2.1.3. Roles of Plant Organellar DEAD-Box RHs in Salinity Response

Another group of proteins involved in the control of OGE, which are hence experimentally related to the stress responses mediated by chloroplasts and plant mitochondria, is that of DEAD-box RNA helicases (DEAD-box RHs), the largest known RNA helicase subfamily. The helicase core of DEAD-box RHs is composed of the N and C terminal domains, which are structurally similar to recombination protein RecA, in which nine conserved motifs can be identified. Motif II of the N-terminal domain contains the amino acid sequence Asp-Glu-Ala-Asp (D-E-A-D), which gave a name to the subfamily. Variations in motif II determine three related RNA helicase subgroups, namely DEAD, DEAH, and Ski2, whose members are referred to as DExD/H-box proteins [68,69].

DEAD-box RHs are present in some viruses, many prokaryotes, and all eukaryotes, and they usually possess ATP-dependent RNA helicase activity, which allows them the localized unwinding of RNAs, and hence, to participate in different RNA metabolism processes, such as ribosome biogenesis, translation initiation, RNA splicing, turnover, and decay [70]. DEAD-box RHs are also involved in OGE in both chloroplasts and mitochondria. In fact, one of the first RH family members to be identified was Mss116, which is required for mitochondrial splicing in yeast [71].

Like mTERFs and PPRs, this family also extends in plants. Accordingly, Arabidopsis and rice show a similar number of presumed DEAD-box RHs, 58 and 60 [24], respectively, while *Drosophila melanogaster* and *Caenorhabditis elegans* show nearly half these numbers with 30 and 34 [72], respectively. By an in silico analysis, Nawaz and Kang [24] found that the numbers of chloroplastic DEAD-box RHs ranged from 7 to 12, and the mitochondrial ones from 4 to 7, in four different plant species.

Nowadays, growing evidence reveals the implication of DEAD-box RHs in plant responses to different kinds of stresses [24]. Regarding chloroplasts and mitochondria, it is known that some nuclear genes that encode organellar-localized DEAD-box RHs are induced by different stresses, including salinity. *HVD1* (*Hordeum vulgare* DEAD box protein1) was the first salt-responsive DEAD-box RH gene to be described encoding a chloroplast protein [73]. The expression of the *HVD1* gene is highly induced under salt and cold stresses, with subsequent recovery after exposure to salinity stress. In a genome-wide in silico study performed later, Umate et al. [74] found that several Arabidopsis DEAD-box RHs located in chloroplasts or mitochondria regulate transcription in response to various abiotic stresses (reviewed in [24]). In addition, the overexpression of some organellar-localized nuclear-encoded DEAD-box RHs has been described to confer resistance to salinity. Accordingly, the overexpression of OsSUV3 (*Oryza sativa* SUPPRESSOR OF VAR3), a mitochondrially-localized DExH/D-box-related DNA/RNA helicase, confers salt tolerance in rice by maintaining photosynthesis and antioxidant machinery [69]. Under salt stress, the levels of the plant hormones GA3 (gibberellic acid 3), zeatin, and IAA (indole-3-acetic acid) increase in the SUV3 overexpressing lines, which, in turn, may induce the expression of genes to help cope with stress [75]. Recently, the chloroplast-targeted DEAD-box RH BrRH22 of *Brassica rapa* has been described to be induced by several abiotic stresses, including salinity and ABA treatment. Likewise, the chloroplast-localized DEAD-box RH OsRH58 of rice is up-regulated by salinity, dehydration, and heat, but not ABA, unlike BrRH22 [76]. Transgenic Arabidopsis OsRH58-expressing plants show increased growth and yield seeds under no stress conditions. The heterologous expression of BrRH22 or OsRH58 in Arabidopsis improves germination and plant growth under salinity conditions. Both proteins may possess RNA chaperone activity, which may, in turn, influence plastid translation under stress conditions, as the Arabidopsis transgenic plants expressing BrRH22 or OsRH58 show larger amounts of some chloroplast genome-encoded proteins when treated with either NaCl or mannitol [76,77].

