Genome-Wide Identiﬁcation and Expression Analysis of the Xyloglucan Endotransglucosylase/Hydrolase Gene Family in Sweet Potato [ Ipomoea batatas (L.) Lam]

: The xyloglucan endotransglucosylase/hydrolase (XET/XEH, also named XTH) family is a multigene family, the function of which plays a signiﬁcant role in cell-wall rebuilding and stress tolerance in plants. However, the speciﬁc traits of the XTH gene family members and their expression pattern in different tissues and under stress have not been carried out in sweet potato. Thirty-six XTH genes were identiﬁed in I. batatas, all of which had conserved structures (Glyco_hydro_16). Based on Neighbor-Joining phylogenetic analysis the IbXTHs can be divided into three subfamilies—the I/II, IIIA, and IIIB subfamilies, which were unevenly distributed on 13 chromosomes, with the exception of Chr9 and Chr15. Multiple cis-acting regions related to growth and development, as well as stress responses, may be found in the IbXTH gene promoters. The segmental duplication occurrences greatly aided the evolution of IbXTHs . The results of a collinearity analysis showed that the XTH genes of sweet potato shared evolutionary history with three additional species, including A. thaliana , G. max , and O. sativa . Additionally, based on the transcriptome sequencing data, the results revealed that the IbXTHs have different expression patterns in leaves, stems, the root body (RB), the distal end (DE), the root stock (RS), the proximal end (PE), the initiative storage root (ISR), and the ﬁbrous root (FR), and many of them are well expressed in the roots. Differentially expressed gene (DEG) analysis of FRs after hormone treatment of the roots indicated that IbXTH28 and IbXTH30 are up-regulated under salicylic acid (SA) treatment but down-regulated under methyl jasmonate (MeJA) treatment. Attentionally, there were only two genes showing down-regulation under the cold and drought treatment. Collectively, all of the ﬁndings suggested that genes from the XTH family are crucial for root speciﬁcity. This study could provide a theoretical basis for further research on the molecular function of sweet potato XTH genes.


Introduction
The cell wall is a crucial characteristic feature of plant cells. Plant cell walls serve as an exterior supporting structure that greatly affects the size and form of the cells during the growth and development of the plant. The cell wall mostly consists of cellulose, hemicellulose, lignin, and pectin. Cell-wall reconstruction is accompanied by breakage and regeneration of the xyloglucan cell wall [1]. Xyloglucan, which is a cellulose chain containing oligosaccharide side chains, is the most important hemicellulose component in the primary cell wall of dicotyledons [2]. XTHs, one of the key enzymes in cell-wall remodeling, is widely present in plant cells; it can catalyze the connection and breakage of xyloglucan molecules and modify the fiber-xyloglucan composite structure by controlling the elasticity and ductility of cells [3].

Phylogenetic Analysis and Classification of IbXTHs
There are 36 XTH gene members in the sweet potato genome; these XTH gene members were designated IbXTH01-IbXTH36 according to their position in the chromosome ( Figure 1). In this study, the BLAST and HMMER search methods were used and then followed by identification using the Pfam tool. A total of 36 XTH genes were identified in the genome of I. batatas and named from IbXTH1 to IbXTH36 by their chromosome position. The chromosomal location of IbXTHs showed that the 36 identified XTH genes were unequally mapped on the 13 chromosomes, except for chromosomes 9 and 15. For instance, chromosome 13 contained the largest number of IbXTHs (8), while chromosomes 1, 5, 6, 8, and 12 only contained one IbXTH gene. Gene cluster exists for chromosome 3, 7, 13, and 14 and was also identified in the chromosome distribution map, especially in the interval 2523570 to 25228579 of Chr13 containing two gene clusters with three and five IbXTHs, respectively, indicating there are tandem duplication events in the IbXTH gene family.

