Genome-Wide Identification of Common Bean PvLTP Family Genes and Expression Profiling Analysis in Response to Drought Stress

Common bean is one of the most important legume crops for human consumption. Its yield is adversely affected by environmental stress. Plant non-specific lipid transfer proteins (nsLTPs) are essential for plant growth, development, and resistance to abiotic stress, such as salt, drought, and alkali. However, changes in nsLTP family genes responding to drought stress are less known. The PvLTP gene family in the common bean was identified by a comprehensive genome-wide analysis. Molecular weights, theoretical isoelectric points, phylogenetic tree, conserved motifs, gene structures, gene duplications, chromosome localization, and expression profiles were analyzed by SignalP 5.0, ExPASy, ClustalX 2.1, MEGA 7.0, NCBI-CDD, MEME, Weblogo, and TBtools 1.09876, respectively. Heatmap and qRT-PCR analyses were performed to validate the expression profiles of PvLTP genes in different organs. In addition, the expression patterns of nine PvLTP genes in common beans treated with drought stress were investigated by qRT-PCR. We obtained 58 putative PvLTP genes in the common bean genome via genome-wide analyses. Based on the diversity of the eight-cysteine motif (ECM), these genes were categorized into five types (I, II, IV, V, and VIII). The signal peptides of the PvLTP precursors were predicted to be from 16 to 42 amino acid residues. PvLTPs had a predicated theoretical isoelectric point of 3.94–10.34 and a molecular weight of 7.15–12.17 kDa. The phylogenetic analysis showed that PvLTPs were closer to AtLTPs than OsLTPs. Conserved motif and gene structure analyses indicated that PvLTPs were randomly distributed on all chromosomes except chromosome 9. In addition, 23 tandem duplicates of PvLTP genes were arranged in 10 gene clusters on chromosomes 1 and 2. The heatmap and qRT-PCR showed that PvLTP expression significantly varied in different tissues. Moreover, 9 PvLTP genes were up-regulated under drought treatment. Our results reveal that PvLTPs play potentially vital roles in plants and provide a comprehensive reference for studies on PvLTP genes and a theoretical basis for further analysis of regulatory mechanisms influencing drought tolerance in the common bean.


Introduction
The common bean (Phaseolus vulgaris L.) is an important legume used as food because it is a major source of proteins, minerals, and vitamins [1]. Thus, it is consumed as part of the traditional diets in Europe (e.g., the Mediterranean region) and the Middle East [2]. than that of drought-sensitive lines. The heterologous expression of TdLTP4 in Arabidopsis led to tolerance to abiotic and biotic treatments [33]. Because of their association with stress resistance-related transcription factors, researchers believe that nsLTP can enhance stress resistance in plants.
Until now, only four nsLTP genes have been found in the common bean. Among these, the expression of PvLTP24 increased significantly in stems and leaves following drought and ABA treatments [34]. The apical cortex specifically expressed PVR3 in the root tip, which could be used as a marker gene for cortex development [35,36]. Additionally, PvLTP1a and PvLTP1b are the two major food allergens, while the functions of the other common bean nsLTP genes are still unclear [37]. We utilized bioinformatics methods to identify the common bean's PvLTP gene family. Chemical and physical properties, as well as a phylogenetic tree, conserved motifs, gene structure, and chromosome location, were analyzed. Moreover, a qRT-PCR assay was used to analyze the expression levels of different PvLTPs in organs and drought stress. Our findings will serve as a foundation for gene function and mechanism studies of the common bean PvLTPs in response to drought stress.

