Transcriptional Regulation of Metabolic and Cellular Processes in Durum Wheat (Triticum turgidum subsp. durum) in the Face of Temperature Increasing

The Yaqui Valley, Mexico, has been historically considered as an experimental field for semiarid regions worldwide since temperature is an important constraint affecting durum wheat cultivation. Here, we studied the transcriptional and morphometrical response of durum wheat at an increased temperature (+2 °C) for deciphering molecular mechanisms involved in the thermal adaptation by this crop. The morphometrical assay showed a significant decrease in almost all the evaluated traits (shoot/root length, biovolume index, and dry/shoot weight) except in the dry root weight and the root:shoot ratio. At the transcriptional level, 283 differentially expressed genes (DEGs) were obtained (False Discovery Rate (FDR) ≤ 0.05 and |log2 fold change| ≥ 1.3). From these, functional annotation with MapMan4 and a gene ontology (GO) enrichment analysis with GOSeq were carried out to obtain 27 GO terms significantly enriched (overrepresented FDR ≤ 0.05). Overrepresented and functionally annotated genes belonged to ontologies associated with photosynthetic acclimation, respiration, changes in carbon balance, lipid biosynthesis, the regulation of reactive oxygen species, and the acceleration of physiological progression. These findings are the first insight into the regulation of the mechanism influenced by a temperature increase in durum wheat.


Introduction
In the Yaqui Valley, the birthplace of the Green Revolution, agriculture is based on intensive practices such as monoculture, mechanization, and large-scale agrochemical applications. This region is responsible for approximately 54% of the national wheat production, mainly durum wheat (Triticum turgidum subsp. durum) [1]. However, this valley is susceptible to elevated temperatures [2,3], which puts the maintenance of food production at risk [4][5][6]. Climate is one of the most important determinants of durum wheat yield and accounts for 30% to 50% of the overall output variability [7]. Therefore, crop growth and yield under rising temperature scenarios are receiving increased attention. For example, the temperature effect has been calculated to reduce the durum wheat yield from 4.1% to 6.4% per Celsius degree increase [8], and a loss of yield of 24% under a growth temperature of 31 • C in the flowering stage [7].
Plants respond to elevated temperatures in various ways; an increase of 2 • C to 5 • C above the optimal conditions can lead to a process known as thermomorphogenesis, which consists of changes in development and morphology influenced by a moderate increase in temperature [9]. The physiological effects of thermomorphogenesis are characterized by a fully expanded leaf structure, a larger root system, and early flowering. These are aimed at reducing the exposure of meristematic tissue to adverse conditions [10]. A temperature above this threshold of 5 • C to 10 • C, from this point called heat stress, can lead to a decline in pollen viability, starch synthesis, and grain filling. Additionally, severe thermal stress conditions can produce seed sterility due to the sensitivity of microspore and megaspore development [11].
Durum wheat cultivars developed in tropical regions are mostly adapted to short periods of temperatures above 20-30 • C; however, the temperature-influenced growth could have negative implications depending on exposure time, i.e., 5 days at temperatures above that range. Thus, an increase in temperature impacts yield-related traits, such as the thousand-kernel weight and reproductive tiller number, explaining approximately 52% of phenotypic variance [12].
In durum wheat, the stages most sensitive to heat stress are anthesis and the vegetative period; both stages are characterized by being crucial for the acquisition of nutrients (vegetative stage) and the grain filling (anthesis) [11]. Physiologically, the effect of heat stress is presented by a reduced photosynthetic rate, accelerated development, reduced flowering time, and fewer grains per spike [13]. It has been observed that the negative effects on yield may be imperceptible up to an upper limit of 31 • C near the flowering stage. This may depend on the genotype, the availability of water, and the stage of development [14]. A temperature above 31 • C could inhibit the transport of photoassimilates and nitrogen compounds from leaves and stems to grains [10,15].
The application of multi-omics approaches has allowed the identification of several molecular mechanisms of response to heat stress in model plants (Arabidopsis spp.), although these mechanisms are diverse and depend on specific conditions. The main temperature sensors in plants have been reported to be phytochrome B (PhyB) and phototropin, which respond to moderate changes in basal temperature and variation in red light. Once warm temperatures and red light are detected, PhyB promotes the accumulation of interactive phytochrome factors (PIF (bHLH family of transcription factors)); PIFs enhance tissue elongation through auxin signaling to result in a larger leaf area to cool the entire plant [10]. Another downstream acclimatization process involves the expression of heat shock proteins (HSP), reactive oxygen species (ROS) signaling, osmotic regulation, water transport, and cell wall modifications, among others [16].
Thus, transcriptional information suggests that temperature sensing is organ specific, with the autonomy of the root tissue triggering the cell elongation process independently of the upper sections of the plant, while the shoot responds locally and systemically [17]. This implies that the thermal detection mechanism in the roots is different (and not characterized) from PhyB and phototropin signaling because they depend on the variation of light, which is absent in the roots [17].
The above suggests that a lack in the fine thermo-sensing processes could be still unknown in the majority of crops of agricultural importance, such as durum wheat. The present study aims to analyze the physiological and transcriptional response of durum wheat seedlings to conditions of increased temperature (+2 • C). The findings constitute the first insight at the regulation of the mechanism influenced by a temperature increase in wheat, which represents valuable information to improve our understanding of durum wheat adaptation strategies and enhance the development of new heat-tolerant varieties to help farmers in vulnerable regions cope with increasing climate risks.