Phylogenetic Analysis and Classification of IbXTHs
There are 36 XTH gene members in the sweet potato genome; these XTH gene members were designated IbXTH01-IbXTH36 according to their position in the chromosome ( Figure 1). In this study, the BLAST and HMMER search methods were used and then followed by identification using the Pfam tool. A total of 36 XTH genes were identified in the genome of I. batatas and named from IbXTH1 to IbXTH36 by their chromosome position. The chromosomal location of IbXTHs showed that the 36 identified XTH genes were unequally mapped on the 13 chromosomes, except for chromosomes 9 and 15. For instance, chromosome 13 contained the largest number of IbXTHs (8), while chromosomes 1, 5, 6, 8, and 12 only contained one IbXTH gene. Gene cluster exists for chromosome 3, 7, 13, and 14 and was also identified in the chromosome distribution map, especially in the interval 2523570 to 25228579 of Chr13 containing two gene clusters with three and five IbXTHs, respectively, indicating there are tandem duplication events in the IbXTH gene family. To clarify the taxonomic and evolutionary relationships of the identified sweet potato IbXTH gene family, an Neighbor-Joining phylogenetic tree was constructed by MEGA7.0 with 36 XTH protein sequences from sweet potato and 33 AtXTH protein sequences (Figure 2). According to the previous subfamily classification of Arabidopsis XTH proteins-in which AtXTH1-AtXTH26 belong to subgroup I/II, AtXTH31 and AtXTH32 belong to subgroup III-A, and the rest of the proteins belong to subgroup III-B-sweet potato XTH proteins were classified into three subfamilies, namely I/II (blue), IIIA (yellow), and IIIB (pink). Among these, the I/II subfamily was the largest group, with 51 XTH protein sequences including 22 XTH genes in sweet potato There were eight XTHs in IIIA, and the IIIB subfamily in which XEH activity has been only reported contained 10 proteins. To clarify the taxonomic and evolutionary relationships of the identified sweet potato IbXTH gene family, an Neighbor-Joining phylogenetic tree was constructed by MEGA7.0 with 36 XTH protein sequences from sweet potato and 33 AtXTH protein sequences ( Figure 2). According to the previous subfamily classification of Arabidopsis XTH proteinsin which AtXTH1-AtXTH26 belong to subgroup I/II, AtXTH31 and AtXTH32 belong to subgroup III-A, and the rest of the proteins belong to subgroup III-B-sweet potato XTH proteins were classified into three subfamilies, namely I/II (blue), IIIA (yellow), and IIIB (pink). Among these, the I/II subfamily was the largest group, with 51 XTH protein sequences including 22 XTH genes in sweet potato There were eight XTHs in IIIA, and the IIIB subfamily in which XEH activity has been only reported contained 10 proteins. Neighbor-Joining phylogenetic tree analysis of XTH proteins from sweet potato and Arabidopsis constructed by MEGA 7.0 with the default parameters. The evolutionary tree was divided into three subfamilies with different colors. According to the previous subfamily classification of Arabidopsis XTH proteins, in which AtXTH1 -AtXTH26 belong to subgroup I/II, AtXTH31 and AtXTH32 belong to subgroup III-A, and the rest of the proteins belong to subgroup III-B, sweet potato XTH proteins were classified into three subfamilies, namely I/II (blue), IIIA (yellow), and IIIB (pink). The blue triangle represents Arabidopsis XTH proteins, while the red circle represents sweet potato XTH proteins.

Identification and Sequence Analysis of XTH Genes in I. batatas
Among the 36 IbXTHs in sweet potato, the shortest (IbXTH13) was of 164 aa, while the longest (IbXTH8) was of 344 aa, and the average length of genes was 280.22 amino acids. The predicted MWs of the IbXTH were as high as 38.73763kDa (IbXTH8) and as low as 3.30876 kDa (IbXTH17). The pI (isoelectric point) values of 21 out of 36 XTH in I. batatas were higher than 7.0, indicating that most of the sweet potato XTHs were alkaline proteins. The pI value of IbXTH04was only 4.69, whereas the pI value of IbXTH21 as the Figure 2. Neighbor-Joining phylogenetic tree analysis of XTH proteins from sweet potato and Arabidopsis constructed by MEGA 7.0 with the default parameters. The evolutionary tree was divided into three subfamilies with different colors. According to the previous subfamily classification of Arabidopsis XTH proteins, in which AtXTH1 -AtXTH26 belong to subgroup I/II, AtXTH31 and AtXTH32 belong to subgroup III-A, and the rest of the proteins belong to subgroup III-B, sweet potato XTH proteins were classified into three subfamilies, namely I/II (blue), IIIA (yellow), and IIIB (pink). The blue triangle represents Arabidopsis XTH proteins, while the red circle represents sweet potato XTH proteins.

