Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Genome-wide identification and characterization of microRNAs differentially expressed in fibers in a cotton phytochrome A1 RNAi line

  • Qing Miao,

    Affiliation Department of Biochemistry, Molecular Biology, Entomology and Plant Pathology, Mississippi State University, Mississippi State, MS, United States of America

  • Peng Deng,

    Affiliation Department of Pharmacology, Weill Cornell Medical College of Cornell University, New York, NY, United States of America

  • Sukumar Saha,

    Affiliation USDA-ARS, Crop Science Research Laboratory, Mississippi State, MS, United States of America

  • Johnie N. Jenkins,

    Affiliation USDA-ARS, Crop Science Research Laboratory, Mississippi State, MS, United States of America

  • Chuan-Yu Hsu,

    Affiliation Institute for Genomics, Biocomputing and Biotechnology, Mississippi State University, Mississippi State, MS, United States of America

  • Ibrokhim Y. Abdurakhmonov,

    Affiliation Center of Genomics and Bioinformatics, Academy of Sciences of Uzbekistan, Tashkent, Uzbekistan

  • Zabardast T. Buriev,

    Affiliation Center of Genomics and Bioinformatics, Academy of Sciences of Uzbekistan, Tashkent, Uzbekistan

  • Alan Pepper,

    Affiliation Department of Biology, Texas A & M University, College Station, TX, United States of America

  • Din-Pow Ma

    dma@bch.msstate.edu

    Affiliation Department of Biochemistry, Molecular Biology, Entomology and Plant Pathology, Mississippi State University, Mississippi State, MS, United States of America

Genome-wide identification and characterization of microRNAs differentially expressed in fibers in a cotton phytochrome A1 RNAi line

  • Qing Miao, 
  • Peng Deng, 
  • Sukumar Saha, 
  • Johnie N. Jenkins, 
  • Chuan-Yu Hsu, 
  • Ibrokhim Y. Abdurakhmonov, 
  • Zabardast T. Buriev, 
  • Alan Pepper, 
  • Din-Pow Ma
PLOS
x

Abstract

Cotton fiber is an important commodity throughout the world. Fiber property determines fiber quality and commercial values. Previous studies showed that silencing phytochrome A1 gene (PHYA1) by RNA interference in Upland cotton (Gossypium hirsutum L. cv. Coker 312) had generated PHYA1 RNAi lines with simultaneous improvements in fiber quality (longer, stronger and finer fiber) and other key agronomic traits. Characterization of the altered molecular processes in these RNAi genotypes and its wild-type controls is a great interest to better understand the PHYA1 RNAi phenotypes. In this study, a total of 77 conserved miRNAs belonging to 61 families were examined in a PHYA1 RNAi line and its parental Coker 312 genotype by using multiplex sequencing. Of these miRNAs, seven (miR7503, miR7514, miR399c, miR399d, miR160, miR169b, and miR2950) were found to be differentially expressed in PHYA1 RNAi cotton. The target genes of these differentially expressed miRNAs were involved in the metabolism and signaling pathways of phytohormones, which included Gibberellin, Auxin and Abscisic Acid. The expression of several MYB transcription factors was also affected by miRNAs in RNAi cotton. In addition, 35 novel miRNAs (novel miR1-novel miR35) were identified in fibers for the first time in this study. Target genes of vast majority of these novel miRNAs were also predicted. Of these, nine novel miRNAs (novel-miR1, 2, 16, 19, 26, 27, 28, 31 and 32) were targeted to cytochrome P450-like TATA box binding protein (TBP). The qRT-PCR confirmed expression levels of several differentially regulated miRNAs. Expression patterns of four miRNAs-targets pairs were also examined via RNA deep sequencing. Together, the results imply that the regulation of miRNA expression might confer to the phenotype of the PHYA1 RNAi line(s) with improved fiber quality.

Introduction

Cotton (Gossypium hirsutum L.) is one of the most important cash crops around the world, providing the largest renewable natural fiber for the textile industry. The cotton fiber is a trichome cell derived from the epidermal layer of the ovule, and its development consists of four overlapping stages: initiation, elongation (primary wall biosynthesis), secondary wall biosynthesis and maturation [1]. Previous studies showed that silencing phytochrome A1 gene (PHYA1) by RNA interference generated RNAi cotton lines with improved fiber quality (longer, stronger and finer fiber) and vigorous shoot and root development, without adverse effects on the key agronomic traits (maturity and productivity) when compared to non-transformed Coker 312 [2]. In addition, field-trials of RNAi genotypes revealed increased photosynthesis and better adaptations to abiotic environmental stresses of RNAi genotypes compared to wild-types, a finding that may be associated with improved root development in PHYA1 RNAi cotton plants [3].

Phytochromes are red and far-red light photoreceptors, and a high ratio of far-red light to red light perceived by the photoreceptors increased fiber length [4]. Previous molecular mapping using phytochrome-based cleaved amplified polymorphisms (CAPS and dCAPs) markers suggested an importance of cotton phytochromes on regulation of some fiber quality traits including fiber length [2, 5, 6]. Phytochromes were reported to mediate plant hormone signaling through direct interaction between their negative regulators phytochrome-interacting factors (PIFs) and certain phytohormones signaling components, such as DELLA proteins [7]. It was well known that phytohormones play central roles in the regulation of cotton fiber development. Auxin, gibberellins (GAs), bassinosteroids (BR), and ethylene (ET) have been shown to promote fiber cell initiation and fiber development [811], whereas abscisic acid (ABA) and cytokines (CK) negatively affect fiber development [12]. Over the past decades, many genes have been reported to be involved in cotton fiber development. Genes encoding MYB transcription factors and sucrose synthase play important roles in fiber cell fate determination and fiber development [13]. Genes for cellulose synthases [14], vacuolar invertase, aquaporin, and lipid transfer proteins also play critical roles in the process of rapid fiber elongation [13] [1517]. Calcium and the second messenger molecule H2O2 might function as the terminal signal to the fiber elongation [18].

MicroRNAs are a family of small (21–25 nt) noncoding RNA molecules that regulate eukaryotic gene expression post-transcriptionally in a sequence-specific manner. The miRNA gene (MIR) is first transcribed into a larger primary transcript (pri-miRNA) by RNA polymerase II (pol II) [19]. The pri-miRNA is then processed into a 60–120 nt pre-miRNA with a stem-loop structure [20]. The pre-miRNA is further processed to form a miRNA:miRNA* duplex. In plants, these two successive cleavages occur in the nucleus and are facilitated by the dicer enzyme. The miRNA:miRNA* duplex is exported into cytoplasm by HASTY [21]. The RNA strand with the weakest 5´ end base pairing in the duplex is then selected as the mature miRNA, and the other strand, called miRNA* is degraded [22]. The mature miRNA is loaded to RISC (RNA-induced silencing complex) and guided by ARGONAUTE proteins (as parts of RISC) to target mRNA complementary sequences and trigger mRNA cleavage or translational inhibition [23]. miRNAs play regulatory roles in many aspects of plant development, hormone signaling, abiotic stress resistance such as heat, salinity and drought [24, 25]. More interestingly, miRNAs have also been shown to play crucial roles in cotton fiber development [26].

