TY - JOUR
AU - Van den Ende, Wim
AB - Abstract Vacuolar invertases (VIs) are highly expressed in young tissues and organs. They may have a substantial regulatory influence on whole-plant metabolism as well as on photosynthetic efficiency. Therefore, they are emerging as potentially interesting biotechnological targets to increase plant biomass production, especially under stress. On the one hand, VIs are well known as soluble and extractable proteins. On the other hand, they contain complex N-terminal propeptide (NTPP) regions with a basic region (BR) and a transmembrane domain (TMD). Here we analyzed in depth the Arabidopsis thaliana VI2 (AtVI2) NTPP by mutagenesis. It was found that correct sorting to the lytic vacuole (LV) depends on the presence of intact dileucine (SSDALLPIS), BR (RRRR) and TMD motifs. AtVI2 remains inserted into membranes on its way to the LV, and the classical sorting pathway (endoplasmic reticulum→Golgi→LV) is followed. However, our data suggest that VIs might follow an alternative, adaptor protein 3 (AP3)-dependent route as well. Membrane-anchored transport and a direct recognition of the dileucine motif in the NTPP of VIs might have evolved as a simple and more efficient sorting mechanism as compared with the vacuolar sorting receptor 1/binding protein of 80 kDa (VSR1/BP80)-dependent sorting mechanism followed by those proteins that travel to the vacuole as soluble proteins. Introduction Sucrose (Suc) is the cornerstone of higher plant metabolism. Utilization of Suc as a source of carbon and energy depends on its cleavage into glucose and fructose. Therefore, Suc-splitting enzymes such as invertases and sucrose synthases fulfill important roles in source–sink relations and carbohydrate partitioning (Ruan et al. 2010, Vargas and Salerno 2010). The action of Suc-splitting enzymes probably modulates the ratio of Suc to hexose (Hex), thereby influencing sugar-specific signaling pathways, controlling plant growth, development, photoperiodic flowering and circadian clock regulation (Dalchau et al. 2011, Seo et al. 2011, Xiang et al. 2011). Plant vacuolar invertases (VIs) are well-known sink strength mediators, especially in young sink tissues such as seedling hypocotyls, developing fruits and expanding tissues (Sergeeva et al. 2006, Bonfig et al. 2007, Nie et al. 2010). High VI gene expression and increased VI activities are typically observed in stressed plants (Kim et al. 2000, Reinders et al. 2005, Ji et al. 2007, Ruan et al. 2010, Yamada et al. 2010). The model plant Arabidopsis contains only two VI isoforms termed AtVI1 (Atβfruct3, At1g62660) and AtVI2 (Atβfruct4, At1g12240). The function of AtVI2 has been studied in greater detail. Bioinformatics and in vivo studies demonstrated that AtVI2 fulfills a prominent role in plant growth and the generation of biomass (Nägele et al. 2010, Wang et al. 2010). Plant vacuoles are multifunctional organelles, serving physical and metabolic functions essential to plant life and fulfilling crucial roles in cellular responses to abiotic/biotic stresses (Bassham and Raikhel 2000). Both lytic vacuoles (LVs) and protein storage vacuoles (PSVs) can be discerned (Rogers 2008), and plants developed separate trafficking pathways to these vacuoles (Zouhar and Rojo 2009, Xiang et al. 2013). Vacuolar delivery typically depends on the presence of vacuolar sorting determinants (VSDs). Three types are discerned in plants (Xiang et al. 2013, and references therein): (i) sequence-specific VSDs, often situated at the N-terminus of a protein; (ii) C-terminal VSDs; and (iii) protein structure-dependent VSDs. Most proteins travel to these vacuoles in a soluble form, unless they are destined to function in the tonoplast (Wolfenstetter et al. 2012, Xiang et al. 2013). After passing through the Golgi, typical storage proteins reach the PSVs via dense vesicles and multivesicular bodies/pre-vacuolar compartments (MVBs/PVCs) by bulk flow or aggregation. Other soluble proteins (such as aleurain; Di Sansebastiano et al. 2001) reach the LVs via another type of MVBs/PVCs (termed PVCs from this point on). In this process, the VSDs in these soluble proteins are recognized by vacuolar sorting receptors (VSRs) such as binding protein of 80 kDa (BP80), a type I membrane protein with an Nout/Cin (N-terminus out of the cytoplasm, C-terminus in the cytoplasm) topology (Paris and Neuhaus 2002, Neuhaus and Paris 2005, Zouhar and Rojo 2009). In their turn, VSRs are recognized by adaptor protein complexes (APs: AP1–AP5). In addition to the classical pathway [endoplasmic reticulum (ER)→Golgi→PVC], an alternative pathway was reported for the sorting of soluble proteins to the vacuole/lysosome in yeast and in mammals (Seaman et al. 1997, Wang et al. 2011). It is likely that such an alternative pathway might also exist in plants (Dell’Angelica 2009, Feraru et al. 2010, Zwiewka et al. 2011, Wolfenstetter et al. 2012, Xiang et al. 2013). In contrast to cell wall invertases, it was noticed that VIs contain a C-terminal extension (Sturm 1999) and a much longer and complex N-terminal propeptide region (NTPP), containing a transmembrane domain (TMD) (Ji et al. 2005, Ji et al. 2007). Moreover, these authors already detected a considerable homology between the NTPPs of plant VIs and yeast