Keywords

1 Introduction

Biotic and abiotic stresses, global climate change, and environmental pollution have a significant negative impact on crop yields. Together with rapid global population growth, these factors threaten global food security. The current world population of 7.3 billion is expected to reach 8.5 billion by 2030, 9.7 billion in 2050 and 11.2 billion in 2100, according to the most recent UN DESA report, “World Population Prospects: The 2015 Revision” [1]. Moreover, food demand is expected to increase by 59–98% between 2005 and 2050 [2]. In order to mitigate and control these diverse agricultural productivity constraints, extensive effort has been put into improving crop yield and stress-tolerance through traditional and molecular plant breeding. Traditional or conventional plant breeding is a time-consuming and labor-intensive approach. It is limited to the exchange of genes between fairly closely related species [3]. In order to overcome these hurdles, molecular breeding through gene manipulation has widely been used to develop new high-yielding, stress-tolerant crop varieties [4]. However, this technology has some limitations such as non-targeted and unanticipated effects [5]. Recently, new technologies such as Transcription Activator-Like Effector Nucleases (TALENs) and Clustered Regulatory Interspaced Short Palindromic Repeats (CRISPR)/Cas systems have emerged for genome editing [6]. However, harnessing the benefits of these technologies requires a complete understanding of the complexity of plant defense mechanisms and stress signaling pathways at the molecular level. This knowledge is crucial to enable development of high-yielding, stress-resistant crop plants with minimum yield penalty through selective genetic engineering and precise gene editing techniques [7].

Over the past 20 years, the field of gene expression profiling has undergone a dramatic revolution. Transcriptomics has witnessed remarkable success due to major advances in transcriptome sequencing and analysis technologies. A wide variety of molecular biology techniques have been used for expression profiling and transcription quantification. Traditional techniques such as northern blotting and in situ hybridization [8] and reverse transcription polymerase chain reaction (RT-PCR) [9] allow only single transcripts or small groups of transcripts to be analyzed at once. The real-time RT-PCR method is a medium-throughput and very sensitive technique for the detection of low-abundance mRNA. It has been widely used for absolute and relative gene expression quantification [10,11,12]. The development of microarrays in the mid-1990s revolutionized gene expression studies and provided a new tool for genome expression profiling by allowing large-scale analysis of thousands of genes simultaneously [13]. Microarrays have widely been employed to understand molecular mechanisms underlying plant development and response to a multitude of stresses [14,15,16,17]. More recently, next-generation sequencing (NGS) has essentially provided the second revolution since the development of microarrays. Through high-throughput sequencing, it has remarkably improved our understanding and knowledge of gene regulatory mechanisms and epigenetics [18, 19]. NGS-based RNA sequencing (RNA-seq) allows detection and quantification of known, novel and rare transcripts, genome annotation, and rearrangement detection to non-coding RNA discovery. Furthermore, it provides greater insights into biological pathways and molecular mechanisms that regulate cell fate, development, and disease progression [6, 18, 20].

Recent advances in transcriptomics technologies shed light on the dark intergenic regions between protein-coding genes traditionally referred to as transcriptional “noise,” “junk DNA,” or experimental artifact. In 2012, ENCODE (Encyclopedia of DNA Elements) declared that 80% of the human genome has a biochemical function. However, scientists of the ENCODE project recently got together in Potomac, MD, USA and claimed that ~50% is functional [21]. By contrast, Rands and Colleagues claim that 8.2% of the human genome is likely to be functional [22]. In comparison, only a tiny portion of the transcribed human genome (~1–2%) codes for proteins [23]. The number of protein-coding genes in the human genome is reported to be fewer than 20,000 genes and has continued to shrink [24]. According to the most recent estimate of the GENCODE annotation of the human genome, GENCODE release 24 (09.12.2015) corresponds to Ensembl 83, 84, and the human genome encompasses 19,815 protein-coding genes and 25,823 non-coding RNA genes (ncRNAs). Of these, 15,941 and 9,882 are long and small non-coding RNA genes, respectively. Moreover, many new non-coding RNA classes have been identified in the last few years and classified based on their distinct biogenesis pathways. Although little is known about plant genomics and plant genome composition, several recent studies have identified different types of non-coding RNAs with diverse functions in plants [25]. In this chapter, we discuss recent advances in our knowledge of the biological functions of mRNAs (with a particular focus on alternatively spliced mRNAs and polyadenylation) and ncRNAs (with a particular focus on miRNAs and siRNAs) in plant development and response to biotic and abiotic stresses. The purpose of this chapter is to provide an overview of important regulatory components, apart from pure mRNA expression, which has been studied for several decades.

