Transcriptome analysis reveals candidate genes related to phosphorus starvation tolerance in sorghum

Abstract

Background: Phosphorus (P) deficiency in soil is a worldwide issue and a major constraint on the production of sorghum, which is an important staple food, forage and energy crop. The depletion of P reserves and the increasing price of P fertilizer make fertilizer application impractical, especially in developing countries. Therefore, identifying sorghum accessions with low-P tolerance and understanding the underlying molecular basis for this tolerance will facilitate the breeding of P-efficient plants, thereby resolving the P crisis in sorghum farming. However, knowledge in these areas is very limited.

Results: The 29 sorghum accessions used in this study demonstrated great variability in their tolerance to low-P stress. The internal P content in the shoot was correlated with P tolerance. A low-P-tolerant accession and a low-P-sensitive accession were chosen for RNA-seq analysis to identify potential underlying molecular mechanisms. A total of 2089 candidate genes related to P starvation tolerance were revealed and found to be enriched in 11 pathways. Gene Ontology (GO) enrichment analyses showed that the candidate genes were associated with oxidoreductase activity. In addition, further study showed that malate affected the length of the primary root and the number of tips in sorghum suffering from low-P stress.

Conclusions: Our results show that acquisition of P from soil contributes to low-P tolerance in different sorghum accessions; however, the underlying molecular mechanism is complicated. Plant hormone (including auxin, ethylene, jasmonic acid, salicylic acid and abscisic acid) signal transduction related genes and many transcriptional factors were found to be involved in low-P tolerance in sorghum. The identified accessions will be useful for breeding new sorghum varieties with enhanced P starvation tolerance.

Keywords: Candidate genes; Malate; Phosphorus starvation; Root development; Sorghum; Transcriptome analysis.

Fig. 1

Physiological changes of sorghum accessions in response to low-P stress. a and b Relative dry weight of shoot (a) and relative P content of shoot (b) of different accessions from 2015 and 2016 experiments. Asterisks denote statistically significant differences between the results from two independent experiments (P < 0.05). Bars represent the standard error of the mean (n = 3)

Fig. 2

Effect of low-P stress on root morphologies of sorghum accessions 12484 and 13443 grown in soil. Asterisks denote statistically significant differences between the two accessions (P < 0.05) by Student’s t-test. Bars represent the standard error of the mean (n = 3)

Fig. 3

Physiological responses of sorghum accessions 12484 and 13443 to low-P stress in the hydroponic system. a and b Root/shoot dry weight ratio of accession 12484 (a) and accession 13443 (b). c Photographs of representative plants. Asterisks denote significant differences between control and treatment groups by Student’s t-test (* means P < 0.05; ** means P < 0.01). Bars represent the standard error of the mean (n = 3)

Fig. 4

Comparison of the number of DEGs. 1: low-P-sensitive accession 13443; 2: low-P-tolerant accession 12484; L: low P; C: sufficient P. a The number of differentially expressed genes in each part. b Venn diagram illustrating the genes of the two accessions in response to low-P stress. c Response of DEGs to low-P stress and different accessions under low-P conditions

Fig. 5

Validation of transcript abundance obtained from RNA-seq using qRT-PCR. Twenty randomly chosen genes were used for validation. a and b Relative expression by RNA-seq and qRT-PCR of accession 13443 (a) and accession 12484 (b). Bars represent the standard error of the mean (n = 3)

Fig. 6

Heatmap of the candidate genes involved in plant hormone signal transduction. The log2 fold change of the candidate genes involved in plant hormone signal transduction under low-P conditions compared with that under sufficient-P conditions in each section is represented by a color scale consisting of red (upregulated), white (not regulated) and green (downregulated)

Fig. 7

Effect of malate application on root development. a Expression profile of the malate metabolism-related genes XLOC_030929 (Sb07g023910), XLOC_027815 (Sb06g020720) and XLOC_035509 (Sb09g005810). b Effect of applying malate on the length of the primary root. c Effect of applying malate on the number of root tips. The different lowercase letters denote significant differences among different treatments. Bars represent the standard error of the mean (n = 3)