MicroRNA identification in cotton is lagging behind other plant species such as Arabidopsis and rice, which might be due to polyploidization events occurred in the cotton genome. Gossypium hirsutum, also known as upland cotton, is an allotetraploid (AADD, 2n = 4x = 52), arose from hybridization between a transoceanic dispersal of an A-genome progenitor (Gossypium arboretum, AA, 2n = 2x = 26) and a local D-genome progenitor (Gossypium raimondii, DD, 2n = 2x = 26). The genome sequences of two diploid cotton Gossypium raimondii [27] and Gossypium arboretum [28], and the allotetraploid Gossypium hirsutum [29, 30] have been determined. Whole genome sequences will definitely accelerate the identification and annotation of cotton miRNAs, especially cotton-specific miRNAs. Cotton miRNAs were first identified in 2007 by two groups using a comparative genomic approach [31, 32]. Thirty and thirty-seven miRNAs candidates were identified by the two groups, respectively. Besides homology searches on database, Abdurakhmonov et al. [33] used a direct cloning technique to identify three miRNAs in cotton. Recently high-throughput RNA sequencing have been adopted for identification and expression analysis of miRNAs in cotton. Through small RNA sequencing, 34 conserved miRNAs families were identified in G. hirsutum [34] and 22 conserved miRNA families were expressed in developing cotton ovules [35]. By using the same approach, Pang et al. identified 4 novel miRNA families and proved that miRNAs play a role in cotton fiber development [26]. Recently, 65 conserved miRNA families were identified in cotton leaf and ovule tissues and forty of the identified miRNAs were ovule specific, suggesting that these miRNAs may play important roles in ovule and fiber development [34, 36]. Using short fiber mutants, Naoumkina et al. [37] identified 24 conserved and 147 novel miRNA families and revealed that 4 miRNAs were involved in fiber elongation. In comparison to other plant species such as Arabidopsis or rice, less miRNAs had been identified in cotton. With its larger genome size, it is predicted that the cotton genome may encode more than the typical number of miRNAs, including some with unique roles in fiber development.

With the aim to better understand the altered molecular pathways in the PHYA1 RNAi line of tetraploid cotton, and because of potential importance of miRNAs in the complex developmental processes such as fiber quality, miRNA libraries were constructed in this study using fiber RNAs (at different development stages) extracted from the PHYA1 RNAi cotton line and its non-transgenic background Coker 312. The libraries were subjected to multiplex sequencing on the Illumina platform, and conserved and novel miRNAs were then identified and used for differential expression analysis. As results, a total of 77 known miRNAs from 61 different families and 35 novel miRNAs were identified in developing cotton fibers. Of these, 7 miRNAs were differentially expressed in the PHYA1 RNAi line. Our results implied that the regulation of miRNA expression might play an important role in cotton fiber development.

Materials and methods

Plant materials and plant growth

Gossypium hirsutum L. cv. Coker 312 and its phytochrome A1 (PHYA1) RNAi plants were grown in the field at the North Farm R. R. Foil Plant Science Research Center of Mississippi State University. The seeds of these RNAi cotton plants [2] were provided by Uzbekistan collaborator of this study through Technology Transfer Office of USDA-ARS, USA under the USDA-Uzbekistan cooperation programs. Flowers were tagged on the day of anthesis (0 DPA). Cotton bolls (5, 10, 15 DPA) were harvested with three biological replications. Fibers were carefully dissected from cotton bolls, immediately frozen in liquid nitrogen, and then stored at -80°C for small RNA extraction.

Isolation of small RNAs and library construction

The small RNAs were isolated from fiber tissues by using the mirPremier™ MicroRNA Isolation Kit (Sigma-Aldrich) according to the manufacturer’s instructions. The RNA concentration and purity were determined by using a Nanodrop 2000c spectrophotometer (ThermoFisher Scientific). The Next Multiplex Small RNA Library Prep Set (New England Biolabs) was used to convert isolated RNAs (500 ng) into barcoded small RNA libraries according to the manufacturer’s instructions. To determine the differential expression of miRNAs in fibers between the PHYA1 RNAi line and its non-transgenic background line Coker 312, 18 barcoded small RNA libraries with three biological replicates were constructed by using miRNAs extracted from fibers of the two lines at three different developmental stages of 5-, 10-, and 15-DPA, respectively. Size selection (~150 nt) of the libraries was performed using AMPure XP beads. The concentration and quality of the libraries were determined with a Qubit Fluorimeter (ThermoFisher Scientific) and the Bioanalyzer 2100 (Agilent Technologies). The libraries were pooled and single-end (1 X 50) sequenced on an Illumina HiSeq 2500 system performed by BGI. The sequencing reads were deposited into NCBI Sequence Read Archive (SRA) with accession number SRX2467209-17 (Gossypium hirsutum L. cv. Coker 312) and SRX2467222-30 (RNAi line), respectively. Data files are available at https://www.ncbi.nlm.nih.gov/Traces/study/?acc=SRP096134 (Coker 312) and https://www.ncbi.nlm.nih.gov/Traces/study/?acc=SRP096136 (RNAi line).

MicroRNA identification and target prediction

Total raw sequences generated from Illumina instrument were trimmed to remove low-quality reads and adaptor sequences using Trimmomatic [38], and FastQC (http://www.bioinformatics.bbsrc.ac.uk/projects/fastqc/) was used to test data quality. The miRBase database release 21 (http://www.mirbase.org/) and the Gossypium hirsutum L. acc. TM-1 genome sequence [30] were used as references for known conserved miRNA mapping and novel miRNA identification, respectively. SeqMan NGen and ArrayStar (DNASTAR, Inc.) were used for mapping of clean short reads (18–44 nt) to the reference and expression level analysis. The miRNA expression was normalized by the Reads Per Kilobase per Million mapped reads (RPKM) method [39].

Conserved miRNAs were identified by comparing to miRBase with the criterion that tags were similar to their homologues within two mismatches and without gaps [40]. Potential novel miRNAs were firstly BLAST verified against non-coding rRNA, tRNA, small nuclear RNA (snRNA), and small nucleolar RNA (snoRNA) in GenBank, tRNA database (http://gtrnadb.ucsc.edu), and Rfam database [41]. Sequences with no hits to known miRNAs or other cellular RNAs were mapped to the Gossypium hirsutum L. acc. TM-1 genome. The mireap 0.2 (https://sourceforge.net/projects/mireap/) was used to analyze pre-miRNAs [42]. The psRNATarget was used for target prediction [43]. Sequence folding prediction were made by MFOLD [44]. miRNAs with negative hairpins folding energy from -25.7 to -230.5 kcal mol-1 [45] were selected for further analysis.