2 Alternative Splicing, Alternative Polyadenylation, and Other Modifications of mRNA

The pre-mRNA containing introns can be alternatively spliced to generate multiple transcripts from a single gene through the differential use of splice sites that increase the transcriptome and proteome complexity of the cells and tissues [26, 27]. Eighty percent of the genes in plants and animals contain introns. Splicing, which is the removal of introns, is carried out by the spliceosome that surrounds the splice sites at each intron. The pre-mRNA (primary transcript) structure includes cis-elements such as the 5′ splice site, the branch-point that is close to the 3′end splice site, the polypyrimidine ring tract, and the 3′ splice site, which are required for splicing. The splice sites contain consensus sequences that are recognized by the spliceosome. The spliceosome, which is a complex ribonucleoprotein mega particle, consists of small nuclear ribonucleoproteins (snRNPs)- U1, U2, U4, U5, U6, and auxiliary factors, U2AF65 and U2AF35. Spliceosomes have a stronger affinity for some splice sites and a weaker affinity for others, and this phenomenon is important in alternative splicing (AS) [28]. The spliceosome recognizes these features and it applies two sequential trans-esterification reactions that ligate the selected exon sequences and remove the introns. The first step involves the nucleophilic attack by the 2′OH group of an important adenosine in the branch consensus site on the 5′ splice site, forming a branched RNA intermediate called intron lariat. After this, some of the snRNPs are released. In the second step, the 3′OH group of the upstream exon targets the 3′ splice site. This reaction results in the spliced mRNA and the intron lariat is removed and degraded [28]. The cis-regulatory sequences in the pre-mRNA (exon splicing enhancers (ESE), exonic splicing silencers (ESSs), intronic splicing enhancers (ISEs), and intronic splicing silencers (ISSs)) as well as the AS regulatory proteins (Ser/Arg-rich proteins (SRs), and heterogeneous nuclear ribonucleoproteins (hnRNPs)) are important regulators of the splicing process [29, 30]. The exon-exon junction complex (EJC) accumulates 24 nt upstream of the exon-exon junction for exportation of mRNA from the nucleus and for the cytoplasmic mRNA control [31]. The binding of the serine/arginine-rich (SR) family of splicing factors to ESE helps in the recruitment of the splicing components and prevents “exon skipping” [32]. The key studies on mRNA modification are summarized in Table 1.

Table 1 Key studies on mRNA modification, and miRNA and siRNA biogenesis
Full size table

2.1 Types of Alternative Splicing (AS)

Using different splice sites, the AS generates two or more mRNAs from the same pre-mRNA. Based on the type of AS, sequences of exons and introns are either included or excluded from the mRNA. The cassette exon is an exon that is included or excluded from the mRNA. Mutually exclusive exons refer to the splicing of the adjacent exon, causing only one of them to be included at a time in the mRNA. The alternative 5′ splice site involves the use of the distal or proximal 5′ splice site producing mRNAs of different size. The alternative 3′ splice site involves the exploitation of the 3′ proximal and distal splice sites, resulting in the production of mRNAs of different sizes. The final type of AS is the retention of an intron where the intron is retained or removed from the mRNA [27].

2.2 Coupling of Transcription with AS

More recently, it has become widely accepted that AS is a co-transcriptional event in which crosstalk is involved [33, 50]. The carboxy-terminal domain (CTD) involved in the coupling of transcription and processing steps is required for the recruitment of Ser/Arg-rich splicing factor 3 (SRSF3), which then inhibits the inclusion of alternative exons [51]. The mediator joins with the general transcription factors (GTFs) at promoters and specific TFs that are bound to gene enhancers, and recruits the negative splicing factor hnRNPL. This process causes hnRNPL to inhibit the inclusion of an alternative exon during splicing [52].