Identification and Bioinformatic Analyses of PvLTP Family in Common Bean
The genome and proteome sequences of common bean, Oryza sativa, and Arabidopsis thaliana were downloaded from Phytozome (http://phytozome.jgi.doe.gov/, accessed on 1 March 2021), RAP (https://rapdb.dna.affrc.go.jp/, accessed on 1 March 2021), and TAIR (https://www.arabidopsis.org/, accessed on 1 March 2021) databases, respectively [38][39][40]. The NsLTP sequence ID lists of Arabidopsis and rice were obtained from Boutrot et al. [19]. Candidate nsLTP genes of the common bean were obtained using BLASTp with AtLTP and OsLTP sequences as queries to blast against the common bean protein database with an E-value of 1 × 10 −5 . Proteins with the HMM domain PF00234 were further retrieved using HMMER 3.3 with default parameters to avoid missing nsLTP genes. After eliminating redundant sequences, all proteins of putative nsLTP genes lacking the essential ECM domain were removed manually. Subsequently, the signal peptide cleavage sites were analyzed using Signal 5.0 (https://services.healthtech.dtu.dk/service.php?SignalP-5.0, accessed on 5 March 2021). Proteins without N-terminal signal sequences and the deduced hybrid proline-rich proteins were discarded. Moreover, the rest of the candidate sequences were subjected to BLASTp to exclude potential α-amylase inhibitors and cereal storage proteins using RATI and 2S-albumin as queries, respectively [41,42]. The remaining predicted proteins were uploaded to the NCBI-CDD (https://www.ncbi.nlm.nih.gov/cdd, accessed on 7 March 2021) and Pfam (http://pfam.xfam.org/, accessed on 7 March 2021) websites to identify the conserved LTP domain. Finally, the mature proteins containing more than 120 amino acid residues were eliminated. The identified PvLTPs were named according to Boutrot's method and their orders on chromosomes [19].

Sequence Alignment and Phylogenetic Analysis
The ECM domains of PvLTPs from Arabidopsis, rice, and the common bean were subjected to multiple alignments by ClustalX 2.1 software with default parameters [43]. After that, they were utilized to build a phylogenetic tree based on the Neighbor-Joining model with MEGA 7.0 software [44,45]

Conserved Motifs and Gene Structure Analysis
To identify conserved motifs, PvLTPs were uploaded to the MEME (https://memesuite.org/meme/tools/meme, accessed on 10 March 2021) website with the motif width setting as 6-50 amino acid residues and the maximum motif number setting as 10 [46]. Gene structures were analyzed and visualized using TBtools 1.09876 [47]. The WebLogo

Chromosomal Localization and Gene Duplication
The common bean v2.1 gff3 file was retrieved from the Phytozome website [38,49]. The relative distances and positions of PvLTPs were obtained according to the genome annotation information and drafted on the 11 chromosomes using TBtools 1.09876 [47]. The duplication events of PvLTP gene family members in the common bean were identified using TBtools 1.09876 with the default parameters.

Plant Materials
Common bean cultivars (Pinjinyun 3) were planted in the field at the Center for Agricultural Genetic Resources Research at Shanxi Agricultural University (China, Shanxi, Latitude 112 • 58 E, 37 • 78 N, Figure S1). The roots, stems, leaves, flowers, seeds, and pods were collected and fast-frozen in liquid nitrogen before being preserved at −80 • C. Moreover, the cultivar seeds were cultivated at 25 • C with 14 h of light and 20 • C with 10 h of darkness in a growth chamber ( Figure S2). The seedlings were treated with 20% PEG6000 for 24 h at the trifoliate leaves stage. The leaves were collected after 0, 6, 12, and 24 h under drought treatment, and then samples were stored at −80 • C before RNA extraction. For each organ, three separate biological replicates and technical replicates were performed.

Quantitative Real-Time PCR Analysis
Total RNA was extracted from roots, stems, flowers, leaves, seeds, and pods using SV Total RNA Isolation System (Promega, Madison, WI, USA) as directed by the manufacturer's instruction. Precisely 1 µg of total RNA was reverse-transcribed by PrimeScript RT Master Mix (TaKaRa, Otsu, Japan). Then, the cDNA was amplified on a Quantstudio 6 thermal cycler (Applied Biosystems, Waltham, MA, USA) using TB Green Premix Ex Taq II (TaKaRa, Otsu, Japan). The qRT-PCR amplification procedure was 95 • C for 30 s, performed with 40 cycles of 95 • C for 5 s and 60 • C for 34 s. We used the common bean Actin gene as a reference control. For each experiment, three separate technical and biological replicates were performed. The 2 −∆∆Ct analysis method was used to calculate the relative transcription levels (Table S1) [50]. All primers are listed in Table S2.