Quantitative real-time PCR

Quantitative Real time PCR was performed to validate the expression of miRNAs identified from RNA seq. The isolated fiber small RNAs (500 ng) were first reverse-transcribed to generate cDNAs using a Mir-X miRNA First-Strand Synthesis Kit (Clontech Laboratories). The cDNAs were then used as template for quantitative real-time PCR on LightCycler 480 (Roche Holding AG) by using SYBR Advantage qPCR Premix (Clontech Laboratories). The qRT-PCR was carried out by pre-denaturation at 95°C for 15 seconds, following by 45 cycles of 3-step PCR with denaturation at 95°C for 5 seconds, annealing at 55–60°C for 15 seconds, and final extension at 72°C for 15 seconds. Gene expression levels were presented as fold-change and calculated using the comparative threshold cycle (CT) method as described [46] with U6 snRNA as the internal reference. The expression data of four target genes (Gh_Sca142710G01, Gh_Sca006071G01, Gh_A05G2828 and GhD07G0477) were extracted from high throughput RNA-seq for 10 DPA fibers of RNAi line and Coker 312 (unpublished data). All the primers used for miRNA qRT-PCR were listed in Supplementary S1 Table.

Results

High-throughput deep sequencing of 5-, 10- and 15-DPA fiber small RNA libraries from both Coker 312 and PHYA1 RNAi plants

The libraries were subject to next generation sequencing on HiSeq 2500. Libraries had a minimum 8,052,792 and a maximum 12,947,430 short reads extracted from raw data (Table 1). Adaptor sequences and low-quality reads composed of 39.54% ~ 65.63% of raw reads were removed, and the average length of short reads after trim was from 22.5 to 24.2 nt (Table 1). Within the trimmed reads, 8.9%, 8.9%, 9.6%, 10.3%, 10.7% and 10.4% were annotated for 5 DPA (CF5), 10 DPA (CF10) and 15 DPA (CF15) fiber from Coker 312 and 5 DPA (RF5), 10 DPA (RF10) and 15 DPA (RF15) fiber from the PHYA1 RNAi line, respectively (Table 2). Among all the reads, 6,566,588; 6,588,884; 5,025,362; 8,286,014; 6,426,397; and 4,741,820 unique reads were annotated as small RNAs, respectively (Table 2). Most of these reads were clustered into rRNA, tRNA, snRNA, snoRNA and scRNA and removed before miRNA analyses. Among the rest of the annotated reads, 7,004, 7,291, 5,704, 8,698, 6,942, and 5,559 reads were matched to miRBase database (Table 2). The size distribution of these small RNAs ranged from 15 to 30 nt, and the most abundant length of them was 24 nt in both Coker 312 and RNAi lines (S1 Fig).

Identification of differentially expressed known miRNAs in PHYA1 RNAi cotton fibers

Known and conserved miRNAs were identified by mapping to the miRBase database (release 21), and 77 known miRNAs belonging to 61 miRNA families were found and they are all present in both of the two cotton lines. Of these, 34 miRNA families were cotton-specific, and 7 miRNAs were differentially expressed (≥ 2-fold change, FDR ≤ 0.05) in the PHYA1 RNAi cotton. For majority of the miRNA families, only one member was identified. However, multiple members were identified in some miRNA families, which included miR7495, 390, 2949, 167, 399, 7486, 394, 7484, 396, 7492, 156, 827, and 7496 (Fig 1A).

thumbnail
Fig 1.

Differentially-expressed-known miRNA families (A) and novel miRNAs (B) in PHYA1-RNAi cotton (RF) relative to the non-transformed Coker 312 (CF) in fibers. Those with a fold change greater than 2 are displayed by darker columns.

https://doi.org/10.1371/journal.pone.0179381.g001

Among these conserved miRNAs, 39 miRNAs were up-regulated in PHYA1 RNAi compared to Coker 312, and 38 miRNAs were down-regulated in the RNAi line (Fig 1A). Seven of conserved miRNAs, including miR7503, miR7514, miR399c, miR399d, miR160, miR169b and miR2950, were differentially expressed (fold change ≥ 2) in the PHYA1 RNAi line compared to Coker 312. It will be interesting to determine whether miRNA-mediated gene regulation could confer to the phenotype of the RNAi line with improved fiber quality.

Identification of novel miRNAs in PHYA1 RNAi cotton fibers

In addition to successful identification of 77 conserved miRNAs in the PHYA1 RNAi and wild type lines, novel miRNAs in fibers were also identified in both RNAi and wild-type cotton lines. After removing known miRNAs and other small RNAs, the stem-loop secondary structure of novel pre-miRNAs was predicted by MFOLD. The stability of an RNA secondary structure is affected by the minimal folding free energy (MFE). The lower the MFE value, the more stable the RNA structure. One of the criteria to annotate a novel miRNA is that it was generated by precise excision from the stem of a stem-loop precursor [47]. A total of 35 novel miRNAs (named as novel-miR 1 to 35) were identified in the 18 fiber libraries (Table 3). The sizes of these novel miRNAs ranged from 21 to 24 nt. The 24 nt miRNAs were the most abundant with 27 newly discovered miRNAs. Five, two and one novel miRNAs were found to have the sizes of 21, 22 and 23 nt, respectively. The stem-loop precursors of these novel miRNAs predicted by MFOLD have negative free folding energies from -45.7 to -230.1 kcalmol-1, which are lower than those of tRNAs (-27.5 kcal mol-1) or rRNA (-33 kcal mol-1) (Table 3). All miRNAs* of these novel miRNAs have been identified in RNA sequencing. Among these novel miRNAs, none were significantly differentially expressed in PHYA1 RNAi fibers (≤ 2 fold change) (Fig 1B). The 35 novel miRNAs were identified in cotton fibers for the first time, and their mature miRNA sequences were listed in Table 3.

Prediction and annotation of targets for novel miRNAs

A plant miRNA generally has perfect or near perfect complementarity to its target, and the cleavage happens between the 10th and 11th nucleotides from the 5´ end of the miRNA. These characters were used to identify the targets of novel miRNAs [48]. By using psRNATarget, a total of 97 targets were predicted for the 35 novel miRNAs (Table 3). Among them, the targets of three novel miRNAs novel-miR14, 22, and 24 were not identified. The top-rated target genes for each of the novel miRNAs were listed in Table 3. The details of the miRNA silencing its target and the predicted cleavage/translation inhibition sites were also presented in Table 3. Among these targets, the gene encoding cytochrome P450-like TATA box binding protein (cytochrome P450-like TBP) has the highest hits, being the target of 9 different novel miRNAs which included novel-miR1, 2, 16, 19, 26, 27, 28, 31 and 32. Plant cytochrome P450s are involved in a wide range of biosynthetic reactions, including fatty acid conjugation, hormones synthesis, and generating defensive compounds. Of these 9 novel miRNAs, only novel-miR27 and novel-miR32 were accumulated slightly higher in the PHYA1RNAi line, and the other 7 miRNAs, novel-miR1, 2, 16, 19, 26, 28, and 31, were all slightly lower in RNAi cotton compared to Coker 312. The novel miRNAs were also predicted to silence an rRNA promoter binding protein and some transcription factors and related proteins, such as a bZIP transcription factor, a global transcription factor group isoform 1 and a WD40-like transcription factor. Genes for ribosomal proteins were also targeted by these novel miRNAs. In addition, many proteins related to translation, such as translation initiation factor and elongation factor 1α were predicted to be the targets of novel-miR2, 10, 26 and 27. The predicted targets of these novel miRNAs also include many enzymes such as ATP synthase, serine threonine-protein kinase, peroxidase, acetolactate synthase, lipase, endo-β-glucanase, E3 ubiquitin protein ligase, receptor-like protein kinase, CoA reductase, xyloglucan glycosyltransferase, methylase, and some functional proteins, like chaperons, senescence-associated proteins, calcium-binding proteins and the 14-3-3-like protein (Table 3).