2.3 AS in Plants

AS is uncommon in unicellular eukaryotes and commonly found in multicellular eukaryotes and differs greatly among tissues and species [28, 53]. Only about 4% of the genes in budding yeast contain introns and AS is uncommon [54]. In comparison, the RNA structures containing exon-intron precincts, spliceosome components, and other splicing factors are commonly found in plants [27]. However, splicing in plants is unique due to their shorter introns compared to animals. Furthermore, intron retention is a common method of AS in plants and they contain more genes encoding Ser/Arg-rich (SR) proteins. The pre-mRNA of the spliceosomal proteins, particularly SR proteins, which have a key role in spliceosome assembly and splicing regulation, are extensively spliced. The availability of plant genome and transcript sequence data has allowed the global analysis of AS in many plant species. Genome-wide analysis of AS has been performed in model plants such as Arabidopsis [55], Brachypodium [56], and in crop plants such as rice [57] and soybean [58]. These studies have shown that plant genes have one or more alternative transcript isoforms (~20% of the genes) [59]. Studies have shown that nearly 61% of multiexonic genes in Arabidopsis and nearly 33% of rice genes are alternatively spliced [55, 57]. The AS of genes has been studied to understand their role in plant growth development, environmental changes, and stress responses [60,61,62]. The studies mentioned above and others support the importance of intron retention in plants.

2.4 Database Resources of Plant Spliceosomal Proteins and AS

There are several resources available in relation to plant spliceosomal proteins and AS. Many of them are based on the model plant, Arabidopsis, such as the Arabidopsis slicing-related genes (ASRG) [63], and The Arabidopsis Information Resource (TAIR) [64]. Others include the AS in plants (ASIP), which is available for Arabidopsis and rice [65].

2.5 Role in Plant Development

Whole transcriptome profiling using RNA-seq was useful in enhancing the understanding of the gene expression of key genes and the coordinated expression of related genes during early somatic embryogenesis in maize [66]. AS is important in photosynthesis as it generates two protein products from the Rubisco activase gene, which is a nuclear-encoded chloroplast protein that mediates light activation of ribulose 1,5-biphosphate carboxylase/oxygenase (Rubisco) [67]. Based on the cDNAs and ESTs of Arabidopsis and rice analyzed using genome-wide computational analysis, AS has been shown to be common during flowering [65]. The alteration from the vegetative to the reproductive developmental stages is regulated by the alternative processing of the FCA pre-RNA [68]. Further, AS of the transcripts controls the spatial and temporal production of the FCA protein that regulates flowering.

2.6 Role in Biotic and Abiotic Stress Response

The pre-mRNAs of spliceosomal proteins are severely affected by biotic and abiotic stresses leading to AS [69, 70]. For example, OSDREB2B was shown to be regulated by stress-inducible AS [71]. Furthermore, the AS of the NRR transcript (related to root growth) identified in rice, produced two 5′ co-terminal transcripts (NRRa and NRRb), and both products were shown to possess negative regulatory roles [72]. In another example, the TIR-NBS-LRR gene that is involved in tobacco mosaic virus (TMV) resistance produced two transcripts, NS and NL, through AS. Expression of both transcripts were shown to be required for complete resistance to the virus [73]. Similarly, the combined presence of RPS4 transcripts containing both full-length and truncated open reading frames was required to mediate disease resistance [74]. Abiotic stresses have been shown to affect the AS of the pre-mRNA in several SR genes [75]. The AS regulators such as the SR proteins, hnRNPs, and protein kinases have been suggested to play significant roles in stress responses [27]. They have been suggested to allow plants to react promptly in regulating splicing and gene expression.

2.7 Alternative Polyadenylation

The regulatory role of polyadenylation in eukaryotic gene expression involves alternative polyadenylation (APA) sites that produce different transcripts with altered coding capacity for proteins and/or RNA [34]. APAs have been reported in plants, in relation to flowering time control pathways [76], seed dormancy [77], and stress responses [78, 79]. It has been exhibited through global profiling methods that plants exploit APA for diversity generation in their transcriptomes. Through genome-wide analysis in Arabidopsis, the HLP1 protein was identified to regulate the pre-mRNA 3′-end processing and targets APA. It was enriched at transcripts involved in metabolism and flowering [76]. Another genome-wide study in Arabidopsis showed that the CPSF30 is associated with APA in response to oxidative stress [80]. The Plant APA is a recently developed database for the visualization and analysis of APA [81]. The role of AS in plant development and response to biotic and abiotic stresses is summarized in Table 2.

