Comparative Transcriptomic Analyses of Nitrate-Response in Rice Genotypes With Contrasting Nitrogen Use Efficiency Reveals Common and Genotype-Specific Processes, Molecular Targets and Nitrogen Use Efficiency-Candidates

Abstract

The genetic basis for nitrogen (N)-response and N use efficiency (NUE) must be found in N-responsive gene expression or protein regulation. Our transcriptomic analysis of nitrate response in two contrasting rice genotypes of Oryza sativa ssp. Indica (Nidhi with low NUE and Panvel1 with high NUE) revealed the processes/functions underlying differential N-response/NUE. The microarray analysis of low nitrate response (1.5 mM) relative to normal nitrate control (15 mM) used potted 21-days old whole plants. It revealed 1,327 differentially expressed genes (DEGs) exclusive to Nidhi and 666 exclusive to Panvel1, apart from 70 common DEGs, of which 10 were either oppositely expressed or regulated to different extents. Gene ontology analyses revealed that photosynthetic processes were among the very few processes common to both the genotypes in low N response. Those unique to Nidhi include cell division, nitrogen utilization, cytoskeleton, etc. in low N-response, whereas those unique to Panvel1 include signal transduction, protein import into the nucleus, and mitochondria. This trend of a few common but mostly unique categories was also true for transporters, transcription factors, microRNAs, and post-translational modifications, indicating their differential involvement in Nidhi and Panvel1. Protein-protein interaction networks constructed using DEG-associated experimentally validated interactors revealed subnetworks involved in cytoskeleton organization, cell wall, etc. in Nidhi, whereas in Panvel1, it was chloroplast development. NUE genes were identified by selecting yield-related genes from N-responsive DEGs and their co-localization on NUE-QTLs revealed the differential distribution of NUE-genes between genotypes but on the same chromosomes 1 and 3. Such hotspots are important for NUE breeders.

Keywords: QTLs; networks; nitrate; nitrogen use efficiency; rice; transcriptome.

FIGURE 1

Nitrate-responsive transcriptomes in Nidhi and Panvel1. (A) Representative image of 21 days old Nidhi and Panvel1 plants grown in nutrient-depleted soil supplemented with AH media containing normal (15 mM) and low (1.5 mM) nitrate. (B) Leaf chlorophyll content (SPAD value) was estimated earlier in Nidhi and Panvel1 (Sharma et al., 2021). Quantitative RT-PCR was used to calculate the relative mRNA expression of nitrate reductase (C) and nitrite reductase (D) genes in 21 days old Nidhi and Panvel1 plants. Test samples were low nitrate, whereas normal nitrate samples were used as control. The actin gene was used as a reference gene to normalize the expression data. Data represent the mean ± SE of three technical replicates. An unpaired t-test was performed in GraphPad Prism. Experiments were performed repeatedly with two independent biological replicates. Volcano plots for differential gene regulation are shown for Nidhi (E) and Panvel1 (F). Scattered dots represent different transcripts and the horizontal dashed line corresponds to the P-value cut-off (P = 0.05). Red scattered dots represent the mapping of upregulated genes, whereas downregulated genes are by green dots. (G) Venn diagram shows commonly and uniquely up or downregulated DEGs between Nidhi and Panvel1. (H) Predicted subcellular localization of DEGs encoded proteins in Nidhi and Panvel1. **P < 0.01, ***P < 0.001, ****P < 0.0001.

FIGURE 2

Heat map was constructed using Heatmapper, which represents enriched top ten gene ontology (GO) terms (biological processes) for contrasting rice genotypes Nidhi and Panvel1. Minus log P values were plotted against the respective GO term.

FIGURE 3

Validation of expression profile of nitrate responsive genes by RT-qPCR. Relative change in the gene expression was calculated by the comparative Ct value method and the actin gene was used for data normalization. The control values were taken as zero and the test values are shown as the average of three technical and two independent biological replicates (+SE) except gene BGL1 for which the calculations were done based on three technical replicates of a biological replicate. Each sub-figure compares gene expression of RT-qPCR and microarray for Nidhi versus Panvel1 for gene OsBGl1 (A), OSBBX19 (B), OsSULTR3;2 (C), OsCP29 (D), OsSRP43 (E), and OsPCL1 (F).