Validation of differentially expressed miRNAs in PHYA1 RNAi line by qRT-PCR

The expression of twelve known miRNAs and seven novel miRNAs in 10-DPA fibers of PHYA1 RNAi and Coker 312 lines was also determined by qRT-PCR (Figs 2 and 3, S1 Table). The abundances of these miRNAs were relatively high in majority of the fiber libraries (Fig 1). The results showed that these 19 miRNAs were successfully detected by qRT-PCR and showed consistency with the expression profiles analyzed by small RNA sequencing (Figs 2 and 3). Among the known miRNAs, miR162a, 396a/b, 2950 and 160 were up-regulated, and miR172, 399d, 167a/b, 390a/b/c, 164, 166b, 399c, and 169b were down-regulated (Fig 2). For the novel miRNAs, novel-miR22 and novel-miR8 were up-regulated, while novel-miR5, 3, 6, 2, and 4 were down-regulated (Fig 3). The expression level of these novel miRNAs (except for novel-miR4) in the PHYA1 RNAi line was less than two-fold change when compared to Coker 312.

thumbnail
Fig 2. Validation of expression patterns of differentially-expressed-known miRNAs in 10 DPA-fiber of PHYA1-RNAi line by qRT-PCR.

The miRNA expression was represented as relative fold change 2-ΔΔCT (ΔΔCT = ΔCT RNAi - ΔCT Coker 312). The expression levels of miRNA were normalized by using U6 snRNA as a reference. Three biological replicates and three technical replicates were used for qRT-PCR analysis. The error bar represent the confidence limits. RF and CF denote fibers from RNAi and Coker 312 lines, respectively. MIR 169b was not detected in 10-DPA fiber by MIR Seq.

https://doi.org/10.1371/journal.pone.0179381.g002

thumbnail
Fig 3. Validation of expression patterns of novel miRNAs in 10-DPA fiber of PHYA1- RNAi line by qRT-PCR.

The miRNA expression was represented as relative fold change 2-ΔΔCT (ΔΔCT = ΔCT RNAi - ΔCT Coker 312). The expression levels of miRNA were normalized by using U6 snRNA as a reference. Three biological replicates and three technical replicates were included for qRT-PCR assays. The error bar represent the confidence limits. RF and CF denote fibers from RNAi and Coker 312 lines, respectively.

https://doi.org/10.1371/journal.pone.0179381.g003

Inverse expression patterns between four pairs of miRNAs and their targets.

Because miRNAs control the degradation of mRNA transcripts, the expression levels of miRNA and its target mRNA will be negatively correlated. To test this, the expression patterns of predicted targets of four novel miRNAs including novel-miR1, 2, 4 and 8 in 10 DPA fiber were analyzed by RNA sequencing and validated by qRT-PCR (Fig 4A). The results showed consistency between these two assays, and the relative expression levels of these miRNA targets were inversely correlated with the accumulation levels of corresponding miRNAs (Fig 4). For example, Gh_Sca142710G01 encoding an rRNA promoter binding protein was up-regulated in the PHYA1 RNAi plant (Fig 4A) compared to Coker 312, and its corresponding miRNA novel-miR1 was down-regulated (Fig 4B). Novel-miR1 also targeted to cytochrome P450_TBP (The counts was too low to count in RNA Seq), which plays a crucial role in photosynthesis [49]. In addition to novel-miR1, eight of the novel miRNAs identified in upland cotton were predicted to target cytochrome P450 TBP (Table 3). This implied that the better quality fiber of the PHYA1 RNAi line may be partially conferred through the miRNA regulation of cytochrome P450s, which are involved in a wide range of biosynthetic reactions, including fatty acid conjugation, hormones synthesis, and generating defensive compounds [50]. Thus, miRNA regulation on cytochrome P450 might play a positive role during fiber development.

thumbnail
Fig 4. Inverse correlation between the expression of four novel miRNAs and their target genes.

(A) RNA Sequencing and qRT-PCR analysis of predicted target genes of four novel miRNAs in 10 DPA fiber; (B) miRNA Sequencing and qRT-PCR analysis of four novel miRNAs in 10 DPA fiber. Gene expression was represented as fold change 2-ΔΔCT (ΔΔCT = ΔCT RNAi - ΔCT Coker 312). The expression levels of the genes were normalized by UBQ7 (Gen bank accession NO. DQ116441) and U6 snRNA as internal references for mRNAs and miRNAs, respectively. Three biological replicates and three technical replicates were included for qRT-PCR assays. The error bar represent the confidence limits.

https://doi.org/10.1371/journal.pone.0179381.g004

Discussion

Upland cotton (Gossypium hirsutum L.) produces the most commonly used textile fiber in the world. The improvement of cotton fiber quality using different genetics tools has long been a key interest for cotton breeders. Using RNAi technology, Abdurakhmonov et al. [2] had developed PHYA1 RNAi cotton plants with improved fiber length, strength and fineness. Recently, the GhMYB25 genes in allotetraploid cotton genome have been successfully targeted for mutagenesis using the CRISPR/Cas9 technology [51]. This finding indicates that the CRISPR/Cas9 system will be an excellent tool for functional analysis of cotton genes in A and D subgenomes and improvement of cotton fiber quality. In this study, a genome-wide miRNAome analysis was performed to identify differentially expressed miRNAs, which might reveal the molecular mechanisms of the fiber quality improvement by PHYA1RNAi in cotton. A total of 77 known miRNAs were identified in cotton fibers. Of these, 7 known miRNAs were differentially expressed (fold change ≥ 2) in the PHYA1 RNAi plant. Using the published draft genome sequence of the allotetraploid cotton TM-1 line [30] as a reference, 35 novel cotton-specific miRNAs had been also identified and their target prediction was achieved.

