Genome-wide comparative analysis of annexin superfamily in plants

Abstract

Most annexins are calcium-dependent, phospholipid-binding proteins with suggested functions in response to environmental stresses and signaling during plant growth and development. They have previously been identified and characterized in Arabidopsis and rice, and constitute a multigene family in plants. In this study, we performed a comparative analysis of annexin gene families in the sequenced genomes of Viridiplantae ranging from unicellular green algae to multicellular plants, and identified 149 genes. Phylogenetic studies of these deduced annexins classified them into nine different arbitrary groups. The occurrence and distribution of bona fide type II calcium binding sites within the four annexin domains were found to be different in each of these groups. Analysis of chromosomal distribution of annexin genes in rice, Arabidopsis and poplar revealed their localization on various chromosomes with some members also found on duplicated chromosomal segments leading to gene family expansion. Analysis of gene structure suggests sequential or differential loss of introns during the evolution of land plant annexin genes. Intron positions and phases are well conserved in annexin genes from representative genomes ranging from Physcomitrella to higher plants. The occurrence of alternative motifs such as K/R/HGD was found to be overlapping or at the mutated regions of the type II calcium binding sites indicating potential functional divergence in certain plant annexins. This study provides a basis for further functional analysis and characterization of annexin multigene families in the plant lineage.

Figure 1. Phylogenetic trees showing the evolutionary relationship of 149 deduced annexin proteins from Viridiplantae by maximum likelihood method in RaxML (A) and Bayesian inference in MrBayes (B).

The multiple sequence alignment was done for the deduced protein sequences using multiple sequence and structure alignment program PROMALS3D. The numbers at the nodes indicates the statistical support as obtained by 100 bootstrap RaxML replicates (likelihood of −51862.57) and Bayesian posterior probabilities (likelihood of −52897.07). The red lettered taxon labels represent the segmentally duplicated paralogous annexin sequences. We used algae as outgroup. The bar indicates amino acid substitutions per site.

Figure 2. Sequence logos of four annexin domains.

The sequence logos were generated by amino acid alignment of individual domains from 149 annexins using WebLogo. The taxon-specific indels were removed to optimize the alignments. The height of letter designating the amino acid residue at each position represents the degree of conservation. The GXGT and D/E, IRI and DXXG motifs are represented on the top of each plot. The conserved His residue in the heme motif is indicated by an arrow. The residues thought to be involved in ion channel activity are represented as diamonds. Asterisks (*) indicated the conserved residues observed in the alignment of 149 annexins. The numbers on the x-axis represent the sequence positions in annexin domains. The y-axis represents the information content measured in bits.

Figure 3. Chromosomal localization of annexin genes in (A) rice, (B) Arabidopsis and (C) poplar.

The number indicated at the top represents the chromosome number. The tandemly duplicated genes are indicated as vertical lines and the segmental duplicated genes by dotted lines. The scale represents a 5 Mb chromosomal distance. The numbers in brackets represents the corresponding chromosome size.

Figure 4. Exon-intron organizations of annexin genes from genomes of moss, spike moss, Arabidopsis and rice.

Exons and introns are indicated as open boxes and dotted lines respectively. The intron phases are depicted as 0 and 1 at the top. Numbers at the left show intron-exon patterns and those at the right show the type of pattern observed in different genes across genomes. Numbers within the boxes represent exon sizes. The exons and introns are not drawn to scale.

Figure 5. Gene structures of annexins from four representative genomes (moss, spike moss, Arabidopsis and rice) in plant lineage.

The intron phases are highly conserved in all the genomes and represented in the maximum likelihood phylogeny tree. As shown in the legend, the intron phases in between exon-intron junctions are given as 0 and 1, exons are represented by green filled boxes, introns by black lines and untranslated regions (UTR) by blue filled boxes. The scale bar represents 0.2 amino acid substitutions per site. The gene structures were drawn using online tool Gene Structure Display Server (http://gsds.cbi.pku.edu.cn/).

Figure 6. Domain organizations showing the presence or absence of CBS.

Anx domain structures showing CBS and K/H/RGD motifs. The occurrence of these motifs in different genomes was analyzed from all deduced proteins. Dark circles represent the absence of CBS or K/H/RGD motifs. Domain structures are not drawn to scale.