The XTH Gene Family in Cassava: Genomic Characterization, Evolutionary Dynamics, and Functional Roles in Abiotic Stress and Hormonal Response

Abstract

Xyloglucan endotransglucosylases/hydrolases (XTHs) are key enzymes involved in cell wall remodeling that play roles in plant responses to environmental stress. Despite their importance, a comprehensive investigation of the XTH gene family in cassava (Manihot esculenta Crantz), a crucial drought-tolerant crop in tropical and subtropical regions, has not yet been conducted. In the present study, we identified 37 XTH genes (MeXTH1-37) within the cassava genome, and most of them contain two conserved structures (Glyco_hydro_16 and XET_C domain). Phylogenetic analysis grouped 37 MeXTH genes into three distinct clades, a classification further supported by exon–intron organizations and the conserved protein motif architectures. Duplication events, particularly segmental duplication, were identified as the main driving force for MeXTH gene expansion in cassava. Comparative synteny analysis revealed orthologous relationships between MeXTH genes and XTH-related genes in seven other plant species, including soybean, poplar, tomato, Arabidopsis, maize, wheat, and rice. Global expression analysis revealed that MeXTH genes display different expression patterns in various cassava tissues, shedding light on their potential biological functions. Furthermore, quantitative real-time PCR (qRT-PCR) analysis of 12 representative MeXTH genes under salt and osmotic stress, as well as salicylic acid (SA) and methyl jasmonate (MeJA) treatments, demonstrated their differential responses to these stimuli. These results provide novel insights into the role of the MeXTH gene family in enhancing cassava’s tolerance to abiotic stress.

1. Introduction

The cell wall, a defining feature of plant cells, is characterized by a complex matrix of polymers that plays an essential role in growth and development. It not only influences cell morphology and size but also serves as a barrier against environmental stresses [1,2]. The primary cell wall is primarily composed of cellulose, hemicellulose, pectin, and glycoproteins [3,4]. Among these, xyloglucan is a major hemicellulosic component that interacts with cellulose microfibrils and links the main structural network to the pectin matrix, contributing significantly to wall integrity and flexibility [3,4,5]. Xyloglucan endotransglucosylases/hydrolases (XET/XEH), also known as XTHs, are enzymes that modify xyloglucans. The xyloglucan backbones within plant cell walls are responsible for cell wall synthesis and reconstruction [6,7,8]. XTHs belong to the glycoside hydrolase 16 (GH16) family, which includes enzymes that act on various substrates such as keratan sulfate, β-1,3-glucans, mixed-linkage β-1,3(4)-glucans, xyloglucans, j-carrageenan, and agarose [8,9,10]. The XTH family possesses two catalytic functions: xyloglucan endotransglucosylase (XET) activity, which enables the restructuring of xyloglucan chains, and xyloglucan endohydrolase (XEH) activity, which breaks glycosidic bonds to promote cell wall expansion and morphogenesis [11,12].

The first XTH gene was identified in cowpea (Vigna unguiculata) in 1992, and it was found to play a crucial role in reconstructing the cell wall matrix by reconnecting xyloglucan, contributing to morphological changes through chemical creep [13]. With the advancement of plant genome sequencing, the XTH gene family has been extensively annotated across multiple plant species [14,15]. For instance, Arabidopsis thaliana contains 33 XTH genes [14], rice (Oryza sativa) has 29 [16], wheat (Triticum aestivum) has over 57 [17], sorghum (Sorghum bicolor) has 35 [18], tobacco (Nicotiana tabacum) has 56 [19], soybean (Glycine max) has 61 [20], tomato (Solanum lycopersicum) has 25 [21,22], and poplar (Populus trichocarpa) has 38-43 [2,3,23]. Most XTH proteins contain two conserved domains, Glyco_hydro_16 and XET_C, which facilitate xyloglucan network remodeling and play a critical role in regulating cell wall flexibility and expansion throughout plant growth and stress responses. Initial classification of the A. thaliana XTH family divided it into three major groups: I, II, and III [14]. Subsequent analyses in rice reclassified these genes into two main subgroups, I/II and III, due to a lack of clear distinction between groups I and II [16]. Additionally, a small outlier group near the root of the phylogenetic tree has been described as an ancestral lineage [24,25]. Group III was further subdivided into subgroups III-A and III-B based on three-dimensional structural differences [12,24]. Thus, the XTH gene family in plants is typically classified into I/II, IIIA, IIIB, and an ancestral group. Notably, members in group I/II are predominantly associated with transglycosylase activity, while those in group III exhibit more hydrolytic activity [11,25,26].

Genetic evidence suggests that XTH family genes play diverse roles in plant growth and development by modeling cell wall properties. They are involved in root development [14,27], fruit development [22,28,29], fiber elongation [1], hypocotyl growth [30], and flower opening [31]. Importantly, XTH genes have also been implicated in responses to abiotic stress. For example, in A. thaliana, an AtXTH31 mutant exhibited reduced xyloglucan content, thereby decreasing aluminum uptake and enhancing aluminum tolerance [32]. Similarly, the xth17 and xth15 mutants displayed increased resistance to aluminum stress [33,34]. AtXTH19 and AtXTH23 regulate lateral root development and adaptation to salt stress via the BES1-dependent pathway [35]. In addition, xth19 mutants were more susceptible to freezing following cold and sub-zero acclimation due to changes in cell wall structure [36]. Homologues such as CaXTH1, CaXTH2, and CaXTH3 in hot pepper (Capsicum annuum) are associated with enhanced tolerance to drought, salinity, and low temperature [37]. Overexpression of DkXTH1 from Diospyros kaki improved tolerance to salt, drought, and ABA (abscisic acid) in transgenic A. thaliana [38]. Overexpression of CaXTH3 in pepper also conferred improved drought and salt resistance in transgenic A. thaliana plants, albeit accompanied by severe leaf folding [39]. Moreover, XTH gene expression is modulated by various hormones, such as gibberellic acid (GA₃), brassinolide (BR), jasmonic acid (JA), and ABA, which jointly regulate growth, defense, and cell wall dynamics [38,40,41,42,43].

