Table 1.
The number of Hsfs identified in the legume species and lower plants.
Figure 1.
Phylogenetic tree of 140L. japonicus, M. truncatula, C. arietinum, G. max, C. cajan and P. vulgaris.
This tree was constructed based on amino acid sequence comparison of the conserved N-terminal regions of Hsfs including the DNA-binding domain, the HR-A/B region and parts of the linker between them, using the neighbor-joining method with 1,000 bootstrap replicates. The Hsf of Saccharomyces cerevisiae (ScHsf1) and the Hsfs of C. reinhardtii, S. moellendorffii and P. patens were used as the outgroup. The colors indicate the species background of the Hsfs. The tree was divided into 18 shared clades (Clades 1–18) according to evolutionary distances. The bootstrap values of both neighbor-joining (NJ) tree (first number; 1000 replicates) and maximum likelihood (ML) tree (second number; 1000 replicates) were shown on the branches leading to each of the clades. The clades were supported by high bootstrap values in neighbor-joining and maximum likelihood analyses. Different subclasses of Hsfs are indicated in brackets. Gene names are presented in Table S2. The scale bar represents 0.1 amino acid changes per site.
Figure 2.
Idealized gene trees of the duplication groups of Hsf genes in G. max, L. japonicus, M. truncatula and C. cajan.
Each tree represents a duplication group from large-scale gene duplication. As shown in the trees, every Hsf gene of G. max was expected to be present in four copies after two rounds of whole-genome duplication (early and recent). Similarly, the number of Hsf genes in L. japonicus, M. truncatula and C. cajan will have doubled after the early-legume whole-genome duplication. The five duplicated gene pairs (LjHsf-06/LjHsf-11, GmHsf-09/GmHsf-34, GmHsf-18/GmHsf-24, GmHsf-18/GmHsf-46 and GmHsf-21/GmHsf-45) were classified in the flexible set. The question marks indicate possible gene loss events. The GmHsf-18/GmHsf-38 pair could be formed by a segmental duplication that predated the recent whole-genome duplication, and both the GmHsf-18 and GmHsf-38 lost homoeologs from the recent whole-genome duplication.
Table 2.
Estimates of the dates for the large scale duplication events in legume species.
Figure 3.
Estimates of Ks and Ka/Ks ratios in pairwise comparisons.
(A) Distribution of synonymous distances (Ks) between paralogous genes flanking duplicated Hsf genes in G. max. The histogram shows the number of duplicate gene pairs (y-axis) versus synonymous distance between pairs (x-axis). The Ka/Ks ratios of the duplicated Hsf genes (B) and their flanking paralogs (C) in G. max are shown in the scatter plots; the y and x axes denote the Ka/Ks ratio and synonymous distance for each pair, respectively.
Figure 4.
Extensive microsynteny of Hsf regions across L. japonicus, M. truncatula, G. max and C. cajan.
G. max chromosomes, labeled Gm, are indicated by red boxes. The L. japonicus, M. truncatula and C. cajan chromosomes, shown in different colors, are labeled Lj, Mt and Cc, respectively. Numbers along each chromosome box indicate sequence lengths in megabases. The whole chromosomes of these four legumes, harboring Hsf regions, are shown in a circle. Black lines represent the syntenic relationships between Hsf regions.
Figure 5.
Comparative maps of representative Hsf genes and their flanking genes within syntenic chromosomal intervals across selected legume species.
The relative positions of all flanking protein-coding genes were defined by the anchored Hsf genes, highlighted in red. The chromosome segments are shown as gray horizontal lines, with arrows corresponding to individual genes and their transcriptional orientations. All genes are numbered from left to right, in order, for each segment. Where several duplicated genes were present within a region, these genes were given the same number, with the letters a, b, c… appended in order. Conserved gene pairs among the segments are connected with lines. (A) The syntenic chromosomal intervals containing MtHsf-17, LjHsf-07, CcHsf-09, GmHsf-13 and GmHsf-15 across M. truncatula, L. japonicus, C. cajan and G. max. (B) The syntenic chromosomal intervals containing MtHsf-03, MtHsf-16, LjHsf-06, LjHsf-11, GmHsf-12, GmHsf-17, GmHsf-32 and GmHsf-36 across M. truncatula, L. japonicus, and G. max. (C) The syntenic chromosomal intervals containing MtHsf-12, GmHsf-01 and GmHsf-20 across M. truncatula and G. max. The full microsynteny maps of the regions containing Hsf genes within M. truncatula, L. japonicus, C. cajan and G. max are shown in Figure S5.
Table 3.
The synteny quality of regions orthologous across L. japonicus, M. truncatula, G. max and C. cajan.
Figure 6.
Sliding window plots of representative duplicated Hsf genes in G. max.
As shown in the key, the gray blocks, from dark to light, indicate the positions of the DBD domain, HR-A/B region, NLS, NES and AHA motifs, respectively. The window size is 150 bp, and the step size is 9 bp. The data for all pairs of duplicated Hsf genes of soybean are shown in Figure S6.
Figure 7.
L. japonicus Hsf genes expression in various plant tissues.
The type of tissue (nodule, root, stem, leaf and flower) and the gene name are shown on the y-axis and x-axis, respectively. Hierarchical clustering based on average log signal values in various tissues grouped 10 of the L. japonicus Hsf genes into four types (A–D).
Figure 8.
Expression of L. japonicus Hsf genes in response to abiotic stress measured by quantitative real-time PCR.
The mRNA level of each gene in L. japonicus seedlings given heat (HS: 42°C), cold (4°C) and oxidative (OS: 10 mM H2O2) stress for 1 h was plotted relative to the value obtained for the unstressed contral. Error bars represent standard errors.