Table 2 Role of alternate splicing, miRNA and siRNAs in plant development, and response to biotic and abiotic stresses
Full size table

2.8 Modifications in mRNA

Recently, modifications of mRNA with N6-methyladenosine (m6A), 5-methylcytosine (m5C), and pseudouridine (ψ) have been revealed through technical advances. The m6A mRNA was the first internal mRNA modification to be identified. Due to its abundance it has been easily detected through bulk mRNA analysis, and NGS approaches have allowed the mapping of its locations [36]. Recently, transcriptome-wide m6A profiling of rice callus and leaf [35] and shoot [37] have been reported. There are, however, other RNA methylation events that have been found in other organisms and are thought to occur in plants. Subsequent studies will reveal their frequency of occurrence in plants and if they have any role in development or response to biotic/abiotic stresses.

2.9 Stress Response Mechanism and the Cytoplasmic RNA-Containing Granules

It is important in transcriptomics to study the mRNAs that are translated, degraded, or stored temporarily during stress [97]. Based on the environmental or developmental conditions, the messenger ribonucleoprotein complexes (mRNPs) are formed through transcribed mRNA. While the polysome-associated mRNAs are translated, the non-translated mRNAs are localized on either the mRNA processing body (PB) or stress granules (SG), which are cytoplasmic mRNP granules. The PB (identified in yeast and mammals) contains RNA decay machinery for destroying unwanted mRNA in the 5′-3′ direction. The SG store the non-translated mRNA that is stalled during initiation of translation and under stress conditions that cause the SG numbers to increase and accumulate. Several studies have suggested that SGs and PBs are an essential cytoplasmic structure that control gene expression during plant stress responses [98, 99].

3 microRNAs (miRNAs) and Small Interfering RNAs (siRNAs)

microRNAs (miRNAs, 19–25 nt) and small interfering RNAs (siRNAs, 21–22 nt) are small non-coding RNAs with important regulatory functions. Though miRNAs and siRNAs share a number of features in size, structure, and molecular function, they differ in biogenesis pathway and precursor structure [39, 49, 100, 101]. Both, miRNAs and siRNAs are capable of producing a gene silencing effect at the post-transcriptional and transcriptional (epigenetic regulation) levels [38, 47, 102]. In contrast to miRNAs that are derived from either double-stranded or hairpin-like (60–70 nt) RNA precursors in almost all eukaryotes [38], siRNAs are generated from long double-stranded RNAs [103]. The miRNAs are endogenous, encoded by the host genome, while siRNAs can be exogenous or endogenous in origin. The former is originally derived from the transcription of viruses, transposons, repetitive DNA sequences, or transgene trigger [104, 105]. The miRNAs have numerous targets and regulate the expression of large numbers of target mRNAs. In contrast, the siRNAs are specific and mostly regulate the same genes they originate from [106]. Another major difference between miRNAs and siRNAs is that siRNAs base-pair to their target gene and exert targeted gene knockdown through the siRNA-induced mRNA cleavage, translational repression, and DNA methylation, whereas the former are partially complementary to target mRNAs and mediate post-transcriptional gene regulation through either mRNA cleavage or translational repression [47, 106].

3.1 Biogenesis of miRNAs in Plants

miRNAs are evolutionary highly conserved RNA molecules [39]. In the nucleus, miRNAs are transcribed by RNA polymerase II into primary miRNA (pri-miRNA), which are capped and polyadenylated. Subsequently, the pri-miRNAs are processed by the ribonuclease III enzyme, Dicer like 1 (DCL1), in the Dicer family, to a smaller stem-loop structure called precursor miRNAs (pre-miRNAs) [38, 39]. The pre-miRNAs are further processed again by DCL1 into the mature miRNA: miRNA* duplexes that carry 5′ phosphates and 2-nt overhangs on their 3′ end that are not fully complementary [40]. Next, they are exported from the nucleus to the cytoplasm by HASTY, the Arabidopsis homolog of exportin 5 in animals [41, 42]. In the cytoplasm, the mature miRNA strand, the so-called guide strand, is subsequently incorporated into the RNA-induced silencing complex (RISC, or miRISC for miRNA-containing RISC), where it is loaded onto a member of the Argonaute protein family and guides effector RISC to the target mRNA. The miRNA*, which is derived from the other strand known as the passenger strand, can be either degraded or functional [43, 44]. The key studies on miRNA and siRNA biogenesis are summarized in Table 1.