Identification of novel miRNAs in cotton

The utilization of allotetraploid TM-1 cotton genome as a reference is very useful for miRNA identification and target prediction. Without the complete genome sequence of Gossypium hirsutum, it is difficult to conduct miRNAome analysis through bioinformatics prediction. First of all, small RNA read libraries generated by previous studies might contain errors [52]. Secondly, it is hard to evaluate the possibility of MIR gene loci duplication in allotetraploid. Thirdly, the isolation/identification of miRNAs in cotton is lagging when compared to other plant species due to the complexity of cotton genome compensation. The major criterion for novel miRNA prediction is the hairpin structure formed by precursor miRNAs. Due to incomplete upstream and downstream sequences, it is difficult to predict the secondary structure of miRNAs [37]. Furthermore, the available transcript libraries in cotton research were mostly from Gossypium raimondii, and thus target prediction had generated extensive redundancy. In this study, the precursors were predicted and identified for all novel miRNAs by using the psRNATarget, and the redundancy of the targets was reduced by the second blasting with TM-1 as the reference, because many of the hits from distinct miRNAs turned out to be the same gene.

Using high-throughput small RNA sequencing, 89 conserved and 8 novel miRNAs have been recently identified in cotton [37]. Based on the annotation of whole genome sequence of the cultivated Gossypium hirsutum TM-1 upland cotton, 602 miRNAs [29] and 301 miRNAs [30] were identified, respectively, in the allotetraploid cotton by two research groups. As a result, further identification and functional validation of cotton miRNAs are still needed, particularly, for cotton specific miRNAs and fiber-development-related miRNAs. In this study, a total of 35 novel miRNAs were identified in the small RNA libraries of both RNAi and control genotypes. None of these novel miRNAs were differentially expressed in RNAi lines or was specific to RNAi genotype although there were subtle difference in expression levels. This is desired objective of any RNAi crop development that positively contributes to its biosafety concerns because if novel non-existed signature is generated by RNAi hairpins incretion into a genome, as a first step alert, RNAi product must be subjected for further risk assessment analyses [3].

Of the novel miRNAs, 16 miRNAs were in the D-subgenome, and another 15 miRNAs were in the A-subgenome. Most of miRNAs identified previously were located in the D-genome due to limited sequence information of upland cotton genomes [37]. An enlarged pool of sequenced genomes has increased the accuracy and coverage for bioinformatics prediction and analysis of miRNAs. The 24 nt miRNAs were the most abundant among the 35 novel miRNAs, which is consistent with previous studies [53, 54]. Unlike the 21 nt miRNAs associated with AGO1, the 24 nt miRNAs are loaded onto AGO4, which has a preference for sRNAs with 5´ terminal adenine [55, 56]. As expected, most of the 24 nt miRNAs (9) had a 5´ terminal adenine nucleotide, which is consistent with the previous study [54]. The predicted hairpin precursors had negative folding free energies from -45.7 to -230.1 kcal mol-1, which were in agreement with previous studies [55]. By base pairing to mRNAs, microRNAs mediate mRNA degradation or translational repression. Most of the novel miRNAs in this study inhibit target expression through mRNA degradation, and a few of them regulate the expression of the target genes through translational repression.

Cotton fiber elongation-related miRNAs

In this study a total of 77 known miRNAs belonging to 61 miRNA families were identified in elongating fibers. Many of these conserved miRNAs were present at lower levels in RNAi cotton compared to Coker 312, including Gh-miR2950, 169b, 160, 399c and 399d. In contrast, Gh-miR7514, a cotton specific miRNA, was highly enriched in RNAi cotton. The target (ES793451) of Gh-miR2950, also a cotton specific miRNA, was predicted to encode a putative gibberellin 3 hydroxylase, which accumulated at higher levels in fibers [26]. Gibberellin 3 hydroxylase had been shown to control internode elongation in pea [57]. It was reported that Gh-miR2950 might affect fiber cell elongation via GA signaling [26]. The results in this study are consistent with this hypothesis, in that down-regulation of Gh-miR2950 promotes fiber elongation in the PHYA1 RNAi cotton by increasing Gibberellin 3 hydroxylase activity and therefore increasing the levels of the biologically active gibberellin GA1 [58]. miRNA169, 160, 399 are highly conserved miRNAs in plants. It was reported that miR160 and miR169 were significantly expressed at low levels in fibers than in seedlings [59]. miRNA160 had been shown to target three auxin response factors ARF10, 16, and 17 [26]. It is well known that auxin plays an essential role in cotton fiber development [10, 11, 60]. The results in this study suggested that miR160 can promote fiber development via the auxin signaling transduction pathway through increased expression of ARF10, 16 and 17. miR399 was predicted to target a MYB transcription factor [54]. Many MYBs, such as MYB25 and 109, were reported to promote fiber elongation and their suppression caused shorter fibers. The ABA 8´ -hydroxylase catalyzed the first step of inactivation of ABA, and the active levels of ABA might increase by decreasing the expression of ABA 8´ -hydroxylase in the RNAi line. In this study, miR7514 was accumulated at higher levels in RNAi cotton, and it targeted the gene for Rho guanyl-nucleotide exchange factor 7 (RhoGEF7) [61]. The functional analysis of RhoGEF7 in plants has not been documented so far. However, it was reported that ABA could induce the degradation of another Rho guanyl-nucleotide exchange factor RhoGEF1 [62]. Taken together, the results in this study imply that miRNA mediated gene regulation may be involved in fiber improvement of the PHYA1 RNAi cotton. Although the seven known miRNAs were differentially regulated in fiber cells and possibly involved in fiber development, some of them were found to play roles in biotic and abiotic stress responses. In wheat, miR160 was regulated under drought stress [63], miR164 was involved in salt stress response [64], and miR169b was differentially expressed in response to fungal infection [65]. In addition, miR396 in opium poppy was involved in regulation of secondary metabolite synthesis [66]. Another miRNA miRRNVL5 in G. hirsutum was found to play a role in regulation of cotton response to salt stress [67].

Cytochrome P450 TBP has the most abundant hits when the genome was screened with psRNATarget, and it was the target of 9 novel miRNAs identified in this study. Of these 9 miRNAs, 7 of them had lower levels in RNAi cotton compared to Coker 312. Cytochrome P450 is involved in the biosynthesis of plant hormones (such as ABA, GA, BR) and secondary metabolites (e.g. phenylpropanoids, alkaloids, terpreoids) [68]. Cytochrome P450 enzymes also play critical roles in response to different abiotic and biotic stresses. The cytochrome P450 involved pathways were abundantly present in RNAi cotton, and miR172, a stress related miRNA was reported to target Cytochrome P450 in cotton and corn [69, 70]. Taken together, although the expression of these novel miRNAs were not significantly changed in developing fibers, the observation of cytochrome P450 TBP as the target of 9 miRNAs suggests that miRNA mediated gene regulation on cytochrome P450 associated pathways might play a role in improving fiber quality in the PHYA1 RNAi cotton.

