Genome-Wide Analysis of the Trehalose-6-Phosphate Synthase ( TPS ) Gene Family and Expression Proﬁling of ScTPS Genes in Sugarcane

: The trehalose-6-phosphate synthase ( TPS ) gene family plays important roles in conferring plant stress resistance, but a comprehensive analysis of the gene family is lacking for sugarcane ( Saccharum spp. hybrids). The objective of this study is to document functional classiﬁcation, evolutionary characterization, and expression proﬁling of sugarcane TPS gene ( ScTPS ) family. Nine putative ScTPS genes were identiﬁed and assigned to two distinct classes based on gene structure and phylogeny. Phylogenetic analysis showed that 31 TPS genes from Arabidopsis , rice and sugarcane could be divided into ﬁve distinct clades, suggesting that there were at least ﬁve orthologous groups in monocot and dicot plants. Evolution analysis of TPS genes revealed that TPS family members appeared to have undergone strong negative selection. The strength of the selective pressure di ﬀ ered in most clades, especially Class I TPS genes, experienced signiﬁcantly stronger negative selection pressure than Class II TPS genes. There were also cis -regulatory elements related to phytohormones and abiotic stresses. Additionally, ScTPS genes were found to exhibit divergent expression in response to simulated drought, salinity, and ABA stresses. Since ScTPS genes function in sugarcane adaptation to environmental stimuli, it might be used as a molecular marker in screening sugarcane germplasm for increased stress resistance.


Introduction
Trehalose is a non-reducible disaccharide commonly found in many plant species. It is an energy carrier and an irreplaceable hydrophilic solute that protects cellular proteins and membranes from adverse environmental stresses such as drought, high salinity, and extreme temperatures. A high level of trehalose is thought to help Selaginella lepidophylla (Hook. & Grev.) Spring survive under extreme drought stress [1]. However, in Arabidopsis thaliana (L.) Heynh. and other drought-resistant species, only trace quantities of trehalose are detected despite the presence of numerous gene families that encode enzymes involved in trehalose biosynthesis [2]. Trehalose metabolism is an important target for genetic manipulation to enhance stress tolerance in plants [3].

Screening and Classification of TPS Gene Family Members in Sugarcane
Using the acquired amino acid sequences of TPS proteins in Oryza sativa Indica Group and A. thaliana as templates, TBLASTN was used to blast the sugarcane cultivar SP80-3280 whole-genome shotgun contigs (WGS) database (Accession:PRJNA431722) of NCBI to find all available sequences of sugarcane TPS genes. For BLAST searches, protein sequences of queries were used as inputs in the BLASTp tool, and the resulting hits were filtered by E-value (1e −5 ). Only the longest sequence was retained if several results were found for the same gene. The open reading frames were sought using the Open Reading Frame Finder program. The TPS and TPP domains were further validated by NCBI-CDD [28] and PFAM database [29]. Pseudogenes that overlapped less than 50% of the complete TPS and TPP domains were discarded. The gene structure of all sugarcane TPS gene family members was analyzed using Gene Structure Prediction [30]. BioEdit v7.0 [31] was used to analyze the homology and structural characteristics of all TPS gene family members and to systematically classify and name these TPS gene family members in reference to the rice TPS gene family. All sugarcane TPS protein sequences were analyzed by ExPASy to obtain their basic physical and chemical properties, such as molecular weight (MW), isoelectric point (pI), amino acid composition, instability coefficient, liposolubility index, and total average hydrophilicity. The protein subcellular localization was predicted by CELLO v.2.5 [32].

Motif Analysis of Sugarcane TPS Proteins
The motif structure of sugarcane TPS family members was analyzed by the MEME program (Version 5.1.1), and there were no gaps in the protein sequences from MEME analysis. Letters that appeared in each position constituted a position-specific probability matrix (PSPM), which could be used to judge the possible motifs in the sequence group. In parameter design, the minimum length of a motif sequence was 10, the maximum was 60, and the maximum number of conservative sequence discoveries was six.

Phylogenetic Analysis of TPS Proteins
Multiple amino acid sequence alignment of selected 31 TPS subfamily members from A. thaliana, O. sativa, and sugarcane was performed using ClusterW. MEGA 7.0 was used to construct a phylogenetic tree using Neighbor-Joining method and repeated 1000 times to obtain a bootstrap value [32].