Genome-Wide Identification of Putative nsLTPs in Common Bean
To investigate all putative nsLTPs in the common bean, we performed BLASTp using AtLTP and OsLTP sequences as queries to search against the common bean proteome database. A total of 73 candidate nsLTP genes were obtained. Another ten nsLTP genes were retrieved from the HMM domain profiles of PF00234 and PF14368. After integrating the above results, three proteins lacking the essential ECM domain were manually removed. In addition, six proteins without N-terminal signal sequences were identified using Signal 5.0 and were deleted. Five hybrid proline-rich proteins were not taken into consideration. Proteins with high similarity to α-amylase inhibitors or cereal storage proteins were not found. The remaining 69 candidate protein sequences were uploaded to the Pfam and NCBI-CDD websites to identify the conserved LTP domains. All candidates possessed the LTP domains. Because all mature nsLTPs have low molecular weights, eleven of the sixty-nine candidates with more than 120 amino acid residues were discarded. As a result, 58 genes coding nsLTPs were designated as PvLTPs.

Phylogenetic Analysis of PvLTPs
To demonstrate the evolutionary relationship of nsLTP genes among the common bean, Arabidopsis, and rice, phylogenetic analysis was completed using ClustalX 2.1 and MAGA 7.0 software. As shown in Figure 1, the 152 nsLTP genes were divided into nine distinct clades, with 77, 33,5,14,9,8,1,3, and 2 nsLTP genes from types I-IX in each clade, respectively. Obviously, the largest portions of common bean PvLTPs and Arabidopsis AtLTPs were clustered together in clade I, whereas rice OsLTPs were mainly gathered in clade II. In addition, PvLTPs were closer to AtLTPs than OsLTPs in each clade of the phylogenetic tree. The cluster separation revealed a closer genetic lineage-specific relationship between the common bean and Arabidopsis during dicotyledon evolution.

Conserved Motifs and Structure of PvLTPs
CDD and Weblogo websites showed the conserved domains in the PvLTP gene family ( Figure 2). PvLTP family members contained an ECM domain, which belonged to the α-amylase inhibitors and lipid transfer/seed storage proteins (AAI-LTSS) superfamily. A total of ten distinct conserved motifs (motif1-motif10) were identified using MEME tools ( Figure 3) to analyze the relationship between the structural features and diversity of   "X" indicates any amino acid, and the Arabic digits following "X" indicate the number of amino acids.

Conserved Motifs and Structure of PvLTPs
CDD and Weblogo websites showed the conserved domains in the PvLTP gene family (Figure 2). PvLTP family members contained an ECM domain, which belonged to the α-amylase inhibitors and lipid transfer/seed storage proteins (AAI-LTSS) superfamily. A total of ten distinct conserved motifs (motif1-motif10) were identified using MEME tools ( Figure 3) to analyze the relationship between the structural features and diversity of PvLTP members further. All PvLTP family members contained motif3 with highly conserved amino acid sequences. Moreover, motif1 and motif4 were presented in type I PvLTPs (except PvLTPI.6), while motif7 and motif10 were presented in type II PvLTPs. The results revealed that motif3 was conserved within the five PvLTP types.  The amount and distribution of exons and introns in PvLTPs were examined to investigate the components of PvLTP gene structure. Among the 58 PvLTP genes, 29 contained one intron and two exons, including 25 type I, one type IV, two type V, and one type VIII PvLTPs. The other 29 had no introns. Interestingly, all type II PvLTP genes had no introns, consistent with the findings in cotton and potato. These findings indicated that the PvLTP gene family had been conserved in plant evolution.