The possible functions of cytochrome P450 in cotton improvement

One possible explanation to the highest hits of cytochrome P450 TBP in miRNA target prediction is the high constitution of cytochrome P450 gene (CYP) family in plant genomes. It was reported that CYP genes occupied approximately 1% of the protein coding genes in six sequenced angiosperms, which include grape (Vitis Vinifera), papaya (Carica papaya), poplar (Populus trichocarpa), rice (Oryza sativa), Arabidopsis (Arabidopsis thaliana) and moss (Physcomitrella patens) and were involved in almost every aspects of the plant life [50]. Based on evolutionary analysis, the oldest CYPs are involved in biosynthesis of carotenoid and sterol. The middle branches of CYPs mediate the adaptation to environment, including abiotic stresses and biotic defenses. The newest CYPs are involved in the synthesis of structural components such as lignin, fatty acid, pigments, odorants, and signaling compounds [50]. Besides with improved fiber quality, PHYA1 RNAi plants have vigorous root and vegetative growth, and early flowering due to increased photosynthesis [2, 3]. CYPs are light-driven oxidase enzymes in electron transfer chains [49]. Thus this study support the hypothesis that increased photosynthesis of RNAi cotton might be due to miRNA-mediated regulation on the expression of CYPs. In addition, CYPs are also involved in biosynthesis and catabolism of many plant hormones, such as ABA (CYP707), GA (CYP714) and BR (CYP724) [50]. ABA regulates osmotic stress tolerance [71], and GA plays a role in salinity adaptation [72]. As mentioned previously, CYPs also play crucial roles in environmental adaptation, which might explain the better tolerance of RNAi cotton to salinity, drought, and heat stresses.

Conclusion

Eighteen RNA libraries constructed using small RNAs from 5-, 10-, and 15-DPA fibers of PHYA1 RNAi and Coker 312 lines were sequenced using the Illumina HiSeq system. Sixty-one conserved miRNA families and thirty-five novel miRNAs were identified in the upland cotton lines. The targets of 6 conserved miRNAs, which expressed differentially in the RNAi line, were reported to participate in primary cell wall synthesis and phytohormone signaling pathways. The 35 novel miRNAs were identified in cotton for the first time, and their target genes were predicted. Nine novel miRNAs were identified to target cytochrome P450 TBP. Together, the results imply that miRNAs involved in fine-tune gene regulation might confer to the phenotype of the RNAi line with improved fiber quality.

Supporting information

S1 Fig. The small RNA length distribution in cotton fibers from both PHYA1 RNAi line (RF) and Coker 312 (CF).

https://doi.org/10.1371/journal.pone.0179381.s001

(TIF)

Acknowledgments

We thank Academy of Science of Uzbekistan for sharing RNAi cotton seeds to the partner labs of USDA-ARS, USA and Technology Transfer Office of USDA making these seeds available for this study. We also thank two Uzbek scientists, Mr. Bakhtiyor Rahmanov and Mr. Ozod Turaev in partner labs at USDA, who helped with sample preparation and evaluation of plants in the USA environment. We would also like to acknowledge the help and support of Dr. Kater Hake, Cotton Incorporated.

Author Contributions

  1. Conceptualization: SS JJ IA AP.
  2. Data curation: PD.
  3. Formal analysis: PD QM.
  4. Funding acquisition: DM SS JJ.
  5. Investigation: QM.
  6. Methodology: CH QM ZB.
  7. Project administration: DM.
  8. Resources: IA AP ZB.
  9. Supervision: DM.
  10. Validation: CH QM.
  11. Visualization: QM CH DM.
  12. Writing – original draft: QM DM.
  13. Writing – review & editing: DM.

