Li et al. [72] reported that both the concentrations of K, Ca and Mg in wheat organs and the total accumulations of these elements in the plants were significantly decreased by eCO2. Moreover, they also found decreases of the concentration of these minerals in the xylem sap, suggesting that the reduced mineral concentrations were not only because of dilution effect but also due to a reduced nutrient acquisition by roots. Within the functional balance concept mentioned above, decreased specific root activity could also play a role in the reduction of nutrient concentrations in plants through a decreased root uptake rate under eCO2, but the effect is larger for plants grown in soil than in hydroponics [73]. It is reported that the average mean decrease of specific root uptake rate of N (i.e., uptake per unit root mass or length) for plants grown in solid media is 16.4% [74–76]. While the results for root uptake kinetics in solution were quite different and variable [69], but the overall trend was to increase rather than decrease the specific N uptake [77,78]. Therefore, there are some elements influencing the root uptake that are not present in hydroponics. This indicates that the soil microbiomes, rhizosphere conditions and/or root architecture may play a role in nutrient uptake as well.

When it comes to the acquisition of nutrients, there are two aspects: uptake and demand by the plants. On the one hand, eCO<sup>2</sup> could affect the ability of soil-root system to supply N, which refers to source effect. On the other hand, eCO<sup>2</sup> could also increase source use efficiencies, which allows plants to sustain growth in a lower N concentration leading to a lower demand of nutrients (demand effect) [61]. Besides the root specific activity mentioned above, it is widely recognized that the eCO2-induced decrease in plant mineral uptake associated with the reduced mass flow or diffusion of mineral ions from the soil solution to the root surface due to lower transpiration rate which results from a reduced stomatal conductance under eCO2. There is evidence that eCO<sup>2</sup> enhances the root growth, which may enable the plants to acquire more nutrients; while a number of studies have also indicated that eCO<sup>2</sup> depresses root hydraulic conductance [79] probably via downregulating genes encoding aquaporin hereby reducing the mass flow. In addition, it has been proposed that the effect of transpiration rate on mineral uptake is more pronounced with those primarily transported via mass flow (e.g., N, Ca and Mg) than those via diffusion (e.g., P, K and most micronutrients) [80]. This may lead to a shift in the stoichiometry, further affecting nutritive value of the grain [7]. However, to date these possible effects of eCO<sup>2</sup> on root water and mineral uptake have not been fully illustrated.

#### **4. The Effects of Heat Stress on Wheat Grain Quality**

Wheat is a temperate crop adapted to temperatures below 30 ◦C and the threshold temperature during post-anthesis stage is 26 ◦C [18]. Heat stress (over 35 ◦C) has a huge impact on both grain yield and grain quality during anthesis and grain-filling phase; however, most of the studies focused on the heat stress effects on grain yield components [18,81,82], only a few studies investigated the impact of heat stress on grain quality traits [17,18,35,83,84]. Increased temperature affects grain development due to the limitation of assimilate supply, grain-filling duration and rate, and starch biosynthesis and deposition [11]. The effects of heat stress on wheat grain quality are summarized in Table 1. The response of wheat to heat stress varied between genotypes and the degree of heat-caused damage depends on the intensity, duration and frequency of heat stress [83,85].


**Table 1.** The effects of heat stress on different grain quality traits.

The "+" sign indicates an increase and the "−" sign indicates a decrease in the performance of the trait of interest.