Results
The morphometrical assay was carried out to compare the physiological effect of an elevated temperature (+2 • C) vs. the optimal temperature condition for durum wheat production in the Yaqui Valley (28 • C). Thus, a negative significant (Tukey-Kramer test, p-value < 0.05) effect on root length (−20%), dry shoot weight (−24.5%), and biovolume index (−28.2%) was observed by elevated temperatures. However, the root:shoot ratio showed a positive significant effect (57.9%) under that treatment (Table 1). On the other hand, no significant differences were observed in the shoot length (−6.7%), and dry root weight (10.8%). However, it was also observed that the dry root weight and root:shoot ratio tended to increase in seedlings under increased temperature (Table 1). Means (n = 21) ± standard deviation with an asterisk (*) are significantly different (increased vs. optimal temperature treatments), according to the Tukey-Kramer test (p < 0.05).

Validation of RNA Extraction from Durum Wheat Seedlings, Quality, and Sequencing
The RNA extraction from whole durum wheat seedlings (root and foliar tissue) had high-quality RNA, 92.73 ± 24.1 ng/µL, for seedlings growing at the optimal temperature (T_optimal), and 109.73 ± 32.4 ng/µL for seedlings under increased temperature (+2 • C) (T_heat). Additionally, all RNA Integrity Number (RIN) values were at 8.0-9.0, which indicates that isolated RNA is suitable for RNA-seq analysis [18].

Morphometrical Influence of An Elevated Temperature
In this study, we observed a significant decrease in several morphometrical traits in plants grown under an increase in temperature, i.e., root length (−20%), dry shoot weight (−24.5%), and biovolume index (−28.2%) ( Table 1). This coincides with a significant loss of biomass in maize and broccoli seedlings at a temperature above 30 • C in the vegetative stage, previously reported by Hatfield and Prueger (2015) [21]. Such effects could be attributed to the inhibition of the transport of photoassimilates and nitrogen compounds through plant structures and protein denaturation [10,15].
However, a significant increase of 57.9% in the root:shoot ratio (dry mass) was also observed; this suggests a mobilization and reprogramming of nutrients from the upper tissues to the root system, as an early response of adaptation to the increases in temperature [9,22]. Such behavior was also described by Zhang et al. (2015) [23]. in A. thaliana under heat stress of 42 • C to 45 • C, with and without a gradual acclimation process. This gradual increase in temperature maintains or increases the root growth and surviving rate. This strategy has been reported in other Arabidopsis accessions under high ambient temperatures, suggesting that root nutrient mobilization is required to move sensitive and active meristematic tissue away from the shallow soil, which absorbs heat and can promote cooling by allowing better access to water [15]. Additionally, adaptation to a high ambient temperature also involves physiological processes, such as photosynthetic acclimation, respiration, carbon balance changes, cell-wall modifications, and reactive oxygen species (ROS) regulation.

