Abstract Nutrient boron (B) is important for crop development. The absorption and transport of B ions are regulated by the B transport (BOR) gene family. Although some members of the BOR gene family have been heavily researched, to the best of our knowledge, no comprehensive research on the structural, evolutionary and functional relationships among crops has been reported. In this study, comparative genomic research of the BOR gene family was performed in five crops, 40 BOR genes were identified, and analyses of phylogenetics, structure, conserved motifs, and transmembrane topology were performed. These genes had highly similar physicochemical properties, structure and motif distribution. Specific elements in the C-terminus and functional differences were also found among the genes. In addition, microsynteny and evolutionary analysis suggested that large-scale replication events and purifying selection played essential roles during the OsBOR gene evolutionary process in rice. Moreover, expression pattern of OsBORs were also analyzed. Our research provides comprehensive and detailed information on BOR genes in five crops and establishes a foundation for further functional exploration of these genes in B metabolic pathways.
　　Key words Boron transport; Functional differentiation; Crops; Genome-wide analysis
　　Received: January 10, 2021  Accepted: March 17, 2021
　　Supported by National Science Foundation of Anhui Province (grant number 1908085QC135); Anhui Science and Technology Department, International Science & Technology Cooperation Plan (grant number 1804b06020344); Anhui Science and Technology Department; Major Science and Technology Projects (grant number 201903a06020012).
　　Quan GAN (1990-), male，P. R. China,, assistant scientist, devoted to research about molecular breeding of rice.
　　*Correspondence: E-mail: dahuni1974@163.com.
　　Introduction
　　Nutrient boron (B) plays an important role in both vegetative and generative development for crops［1-2］. As a receptor, B-mediated rhamnogalacturonan-II (RG-II) dimer formation is essential for the maintenance of cellular structure to shoot development［3］, root elongation［4］, pollen sprout and growth［5-6］. In contrast to other indispensable mineral nutrients, most crops have a relatively strict concentration demand for boron in soil［7］. On the one hand, excess B will over accumulate reactive oxygen species (ROS) and generate oxidative damage in the cell membrane［8］. On the other hand, B deficiency restrains the development of essential tissues in plants, such as roots, stems and leaves［9-10］. In addition, reproductive development is also affected by B deficiency, leading to fewer pods, dried floral buds and low seed yield［11］. Thus, it is crucial for crops to regulate the B concentration in different tissues during development, which is driven by the B transport (BOR) gene family［12］. 　　Some genes participating in the B transportation process have been identified as necessary for plant development. The B transport gene was first identified in Helianthus annuus［13］. Subsequently, the Arabidopsis gene AtBOR1 was found to prevent B deficiency［14-15］. Notably, when environmental B is in excess, AtBOR1 activity is reduced to prevent B toxicity through ubiquitination-mediated vacuolar trafficking［16-17］. BOR genes have also been identified in other plants, including Oryza sativa (rice)［18］, Glycine soja (soybean)［19］, Zea mays (maize)［20］, Brassica napus［21］ and Gossypium hirsutum (cotton)［22］. In rice, four BOR genes were identified, and OsBOR1 was predicted to encode 711 amino acids and contains 10 transmembrane domains, as does AtBOR1. Primary phylogenetic analysis between rice and Arabidopsis suggested that OsBOR1 was most similar to AtBOR1, and OsBOR2, -3, and -4 were more distantly related to AtBOR1 than OsBOR1, implying functional differentiation［18］. These studies revealed that the BOR gene family is pervasively involved in crops to regulate B nutrition assimilation and transportation, even though most BOR-specific functions and differences among family numbers have not been determined.