To our knowledge, Arabidopsis *AtRH3* is the only gene to encode an organellar-localized DEAD-box RH, whose involvement in the response to salinity has been reported by the isolation and characterization of loss-of-function mutant alleles. The impaired function of chloroplast targeted AtRH3 causes a range of effects, from embryonic lethality to delayed growth, as well as reduced greening of vegetative tissues, depending on the mutant allele's strength [54]. Lee and colleagues [55] showed that young viable *atrh3-4* mutant seedlings were impaired in chloroplast biogenesis as these are smaller, and include fewer and smaller chlororibosomes than the wild type. The defective chloroplast development noted in *atrh3-4* seedlings gives rise to ABA-deficient and salt- and cold-sensitive phenotypes (Table 1) [55,56]. The function of AtRH3 is needed for the splicing of most intron-containing chloroplast genes [54,56] and the spliced/unspliced transcript ratio is lower in *atrh3* mutants (Table 1) [56]. Stress conditions do not affect this RNA processing defect, except for genes *ndhA* and *ndhB*, as the spliced/unspliced transcript ratio in *atrh3* mutants is significantly lower, especially under salt or cold stress, but not after dehydration stress [56]. The correct splicing of chloroplast transcripts is probably mediated by the RNA chaperone activity of AtRH3 [56]. More recently, loss of function mutations of *AtRH50*, another Arabidopsis gene that codes for a chloroplast-localized DEAD-box RH, has been shown to increase cold sensitivity as a result of the defective processing of chloroplastic rRNA, and hence inefficient plastid translation [78]. The response of *ath50* mutants to other abiotic stress, such as salinity, remains to be tested.

## 2.1.4. Mutations in Plastid Ribosomal Protein PSRP2 Alter Salinity Tolerance

The previous sections noticeably illustrate the relationship between changes in the activity of nuclear genes involved in the post-transcriptional regulation of OGE at the RNA level and salt stress phenotypes. The involvement of plastidial translational machinery in plant salt responses has also been investigated. Accordingly, salt stress causes a transient suppression of the de novo synthesis of proteins in Arabidopsis suspension cell cultures [79]. Furthermore, Omidbakhshfard and colleagues [80] tested the expression by qRT-PCR of 170 genes related to protein synthesis in Arabidopsis leaves after NaCl exposure at different time points. Some genes showed up-regulation under salinity stress, and coded for chloroplast-located translation-related proteins, such as ribosomal proteins L11 (PRPL11) and L9-1 (also known as PIGGYBACK 2), ATAB2, an A/U-rich RNA-binding protein, which likely functions as an activator of translation, and PDF1B, a peptide deformylase required to remove the N-formyl group from nascent peptides. Interestingly, these genes have been previously described as being important for chloroplast development, and have been proposed to represent potential biotechnological targets for plant salt tolerance optimization [80].

As far as we know, only a salt-stress related phenotype has been described for a mutant affected in a plastid translational protein. Accordingly, the Arabidopsis *psrp2* mutant defective in PLASTID-SPECIFIC RIBOSOMAL PROTEIN 2 (PSRP2) of the 30S ribosomal subunit, one of the six Arabidopsis PSRP proteins, shows enhanced seedling growth under salinity stress (Table 1) [57]. On the contrary, the transgenic plants overexpressing PSRP2 display delayed germination and reduced seedling growth in response to salinity compared to the wild type (Table 1). These results suggest that PSRP2 functions as a negative regulator of germination and seedling growth under salinity conditions. Interestingly, PSRP2 contains two RNA recognition motifs (RRM), it can bind RNA and ssDNA, and it possesses RNA chaperone activity. All this suggests a connection between plastid translation and regulation of RNA metabolism [57].

#### *2.2. Transcriptional Regulation of OGE and Plant Salt Tolerance*

## 2.2.1. Arabidopsis SIG5 Protects Chloroplasts from Abiotic Stress Damage

Transcriptional regulation is also fundamental for OGE in plants, and mutations in the genes involved in this process can lead to a salt stress phenotype. Along these lines, a small family of nuclear genes, called *RpoT* (*RNA polymerase T7 phage-type*), encoding monomeric RNA polymerases, is responsible for the transcription of chloroplast or mitochondria genes [81]. Furthermore, the transcription of plastid genes requires a second type of DNA-dependent RNA polymerase, similar to those of prokaryotes, which is

encoded by the organellar genome and dubbed PEP (plastid-encoded RNA polymerase). PEP comprises four catalytic core subunits (α, β, β', and β", respectively, encoded by plastid genes *rpoA, rpoB, rpoC1*, and *rpoC2*) and requires nucleus-encoded sigma factors (SIG) to recognize specific promoter sequences to initiate plastid gene transcription [82].

