Introduction

The plant vacuole and its function in the cell

Unlike cells from other organisms, plant cells have a uniquely large and prominent organelle called the vacuole, occupying 90–95% of the cell’s volume (Owens and Poole 1979). The vacuole has important physiological functions, one of the main being the preservation of turgor pressure against the cell wall, thus supporting the structural stability of the cell and of the surrounding tissue (Marty 1999). It also serves as a storage tank holding many different materials needed by the cells, including but not limited to: sugars, metabolites, carbohydrates, lipids, amino acids, enzymes, proteins and anthocyanins (Marty 1999; Paris et al. 1996). In addition, the vacuole stores toxic ions and many other compounds that play a role in the defense against bacterial pathogens and herbivores (Martinoia et al. 2012; Van der Hoorn and Jones 2004).

Depending on tissue and cell types, vacuoles are diverse in their morphologies and functions (Swanson et al. 1998). Two main types are found in plants, the protein storage vacuoles (PSV) and the lytic vacuoles (LV) and they are functionally distinct (Hoh et al. 1995; Paris et al. 1996; Robinson et al. 1995). The typical storage compartment PSVs are most abundant in seeds (Epimashko et al. 2004; Hoh et al. 1995; Otegui et al. 2005; Paris et al. 1996; Swanson et al. 1998) and are formed during seed development and maturation. They accumulate large amounts of storage proteins, which are synthesized in the endoplasmic reticulum (ER) and delivered into vacuoles via the prevacuolar complex (Sansebastiano et al. 2017), where they remain stored until they are mobilized during germination and seedling growth (Bewley and Black 1994). PSVs also store defense proteins for the response against microbial pathogens and herbivores. After seed imbibition, PSVs are converted into lytic vacuoles (Bewley and Black 1994; Wang et al. 2007), which are the predominant compartments occurring in vegetative cells and are also called vegetative vacuoles. The identity of each vacuole is determined by its pH and by the presence of specific proteins known to be localized to PSVs or LVs, such as tonoplast intrinsic proteins (TIPs), which are integral membrane proteins found in specific vacuolar membranes (Jauh et al. 1999; Johnson et al. 1989). For example, the PSVs have a neutral pH and are defined by the presence of α- and δ-tonoplast intrinsic protein (TIP), whereas the LVs have an acidic pH and are marked by γ-TIP (Jiang et al. 2000; Jauh et al. 1999).

Vacuoles are increasingly recognized for their role in cellular signaling during growth (Zhang et al. 2014), the immune responses (Hatsugai and Hara-Nishimura 2010) and the regulation of cell death (CD) (Hara-Nishimura and Hatsugai 2011; Koyano et al. 2014). However, little is known about vacuolar structures and their roles in the defense response against pathogens. Several reviews have summarized the interactions between the plant vacuole and pathogenic microbes (Hara-Nishimura and Hatsugai 2011; Hatsugai et al. 2006). Here, the first section of this review focuses on the most up-to-date insights about the plant vacuolar structures and dynamics and vacuolar markers. The last section focuses on the role of the vacuole and vacuolar structures in plant-pathogen interactions.

Vacuolar structures

Recent advances in the visualization of the vacuole together with developments in image analysis has revealed the highly organized and complex morphology of the vacuole as well as its dynamics. The plant vacuole is surrounded by a membrane barrier known as the tonoplast, which separates the vacuolar content from the cytoplasm (Fig. 1). The semi-permeable tonoplast maintains a balance of nutrients and ions inside and outside of the vacuole, thus keeping a suitable turgor pressure in the plant cell.