　　Moreover, although the BOR genes in some model plants, such as Arabidopsis and rice, have been identified, the specific structure, functional domain information, and molecular evolution mechanisms of these genes have not been elucidated. Therefore, in this research, we analyzed the BOR genes from Arabidopsis and five crops, including rice, soybean, maize, B. napus and cotton. The gene locations on chromosomes, phylogenetics, exon/intron structure, motif distribution, topological structure of the transmembrane domain and microsynteny of the BOR genes in these species were analyzed. Specifically, the environmental selection and expression patterns of OsBORs was further investigated in rice. Our research provides comprehensive and detailed information on the BOR genes in five crops and establishes a foundation for further functional exploration of these genes in boron metabolic pathways.
　　Materials and Methods
　　Identification and characterization of BORs
　　In this research, the genome parameters of six plants were obtained from their genome database: rice from the RAP, Arabidopsis from the TAIR, soybean from the SoyBase, maize from the MaizeGDB, B. napus from the Genoscope and cotton from the Cottongen. The Arabidopsis B transport proteins were used as normative sequences to identify BOR protein sequences in these species by the BLASTP program (e-value<10-10). The physicochemical characteristics for each BOR protein were calculated by the ExPASy online tool. Protparams were used to calculate the GRAVY (grand average of hydropathicity) values［23］. Physical location was visualized using MapInspect software［24］. 　　Phylogenetic analysis
　　The multiple alignment of BOR proteins in six plants (Arabidopsis, rice, soybean, maize, B. napus and cotton) was performed by CLUSTAL_X software［25］. Then, the data were transformed to a phylogenetic tree with the neighbor-joining (NJ) method by MEGA 5.5 software. To construct parameters, 1 000 was set for bootstrap value, and pairwise deletion and poison correction were also selected［26］.
　　Structure analysis and motif identification
　　According to the DNA sequences and CDS sequences of the BOR gene family, the exon-intron structure was generated through the GSDS program. Sequences and motifs of all BOR proteins were analyzed using the MEME website. The motif width was set from 6 to 200, and the maximum number was set to 20. Additionally, annotation of the structural motifs was performed by SMART databases［27］.
　　Intraspecies microsynteny analysis
　　All of the BOR genes in Arabidopsis, rice, soybean, B. napus and cotton were set as the initial anchor points based on their chromosomal locations. Next, in the range of 100 kb upstream and downstream, every flanking gene was compared to others to identify the homology relationship through the BLASTP program (e-value<10-10)［28］. If more than four homologous gene pairs were identified, this region would be considered to result from the large-scale duplication event［29］.
　　Environmental selection analysis of BOR genes
　　The nucleotide coding sequences from duplicated pairs were aligned by ClustalX［30］. Then, the nonsynonymous (Ka) and synonymous (Ks) substitution rates of the gene pairs in the duplication blocks were calculated through the DnaSP program［31-32］. Furthermore, the Ka/Ks ratios of OsBORs were calculated to evaluate the selective pressure experienced through the sliding window method, and the window-step sizes were set to 150-9 bp［23］.
　　Results
　　Identification and distribution of the BOR gene family
　　For the identification of boron transport genes, BLASTP searches were performed according to Arabidopsis BOR proteins in five crops. As a result, 4, 10, 4, 6 and 9 classic BOR genes were identified in rice, soybean, maize, B. napus and cotton. The BOR genes that were not reported previously were named according to the sequence similarity compared to AtBORs. The detailed parameters of these genes are listed in Table 1, including their names, descriptions and physicochemical parameters. To encode amino acid length, BOR proteins exhibited relatively small variations that varied from 652 aa (GsBOR2c) to 737 aa (BnBOR3s) and putative protein weights ranging from 73.03 to 81.43 kDa, except ZmBOR4 (1 023 aa, 112.92 kDa). For isoelectric points (pIs), 40 BOR proteins showed relatively high parameters and ranged from 6.24 to 9.26. For GRAVY values, almost all BOR proteins were hydrophilic and ranged from 0.094 to 0.266, which was in keeping with the specificity of membrane proteins to transport ions. 　　Next, according to genomic annotation information, the distributions of 40 BOR genes on their chromosomes were identified (Fig. 1). Relatively evenly distributed chromosomes were discovered in each of six plants. Except for several pairs of BOR genes, most of them were distributed separately on a single chromosome. In particular, there was a distance of only 10 kb between OsBOR2 and OsBOR3, which indicated that they were tandem duplications. Our results suggested that the members of the BOR gene family had similar protein sizes, stable physicochemical characteristics and a balanced chromosomal distribution, which implied their conserved function in crops.