Identification and Sequence Analysis of XTH Genes in I. batatas
Among the 36 IbXTHs in sweet potato, the shortest (IbXTH13) was of 164 aa, while the longest (IbXTH8) was of 344 aa, and the average length of genes was 280.22 amino acids. The predicted MWs of the IbXTH were as high as 38.73763 kDa (IbXTH8) and as low as 3.30876 kDa (IbXTH17). The pI (isoelectric point) values of 21 out of 36 XTH in I. batatas were higher than 7.0, indicating that most of the sweet potato XTHs were alkaline proteins. The pI value of IbXTH04 was only 4.69, whereas the pI value of IbXTH21 as the biggest was of 9.23, and others distributed between 4.69 and 9.38. There was a 50/50 split between stable proteins (<40) and unstable proteins. GRAVY (Grand average of hydropathicity) showed that all IbXTH proteins were hydrophilic, with an average GRAVY of −0.3595. Subcellular localization analysis showed that, 36.1% IbXTH proteins were extracellular, while 63.8% were predicted to be located in the plasma membrane. The aliphatic index of the peptides ranged from 57.91 (IbXTH18) to 101.33 (IbXTH13) ( Table 1).

Gene Structure Analysis of Sweet Potato XTHs
Generally, genes clustered in a subfamily have similar structures. In this research, XTHs in the I/II subfamily had one to three introns, except for IbXTH08 and IbXTH23 with four introns and IbXTH15 (zero introns). There are three to four introns in IIIA and IIIB subfamilies, except for IbXTH05, which has two introns belongs to the IIIA subfamily.
To further explore the phylogenetic relationships of the IbXTH family and to investigate their potential functions, PFAM data were used to analyze the conserved domains of 36 XTH protein sequences. Results showed that the Glyco_hydro_16 domain was found in all IbXTHs, and 29 IbXTHs contained both the Glyco_hydro_16 and the XET_C domains, holding 80.56% in all genes. In addition, to further reveal the structural diversity of IbXTH protein sequences, MEME online software combined with the visualized motif pattern function of TBtools was used to analyze the related motifs. As shown in Figures 3 and 4, the motif number of the IbXTH family members ranged from four to nine. The XET_ C domain mainly covered motif 5. Motif 4 and motif 5 were the longest motifs containing 41 aa, while the shortest contained only 10aa (motif 9). Motif 1 was widely present in I/II and IIIA and IIIB, which contains the active site DEIDFEFLG (E is the catalytic point). The IbXTH5, IbXTH12, and IbXTH34 had the highest number of motifs, with nine motifs which had a consistent arrangement order, starting with motif 10 and ending with motif 5. However, IbXTH13 and IbXTH31 had the least number of motifs, with only four motifs. Furthermore, we also found that motif 10 only existed in the IIIA subfamily, suggesting that motif 5 might have evolutionary specificity. Except for IbXTH9, all IbXTH gene members of the IIIA and IIIB subfamilies had motif 9.     (Table S1); the diagram was created by MEME online program. The X-axis shows the conserved sequences of the motifs, and the conservation of residues are indicated by the height of the letters. The Y-axis represents the conservation of the amino acid.  (Table S1); the diagram was created by MEME online program. The X-axis shows the conserved sequences of the motifs, and the conservation of residues are indicated by the height of the letters. The Y-axis represents the conservation of the amino acid.

Chromosomal Distribution and Synteny Analysis of IbXTHs
It was discovered that four pairs of IbXTHs, which belong to inter-chromosome segmental duplications, had a collinear (red) relationship between and within chromosomes in sweet potato based on the collinearity analysis performed by MCScanX ( Figure 5). We inferred that segmental gene duplication might play a significant role in the growth of the IbXTH gene. Furthermore, the duplicated genes belonged to the same subfamily, and the four pairs of genes were found to have strong collinearity. IbXTH18 and IbXTH19, IbXTH17, and IbXTH32 were in the I/II subfamily; IbXTH12 and IbXTH34 were in the IIIB subfamily; and IbXTH20 and IbXTH21 belonged to the IIIB subfamily. In addition, in order to comprehend the mechanism of gene divergence and evolutionary pressure, the Ka/Ks ratios were also determined. The gene is more susceptible to nonsynonymous mutation if the Ka/Ks ratio is >1, which denotes positive selection. The ratio indicatesneutral selection if it is equal to 1, while indicated purifying selection. I Ifthe ratio is <1, and synonymous mutation is more likely in the gene. All Ka/Ks values were less than 0.5, meaning that XTH family genes have gone through a strong purifying selection during evolution ( Table 2).  To further explore the gene duplication events of the XTH gene family, collinearity and replication analysis of all genes were carried out using TBtools software, and three comparative synthetic maps of three representative species with sweet potato were constructed, including two dicotyledonous plants (A. thaliana, G. max) and one monocotyledonous plant (O. sativa) ( Figure 6). The collinearity analysis results showed that a total of 63 covariate pairs were found, including 12 pairs in A. thaliana (green line); 49 XTH gene pairs exits in G. max (purple line); and only one pair in O. sativa (red line) compared to I.batatas. In addition, IbXTH24 showed a high collinear relationship with three comparative species (two in A. thaliana, four in G. max, and one in O. sativa), suggesting that IbXTHs showed higher evolutionary differentiation in dicotyledonous plants than in monocotyledonous plants. . Collinearity analysis of the XTHS from sweet potato and three other species which were analyzed and visualized by Tbtools. The green lines represent XTH syntenic gene pairs between sweet potato and Arabidopsis. The purple lines represent XTH syntenic gene pairs between sweet potato and bean. The red lines represent XTH syntenic gene pairs between sweet potato and rice, and the gray lines represent orthologous genes of sweet potato with three other species.