Selection Assessment and Testing
The program PAML version 4.9j was used to test the selection pressure by assessing the values of nonsynonymous substitutions (dN), synonymous substitutions (dS) and dN/dS ratio (ω). [33]. The branch model 2 (two ratios) and branch-site model A was used on CodeML module in PAML [34]. The branch model 2, allowing the foreground branch to evolve under a different rate, was compared to null model (model 0: one ratio) assuming that all branches had been evolving at the same rate. Branch-site model A was used to search for positive selection in codon sites, which allowed individual site and different setups of foreground lineages to be tested depending on the gene substitutions. The likelihood ratio test (LRT) was used to evaluate the fit of the alternative model comparing with the null model. The branch with a p value less than 0.05 and a higher ω value for the foreground branch than the background branch was considered as evolving with a significantly faster rate in the foreground branch. The sites under positive selection were further identified using naive empirical Bayes (NEB) method and Bayes empirical Bayes (BEB) method [35].

Cis-Regulatory Elements Analysis of Sugarcane TPS Gene Families
A 2000-bp upstream sequence of the coding region of each sugarcane TPS gene family was analyzed for the promoter region. The cis-acting elements were analyzed using online data analysis software Plantcare.

Plant Growth and Treatments
As described previously [36], drought stress, salt stress, and ABA treatments were simulated using PEG-6000, NaCl, or ABA, respectively. Sugarcane seedlings of YZ05-51 were grown hydroponically in Hoagland solution containing 10% (w/v) PEG-6000, 200-mM NaCl, or 30-nM abscisic acid, respectively. Control groups were collected right before treatment, and the exponential groups were collected at 6, 12, 24, and 48 h after the stress treatments were imposed. Two seedlings were mixed into one sample, and three biological replicates were adopted at each condition. Samples were frozen in liquid N 2 for further analysis.

Statistical Analysis
Regarding the data distribution and the homogeneity of variance, the Kruskal-Wallis H test was used to analyze the difference of protein sequence identity in the pairwise comparisons of TPS gene members from A. thaliana, O. sativa and sugarcane at p < 0.05 (SPSS 10.0, Inc., Chicago, IL, USA). The Bonferroni method was further used for multiple comparison when there was statistical difference between groups. In the experiment of reverse transcription polymerase chain reaction analysis, the difference between relative gene expressions was analyzed using one-way ANOVA test, LSD-t method was used for further comparison between two groups at p < 0.05 (SPSS 10.0, Inc., Chicago, IL, USA). The data of each time points were independent by the chi-square test.

Screening TPS Genes for Basic Physic-Chemical Properties of ScTPS Coding Proteins
Nine genes of the TPS family were identified from the sugarcane genome database. According to the homology of TPS proteins related to O. sativa, the nine sugarcane TPS genes were named ScTPS1 to ScTPS9 (Table 1). A protein domain analysis showed that the nine ScTPS gene-coded proteins typically contained two domains: TPS (Pfam: Glyco-transf-20) and TPP (Pfam: Trehalose-PPase). The nine ScTPS proteins had as few as 818 (ScTPS5) to as many as 976 (ScTPS1) amino acids. The molecular weights ranged from 91.05 (ScTPS5) to 108.39 (ScTPS1) kD. The isoelectric points ranged from 5.19 (ScTPS2) to 6.63 (ScTPS5) ( Table 1). The ScTPS proteins were hydrophilic (hydrophobicity was <0). Except for ScTPS9, the other eight ScTPS proteins were unstable (instability coefficient > 40). Seven ScTPS proteins were localized in the cytoplasm, ScTPS2 was located in the cell membrane, and ScTPS7 in the nucleus.

Gene Structures and Protein Domains Analysis of ScTPSs
The nine ScTPS genes could be divided into two subfamilies (I and II) based on gene structures and protein homology ( Figure 1). Subfamily I consisted of ScTPS1, with 16 introns. Subfamily II consisted of the other eight ScTPS genes. Except for ScTPS8 with three introns, the other ScTPS genes contained just two introns. Multiple sequence alignment of nine TPS protein sequences showed that the average amino acid sequence identity among nine full-length ScTPS proteins was 53.73%, with the highest identity between ScTPS2 and ScTPS6 (79.16%), and the lowest identity between ScTPS1 and ScTPS3 (32.40%). The average identities of amino acid sequence of TPS and TPP domains were 57.3% and 56.2%, respectively, with 35.30% of the sequences outside the domains. The amino acid identities between Class I (ScTPS1) with Class II (ScTPS2-9) were ranged from 32.40% to 35.27%, while the values were varied from 51.42% to 79.16% within Class II members.

