Movement of small RNAs in and between plants and fungi

Abstract RNA interference is a biological process whereby small RNAs inhibit gene expression through neutralizing targeted mRNA molecules. This process is conserved in eukaryotes. Here, recent work regarding the mechanisms of how small RNAs move within and between organisms is examined. Small RNAs can move locally and systemically in plants through plasmodesmata and phloem, respectively. In fungi, transportation of small RNAs may also be achieved by septal pores and vesicles. Recent evidence also supports bidirectional cross‐kingdom communication of small RNAs between host plants and adapted fungal pathogens to affect the outcome of infection. We discuss several mechanisms for small RNA trafficking and describe evidence for transport through naked form, combined with RNA‐binding proteins or enclosed by vesicles.


| INTRODUC TI ON
Small RNAs were first discovered in Escherichia coli in 1984 (Mizuno et al., 1984). Subsequently, they have been found in all kingdoms of life operating as noncoding RNA with diverse functions (Wassarman et al., 2001;Saito, Kakeshita, and Nakamura, 2009;Pantaleo et al., 2010;Li et al., 2016). Most small RNAs serve as regulators of gene expression (Hammond et al., 2000;McCaffrey et al., 2002;Paul et al., 2002). In eukaryotes, small RNAs induce silencing of target genes, known as RNA interference (RNAi), at both transcriptional and post-transcriptional levels. Here, their defining features are short length (c.20-30 nucleotides) and association with proteins of the Argonaute family, with whose help they can recognize target mRNAs and lead to their reduced expression (Ghildiyal and Zamore, 2009).
siRNA, miRNA as well as piRNA all act to control gene expression and play important roles in many fundamental biological processes in eukaryotic organisms. They have been tied to vital processes such as cell growth, tissue differentiation, heterochromatin formation, cell proliferation, and disease resistance (Blair and Olson, 2015;Yuan et al., 2015;Tassetto et al., 2017;Czech et al., 2018;Mondal et al., 2018;Almeida et al., 2019). Research over the past few decades has led to powerful insight into the structure and function of small RNAs, which has been summarized in several reviews (Eamens et al., 2008;Ghildiyal and Zamore, 2009;Peters and Meister, 2007;Pratt and MacRae, 2009;Holoch and Moazed, 2015;Quinn and Chang, 2016;Zhang, Cozen et al., 2016;Zhang et al., 2019). The purpose of this review, however, is to highlight what is known and not known about the mechanisms of how small RNAs move within and between organisms. Indeed, small RNAs can travel both short and long distances in plants, as well as in fungi. Below, we summarize their movement in plants and fungi before considering how small RNAs move between fungi and plants.

| S HORT-AND LONG -DIS TAN CE MOVEMENT OF S MALL RNA S IN PL ANTS AND FUNG I
In plants, small RNAs are produced to coordinate plant development, maintain genome integrity, and combat adverse environmental conditions (Buchon and Vaury, 2006;Chen, 2009;Ruiz-Ferrer and Voinnet, 2009). The mobility of small RNAs was presumed to be a prerequisite for carrying out these functions.
Evidence now shows that small RNAs can move both short and long distances in plants (Sarkies and Miska, 2014). Primary siRNA can spread 10-15 cells without producing secondary siRNA (Kim, 2005), while long-distance small RNA movement involves amplification of silencing signals through RNA-dependent RNA polymerases (RDRPs) that are transported primarily through the phloem (Wassenegger and Krczal, 2006). Transitivity and secondary siRNA production amplify the RNAi so silencing persists even in the absence of the initiator double-stranded RNA (dsRNA) (Baulcombe, 2004). As early as 1928, Wingard found the upper leaves of a tobacco plant whose lower leaves had been inoculated with tobacco ringspot virus and showed strong symptoms became resistant to the same virus (Wingard, 1928). We now know that the recovery from virus disease involves small RNAs derived from the virus moving from the infection site to upper leaves and conferring small RNA-mediated resistance in the distal tissues (Ratcliff, 1997;Baulcombe, 2004).