Fig. 1
figure 1

Schematic diagram presents the vacuole and vacuolar structures in plant cells. The vacuole structures tonoplast, TVS, bulb, IVSP, nucleus, nucleolus, and plasma membrane are shown with arrowheads and the cytosolic flow with arrows. TVS transvacuolar structures, IVSP intravacuolar spherical structures

Full size image

The tonoplast not only surrounds the typical large vacuole but also other transient and mobile structures such as transvacuolar strands (TVS) and bulbs, represented in Fig. 1 (Ruthardt et al. 2005; Uemura et al. 2002). TVSs are dynamic thin tubular structures that traverse the central vacuole, containing cytoplasm and even small organelles (Uemura et al. 2002). Moreover, TVSs provide a direct connection between the perinuclear cytoplasm and the cortical cytoplasm of the cell and as such they act as an important transport route for the distribution of the cytoplasmic content, including the smaller organelles (Grolig and Pierson 2000; Nebenführ et al. 1999). Indeed, it was observed that Golgi bodies (Nebenführ et al. 1999), mitochondria (Van Gestel et al. 2002), endosomes (Ovečka et al. 2005; Ruthardt et al. 2005), and amyloplasts (Saito et al. 2005) dynamically move through the TVSs. Additionally, they play a role in the positioning of the nucleus (Katsuta et al. 1990; Kumagai and Hasezawa 2001; Williamson 1993). The dynamics of the TVS depends on the actin cytoskeleton; consequently, the disruption of the actin filaments leads to the loss of the transvacuolar strands and inhibition of their movement (Kovar et al. 2000; Kutsuna et al. 2003; Tominaga et al. 2000).

On the other hand, the bulbs are highly dynamic spherical structures between 1 and 22 µm of diameter located in the vacuolar lumen (Madina et al. 2018; Saito et al. 2002). 3-D reconstruction of electronic microscopic images indicates that the bulbs are formed of a double membrane and cytoplasmic material is detected between the two lipid bilayers (Saito et al. 2002). This double membrane of the bulbs is responsible for their brighter fluorescence signal compared to that of the tonoplast membrane (Saito et al. 2002). Similar to the dynamics of the TVSs, the movement of the bulbs is dependent on actin (Beebo et al. 2009; Uemura et al. 2002). However, whether bulbs are naturally occurring structures has recently been questioned by Segami et al. (2014), as they proposed that some bulbs are artifacts due to the dimerization of the GFP moiety of tagged tonoplastic proteins, while intravacuolar spherical structures (IVSP) form naturally. The IVSPs are different from bulbs in florescence intensity (twofolds the fluorescence intensity of the tonoplast vs. 3 or more folds for the bulbs) and thickness of the double membranes (Segami et al. 2014). These structures are also less abundant in cells and are believed to temporarily store membrane components, however, whether IVSPs are independent structures remains to be resolved (Segami et al. 2014). This model also needs to be further explored as GFP dimerization alone cannot explain the accumulation of bulbs in the YFP-2xFYVE A. thaliana line, as this marker binds to the membrane via a protein-lipid interaction and not a transmembrane domain (Saito et al. 2011). Two functions have been proposed for bulbs; first they may be involved in tonoplastic proteins degradation inside of the vacuole (Maîtrejean et al. 2011; Saito et al. 2002). On the other hand, they may be useful for rapid cell expansion by serving as membrane reservoirs (Saito et al. 2002). Recently, Han et al. (2015) even proposed that bulbs may not have specific functions other than the transient accumulation of tonoplast and cytoskeletal proteins until the bulb membrane is reabsorbed back into the tonoplast. Thus, the definitive function and biogenesis mechanism of bulbs still need to be fully elucidated.

Markers to study the vacuolar structures

The large central vacuole can be easily detected as a large transparent region in the plant cell as seen in light microscopy (Marty 1999), whereas the fine structure of the tonoplast and the intravacuolar compartments had mostly been studied by electron microscopy (Gaffal et al. 2007; Gao et al. 2015; Morita et al. 2002; Saito et al. 2002). The development of chemicals and fluorescent protein markers made possible the use of live-cell imaging to study the vacuole in a more detailed manner, which allows a greater understanding of its structure and dynamics under different growth conditions and various stress types (Reisen et al. 2005). Two different approaches for visualizing the plant vacuolar structures are presented here in below: chemical and protein markers. For selecting a particular technique, it is important to be aware of its advantages and limitations.