Gene Structures and Protein Domains Analysis of ScTPSs
The nine ScTPS genes could be divided into two subfamilies (I and II) based on gene structures and protein homology ( Figure 1). Subfamily I consisted of ScTPS1, with 16 introns. Subfamily II consisted of the other eight ScTPS genes. Except for ScTPS8 with three introns, the other ScTPS genes contained just two introns. Multiple sequence alignment of nine TPS protein sequences showed that the average amino acid sequence identity among nine full-length ScTPS proteins was 53.73%, with the highest identity between ScTPS2 and ScTPS6 (79.16%), and the lowest identity between ScTPS1 and ScTPS3 (32.40%). The average identities of amino acid sequence of TPS and TPP domains were 57.3% and 56.2%, respectively, with 35.30% of the sequences outside the domains. The amino acid identities between Class I (ScTPS1) with Class II (ScTPS2-9) were ranged from 32.40% to 35.27%, while the values were varied from 51.42% to 79.16% within Class II members.

Evolution Analysis of ScTPS Genes
Pairwise comparisons were analyzed among the 31 full-length TPS protein sequences from Arabidopsis, rice and sugarcane. The result revealed that the protein sequence identities were differed between different groups ( Figure 3). The full-length of 31 TPS proteins share 30.03-89.99% identities, while the average pairwise sequence identities were 67.88%, 58.86% and 34.03% in class I, class II and the sequences between the two TPS classes (Class I-II), respectively. Pairwise comparisons in both Class I and Class II TPS proteins groups were significantly higher than that of Class I-II groups (p < 0.0001). It indicated that there was clearly an evolutionary divergence between class I and class II TPS subfamilies.
A phylogenetic tree of 31 TPS genes was further constructed for the characterization of their evolutionary relationships (Figure 4). These TPS genes were divided into two distinct subfamily with 100% bootstrap support, each contained 25 Class II and 6 Class I TPS genes, respectively. In order to analyze orthologous relations in each subfamily, the 25 Class II TPS genes could be further divided into four groups (A, B, C and D) with high bootstrap support. The number of ScTPS genes, OsTPS genes and AtTPS genes in each of groups were A (5, 5, 3), B (2, 2, 3), C (0, 0, 1), D (2, 2, 0) and E (1, 1, 4), respectively. Clades A, B and E all contained at least one TPS gene of the three species, while only AtTPS gene was present in Clade C and no AtTPS genes belonged to Clade D. The six Class I TPS genes could be further assigned to two groups, the three TPS1 genes (AtTPS1, OsTPS1, and ScTPS1) subclade I and three other Arabidopsis Class I TPS genes (AtTPS2, AtTPS3, and AtTPS4) subclade II. The quantity variance of Class I TPS orthologous may indicate that subclade II TPS genes had been lost from the rice and sugarcane genome. There was a consistent evolutionary relation between sugarcane and rice, and the branches identified in the purple line represented the most recent split between monocot and dicot plants. The nodes that divided into purple lines and black lines were designated as the monocot-dicot common ancestors. Therefore, it was presumed that there were at least five orthologous groups in monocot and dicot plants.