Cassava (Manihot esculenta Crantz) is a major tropical crop and a dietary staple for over 700 million people worldwide [44]. Cultivated primarily for its starchy roots, cassava plays an essential role in food security and is widely used in both the food industry and bio-based applications [45,46]. Its foliage boasts a high protein content (ranging from 16.41% to 22.68%), along with essential minerals and gross energy [47,48]. In many countries, cassava foliage is utilized as animal feed, showing promise for enhancing livestock production [49,50]. Notably, cassava foliage harvests exhibit seasonality, with greater biomass availability during the summer or rainy seasons [51]. Recognized by the FAO as Africa’s most crucial root crop, it holds special importance in underdeveloped countries [47]. Like other staple crops, cassava frequently encounters environmental challenges such as drought, salt, and pathogen attacks. Understanding the molecular basis of stress responses in cassava is crucial for breeding stress-resilient varieties. The availability of its complete genome offers a valuable resource for exploring gene families involved in these responses, including XTHs [52]. Thus, this study conducted a genome-wide identification and systematic analysis of the xyloglucan endotransglucosylase/hydrolase (XTH) gene family in cassava (MeXTH), investigating their phylogenetic relationships, conserved motifs, gene structures, chromosomal distributions, duplication events, cis-regulatory elements, and tissue-specific expression patterns. To evaluate the potential role of these genes in stress resilience, which represents a critical objective for breeding programs, we selected two major abiotic stresses threatening cassava productivity: drought (simulated by osmotic stress/PEG) and salinity (NaCl). Additionally, treatments with salicylic acid (SA) and methyl jasmonate (MeJA) were employed to probe hormone-responsive expression patterns, given their established roles in stress signaling and defense regulation. The expression dynamics of selected MeXTH genes under these conditions were profiled to pinpoint robust stress-responsive candidates. The results offer valuable evolutionary and functional insights into the MeXTH family, promising genetic targets for improving cassava’s tolerance to adverse environmental conditions through molecular breeding.

2. Materials and Methods

2.1. Genome-Wide Identification of Cassava XTH Genes

The genome sequences and General Feature Format (GFF) files for cassava (Mesculenta_305_v6.1) were retrieved from Phytozome v13 (https://phytozome-next.jgi.doe.gov/, access date: 20 October 2023), while protein sequences for the A. thaliana XTH (AtXTH) family were sourced from the TAIR database (http://www.arabidopsis.org/, access date: 20 October 2023) [53]. To identify the target proteins, two complementary strategies were employed. Initially, HMM profiles PF00722 (Glyco_hydro_16) and PF06955 (XET_C) were obtained from Pfam (http://pfam.xfam.org/, access date: 8 November 2023) and utilized to query the cassava genome using HMMER v3.0 [54]. Additionally, a two-way BLAST alignment was performed using the Two Sequence Files tool in TBtools-II.v2.119, aligning AtXTH protein sequences with cassava genome sequences to identify potential members of the XTH family [55]. The conserved XTH domains Glyco_hydro_16 and XET_C were further validated using SMART (http://smart.embl-heidelberg.de/, access date: 10 November 2023) [2]. Proteins containing these domains were subsequently verified through the Conserved Domain Database (CDD) (https://www.ncbi.nlm.nih.gov/cdd/, access date: 10 November 2023) [56] and InterProScan (https://www.ebi.ac.uk/interpro/, access date: 23 November 2023) [57].

2.2. Phylogenetic Analysis

A multiple sequence alignment of the cassava, rice, A. thaliana, and poplar XTH protein sequences (Table S1) and the reference data was conducted using ClustalW with default parameters in MEGA-X [2]. A phylogenetic tree was then constructed using the Maximum-Likelihood (ML) method with a bootstrap value of 1000. The phylogenetic tree was further refined using Evolview v3.0 online software (https://www.evolgenius.info/evolview-v3/#login, access date: 6 June 2025) [58].

2.3. Conserved Motif and Gene Structures Analyses

Conserved motifs in the MeXTH proteins were identified via MEME Suite (http://meme-suite.org/tools/meme, access date: 6 June 2025), with the maximum number of motifs set to 10 [59], and all other parameters were default settings. Intron–exon structures were examined using the Amazing Optional Gene Viewer feature in TBtools-II.v2.119 [55].

2.4. Sequence Analysis

The fundamental physicochemical properties of the identified MeXTH proteins, such as amino acid count, molecular weight, protein length, pI, GRAVY, and instability index, were predicted using the ProtParam tool on the ExPASy server (https://web.expasy.org/protparam/, access date: 17 November 2023) [60]. Additionally, subcellular localization predictions for the MeXTH proteins were made with WoLF PSORT (https://wolfpsort.hgc.jp/, access date: 17 November 2023) [61]. Secondary structural features of MeXTH proteins were analyzed via the ESPript server (https://espript.ibcp.fr/ESPript/ESPript/, access date: 23 November 2023) [62]. Protein sequences were aligned with reference structures TmNXG1 (PDB ID: 2UWA) [10] and PttXET16-34 (PDB ID: 1UN1) [5] to visualize conserved structural domains. SWISS MODEL software (https://swissmodel.expasy.org/, access date: 8 June 2025) was used to predict the tertiary structure.

2.5. Chromosomal Location and Collinearity Analysis

Chromosomal location data for the MeXTH genes were obtained from the cassava GFF annotation files. TBtools-II.v2.119 software was used for chromosome localization and gene structure analysis maps [55]. Gene duplication events were examined using MCScanX with default settings [63]. Synteny analyses within cassava and across multiple species, including A. thaliana, poplar, soybean, tomato, wheat, maize, and rice, were performed using dual systemic plots and Advanced Circos in TBtools-II.v2.119 [55]. The Ka Ks substitution rates and their Ka/Ks ratios for MeXTH gene pairs were calculated using the Simple Ka/Ks Calculator tool in TBtools-II.v2.119 [55].

