Genome-wide identification of lncRNAs
To systematically identify and characterize lncRNAs in rice, ssRNA sequencing was performed on shoot and root samples from rice seedlings grown in Fe-sufficient and -deficient conditions. After 10 days of Fe-deficient growth, rice plants showed significant chlorosis and lower chlorophyll content in the young leaves (Fig. 1A and 1B). The expression of typic Fe-deficiency responsive genes, such as the iron-related bHLH transcription factor 2 (IRO2), nicotianamine synthases 1 and 2 (NAS1/2), Fe(III)-DMA transporters (YSL15/16) and Iron-Regulated Transporter 1 (IRT1), were significantly increased (Fig. 1C), indicating that the rice seedlings is under iron deficiency condition and at the sampling time.
The pipeline for lncRNA identification and characterization is shown in Figure S1 (see methods). Using this pipeline, the approximately 700 million 150-bp pair-end reads were assembled into 31947 transcripts using Cufflinks. The Coding Potential Calculator (CPC) was used to evaluate the protein-coding potential of the transcripts to distinguish protein coding transcripts and lncRNAs. Transcripts more than 200 bp in length with CPC scores < 0 were defined as lncRNAs, the remaining transcripts were classified as protein-coding transcripts (mRNAs). Using this method, 25470 mRNAs and 6477 lncRNAs were identified.
Fe-deficiency responsive lncRNAs and mRNAs in rice shoot and root
To identify the lncRNAs and mRNAs that are differentially expressed in response to Fe deficiency, the normalized expression levels (in fragments per kilobase of exon per million fragments mapped, FPKM) of lncRNAs or mRNAs were compared between the Fe-deficient and Fe-sufficient treatments (Figure S2). In shoots, 80 differently expressed lncRNAs were identified (Log2 (fold change) ≥ 1 or ≤–1, probability > 0.8). Among them, 47 lncRNAs were up-regulated and 33 lncRNAs were down-regulated (Figure S2A and S3A; Table S2A). In roots, 89 lncRNAs were up-regulated and 32 lncRNAs were down-regulated under Fe deficiency (Fig. S2B and S3A; Table S2B). In addition, 394 and 841 mRNAs were differentially expressed in either roots or shoots due to Fe deficiency, respectively (Log2 (fold change) ≥ 1 or ≤–1, probability > 0.8). In shoots, 240 mRNAs were up-regulated and 154 were down-regulated (Figure S2C and S3B; and Table S2C), while in roots, 536 mRNAs were up-regulated and 305 mRNAs were down-regulated (Figure S2D and S3B; Table S2D).
The differentially expressed lncRNAs and mRNAs were used to generate a heat map (Fig. 2). Classes I and III contained lncRNA and mRNA transcripts that were expressed significantly higher in Fe-sufficient than in Fe-deficient conditions in either roots (Class I) or shoots (Class III), respectively. In contrast, transcripts in Classes Ⅱ and Ⅳ had higher expression in roots or shoots under Fe-deficient conditions, respectively. Transcripts in Class Ⅴ were more highly expressed in both shoots and roots under Fe-deficient conditions. Among the five groups, Class Ⅱ, the transcripts induced under Fe-deficient roots, contained the largest number of both lncRNAs (Fig. 2A) and mRNAs (Fig. 2B). In total, 171 lncRNAs and 1001 mRNAs were differentially expressed under different Fe supply conditions (Fig. 2; Table S3).
Verification of lncRNAs responding to Fe deficiency using quantitative RT-PCR
Quantitative RT-PCR (qRT-PCR) was performed to verify the accuracy of the RNA-seq data for the lncRNAs. Nine intergenic lncRNAs were picked for the verification. Expression of the Class IV lncRNAs XLOC_006153 and XLOC_028199 were induced in shoots but not detected in roots regardless of the Fe supply status. The Class II lncRNAs XLOC_052823 and XLOC_007199 were up-regulated by Fe deficiency in the roots. The remaining 5 lncRNAs belonged to Class Ⅴb, which were induced upon Fe deficiency in both shoots and roots (Fig. 3). Thus, qRT-PCR results were consistent with the RNA-Seq results.
Distribution of lncRNAs and mRNAs in the rice genome
Based on their relative position to protein-coding genes, lncRNAs can be classified into three types. Intergenic lncRNAs have no overlap with any protein-coding sequences, while sense lncRNAs and anti-sense lncRNAs overlap with one or more exons of another transcript on the same or opposite strand, respectively [20]. Among the 6477 lncRNAs identified in this work, 3730 (57%) were intergenic lncRNAs, 1696 were cis-lncRNAs, and 1051 were antisense lncRNAs (Figure S4).
Recent studies have shown that lncRNAs regulate the expression of genes via either cis- or trans-acting modes based on their genomic proximity to protein-coding genes [29]. To determine the modes of action of the lncRNAs in response to Fe deficiency, the genomic locations of the Fe-related lncRNAs and mRNAs mapped to each chromosome of the rice genome (Fig. 4). No significant genomic proximities were found between the Fe-responsive mRNAs and lncRNAs. Fe-related lncRNAs showed a higher degree of clustering in the genome than did the protein coding transcripts.
The conservation of lncRNAs
To investigate the conservation of the lncRNAs in the Oryza genus, putative lncRNAs were aligned with lncRNAs from 8 species of Oryza (O. barthii, O. brachyantha, O. glaberrima, O. glumipatula, O. meridionalis, O. nivara, O. punctata and O. rufipogon). Most of the lncRNAs were highly conserved among the Oryza species. In total, 1662 lncRNAs were detected in more than 7 species, while 860 lncRNAs were conserved in all 8 species (Fig. 5A).
The 860 conserved lncRNAs were aligned with lncRNAs from the monocots Zea mays, Hordeum vulgare, Sorghum bicolor, and Triticum aestivum. Among these 5 grass species, 88 of the lncRNAs were conserved (Fig. 5B; Table S4). However, only 3 of these conserved lncRNAs showed significant response to Fe deficiency (Fig. 5C). The results demonstrated that the lncRNAs were weakly conserved between Oryza and other gramineous plants.
Prediction of potential miRNA precursors and miRNA target mimics
miRNAs regulate key aspects of development, cell signaling, and responses to various biotic and abiotic stresses via binding to specific complementary transcripts, including protein coding or non-coding sequences, resulting in the degradation or translational repression of the target. LncRNA have been shown to function as precursors of miRNA in many studies [18, 30]. In this study, 23 lncRNAs were identified to be precursors of 29 known miRNAs in rice (Table S5). TCONS_00034827 was predicted to be the precursor of four miRNAs (osa-miR1428b, osa-miR1428c, osa-miR1428d, osa-miR1428e). TCONS_00000537 was predicted to generate two miRNAs (osa-miR156b and osa-miR156c). The results indicated the lncRNAs are quite complex and versatile and that at least some Fe-responsive lncRNAs may function through miRNAs.
In addition to generating miRNAs, lncRNAs are also targets of miRNAs. In this case, lncRNAs function as target mimicry with the sequestered transcript known as an endogenous target mimic (eTM) to inhibit miRNA activity [26]. In order to further verify whether the target mimicry mechanism is involved in Fe regulation in rice, any potential interactions between the Fe-responsive lncRNAs and known Fe-related microRNAs were investigated. Two endogenous target mimics (eTMs), osa-eTM159 and osa-eTM408, were identified. The lncRNAs, XLOC_012715 and XLOC_054182, were predicted to bind the miR159 and miR408, respectively (Fig. 6). The results demonstrated that target mimicry might be a part of the regulation of Fe uptake.