Evolution Analysis of ScTPS Genes
Pairwise comparisons were analyzed among the 31 full-length TPS protein sequences from Arabidopsis, rice and sugarcane. The result revealed that the protein sequence identities were differed between different groups (Figure 3). The full-length of 31 TPS proteins share 30.03-89.99% identities, while the average pairwise sequence identities were 67.88%, 58.86% and 34.03% in class I, class II and the sequences between the two TPS classes (Class I-II), respectively. Pairwise comparisons in both Class I and Class II TPS proteins groups were significantly higher than that of Class I-II groups (p < 0.0001). It indicated that there was clearly an evolutionary divergence between class I and class II TPS subfamilies.
A phylogenetic tree of 31 TPS genes was further constructed for the characterization of their evolutionary relationships (Figure 4). These TPS genes were divided into two distinct subfamily with 100% bootstrap support, each contained 25 Class II and 6 Class I TPS genes, respectively. In order to analyze orthologous relations in each subfamily, the 25 Class II TPS genes could be further divided into four groups (A, B, C and D) with high bootstrap support. The number of ScTPS genes, OsTPS genes and AtTPS genes in each of groups were A (5, 5, 3), B (2, 2, 3), C (0, 0, 1), D (2, 2, 0) and E (1, 1, 4), respectively. Clades A, B and E all contained at least one TPS gene of the three species, while only AtTPS gene was present in Clade C and no AtTPS genes belonged to Clade D. The six Class I TPS genes could be further assigned to two groups, the three TPS1 genes (AtTPS1, OsTPS1, and ScTPS1) subclade I and three other Arabidopsis Class I TPS genes (AtTPS2, AtTPS3, and AtTPS4) subclade II. The quantity variance of Class I TPS orthologous may indicate that subclade II TPS genes had been lost from the rice and sugarcane genome. There was a consistent evolutionary relation between sugarcane and rice, and the branches identified in the purple line represented the most recent split between monocot and dicot plants. The nodes that divided into purple lines and black lines were designated as the monocot-dicot common ancestors. Therefore, it was presumed that there were at least five orthologous groups in monocot and dicot plants.  Parameter estimates and likelihood ratio tests for the branch models were performed to evaluate the selective pressure on the five clades of TPS genes. The ω value of the one-ratio model (M0) was 0.10310, indicating a strong purifying selection pressure experienced by the TPS gene family ( Table  2). Except for Clade A, there were significant diverse negative pressures on the other four clades by comparing M0 with M2. By comparing the values of background ω0 and foreground ω1, it was found that Clades B, C, D and E had clearly experienced stronger purifying selection. The tests also revealed that Class I TPS genes (all in Clade E) were under a stronger purifying selection than Class II TPS genes. Parameter estimates and likelihood ratio tests for the branch models were performed to evaluate the selective pressure on the five clades of TPS genes. The ω value of the one-ratio model (M0) was 0.10310, indicating a strong purifying selection pressure experienced by the TPS gene family ( Table 2). Except for Clade A, there were significant diverse negative pressures on the other four clades by comparing M0 with M2. By comparing the values of background ω 0 and foreground ω 1 , it was found that Clades B, C, D and E had clearly experienced stronger purifying selection. The tests also revealed that Class I TPS genes (all in Clade E) were under a stronger purifying selection than Class II TPS genes.
The branch-site model, which had the advantage of considering different branches with different selection pressures, was also applied to identify positive selective sites for five clades of TPS genes (Table 3) 32.3% in Clades A, C, D and E, respectively. There are three positive sites identified in both Clades A and C by NEB test at posterior probabilities (p) > 0.95 level. However, only one positive site in Clade A was significant in the BEB test at 0.95 cut-off, which was recommended by PAML [35]. This positively selected site identified in Clade A was 108R according to the references to first sequence of AtTPS1. Only a few positive selection sites were detected, further confirming that TPS genes were mainly affected by negative selection pressure.

Cis-Regulatory Element Analysis of ScTPS Genes
The promoter regions of the ScTPS gene family are rich in both plant hormone and abiotic stress-related cis-regulatory elements, and also contain an ABA response element (ABRE) ( Figure 5). Eight ScTPS genes, except for ScTPS9, contained the methyl jasmonate reaction element (CGTCA-motif Agronomy 2020, 10, 969 9 of 15 and TGACG motif) and salicylic acid response element (as-1). The promoter of four ScTPS genes contain an auxin response element (TGA-motif).
The promoter regions of the TPS gene also contain various cis-acting elements related to environmental and signal molecule responses. Within the promoter region, six ScTPS genes contain light response elements (G-box), seven ScTPS genes contain damage response elements (WRE3), eight ScTPS genes contain stress response elements (STRE), and nine ScTPS genes contain MYB and MYC transcription factor elements related to ABA or drought-induced expression regulation. Thus, the expression of ScTPS genes could be induced by various stresses, which might confer a homeostatic response to maintain normal growth and development under stress. Figure 5. Cis-regulatory elements for the trehalose-6-phosphate synthase (ScTPS) gene family in sugarcane. The length of the bar represents the number of elements from 1 to 9. For reference, bars for cis-regulatory element MYB ScTPS8 and ScTPS7 represent 1 and 9 elements, respectively.
The promoter regions of the TPS gene also contain various cis-acting elements related to environmental and signal molecule responses. Within the promoter region, six ScTPS genes contain light response elements (G-box), seven ScTPS genes contain damage response elements (WRE3), eight ScTPS genes contain stress response elements (STRE), and nine ScTPS genes contain MYB and MYC transcription factor elements related to ABA or drought-induced expression regulation. Thus, the expression of ScTPS genes could be induced by various stresses, which might confer a homeostatic response to maintain normal growth and development under stress.

