Genome-wide analysis of the rice and Arabidopsis non-specific lipid transfer protein (nsLtp) gene families and identification of wheat nsLtp genes by EST data mining

Abstract

Background: Plant non-specific lipid transfer proteins (nsLTPs) are encoded by multigene families and possess physiological functions that remain unclear. Our objective was to characterize the complete nsLtp gene family in rice and arabidopsis and to perform wheat EST database mining for nsLtp gene discovery.

Results: In this study, we carried out a genome-wide analysis of nsLtp gene families in Oryza sativa and Arabidopsis thaliana and identified 52 rice nsLtp genes and 49 arabidopsis nsLtp genes. Here we present a complete overview of the genes and deduced protein features. Tandem duplication repeats, which represent 26 out of the 52 rice nsLtp genes and 18 out of the 49 arabidopsis nsLtp genes identified, support the complexity of the nsLtp gene families in these species. Phylogenetic analysis revealed that rice and arabidopsis nsLTPs are clustered in nine different clades. In addition, we performed comparative analysis of rice nsLtp genes and wheat (Triticum aestivum) EST sequences indexed in the UniGene database. We identified 156 putative wheat nsLtp genes, among which 91 were found in the 'Chinese Spring' cultivar. The 122 wheat non-redundant nsLTPs were organized in eight types and 33 subfamilies. Based on the observation that seven of these clades were present in arabidopsis, rice and wheat, we conclude that the major functional diversification within the nsLTP family predated the monocot/dicot divergence. In contrast, there is no type VII nsLTPs in arabidopsis and type IX nsLTPs were only identified in arabidopsis. The reason for the larger number of nsLtp genes in wheat may simply be due to the hexaploid state of wheat but may also reflect extensive duplication of gene clusters as observed on rice chromosomes 11 and 12 and arabidopsis chromosome 5.

Conclusion: Our current study provides fundamental information on the organization of the rice, arabidopsis and wheat nsLtp gene families. The multiplicity of nsLTP types provide new insights on arabidopsis, rice and wheat nsLtp gene families and will strongly support further transcript profiling or functional analyses of nsLtp genes. Until such time as specific physiological functions are defined, it seems relevant to categorize plant nsLTPs on the basis of sequence similarity and/or phylogenetic clustering.

Figure 1

Organization of nsLtp genes in rice and arabidopsis genomes. Positions of nsLtp genes are indicated on chromosomes (scale in Mbp).

Figure 2

Multiple sequence alignment of rice nsLTPs. Amino acid sequences were deduced from nsLtp genes identified from the TIGR Rice Pseudomolecules release 4 (Table 1). Sequences were aligned using HMMERalign to maximize the eight-cysteine motif alignment, and manually refined. The conserved cysteine residues are black boxed and additional cysteine residues grey boxed.

Figure 3

Multiple sequence alignment of arabidopsis nsLTPs. Amino acid sequences were deduced from nsLtp genes identified from the TAIR arabidopsis genome database (TAIR release 6.0) (Table 2). Sequences were aligned using HMMERalign to maximize the eight-cysteine motif alignment, and manually refined. The conserved cysteine residues are black boxed and additional cysteine residues grey boxed.

Figure 4

Multiple sequence alignment of wheat nsLTPs. Amino acid sequences were deduced from genes or ESTs indexed in the NCBI database. Amino acid sequences were aligned using HMMERalign to maximize the eight-cysteine motif alignment, and manually refined. For each nsLTP subfamily, one sequence is presented and the number of putative members identified is indicated between parentheses. The conserved cysteine residues are black boxed and additional cysteine residues grey boxed. Accession numbers are given in Additional file 2 and amino acid sequence of mature nsLTPs in Additional file 3.

Figure 5

Diversity of the eight cysteine motif in rice, arabidopsis and wheat nsLTP types. The consensus motif of each nsLTP type was deduced from the analysis of the matures sequences of the 52 rice nsLTPs, the 49 arabidopsis nsLTPs and the 156 wheat nsLTPs presented in Table 1, Table 2, and Additional file 2, respectively. AtLTPII.8 that appears to be more distantly related to other type II sequences (see the phylogenetic analysis) was excluded. The values allowing direct identification of the nsLTP type are grey boxed. ^a cysteine residue number 6 is missing in AtLTPII.10. ^b cysteine residue number 7 is missing in TaLTPVIa.5. ^c cysteine residue number 8 is missing in AtLTPI.1. ^d AtLTPII.10, OsLTPVI.1, OsLTPVI.2, OsLTPVI.4, and TaLTPVIa subfamily members harbor an extra cysteine residue. All type VI contain a Val 4 aa before Cys7 and a Met 10 aa before Cys7 allowing a distinction between type IV and type VI. ^e AtLTPII.6 harbors an extra cysteine residue. ^f TaLTPIVc.1 and TaLTPIVa subfamily members harbor an extra cysteine residue. ^g 12 amino acid residues were counted for the TaLTPIVd.1 that displays no CXC motif. ^h OsLTPVII.1 and TaLTPVIIa.1 subfamily members harbor an extra cysteine residue.

Figure 6

Unrooted phylogenetic tree between rice, arabidopsis and wheat nsLTP gene families. The mature sequences of the 122 non-redundant wheat nsLTPs, the 49 rice nsLTPs, and the 45 arabidopsis nsLTPs were aligned using HMMalign and then manually refined. The phylogenetic tree was built from the protein alignment (Additional file 3) with the maximum-likelihood method using the PHYML program [75]. When possible, subtrees including sequences of the same type are grouped and represented by a grey triangle close to which is indicated, in brackets, the number of sequences of arabidopsis, rice and wheat respectively. Subtrees are detailed in Figure 7. Bootstrap values (% of 100 re-sampled data set) are indicated for each node.

Figure 7

Rooted phylogenetic subtrees detailed from unrooted phylogenetic tree between rice, arabidopsis and wheat nsLTP gene families. Each subtree represented by a grey triangle in Figure 6 is detailed and rooted on the remaining parts of the tree. Wheat nsLTPs are in black, rice nsLTPs in red and arabidopsis nsLTPs in blue. Monophyletic subfamilies are indicated by solid brackets, paraphyletic subfamilies by dotted brackets. Black brackets indicate the wheat subfamily in which a potential rice ortholog nsLTP gene is present, and green brackets indicate wheat-specific subfamilies. Bootstrap values (% of 100 re-sampled data set) are indicated for each node.