Figure 1.
Transcriptomes of sorghum and Bipolaris sorghicola were analyzed simultaneously by using different approaches. The mixed transcriptome obtained from B. sorghicola-infected sorghum leaves was sequenced by using Illumina mRNA-Seq technology. The sequenced reads contain the reads derived from sorghum (green bars), B. sorghicola (purple bars), and other organisms such as normal inhabitants of plant tissue (red bars). The reads were aligned to the sorghum reference genome (black bars), and the aligned and unaligned reads were used to analyze gene expression in sorghum (left) and B. sorghicola (right), respectively. Gene expression in sorghum was analyzed for each transcript (green arrows), including transcripts annotated with Phytozome and unannotated transcripts identified on the basis of the piling-up of aligned reads by using the Cufflinks program. For B. sorghicola, unaligned reads were assembled by using the Oases program to retrieve the pathogen transcripts expressed during growth of the fungus in the plant. Expression of the assembled transcripts was analyzed by aligning the reads back to the transcripts. The assembled transcripts contained not only the transcripts of B. sorghicola (purple arrows) but also those of other organisms (red arrows). The transcripts of other organisms were removed as contaminants, and those of B. sorghicola was validated experimentally.
Table 1.
Transcripts of Bipolaris sorghicola expressed during its growth in Sorghum bicolor plants.
Figure 2.
Pathogen (Bipolaris sorghicola)-induced genes encoding receptors in sorghum.
(A) Fold changes of pathogen-induced transcripts for each family. RPKM fold changes at 24 h were calculated for infected samples compared with mock-infected samples. Genes with a fold change value of ≥3 were shown. To avoid division by 0, 1 was added to the RPKM values of mock-infected samples. Families of the receptors, shown on the top, were defined on the basis of the Pfam domain structures (See Materials and Methods). (B) Domain structures of the 2 receptors for which the genes were highly induced. Predicted structures included SP, signal peptide; LRRNT_2, Leucine rich repeat N-terminal domain (PF08263); LRR_4, Leucine Rich repeats (2 copies) (PF12799); and LRR_8, Leucine rich repeat (PF13855). (C) Phylogenetic tree of the 2 receptors for which the genes were particularly strongly induced, along with their homologous proteins. The amino acid sequences of LRRNT_2 domain (PF08263) of each protein were aligned by using ClustalW and the tree was created by using MEGA5. Red underlines show the 2 receptors for which the genes were particularly strongly induced. Abbreviations are as follows: Sb, Sorghum bicolor; Os, Oryza sativa (rice); At, Arabidopsis thaliana; Nt, Nicotiana tabacum (tobacco); Ca, Capsicum annuum (pepper); Sp, Solanum pimpinellifolium (currant tomato).
Table 2.
Pathogen (Bipolaris sorghicola)-induced genes encoding proteins for signaling cascade in Sorghum bicolor.
Figure 3.
Pathogen (Bipolaris sorghicola)-induced genes encoding transcriptional factors in sorghum.
(A) Number of pathogen-induced transcripts for each family. Families of the transcriptional factors (TFs), shown at left, were defined on the basis of a homology search in the Plant Transcription Factor Database [22]. (B) Expression of the 10 induced WRKY TFs. RPKMs of each transcript were compared in mock- and pathogen-infected leaves by using the disease-resistant cultivar SIL-05 in the early stages of infection (24 h) (this study) and a related cultivar, BTx623, at a relatively late infection stage (7 days) (our previous study [13]) (C) Phylogenetic tree of the 2 WRKY TFs for which the genes were expressed at relatively high levels. The amino acid sequences of WRKY domain (PF03106) of Sb08g005080 (red underline in right panel) and Sb04g005520 (red underline in left panel), and their best 10 BLAST hits in the Phytozome sorghum protein database [16] and the Rice Annotation Project (RAP) protein database (http://rapdb.dna.affrc.go.jp), were aligned by using ClustalW. Phylogenetic trees were constructed by using MEGA5. For Sb04g005520, HvWRKY1/2 and ATWRKY40/60 were also aligned. Abbreviations are as follows: Sb, Sorghum bicolor; Os, Oryza sativa (rice); Hv, Hordeum vulgare; At, Arabidopsis thaliana.
Figure 4.
Pathogen (Bipolaris sorghicola)-induced genes encoding peroxidases in sorghum.
(A) Pathogen-induced fold changes in expression. RPKM fold changes at 24 h were calculated for infected samples compared with those of mock-infected samples. To avoid division by 0, 1 was added to the RPKM values of mock-infected samples. (B) Phylogenetic tree of peroxidases for which the encoding genes were strongly induced. The amino acid sequences of peroxidase domain (PF00141) of Sb02g042860 (SbPrx18) (red underline) and its best 10 BLAST hits in the Phytozome sorghum protein database [16] and the Rice Annotation Project (http://rapdb.dna.affrc.go.jp) protein database, as well as the sequences of wheat TaPrx103 and barley HvPrx08, were aligned by using ClustalW. Phylogenetic tree was constructed by using MEGA5. Abbreviations are as follows: Sb, Sorghum bicolor; Os, Oryza sativa (rice); Hv, Hordeum vulgare; Ta, Triticum aestivum.