Chemical markers

To date, a number of fluorescent/chemical dyes have been identified for staining the tonoplast and vacuolar membranous structures (presented in Table 1). For example, the well-known amphiphilic styryl dyes, FM1-43 (N-(3-triethylammoniumpropyl)-4-(4-(dibutylamino)styryl) pyridinium dibromide) and FM4-64 (N-(3-triethylammoniumpropyl)-4-(4-diethylaminophenylhexatrienyl) pyridinium dibromide), are valuable and frequently used chemical dyes for vital staining of the tonoplast (i.e. staining in live cells) (Kim et al. 2001; Kutsuna and Hasezawa 2002; Kutsuna et al. 2003; Leshem et al. 2006; Okubo-Kurihara et al. 2008; Parton et al. 2001; Silady et al. 2008; Tanaka et al. 2007). The FM4-64 has also been reported for staining transvacuolar strands (Kutsuna and Hasezawa 2002; Kutsuna et al. 2003; Silady et al. 2008; Tanaka et al. 2007) and bulbs (Kim et al. 2001; Silady et al. 2008; Tanaka et al. 2007). The FM dyes initially stain the plasma membrane, then small cytoplasmic compartments, and finally reach the tonoplast in a process that is time, temperature, and energy-dependent. In general, to label the tonoplast with FM dyes, the cells are pulsed labeled for several minutes and then chased for several hours in fresh medium. The optimal duration of the chase period depends on the trafficking activity of the cells. In the initial stage of the chase period, the dyes simultaneously label the tonoplast and the other endomembrane components, including the endosomal organelles and developing cell plates (Higaki et al. 2008; Kutsuna and Hasezawa 2002; Ovečka et al. 2005; Parton et al. 2001; Tanaka et al. 2007; Ueda et al. 2001; Vermeer et al. 2006), whereas a longer chase period is required to visualize only the tonoplast (Emans et al. 2002; Kutsuna and Hasezawa 2002; Ueda et al. 2001). In addition to the FM dyes, BCECF (2′,7′-bis-(2-carboxyethyl)-5-(and-6)-carboxyfluorescein) fluorescently labels the vacuolar lumen, the tonoplast, the transvacuolar strands, and the bulbs (Higaki et al. 2007; Kutsuna and Hasezawa 2002; Mitsuhashi et al. 2000; Swanson et al. 1998; Toyooka et al. 2006). Vital staining using dyes is rapid, simple to perform, and compatible with the concomitant visualization of fluorescently tagged protein markers. However, this method has several limitations, in either sensitivity or specificity, and vital dyes themselves can sometimes introduce artifacts that must be taken care of during sample preparation or live cells imaging (Melan 1999; Schnell et al. 2012).

Table 1 Chemical and protein markers for labeling tonoplast, TVS, and bulbs
Full size table

Protein markers

In addition to chemical markers, many marker proteins tagged with fluorescent proteins are used to label the tonoplast and the vacuolar structures in living cells (Table 1). Most of those markers are integral membrane proteins, with the exception of YFP-2xFYVE which lacks a transmembrane domain and binds to phosphatidylinositol 3-phosphate (Saito et al. 2011). As YFP-2xFYVE labels bulbs more intensely than the tonoplast, it suggests that PI3P is more concentrated in the bulbs (Vermeer et al. 2006; Saito et al. 2011). However, not all tonoplast markers can also label bulbs, showing that bulb membranes are qualitatively different than tonoplast membranes (Saito et al. 2002). For example, γ-TIP-GFP was found to be concentrated in tonoplast and bulbs, whereas GFP-AtRAB7c, another tonoplast marker, did not give any fluorescent signal in bulbs from transgenic plants, although the presence of bulbs in these lines was confirmed by transmission electron microscopy (Saito et al. 2002). Some aquaporin isoforms also specifically label tonoplast or bulbs, for example TIP2;1-GFP exclusively localizes in the tonoplast but does not label the bulbs in salt-treated root cells, whereas TIP1;1 relocalized into intracellular spherical structures hypothesized to be tonoplast compartmentalization domains specialized for degradation of this isoform (Boursiac et al. 2005). In addition, two candidate effectors of pathogens have been observed to target the tonoplast and tonoplast-derived structures (TVSs and bulbs): HaRxLR17 from the oomycete Hyaloperonospora arabidopsidis (Hpa) (Bozkurt et al. 2015; Caillaud et al. 2012) and Mlp124357 from the fungus Melampsora larici-populina (Mlp) (Madina et al. 2018), and they can be used to study those organelles. The GFP-tagged marker proteins allow the visualization of the vacuole and vacuolar structures without further experimental manipulations once the transgenic lines are constructed. Although this is a powerful technique, it is laborious, time-consuming, and often not practical for many laboratories (Melan 1999; Schnell et al. 2012).

