THB1, a putative transmembrane protein that causes hybrid breakdown in rice

Abstract

Hybrid breakdown is a post-zygotic reproductive isolation that hinders genetic exchange between species or populations in both animals and plants. Two complementary recessive genes, temperature sensitive hybrid breakdown1 (thb1) and thb2, cause hybrid breakdown in rice (Oryza sativa). The present study delimited the THB1 locus to a 9.1-kb sequence, containing a single gene encoding a putative transmembrane protein with unknown functions. Haplotype analysis of THB1 in the two core collections of 119 accessions revealed that these accessions were divided into 22 haplotypes. A test cross with thb2 carrier showed that haplotype2 (H2) was assigned to thb1 and was restricted to temperate japonica. A nonsynonymous nucleotide polymorphism (SNP) specific to H2 was identified as a causal mutation in thb1. A test cross with thb1 carrier indicated that six accessions, including temperate japonica, tropical japonica, and indica, carried thb2. These results suggest that thb1 has recently evolved in temperate japonica, whereas thb2 arose in an ancient japonica and introgressed into the present three subgroups. Furthermore, we developed a derived cleaved amplified polymorphic sequence (dCAPS) marker to detect causal SNP in THB1. Our findings provide new insights into reproductive isolation and may benefit rice breeding.

Keywords: derived cleaved amplified polymorphic sequence marker; haplotype; hybrid breakdown; putative transmembrane protein; reproductive isolation; rice.

Fig. 1.

Map-based cloning of THB1. (A) High-resolution linkage map of THB1 using 72 recombinants between YK3InDel06-845078_2 and YK3InDel06-646646 from 3,754 F₂ individuals in the present study. (B) Structures of alternative splicing variants of LOC_Os06g02370. The full-length protein encoded by the longest transcript variant (LOC_Os06g02370.1) is 584 amino acids in length and was taken as reference for intron/exon numbering. LOC_Os06g02370.3 was generated by miss-splicing of intron resulting in less intron and changing the position of start codon, which causes lack of 1^st exon and reducing 5ʹ site of 2^nd exon (332 amino acids). White boxes represent 5ʹ-UTR; white pentagons represent 3ʹ-UTR; black boxes represent exon; and horizontal lines between them represent introns. The vertical arrows indicate the polymorphisms between ‘Yukihikari’ and ‘Kirara397’ in LOC_Os06g02370. Two sets of short horizontal arrows represent the positions of primers, thb1-1 and thb1-2, for expression analyses.

Fig. 2.

Phylogenic tree and multiple sequences alignment of THB1 (LOC_Os06g02370.1) with its highly similar protein sequences. Amino acid sequences of XP_015642768.1 (THB1, LOC_Os06g02370.1) from Oryza sativa ssp. japonica group; XP_015694070.2 from Oryza brachyantha; XP_025811579.1 from Panicum hallii; XP_003557155.1 from Brachypodium distachyon; XP_039842111.1 and XP_039805188.1 from P. virgatum; XP_051197997.1 from Lolium perenne; XP_047075120.1 from L. rigidum; XP_004964339.1 from Setaria italica; XP_037423695.1 and XP_037458488.1 from Triticum dicoccoides; XP_044426481.1, XP_044443345.1, and XP_044426478.1 from T. aestivum; NP_001144729.1 from Zea mays; XP_002437707.1 from Sorghum bicolor; XP_044959682.1 from Hordeum vulgare ssp. vulgare, XP_020147246.1 from Aegilops tauschii ssp. strangulate; XP_020112363.1 from Ananas comosus; XP_008813158.2 from Phoenix dactylifera; XP_010909418.1 from Elaeis guineensis; XP_009383130.2 from Musa acuminata ssp. malaccensis; XP_060211170.1 from Lycium barbarum; and XP_059318710.1 from L. ferocissimum were compared. (A) Phylogenic tree of THB1-like proteins. (B) Amino acid sequence alignment of 17 THB1-like proteins (amino acid identity >87.1% to LOC_Os06g02370.1). The amino acid sequence of LOC_Os06g02370.3 is indicated by white box. Predicted transmembrane domains (TMD) are indicated with red boxes. Identical and similar amino acid residues are shaded in black and gray, respectively. A nonsynonymous mutation site is indicated using a triangle.

Fig. 3.

Phylogenic and network analysis of THB1 gene. (A) Haplotype network of the THB1 in WRC and JRC. The size of each circle is proportional to the accession numbers encompassed, and different colors indicate the subgroups. The number of hatch marks reflects the number of nucleotide differences detected between haplotypes. (B) Phylogenic tree of THB1 based on nucleotide sequences of WRC and JRC. Arrowheads indicate the JRC ID used in test-cross experiments.

Fig. 4.

Gene structure and sequence alignment of the THB1 region. H4 is reference haplotype being carried by 51 accessions from 119 accessions from JRC and WRC. White cells indicate the same nucleotide as that of the reference haplotype. Black cells indicate variants. a indicates physical position on chromosome 6 (Nipponbare IRGSP-1.0 reference genome). b indicates 40-bp indel polymorphism of AAGCTACACAAACACAACGGAATGGTGAAGGTATAGAGAG.

Fig. 5.

Relative expression level of THB1 in nine accessions in 10 days after germination. (A) Relative expression levels of the LOC_Os06g02370.1 gene using primer set of thb1-1 were detected using quantitative RT-PCR. (B) Combined relative expression level of LOC_Os06g02370.1 and LOC_Os06g02370.3 using the primer set of thb1-2 were detected using quantitative RT-PCR. All data were normalized to the expression level of OsUBQ1 (Os03g0234200) (Yamamoto et al. 2010); error bars are SD for three independent experiments. Different letters indicate significant difference at p < 0.05 (Tukey’s test).