Chromosomal Localization and Gene Duplication of PvLTPs
To better obtain the distribution and exact position of PvLTPs on chromosomes, a detailed chromosome map was constructed. The 58 PvLTP genes were unevenly dispersed on ten chromosomes ( Figure 4). Among them, twenty, seventeen, five, four, four, three, one, one, and one PvLTPs were located on chromosomes 8, 6, 10, 1, 7, 4, 2, 3, and 11, respectively, while there was no PvLTP on chromosome 9.
Gene duplication analysis showed that, among the 58 PvLTPs, 23 genes were arranged in ten gene tandem duplicate clusters and distributed on two chromosomes. Among the ten gene clusters, nine clusters contained 21 genes and were located on chromosome 1, and the other cluster was located on chromosome 2. The large proportion (39.66%) of tandem duplicate clusters indicated that tandem duplications might cause PvLTP gene family expansion in the common bean genome.

Expression Pattern of PvLTPs in Different Organs
The expression profile of PvLTPs in different organs was visualized in a heatmap using RNA-seq data from Phyzotome to investigate the functions of PvLTPs during common bean growth. Figure 5 showed that 19 PvLTPI genes were highly expressed in root nodules but lower in other tissues. Moreover, the rest of the PvLTP genes showed a diverse expression profile in different tissues.   To further validate the above results, the expression of nine PvLTPs in roots, stems, leaves, flowers, seeds, and pods was detected using qRT-PCR ( Figure 6). PvLTPI.27, PvLTPI.28, and PvLTPI.42 were highly expressed in flowers, while PvLTPI.44 and PvLTPV.1 were highly expressed in seeds, consistent with the results shown in the heatmap. In addition, PvLTPII.3 and PvLTPV.2 were predominantly expressed in stems, while PvLTPI.45 was strongly expressed in pods. These findings suggested that the differentially expressed PvLTPs might play various biological functions in the growth and development of the common bean.

Chromosomal Localization and Gene Duplication of PvLTPs
To better obtain the distribution and exact position of PvLTPs on chromosomes, a detailed chromosome map was constructed. The 58 PvLTP genes were unevenly dispersed on ten chromosomes (Figure 4). Among them, twenty, seventeen, five, four, four, three, one, one, and one PvLTPs were located on chromosomes 8, 6, 10, 1, 7, 4, 2, 3, and 11, respectively, while there was no PvLTP on chromosome 9. Gene duplication analysis showed that, among the 58 PvLTPs, 23 genes were arranged in ten gene tandem duplicate clusters and distributed on two chromosomes.

Expression Analysis of PvLTPs under Drought Stress
Abiotic environmental conditions have direct impacts on plant growth. Therefore, it is significant to analyze how relatively resistant genes change under different stresses. The expression levels of 9 PvLTPs after drought treatment were examined using qRT-PCR to explore the functions of PvLTPs involved in drought stress. The results showed that all PvLTPs were up-regulated after drought stress (Figure 7). The relative expression levels of PvLTPI.28, PvLTPI.41, PvLTPI.42, PvLTPI.44, PvLTPI.45, and PvLTPII.3 showed a similar trend of first increasing and then decreasing, reaching the maximum at 12 h. PvLTPI.44 especially exhibited more than 115-fold up-regulation. Moreover, the relative expression levels of the other three PvLTPs, PvLTPI.27, PvLTPV.1, and PvLTPV.2 showed a trend of gradual increase and reached the highest of 37-, 78-, and 142-fold up-regulation at 24 h compared with the control, respectively. These findings displayed that PvLTPs might function as important positive regulators influencing plant tolerance towards drought stress.
(39.66%) of tandem duplicate clusters indicated that tandem duplications might cause PvLTP gene family expansion in the common bean genome.