　　Phylogenetic analysis
　　To investigate the phylogenetic relationship among 40 BOR proteins in six plants, a phylogenetic tree was constructed using the NJ method (Fig. 2). ME and ML phylogenetics were also generated to validate the topologies (Fig. S1 and S2). As a result, all the members of the BOR gene family were grouped into four subfamilies, named I, II, III and IV. In general, subfamily I had the maximal 20 BOR members, subfamily II contained the minimal two BOR members, subfamily III had five BOR members and subfamily IV had 13 BOR members. Each of the six plants included at least one BOR gene member in subfamily I, whereas the members of subfamilies II and III included only two species (Arabidopsis, B. napus and rice, maize). According to these results, we concluded that this phylogenetic relationship might be induced by the gene expansion event (obtained or lost) in the BOR gene family evolution process. In addition, ten pairs of paralogous genes were found among the members of the BOR gene family in the phylogenetic tree. For orthologous genes, six pairs of BOR genes were shared by Arabidopsis and B. napus, and one pair of BOR genes was found in soybean and cotton. No orthologous BOR genes were found in other species (Fig. 2). These results suggested that most members of the BOR gene family appeared as paralogous gene pairs and that BnBORs might be functionally similar to AtBORs as B transporters.
　　Structure, conserved motif and topologies analysis
　　Previous studies have suggested that gene structure diversity is an important resource for the evolution of multigene families［33］. To elucidate the structural diversity of the BOR genes in these species, schematic diagrams of exons and introns were deduced. As shown in Fig. 3a, different numbers of exons were found in 40 BOR genes, varying from 10 to 18. Specifically, 22 BOR genes had 12 exons, 12 BOR genes had 13, two genes had 10 exons and three genes had 14. For ZmBOR4, 18 exons and 17 introns were detected because of the large genomic component of this gene. Subsequently, an analysis of BOR paralogous and orthologous gene pair structures was performed. Among these genes, seven gene pair exon numbers were changed, and one or two exons were lost or obtained, including GhBOR1a-GhBOR2a, AtBOR1-BnBOR1s, AtBOR3-BnBOR3s, OsBOR4-ZmBOR4, GsBOR4a-GhBOR4c, AtBOR6-BnBOR6s, and AtBOR7-BnBOR7s. These changes may occur over the long course of evolution, in which the introns specifically inserted and remained in the genome of crops［34,35］, and indicate that both exon obtain and loss existed in the BOR gene family evolutionary period and helped to expand the functional differentiation of the BOR gene family, even though their structures were still highly similar in general. 　　To explore the differences in protein composition, 20 conserved motifs were identified among the BOR proteins by MEME tools (Table 2, Fig. 3b). The composition and arrangement of these motifs were largely consistent with the phylogenetic analysis. Most motifs were annotated to HCO3 cotransporters by the SMART and Pfam databases, which agreed with their function of B transport in biomembranes. In particular, motifs 6, 3, 8 and 2 (from left to right) were present in all BOR proteins, as shown in Fig. 3c. These motifs were arranged unusually in the N-terminus of BOR proteins, and we called them "general motifs" due to the majority distribution (>80%) in BOR proteins. The remaining motifs were classified as "specific motifs", which were arranged in the C-terminus of BOR proteins with unknown function. These properties might help to explain the functional differentiation of BOR proteins among these different species.