Acceleration of Development
The progression of the developmental stage was evidenced by the overrepresentation of ontologies associated with this process, i.e., inflorescence development (GO:0010229 (FDR: 0.0121)), sporopollenin biosynthetic process (GO:0080110 (FDR: 0.0140)), photoperiodism and flowering (GO:0048573 (FDR: 0.0143)), regulation of timing of transition from the vegetative to reproductive phase (GO:0048510 (FDR: 0.0249)), spindle (GO:0005819 (FDR: 0.0363)), and phragmoplast (GO:0009524 (FDR: 0.0309)) ( Figure 3). In durum wheat, the stages most sensitive to heat stress are anthesis and the vegetative period; both stages are characterized by being crucial for the acquisition of nutrients (vegetative stage) and the grain filling (anthesis) [11]. Physiologically, the effect of heat stress is presented by a reduced photosynthetic rate, accelerated development, reduced flowering time, and fewer grains per spike [13]. It has been observed that the negative effects on yield may be imperceptible up to an upper limit of 31 • C near the flowering stage. This may depend on the genotype, the availability of water, and the stage of development [14]. This acceleration of plant development has been proposed as an effect of the modification of the circadian clock [24]; some of the signals regulating this mechanism include light and temperature and are associated with the compensation of plant growth in seasonal changes [25]. A study in A. thaliana reported that circadian modifications and their physiological effects, such as an accelerated change to the reproductive stage, are more significant in the increase in temperature than in the decrease [26].

Photosynthesis and ATPase Activity
At the transcriptional level, we highlight the DEGs that belong to the ontologies below. The genes in Bin 1 "Photosynthesis" were upregulated (UR), which included the CGL160 factor (TRIDC3BG045570, (FC: 1.689)), the light-harvesting complex LHCb1/2/3 (TRIDC5BG070090 (FC: 1.420)), and Kinesin-like protein KIN-7D (TRIDC7AG021480 (1.375)) (Figures 2 and 4). These genes are related to ATPase activity and have been reported to actively regulate photosynthetic activity under different luminosity conditions [27][28][29][30]. Previous studies in A. thaliana have described that the molecular response of the upper sections of plants to thermal increase is highly related to light sensors as such increases usually depend on the luminous intensity [18,24].
On the other hand, photosynthesis activates the phosphorylation and accumulation of ATPase, which have effects on the maintenance of lipid membrane integrity. It has been reported that the overexpression of this process has positive implications for thermal tolerance in Arabidopsis [31,32]. The aforementioned is suggested by the overrepresentation of ontologies related to photosynthetic activity and thylakoid membrane to compensate the effect of increases in temperature, such as chloroplast thylakoid membrane protein complex (GO:0098807 (FDR: 0.008)), cellular response to light stimulus (GO:0071482 (FDR: 0.0148)), chloroplast (GO:0009507 (FDR: 0.0219)), chloroplast envelope (GO:0009941 (FDR: 0.0492)), and plastoglobule (GO:0010287 (FC: 0.0009)) ( Figure 3). The activation of photosynthesis has been proposed as one of the first responses to thermal sensitivity and carries out the acclimatization process; this activation involves ATPase accumulation, stomatic opening, and light detection (due to the light and heat share sensing mechanism) as strategies to cope with the thermal increase [10]. It is also a requirement to accelerate physiological development to accomplish the biological cycle of plants [33].

Regulation of Lipid Biosynthesis
Another behavior observed in plants at the threshold of thermal stress is the increase in water accumulation through the synthesis of lipids and the reinforcement of the cell wall [34]. High temperatures affect the lipid composition and viscosity of the membrane, the hotter temperature made more fluid cell membranes. This phenomenon is attributed to the activation of the accumulation and remodeling of the lipid composition of the membrane ( Figure 4) [35,36].
The main strategy that plants use to lessen the effects of oxidative stress is peroxidase activity. This process was highly represented in the ontology enrichment analysis, specifically the hydrogen peroxide catabolic process (GO:0042744 (FDR: 0.0073)), oxidation-reduction process  (Figures 3 and 4).
Additionally, the influence of post-transcriptional regulation through non-coding RNAs (miRNA, siRNA, lncRNA) is well known. It has been reported that the target genes of such RNAs are generally those with ontologies related to cell wall architecture, ROS regulation, and the development of the next phase, among others [40,41]. Similarly, epigenetic mechanisms have been found to be involved in the response to thermal stress, such as DNA methylation, histone modifications, and chromatin remodeling [42]. For the above, the struggles to fully decipher the response to thermal stress will encompass various levels of molecular organization.
Two treatments were evaluated (n = 66 seedlings per treatment (21 for morphometric assay and 15 to RNASeq (in triplicate)): (i) seedlings grown at optimal temperature conditions (28 • C days and 15 • C night) for durum wheat production in the Yaqui Valley, called "T_optimal" and (ii) seedlings grown under conditions of an elevated temperature (+2 • C) compared to the optimal conditions (30 • C days and 17 • C night), called "T_heat". These seedlings growing in sterilized hermetic containers were placed in a growth chamber (Biobase #BJPX-A450, Shandong, China) with the following parameters according to durum wheat requirements: photoperiod day/night 14:10 for 45 days (development stage   [48,49].