| Cell-to-cell (short-range) movement in plants
The early clear evidence for mobile small RNAs was reported using Nicotiana benthamiana plants expressing the GFP transgene. Leaf infiltration with Agrobacterium also expressing GFP resulted in a ring of GFP silencing that was consistently observed spreading over 10-15 cells beyond the agroinfiltration zone without triggering small RNA amplification. When an RNA silencing suppressor was co-infiltrated, GFP silencing was abolished (Johansen, 2002). In addition to siRNA generated by transgenes, endogenous miRNAs have also been observed to spread from cell to cell. For example, when miR390 precursor loci were transcribed in the vascular system and pith region of Arabidopsis, mature miR390 were found only in the shoot apical meristem and young leaf primordia where their precursors were not detected (Chitwood et al., 2009). Similarly, miR165/166 precursors were transcribed mainly in the endodermis of Arabidopsis root, but mature miR166 were observed in adjacent cell layers (Carlsbecker et al., 2010). These and other examples are consistent with the cell-to-cell movement of miRNA (Chitwood et al., 2009;Martínez et al., 2016;Wu and Zheng, 2019) (see Figure 1a).

| Long-range movement in plants
Long-range systemic movement was first demonstrated by Dalmay in 2000 using a phloem-restricted virus expressing a GFP reporter gene. The virus was applied to GFP expressing plants and GFP silencing was observed for entire leaves (Dalmay et al., 2000). Later, Pant and colleagues demonstrated the long-range movement of miRNA through micrografting Arabidopsis plants. In grafted plants with miR393 overexpressing shoots and wild-type roots, high levels of miR393 accumulated in the roots, suggesting the long-range movement (shoot to root) of miR393 (Pant et al., 2008). Molnar demonstrated that both exogenous and endogenous small RNAs could pass through the graft union (Molnar et al., 2010). Other studies using grafted Nicotiana tabacum as well as Arabidopsis showed small RNAs can transfer from source tissue (leaves) to meiotically active cells such as anthers and flowers ; see Figure 1b).

| Movement of small RNAs in fungi
In contrast to plants, fungi are simple organisms and lack defined cellular transportation systems for the movement of nutrients and metabolites. Fungi may exist as unicellular forms or as extensive multicellular hyphal branched networks. A number of fungi, including Zygomycota, are usually aseptate; in contrast, other fungal divisions like Ascomycota and Basidiomycota hyphae are separated by septa, which usually have pores. Small RNAs have also been well characterized in fungi (Drinnenberg et al., 2009;Nicolas et al., 2010;Nunes et al., 2011;Mueth et al., 2015;Campo et al., 2016;Donaire and Ayllón, 2017). In 1992, small RNAs were first demonstrated to mediate gene silencing, termed quelling, in Neurospora crassa (Romano and Macino, 1992). Subsequently, similar phenomena were reported in many fungal phyla, including Ascomycetes and Basidiomycetes, as well as in fungal-like Oomycota (Nicolás, Torres-Martínez, and Ruiz-Vázquez, 2003;Latijnhouwers et al., 2004;Wang et al., 2010;Nunes et al., 2011). Studies of the direct movement of small RNAs within fungal colonies and tissues are largely absent. However, transfection of protoplasts with dsRNA can lead to targeted gene silencing that is maintained for several months across a growing colony, suggestive of both amplification and movement (Caribé dos Santos et al., 2009;Saraiva et al., 2014).

| TR AN S P ORTATI ON PATHWAYS OF S MALL RNA S IN PL ANTS AND FUNG I
Conceptually, molecules, including small RNAs, can be transported between cells and tissues within an organism via two principal F I G U R E 1 Short-and long-distance transportation of small RNAs in plants and fungi. (a) Cell-to-cell movement in plants: 1, naked small RNAs, small RNAs bound to RNA-binding proteins (RBP), and small RNAs enclosed in vesicles can move from cell to cell through spaces between the plant plasma membrane (PM) and desmotubule (DM); 2, small RNAs can be transported through the DM, which connects the endoplasmic reticulum (ER) of two adjacent cells; 3, small RNAs can be secreted from the PM and travel through the plant cell wall (CW) to extracellular spaces, and small RNAs can also be taken up by other cells (multiple vesicle bodies, MVB). Note: Vesicle transport through plasmodesmata by active gating is hypothetical at this time. (b) Long-distance movement in plants: 1, naked small RNAs, small RNAs bound to RBP, and small RNAs inside vesicles can be transported from source cells (SC) to companion cells (CC) and then to sieve tube elements (SE) through plasmodesmata; 2, small RNAs can be secreted out of PM and travel through the plant cell wall (CW) to extracellular spaces and subsequently be absorbed by other cells; 3, small RNAs can be transported to distal plant cells through the sieve tube elements (sieve tube plates, SP). (c) Movement in fungi: 1, naked small RNAs, small RNAs bound to RBP, and small RNAs inside vesicles can be transported short distances cell to cell through the septal pore (SP); 2, small RNAs can be secreted out of fungal plasma membrane (FPM) and travel through the fungal cell wall (FCW) to extracellular spaces. Later, small RNAs can be absorbed by distal fungal cells and in this way small RNAs can be dispersed systemically throughout the whole fungal colony; 3, small RNAs can be transferred through the FPM. Unlike nonselective transportation through septal pores, FPM can conduct selective transportation by binding, fusion, and secretion. Note: small RNA movement in fungi needs more evidence.