References

  1. 1. Basra AS, Malik C. Development of the cotton fiber. Int Rev Cytol. 1984;89(1):65–113.
  2. 2. Abdurakhmonov IY, Buriev ZT, Saha S, Jenkins JN, Abdukarimov A, Pepper AE. Phytochrome RNAi enhances major fibre quality and agronomic traits of the cotton Gossypium hirsutum L. Nat Commun. 2014;5:3062. Epub 2014/01/17. pmid:24430163.
  3. 3. Abdurakhmonov IY, Ayubov MS, Ubaydullaeva KA, Buriev ZT, Shermatov SE, Ruziboev HS, et al. RNA Interference for Functional Genomics and Improvement of Cotton (Gossypium sp.). Front Plant Sci. 2016;7.
  4. 4. Kasperbauer MJ. Cotton fibre length is affected by far-red light impingping on developing bolls. Crop Sci. 2000; 40:1673–1678.
  5. 5. Abdurakhmonov IY. Molecular cloning and characterization of genomic sequence tags (GSTs) from the PHYA, PHYB, and HY5 gene families of cotton (Gossypium species): Texas A&M University; 2001.
  6. 6. Kushanov FN, Pepper AE, John ZY, Buriev ZT, Shermatov SE, Saha S, et al. Development, genetic mapping and QTL association of cotton PHYA, PHYB, and HY5-specific CAPS and dCAPS markers. BMC Genetics. 2016;17(1):141. pmid:27776497
  7. 7. Wang Q, Zhu Z, Ozkardesh K, Lin C. Phytochromes and phytohormones: the shrinking degree of separation. Mol Plant. 2013;6(1):5–7. pmid:22973064
  8. 8. Fang L, Tian R, Li X, Chen J, Wang S, Wang P, et al. Cotton fiber elongation network revealed by expression profiling of longer fiber lines introgressed with different Gossypium barbadense chromosome segments. BMC Genomics. 2014;15:838. pmid:25273845; PubMed Central PMCID: PMC4190578.
  9. 9. Beasley C, Ting IP. The effects of plant growth substances on in vitro fiber development from fertilized cotton ovules. Amer J Bot. 1973; 60(2):130–139.
  10. 10. Sun Y, Veerabomma S, Abdel-Mageed HA, Fokar M, Asami T, Yoshida S, et al. Brassinosteroid regulates fiber development on cultured cotton ovules. Plant Cell Physiol. 2005;46(8):1384–1391. pmid:15958497
  11. 11. Shi Y-H, Zhu S-W, Mao X-Z, Feng J-X, Qin Y-M, Zhang L, et al. Transcriptome profiling, molecular biological, and physiological studies reveal a major role for ethylene in cotton fiber cell elongation. Plant Cell. 2006;18(3):651–664. pmid:16461577
  12. 12. Lee JJ, Woodward AW, Chen ZJ. Gene expression changes and early events in cotton fibre development. Annals of Botany. 2007;100(7):1391–1401. pmid:17905721
  13. 13. Guan X, Song Q, Chen ZJ. Polyploidy and small RNA regulation of cotton fiber development. Trends Plant Sci. 2014;19(8):516–528. pmid:24866591
  14. 14. Paterson AH, Wendel JF, Gundlach H, Guo H, Jenkins J, Jin D, et al. Repeated polyploidization of Gossypium genomes and the evolution of spinnable cotton fibres. Nature. 2012;492(7429):423–427. pmid:23257886
  15. 15. Ma D-P, Liu H-C, Tan H, Creech RG, Jenkins JN, Chang Y-F. Cloning and characterization of a cotton lipid transfer protein gene specifically expressed in fiber cells. Biochim Biophys Acta (BBA)-Lipids and Lipid Metabolism. 1997;1344(2):111–114.
  16. 16. Ma D-P, Tan H, Si Y, Creech RG, Jenkins JN. Differential expression of a lipid transfer protein gene in cotton fiber. Biochimica et Biophysica Acta (BBA)-Lipids and Lipid Metabolism. 1995;1257(1):81–84.
  17. 17. Wang QQ, Liu F, Chen XS, Ma XJ, Zeng HQ, Yang ZM. Transcriptome profiling of early developing cotton fiber by deep-sequencing reveals significantly differential expression of genes in a fuzzless/lintless mutant. Genomics. 2010;96(6):369–376. pmid:20828606
  18. 18. Potikha TS, Collins CC, Johnson DI, Delmer DP, Levine A. The involvement of hydrogen peroxide in the differentiation of secondary walls in cotton fibers. Plant Physiol. 1999;119(3):849–858. pmid:10069824
  19. 19. Cai X, Hagedorn CH, Cullen BR. Human microRNAs are processed from capped, polyadenylated transcripts that can also function as mRNAs. RNA. 2004;10(12):1957–1966. pmid:15525708
  20. 20. Burke JM, Bass CR, Kincaid RP, Sullivan CS. Identification of tri-phosphatase activity in the biogenesis of retroviral microRNAs and RNAP III-generated shRNAs. Nucleic Acids Res. 2014. pmid:25428356
  21. 21. Park MY, Wu G, Gonzalez-Sulser A, Vaucheret H, Poethig RS. Nuclear processing and export of microRNAs in Arabidopsis. Proc Nat Acad Sci USA. 2005;102(10):3691–3696. pmid:15738428
  22. 22. Tomari Y, Matranga C, Haley B, Martinez N, Zamore PD. A protein sensor for siRNA asymmetry. Science. 2004;306(5700):1377–1380. pmid:15550672
  23. 23. Bushati N, Cohen SM. microRNA functions. Annu Rev Cell Dev Biol. 2007;23:175–205. Epub 2007/05/18. pmid:17506695
  24. 24. Khraiwesh B, Zhu J-K, Zhu J. Role of miRNAs and siRNAs in biotic and abiotic stress responses of plants. Biochim Biophys Acta (BBA)-Gene Regulatory Mechanisms. 2012;1819(2):137–148.
  25. 25. Chen X. Small RNAs and their roles in plant development. Annu Rev Cell Dev Biol. 2009;25:21–44. pmid:19575669
  26. 26. Pang M, Woodward AW, Agarwal V, Guan X, Ha M, Ramachandran V, et al. Genome-wide analysis reveals rapid and dynamic changes in miRNA and siRNA sequence and expression during ovule and fiber development in allotetraploid cotton (Gossypium hirsutum L.). Genome Biol. 2009;10(11):R122. pmid:19889219
  27. 27. Wang K, Wang Z, Li F, Ye W, Wang J, Song G, et al. The draft genome of a diploid cotton Gossypium raimondii. Nature Genetics. 2012;44(10):1098–1103. pmid:22922876
  28. 28. Li F, Fan G, Wang K, Sun F, Yuan Y, Song G, et al. Genome sequence of the cultivated cotton Gossypium arboreum. Nature Genetics. 2014;46(6):567–572. pmid:24836287
  29. 29. Li F, Fan G, Lu C, Xiao G, Zou C, Kohel RJ, et al. Genome sequence of cultivated Upland cotton (Gossypium hirsutum TM-1) provides insights into genome evolution. Nature Biotechnol. 2015;33(5):524–530.
  30. 30. Zhang T, Hu Y, Jiang W, Fang L, Guan X, Chen J, et al. Sequencing of allotetraploid cotton (Gossypium hirsutum L. acc. TM-1) provides a resource for fiber improvement. Nature Biotechnol. 2015;33(5):531–537.
  31. 31. Zhang B, Wang Q, Wang K, Pan X, Liu F, Guo T, et al. Identification of cotton microRNAs and their targets. Gene. 2007;397(1):26–37.
  32. 32. Qiu CX, Xie FL, Zhu YY, Guo K, Huang SQ, Nie L, et al. Computational identification of microRNAs and their targets in Gossypium hirsutum expressed sequence tags. Gene. 2007;395(1):49–61.
  33. 33. Abdurakhmonov IY, Devor EJ, Buriev ZT, Huang L, Makamov A, Shermatov SE, et al. Small RNA regulation of ovule development in the cotton plant, G. hirsutum L. BMC Plant Biol. 2008; 8:93. pmid:18793449
  34. 34. Ruan M-B, Zhao Y-T, Meng Z-H, Wang X-J, Yang W-C. Conserved miRNA analysis in Gossypium hirsutum through small RNA sequencing. Genomics. 2009;94(4):263–268. pmid:19628031
  35. 35. Kwak PB, Wang QQ, Chen XS, Qiu CX, Yang ZM. Enrichment of a set of microRNAs during the cotton fiber development. BMC Genomics 2009; 10:457. pmid:19788742
  36. 36. Xie F, Jones DC, Wang Q, Sun R, Zhang B. Small RNA sequencing identifies miRNA roles in ovule and fibre development. Plant Biotechnol J. 2015;13(3):355–369. pmid:25572837
