Identification of ankyrin-transmembrane-type subfamily genes in Triticeae species reveals TaANKTM2A-5 regulates powdery mildew resistance in wheat

Abstract

The ankyrin-transmembrane (ANKTM) subfamily is the most abundant subgroup of the ANK superfamily, with critical roles in pathogen defense. However, the function of ANKTM proteins in wheat immunity remains largely unexplored. Here, a total of 381 ANKTMs were identified from five Triticeae species and Arabidopsis, constituting five classes. Among them, class a only contains proteins from Triticeae species and the number of ANKTM in class a of wheat is significantly larger than expected, even after consideration of the ploidy level. Tandem duplication analysis of ANKTM indicates that Triticum urartu, Triticum dicoccoides and wheat all had experienced tandem duplication events which in wheat-produced ANKTM genes all clustered in class a. The above suggests that not only did the genome polyploidization result in the increase of ANKTM gene number, but that tandem duplication is also a mechanism for the expansion of this subfamily. Micro-collinearity analysis of Triticeae ANKTMs indicates that some ANKTM type genes evolved into other types of ANKs in the evolution process. Public RNA-seq data showed that most of the genes in class d and class e are expressed, and some of them show differential responses to biotic stresses. Furthermore, qRT-PCR results showed that some ANKTMs in class d and class e responded to powdery mildew. Silencing of TaANKTM2A-5 by barley stripe mosaic virus-induced gene silencing compromised powdery mildew resistance in common wheat Bainongaikang58. Findings in this study not only help to understand the evolutionary process of ANKTM genes, but also form the basis for exploring disease resistance genes in the ANKTM gene family.

Keywords: ankyrin-transmembrane protein; evolutionary progress; powdery mildew; virus-induced gene silencing; wheat.

Figure 1

Phylogenetic relationship analysis of 381 ANKTM proteins from Triticum aestivum, Triticum dicoccoides, Aegilops tauschii, Triticum urartu, H. vulgare and Arabidopsis. The phylogenetic tree was built using the Maximum likelihood method with 1,000 bootstrap replicates by MEGA X. The diverse classes of ANKTM proteins were marked with different colors. The ANKTM proteins of T. aestivum, T. dicoccoides, A. tauschii, T. urartu, H. vulgare, and Arabidopsis were represented by red triangles, blue circles, yellow squares, red stars, black triangles and blue stars, respectively. Gene IDs of all the analyzed genes can be found in Supplementary Table 2.

Figure 2

Homologous gene pairs, tandem duplication and location of ANKTM genes in wheat and T. dicoccoides genomes. All ANKTM genes were mapped to their respective loci in the wheat (A) and T. dicoccoides (B) genome in a circular diagram using shinyCircos. Homologous genes were inferred by TGT and linked with specific colors. Tandem duplication was analyzed by MCScanX, and the genes of red font indicated these genes were produced by tandem duplications. Subgenomes A and B are indicated by light blue and dark green, respectively; subgenomes D of wheat is indicated by dark blue.

Figure 3

Micro-collinearity analysis by TGT to track the evolutionary history of the tandem duplicated TuANKTM homologs. (A) The tandem repeats TuANKTM1A-4 and TuANKTM1A-5 were used as query genes. The micro-collinearity relationship showed that homologs of the analyzed TuANKTMs were found in all investigated genomes, and the homologous of TuANKTM1A-4 and TuANKTM1A-5 was TRIDC1AG012570 in T. dicoccoides and TraesCS1A02G088100 in wheat, respectively. (B) Conservative domain prediction of TRIDC1AG012570 and TraesCS1A02G088100 proteins. The conserved domain prediction indicates that TRIDC1AG012570 and TraesCS1A02G088100 are ANK proteins with a RING finger domain. Abbreviations: ANK, ankyrin repeat; TM, transmembrane domain; RING, RING finger domain. (C) The tandem repeats TuANKTM3A-9 and TuANKTM3A-10 were used as query genes. TuANKTM3A-9 and TuANKTM3A-10 from T. urartu had a “1-to-many” pairwise homology with TdANKTM3A-4, TdANKTM3A-5 and TdANKTM3A-6 in the A-subgenome of T. dicoccoides; meanwhile the homologous of the above genes in the collinearity regions of common wheat was just TaANKTM3A-5. Blackline, 1-to-1-mutual-best. Greenline, 1-to-its-best. Yellowline, 1-to-many. RBH, reciprocal best hits; SBH, single-side best hits.

Figure 4

Heat map of the expression profiling of wheat ANKTM genes under different stress. The color scale bar represents the expression values (in log2-based tags per million values) of the genes, and the values in square frames represent the tags per million values. The phylogenetic tree was constructed using the neighbor-joining method with 1,000 bootstrap replicates by MEGA X. Bgt, Blumeria graminis f. sp. tritici; Pst, Puccinia striiformis f. sp. tritici; Mock-N, Disease-resistant wheat varieties N9134 without the infection of Bgt and Pst.

Figure 5

Expression profiling of three TaANKTMs in common wheat AK58 after Bgt inoculation by qRT-PCR. Relative expression of three TaANKTMs in response to Bgt. Data were normalized to the TaTubulin gene. The values are the means of three technical replicates of one biological experiment. Asterisks indicate significant differences (assessed using Duncan’s honestly significant difference test), **p < 0.01. All the raw data for qRT-PCR are listed in Supplementary Table 3.

Figure 6

Functional analysis of TaANKTM2A-5 by Barley stripe mosaic virus-based virus-induced gene silencing (BSMV-VIGS) in AK58. (A) BSMV: TaANK2A-5 infected plants were inoculated with Blumeria graminis f. sp. Tritici, and their leaves were photographed 6 days post-inoculation. BSMV:γ infected plants were performed as controls. The experiment was repeated independently three times, and the same results were obtained. (B) Expression of TaANKTM2A-5 in BSMV: TaANKTM2A-5-infected leaves was compared with that in BSMV:γ-infected controls of AK58. CK represents plants inoculated with BSMV:γ, and 1–3 represents plants inoculated with BSMV: TaANK2A-5. Significant differences assessed using Duncan’s honestly significant difference test, **p < 0.01.