As aforementioned, plant viruses can move their RNA and DNA genomes from cell to cell and over a long distance. This is determined (at least in part) by virus-encoded movement proteins. Thus, defects in viral movement proteins can rid viruses of intercellular and systemic mobility whilst such movement-deficient viruses can still replicate in single infected cells. Based on these, two plant RNA viruses, i.e., Potato virus X (PVX, Chapman et al., 1992) and Turnip crinkle virus (TCV; Ryabov et al., 2004), were modified as RNA mobility Assay (RMA) vectors in which the coat protein gene was deleted from each virus genome. The resultant virus-based RMA vectors PVX/ΔCP and TCV/ΔCP, can infect, but are restricted within individual leaf epidermal cells (Li et al., 2009). By engineering the FT RNA into the two RMA vectors, it restores cell-to-cell and long-distance movement of PVX/ΔCP and TCV/ΔCP RNA. Such virus-based RMAs provide compelling answers to three questions on FT mRNA movement. Firstly, FT mRNA can move long distances independent of FT protein. This conclusion was also confirmed by virus-free RMA in which FT RNA produced via agro-infiltration of local leaf tissues can spread to distal non-infiltrated newly growing leaves (Li et al., 2009). Secondly, the core mobile determinant consists of 102-nucleotides at the 5′ end of the FT mRNA (Li et al., 2009). Thirdly, FT mRNA can facilitate PVX entry of SAM where viruses are usually excluded (Wu et al., 2020), and leads to virus-induced gene silencing in SAM. Moreover, the long-distance FT RNA movement is also shown to enhance early flowering (Li et al., 2011). It is worthwhile noting that the virus-based RMA is also used to show that long-distance movement of the BEL5 mRNA contributes to tuberization in potato (Cho et al., 2016).

Grafting is one of the gold standard methods to study long-distance RNA movement (Gaut et al., 2019). Using heterografting technique coupled with RNAseq, 2,006 genes producing mobile mRNAs were identified in two different A. thaliana ecotypes (Thieme et al., 2015), and 138 Arabidopsis mobile mRNAs were found in A. thaliana (as stock) - Nicotiana benthamiana (as scion) system (Notaguchi et al., 2015). However, it is conceivable that this method is more suitable to identify mobile RNAs of high abundance. Therefore, it is unsurprising that FT mRNA with limited and dynamic expression pattern cannot be easily detected in previous studies (Corbesier et al., 2007; Tamaki et al., 2007). To avoid this issue, some mature leaves of the ft-3 scion grafted onto the wild-type stock were removed in order to enrich potential wild-type FT mRNAs originated from wild-type stock. This indeed allows a positive detection of the wild-type FT mRNA movement from stock to scion through grafting junction (Lu et al., 2012). Furthermore, such grafting experiment also led to identify the 210-nt sequence at the 5′ end of FT mRNA as the mobility determinant (Lu et al., 2012), consistent with the virus-based RMA’s findings (Li et al., 2009).

The bacteriophage coat protein MS2 can bind its target RNAs “stem-loop repeats (SL).” The MS2-GFP fusion protein is often used to label and image RNAs in different organisms (Querido and Chartrand, 2008). However, due to its high background fluorescent noise in the cytoplasm, this method is not extensively used in plants. Nevertheless, MS2 was found to be very specific nucleus localized when it is fused with transcription factor FD (Luo et al., 2018). The nuclear retention of MS2_FD-GFP was much longer than MS2_SV40-GFP. Luo et al. (2018) then co-expressed the MS2_FD-GFP, and chimeric SL-FT mRNA. Through fluorescent imaging, the SL-FT mRNA was found to move intracellularly and mainly accumulated at the plasmodesmata sites; however, the dynamic process of the SL-FT mRNA moving from one cell to another was not observed (Luo et al., 2018).

Clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (CRISRP/Cas9) gene editing system has revolutionized targeted gene editing (Jinek et al., 2012; Cong et al., 2013; Feng et al., 2013). This gene editing tool contains three main components, Cas9 enzyme, spacer sequence and sgRNA (Jinek et al., 2012). Spacer and sgRNA are usually taken as a whole. As for plants, agrobacterium-mediated CRISPR/Cas9 technology is widely used. However, this technology is often involved in genetically modified plants and time-consuming screen of homologous lines with the edited target gene. These issues can be partially avoided by a technology so called virus-induced gene/genome editing (ViGE; Ali et al., 2015).

However, gene editing resulted from ViGE is often not heritable to next-generation due to virus exclusion from SAM, thus its use is limited (Zhang et al., 2020). How to generate heritable ViGE is the issue that needs to be resolved. Recently, Ellison et al. (2020) have elegantly exploited the fact that the mobile translatable and non-translatable FT RNA can facilitate RNA virus entering SAM (Li et al., 2011), and used the Tobacco rattle virus (TRV) to deliver sgRNA-tagged with the mobile FT RNA into SAM. In the way, the mutant progeny of Nicotiana benthamiana were recovered in next-generation at frequencies ranging from 65 to 100%; and due to sgRNA-targeting of PDS gene, the transgenerational seedlings with albino phenotype resulted from the PDS gene editing via TRV/(m)FT-sgRNA were statistically different compared with the mock control (Ellison et al., 2020). This demonstrated that the mobile FT RNA can promote the PDS sgRNA into SAM and enhances the progeny gene editing efficiency of the CRISPR/Cas9 system (Ellison et al., 2020). Similar results have been also reported using PVX to deliver FT RNA-tagged sgRNA to plants (Uranga et al., 2021).