  37. 37. Naoumkina M, Thyssen GN, Fang DD, Hinchliffe DJ, Florane CB, Jenkins JN. Small RNA sequencing and degradome analysis of developing fibers of short fiber mutants Ligon-lintles-1 (Li 1) and− 2 (Li 2) revealed a role for miRNAs and their targets in cotton fiber elongation. BMC Genomics. 2016;17(1):360. pmid:27184029
  38. 38. Bolger AM, Lohse M, Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics. 2014;30(15):2114–2120. Epub 2014/04/04. pmid:24695404
  39. 39. Mortazavi A, Williams BA, McCue K, Schaeffer L, Wold B. Mapping and quantifying mammalian transcriptomes by RNA-Seq. Nat Methods. 2008;5(7):621–628. Epub 2008/06/03. pmid:18516045
  40. 40. Yang X, Wang L, Yuan D, Lindsey K, Zhang X. Small RNA and degradome sequencing reveal complex miRNA regulation during cotton somatic embryogenesis. J Exp Bot. 2013;64(6):1521–1536. Epub 2013/02/06. pmid:23382553
  41. 41. Griffiths-Jones S, Bateman A, Marshall M, Khanna A, Eddy SR. Rfam: an RNA family database. Nucleic Acids Res. 2003;31(1):439–441. pmid:12520045
  42. 42. Li Y, Zhang Z, Liu F, Vongsangnak W, Jing Q, Shen B. Performance comparison and evaluation of software tools for microRNA deep-sequencing data analysis. Nucleic Acids Res. 2012;40(10):4298–305. pmid:22287634
  43. 43. Dai X, Zhao PX. psRNATarget: a plant small RNA target analysis server. Nucleic Acids Res. 2011;39(Web Server issue):W155–9. Epub 2011/05/31. pmid:21622958
  44. 44. Zuker M. Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res. 2003;31(13):3406–3415. pmid:12824337
  45. 45. Wang ZM, Xue W, Dong CJ, Jin LG, Bian SM, Wang C, et al. A comparative miRNAome analysis reveals seven fiber initiation-related and 36 novel miRNAs in developing cotton ovules. Mol Plant. 2012;5(4):889–900. Epub 2011/12/06. pmid:22138860
  46. 46. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2− ΔΔCT method. Methods. 2001;25(4):402–408. pmid:11846609
  47. 47. Meyers BC, Axtell MJ, Bartel B, Bartel DP, Baulcombe D, Bowman JL, et al. Criteria for annotation of plant MicroRNAs. Plant Cell. 2008;20(12):3186–3190. pmid:19074682
  48. 48. Gunawardane LS, Saito K, Nishida KM, Miyoshi K, Kawamura Y, Nagami T, et al. A slicer-mediated mechanism for repeat-associated siRNA 5'end formation in Drosophila. Science. 2007;315(5818):1587–1590. pmid:17322028
  49. 49. Jensen K, Jensen PE, Møller BL. Light-driven cytochrome p450 hydroxylations. ACS Chemical Biology. 2011;6(6):533–539. pmid:21323388
  50. 50. Nelson DR, Ming R, Alam M, Schuler MA. Comparison of cytochrome P450 genes from six plant genomes. Trop Plant Biol. 2008;1(3–4):216–235.
  51. 51. Li C, Unver T, Zhang B. A high-efficiency CRISPR/Cas9 system for targeted mutagenesis in cotton (Gossypium hirsutum L.) Sci Rep. 2017; 7: 43902. pmid:28256588
  52. 52. AC't Hoen P, Friedländer MR, Almlöf J, Sammeth M, Pulyakhina I, Anvar SY, et al. Reproducibility of high-throughput mRNA and small RNA sequencing across laboratories. Nature Biotechnol. 2013;31(11):1015–1022. pmid:24037425
  53. 53. Wang Z-M, Xue W, Dong C-J, Jin L-G, Bian S-M, Wang C, et al. A comparative miRNAome analysis reveals seven fiber initiation-related and 36 novel miRNAs in developing cotton ovules. Mol Plant. 2012;5(4):889–900. pmid:22138860
  54. 54. Xue W, Wang Z, Du M, Liu Y, Liu J-Y. Genome-wide analysis of small RNAs reveals eight fiber elongation-related and 257 novel microRNAs in elongating cotton fiber cells. BMC Genomics. 2013;14(1):1. pmid:24044642
  55. 55. Bonnet E, Wuyts J, Rouzé P, Van de Peer Y. Evidence that microRNA precursors, unlike other non-coding RNAs, have lower folding free energies than random sequences. Bioinformatics. 2004;20(17):2911–2917. pmid:15217813
  56. 56. Mi S, Cai T, Hu Y, Chen Y, Hodges E, Ni F, et al. Sorting of small RNAs into Arabidopsis argonaute complexes is directed by the 5' terminal nucleotide. Cell. 2008;133(1):116–127. pmid:18342361
  57. 57. Lester DR, Ross JJ, Davies PJ, Reid JB. Mendel's stem length gene (Le) encodes a gibberellin 3 beta-hydroxylase. Plant Cell. 1997;9(8):1435–1443 pmid:9286112
  58. 58. Hedden P, Kamiya Y. Gibberellin biosynthesis: enzymes, genes and their regulation. Annu Rev Plant Biol. 1997;48(1):431–460. pmid:15012270
  59. 59. Xie F, Zhang B. microRNA evolution and expression analysis in polyploidized cotton genome. Plant Biotechnol J. 2015;13(3):421–434. pmid:25561162
  60. 60. Kim HJ, Triplett BA. Cotton fiber growth in planta and in vitro. Models for plant cell elongation and cell wall biogenesis. Plant Physiol. 2001;127(4):1361–1366. pmid:11743074
  61. 61. Zhang Y, Wang W, Chen J, Liu J, Xia M, Shen F. Identification of miRNAs and Their Targets in Cotton Inoculated with Verticillium dahliae by High-Throughput Sequencing and Degradome Analysis. Int J Mol Sci. 2015;16(7):14749–14768. pmid:26133244.
  62. 62. Li Z, Waadt R, Schroeder JI. Release of GTP Exchange Factor Mediated Down-Regulation of Abscisic Acid Signal Transduction through ABA-Induced Rapid Degradation of RopGEFs. PLoS Biol. 2016;14(5):e1002461. pmid:27192441
  63. 63. Akdogan G, Tufekci ED, Uranbey S, Unver T. miRNA-based drought regulation in wheat. Funct Integr Genomics. 2016; 16: 221–233. pmid:26141043
  64. 64. Eren H, Pekmezci MY, Okay S, Turktas M, Inal B, Ilhan E, Atak M, Erayman M, Unver T.Hexaploid wheat (Triticum aestivum) root miRNome analysis in response to salt stress. Ann Appl Biol 2015; 167: 208–216.
  65. 65. Inal B, Türktas M, Eren H, Ilhan E, Okay S, Atak M, Erayman M, Unver T. Genome-wide fungal stress responsive miRNA expression in wheat. Planta. 2014;240:1287–1298. pmid:25156489
  66. 66. Boke H, Ozhuner E, Turktas M, Parmaksiz I, Ozcan S, Unver T. Regulation of the alkaloid biosynthesis by miRNA in opium poppy. Plant Biotechnol. J. 2015; 13:409–420. pmid:25735537
  67. 67. Gao S, Yang L, Zeng HQ, Zhou ZS, Yang ZM, Li H, Sun D, Xie F, Zhang B. A cottton miRNA is involved in regulation of plant response to salt stress. Sci Rep. 2016; 6:19736. pmid:26813144
  68. 68. Ramamoorthy R, Jiang SY, Ramachandran S. Oryza sativa cytochrome P450 family member OsCYP96B4 reduces plant height in a transcript dosage dependent manner. PLoS One. 2011;6(11):e28069. pmid:22140509
  69. 69. Zhang L, Chia J-M, Kumari S, Stein JC, Liu Z, Narechania A, et al. A Genome-Wide Characterization of MicroRNA Genes in Maize. PLoS Genetics. 2009;5(11):e1000716. pmid:19936050
  70. 70. An W, Gong W, He S, Pan Z, Sun J, Du X. MicroRNA and mRNA expression profiling analysis revealed the regulation of plant height in Gossypium hirsutum. BMC Genomics. 2015;16:886. pmid:26517985
  71. 71. Chinnusamy V, Schumaker K, Zhu JK. Molecular genetic perspectives on cross-talk and specificity in abiotic stress signalling in plants. J Exp Bot. 2004;55(395):225–36. pmid:14673035
  72. 72. Zhu W, Miao Q, Sun D, Yang G, Wu C, Huang J, et al. The mitochondrial phosphate transporters modulate plant responses to salt stress via affecting ATP and gibberellin metabolism in Arabidopsis thaliana. PloS One. 2012;7(8):e43530. pmid:22937061