3.2 Biogenesis of siRNAs in Plants

In contrast to miRNAs, siRNAs are derived from double-stranded RNAs originating from protein-coding genes, non-coding transcripts, and transposable elements with perfect base-pairing complementarity to target mRNAs [47]. The DCL2, DCL3, and DCL4 sequentially cleave the long duplex structure into 22-nt, 24-nt, and 21-nt siRNAs, respectively [39, 45, 46]. Short RNA duplexes are similar to the miRNA:miRNA* duplexes but are fully based-paired along the length. Once small RNA duplexes are generated, they are also loaded on an Argonaute protein. Next, the passenger strand is removed and the remainder forms the effector RNAi complex RISC (siRISC, which is loaded with siRNA [39]). Besides the RNA-dependent RNA polymerase 2 and 6 (RDR2, RDR6), SUPPRESSOR OF GENE SILENCING 3 (SGS3), and dsRNA-BINDING 4 (DRB4) are also implicated in siRNA biogenesis [107, 108].

Endogenous siRNAs in plants have been characterized based on their characteristics and biogenesis pathways into hairpin-derived siRNAs (hp-siRNAs, 21–24 nt), natural antisense siRNAs (nat-siRNAs, 21–24 nt), secondary siRNAs, and heterochromatic siRNAs (het-siRNAs, 24 nt) [47,48,49]. Secondary siRNAs could be subclassified into trans-acting siRNAs (tasiRNAs), phased siRNAs (phasiRNAs), and epigenetically activated siRNAs (easiRNAs) [45, 47].

3.3 Functions of miRNAs and siRNAs in Plants

A large number of miRNAs have been identified and characterized in plant genomes with diverse functions. Several miRNAs have been identified to have a crucial regulatory role in a wide range of biological processes in diverse plant species including but not limited to, cell signaling, differentiation, heterosis, DNA damage repair, hormone signaling, organ development, and response to biotic and abiotic stresses (reviewed in [47, 109, 110]).

The siRNA are widespread and numerous endogenous siRNAs have been discovered in plants by deep sequencing. The results of several studies revealed that siRNA is implicated in the morphological control of leaf [111]; developmental timing (temporal regulation) [112, 113]; hormone signaling [114]; fertility and reproductive function [115]; maintenance of genomic integrity, and developmental patterning [116].

Furthermore, both miRNAs and siRNAs, as part of multi-layered sophisticated defense mechanisms, play pivotal roles in regulating immune responses to environmental stresses [109, 116,117,118]. The role of miRNAs and siRNAs in plant development, and response to biotic and abiotic stresses is summarized in Table 2.

3.4 Role of miRNAs in Plant Stress Responses

Plants as sessile organisms are continuously and simultaneously challenged by multiple biotic and abiotic stressors. They have evolved sophisticated defense mechanisms and intricate regulatory networks to perceive their attackers. The miRNAs as critical regulators of gene expression, fine-tune defense responses by regulating the expression of their stress/defense-related target genes. Thousands of miRNAs responsive to biotic and abiotic stresses and their targets have been identified in diverse plant species using deep sequencing technologies and degradome sequencing, respectively [119].

3.5 miRNAs in Biotic Stress

miRNAs orchestrate plant adaptive response to pathogens as the key players in hormone signaling pathways and plant immunity [89, 95, 117, 120]. Thus far, numerous biotic stress-responsive miRNAs have been identified in different plants [93, 94, 120,121,122]. Recently, several miRNAs such as miR156, miR159, miR172, miR319, and miR393 were found to be responsive to Cucumber mosaic virus in tomato [93]. In response to the fungal pathogen Verticillium longisporum, several miRNAs including the miR319 family have been identified in oilseed rape (Brassica napus) [121]. Flagellin-22 triggered miR393 expression and conferred resistance to Pseudomonas syringae in Arabidopsis [95]. In line with this, several miRNAs have been identified to be differentially expressed in Arabidopsis in response to a bacterial pathogen, P. syringae pv. tomato, using deep sequencing [122]. The results of these studies indicate that miRNAs target genes that are related to hormone signaling pathways and negatively regulate their target genes to enhance plant resistance to bacterial infection. The miR393 was reported to target TIR1 (Transport Inhibitor Response 1) and its functional paralogs, AFB2 and AFB3 (Auxinsignaling F-Box proteins 2 and 3). Whereas miR160 and miR167 target ARF8, ARF10, ARF16, and ARF17 to repress auxin signaling. Therefore, miRNAs confer a high degree of resistance to the bacterial pathogen P. syringae through miRNA-mediated suppression of auxin signaling [95, 122]. A recent study found that there is a negative correlation between miR319 and its target TEOSINTE BRANCHED1/CYCLOIDEA/PROLIFERATING CELL FACTOR 4 (TCP4) in response to root-knot nematode (RKN, Meloidogyne incognita) invasion in tomato (Solanum lycopersicum var Castlemart) [94]. The TCP genes encode plant-specific transcription factors that positively regulate jasmonic acid (JA) biosynthesis genes and JA levels in plants. The expression of miR319b was repressed, while the expression of its target, TCP4, was increased under JA treatment. On the other hand, the expression levels of all miR319-targeted TCP genes were significantly decreased in transgenic tomato plants overexpressing miR319 [94]. The results of this study showed that miR319 negatively regulates RKN resistance and JA-mediated miR319 confers systemic resistance to RKN infection. Additionally, crosstalk between miRNAs and hormone signaling pathways was revealed.