Cis-Acting Elements in IbXTHs
To predict the function of IbXTH gene family, the 2000bp upstream CDS sequences were extracted from the promoter regions of 36 IbXTHs of sweet potato by using Plant-CARE database (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/ (accessed on 15 September 2022). It was found that the promoter region of the sweet potato XTH family is abundant in cis-acting elements, which are divided into the following six categories: light-response, hormonal-response, environmental-stress-related, developmentalrelated, binding-site-related, and other elements (Figure 7). All XTHs of sweet potato contained 5 to 19 light-responsive elements, including the GTGGC-motif, TCCC-motif, and TCT motif. Five hormonal-related response elements were invested in all IbXTHs, ABA(ABRE), GA (GARE-motif, TATC-box, and P-box), IAA (AuxRR-core, TGA-element, and TGA-box), MeJA (CGTCA-motif and TGACG-motif), and SA (TCA-element and SARE). Endosperm expression (GCN4_motif), meristem expression elements (CAT-box), and seed-specific regulation (RY-element) are development-related elements, which were also determined to exist in half of the IbXTH gene. Among the elements related to environmental stress (Figure 7c), anaerobic induction (ARE) and low-temperature responsiveness (LTR) was found in 31 IbXTHs. In addition, there are 42 MYB response elements

Cis-Acting Elements in IbXTHs
To predict the function of IbXTH gene family, the 2000bp upstream CDS sequences were extracted from the promoter regions of 36 IbXTHs of sweet potato by using PlantCARE database (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/ (accessed on 15 September 2022). It was found that the promoter region of the sweet potato XTH family is abundant in cis-acting elements, which are divided into the following six categories: light-response, hormonal-response, environmental-stress-related, developmental-related, binding-site-related, and other elements (Figure 7). All XTHs of sweet potato contained 5 to 19 light-responsive elements, including the GTGGC-motif, TCCC-motif, and TCT motif. Five hormonal-related response elements were invested in all IbXTHs, ABA(ABRE), GA (GARE-motif, TATC-box, and P-box), IAA (AuxRR-core, TGA-element, and TGA-box), MeJA (CGTCA-motif and TGACG-motif), and SA (TCA-element and SARE). Endosperm expression (GCN4_motif), meristem expression elements (CAT-box), and seed-specific regulation (RY-element) are development-related elements, which were also determined to exist in half of the IbXTH gene. Among the elements related to environmental stress (Figure 7c

Expression Profiles of XTH Genes in Sweet Potato
To investigate the potential biological functions of IbXTHs in the growth and development of sweet potato, the expression profiles of the 36 IbXTHs were detected in seven different tissues by transcriptome data of "Xushu 18" retrieved from the GEO database (PRJNA511028). The expression of IbXTHs in each tissue of sweet potato was obtained, and a heat map was drawn according to FPKM (Fragments Per Kilobase of exon model per Million mapped fragments). If the FPKM of a gene is less than 1, it is considered that the gene lacks expression data.In this study, The FPKM ofIbXTH5, IbXTH7, IbXTH9, IbXTH13, IbXTH15, and IbXTH36 in all tissues were less than 1, indicating that these genes had no function or didn't expressed. On this basis, the remaining 30 IbXTHs were expressed in at least one tissue (Table S2).
The IbXTHs had different expression patterns in various tissues (Figure 8a), which implied that the IbXTHs might have played different roles in developmental processes. Generally, almost half of the identified IbXTHs (14 genes) in sweet potato were unregulated in all of the detected tissues. The others were highly expressed in the root body (RB), distal end (DE), root stock (RS), proximal end (PE), and initiative storage root (ISR), while they were downregulated in the leaf, stem, and fibrous root (FR), which showed strong tissue-specific expression. Compared to other family members in the XTH gene family of sweet potato, it is noteworthy that IbXTH3, IbXTH22, IbXTH28, IbXTH29, and IbXTH30 have relatively high expression levels. Among them, IbXTH28, IbXTH29, and IbXTH30 have high expression levels in root tissues other than FR, indicating that these genes may be necessary for the development of root tissues.
Furthermore, the FB expression patterns of IbXTHs were analyzed under various abiotic stresses and different hormone treatments (Figure 8b). Nine IbXTHs were identified