2.6. Prediction of Cis-Acting Elements in the Promoter Regions of MeXTHs

Genomic DNA fragments, spanning 2.0 kb upstream of the ATG start codon, were extracted from cassava genome sequences using TBtools-II.v2.119 software based on the GFF file [55]. The PlantCARE online search tool (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/, access date: 26 November 2023) was used to predict the cis-elements that may be involved in the regulation of MeXTH gene expression in cassava. The identified cis-acting elements were categorized according to the classification system outlined by Liu et al. [64]. Visualization of the results was performed using the HeatMap tool in TBtools-II.v2.119 software [55].

2.7. Protein–Protein Interaction (PPI) Network Construction and Gene Ontology

To predict the protein–protein interaction network and assess the functional relationships among cassava XTH proteins, A. thaliana interaction data were utilized. The protein interactions of MeXTH were predicted using the STRING database (http://string-db.org/, access date: 9 February 2024), with a confidence threshold set at 0.15 [65]. GO enrichment analysis, encompassing MF, BP, and CC, was performed using the Pannzer2 online tool (http://ekhidna2.biocenter.helsinki.fi/sanspanz/, access date: 9 February 2024) to investigate the functional roles of the 37 MeXTH proteins. All amino acid sequences were submitted in FASTA format.

2.8. Gene Expression Analysis Based on Transcriptome Data

Transcriptome data were obtained from the NCBI database (accession numbers: PRJNA324539, GEO dataset: GSE82279), as reported by Wilson et al. [66]. These datasets were employed to analyze the expression profiles of MeXTH genes across a range of tissues, including leaves, midveins, lateral buds, OES, FEC, FR, SR, stems, petioles, RAM, and SAM. Differential expression analysis was performed using Cuffdiff v.2.2.1 to identify significantly expressed genes, with a cut-off value of |log₂(fold change)| > 2. Gene expression levels were quantified by calculating fragments per kilobase of transcript per million mapped reads (FPKM). Log2-transformed heatmaps of all MeXTH genes were generated using TBtools-II.v2.119 software [55].

2.9. Plant Material, Growth Conditions, and Treatments

The cassava cultivar SC9 (South China 9) was sourced from the National Cassava Germplasm Nursery in Danzhou, China. Stem cuttings approximately 15 cm in length, each containing two to three buds, were cultured for 30 days in 1/2-strength Hoagland nutrient solution at 25 °C under a 16 h light/8 h dark cycle at Hainan University (Haikou, China). Uniform seedlings were selected for the analysis of MeXTH gene transcriptional responses under various abiotic stress and hormone treatments. The seedlings were exposed to 1/2-strength Hoagland nutrient solution supplemented with 30% polyethylene glycol (PEG) 6000, 400 mM NaCl, 100 µM SA, and 100 µM MeJA for 4, 12, and 24 h. The seedlings grown in normal 1/2-strength Hoagland solution without any treatment were defined as the control group (representing the 0 h time point). At each predetermined time point, fully expanded leaves from the middle section of the seedlings were collected. Prior to RNA extraction, leaf samples were collected, immediately frozen in liquid nitrogen, and stored at −80 °C.

2.10. RNA Extraction and Quantitative Real-Time PCR (qRT-PCR)

Total RNA was extracted from the collected samples using the RNAprep Pure Plant Plus Kit (TIANGEN Biotech Co., Ltd., Beijing, China), following the manufacturer’s protocol. Approximately 1 µg of total RNA was used for complementary DNA (cDNA) synthesis using a reverse transcriptase kit (TIANGEN Biotech Co., Ltd., Beijing, China). The first strand of cDNA was synthesized using the M1631 reverse transcriptase kit (Thermo Fisher Scientific, Waltham, MA, USA). qRT-PCR was performed using a 7500 Real-Time PCR System (Thermo Fisher Scientific) in a total reaction volume of 10 µL, consisting of 1 µL cDNA template (100 ng/µL), 0.5 µL forward primer (10 µmol/L), 0.5 µL reverse primer (10 µmol/L), 5 µL qPCR Master Mix (TB GREEN PREMIX EX TAQ II, Takara, Shiga, Japan), and 3 µL sterilized ddH₂O. The PCR amplification protocol followed that outlined by Cao et al. [67]. The elongation factor 1α (EF1α) gene was used as the internal control. The relative expression levels of MeXTH genes were calculated using the 2^−ΔΔCT method as described by Schmittgen and Livak [68]. Each experiment included three biological replicates, with three technical replicates for each. The primer sequences used in this study were designed using Primer3 (https://primer3.org/, access date: 18 February 2024) and are listed in Table S2.

2.11. Statistical Analyses

Statistical analyses were performed using IBM SPSS v26 Statistics software. Data are presented as mean ± standard deviation (SD). The data were analyzed with one-way analysis of variance (ANOVA). Group differences were assessed using Duncan’s multiple range test, with a p-value of ≤ 0.05 considered statistically significant.

3. Results

3.1. In-Silico XTH Proteins Discovery in Cassava

Through HMM (Hidden Markov Model) searches utilizing XTH protein domains (PF00722 and PF06955) and BLAST alignment with 33 A. thaliana XTH (AtXTH) proteins, a total of 37 MeXTH proteins were identified in the cassava genome and named based on its chromosome location (

Table 1. Basic information on the xyloglucan endotransglucosylase/hydrolase (XTH) family in cassava.