Although extremely high temperatures have more detrimental impact on grain quality, within moderate or chronic temperature range (15–35 ◦C), there are also marked changes in grain quality. For example, dough strength (measured by resistance to extension of a dough piece in the Babender Extensograph) was increased with rise in daily average temperature up to about 30 ◦C; however, when the temperature was above this threshold value to max. 36 ◦C, even applied for only 3 days, it tended to decrease dough strength. This occurred independent of the timing of the stress, but the degree of reduction varied with different growth stages [89]. In addition, the protein content per grain increases without a change for starch when the temperature rises between 15 and 21 ◦C during grain-filling stage; however, when temperature reaches 30 ◦C, the deposition of both protein and starch reduces but with more pronounced decrease in starch than protein [86]. These discrepancies indicate that different growth stages have different threshold and sensitivity to high temperatures. However, the overall higher temperatures generally reduce wheat grain yield through producing smaller grains, reducing grain number and lowering kernel weight [15,90,91], and alter grain quality by increasing GPC but decreasing starch deposition and functional properties of wheat flour [18,35,87], although different wheat genotypes respond differently under heat stress [84,92,93]. For example, Castro et al. [84] evaluated 14 spring wheat genotypes to characterize their response to high temperatures and detected a significant genotype × treatment interaction, suggesting that varieties possess a thermos tolerant response could be used as genetic sources for breeding heat tolerance wheat cultivars.

GPC is the most important characteristic determining wheat grain quality and is in essence determined by the relative rates and durations of protein and starch synthesis [88]. Starch accounting for 65–75% of wheat grain dry weight and over 80% of endosperm weight, which is a decisive factor of grain yield and flour quality [94]. During grain development, starch is deposited into three types of granules differing in size and their formative period in amyloplasts. Large lenticular A-type granules with diameters greater than 15.9 µm are synthesized early during endosperm development; spherical B-type granules with diameters between 5.3 µm and 15.9 µm are produced during mid-development and smaller C-type granules with diameters less than 5.3 µm are initiated late in development [95]. It is reported that high temperatures applied post anthesis reduced the duration of starch accumulation and starch content and modified the size distribution of starch granules with less B-type granules produced in the grain under higher temperatures [88,96]. For the functional properties, starch is composed of two classes of glucose polymers: straight-chained amylose, which is an almost linear α-1,4 glucan molecule comprising 25–30% of grain starch, and highly branched amylopectin, which constitutes 70–75% grain starch [97,98]. Under heat stress during grain-filling stage, the amylose to amylopectin ratio increases, leading to a reduction in dough elasticity [99]. Extension of α-1,4 glucan chains is catalyzed by starch synthases, which are sensitive to heat stress [100], indicating that high temperatures decrease metabolism and enzyme activities involved in starch biosynthesis and reducing the rate of conversion of sucrose to starch [101]. The decreasing starch deposition

affects protein concentration by allowing more N per unit of starch [91], leading to smaller grain size, which as a result causing a decrease in milling quality [83].

Under heat stress, the reduction of dough properties of wheat is mainly associated with the reduction of glutenins in gluten proteins especially the HMW-GS [102] that accounts for the genotypic variance in wheat quality [35]. Although there is a general increase in grain protein proportion relative to starch content under high temperatures, due to less temperature sensitivity of N accumulation than starch deposition [88], the protein composition alters towards a poorer flour quality by the following reasons summarized by Blumenthal et al. [102]. First, heat stress decreases synthesis of glutenins therefore leading to the reduction in glutenin/gliadin ratio with gliadin synthesis being maintained or increased [103]. This is explained by the molecular mechanism that there are heat-shock elements (HSE) in the upstream of coding regions of gliadin but not for glutenin [104,105]. However, the effect also depends on genotypes. Stone and Nicolas [18] reported variety difference in glutenin/gliadin ratio in response to heat stress where only one variety (Oxley) showed a decrease in this ratio and Sun 9E-16 increased in this component while the other three varieties had no significant response to high temperature. This suggested that wheat varieties vary in their response of gliadin synthesis to heat stress. The second hypothesis is that heat shocks lower the degree of polymerization of glutenin subunits and reduce the large sized glutenin polymers by altering the formation of disulphide bonds between glutenin peptides therefore weakening the dough properties [106]. In addition, heat-shock proteins (HSP) play an important role in determining the dough-protein function. On one hand, HSP per se is involved in guiding the formation, folding and polymerization of peptides in the grain and can disaggregate and hydrolyze deformed proteins under stress conditions. Therefore, under heat stress, HSP may break off the glutenin synthesis and polymerization thus influencing dough structure [99]. On the other hand, the synthesis of HSP and their prevailing presence in the mature grain could result in the heat-related loss of dough quality. However, according to Blumenthal et al. [102], although the concentration of HSP 70 in mature grain increased under a few days' heat treatment, there was no strong correlation between the amount of HSP 70 and the loss of dough strength. Moreover, they failed to find the presence of HSE in the upstream coding region of the genes for HMW-GS, suggesting that the weakening of dough properties under heat stress may be more relevant to the degree of polymerization of glutenin molecules and the roles of HSP family during grain development.