Protein distribution and motility of tonoplast and bulbs

Protein distribution

The complexity and dynamic changes in vacuolar lumen content raise the question of how cells regulate the movement of these materials between the cytosol and the vacuole. A controlled transport across the tonoplast is essential for appropriate plant responses to environmental conditions and for adequate intracellular signaling. The tonoplast contains numerous proteins that facilitate the transport of water, ions and metabolic products across the membrane (Martinoia et al. 2012; Zhang et al. 2014). In response to variation in the cytoplasmic environments, the activity of tonoplast enzymes, transporters, and channels is changed and thus they regulate the material exchange between the cytoplasm and the vacuolar lumen, maintaining cellular homeostasis. For example, H+ -ATPase (V-ATPase), H+ -pyrophosphatase (V-PPase), and water channels (aquaporins) have been characterized and their roles in the regulation of transport across the membrane have been discussed (Hedrich 2012; Martinoia et al. 2012; Neuhaus and Trentmann 2014).

On the other hand, the lipids of the tonoplast provide an essential molecular environment for the activity of the membrane proteins and serve as a barrier between the cytoplasm and the vacuolar lumen. Interestingly, lipids and proteins in the tonoplast are not always uniformly distributed and tend to be enriched in particular regions, termed membrane micro-domains, that depend on sphingolipids and sterols (Kusumi et al. 2005; Lillemeier et al. 2006; Minami et al. 2009). For example, the tonoplast of Arabidopsis suspension cultured cells contains micro-domains with higher ratios of the saturated phospholipids phosphatidylcholine (PC) and phosphatidylethanolamine (PE) in which the vacuolar-type proton ATPase (V-ATPase) was more abundant in detergent-resistant microdomains and appeared to be unevenly distributed in the tonoplast, whereas the vacuolar-type proton pyrophosphatase (V-PPase) was distributed evenly (Yoshida et al. 2013). Another study showed a similar non-uniform distribution of the V-ATPase in the tonoplast of isolated maize root cells (Kluge et al. 2004) and detergent-resistant micro-domains containing a high percentage of sphingolipids, free sterols and saturated fatty acids were described in the tonoplast of sugar-beet roots (Ozolina et al. 2011). The causes of this precise distribution of lipids and proteins in the tonoplast membrane are not fully elucidated yet; however, the environmental conditions seem to act as important regulators. For example, under phosphate deficiency the Arabidopsis phospholipase D PLDζ2 adopts an uneven distribution in which the higher concentrations of PLDζ2 were preferentially positioned close to mitochondria and chloroplasts and thus facilitated transfer between them and the tonoplast (Yamaryo et al. 2008).

Recently, it was observed that an overexpression in Arabidopsis of tonoplast intrinsic protein 1;1 fused with GFP (AtTIP1;1-GFP) labels both the tonoplast and the bulbs of the central vacuole but that the distribution of the GFP fusion protein was uneven along the tonoplast (Beebo et al. 2009). We also observed an uneven distribution of a fluorescently tagged protein on the bulb membrane (Video S1). Our recent report showed that the candidate effector protein Mlp124357-eGFP localized in the tonoplast, the TVS, and the bulbs of the vacuolar lumen in Arabidopsis (Madina et al. 2018). With the help of high-resolution 3-D imaging, we observed two different distribution patterns on the bulb membranes for both Mlp124357-eGFP and the well-known tonoplast marker γ-TIP-YFP. This suggests the existence of two distinct bulb types in these cells, some with a regular marker protein distribution and some with exclusion spots.

