1. Introduction
Transcription in bacteria and archaea is carried out by a single multimeric RNA polymerase, while most eukaryotes contains three multimeric complexes (RNA pol I, II, and III) [
1,
2,
3]. Furthermore, plants contain two additional RNA pols (IV and V) that have evolved from RNA pol II [
4,
5,
6,
7,
8,
9,
10,
11,
12]. RNA pol I consists of 14 subunits and synthesizes precursor rRNA 45S (35S in yeast) of the three largest rRNAs [
3,
13,
14,
15]. RNA pol III contains 17 subunits and transcribes tRNAs, 5S rRNA and other non-coding RNAs [
13,
16,
17,
18]. RNA pol II is composed of 12 subunits and synthesizes mRNAs and some non-coding RNAs [
13,
19,
20,
21]. Plant-specific RNA pol IV and V, which have evolved from RNA pol II through duplication and functional divergence, also contain 12 subunits. These two enzymes, that are involved in epigenetic regulation synthesize siRNAs, which play roles in transcriptional silencing via RNA-directed DNA methylation (RdDM), and non-coding RNAs participating in plant growth, development, response to environmental changes or plant immunity [
4,
5,
6,
7,
8,
9,
10,
11,
12,
22,
23,
24].
RNA pol II, IV and V contain specific conserved subunits that may specialise some of their functions [
5,
22,
25]. This is the case of subunits NRPD1, NRPE1 and NRPB1 which correspond to RNA pols IV, V and II, respectively, in
Arabidopsis thaliana. In addition, there are other subunits that are common to RNA pol IV and V, but are conserved in RNA pol II, such as the subunits NRPDE2 and NRPB2, RNPDE4 and NRPB4, and NRPDE7 and NRPB7, which are shared by RNA pol IV and V and conserved with RNA pol II [
5,
22,
25,
26,
27]. Furthermore, several isoforms of the common subunit five, shared by all the RNA pols have been described, while a specific isoform, NRPE5, has been found for RNA pol V [
28]. In addition, several paralogues have been described for these and other subunits in different plants [
22,
25,
26,
28,
29,
30]. Based on the existence of these paralogues, it has been proposed that these may perform new functions or be subject to different regulation. This is indeed the case of the distinct isoforms of the shared subunits from the RNA pols in cultivated olive trees ‘Picual’ (
Olea europaea L. cv. Picual) [
28].
RNA pol IV and V have been reported to be involved in the biogenesis and functionality of 24-nt siRNA, which participates in RdDM [
9,
12,
25,
30]. RNA pol IV and V have also been proposed to participate in the transcription of long non-coding RNAs (lncRNAs) [
24,
31,
32,
33,
34,
35,
36,
37,
38]. Some of these lncRNAs are the intermediary of siRNA and are found within intergenic regions [
34]. Although lncRNAs are also transcribed by RNA pol II, those synthesised by RNA pol IV and V show some structural differences as regards the RNA pol II ones, such as lack of poly-A at the 3’ end region or lack of introns [
4]. lncRNAs transcribed by RNA pol IV and V are poorly characterized, in part because of their low expression and instability [
34,
39]. However, well-studied examples of non-polyA lncRNAs have been reported [
40,
41,
42]. Notably, the synthesis of non-polyA lncRNAs can be regulated by environmental conditions, as demonstrated in
A. thaliana under abiotic stress [
43,
44,
45].
In this work we searched for genes that putatively encode specific subunits of RNA pol IV and V, and for those corresponding to RNA pol II, in the olive ‘Picual’ cultivar given its economic, agronomic and agro-ecological importance as one of the most important fruit trees in the Mediterranean Basin [
28,
46,
47,
48]. The analysis allowed to identify paralogues for NRPD1, NRPE1 and NRPB1, NRPDE2 and NRPB2, NRPDE4 and NRPB4, NRPDE7 and NRPB7, and also for NRPB7-like, in addition to the putative pseudogenes, according to our transcriptomic analyses. The transcriptional studies from RNA-Seq data evidenced an increase for the corresponding RNA pol IV and V genes during fruit development. Furthermore, given the known role of RNA pol IV and V in the transcription of lncRNAs, we studied the lncRNA transcriptome during fruit development, which revealed important changes in their expression and differed based on the analysed lncRNA type. Accordingly, our data point to RNA pol IV and V involved in the regulation of lncRNAs during fruit development.
4. Discussion
The role of RNA pol IV and V in plants is still under study and is not fully understood. The olive tree is an important crop and is more complex than the model plant A. thaliana. In this work, we searched for the genes encoding the specific subunits of RNA pol IV and V, as well as the RNA pol II from which they have evolved. A comprehensive analysis of the expression profile of these genes was also performed, revealing an increase of mRNA expression for RNA pol IV and V subunits during fruit development. Furthermore, as RNA pol IV and V mediate lncRNA transcription, we analysed lncRNA transcriptome genome-wide and found a positive correlation. In concordance, our data suggest a role for RNA pol IV and V during fruit development through the expression of lncRNAs.