FIGURE 4

(A) Heat map was constructed using Heatmapper, which represents the expression pattern of nitrate regulated differentially expressed genes (DEGs) of the top two transporters’ families for Nidhi and Panvel1 rice contrasting genotypes. For Nidhi, the top two transporter families are Amino Acid/Auxin Permease (AAAP) Family and Amino Acid-Polyamine-Organocation (APC) Family, while for Panvel1 they are Drug/Metabolite Transporter (DMT) Superfamily and Major Intrinsic Protein (MIP) Family. (B) Expression pattern of nitrate regulated DEGs of top two transcription factors’ families for Nidhi and Panvel1 rice contrasting genotypes. For Nidhi, the top two TFs families are WRKY and ARF, while for Panvel1 they are AP2 and bHLH.

FIGURE 5

Predicated nitrate-responsive transcriptional regulatory network (TRN) in Nidhi (A) and Panvel1 (B). Nitrate-regulated Arabidopsis TRNs (Gaudinier et al., 2018) were used to identify the DEGs-associated interactors in rice. Orthologous information was retrieved from the PlantGDB database and networks were constructed in Cytoscape ver 3.9.0. Expression values of DEGs were mapped onto the networks where red color nodes represent upregulated DEGs and blue color nodes correspond to downregulated DEGs. Light gray color nodes represent interactors but not DEGs in Nidhi and Panvel1.

FIGURE 6

Nitrate-responsive protein-protein interaction (PPI) sub-clusters/molecular complexes in Nidhi and Panvel1. DEGs-associated interactors were retrieved by STRING, BioGRID, PRIN, and MCDRP databases. Experimentally validated interaction pairs were used to construct the PPI networks in Cytoscape (Supplementary Figures S2, S3). Molecular complexes/sub-clusters of PPI networks were identified using the MCODE plugin in Cytoscape. Thirteen and six sub-clusters/molecular complexes were detected in Nidhi and Panvel1, respectively. Important nitrate-responsive sub-clusters/molecular complexes identified in Nidhi and Panvel1 are shown (A,B) and the remaining are given in Supplementary Figure S4. Red and blue color nodes correspond to the up and downregulated DEGs, respectively. Light gray color nodes represent the interactors, which are not DEGs.

FIGURE 7

(A) Venn selection of yield-related and Nidhi nitrate genes revealed 188 NUE-genes. Among them, only 36 NUE-genes colocalized onto 9 NUE-QTLs. (B) Venn selection of yield-related and Panvel nitrate genes revealed 98 NUE-genes. Among them, only 26 NUE-genes colocalized to 16 NUE-QTLs. (C,D) Representative figure of Nidhi nitrate NUE genes colocalized on chromosomes 1 and 3. (E,F) Representative figure of Panvel nitrate NUE genes colocalized on chromosomes 1 and 3. Gene id is given on the right side of the map and the physical location of genes is given on the left side of the map (in mb).

FIGURE 8

Validation of biological processes: Validation was done using Licor instrument 6400XT (LI-COR, Lincoln, NE, United States) on 21 old days grown plants. Plants were grown in nutrient-depleted soil and fertilized with Arnon Hoagland medium having nitrate as the sole source of N with 15 mM concentration as control while 1.5 mM, was used as a test. Measurement was done in five biological replicates. Percent increase or decrease (relative measurement) for each of the measurements was calculated in low nitrate over normal nitrate. (A) Relative photosynthesis was measured in terms of μmol (CO₂) m^–2s^–1. (B) Relative transpiration was measured in terms of mol (H₂O) m^–2s^–1, and (C) Relative stomatal conductance was measured in terms of mol (H₂O) m^–2sec^–1). (D) Relative carboxylation efficiency was measured in terms of μmol (CO₂)/m²s¹/Ci as the ratio of photosynthesis and internal CO₂ concentration, (E) Relative internal water use efficiency was measured in terms of μmol (CO₂)/mol (H₂O), and (F) Relative transpiration efficiency was measured in terms of μmol (CO₂)/mmol H₂O m^–2s^–1. The test of significance for low nitrate over normal nitrate for each of the individual bars has been shown as star (P < 0.05), while NS represents non-significance.