Protein mobility

Previous description of bulbs and prevailing models indicate they are connected to the tonoplast (Saito et al. 2002). On the other hand, a report pointed out the qualitative differences in protein content between the bulb membranes and the tonoplast (Saito et al. 2011). To investigate the connection between the two membranes, protein mobility between the tonoplast and the bulb membrane have recently been assessed using fluorescence recovery after photobleaching (FRAP) of young Arabidopsis leaf epidermal cells expressing Mlp124357-eGFP or γ-TIP-YFP. We observed the fluorescence of the bleached region of tonoplast recovered completely within 1 min, whereas the bleached area of the bulb membrane fluorescence did not recover (Madina et al. 2018). This observation can be explained by two mechanisms: (1) something restricts the movement of proteins from the tonoplast to the bulb membrane even though they are still attached or (2) the connection between the bulbs and the tonoplast is severed. Both explanations provide a new perception of the biology of tonoplast-derived substructures, but further investigation is required to validate the molecular mechanisms implicated. Furthermore, as the vacuole plays a crucial role in plant defense, the role of protein distribution and motility in tonoplast and bulbs should also be the focus of more research.

The vacuole in plant-pathogen interaction

The immune system of plants lack antibodies or phagocytosis. As an alternative, they have evolved numerous layers of active defense responses against pathogens including the production of reactive oxygen species (ROS) (Alvarez et al. 1998; Zhang et al. 2003) and of many other defense compounds such as phytoalexins (Neuhaus et al. 1991). Attempted attacks by avirulent pathogens may result in cell death (CD) in the tissues, a reaction known as the hypersensitivity response (HR), which is effective in preventing the spread of pathogens (Mur et al. 2007). These defense responses largely depend on the plant’s vacuole because it constitutes a reservoir for many secondary metabolites, hydrolytic enzymes and defense proteins (Marty 1999; Hara-Nishimura and Hatsugai 2011).

Secondary metabolites

Plant vacuoles accumulate a variety of secondary metabolites including cyanogenic glycosides, benzoxazinoids, and phenolics, some of which are thought to function as direct defenses against pathogens by reducing their performance, survival, and reproduction (Shitan 2016; Steppuhn et al. 2004). For example, the well-known cyanogenic glycosides are stored in the plant vacuole as inactive precursors and are able to form toxic hydrocyanic acid (HCN) in response to tissue damage by different phytopathogens (Vetter 2000; Freeman and Beattie 2008). A recent study reported that mutation of the cyanogenic 4-hydroxyindole-3-carbonyl nitrile (4-OH-ICN) pathway increases susceptibility to the bacterial pathogen Pseudomonas syringae in Arabidopsis, suggesting a role in inducible pathogen defense (Rajniak et al. 2015). Benzoxazinoids are among the most important plant defense compounds for grasses (Poaceae) and are also stored as inactive glucosides in the vacuole to avoid toxicity to the plant itself (Niemeyer 2009; Handrick et al. 2016; Zhou et al. 2018). While some benzoxazinoids are constitutively present, others are only synthesised following pathogen infection. Upon tissue damage, they undergo enzymatic and chemical degradation to become the active benzoxazinoid form (Niemeyer 2009). Benzoxazinoids have also been shown to act as defense signaling molecules and to induce callose deposition in response to the pathogenic fungal elicitor chitosan in maize (Ahmad et al. 2011). Another large group of secondary metabolites implicated in defense are the flavonoids, which are widely distributed in terrestrial plants and serve as defense compounds in the plant–microbe interaction (Harborne and Williams 2000; Taylor and Grotewold 2005; Grotewold 2006). For example, silencing of a G-type ABC transporter of M. truncatula (MtABCG10) results in a lower concentration of isoflavonoids in the roots which in turns results in increased growth of Fusarium oxysporum, indicating that flavonoids play an important role in the defense against root-infecting pathogens (Banasiak et al. 2013). Finally, the seed coat accumulates flavonoids to protect the embryo and the endosperm from external stresses such as UV radiation and pathogen infections (Lepiniec et al. 2006; Shimada et al. 2006, 2018).