Besides GPC, heat stress also affects mineral nutrition of wheat grains with a general decline in the concentration of micronutrients, especially Fe, Zn and Mn [38,107], but the results strongly depend on varieties and meteorological factors. For example, Dias et al. [108] reported that under heat stress, Fe concentration in the stems and leaves decreased in bread wheat but increased in durum wheat, and Mn concentration increased significantly in shoot during grain-filling stage for all genotypes except Golia, which is a less heat tolerant genotype. In addition, Kumar et al. [107] reported that under heat stress, the accumulation of Fe and Mn in flag leaf and spike diminished significantly during booting and grain-filling stages, while some of the varieties showed increased accumulation of micronutrients in either flag leaf or spike. Narendra et al. [109] found a high Zn or Fe content in some heat-tolerant varieties under heat stress, suggesting that these genotypes can be used for future breeding to cope with the problem of malnutrition. Furthermore, it is reported that the proportions of gliadins and polymeric protein were less affected by heat stress in the grain with high Zn concentrations, indicating that grain Zn nutrition may interact with grain-filling temperatures and alter protein composition under heat stress [110].

#### **5. The Interactive Effects of eCO<sup>2</sup> and Heat Stress on Wheat Plants**

So far, there are many studies of plant response to single eCO<sup>2</sup> or high temperature stress, but the research on the combined effect of these two factors on wheat plants is quite limited, especially on grain quality traits. Kadam et al. [111] summarized that the eCO<sup>2</sup> ×

high temperature interaction strongly depends on the growth and developmental stages of plants. During vegetative stage, heat stress decreases net CO<sup>2</sup> assimilation rates in wheat grown under aCO2, but eCO<sup>2</sup> moderates the negative effect of heat stress on canopy photosynthesis. This is because increased temperature will decrease the ratio of solubility of CO<sup>2</sup> and of O<sup>2</sup> in water and reduce the specificity of Rubisco for CO<sup>2</sup> relative to O<sup>2</sup> [112], leading to a preference to oxygenation rather than carboxylation in C<sup>3</sup> photosynthesis. Rising CO<sup>2</sup> concentration can inhibit photorespiration and increase ribulose bisphosphate (RuBp) regeneration capacity thus increasing the net photosynthesis. However, the response varies between species. For example, in rice, leaf photosynthesis increased up to 63%, while in sorghum, the increase was not significant [111]. In wheat, Abdelhakim et al. [113] tested the physiological responses of different spring wheat genotypes of heat-tolerance to eCO<sup>2</sup> × heat interaction and found that under eCO2, all genotypes showed higher photosynthetic rate and maintained maximum quantum efficiency of PSII under heat stress compared to aCO2.