　　A previous study suggested that BOR proteins are transmembrane proteins［18］. Then, topological structures of 40 BOR gene family numbers were predicted by Protter tools (Fig. 4). In general, membrane topologies have similarities with the NJ phylogenetic tree. The number of TMSs (transmembrane segments) was variable and ranged from eight to fourteen (subfamily I: 8 to 12, subfamily II: 11, subfamily III: 9-12, subfamily IV: 8-14). Most TMSs were distributed in the N-terminus in BOR proteins, and only one to three TMSs existed in the C-terminus. In subfamilies I, II, and III, almost all N-termini were intracellularly distributed, except GhBOR2b and OsBOR3. At the C-terminus, two kinds of large loops showed both sides of the membrane in most BOR proteins with various models. This loop comprises sites that are essential for ion transportation, specified splicing and self-inactivation. In addition, BnBORs also exhibited topological structures highly similar to those of AtBORs in four subfamilies.
　　Intraspecies microsynteny analysis
　　Previous research has indicated that the genome duplication process is important for the evolution of gene families, including tandem, segmental and whole-genome duplication［24］. Next, we performed microsynteny analysis to identify the evolutionary history of the BOR gene family in these plants (i.e., rice, Arabidopsis, soybean, B. napus and cotton). If the flanking genes in the chromosome region of the BOR gene included four or more pairs of genes that are collinear, these two regions could be the result of a large-scale duplication event. 　　In rice, eight homologous gene pairs were involved between OsBOR2/OsBOR4 and OsBOR3/OsBOR4 comparation, and this region was considered to originate from a large-scale duplication event (Fig. 5a). In addition, because of the close positions of chromosome one between OsBOR2 and OsBOR3, they were considered to have evolved by tandem duplication. In Arabidopsis, seven BOR genes were observed, and all of them were distributed in duplicated regions of chromosomes (Fig. 5b). The flanking sequence synteny of AtBOR2/AtBOR3 was significant, and large-scale duplication events were considered to have occurred during their evolution. Moreover, nine (90%) and four (67%) BOR genes were found in duplicated regions in soybean and B. napus (Fig. 5c and 5d). Notably, the pairs GsBOR2b/GsBOR2c, GsBOR2b/GsBOR1c, GsBOR4b/GsBOR4c and GsBOR1a/GsBOR1b were notably located in syntenic regions. In contrast, only three (33%) BOR genes were found in the duplication region in cotton, and the synteny of the gene pairs GhBOR1a and -1d/GhBOR2b was not apparent in the genome (Fig. 5e). In general, these results suggested that large-scale duplication actively appeared in the evolution process of the BOR gene family in crops.
　　Strong purifying selection for OsBORs in rice
　　Microsynteny analysis suggested that gene duplication events were the key factor mediating BOR gene expansion during rice evolution. To determine the selection pressures driving BOR gene duplication, the Ka/Ks ratios for 15 flanking gene pairs were calculated in rice. Generally, Ka/Ks=1 means neutral selection, Ka/Ks>1 means positive selection, and Ka/Ks<1 means purifying selection. In this study, most Ka/Ks ratios of flanking gene pairs were less than 0.8, indicating that the OsBOR gene family underwent strong purifying selection and slow evolution at the protein level (Fig. 6a). Next, to further clarify OsBOR protein selection pressures, sliding-window analysis of Ka/Ks for each gene pair was performed (Fig. 6b-f). All Ka/Ks ratios were considerably less than 1 through coding regions, and compared to other structures, the HCO3 cotransport domains generally exhibited lower ratios. In addition, Ka/Ks ratios increased significantly in the C-terminus of OsBORs, which implied that positive selection began to gradually affect these regions. In general, these results revealed that strong purifying selection was an important factor for the evolution of BOR genes in rice, especially in HCO3 cotransport domains, which might be essential for the functional maintenance of boron transportation. 　　Expression pattern analysis of OsBORs in rice
　　To evaluate the functional differentiation of BOR gene family numbers in rice, OsBOR1-4 expression profiling was performed from the RiceXPro database (https://ricexpro.dna.affrc.go.jp/) and heatmaps generated by MeV software (Fig. 7). OsBOR1 and -3 exhibited a wide range of spatiotemporal distributions in rice compared to OsBOR2 and -4. OsBOR1 expression could be detected in almost all tissues, which indicated its essential function in B transportation during the whole growth cycle in rice. For OsBOR2, the expression was mainly focused on the ovary and embryo, and its function has not been determined. OsBOR3 was mainly expressed in anthers, pistils, ovaries, embryos, and OsBOR4 mRNA was only detected in anthers, which suggested their important role in rice reproduction. In general, the diverse patterns and degrees of expression of OsBORs might imply their functional differentiation in rice.