Hydrolytic enzymes

Like animal lysosome, plant vacuole contains hydrolytic enzymes (e.g. aspartate proteinases, cysteine proteinases, and nucleases) that play an important role in the crucial events of plant cell death to prevent the spread of biotrophic pathogens (Boller and Kende 1979; Wada 2013). As first reported by Jones almost two decades ago, the plant vacuole play an important role in the programmed cell death that occurs in response to biotrophic pathogens (Jones 2001). It was also reported that the vacuolar processing enzyme (VPE) is up-regulated during cell death associated with leaf senescence and lateral root formation in Arabidopsis (Hatsugai et al. 2004). The same group confirmed that during the defense response, plants use the vacuole content in both a destructive and a non-destructive way. The destructive pathway is effective against virus infection, during which the tonoplast collapses and releases vacuolar hydrolytic enzymes called vacuolar processing enzymes (VPE) to suppress virus proliferation in the host cytosol (Hatsugai et al. 2004). The non-destructive way is effective against extracellular bacterial infections and involves the fusion of the plasma membrane to the tonoplast, which allows the discharge of the vacuolar content, including the proteasome subunit PBA1, in the apoplast. This leads to a hypersensitive cell death which suppresses the bacterial proliferation (Hatsugai et al. 2009). Interestingly, although structurally unrelated to caspases, both the vacuolar processing enzyme and the proteasome subunit PBA1 exhibit a caspase-like activity (Hatsugai et al. 2009). As it has been observed that proteasome defects impair the vacuole membrane fusion and VPE deficiency prevents virally induced hypersensitive cell death (Hatsugai et al. 2004, 2009), the identification of PBA1 and VPE substrates would help to unravel the molecular mechanisms of tonoplast breakdown or fusion with the PM.

Defense proteins

The defense proteins, including pathogenesis related proteins (PR proteins) (Neuhaus et al. 1991), myrosinases (Ueda et al. 2006), and lectins (Bowles et al. 1986), are located in the vacuole and act as an effective second line of defense when a pathogen causes tissue damage. For example, overexpression of the pathogen-inducible PR1 genes in tobacco enhances resistance to several fungi, including Peronospora tabacina and Phytophthora parasitica f.sp. nicotianae, and to the bacteria Pseudomonas syringae pv. tabaci (Broekaert et al. 2000). The association between PR-1 proteins and enhanced resistance against oomycetes has also been noted when PR1 expression was transiently silenced by double-stranded RNA interference in barley (Schultheiss et al. 2003). The myrosinases accumulate mainly in the vacuoles of idioblastic myrosin cells, a cell type specific to the abaxial side of the leaf known to accumulate myrosinases (Höglund et al. 1991, 1992; Thangstad et al. 1990, 1991). When plants experience tissue damage, the myrosinases are released from the collapsed vacuoles of the myrosin cells and start the hydrolysis of their glucosinolates substrates to produce isothiocyanates, which are toxic for bacteria. This chemical defense system is known as the myrosinase-glucosinolate system, which is also called the mustard oil bomb (Fuji et al. 2016; Grubb and Abel 2006; Halkier and Gershenzon 2006; Hopkins et al. 2009; Kissen et al. 2009; Rask et al. 2000; Wittstock and Halkier 2002). Moreover, overexpression of Ta-JA1, a jacalin-related lectin gene that resides in vacuole, has been found to increase resistance to bacteria, fungal, and viral pathogens in tobacco plant (Ma et al. 2010).