| Transport as either a naked form or encased in vesicles
Evidence for transport of naked forms is primarily inferred from Production of secondary siRNA may occur to amplify the silencing effect and small RNAs may move through the whole fungal colony.
In Saprolegnia parasitica, dsRNA-mediated long-term gene silencing has also been reported (Saraiva et al., 2014). Moreover, when artificial synthesized siRNA were co-cultured with the model filamentous fungus Aspergillus nidulans, silencing of the reporter GFP gene as well as endogenous AnrasA & B genes was induced, supporting the possibility that this may be a natural means of small RNA transport in fungi (see Figure 1c) (Kalleda et al., 2013).
Direct application of RNA molecules to plants has been shown to down-regulate endogenous transcript levels. Sammons et al. (2011), in a patent application, showed that direct application of various nucleic acids, including dsRNA and siRNA, down-regulated herbicide resistance (Sammons et al., 2011). Through root soaking, dsRNA targeting Mob1A and WRKY23 was delivered into Arabidopsis and rice tissue. Suppression of root growth, seed germination, and failure of bolt or flower were detected along with silencing of the targeted genes . Besides suppression of plant endogenous genes, a number of studies have demonstrated that direct application of dsRNAs can effectively silence transgenes such as GFP or YFP in plants (Dubrovina et al., 2019).
As an alternative to the naked form, small RNAs can also be transported through a pathway involving vesicular migration from the endoplasmic reticulum (ER) to the Golgi apparatus and then loading to a complex network of vesicles. Small RNAs can be sorted to transporting vesicles fusing with the plasma membrane and then released by exocytosis (Bonifacino and Glick, 2004 (Albuquerque et al., 2008;Rodrigues et al., 2008Rodrigues et al., , 2007. In addition to proteins, neutral lipids, glycans, and pigments, fungal RNA has also been found in EVs (Rodrigues et al., 2007;Oliveira et al., 2010Oliveira et al., , 2009Vallejo et al., 2012;Garcia-Silva et al., 2014). Different types of noncoding small RNAs have been characterized inside EVs from C.
neoformans, Paracoccidiodes brasiliensis, and C. albicans as well as from S. cerevisiae .
As each cell has two endomembrane systems, one for outgoing traffic and the other for incoming traffic (Hilbi and Haas, 2012), small RNAs can be released from the cell through EVs as well as be absorbed by the recipient cell through membrane fusion. This has been demonstrated using synthetic EVs composed of siRNA inside cationic lipid/liposomes (Spagnou et al., 2004). Moreover, the trafficking of EVs by fungal cells is regulated by both cell turgor and cell wall structure (Eisenman et al., 2005;Brown et al., 2015). Thus, the fungal cell wall may play an important role in regulating the movement of small RNAs (via EVs) between fungal cells and to plant hosts ( Figure 2).