Analysis of ScTPSs Expression
It was found that ScTPS genes showed differential expression patterns under different treatments ( Figure 6). Under salt treatment, ScTPS1, ScTPS5, ScTPS7 and ScTPS9 exhibited strong expression, with the highest expression level observed at 48 h post-treatment. In contrast, ScTPS4 and ScTPS8 were down-regulated with the lowest expression level observed at 24 h post-treatment. PEG-6000 treatment induced a decline in expression levels of most ScTPS genes, while ScTPS1, ScTPS5 and ScTPS7 exhibited obviously increased expressions at 24h post-treatment. Under ABA treatment, ScTPS2, ScTPS4, ScTPS6 and ScTPS8 were clearly down-regulated at both 24 and 48 h post-treatment, whereas the expresion of ScTPS1 and ScTPS5 were induced in response to ABA. Some genes with a closer phylogenetic relationship frequently showed a similar expression pattern in different treatments, although the response to each stressor varied somewhat. For instance, high salt, simulated drought, and ABA treatments all induced the expression of ScTPS1 and ScTPS5, and inhibited the expression of ScTPS2 and ScTPS6. treatments ( Figure 6). Under salt treatment, ScTPS1, ScTPS5, ScTPS7 and ScTPS9 exhibited strong expression, with the highest expression level observed at 48 h post-treatment. In contrast, ScTPS4 and ScTPS8 were down-regulated with the lowest expression level observed at 24 h post-treatment. PEG-6000 treatment induced a decline in expression levels of most ScTPS genes, while ScTPS1, ScTPS5 and ScTPS7 exhibited obviously increased expressions at 24h post-treatment. Under ABA treatment, ScTPS2, ScTPS4, ScTPS6 and ScTPS8 were clearly down-regulated at both 24 and 48 h post-treatment, whereas the expresion of ScTPS1 and ScTPS5 were induced in response to ABA. Some genes with a closer phylogenetic relationship frequently showed a similar expression pattern in different treatments, although the response to each stressor varied somewhat. For instance, high salt, simulated drought, and ABA treatments all induced the expression of ScTPS1 and ScTPS5, and inhibited the expression of ScTPS2 and ScTPS6.