Expression Profiles of XTH Genes in Sweet Potato
To investigate the potential biological functions of IbXTHs in the growth and development of sweet potato, the expression profiles of the 36 IbXTHs were detected in seven different tissues by transcriptome data of "Xushu 18" retrieved from the GEO database (PRJNA511028). The expression of IbXTHs in each tissue of sweet potato was obtained, and a heat map was drawn according to FPKM (Fragments Per Kilobase of exon model per Million mapped fragments). If the FPKM of a gene is less than 1, it is considered that the gene lacks expression data.In this study, The FPKM ofIbXTH5, IbXTH7, IbXTH9, IbXTH13, IbXTH15, and IbXTH36 in all tissues were less than 1, indicating that these genes had no function or didn't expressed. On this basis, the remaining 30 IbXTHs were expressed in at least one tissue (Table S2).
The IbXTHs had different expression patterns in various tissues (Figure 8a), which implied that the IbXTHs might have played different roles in developmental processes. Generally, almost half of the identified IbXTHs (14 genes) in sweet potato were unregulated in all of the detected tissues. The others were highly expressed in the root body (RB), distal end (DE), root stock (RS), proximal end (PE), and initiative storage root (ISR), while they were downregulated in the leaf, stem, and fibrous root (FR), which showed strong tissue-specific expression. Compared to other family members in the XTH gene family of sweet potato, it is noteworthy that IbXTH3, IbXTH22, IbXTH28, IbXTH29, and IbXTH30 have relatively high expression levels. Among them, IbXTH28, IbXTH29, and IbXTH30 have high expression levels in root tissues other than FR, indicating that these genes may be necessary for the development of root tissues.
IbXTH29, and IbXTH30 were the most significantly up-regulated under MEJA treatment. FKPM expression had increased by about 72, 104, and 112 times, respectively. In addition, we analyzed the differential expression genes (DEGs) of IbXTHs under various abiotic stresses, such as salt, drought, and cold stress. The IbXTH gene family members of the FR of sweet potato had no up-regulated genes under all abiotic stresses but had two downregulated genes, including IbXTH02 and IbXTH12, which were down-regulated in cold and drought stress. IbXTH12 was also down-regulated under the stress of salt.

Tissue-Specific and Stress-Inducible Expression of IbXTH02 and IbXTH12
Subsequently, the expression patterns of two IbXTHs (IbXTH02 and IbXTH12), which displayed substantial change in the RNA-data and which down-regulated under salt and drought treatment and showed xyloglucan endo-hydrolase activity, were further examined under two abiotic stresses: salt stress (0.2 mol/L NaCl) and PEG-induced drought stress (20% PEG6000) by qRT-PCR assay, and a two-fold cut-off value was explored (Table S5, Figure 9). Under PEG-induced drought stress (Figure 9a,b; Table S5a,b), IbXTH02 and Furthermore, the FB expression patterns of IbXTHs were analyzed under various abiotic stresses and different hormone treatments (Figure 8b). Nine IbXTHs were identified not expressed in all the treatments tested, including IbXTH5, IbXTH7-IbXTH9, IbXTH11, and IbXTH13-IbXTH15. On this basis, the remaining 27 IbXTHs were expressed in at least one treatment. As shown in Figure 8b (Table S3), there were 10 differential expression genes (DEGs) in the FR after SA treatment (9 up-regulated and 1 down-regulated), 13 after MEJA treatment (7 up-regulated and 6 down-regulated), and only 1 down-regulated gene in FR under the treatment of ABA(IbXTH02). It is worth mentioning that the IbXTH28, IbXTH29, and IbXTH30 were the most significantly up-regulated under MEJA treatment. FKPM expression had increased by about 72, 104, and 112 times, respectively. In addition, we analyzed the differential expression genes (DEGs) of IbXTHs under various abiotic stresses, such as salt, drought, and cold stress. The IbXTH gene family members of the FR of sweet potato had no up-regulated genes under all abiotic stresses but had two down-regulated genes, including IbXTH02 and IbXTH12, which were down-regulated in cold and drought stress. IbXTH12 was also down-regulated under the stress of salt.