| Movement of small RNAs via the symplast and apoplast
In plants, for movement through the symplast, small RNAs probably move through the plasmodesmata (PD), a plasma membrane-lined pore acting as an intercellular channel that connects the plant cytoplasm of connected cells (Figure 1a). There are several lines of evidence supporting the symplast route. Mature guard cells that are symplastically isolated from adjacent cells escape transitive GFP silencing (Voinnet et al., 1998;Vatén et al., 2011). The presence of the tobacco mosaic virus movement protein (MP) increased PD aperture size and enhanced the spread of transgene silencing (Bucher et al., 2001;Vogler et al., 2008). Several viruses transfer their RNA genome to plant cells through ER protrusions that extend through the PD (Chou et al., 2013;Pyott and Molnar, 2015).
The PD size exclusion limit is 30-50 kDa and may dictate which forms of small RNAs can move through the PD. Naked small RNAs are around 15 kDa, thus their free diffusion through the PD should not be limited (Crawford and Zambryski, 2000). As the size of plant vesicles (>10 nm) (Huang et al., 2017;Rutter and Innes, 2017) is generally larger than the diameter of PD microchannels (3-4 nm) (Ding et al., 1999;Sager and Lee, 2018), vesicles containing small RNAs may not diffuse freely through the PD. However, PD permeability can be significantly increased through dilation, active gating, and structural remodelling (Lucas and Lee, 2004). Thus, naked as well were found (Turnbull and Lopez-Cobollo, 2013). In 2010, Varkonyi and co-workers also found a subset of miRNAs present in the phloem of apple Varkonyi-Gasic et al., 2010). In addition, Roberts showed that treatment of plants with a nontoxic concentration of cadmium to block phloem transport of specific virus movement also inhibited systemic RNA silencing (Ghoshroy et al., 1998;Ueki and Citovsky, 2001). In vascular plants, phloem is a living tissue that conveys organic compounds made during photosynthesis from source (typically leaves) to sink tissues (such as roots and buds) (Van Bel, 2003). However, in several solanaceous species as well as Arabidopsis, upward long-distance mobile silencing has also been shown to be phloem mediated (Liang et al., 2012).
Proteins may assist in both short-and long-range transportation of naked small RNAs. In plants, RNA-binding proteins (RBPs), which are at the core of ribonucleoprotein complexes (RNPCs), are important for RNA movement (Kedde et al., 2007). Phloem Small-RNA Binding Protein 1 (CmPSRP1) from pumpkin (Cucurbita maxima) phloem binds single-stranded small RNAs moving from cell to cell through the PD (Yoo et al., 2004). This protein can also shuttle small RNAs through the companion cell-sieve element complex (Ham et al., 2014). For fungal cells that are linked to each other, intercellular communication may be achieved via septal pores, similar to plasmodesmata in plant cells (Bloemendal and Kück, 2013). Septal pores were first reported by Bary in 1884(Bary, 1884. Later, in 1893, Wahrlich observed cytoplasmic flow between different fungal compartments (Wahrlich, 1893). Septa can be described as a simple plate with a central pore about 50-500 nm in diameter that allows the passage of cytoplasm and organelles like mitochondria, vacuoles, and nuclei (Gull, 1978;Esser, 1982). Moreover, microtubules have also been found to direct the transport process in filamentous fungi and the range of cargo can be expanded to include endosomes, mRNA, peroxisomes, and secretory vesicles (Egan et al., 2012). It was further demonstrated that tubules can move cargo in either direction across the septal pores as well as transport material between cells (Shepherd et al., 1993). In sum, the septal pore, a plasmodesmata-like structure associated with ER or the desmotubule, a membranous cell wall-spanning structure, may enable small RNAs either in naked form or enclosed in vesicles to move throughout the whole mycelial network (Zarnack and Feldbrügge, 2007; Figure 1c).
Transport via symplastic routes is probably valuable for movement within an organism and where direct cellular connections exist. For

| Endogenous small RNA transfer between plants and fungi
The role of cross-kingdom RNAi for defining interactions between fungal pathogens and plant hosts was pioneered by Hailing Jin's group.
They showed that to promote virulence, the necrotrophic fungal path-  . A growing number of recent studies suggest that both plants and fungi use cross-kingdom RNAi strategies for their own benefit (Table 1).

| HIGS: artificial small RNAs transfer from plants to fungi
Observations that naturally occurring endogenous small RNAs move between organisms led to studies that showed that artificial transgene-derived small RNAs are also able to move between interacting organisms. This has been exploited for the development of host-induced gene silencing (HIGS), a novel RNA-based technology for the efficient control of fungal pathogens and other pests (see Table 1). Conceptually, HIGS involves generating small RNAs targeting a pathogen gene in the host plant, which results in the uptake of small RNAs and gene silencing in the invading pathogen.
HIGS has been demonstrated in a number of diverse fungal pathosystems and provides a promising disease control alternative to chemical control (Nowara et al., 2010;Yin et al., 2011;Zhang et al., 2012;Panwar et al., 2013;Hu et al., 2015;Deising et al., 2016;Song and Thomma, 2018;Zhou et al., 2016;Zhu et al., 2017;Qi et al., 2018). In addition, it also can be used as a tool to screen potentially crucial fungal genes without the need to produce knockout mutants, which is challenging in a number of pathogens (Yin et al., 2015).
Small RNAs have been shown to transfer bidirectionally between plants and fungi; however, the mechanism(s) of how they move remains to be fully determined.