Altogether, miRNAs are responsive to a broad range of biotic stresses and confer resistance to plants against pathogens through complex mechanisms such as miRNA-mediated hormone signaling and/or hormone-mediated miRNA regulation.

3.6 miRNAs in Abiotic Stress

Several studies have shown that miRNA expression is regulated in response to a wide array of abiotic stresses such as drought, salinity, cold, heat, heavy metals, nutrients, oxidation, hypoxia, and UV-B in an miRNA-, stress-, tissue-, and genotype-dependent manner (reviewed in [109, 118, 119, 123]). The miR319 miRNA family is one of the most conserved and ancient miRNA families in plants [83]. miR319 was found to be induced in response to not only different biotic stresses, e.g., bacteria, fungi, viruses, and nematodes [93, 94, 121, 122], but also to multiple abiotic stress factors such as drought, salinity, cold, and aluminum [82, 83, 124,125,126,127]. Constitutive expression of miR319 significantly increased salt and drought tolerance in transgenic creeping bentgrass (Agrostis stolonifera) [82, 83]. Hence, miR319 can be a general multi-stress responsive miRNA. A recent study in Arabidopsis revealed that three miRNAs from the miR319 family, i.e., miR319a/b and miR319b.2, are associated with several abiotic stresses [84]. Interestingly, miR319a and miR319b exhibited the same patterns of expression in response to copper, cadmium, and sulfur deficiency conditions as well as salt stress. Moreover, the expression of miR319b.2 was augmented in response to copper, cadmium, and sulfur deficiency stresses, whilst it was down-regulated in response to drought, heat, and salinity. Similarly, the expression levels of miRNA319a/b were increased under metal stresses. On the other hand, miRNA319a/b was notably up-regulated under salinity stress [84]. These results suggest that miRNAs appear to have a complex regulatory role and orchestrate defense responses to a wide range of abiotic stresses through different regulatory networks.

In addition to the miR319 family, the miR169 family is another highly conserved family that plays a critical role in response to abiotic stresses in several plant species. The results of several studies indicated that miR169 plays an important role in response to several abiotic stresses including drought, salt, cold, abscisic acid, nitrogen starvation, and phosphate deficiency [85, 86, 88, 128,129,130].

The miR169 was repressed under drought and phosphate deficiency in Arabidopsis and nitrogen-starvation in Arabidopsis and maize [85,86,87,88]. In contrast, miR169 was up-regulated in response to cold stress in different plant species [130]. Overexpression of miR169 enhanced drought tolerance in Solanum lycopersicum [122]. The miR169 family members are up-regulated in Arabidopsis, maize, and soybean under cold, drought, and salinity stresses [129]. Their results showed that stress-induced miR169 promotes early flowering by repressing the AtNF-YA transcription factor [129]. Conversely, a recent study in Solanum tuberosum exhibited that downregulation of miR169 enhanced drought resistance through over-expression of StNF-YA genes [90]. Nuclear factor Y (NF-Y) transcription factors are the main targets of miR169. NF-Y encodes a CCAAT-binding transcription factor [86]. These findings revealed that there is a negative correlation between the expression of miR169 and its target NF-YA genes, and the miR169 regulates negatively and/or positively their target expression at the post-transcriptional level to enhance stress tolerance in different plant species [90, 129]. Taken together, these results suggest that multi-stress responsive miR169 may orchestrate the expression of its target genes in a host- and stress-dependent manner. Different signaling pathways are mediated by miR169 and there is a complex crosstalk between the miR169 family members and their target transcription factors.