　　Discussion
　　The boron transport gene family plays an essential role in crop growth and reproduction. In this research, 4, 7, 10, 4, 6 and 9 classic BOR genes were identified in rice, Arabidopsis, soybean, maize, B. napus and cotton, respectively, from their genome databases. A total of 40 BOR genes were grouped into 4 subfamilies according to their phylogenetic analysis (Fig. 2). The differences in phylogenetics and BOR gene numbers might reflect the different metabolic regulation in the B transport system among these plants. Of the 40 BOR genes, 8 were found in monocotyledons (rice, maize), and 32 were found in dicotyledons (Arabidopsis, soybean, B. napus, cotton), which was in keeping with a previous study, which observed that the demand for B nutrition in monocotyledons was generally lower than that in dicotyledons［18］.
　　The diversity of gene structure and motif components is an important factor in the functional differentiation of gene families. In our research, the BOR gene family in 6 plants contained different numbers of exons, ranging from 10 (BnBOR1s and GhBOR1a) to 18 (ZmBOR4) (Table 1 and Fig. 3a). This diversity of BOR gene structure may have been caused by gene replication (loss or gain of introns or exons) during evolution and may contribute to the diversity of protein function. Regarding amino acid composition and protein physicochemical parameters, BOR gene family members shared significant homology with transmembrane proteins (Table 1, Fig. 3). For MEME analysis, most motifs of BORs were annotated to HCO3 cotransporters by SMART and the Pfam database, which was consistent with their function in B transport in the cytomembrane. At the N-terminus, the motif distribution was highly similar among the members of the BOR gene family (Fig. 3). In contrast, the differences in motif distribution in the C-terminus suggested that the BOR gene family members were functionally diversified. Furthermore, topological structure analysis revealed similar results with motif distributions. At the C-terminus, two kinds of large loops showed both sides of the membrane in most BOR proteins with various models, which was essential for ion regulation, sodium-dependent inactivation and alternative splicing (Fig. 4). 　　Previous studies have shown that while there is diversity in the number and size of chromosomes among different plant species, the distributions and orders of some genes remain highly conserved over a million years of evolution［36］. Strong microsynteny was detected in the rice, Arabidopsis, soybean, and B. napus genomes, and a comparison suggested the existence of large-scale genome duplications during evolution (Fig. 5). Interestingly, synteny levels in dicotyledons (Arabidopsis, soybean and B. napus) were considerably higher than those in monocotyledons (rice). These results implied the high conservation of the BOR gene family in dicotyledons and were consistent with the above research. In addition, we did not analyze microsynteny in maize because there were only one or two protein-coding sequences in the 100 kb flanking the anchor point, and these sequences were not suitable for analysis.
　　Genome duplication is essential for plants to adapt to changeable environments during evolution. In our study, the majority Ka/Ks ratios of paralogous gene pairs (14/15) were considerably less than 0.8, implying that purifying selection was the key factor in maintaining OsBOR function in rice (Fig. 6a). For sliding-window analysis, all Ka/Ks ratios were considerably less than 1 through the coding regions, and the levels of the four HCO3 cotransport domains were much lower than those in other regions, which agreed with the functional constraints of these domains. In contrast, the Ka/Ks ratios increased rapidly in the C-terminus and implied functional differentiation in the BOR gene family in rice, which was in keeping with the motif and topological structure analysis performed above (Fig. 6b-f).