Gene ID	Gene Name	Chr.	Protein Length (a.a.)	MW (KDa)	pI	Instability Index	Hydrophobicity Index	Sublocation (WoLF)
Manes.01G241900.1.p	MeXTH1	Chr01	286	32.81	5.75	44.19	−0.41	chlo: 10
Manes.01G262300.1.p	MeXTH2	Chr01	213	24.10	4.93	44.94	−0.46	cyto: 6
Manes.03G024400.1.p	MeXTH3	Chr03	294	33.60	8.53	42.45	−0.35	vacu: 7
Manes.03G079700.1.p	MeXTH4	Chr03	287	32.55	9.04	43.33	−0.27	chlo: 4
Manes.03G146800.1.p	MeXTH5	Chr03	290	33.14	7.14	47.38	−0.52	extr: 6
Manes.04G013900.1.p	MeXTH6	Chr04	290	33.00	5.67	43.10	−0.60	vacu: 7
Manes.04G049900.1.p	MeXTH7	Chr04	279	31.32	4.55	34.81	−0.24	extr: 5
Manes.04G096600.1.p	MeXTH8	Chr04	293	34.04	8.45	45.54	−0.47	chlo: 5
Manes.05G108100.1.p	MeXTH9	Chr05	287	32.18	6.13	29.33	−0.27	extr: 5
Manes.05G199600.1.p	MeXTH10	Chr05	293	33.09	8.65	42.41	−0.29	pero: 9
Manes.07G051100.1.p	MeXTH11	Chr07	294	34.46	8.59	36.91	−0.40	plas: 3.5
Manes.08G011900.1.p	MeXTH12	Chr08	294	34.07	9.46	49.73	−0.42	cyto: 5
Manes.08G099000.1.p	MeXTH13	Chr08	293	34.10	8.47	40.76	−0.45	chlo: 9
Manes.09G064800.1.p	MeXTH14	Chr09	294	33.99	9.42	45.88	−0.45	chlo: 11
Manes.11G072700.1.p	MeXTH15	Chr11	293	34.01	6.16	40.49	−0.40	chlo: 5
Manes.11G115200.1.p	MeXTH16	Chr11	284	32.09	8.24	23.06	−0.31	extr: 4
Manes.11G151500.1.p	MeXTH17	Chr11	291	33.14	7.66	39.38	−0.58	chlo: 7
Manes.12G030000.1.p	MeXTH18	Chr12	315	35.41	6.4	56.91	−0.21	vacu: 5
Manes.12G076800.1.p	MeXTH19	Chr12	288	32.48	5.48	36.66	−0.30	extr: 5
Manes.13G046500.1.p	MeXTH20	Chr13	287	32.76	5.59	32.50	−0.32	E.R.: 3.5
Manes.13G107100.1.p	MeXTH21	Chr13	292	33.45	6.38	51.52	−0.29	vacu: 5
Manes.14G114400.1.p	MeXTH22	Chr14	336	38.14	6.99	31.63	−0.21	chlo: 7
Manes.14G114500.1.p	MeXTH23	Chr14	293	33.55	5.69	33.53	−0.33	vacu: 7
Manes.14G145700.1.p	MeXTH24	Chr14	280	31.71	5.22	33.97	−0.38	extr: 4
Manes.14G145800.1.p	MeXTH25	Chr14	280	31.71	5.22	33.97	−0.38	extr: 4
Manes.14G146000.1.p	MeXTH26	Chr14	285	32.30	4.95	33.74	−0.33	extr: 7
Manes.14G146100.1.p	MeXTH27	Chr14	285	32.28	4.94	36.14	−0.34	extr: 6
Manes.14G152100.1.p	MeXTH28	Chr14	283	31.94	8.12	31.74	−0.42	vacu: 4
Manes.14G152200.1.p	MeXTH29	Chr14	286	32.48	6.89	37.17	−0.39	extr: 7
Manes.15G137800.1.p	MeXTH30	Chr15	333	37.98	5.94	49.03	−0.39	chlo: 4
Manes.15G192300.1.p	MeXTH31	Chr15	342	39.69	9.04	47.83	−0.47	cyto: 5
Manes.16G011700.1.p	MeXTH32	Chr16	297	34.73	5	39.38	−0.57	extr: 6
Manes.17G015100.1.p	MeXTH33	Chr17	348	40.10	9.17	46.91	−0.45	chlo: 5
Manes.17G063600.1.p	MeXTH34	Chr17	295	33.82	8.72	36.6	−0.37	extr: 4
Manes.17G087100.1.p	MeXTH35	Chr17	332	37.74	7.14	50.22	−0.39	chlo: 5
Manes.S055900.1.p	MeXTH36	Scaffold 1044	266	30.20	5.07	33.86	−0.52	nucl: 12
Manes.S061400.1.p	MeXTH37	Scaffold 1087	288	33.20	7.63	39.59	−0.37	chlo: 8

Chr.: chromosomal; MW: molecular weight; pI: predicted isoelectric point; chlo: chloroplast; extr: extracellular; cyto: cytoskeleton; E.R.: endoplasmic reticulum; vacu: vacuole; nucl: nucleolus; plas: plasmodesmata; pero: peroxisome.

and Table S1). The Pfam and CDD databases were employed to investigate the conserved domains of the 37 MeXTH proteins. From the results obtained, one protein, MeXTH2, possessed only the Glyco_hydro_16 domain, and the remaining 36 MeXTH proteins contained both Glyco_hydro_16 and XET_C conserved domains (Figure S1). Multiple sequence alignment revealed that all 37 MeXTH proteins harbor a typical highly conserved Glyco_hydro_16 domain, characterized by a 41-residue consensus motif (REQQFYLWFDPTADFHTYSILWNPQQIIFFVDGVPIREFKN) visualized through sequence logo analysis (Figure S1). In addition, 36 MeXTH proteins also contained the XET_C domain, represented by a 25-residue conserved motif (RRLQWVRKNYMIYBYCTDTKRFPQG) (Figure S1).