Discussion
We used bioinformatics methods to study the sugarcane TPS gene family in reference to O. sativa and A. thaliana genomes. In this study, nine TPS genes were identified in sugarcane, only ScTPS1 belonged to Class I. The number of introns and the length of genes differ between Class I and Class II subfamilies. This is basically in accordance with previous studies that identified 11 TPS genes in each of the Arabidopsis and rice genomes, which contain four (AtTPS1, AtTPS2, AtTPS3, and AtTPS4) and one (OsTPS1) Class I TPS genes, respectively [8,26]. The amino acid sequences of ScTPS genes are conservative. The conservation of TPS protein sequences may be driven by the spatial conformation of a long N-terminal TPS domain (460-500 amino acids in size) and a C-terminal TPP domain (about 240 amino acids in size). This hypothesis is well supported in the results that the average identities of amino acid sequence of TPS and TPP domains are both much higher than that of the sequences outside domains. The average amino acid identity is much lower between Class I-Class II pairwise comparisons than that within Class II pairwise comparisons, suggesting a clear evolutionary divergence. It is further confirmed by the phylogeny analysis on 31 TPS genes from Arabidopsis, rice, and sugarcane.
The evolutionary fate of sugarcane TPS genes is basically the same with those of rice. Their difference mainly embodies in the subclade D, which contains three OsTPS gene, but only one ScTPS gene. The distribution of orthologous genes in each subclade differs largely between monocot and dicot plants, suggesting the ancestral origin of the genes as divergent evolution after the monocot-eudicot separation. These results are also noted by several previous reports [2,10,[38][39][40]. It is noteworthy that four Arabidopsis Class I TPS genes are further divided into two subclades, thus one subclade is likely lost from the rice and sugarcane genome. This feature of the dicots is also found in cassava [38]. More detailed TPS phylogentic analysis is needed to confirm the evolutionary feature of TPS gene. The technologies of accurate genome assemblies and annotations are also imperatively demanded in plant genomics.
The evolution analysis of TPS genes reveals that TPS family members appear to have undergone strong negative selection implying functional conservation of these TPSs. The strength of the selective pressure differs in most Clades, especially Clade E (belong to Class I) that experienced significantly stronger negative selection pressure than most Class II TPS genes. It is generally considered that amino acid sequences with important biological functions are less variable across species due to selection pressure [41,42]. Indeed, it was presumed that only TPS1 gene had TPS enzyme activity, until the discovery that both TPS2 and TPS4 of A. thaliana also had TPS enzyme activity [43]. Plant TPS1 gene is closely related to the growth and stress resistance of plants. Arabidopsis TPS1 was proven to be essential for embryogenic and vegetative growth [44]. Trehalose-6-phosphate, mainly synthesized by TPS synthase 1, is considered as a signaling molecule of carbohydrate status and is linked to the regulation of genes expression related to adverse environmental stresses [45,46]. The TPS class II gene has TPS domain, but has no obvious TPS enzyme activity, and its function is largely unknown. Thus it seems that most Class II gene is under relaxed purifying selection, but a positively selected site was identified in Clade A. This codon site may be functionally important, which may be a force directing the evolution of TPS gene in this Clade.
The distinct differences of expression pattern and function between Class I and Class II TPS genes might be related to wide variation in the exon-intron structures. It is found that certain introns are necessary for alternative splicing of mRNA, which could regulate the structure and function of gene-encoded proteins. Some may enhance the transcription and transport of mRNA, leading distinct tissue-specific patterns and diverse expression levels [47,48]. The loss of introns in Class II seems not to be related to selection pressure according to evolution analysis of ScTPS genes from our study. This could be due to gene mutation and other causations. Further research is needed to explore the introns of TPS genes on evolutionary reasons and their effects to TPS protein functions.
Plant TPS genes encode a key enzyme in trehalose metabolism pathway in response to diverse stressful conditions. The promoter region of ScTPS genes contains many cis-acting elements related to plant hormone and environmental stresses. Expression levels of TPS1 and TPS7 genes in S. tuberosum are significantly improved under high temperature, high salinity, and drought stress, suggesting that TPS genes might be involved in the signal transduction pathway of stress resistance [11]. An important feature of drought and salt stress is the osmotic effect on cells, which leads to ABA accumulation as an adaptive response [49]. The relationship between ScTPS and ABA signaling indicated that ScTPS might play an important role in plants counteracting drought stress.
Due to the protective effect of trehalose on abiotic stresses in plants, many researchers have attempted to introduce bacterial or yeast TPS genes to enhance plant stress tolerance, especially to drought. However, the over-expression of heterologous TPS genes is accompanied usually with abnormal phenotypes. It has been reported that A. thaliana and N. tabacum plants that overexpressed yeast TPS gene both exhibit obvious morphological defects [50,51]. One approach of enhancing stress resistance while avoiding undesirable phenotypic changes is through over-expression of endogenous TPS gene. For example, over-expression of AtTPS1 in A. thaliana [12] and OsTPS1 in O. sativa [13] improve their respective drought tolerance without increasing the frequency of aberrant phenotypes. The mechanism of phenotypic alterations caused by the expression of endogenous or heterologous TPS gene remains unclear. Interestingly, Zhang et al. [24] found that transgenic sugarcane that over-expresses Grifola frondosa trehalose synthase gene has improved drought tolerance without obvious morphological changes and growth inhibition in the field. Recently, Nilson et al. have acquired two sugarcane clones with STPS1 and STPS2 EST sequences that are identical to the ScTPS1 and ScTPS6 genes in this study, respectively [52]. Their results showed that STPS1 expression was up-regulated in drought-tolerant cultivars under water stress, while the expression of STPS2 gene had no significant change under same treatments. In this study, it also found that ScTPS1 gene expression was induced by salt, simulated drought, and ABA, suggesting that sugarcane plants increased trehalose-6-phosphate production to maintain stable intracellular osmotic pressure to mitigate simulated stress. Therefore, it is possible that genetic engineering of endogenous ScTPS1 genes involved in trehalose biosynthesis might confer increased drought tolerance to sugarcane. More research is needed to better understand the molecular regulatory network of ScTPS in sugarcane. Nonetheless, ScTPS marker-assisted selection of stress tolerant sugarcane genotypes might help sustain sugarcane production under stressful field conditions.