| Possible pathways for small RNA crosskingdom movement
Based on studies of small RNA movement in plants and fungi described above, there are several pathways for cross-kingdom small RNA transportation. Because naked small RNAs can move short and long distances in plants (Hyun et al., 2011) and can also be taken up by fungal cells (Wang, Thomas, and Jin, 2017) Gu et al., 2019). Direct application of small RNAs has been referred to as spray-induced gene silencing (SIGS). In most studies, however, direct uptake is limited without tissue wounding .
How these molecules are taken up and first assimilated to the plant before transfer to the pathogen remains to be determined.  (Jiang et al., 2012;Brown et al., 2015). Recently, Arabidopsis cells have been shown to secrete extracellular vesicles to deliver plant small RNAs into the fungal pathogen B. cinerea, resulting in silencing of fungal genes critical for pathogenicity (Cai et al., 2018).
Studies of mammals such as mice also suggested that small

RNAs can be transferred between different species mediated by
EVs (Knip et al., 2014). Such vesicles probably enable genetic communication between phylogenetic distantly related organisms. For example, plant-derived exosome-like nanoparticles have been detected in the guts of mice after consuming plant material. These ingested plant-derived exosome-like nanoparticles contain proteins, lipids, and small RNAs (Mu et al., 2014). Direct evidence for vesicle involvement in plant-pathogen interactions has also been obtained in barley leaves under attack by powdery mildew pathogen Blumeria graminis. Light microscope-visible vesicle-like bodies were observed accumulating around papillae, which formed at sites where the fungal penetration was halted, suggesting such vesicles may be important for host immunity . These vesicles are known to contain antimicrobial compounds, such as phytoalexins, phenolics or reactive oxygen species (Tam et al., 2015).

TA B L E 1 (Continued)
In addition, vesicle-like inclusions have been shown to accumulate around penetration sites in sorghum leaves attacked by the hemibiotrophic fungus Colletotrichum graminicola (Nielsen et al., 2004). In onion, membrane-bound electron-dense vesicles were observed in epidermal cells in response to necrotrophic fungus  (Chowdhury, 2016;Stewart and Mansfield, 1985). Evidence to date also suggests vesicles derived from both plants and fungi can contain small RNAs. However, further research is required to confirm whether vesicles at the fungal-host plant interface do, indeed, contain small RNAs, and that they are released into the extracellular space and are subsequently taken up by the associated partner. The movement of EVs between organisms may be highly regulated and directed rather than occur by simple diffusion (see Figure 2).
Direct evidence for the role of vesicles in cross-kingdom communication could be obtained through the isolation of vesicles derived from HIGS transgenic plants followed by evaluation of the presence of target small RNAs inside vesicles. Such vesicles could then be co-cultured with fungi to confirm the ability to confer RNA silencing. Using fluorescence or radioactively labelled small RNAs would facilitate monitoring of small RNA movement. Chemical inhibitors such as Brefeldin A (Nebenfuhr et al., 2002), prieurianin (Robert et al., 2008;Tõth et al., 2012), and secramine (Pelish et al., 2006) that block vesicle secretion may be valuable to confirm the function of extracellular vesicles.
The possible mechanisms of small RNA absorption remain unknown. Secretion and absorption may involve cell membrane proteins. Currently, there are limited reports regarding protein channels for small RNA movement in plants. In addition, how specific small RNAs are sorted for secretion and absorption remain enigmatic.
Though much work needs to be done, studies of small RNA communication will probably provide applications for enhancing sustainable agriculture. Gene silencing of pathogen genes by HIGS or the direct application of dsRNA, for instance, is a highly promising strategy to provide resistance to plant disease. clear. Answers to these questions and others related to cross-kingdom communication will not only enrich our understanding of plant disease processes but also aid in the development of powerful new tools for disease control.

ACK N OWLED G EM ENTS
We are grateful for financial support from the USDA-NIFA program

DATA AVA I L A B I L I T Y S TAT E M E N T
Data sharing is not applicable to this article as no new data were created or analysed in this study.