The characteristics of the MeXTH proteins were further analyzed using the online tool ExPasy. As summarized in Table 1, the deduced MeXTH proteins vary in length, ranging from 213 amino acids (MeXTH2) to 348 amino acids (MeXTH33). Their corresponding molecular weights (MW) ranged from 24.1 kDa to 40.1 kDa, while the predicted isoelectric points (pIs) ranged from 4.55 (MeXTH7) to 9.46 (MeXTH12). All MeXTH proteins exhibited negative grand average hydropathy (GRAVY) values, indicating their hydrophilic nature. Subcellular localization predictions, performed using the WoLF PSORT program, revealed that 12 MeXTH proteins are localized in the chloroplast, 12 in the extracellular space, 3 in the cytoskeleton, 1 in the endoplasmic reticulum, 6 in the vacuole, 1 in the nucleolus, 1 in the plasmodesmata, and 1 in the peroxisome.

3.2. Phylogenetic Tree Construction of MeXTH Proteins

To better understand the evolutionary relationship of these identified MeXTHs, a phylogenetic tree was constructed to analyze the evolutionary relationships among 37 MeXTHs from cassava, 33 AtXTHs from A. thaliana, 29 OsXTHs from rice, and 38 PtrXTHs from poplar (Table S1; Figure 1). The results revealed that the XTH proteins in both species could be classified into four distinct groups, namely Group I/II, Group IIIA, Group IIIB, and the Ancestral group (Figure 1), which is largely consistent with the phylogenetic groupings previously reported in other plant species [2,3,14]. Group I/II comprised 24 MeXTHs, along with 22 AtXTHs, 18 OsXTHs, and 28 PtrXTHs. Group IIIA included 5 MeXTHs, 3 PtrXTHs, 4 OsXTHs, and 2 AtXTHs, while Group IIIB contained 6 MeXTHs, 6 PtrXTHs, 7 OsXTHs, and 5 AtXTHs. The Ancestral group had the fewest XTH proteins, including 2 MeXTHs, 1 PtrXTHs, and 4 AtXTHs. Baumann et al. [24] identified a small outlying ancestral group near the root of the phylogenetic tree, comprising AtXTH1 to AtXTH3 and AtXTH11 from Arabidopsis thaliana, which is consistent with our findings. Additionally, a phylogenetic tree constructed using only MeXTH proteins demonstrated that genes within each group clustered together (Figure 2A). This tree structure was consistent with that observed in the phylogenetic tree based on XTH protein sequences from the four plant species (Figure 1). The results of the phylogenetic analysis suggest a potential association between the subfamily classification of MeXTH genes and their functional similarities. For example, MeXTH2, MeXTH6, MeXTH12, MeXTH14, and MeXTH17 belonged to Group III-A, which is predicted to be associated with XEH activity. In contrast, MeXTH1, MeXTH18, MeXTH30, MeXTH31, MeXTH33, and MeXTH35 were grouped into Group III-B, a group putatively defined by XET activity [11,69].

3.3. Conserved Motifs and Gene Structure of MeXTHs

To better understand the structural features of MeXTHs, conserved motifs and intron/exon were analyzed according to their phylogenetic relationships (Figure 2). Conserved motif analysis was performed using MEME online software in combination with the motif visualization function of TBtools-II.v2.119. The number of motifs in the MeXTH family members ranged from 4 to 9. Motif 2 was the longest, containing 44 amino acids, while the shortest motif, motif 9, comprised only 6 amino acids. Motif 4, which contains the catalytic active site DEIDFEFLG (with glutamic acid as the catalytic residue) (Figure S2), was consistently found across all subgroups (I/II, IIIA, IIIB, and the Early diverging group). In addition, all MeXTH members contained motifs 1, 2, 3, and 4 (Figure 2B). Among the 37 MeXTHs, MeXTH2, which belong to Group III-A, exhibited the fewest motifs, starting with motif 3 and ending with motif 2. The reduced motif composition in MeXTH2, compared to other XTHs, may imply a distinct functional specialization, such as a narrowed range of substrate recognition or altered enzymatic efficiency. Additionally, motif 10 and motif 8 were absent from subfamilies IIIA and IIIB but were present in subfamily I/II, suggesting their evolutionary specificity. The exon–intron structures of the MeXTH genes were analyzed based on their genomic DNA sequences and corresponding coding sequences. All MeXTH genes contained 2 to 4 exons, except for MeXTH1, which consisted of only one exon (Figure 2C).

3.4. Secondary and Tertiary Structures of MeXTH Proteins

The predicted secondary structures of the MeXTH proteins indicated that approximately 50% of their structures were composed of extended β-strands, followed by random coils accounting for about 30%, while β-turns represented the smallest proportion (Table S3). These predictions were made based on homologous structures of the well-characterized endoxyloglucan endoglycosidase PttXET16-34 (PDB ID: 1UN1) and endoxyloglucan enzyme TmNXG1 (PDB ID: 2UWA), with conserved domains illustrated in Figure S3. Tertiary structure modeling was conducted using the SWISS-MODEL program, applying homology-based modeling techniques to infer the 3D structures of the MeXTH proteins from homologous proteins with known crystal structures (Figure S4). All 37 modeled MeXTH proteins displayed a characteristic β-sandwich fold, consisting of large β-sheets packed in parallel layers—typical of glycoside hydrolase family 16. Interestingly, variations in the catalytic site residues were observed in four MeXTH members (MeXTH8, MeXTH13, MeXTH15, and MeXTH21), which led to subtle structural differences in their active sites. These differences are likely to contribute to variations in substrate specificity or enzymatic activity, suggesting functional diversification among members of the MeXTH family.