Tissue-Specific and Stress-Inducible Expression of IbXTH02 and IbXTH12
Subsequently, the expression patterns of two IbXTHs (IbXTH02 and IbXTH12), which displayed substantial change in the RNA-data and which down-regulated under salt and drought treatment and showed xyloglucan endo-hydrolase activity, were further examined under two abiotic stresses: salt stress (0.2 mol/L NaCl) and PEG-induced drought stress (20% PEG6000) by qRT-PCR assay, and a two-fold cut-off value was explored (Table S5, Figure 9). Under PEG-induced drought stress (Figure 9a,b; Table S5a,b), IbXTH02 and IbXTH12 in roots were down-regulated in all periods, while the expression of IbXTH02 and IbXTH02 were expressed at a high level at 0 h and showed a trend of first decreasing and then increasing after treatment. The expression level of IbXTH12 reached the lowest level at 12 h, which was down-regulated with 604-fold induction. The expression level of IbXTH12 reached the lowest level at 24 h, which was down-regulated with 153-fold induction. IbXTH02 and IbXTH12 were also down-regulated under salt stress (Figure 9c,d; Table S5c,d). It is worth mentioning that IbXTH02 had the lowest relative expression at 12 h after salt treatment (0.0042), with 236-fold induction down-regulated. Moreover, IbXTH12 displayed a tendency of first declining, then increasing, and then decreasing. At 24 h of salt treatment, it had decreased by nearly 267-fold, reaching its lowest expression level. These results indicated that different genes had different response times to drought stress in sweet potato. Collectively, these results implied that IbXTH genes might have different expression under abiotic stress. IbXTH12 in roots were down-regulated in all periods, while the expression of IbXTH02 and IbXTH02 were expressed at a high level at 0 h and showed a trend of first decreasing and then increasing after treatment. The expression level of IbXTH12 reached the lowest level at 12 h, which was down-regulated with 604-fold induction. The expression level of IbXTH12 reached the lowest level at 24 h, which was down-regulated with 153-fold induction. IbXTH02 and IbXTH12 were also down-regulated under salt stress (Figure 9c,d; Table S5c,d). It is worth mentioning that IbXTH02 had the lowest relative expression at 12 h after salt treatment (0.0042), with 236-fold induction down-regulated. Moreover, IbXTH12 displayed a tendency of first declining, then increasing, and then decreasing. At 24 h of salt treatment, it had decreased by nearly 267-fold, reaching its lowest expression level. These results indicated that different genes had different response times to drought stress in sweet potato. Collectively, these results implied that IbXTH genes might have different expression under abiotic stress. Figure 9. Expression profiles of IbXTH02 and IbXTH12 in response to PEG6000 and NaCl treatment (a) Expression patterns of IbXTH02 in response to PEG-induced drought stress in FR of Xushu 18 were examined by a qPCR assay. (b) Expression patterns of IbXTH12 in response to PEG-induced drought stress in FR of Xushu18 were examined by a qPCR assay. (c) Expression patterns of IbXTH02 in response to NaCl stress in FR of Xushu18 were examined by a qPCR assay. (d) Expression patterns of IbXTH12 in response to NaCl stress in FR of 'Xushu 18' were examined by a qPCR assay. The fibrous root was obtained at 0, 3, 6, 12, 24, and 48 h after PEG6000 and NaCl stress. The mean fold changes of each gene between treated and control fibrous root in Xushu18 were used to calculate its relative expression levels. Data are means ± SD of three biological replicates. Means denoted by the same letter do not significantly differ at p < 0.05, as determined by Duncan's multiple range test.

Discussion
Xyloglucan in the cell wall participates in the reconstruction, breakage, and regeneration of the cell wall. The XTH gene family has been identified in several plant species, including Arabidopsis [5], rice [9], barley [22], poplar [23], and tomato [4,7].
In this study, we report the identification and characterization of the sweet potato XTH gene family and constructed a phylogenetic tree using the XTH proteins from sweet potato and Arabidopsis. Expression pattern analysis suggested that IbXTHs may play an important role under various stresses, in which IbXTH02 and IbXTH12 showed downregulation under PEG6000 and NaCl treatment. . Expression profiles of IbXTH02 and IbXTH12 in response to PEG6000 and NaCl treatment (a) Expression patterns of IbXTH02 in response to PEG-induced drought stress in FR of Xushu 18 were examined by a qPCR assay. (b) Expression patterns of IbXTH12 in response to PEG-induced drought stress in FR of Xushu18 were examined by a qPCR assay. (c) Expression patterns of IbXTH02 in response to NaCl stress in FR of Xushu18 were examined by a qPCR assay. (d) Expression patterns of IbXTH12 in response to NaCl stress in FR of 'Xushu 18' were examined by a qPCR assay. The fibrous root was obtained at 0, 3, 6, 12, 24, and 48 h after PEG6000 and NaCl stress. The mean fold changes of each gene between treated and control fibrous root in Xushu18 were used to calculate its relative expression levels. Data are means ± SD of three biological replicates. Means denoted by the same letter do not significantly differ at p < 0.05, as determined by Duncan's multiple range test.