3.7 Role of siRNAs in Plant Stress Responses

Several studies have indicated that natural antisense transcripts (NATs) play a vital role in the regulation of defense signaling pathways and are involved in the response to different environmental stimuli by orchestrating corresponding NAT mRNAs [91, 96, 131].

3.8 siRNAs in Biotic Stress

nat-siRNAATGB2, the first endogenous siRNA, was specifically expressed in Arabidopsis thaliana leaves challenged with a virulent form of the bacterial pathogen Pseudomonas syringae pv. tomato (Pst) carrying effector avrRpt2 [96]. The nat-siRNAATGB2 and its antisense target PPRL were transcribed in the opposite direction and a negative correlation was observed between the nat-siRNA and its antisense target expression in response to P. syringae infection. nat-siRNAATGB2 induced resistance by repressing the expression of PPRL as a negative regulator of the RPS2-mediated resistance against Pst that triggered hypersensitive response (HR) and cell death by recognition of PstavrRpt2 effector [96]. A novel class of siRNAs known as long siRNAs (lsiRNAs, 30–40 nt) was discovered by Katiyar-Agarwal and colleagues in 2007. The results of this study revealed that lsiRNAs are stress-induced and expressed by a bacterial infection. AtlsiRNA-1 was remarkably and specifically over-expressed in response to Pst carrying effector avrRpt2. Overexpression of AtlsiRNA-1 repressed the expression of its target AtRAPmRNA and induced resistance by silencing AtRAP as a negative regulator of plant defense [132].

Using deep sequencing, 17,000 unique siRNAs corresponding to cis-NATs have been found in Arabidopsis thaliana in response to biotic stress in the form of bacterial infection and abiotic stresses in the form of cold, drought, and salt [72]. The results of these studies suggest that siRNAs are stress-induced and regulate defense response in plants through the reprogramming of gene expression.

3.9 siRNAs in Abiotic Stress

A large number of nat-siRNAs have been identified in rice (Oryza sativa cv. japonica) in response to cold, drought, and salt [133]. In Arabidopsis thaliana, the 21-nt nat-siRNAs repressed the expression of Δ1-pyrroline-5-carboxylate dehydrogenase (P5CDH), a stress-related gene, through mRNA cleavage under salt stress. Down-regulation of the P5CDH led to proline accumulation. Proline is as an osmoprotectant and ROS quencher that helps to tolerate salt stress, although under-expression of P5CDH instigated increased ROS production [91]. One study has indicated that four siRNAs were differentially expressed in response to cold, heat, salt, and drought in wheat (Triticum aestivum) [92]. The siRNA 002061_0636_3054.1 was significantly repressed by heat, salt, and drought stress; 005047_0654_1904.1 was strongly over-expressed in response to cold, whilst down-regulated in response to heat, salt, and drought stress; 080621_1340_0098.1 was faintly induced by cold and repressed by heat but not by either salt or drought stress; and 007927_0100_2975.1 was down-regulated by cold, salt, and drought stress [92]. Further, their results revealed that the four siRNAs were preferentially expressed in spikes and uniformly expressed in leaves and roots [92]. Therefore, the results of these studies exhibited that nat-siRNAs respond to biotic and abiotic stress conditions in a stress-specific and developmental stage-dependent manner.

4 Conclusions and Future Prospects

Alternative splicing, miRNAs, and siRNAs play critical regulatory roles in modulating gene expression during plant growth and development, in response to biotic and abiotic stresses, and plant adaptation to an ever-changing environment. Several small regulatory molecules have been identified to have versatile functions in food and feed crops. Manipulating expression levels of miRNAs and siRNAs in economically important crop plants can be an effective strategy to improve desirable traits, stress tolerance, and resiliency in response to environmental stress and pathogen attack in plants. Therefore, miRNAs and siRNAs can be used as new targets for developing trait-improved crop plants and improving plant tolerance to stresses. Two powerful genome editing tools, TALENs and CRISPR/Cas, can be used for targeted genome editing and knockdown/knockout of small RNAs.

In addition to the identification of small regulatory molecules and their transcriptional profiling, it is indispensable for scientific communities to understand the regulatory mechanisms of small RNAs that orchestrate cellular functions and adaptation to environmental stresses to minimize the unintended side effects in modified plants.