3.5. Chromosomal Locations, Duplication Events, and Collinearity Analysis

All of the MeXTH genes were unevenly distributed across 16 cassava chromosomes, with the exception of MeXTH36 and MeXTH37, which were located on unanchored scaffolds. Chr14 harbored the highest number of MeXTH genes (n = 8; 21.6%), followed by Chr3, Chr4, Chr11, and Chr17, each containing three genes (8.1%) (Figure 3A). To explore the evolutionary expansion of these genes, a collinearity analysis was performed using MCScan. This analysis identified 10 segmental duplication events involving 11 MeXTH genes (MeXTH6, MeXTH7, MeXTH8, MeXTH12, MeXTH13, MeXTH14, MeXTH15, MeXTH16, MeXTH17, MeXTH32, and MeXTH36) (Figure 3B, Table S4), suggesting that segmental duplication has played a major role in the expansion of the MeXTH family in cassava. If two paralogous genes are physically close together (less than 100 KB), we might suggest that they arose through tandem duplication [70]. It was found that three gene groups (MeXTH22 and MeXTH23, MeXTH26 and MeXTH27, and MeXTH28 and MeXTH29) were tandemly distributed on Chr14. Furthermore, all duplication pairs exhibited synonymous (Ka)/non-synonymous (Ks) ratios less than 1, indicating that these genes have evolved under strong purifying selection (Table S4). To further examine the evolutionary conservation of the MeXTH genes across species, comparative syntenic maps were created by comparing cassava to seven representative species, including four dicots (A. thaliana, poplar, tomato, and soybean) and three monocots (rice, wheat, and maize). A substantial number of syntenic blocks were identified in these comparisons. Specifically, 22, 39, 23, 60, 9, 9, and 4 cassava orthologous gene pairs were found in A. thaliana, poplar, tomato, soybean, rice, wheat, and maize, respectively (Figure 3C, Table S5).

3.6. Analysis of Cis-Regulatory Elements in MeXTH Genes

To investigate the involvement of MeXTH genes in the regulation of abiotic stresses, the 2000 bp upstream promoter sequences of these genes were analyzed for cis-regulatory elements using the PlantCARE online tool. As shown in Figure 4, the MeXTH genes contained between 13 and 48 cis-regulatory elements, which were classified into five functional categories. The promoter regions of the MeXTH genes were rich in hormone response elements, including those responsive to abscisic acid (ABRE), methyl jasmonate (CGTCA-motif, TGACG-motif), gibberellin (GARE-motif, P-box, TATC-box), salicylic acid (TCA-element), and auxin (AuxRR-core, TGA-element). This distribution supports the role of XTH transcription factors in mediating responses to abiotic stresses. Additionally, the promoter regions contained various stress-responsive elements, such as those linked to heat stress (Box II-like sequence, CCAAT-box, AT-rich), cold stress (LTR), anaerobic response (ARE), and damage and defense responses (WUN-motif, W box). Some MeXTH genes also exhibited regulatory elements related to growth and development, including elements associated with endosperm expression (GCN4), meristem expression (CAT-box), seed expression (RY-element), differentiation (HD-Zip1, HD-Zip3), zein metabolism (O2-site), and circadian regulation (circadian, MSA-like). Among the light-responsive elements, Box 4 was the most abundant, present in 36 MeXTH genes, excluding MeXTH2. Notably, MeXTH17 contained the highest number of Box 4 elements, with 18 occurrences. The occurrence of different regulatory elements varies across genes. For instance, ARE was found in most genes, albeit at different frequencies, while circadian-related elements were detected in only a few MeXTH genes. MeXTH19 contained the largest number of cis-regulatory elements overall. In summary, the identified cis-regulatory elements suggest that MeXTH genes are likely involved in responses to abiotic stresses.

3.7. Analysis of Cis-Regulatory Elements in MeXTH Proteins

Understanding the functional interactions among proteins through protein–protein interaction (PPI) networks can offer insights into the diverse biological activities and dynamic regulatory networks of biomolecules. In this study, the potential functional and physical interactions among MeXTH proteins were examined using the STRING program, based on the A. thaliana association model. In the PPI network, each node represents a protein generated by a single protein-coding locus. As shown in Figure 5, the analysis identified 14 functional XTH proteins (XTH5, XTH6, XTH7, XTH8, XTH15, XTH21, XTH24, XTH23, XTH26, XTH27, XTH28, XTH31, XTH32, and XTH33) and 10 putative interaction proteins (PME22, GATL7, T19K4.70, ENY2, VPS26B, VPS26A, T4O12.10, EXPB1, T20H2.11, and F12F1.21), all directly connected to MeXTH proteins based on the statistical analysis of 33 AtXTH proteins.

3.8. GO Enrichment Analysis of MeXTH

To further characterize the functions of MeXTH genes, Gene Ontology (GO) annotation analysis was conducted using the PANNZER2 online tool. The predicted functions were classified into three categories: biological process (BP), molecular function (MF), and cellular component (CC) (Figure 6 and Table S6). In the BP category, 32 of the 37 MeXTH genes were annotated with involvement in the xyloglucan metabolic process (GO:0010411). Additional enriched processes included cell wall biogenesis (GO:0042546), cell wall organization (GO:0030154), cell wall biogenesis (GO:0042546), fruit ripening (GO:0009835), and carbohydrate metabolic processes (GO:0005975). In the CC category, most MeXTH genes were associated with the membrane (GO:0016020), apoplast (GO:0048046), ribonucleoprotein complex (GO:1990904), ribosome (GO:0005840), cell periphery (GO:0071944), cytoplasm (GO:005737), and organelle (GO:0043226). In the MF category, 32 out of the 37 MeXTH genes exhibited xyloglucan: xyloglucosyl transferase activity (GO:0016762), while all 37 MeXTH genes displayed hydrolase activity specific to O-glycosyl compounds (GO:0004553), which are characteristic features of transcription factors. Some MeXTH genes (MeXTH1, MeXTH4, MeXTH14, MeXTH15, MeXTH23, and MeXTH26) were also associated with ribosomal structure and polysaccharide binding, underscoring their multifunctional roles in growth, development, and stress responses.