　　References
　　［1］ KOBAYASHI M, MATOH T, AZUMA J. Two chains of rhamnogalacturonan II are cross-linked by borate diol ester bonds in higher plant cell walls［J］. Plant Physiology, 1996(110): 1017-1020.
　　［2］ ONEILL M A, ISHII T, ALBERSHEIM P, et al. Rhamnogalacturonan II: structure and function of a borate cross-linked cell wall pectic polysaccharide［J］. Annual review of plant biology, 2004(55): 109-139.
　　［3］ ONEILL M A. Requirement of borate cross-linking of cell wall rhamnogalacturonan II for Arabidopsis growth［J］. Science, 2001(294): 846-849.
　　［4］ MIWA K, WAKUTA S, TAKADA S, et al. Roles of BOR2, a boron exporter, in cross linking of rhamnogalacturonan II and root elongation under boron limitation in Arabidopsis［J］. Plant Physiology, 2013(163): 1699-1709. 　　［5］ CHENG C and RERKASEM B. Effects of boron on pollen viability in wheat［J］. Plant Soil, 1993(313): 155-156
　　［6］ HUANG L B, PANT J, DELL B, et al. Effects of boron deficiency on anther development and floret fertility in wheat (Triticum aestivum L. ‘wilgoyne’)［J］. Annals of Botany, 2000(85): 493-500.
　　［7］ GOLDBERG S. Reactions of boron with soils［J］. Plant Soil, 1997(193): 35-48.
　　［8］ CERVILLA L M, BLASCO B, ROS J J, et al. Oxidative stress and antioxidants in tomato (Solanum lycopersicum) plants subjected to boron toxicity［J］. Annals of Botany, 2007(100): 747-756.
　　［9］ BROWN P H and SHELP B J. Boron mobility in plants［J］. Plant Soil, 1997(193): 85-101.
　　［10］ DELL B and HUANG L. Physiological response of plants to low boron［J］. Plant Soil, 1997(193): 103-20
　　［11］ CHEN H F, ZHANG Q, HE M L, et al. Molecular characterization of the genome-wide BOR transporter gene family and genetic analysis of BnaC04.BOR1;1c in Brassica napus［J］. BMC Plant Biology, 2018(18): 193.
　　［12］ MIWA K and FUJIWARA T. Boron transport in plants: co-ordinated regulation of transporters［J］. Annals of Botany, 2010(7): 1103-1108.
　　［13］ DANNEL F, PFEFFER H, ROMHELD V. Compartmentation of boron in roots and leaves of sunflower as affected by boron supply［J］. Plant Physiol, 1998(153): 615-622
　　［14］ TAKANO J, NOGUCHI K, YASUMORI M, et al. Arabidopsis boron transporter for xylem loading［J］. Nature, 2002(420): 337-340.
　　［15］ TAKANO J, MIWA K, YUAN L, et al. (2005) Endocytosis and degradation of BOR1, a boron transporter of Arabidopsis thaliana, regulated by boron availability［J］. Proc Natl Acad Sci USA, 2005(102): 12276-12281.
　　［16］ TAKANO J, TANAKA M, TOYODA A, et al. Polar localization and degradation of Arabidopsis boron transporters through distinct trafficking pathways［J］. Proc Natl Acad Sci USA, 2010(107): 5220-5225.
　　［17］ KASAI K, TAKANO J, MIWA K, et al. High boron-induced ubiquitination regulates vacuolar sorting of the bor1 borate transporter in Arabidopsis thaliana［J］. J Biol Chem, 2011(286): 6175-6183.
　　［18］ NAKAGAWA Y, HANAOKA H, KOBAYASHI M, et al. Cell-type specificity of the expression of osbor1, a rice efflux boron transporter gene, is regulated in response to boron availability for efficient boron uptake and xylem loading［J］. The Plant Cell, 2007(19): 2624-2635.