One of the six *SIG* Arabidopsis genes, *SIG5*, is up-regulated in response to different stress conditions, including salinity [58]. SIG5 is required for the transcriptional activation of the blue light-responsive promoter of the *psbD* gene that encodes photosystem II (PSII) reaction center protein D2, which binds the essential redox co-factors needed for photosynthetic electron transfer [58]. Besides, seed germination and recovery from damage of PSII after salt stress are delayed in loss-of-function mutant *sig5-2* compared to the wild type (Table 1). This led Nagashima et al. [58] to propose that SIG5 enhances the repair of the PSII reaction center under stress by protecting chloroplasts, and hence, the plants exposed to adverse environmental stress conditions. Interestingly, the expression of the orthologous gene from the liverwort *Marchantia polymorpha* (*MpSIG5*) is induced by blue-light irradiation under several stress conditions, which suggests the conservation of the responsible mechanism [83]. Nevertheless, the expression of the *psbD* gene of Marchantia did not occur in conjunction with *MpSIG5* induction, which indicates that SIG5 might play a divergent physiological role in different plant phyla [83].

Recently, Zhao et al. [84] obtained Arabidopsis *SIG5* overexpression lines, which are more tolerant to salt stress than the wild type. These results are in agreement with those from Nagashima et al. [58] mentioned above. Furthermore, Zhao and colleagues [84] also identified an upstream regulator of *SIG5*, ATHB17 (ARABIDOPSIS THALIANA HOMEOBOX 17), an Arabidopsis HD-Zip transcription factor that binds to the *cis*-elements present in the *SIG5* promoter. Similarly to *SIG5, ATHB17* loss-of-function reduced plant salt tolerance, whereas ATHB17 overexpression enhanced it. Interestingly, these phenotypes were at least partially dependent on SIG5. Accordingly, *ATHB17* overexpression in a *sig5-1* mutant background showed tolerance to a salinity intermediate between that of the *ATHB17* overexpression lines and the *sig5-1* mutant [84]. Under salt stress conditions, various plastid genes regulated by SIG5 were down-regulated in the *ATHB17* overexpression lines, but were up-regulated in the *ATHB17* knockout lines, which suggests the positive regulation of these genes through SIG5 [84].

## **3. Conclusions and Future Perspectives**

The results compiled in this review reveal that accurate OGE regulation in chloroplasts and mitochondria is fundamental for plants to tolerate and adapt to adverse environmental conditions, such as salt stress (Figure 1). Consequently, perturbed OGE homeostasis affects plant responses to salinity, which can be detrimental for their survival. However, we are still far from fully understanding the role that OGE regulation plays in adapting plants to salinity. To make progress, it is necessary to know more details of the molecular functions of several genes involved in the control of OGE, whose disturbance results in stress phenotypes. This is especially relevant for the *mTERF* genes related to stress tolerance, because the mechanistic insight for most of them is lacking. Furthermore, unraveling why particular stressors specifically affect the expression of some OGE-related genes would shed light on their functional roles in plant tolerance to abiotic stresses. If we consider that a stress mutant phenotype has been reported for only a few *PPR*, *mTERF*, or *DEAD-box RH* genes, it is fundamental to identify and characterize mutants in plant model systems and crops affected in novel genes that transcriptionally or post-transcriptionally regulate OGE in plastids and mitochondria. This information is expected to contribute to the comprehensive understanding of the mechanisms regulating OGE in plants, and to provide insights to improve stress tolerance by identifying potential biotechnological targets for better plant growth and crop yields under salinity conditions.

**Figure 1.** Schematic representation of the cellular functions of OGE regulators in plant response to salinity. Salt stress can perturb chloroplast or mitochondrion homeostasis and this would be communicated to the nucleus through retrograde signals, leading to changes in nuclear gene expression. In turn, this would activate the expression of OGE regulators (e.g., mTERFs, PPRs or DEAD-box RHs proteins), which would result in anterograde signalling responses to adjust organellar function to salinity. In the figure, inside a mitochondrion (purple) and chloroplast (green), only those molecularly-characterized OGE regulators whose mutations lead to altered responses to salinity are depicted. The OGE processes affected in these mutants, as well as their enhanced (↑) or reduced (↓) sensitivity to salt stress, are shown (see also Table 1 for further information).

**Funding:** The research conducted in the laboratory of V.Q. has been supported by a grant from the Conselleria d' Educació of the Generalitat Valenciana (Spain) (AICO/2015).

**Conflicts of Interest:** The authors declare no conflict of interest.