3.9. Cassava MeXTH Genes Are Expressed in Different Organs and Tissues in Cassava Plants

To investigate the possible roles of the MeXTH genes in the cassava genome, we analyzed the expression profiles of the 37 MeXTH genes in different organs and tissues using publicly available RNA-seq data (accession numbers: PR-JNA324539, GEO dataset: GSE82279, and cassava variety: TME204) published by Wilson et al. [66]. The expression profiles of all 37 MeXTH genes were assessed using hierarchical clustering, which revealed distinct patterns across various cassava organs/tissues (Figure 7 and Table S7). Notably, MeXTH30 and MeXTH19 exhibited high expression in most tissues, suggesting essential roles in cassava development. Some genes showed preferential expression across the detected organs/tissues. For example, MeXTH17 and MeXTH35 were most highly expressed in FR, while MeXTH8 and MeXTH19 were abundant in RAM and OES, respectively. MeXTH35 and MeXTH37 had peak expression in leaf and midvein tissues, and MeXTH19 was prominent in the lateral bud. Further analysis indicated that MeXTH19, MeXTH17, and MeXTH36 had higher expression in the FEC, SR, and stem tissues, suggesting involvement in growth and functionality in these tissues. In petiole and SAM, MeXTH19 and MeXTH8 were the most expressed. These findings highlight the likely importance of specific MeXTH genes in tissue development and function.

3.10. Response of MeXTH Genes to Abiotic Stresses and Hormone Treatments

To investigate the transcriptional response of MeXTH genes to abiotic stress and phytohormones, we selected 12 representative genes for qRT-PCR analysis under osmotic (PEG), salt (NaCl), salicylic acid (SA), and methyl jasmonate (MeJA) treatments (Figure 8 and Figure 9). The expression patterns revealed that most MeXTH genes are responsive to these stimuli, but with distinct dynamics and specificities.

Under osmotic stress, a general upregulation was observed for some genes, with MeXTH23 exhibiting the most pronounced induction, with its expression at 12 h and 24 h being significantly greater than that at 0 h (p < 0.05) (Figure 8). Notably, MeXTH23, MeXTH26, and MeXTH35 showed a sustained increase throughout the treatment period, suggesting their potential as primary responders to water deficit. Expression of six genes (MeXTH5, MeXTH11, MeXTH12, MeXTH28, MeXTH29, and MeXTH34) peaked significantly at 4 h but then decreased significantly at later time points (12 h and 24 h) (p < 0.05). Under salt stress, MeXTH28, MeXTH29, and MeXTH35 showed a significant increase at 12 h after NaCl treatment (p < 0.05). MeXTH23, MeXTH25, and MeXTH26 reached their highest expression levels at 24 h, which were significantly greater than those at earlier time points (4 h) (p < 0.05). Conversely, the expression of MeXTH12 and MeXTH34 was significantly downregulated by salt stress (p < 0.05). MeXTH25 and MeXTH34 showed an “up-down-up” pattern over 24 h. Certain genes, such as MeXTH23 and MeXTH26, were consistently upregulated under both stresses, suggesting a role in common signaling pathways mediating stress responses (Figure 8).

In response to SA treatment, 12 MeXTH genes displayed dynamic temporal expression patterns. The majority of genes, including MeXTH2, MeXTH5, MeXTH11, MeXTH23, MeXTH25, MeXTH29, and MeXTH34, exhibited a significant induction that peaked at 4 h, followed by a decline at later time points (p < 0.05). MeXTH28 and MeXTH37 also showed a similar trend of early upregulation, reaching their peak expression at 12 h before subsequently decreasing (p < 0.05). MeXTH28 and MeXTH37 were significantly upregulated at 12 h, with MeXTH37 displaying no change from 0 to 4 h before a substantial increase at 12 h. Under MeJA treatment, MeXTH29 levels increased significantly, peaking at 4 h (p < 0.05). MeXTH2, MeXTH11, MeXTH12, MeXTH23, MeXTH26, MeXTH34, and MeXTH37 exhibited significant expression increases after 12 h, followed by a significant decline at 24 h (p < 0.05). MeXTH26 and MeXTH35 showed similar profiles, although the expression of MeXTH35 at 12 h remained relatively low and was not significantly different from its level at 0 h. MeXTH25 and MeXTH28 were significantly downregulated after 4 h, and MeXTH25 exhibited progressive suppression, decreasing significantly by 24 h (Figure 9).

4. Discussion

The adverse effects of environmental stresses on cassava’s growth and development are well-established, with recent studies emphasizing XTHs’ role in mitigating these stresses through cell wall remodeling and enhanced biogenesis [4]. Such properties make XTHs ideal targets for molecular breeding to improve stress resistance in plants. Our analysis of the XTH gene family in cassava identified 37 members in cassava, slightly more than in A. thaliana (33) and O. sativa (29) [14,16], suggesting that cell wall remodeling may be a particularly important adaptive mechanism in cassava. Phylogenetic analysis grouped cassava MeXTHs with A. thaliana, rice, and poplar XTHs into Groups I/II, IIIA, and IIIB and the Ancestral group, which is consistent with the results of other plants, indicating that there are still some commonalities among species [2,3,71]. Analysis of the conserved structural domains showed that the number and type of conserved motifs differed between cassava MeXTH proteins. The foundational XTH core domain, known as motif 4, exhibited its presence across all MeXTH proteins. In addition, structural conservation across subfamilies, especially the ExDxEx motif in active sites, though loop2, loop3, and N-glycosylation site variations were observed, in line with prior studies [2,8]. Site-directed mutagenesis of AtXTH22 identified the ExDxE motif as essential for activity, and MeXTH36 was one of seven BolXTHs lacking this motif [72]. This intriguing observation suggested that MeXTH might assume an indispensable role, considering the diverse functions attributed to XTH. MeXTH gene sequences mainly contained three or four introns, like rice [16], while fewer introns were found in some genes, potentially affecting alternative splicing and gene regulation [73].