Discussion
Xyloglucan in the cell wall participates in the reconstruction, breakage, and regeneration of the cell wall. The XTH gene family has been identified in several plant species, including Arabidopsis [5], rice [9], barley [22], poplar [23], and tomato [4,7].
In this study, we report the identification and characterization of the sweet potato XTH gene family and constructed a phylogenetic tree using the XTH proteins from sweet potato and Arabidopsis. Expression pattern analysis suggested that IbXTHs may play an important role under various stresses, in which IbXTH02 and IbXTH12 showed down-regulation under PEG6000 and NaCl treatment.

Charaterization of Sweet Potato XTHs Gene Family
It is well-known that plant XTHs play a vital role in regulating the extensibility of the cell wall, but there is limited information on the number of XTH gene family members in sweet potato and the evolutionary relationships between IbXTH genes. The number of IbXTH genes discovered was significantly lower than in Nicotiana tabacum [24] and Glycine max [10] and was grouped into three categories by previous phylogenetic studies in Arabidopsis. Furthermore, evolutionary mechanism analysis suggested that the IbXTHs family expanded partly due to segmental duplication events, which further lead to the conserved protein motif and gene structure. These IbXTHs genes showed different expression patterns in different tissues and six treatments. For example, IbXTH17 exhibited the highest expression level in the initiative storage root (ISR), while IbXTH32 expressed highest in the stem. IbXTH20 and IbXTH21 showed the highest expression in the root stock (RS), but lowest in the fibrous root (FR). The expression patterns of these collinear gene pairs implied that the sweet potato XTH gene family might have shown new function during the subsequent evolution. With that considered, the discovery of the IbXTH gene family lays the foundations and offers important insights for further study of the sweet potato XTH gene family.

IbXTH02 and IbXTH12 Gene Expression in Abiotic Stress
More and more evidences point to the role of XTHs as major players in modifying how plants react to various harmful environmental stimuli. In response to drought stress, previous studies have shown that a plant would close the stomates first to prevent water evaporation, and XTH genes are essential for modifying the cell wall's elongation and improving drought resistance [25]. For instance, virus-induced gene silencing of HvXTH1 in barley demonstrated that BSMV: HvXTH1 plants showed an increase in shoot fresh weight and a reduction in water loss compared with BSMV:γ plants under drought stress [26]. The AtXTH18 gene was silenced by the RNAi method, and the elongation of the first rooted epidermal cells was inhibited. XTH3 from Capsicum annuum enhanced drought resistance of transgenic Arabidopsis thaliana and S. lycopersicum when the seedlings of which were exposed to extreme drought stress [27]. Moreover, it is notable that the existence of xylglucan in early terrestrial plants implies that members of the XTH gene family played a crucial role in the transition from wetter to drier habitats [28]. To better understand the possible function of IbXTH genes under PEG-induced drought stress, we also analyzed in detail the expression of XTH genes in sweet potato. The results showed that the IbXTH02 and IbXTH12 expressed were down-regulated in FR.
Under salt stress, a previous studies reported that MtXTH3 in Medicago truncatula was strongly up-regulated at higher NaCl concentration [29]. In our research, the NaCl-induced salt treatment and RT-qPCR analysis showed that IbXTH02 and IbXTH12 were significantly down-regulated, suggesting that the IbXTH02 and IbXTH12 might have similar functions. Taken together, these findings provide novel information about IbXTHs under abiotic stress, especially drought and salt stresses. It can be speculated that the above genes may show increased cell wall-related functions under stress by combining with xyloglucan. Nevertheless, further molecular and genetic identification efforts are needed to verify their functions.

Chromosomal Localization of XTH Genes
The positions of IbXTHs on chromosomes were obtained from the sweet potato genome annotation information, which was obtained from the Ipomoea genome hub (https:// sweetpotao.com/ [31] (accessed on 10 September 2022), and chromosome mapping was performed and visualized by using TBtools [32].