RNA Extraction and RNA-Seq Analysis
Fifteen seedlings were harvested per treatment (in triplicate) and frozen using liquid nitrogen. The samples were then ground in a DEPC-treated mortar and stored at −80 • C until processing [50]. Total RNA was isolated using a modified TRIzol ® (Life Technologies; ThermoFisher, Waltham, MA, USA) method described by Chaparro-Encinas et al. 2020 [51]. The extracted total RNA was treated with DNase I (Ambion™) to remove genomic DNA according to the manufacturer's instructions, and subsequently, RNA Integrity Number (RIN) was obtained using a BioAnalyzer Agilent 2100 (Agilent Technologies, Santa Clara, CA, USA) to evaluate their suitability for downstream applications. The total RNA samples with RIN >8 were considered suitable for library preparation [18], with the TruSeq Stranded Total RNA Library Kit (Illumina, San Diego, CA, USA), and sequenced by Illumina ® 2 × 150 en el NextSeq 500 platform (2 × 150 bp). A yield of approximately 85 million paired-end reads per library was obtained.
Transcript abundance counts were used for gene-level differential expression analysis using the DESeq2 package [56]. A principal component analysis (PCA) was performed with DESeq2 function to reproducibility and biological variations among samples purposes. Differentially expressed genes (DEGs) were defined as those with adjusted p-value (False Discovery Rate, FDR) <0.05 and a fold change of at ± 1.3. Subsequently, DEGs were annotated with Mercator4 and visualized with MapMan4 [19]. Gene identifiers and Log2 fold change values of DEGs were imported into the MapMan4 framework, and transcripts were assigned into Bins.
To enriched-gene ontology analysis, all gene ontology (GO) terms associated with DEGs were extracted from the BioMart database [57]. Then, the R package GOseq [20] was used to perform a gene length bias correction with Wallenius non-central hypergeometric distribution. Only those GO terms with overrepresented FDR <0.05 were considered as significantly enriched.

Statistical Analysis
For durum wheat morphometric data (shoot and root length, dry weight of shoots and roots, and biovolume index (stem circumference x shoot length) the statistical analysis carried out was ANOVA and Tukey-Kramer range test (p < 0.05) [50]. This consisted of two treatments (n = 21 seedlings per treatment): (i) seedlings grown at optimal temperature conditions (28 • C days and 15 • C night) for durum wheat production in the Yaqui Valley and (ii) seedlings grown under conditions of an elevated temperature (+2 • C) compared to the optimal conditions (30 • C days and 17 • C night).
On the other hand, all statistical analyses for RNA-Seq data were carried out according to the software and parameters previously mentioned in Section 4.2. RNA extraction and RNA-Seq analysis.

Conclusions
The main observed response of durum wheat seedlings to the increase in temperature (30 • C vs. 28 • C) was the transcriptional regulation of photosynthesis and ATPase activity; the biosynthesis and remodeling of lipid composition to reinforce the cell wall, water content, and modulate membrane fluidity; and ROS regulation as a signaling and detoxification process induced by thermal stress. These transcription patterns showed physiological signs in the acceleration of phenological progression to the reproductive stage and reprogramming of nutrient mobilization for root development, such patterns suggest a growth-to-escape strategy. These findings complement the state of the art on molecular mechanisms in plants to reduce or even tolerate the impact of climate change.
The arguments presented in this study are based on the observed transcriptomic patterns and constitute the first insights into the mechanisms regulated by the increase in temperature. Therefore, it is necessary to deepen the mechanisms proposed through multi-disciplinary approaches, including qPCR, proteomics, post-transcriptional studies (non-coding RNAs), phenomics, and epigenomics.
Author Contributions: L.A.C.-E., conceptualization, methodology, data curation, formal analysis, visualization, writing-original draft, and review and editing. G.S., conceptualization, methodology, and review and editing. J.J.P.-C., conceptualization, methodology, and review and editing. L.C.-E., conceptualization, methodology and review and editing. F.I.P.-C., conceptualization, methodology, and review and editing. S.d.l.S.-V., conceptualization, methodology, data curation, formal analysis, review, editing, and final approval of the manuscript. All authors have read and agreed to the published version of the manuscript.
Institutional Review Board Statement: Not applicable.

Informed Consent Statement: Not applicable.
Data Availability Statement: The datasets generated during and/or analyzed during the current study are available in the SRA from the NCBI repository under accession number PRJNA780180 (https://www.ncbi.nlm.nih.gov/bioproject/PRJNA780180) (Accessed on 9 December 2021).