Gene duplication is a primary driver of the expansion of gene families, and tandem duplications and segmental duplications are considered the primary duplication modes [74]. It appears that segmental duplication contributes more to cassava XTH expansion than does tandem duplication [75]. Consistent with that which occurs in other species, such as poplar [4], wheat [76], and Schima superba [77]. Moreover, our results showed that many orthologous pairs between cassava and other dicots, while fewer pairs were observed with monocots like rice, likely due to the dramatic divergence of XTH in dicotyledonous and monocotyledonous plants [78,79].

The expression profiles of MeXTH genes across cassava tissues suggest varied roles. Analysis of publicly available microarray data [66] showed that genes like MeXTH10, MeXTH13, MeXTH19, MeXTH26, MeXTH30, and MeXTH36 are highly expressed across 11 tissues, suggesting involvement in multiple developmental processes. Tissue-specific expression patterns were also noted; for example, MeXTH17 was more highly expressed in the roots, indicating a possible role in root growth. Similar patterns were seen in A. thaliana, where XTH19 and XTH23 contribute to lateral root development via the brassinosteroid pathway [35]. Drought and salinity are key abiotic stresses affecting plant growth [80,81,82], with evidence supporting XTH genes’ regulatory role in adapting to these stresses. XTH genes are crucial for modulating cell wall extension to enhance drought tolerance [83]. In barley, HvXTH1 silencing via Virus-Induced Gene Silencing (VIGS) led to increased shoot weight and reduced water loss under drought [12], while PeXTH expression in tobacco improved osmotic tolerance through reduced water loss [67]. AtXTH19 and AtXTH23 are also salt stress-responsive in A. thaliana [37], and overexpressing the PeXTH gene in tobacco increased water retention, conferring salinity tolerance [84]. The rapid induction of most MeXTH genes within 4 h of PEG treatment implicates them in the initial stress response, potentially mediating adaptive cell wall remodeling and inhibition of cell elongation to cope with water deficit. This pattern mirrors the function of CaXTH3 in hot pepper, which imparts drought tolerance through cell wall remodeling to strengthen the wall layers [39]. The sustained upregulation of MeXTH23 under prolonged salt stress (24h) suggests that it is not involved in an immediate shock response but rather in the ongoing adaptation to ionic and osmotic stress. We hypothesize that MeXTH23-mediated cell wall remodeling is crucial for maintaining root architectural integrity under high salinity, potentially by enhancing wall flexibility to allow for controlled cell expansion under osmotic duress. The concerted upregulation of MeXTH2, MeXTH23, and MeXTH26 under both drought and salt stress suggests that their role is central to a generalized osmotic stress adaptation strategy. This finding is consistent with prior research in which overexpression of DkXTH1 was shown to confer enhanced tolerance to salt and drought stresses in transgenic Arabidopsis [38]. In contrast, it was revealed that some MeXTH genes play negative roles in salt stress, such as MeXTH12 and MeXTH34. Similarly, AtXTH30 in Arabidopsis negatively regulates salt tolerance by reducing the crystalline cellulose content and promoting microtubule depolymerization [85]. A possible explanation for these opposing roles is their possession of divergent enzyme activities (XET and XEH).

Extensive studies have revealed that multiple phytohormones regulate plant stress responses. In A. thaliana, several XTH genes respond to hormones like auxin, BR, GA, and ABA [14]. For instance, AtXTH31/XTR8, specific to endosperm, is induced by SA [70], with mutations affecting ABA sensitivity and germination [86]. In kiwifruit, ethylene exposure triggers XET gene expression [87]. Xu et al. [35] demonstrated that the brassinosteroid-responsive and salt-inducible genes AtXTH19 and AtXTH23 participate in lateral root development under salt stress in Arabidopsis. These findings collectively indicate that most MeXTH genes are upregulated by the phytohormones SA and MeJA, suggesting that their role in salt and drought stress responses is mediated through these hormonal pathways. Moreover, the promoters of MeXTH genes are enriched in stress- and hormone-responsive elements, such as ABRE, ARE, and LTR, and TGACG-motifs [87,88]. This in silico prediction is strongly corroborated by our qRT-PCR results, which show that genes containing these specific elements are indeed significantly induced by the corresponding stresses/hormones. For example, the promoter of MeXTH2 is rich in ABRE, ARE, and LTR elements, directly aligning with its role as a broad-spectrum stress responder. As ABA is a primary signal under water stress, this suggests that MeXTH2 is a direct transcriptional target of the ABA signaling pathway, positioning it as a key effector downstream of stress hormone perception. The induction of most MeXTH genes by MeJA is consistent with the prevalence of JA-responsive cis-elements (TGACG-motifs) identified in their promoter regions. This suggests that these genes are likely components of the jasmonate signaling pathway, contributing to stress-responsive cell wall dynamics [89]. The precise functions of these cis-elements and the regulatory relationships inferred from these correlations require further experimental validation, such as luciferase reporter assays and targeted gene mutagenesis.

5. Conclusions

In this study, we identified 37 MeXTH genes in the cassava genome, classifying them into three distinct subfamilies through phylogenetic analysis. Our results showed significant variation in the physicochemical properties of the MeXTH family. Structure-based alignment revealed that all MeXTH proteins contain conserved domains and N-glycosylation sites. Collinearity analysis indicated that all duplicated gene pairs underwent strong purifying selection, suggesting evolutionary pressure to maintain their functions. The analysis of cis-acting elements and protein interaction predictions further highlighted the role of MeXTH genes in various biological processes, including growth, development, phytohormone response, and stress adaptation. Expression profiling by qRT-PCR under osmotic stress, salt stress, SA, and MeJA treatments revealed that most MeXTH genes are responsive to environmental and hormonal cues. Notably, MeXTH23 and MeXTH26 were consistently upregulated under both osmotic and salt stresses, warranting further investigation for improving cassava’s abiotic stress tolerance. This comprehensive analysis provides a foundation for future research on XTH genes in cassava, with potential applications in the genetic improvement of cassava and related species.