　　［19］ DUAN X B, YU Y, ZHANG Y, et al. Potential efflux boron transporter gene GsBOR2, positively regulates Arabidopsis bicarbonate tolerance［J］. Plant Sci, 2018(274): 284-292. 　　［20］ DURBAK A R, PHILLIPS K A, PIKE S, et al. Transport of boron by the tassel-less1 aquaporin is critical for vegetative and reproductive development in maize［J］. The Plant Cell, 2014(26): 2978-2995.
　　［21］ ZHANG Q, CHEN H, HE M, et al. The boron transporter BnaC4.BOR1; 1c is critical for inflorescence development and fertility under boron limitation in Brassica napus［J］. Plant Cell Environ, 2017(40): 1819-1833.
　　［22］ LI M F, ZHAO Z Q, ZHANG Z H, et al. Effect of boron deficiency on anatomical structure and chemical composition of petioles and photosynthesis of leaves in cotton (Gossypium hirsutum L.)［J］. Sci Rep, 2017(7): 4420.
　　［23］ CAO Y P, HAN Y H, JIN Q, et al. Comparative genomic analysis of the GRF genes in chinese pear (Pyrus bretschneideri Rehd), poplar (Populous), grape (Vitis vinifera), Arabidopsis and rice (Oryza sativa)［J］. Frontiers in Plant Science, 2016(7): 1750.
　　［24］ CAO Y P, HAN Y H, MENG D D, et al. Genome-wide analysis suggests high level of microsynteny and purifying selection affect the evolution of EIN3/EIL family in rosaceae［J］. PeerJ, 2017(5): e3400.
　　［25］ THOMPSON J D, GIBSON T J, PLEWNIAK F, et al. The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools［J］. Nucleic Acids Res, 1997(25): 4876-4882.
　　［26］ TAMURA K, PETERSON D, PETERSON N, et al. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods［J］. Mol. Biol. Evol, 2011(28): 2731-2739.
　　［27］ LETUNIC I, DOERKS T, BORK P. SMART 7: recent updates to the protein domain annotation resource［J］. Nucleic Acids Res, 2012(40): D302-D305.
　　［28］ DELEU W, GONZáLEZ V, MONFORT A, et al. Structure of two melon regions reveals high microsynteny with sequenced plant species［J］. Mol. Genet. Genomics, 2007(278): 611-622.
　　［29］ SATO S, NAKAMURA Y, KANEKO T, et al. Genome structure of the legume. Lotus japonicus［J］. DNARes, 2008(15): 227-239.
　　［30］ EDGAR R C. MUSCLE: multiple sequence alignment with high accuracy and high throughput［J］. Nucleic Acids Res, 2004(32): 1792-1797.
　　［31］ SUYAMA M, TORRENTS D, BORK P. PAL2NAL: robust conversion of protein sequence alignments into the corresponding codon alignments［J］. Nucleic Acids Res, 2006(34): W609-W612.
　　［32］ LIBRADO P and ROZAS J. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data［J］. Bioinformatics, 2009(25): 1451-1452.
　　［33］ LIU S L, ZHUANG Y, ZHANG P, et al. Comparative analysis of structural diversity and sequence evolution in plant mitochondrial genes transferred to the nucleus［J］. Mol. Biol. Evol, 2009(26): 875-891.
　　［34］ ROGOZIN I B, WOLF Y I, SOROKIN A V, et al. Remarkable interkingdom conservation of intron positions and massive, lineage-specific intron loss and gain in eukaryotic evolution［J］. Curr. Biol, 2003(13): 1512-1517.
　　［35］ CARMEL L, WOLF Y I, ROGOZIN I B, et al. Three distinct modes of intron dynamics in the evolution of eukaryotes［J］. Genome Res, 2007(17): 1034-1044.
　　［36］ DEVOS K M and GALE M D. Genome relationships: the grass model in current research［J］. The Plant Cell, 2000(12): 637-646.