Several genes for the specific RNA pol IV and V subunits, and their RNA pol II counterparts, were identified by a blast-p search with the corresponding
A. thaliana subunits [
7,
12,
22]. Protein identity with
A. thaliana homologues varied within the 40-55% range for RNA pol IV and V subunits, except for NRPDE2 with 73% identity (
Figure 2). In contrast, the NRPB homologues exhibited notably higher identity levels (79-95%) indicating greater conservation compared to the NRPD and NRPE subunits that have allowed major variation during evolution. RNA pol IV and V have evolved from RNA pol II [
4,
5,
22,
62,
63], and have apparently evolved more rapidly than the RNA pol II because their
A. thaliana and
O. europaea sequences have diverged more.
Several paralogues of the different specific RNA pol II, IV and V subunit genes were found for all the subunits, except for RNA pol IV/V subunit NRPDE2, which had only a single gene (
Table 1). This was not the only case for the NRP1, NRP2, NRP4 and NRP7 subunits identified in this work, but also for an additional specific RNA pol V subunit (the previously described NRPE5) which has a paralogues for additional common subunits for all the RNA polymerases [
28]. The presence of several paralogues is found for many genes in olive, as it is case for the RNA pol subunits shared by the five RNA pols [
28]. These results are consistent with the olive cultivar genome that results from two independent whole-genome duplication (WGD) events, in addition to recent partial genome duplications [
52,
59]. In addition, several paralogues for RNA pols subunits have been identified in other organisms [
1,
5,
22,
25,
64].
It has been demonstrated that RNA pol IV and V play a role in silencing, plant growth, development, response to environmental changes or plant immunity [
4,
5,
6,
7,
8,
9,
10,
11,
12,
22,
23,
24]. Therefore, we can speculate that the regulation of the gene expression of these RNA pols could be expected in response to some growth conditions. The transcriptomic analysis performed by several RNA-Seq experiments during stress or developing processes using the olive cultivar ‘Picual’ [
50,
51,
61,
65] has shown that all the genes coding for the specific subunits identified in this work are expressed under these conditions. Similarly, the genes corresponding to the RNA pol II subunits were also expressed. Although no common pattern for the changes in the expression of the RNA pol II, IV or V genes was observed, some cases of regulation by plant tissue were evident. Remarkably, two RNA pol II (NRPB7) genes presented strict organ specificity and were expressed only in flowers (
Figure 3 and
Figure 4). Furthermore, the response to biotic
V. dahliae infection [
61] or abiotic stresses like cold [
51] or root injury [
61] showed null or weak changes in the expression profile of most genes.
However, and notably, clear gene regulation occurred for the RNA pol IV and V subunits during fruit development. A consistent and significant overexpression of the RNA pol IV and V genes was observed at the beginning of fruit development, contrasting with the behaviour of the RNA pol II subunits. This observation suggests a possible role of RNA pol IV and V during fruit development in agreement with the role of RNA pol IV and V during plant development and plant growth [
12]. According to the role of RNA pol IV and V during the synthesis of ncRNAs, relevant changes in the lncRNAs expression pattern were observed. In line with this, the synthesis of the non-polyA lncRNAs has been demonstrated to be regulated by environmental conditions in
A. thaliana under abiotic stress [
43,
44,
45]. Indeed, we identified 2,303 lncRNAs (non-polyA) transcripts in flowers and fruit at 15 days AFB. The majority of these transcripts were intergenic, while intronic and antisense lncRNAs were less frequent (
Figure 5). Relevant changes in expression were found, with 284 lncRNA transcripts expressed only in flowers and 264 only in fruit at 15 days AFB (
Figure 6). Furthermore, a tendency toward an increasing expression was noted in the intronic and intergenic lncRNAs, although this was not observed in the antisense lncRNAs. Considering the role of RNA pol IV and V in the synthesis of the small ncRNAs involved in silencing, we cannot rule out that some of these lncRNAs could be processed to small ncRNAs, which has been reported for other plants [
40,
66]. Indeed this fact has been observed for intergenic lncRNAs in
A. thaliana [
40] and, in our case, in 39 intergenic transcripts in olive, which were also identified as putative siRNAs.
In summary, our study identified the genes that code for specific RNA IV and V subunits, and the corresponding ones in RNA pol II, in olive cultivar ‘Picual’. The expression analysis performed of different organs/tissues, responses to biotic and abiotic stresses and of development process revealed that the expression of the RNA pol IV and V genes was induced during the early stages of fruit development. This induction was accompanied by relevant changes in the expression of lncRNAs, particularly an increase in the intergenic and intronic lncRNAs. These changes in the expression of lncRNAs may be important for controlling gene expression during fruit development. In addition, certain intergenic transcripts are susceptible to be processed and to become siRNAs, which are known to play a role in the gene expression control. This reinforces the hypothesis that RNA pol IV and V may contribute to the process of fruit development throughout the synthesis of lncRNAs.