Conserved Motif and Gene Structure Analyses
The MEME online tools (http://meme-suite.org/ [33] (accessed on 18 September 2022)) were used to identify the conserved motif, with ten discovery motif numbers and other default parameters. The gff3 file of the I. batatas genome contains information about the gene structure, which can be used to identify the exons and introns of XTH. The conserved motif and gene structure can be visualized by TBtools [32].

Phylogenetic Tree Construction
The XTH protein sequences of A. thaliana (At) and I. batatas were aligned by using Clustal Musle [38], with the default parameters used to generate a neighbor-joining (NJ) phylogenetic tree [39] using MEGA 7.0 software [40]. The tree was constructed with 1000 replicates bootstrap analysis, the Jones-Taylor-Thornton (JTT) model, and the missing data treatment option set at partial deletion and beautified by Chiplot (https://www. chiplot.online/ (accessed on 30 September 2022)) online tools. The XTH gene subfamily of I. batatas was categorized based on the A. thaliana XTH gene subfamily.

Collinearity Analysis and Ka/Ks Calculation
The duplication types and intraspecific covariance of XTH family members were analyzed using the One Step MCSanX-Super Fast function in TBtools [32]. The Advanced Circle function in TBtools [32] was adopted to plot the collinearity relationship of XTH genes, and the collinearity relationship between I. batatas and three other species, including G. max, O. sativa, and A thaliana, and the genome sequences and general feature formats of which were derived from Ensemble Plants (https://plants.ensembl.org/index.html [41] accessed on 20 September 2022).

Cis-Acting Elements Analyses
The 2000 bp promoter sequences of the sweet potato XTH gene were submitted to the PlantCare website (http://bioinformatics.psb.ugent.be/webtools/plantcare/html/ [42] (accessed on 25 September 2022)) for cis-acting regulation element prediction based on the I. batatas genome database file. The results of the PlantCare analysis were simplified and then visualized by using TBtools [32].

Expression Analysis of XTH Genes
To explore the expression patterns of sweet potato XTH genes, the RNA-seq expression profiles of the XTH genes were mined from the sequence read archive (SRA) of NCBI under the project of PRJNA511028, which was a transcriptome analysis of the different tissues and stress and hormone treatments to the fibrous root (FR) of I. batatas 'Xushu18'. The heat maps of the expression levels of the XTH genes were visualized using the Heatmap illustrator program in the toolkit of TBtools [32].

Plant Materials and Abiotic Stresses
The cultivated sweet potato variety 'Xushu18' was used in this study. The seedlings of this variety were collected from Zhejiang A&F University, China. The two cultivars' uniform seedlings were raised and naturalized in the Hoagland solution at 26 • C with a 16/8 photoperiod for 3 days. The seedlings were treated with 0.2 mol/L NaCl and 1/2 Hoagland solution containing 20% PEG6000 mass/volume fraction, respectively, to simulate drought and salt abiotic stress. The fibrous root collected after 0 (CK), 3, 6, 12, 24, and 48 h.
Total RNA from all the collected fibrous root was extracted using SteadyPure Plant RNA Extraction Kit (, Accurate Biotechnology, Hunan, China.) in accordance with the manufacturer's instructions to validate the RNA-seq data. Reverse transcription was performed on 1 µg of RNA from each sample using the Evo M-MLV RT Mix Kit with gDNA Clean for qPCR (Accurate Biotechnology, Hunan, China). The qRT-PCR assay was conducted by a CFX Connect Real-Time System (Bio-Rad, Veenendaal, UT, USA) using the SYBR Green Premix Pro Taq HS qPCR Kit Accurate Biotechnology, Hunan, China). The specific primer of gene IbXTH02 and IbXTH12 for qRT-PCR detection are designed by Primer6 software listed in the Table S4. The sweet potato β-Actin gene (GenBank, AY905538) was applied as the internal control. The experiments were conducted for three replicates for each gene, and the 2 −∆∆CT method was used to calculate the results [33]. We analyzed the data and compared the means using LS at a 0.01 level of significance.

Conclusions
We identified 36 IbXTH genes and then analyzed their physicochemical properties. Based on the phylogenetic analysis, the IbXTHs were divided into three subfamilies. We also investigated their promoter regions and collinearity relationships with three species. In addition, we systematically investigated the expression profiles in different tissues, and different development stages of storage roots, as well as the expression of the IbXTHs under NaCl-indued salt stress and PEG-induced drought treatment. The IbXTH-mediated stressresponse mechanism in sweet potato can be furtherresearched due to the diversification of the IbXTH genes.