Despite the advantage brought by eCO<sup>2</sup> on the vegetative tissue, during the reproductive phase (especially anthesis and grain-filling stage), temperature is the main factor determining the wheat grain yield and quality when exposed to both eCO<sup>2</sup> and heat stress [111]. This is due to the irreversible damage of high temperatures to the anabolic and metabolic processes in wheat flowers and grains, which has been mentioned in other studies only focusing on the heat stress effects [11,25]. Chavan et al. [81] reported that although eCO<sup>2</sup> alleviates the negative impact of heat stress on photosynthesis in wheat, grain yield was reduced equally under aCO<sup>2</sup> and eCO<sup>2</sup> by heat stress due to grain abortion and shortened grain filling duration. Moreover, Macabuhay et al. [114] found an increase in stem water-soluble carbohydrates (WSC) under eCO<sup>2</sup> in both control and heat stressed wheat plants; however, there is limitation of WSC translocation to grains under heat stress, which decreases WSC remobilization. Therefore, the overriding impact of heat stress on crop production seems to limit the potential benefits provided by carbon fertilization of eCO2. However, there are some positive responses of wheat in the combined situation (reviewed by Kadam et al. [111]) such as increase in grain yield, number of ears and harvest index. While for thousand grain weight, number of spikelets and grain number, the results were inconsistent, indicating a partial ameliorative effect of eCO<sup>2</sup> on grain yield when combined with heat stress. Furthermore, varietal difference, seasonal conditions and different experimental treatment applied also contribute to the inconsistency of the plant response under such situation. Besides, the interaction of weeds such as little seed canary grass (*Phalaris minor*) and common lambsquarters (*Chenopodium murale*) in wheat field could accelerate the yield loss under eCO<sup>2</sup> and thermal scenario due to the similar photosynthetic pathway and nutritional level in weeds and main crop, and the greater response of weeds than crop to eCO<sup>2</sup> condition [115].

Although eCO<sup>2</sup> cannot buffer the negative impact of heat stress on reproductive stage, the reduction of grain N concentration under eCO<sup>2</sup> could be slightly alleviated in heat-treated plants because N accumulation is less sensitive to temperature than C in wheat grain resulting in less carbohydrates relative to N [114]. However, overall, the grain quality seems to decrease in the case of eCO<sup>2</sup> and high temperature combined situation mainly due to the reduction of storage protein under both conditions [33,116], especially the synthesis of GLUT polymers leading to a diminished dough functionality and baking quality. The effect of eCO<sup>2</sup> and heat stress interactions on mineral nutrient composition of wheat grains was rarely investigated. However, a study in rice under field conditions showed a significant decrease in grain mineral content, including Ca, Mg, Cu, Fe, Mn and Zn under combined eCO<sup>2</sup> and heat stress situation across different cultivars. Moreover, some of the micronutrients (Cu, Fe and Zn) were reduced more prominently as compared to eCO<sup>2</sup> alone [117], suggesting that heat stress may further exacerbate the negative effect of eCO<sup>2</sup> on grain mineral nutrition.

#### **6. Conclusions**

Although eCO<sup>2</sup> increases wheat grain yield, its negative impact on grain quality poses a great threat to human nutrition. This is mainly due to the dilution effect by higher accumulation of photosynthetic assimilates of C and the decreased root N uptake associated with lower transpiration rate and mass flow under eCO2. Heat stress generally counteracts the positive effect of eCO<sup>2</sup> on yield components and may aggravate the negative effect of eCO<sup>2</sup> on grain quality because wheat is quite sensitive to high temperature stress especially during anthesis and grain-filling stage, which leads to permanent and irreversible damage during flower and grain development. However, grain quality is strongly dependent on variety and environment, and different quality attributes show diverse responses to abiotic stresses. Since eCO<sup>2</sup> cannot protect wheat plants from high temperature stress, other environmental factors should be taken into consideration, such as enhancing nutrient fertilizing and improving soil water content, etc. Selecting and breeding new genotypes of heat tolerance could be another way to deal with climate change and the increasing demand for food. Moreover, the contemporary crop models for evaluating the effects of different environmental conditions on wheat quality can provide new insights into adaptation strategies to cope with the impacts of climate change on global crop production and grain quality.

**Author Contributions:** Conceptualization, X.W. and F.L.; writing—original draft preparation, X.W.; writing—review and editing, F.L.; supervision, F.L. 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.

**Data Availability Statement:** The article does not contain original data.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