Moreover, a cotton leaf crumple virus (CLCrV)-mediated ViGE system was also developed. In this case, sgRNAs were fused to the 102-nt FT mRNA, then expressed by CLCrV in transgenic Cas9 A. thaliana. The enhanced gene editing efficiency of 4.35–8.79% was found in progeny plants free of the CLCrV genome (Lei et al., 2021). Together, all these latest evidence shows that FT RNA is mobile and can enter SAM, as well as that such mobile RNA element has a high potential of biotechnological application in inheritable and transgenerational genome editing in plants and crops.

Flowering Locus T is one of six phosphatidyl ethanolamine-binding protein (PEBP) family members and the other five are TERMINAL FLOWER1 (TFL1) LIKE, MOTHER OF FT AND TFL1 (MFT), BROTHER OF FT AND TFL1 (BFT), ARABIDOPSIS THALIANA CENTRORADIALIS (ATC), and TWIN SISTER OF FT (TSF) in Arabidopsis (Karlgren et al., 2011). Phylogenetic analysis indicates that this small multigene family consists of three classes, FT-LIKE (FT and TSF), MFT, and TFL1-LIKE (ATC, BFT, and TFL1) (Chardon and Damerval, 2005). FT and TFL1 are two highly conserved homologous proteins, which have opposite functions but compete to regulate the initiation of plant flowering (Hanzawa et al., 2005). So far, there is no evidence that TFL1 RNA can move long distance. On the other hand, CET1 mRNA, an ortholog of the Arabidopsis antiflorigen ATC, is mobile, as revealed in tobacco/Arabidopsis grafting experiments. Its non-cell-autonomously movement is also confirmed in heterograft of tobacco and tomato (Huang et al., 2018). Other FT ortholog gene mRNAs were also found moving across the tomato-tobacco heterograft junction (Huang et al., 2018).

Synonymous codon substitution in FT (synFT) does not change the FT protein amino acid sequence but alters the RNA sequence. Notaguchi et al. (2008) transformed ft-1 with an expression cassette in which synFT contains 171 (of 175) codon substitutions. In grafting experiments where ft-1 was used as a recipient stock and transgenic synFT/ft-1 as a donor scion, the synFT mRNA was not detectable in the ft-1 shoot apical region. These experiments also suggest that changes in the FT RNA sequence did not affect the FT protein movement (Notaguchi et al., 2008). However, unlike the later grafting study (Lu et al., 2012), no enrichment of potential mobile synFT mRNA was done. This may explain why the synFT RNA could not be detected (Notaguchi et al., 2008). On the other hand, changes of primary mRNA sequence may lead to alternation of secondary structures (see below), which may be required for systemic FT RNA movement.

A high number of tRNAs were detected in the phloem sap of pumpkin (Cucurbita maxima). Among these mobile tRNAs, their distributions are uneven. For example, no Ile-tRNA or very few Arg-tRNA but a few Asp-tRNA molecules were detected; however, Cys, Leu, Phe, Try, Trp, and Ser tRNAs were predominantly present in the phloem sap (Zhang et al., 2009). These tRNAs have been predicted to serve as long-distance signals (Zhang et al., 2009, 2016; Lezzhov et al., 2019; Wang et al., 2021), suggesting that tRNA or tRNA-like structure could be an important factor, if not the determinant for systemic RNA movement. Indeed, in the ViGE systems, like the FT mRNA/102 nt-FT RNA element, Met, Gly, and Ile tRNAs were also found to improve transgenerational gene editing when they were fused to sgRNA (Ellison et al., 2020). It is possible that the FT mRNA may form unique structure, or tRNA/tRNA-like secondary or even tertiary structures that are important for its mobility.

In plants, it is well-established that the movement of RNA requires RNA binding proteins. For example, KNOTTED1 (KN1) (Lucas et al., 1995) and a viral movement paralog protein in Cucurbita maxima, CmPP16 have been reported to be involved in cellular RNA trafficking (Xoconostle-Cázares et al., 1999). In Arabidopsis, a conserved SMALL RNA-BINDING PROTEIN 1 (SRBP1) family member, AtSRBP1, was found to mediate small RNA movement. AtSRBP1, a glycine-rich (GR) RNA-binding protein, also named AtGRP7, can bind to single-stranded siRNA (Yan et al., 2020). It is plausible that FT mRNA movement may also involve certain RNA binding protein(s) (Jackson and Hong, 2012).

Numerous RNA modifications have been reported (Boccaletto et al., 2018) and some specific modifications are closely associated with biological functions (Chmielowska-Bąk et al., 2019). In plants, mRNA N6-methyladenosine (m⁶A) and 5-methylcytosine (m⁵C) play crucial and dynamic roles in embryo development, leaf morphogenesis, and root development (Liang et al., 2020). Recently, epigenetic modifications such as m⁵C have been found to be linked with RNA mobility and m⁵C RNAs are predominant in the phloem (Yang et al., 2019; Wang et al., 2021). Interestingly, in RNA m⁵C methylation-deficient mutants, mobile TRANSLATIONALLY CONTROLLED TUMOR PROTEIN 1 and HEAT SHOCK COGNATE PROTEIN 70.1 mRNAs become immobile (Yang et al., 2019). We speculate that m⁵C or other types of epigenetic modifications may play a role in FT mRNA mobility.