Genome-wide identification and expression analyses of the homeobox transcription factor family during ovule development in seedless and seeded grapes

Abstract

Seedless grapes are of considerable importance for the raisin and table grape industries. Previous transcriptome analyses of seed development in grape revealed that genes encoding homeobox transcription factors were differentially regulated in seedless compared with seeded grape during seed development. In the present study, we identified a total of 73 homeobox-like genes in the grapevine genome and analyzed the genomic content and expression profiles of these genes. Based on domain architecture and phylogenetic analyses grape homeobox genes can be classified into eleven subfamilies. An analysis of the exon-intron structures and conserved motifs provided further insight into the evolutionary relationships between these genes. Evaluation of synteny indicated that segmental and tandem duplications have contributed greatly to the expansion of the grape homeobox gene superfamily. Synteny analysis between the grape and Arabidopsis genomes provided a potential functional relevance for these genes. The tissue-specific expression patterns of homeobox genes suggested roles in both vegetative and reproductive tissues. Expression profiling of these genes during the course of ovule development in seeded and seedless cultivars suggested a potential role in ovule abortion associated with seedlessness. This study will facilitate the functional analysis of these genes and provide new resources for molecular breeding of seedless grapes.

Figure 1

Phylogenetic analysis of HB proteins from grape and Arabidopsis. The phylogenetic tree was constructed with alignments of HD sequences from each gene using MEGA5.0 software. Circles represent Arabidopsis proteins and triangles represent grape proteins. Different colors of branch lines of subtrees indicate different subfamilies. Numbers at the nodes indicate bootstrap values.

Figure 2

The structure analysis of grape HB genes. (a) Phylogenetic analysis and classification of grape genes. The phylogenetic tree was constructed with alignments of HD sequences from each gene using MEGA 5.0 software. Subtree branch lines are colored indicating different subfamilies. Numbers near the tree branches indicate bootstrap values. (b) Motif analysis of grape proteins. Motifs, numbered 1-20, were identified using MEME 4.11.2 software and are distinguished by color. Peptide sequence of each motif is provided in Supplementary Table S2. (c) Exon-intron structures of grape HB genes. Exons are marked as yellow boxes, and introns are represented by black lines connecting two exons. Upstream/downstream sequences are shown as blue boxes. Only exons and upstream/downstream sequences are drawn to scale.

Figure 3

Diagrammatic representation of the characteristic domains of representatives of the 11 subfamilies of grape HB proteins. The following domains and motifs are indicated: HD (homeodomain), LZ (leucine-zipper), ZIBEL motif, CPSCE motif, CESVV motif, START domain, HD-SAD domain (HD-START associated domain), MEKHLA domain, WUS Box, DDT domain, DDT domain, PHD domain, SAWADEE domain, PINTOX domain, POX domain, KNOX domain (1 and 2), ELK motif. Characteristic motifs and domains were identified using SMART and MEME software.

Figure 4

Synteny analysis and chromosomal distribution of grape HB genes. Locations of the grape genes are indicated by orange lines on the grape chromosomes. Colored bars connecting two chromosomal regions denote syntenic regions and the corresponding genes on two chromosomes were regarded as segmental duplications. Chr, chromosomes.

Figure 5

Analysis of synteny of grape HB genes between grape and Arabidopsis. The locations of grape and Arabidopsis genes are indicated by orange lines on respective chromosomes. Colored bars denote syntenic regions between grape and Arabidopsis genes. Chr, chromosomes.

Figure 6

Tissue-specific expression analysis of grape HB genes. Seedless grape cultivar ‘Thompson Seedless’ is denoted as ‘T.S’ and seeded grape cultivar ‘Red Globe’ is denoted as ‘R.G’. (a) Semi-quantitative RT-PCR analysis. For each gene, yellow and blue color scale indicates high and low expression levels respectively. Transcripts were normalized to the expression of the ACTIN1 gene and EF1-α gene, and results are shown in Supplementary Figs S3 and S5. Semi-quantitative RT-PCR was analyzed using GeneTools software, and expression values were normalized based on the mean expression value of each gene in all tissues. The heat map was analyzed using MeV 4.8 software. (b) Real-time PCR validation of twelve genes (9 differentially expressed genes and 3 randomly selected genes) expressed in different tissues. Transcripts were normalized to the expression of the ACTIN1 gene; the mean ±SD of three biological replicates is presented.

Figure 7

Expression profiles of grape HB genes during ovule development in four grape cultivars. Two seedless grape cultivars, ‘Thompson Seedless’ and ‘Flame Seedless’, are denoted as ‘T.S’ and ‘F’ respectively. Two seeded grape cultivars, ‘Red Globe’ and ‘Kyoho’, are denoted as ‘R.G’ and ‘K’ respectively. (a) Semi-quantitative RT-PCR analysis. For each gene, yellow and blue color scale indicates high and low expression levels, respectively. Numbers indicate the number of days after full bloom (DAF). Transcripts were normalized to the expression of the ACTIN1 gene and EF1-α gene and the results are shown in Supplementary Figs S4 and S5. Semi-quantitative RT-PCR measurements were analyzed using GeneTools software, and data was normalized based on the mean expression value of each gene in all development stages. The heat map was analyzed using MeV 4.8 software. (b) Real-time PCR validation of twelve genes (9 differentially expressed genes and 3 randomly selected genes) expressed during ovule development in four grape cultivars. Transcripts were normalized to the expression of the ACTIN1 gene; the mean ± SD of three biological replicates is presented.