Expression Pattern of PvLTPs in Different Organs
The expression profile of PvLTPs in different organs was visualized in a heatmap using RNA-seq data from Phyzotome to investigate the functions of PvLTPs during common bean growth. Figure 5 showed that 19 PvLTPI genes were highly expressed in root nodules but lower in other tissues. Moreover, the rest of the PvLTP genes showed a diverse expression profile in different tissues. To further validate the above results, the expression of nine PvLTPs in roots, stems, leaves, flowers, seeds, and pods was detected using qRT-PCR ( Figure 6). PvLTPI.27, PvLTPI.28, and PvLTPI.42 were highly expressed in flowers, while PvLTPI. 44 and PvLTPV.1 were highly expressed in seeds, consistent with the results shown in the heatmap. In addition, PvLTPII.3 and PvLTPV.2 were predominantly expressed in stems, while PvLTPI.45 was strongly expressed in pods. These findings suggested that the differentially expressed PvLTPs might play various biological functions in the growth and development of the common bean.

Expression Analysis of PvLTPs under Drought Stress
Abiotic environmental conditions have direct impacts on plant growth. Therefore, it is significant to analyze how relatively resistant genes change under different stresses. The expression levels of 9 PvLTPs after drought treatment were examined using qRT-PCR to explore the functions of PvLTPs involved in drought stress. The results showed that all PvLTPs were up-regulated after drought stress (Figure 7). The relative expression levels of PvLTPI. 28,PvLTPI.41,PvLTPI.42,PvLTPI.44,PvLTPI.45, and PvLTPII.3 showed a similar trend of first increasing and then decreasing, reaching the maximum at 12 h. PvLTPI.44 especially exhibited more than 115-fold up-regulation. Moreover, the relative expression levels of the other three PvLTPs, PvLTPI.27, PvLTPV.1, and PvLTPV.2 showed a trend of gradual increase and reached the highest of 37-, 78-, and 142-fold up-regulation at 24 h compared with the control, respectively. These findings displayed that PvLTPs might function as important positive regulators influencing plant tolerance towards drought stress.

Discussion
Plants have different types and numbers of the nsLTP gene family. Brassica rapa has nine types (I-VI, VIII, IX, and XI) with 63 members, while Six Solanaceae species have five types (I, II, IV, IX, and X) with 135 members [20,21]. In this work, whole-genome identification of the common bean obtained 58 PvLTP family members unevenly distributed across ten chromosomes, except for chromosome 9. We found that these PvLTP family members could be categorized into five types (I, II, IV, V, and VIII). Nevertheless, the ab-