Vacuole dynamics

The relationship between vacuole dynamics and cell death during the defense response has been discussed for over a decade (Jones 2001). For example, during cell death of the tracheary element in Zinnia elegans, disintegration of the tonoplast was observed (Obara et al. 2001). Recent studies have demonstrated that during cell death induced by a pathogenic signal such as the oomycete elicitor cryptogein from Phytophthora cryptogea or the bacteria Erwinia carotovora, the complex vacuole of BY-2 cells simplified and then ruptured (Higaki et al. 2007; Hirakawa et al. 2015). This seem to be regulated by the vacuolar-localized protease called VPE (Higaki et al. 2011). The structural simplification of vacuoles is also observed in various processes involving program cell death, such as gibberellin-mediated cell death in central aleurone cells (Gao et al. 2015), embryogenesis of gymnosperms (Smertenko et al. 2003), and leaf formation in lace plants (Gunawardena 2007). However, it remains elusive whether simplification of the vacuole is a typical process in defense-related CD.

Tonoplast, TVS, and bulbs in plant-pathogen interaction

Although the vacuole plays an important role in plant defense, little is known about the manipulation of plant vacuolar structures by pathogens. To date, only two effector proteins, one from H. arabidopsidis and the other from M. larici-populina, have been found to reside in the host tonoplast, TVS, and bulbs (Bozkurt et al. 2015; Caillaud et al. 2012; Madina et al. 2018). Such localization of effector proteins may indicate a pathogenic strategy to modulate the host vacuole and vacuolar structures to suppress the vacuole-mediated defense response. For example, dynamic changes of the vacuolar structures are observed during cryptogein-induced PCD in tobacco BY-2 cells (Higaki et al. 2007). As illustrated in Fig. 2, immediately after cryptogein-treatment, no prominent structural changes of the cells were observed, rather, a complex vacuole structures featuring many transvacuolar strands were seen and the nuclei localizes in the central region of the cell (Fig. 2a). However, after some time the TVSs are gradually decreased in number and complexity and bulb structures appeared (Fig. 2b). Subsequently, bulb structures disappeared resulting the simpler vacuolar structure before the cell death (Fig. 2c, d). The simple-shaped vacuole weakened the tonoplast, leading to its ruptures, which sequentially would be lethal for the cell since it would dilute and acidify the cytosol and release cellular reactive secondary metabolites and hydrolytic enzymes (Higaki et al. 2011). A similar kind of vacuolar simplification was observed in BY-2 cells treated with culture filtrates of Erwinia carotovora, a plant pathogenic bacterium (Hirakawa et al. 2015), but in that situation the lysis of the plasma membrane seemed to occur while the vacuole was still intact. It is believed that the simplification of the vacuole is crucial for the vacuolar breakdown, but how these phenomena are orchestrated remains to be elucidated.

Fig. 2
figure 2

A schematic illustration of the reorganization of the vacuolar structures during pathogen-induced cell death in BY-2 cells. a After cryptogein treatment at 0 h, the vacuole contains many transvacuolar strands and the nucleus localizes at the central region of the cell. b After 12 h, the nucleus moves from the center to the periphery of the cell and bulb-like structures appear within the vacuole while transvacuolar strands are gradually disrupted. c After 18 h, the bulb-like structures disappear, and small spherical vacuoles appear. d After 24 h, the VM and PM lose their integrity. Dashed lines indicate broken membranes

Full size image

Conclusions and future perspectives

Although the role of the vacuole in plant-pathogen interactions has recently attracted much attention, our knowledge of the pathogen manipulation of host vacuolar membrane structures and trafficking remains minimal. As we have reviewed here, plant vacuolar structures are highly organized and dynamic, and they are involved in many cellular processes including plant-pathogen interactions. However, further studies are necessary to characterize the biogenesis mechanisms of bulbs and transvacuolar strands and to elucidate which physiological roles they play in the environmental responses of plants. Live-cell imaging using chemical dyes or fluorescently tagged proteins will undoubtedly serve as a critical technique to reveal the underlying mechanisms between vacuolar structural trafficking and plant-pathogen interactions and the use of plants defective in specific transport steps in vacuolar trafficking will help to elucidate the intracellular itinerary of still uncharacterized bulb membrane proteins.