Discussion
Plants have different types and numbers of the nsLTP gene family. Brassica rapa has nine types (I-VI, VIII, IX, and XI) with 63 members, while Six Solanaceae species have five types (I, II, IV, IX, and X) with 135 members [20,21]. In this work, whole-genome identification of the common bean obtained 58 PvLTP family members unevenly distributed across ten chromosomes, except for chromosome 9. We found that these PvLTP family members could be categorized into five types (I, II, IV, V, and VIII). Nevertheless, the absence of some types indicated deletion during the evolution of the common bean. The phylogenetic tree results showed that the common bean PvLTPs were closer to Arabidopsis AtLTPs than rice OsLTPs. The relationship distance between the dicotyledons and monocotyledons in the plants' classification was consistent with the previous studies [51], indicating that the nsLTP genes existed before the separation of monocotyledonous plants and were conserved during evolution. The internal structures of types I and II had similarities, suggesting that the same category of genes may have similar biological functions.
Studying gene expression patterns can provide a theoretical basis for gene functions. Results have shown that the nsLTP gene responded to drought and other abiotic stresses. In Arabidopsis, LTP3 was highly expressed after 6 h of drought treatment. AtLTP3 overexpression significantly enhanced tolerance to drought stress. Further studies showed that the upstream transcription factor, MYB96, regulated the expression of AtLTP3 by binding to the AtLTP3 promoter, thus enabling AtLTP3 to participate in the drought resistance signaling pathway [26]. A comparison between AtLTP3 and PvLTPI.45 revealed that their amino acid sequences are approximately 41.53% similar. Additionally, AtLTP3 was expressed at high levels in the leaves, flowers, and siliques, with peak levels detected in the siliques. The PvLTPI.45 expression pattern was similar to that of AtLTP3, with the highest expression level detected in the pods, followed by the flowers and leaves. Under drought conditions, the AtLTP3 and PvLTPI.45 expression levels increased by about 4-fold and then decreased, suggesting they may have similar functions in plant responses to drought stress. Several findings indicated that the expression levels of nsLTP genes, including OsDIL (rice), ScLTP (sugarcane), and NtLTP4 (tobacco), were up-regulated or down-regulated to varying degrees after drought, salt, and other abiotic stress treatments [31,52,53]. For example, StnsLTP1 expression in potatoes increases in response to excessive heat, salt, and drought. The overexpression of StnsLTP1 in potatoes led to enhanced tolerance to abiotic stresses because it activated antioxidative defense mechanisms and up-regulated the expression of stress-related genes [54]. These observations imply that nsLTP proteins are involved in multiple stress-induced signaling pathways. In this research, the expression levels of PvLTPs were increased to varying degrees after the drought stress treatment. For instance, PvLTPI.44 and PvLTPV.2 increased 115 and 142 times, respectively, compared with the control. Therefore, it can be speculated that PvLTPs play a vital function in drought stress response.
Gene tandem duplication during genomic DNA replication and recombination is the key driver for gene family amplifications [55][56][57]. In Arabidopsis and rice genomes, 15-20% of genes were composed of tandem repeats of gene clusters considered crucial for evolution, plant disease resistance, and abiotic stress responses [58]. Several tandem repeats of genes were found in the nsLTP gene family of angiosperms, including 47.82% (66/138) in cotton, 51.43% (36/70) in barley, and 53.01% (44/83) in potato [59][60][61]. This indicated that tandem repeats played essential roles in gene amplification during the evolution of the nsLTP gene family. In this work, chromosome mapping and gene structure revealed that the gene replication events occurred during genome amplification and evolution of the common bean. Twenty-three genes of the PvLTP gene family were categorized into 10 tandem repeats of gene clusters on chromosomes 1 and 2, accounting for 39.66% of the family members. Most of the genes exhibiting the same tandem repeats had a high similarity, suggesting that they might have functional similarities. For instance, the nucleic acid sequence similarity of PvLTPI.12/PvLTPI.13/PvLTPI.14/PvLTPI.15 was 98.62%, and the protein similarity was 96.71%. However, the nucleic acid sequence similarity of PvLTPI.41/PvLTPI.42 was only 52.18%, with a protein similarity of 58.47%. This implied that, although the PvLTP family was derived from the same ancestor, it changed when plants adapted to the external environment during evolution. Moreover, we found that PvLTPI.42 was more highly expressed in flowers than in other tissues, while the PvLTPI.41 expression was higher in the tissues except for roots. The different expression patterns suggest that the evolution of duplicate genes might lead to the emergence of functional diversity.

Conclusions
In summary, whole-genome identification of the common bean identified 58 members of the PvLTP gene family, which were divided into five types (I, II, IV, V, and VIII). Each member contained the conserved ECM domain. PvLTP genes were randomly distributed on ten chromosomes, except for chromosome 9, of which 23 members formed ten sets of tandemly repeating gene clusters. The organ-specific expression profiles of nine PvLTP genes differed significantly. In addition, exposure to drought stress up-regulated the expression of these genes. These findings provide important insights into the PvLTP genes in the common bean and form the theoretical basis of future research on PvLTP functions related to plant development and stress responses. Furthermore, our study will contribute to investigating the mechanisms underlying the effects of nsLTPs on plant responses to abiotic stresses.

Supplementary Materials:
The following supporting information can be downloaded at: https: //www.mdpi.com/article/10.3390/genes13122394/s1, Table S1: The Ct value of qRT-PCR used in this study. Table S2: qRT-PCR primers of PvLTP genes used in this study. Figure S1: The geo-tagged photograph of the common bean cultivar in the field. Figure S2: The geo-tagged photograph of the seedlings grown in the growth chamber.