alkaline phosphatase, suggesting that plant VIs may not travel to the vacuole as soluble proteins. Instead, it was postulated that they reach the tonoplast as type II transmembrane proteins with an Nin/Cout topology, following the same path as described before for a modified yeast invertase with an alkaline phosphatase sorting signal in transgenic tobacco plants (Chrispeels and Barrieu 1999). Recent data indeed demonstrated membrane-anchored transport all the way to the tonoplast, after which rapid processing by proteases probably occurs (Jung et al. 2011, Rae et al. 2011). However, these authors did not focus on the visualization of the membrane localization during the sorting process. Previously, it was also reported that AtVI2 is localized in the central vacuole (CV) of Arabidopsis roots (Rojo et al. 2003) and in leaf mesophyll cells (Carter et al. 2004), while in seedling roots, it localized in ER bodies prior to arrival at the CV (Rojo et al. 2003). Here, it is reported that correct vacuolar sorting of AtVI2 in Arabidopsis leaf protoplasts requires a dileucine core-containing motif SSDALLPIS as well as an intact BR and a TMD. For the first time, AtVI2–green fluorescent protein (GFP) fusion proteins were visualized at the membranes of punctate structures, on their way to the LV. Furthermore, we demonstrate that the sorting process occurs independently of VSR1/BP80. Moreover, our studies on AP3β knock-out lines suggest that AtVI2 might, at least partially, follow an alternative pathway, in addition to the classic pathway via the PVC. Results AtVI2 is targeted to the vacuole by a complex N-terminal propeptide VIs typically contain long, complex NTPPs containing a dileucine motif, an YXXΦ (X is any amino acid, Φ is a hydrophobic residue) motif, a BR and a TMD (Table 1). The subcellular localizations of full-length (AtVI2) and N-terminal (NAtVI2) GFP fusion proteins were determined in Arabidopsis protoplasts (18 h post-transformation). Both fusion proteins reached the CV (Table 2), confirming that the NAtVI2 is sufficient for vacuolar targeting (Fig. 1A). GFP-only constructs failed to reach the CV, but localized in the cytosol (not shown). Similarly, AtVI1 and NAtVI1 also reached the CV, but at an earlier time point (at 15 h post-transformation, Fig. 1B). After 15 h, most of the NAtVI2 had not yet reached the CV, and localized in punctate structures (Fig. 1C). In all cases, fluorescence could only be detected in the vacuolar lumen, and not at the tonoplast. Fig. 1 View largeDownload slide Subcellular localization of AtVI–GFP/NAtVI–GFP fusion proteins. Full-length–GFP (AtVI1, AtVI2) and NTPP–GFP fusion proteins (NAtVI1, NAtVI2) were produced and then transformed into Arabidopsis mesophyll protoplasts. (A) Pictures were taken under GFP fluorescence and bright field (BF) at 18 h post-transformation for AtVI2 and NAtVI2 GFP fusions; (B) pictures were taken at 15 h post-transformation for AtVI1 and NAtVI1 GFP fusions; (C) pictures were taken at 15 h post-transformation for the AtVI2 and NAtVI2 fusion proteins. Scale bar, 10 µm. Fig. 1 View largeDownload slide Subcellular localization of AtVI–GFP/NAtVI–GFP fusion proteins. Full-length–GFP (AtVI1, AtVI2) and NTPP–GFP fusion proteins (NAtVI1, NAtVI2) were produced and then transformed into Arabidopsis mesophyll protoplasts. (A) Pictures were taken under GFP fluorescence and bright field (BF) at 18 h post-transformation for AtVI2 and NAtVI2 GFP fusions; (B) pictures were taken at 15 h post-transformation for AtVI1 and NAtVI1 GFP fusions; (C) pictures were taken at 15 h post-transformation for the AtVI2 and NAtVI2 fusion proteins. Scale bar, 10 µm. Table 1 Alignment of NTPP sequences of VIs from different plant species     Basic regions are underlined and immediately followed by their adjacent TMDs. Acidic motifs are in gray. GVSEK consensus motifs are in bold and italic. Bo, Brassica oleracea; Os, Oryza sativa; Ib, Ipomoea batatas; Pc, Prunus cerasus; Bv, Beta vulgaris; Ps, Pisum sativum. BoVI1, AF274298; OsVI1, AF019113; IbVI2, AY037937; PcVI2, AY048579; OsVI3, AF276704; BvVI1, AJ277455; PsVI1, AAM526062; OsVI2, AY037871. View Large Table 2 List of NAtVI2 mutants and their subcellular localization         AtVI2 is a type II transmembrane protein. Its transmembrane domain (TMD, shadow) remains inserted in membranes during sorting. The dileucine consensus motif (bold) and the positively charged residues in the basic region (BR, bold and italic) that are essential for AtVI2 sorting to the LV are indicated. Motifs that are not necessary but can partially affect AtVI2 sorting efficiency are indicated in italic. Cyt, cytosol; ER, endoplasmic reticulum; Vac, vacuole; Ps, punctate structures; X, amino acid deletion. View Large AtVI2 reaches the lytic vacuole but not the protein storage vacuole Di Sansebastiano et al. (2001) investigated protein trafficking to LVs and PSVs by using distinct vacuolar GFP fusion proteins. In their work, the C-terminal VSD of tobacco chitinase A (GFP–Chi) was used as a PSV marker, and the NTPP of barley aleurain (Aleu–GFP) was used as an LV marker. Here, the co-expression of the NAtVI2–red fluorescent protein (RFP) fusion protein and Aleu–GFP showed partial co-localization in some punctate structures (Supplementary Fig. S1A, arrows). Such co-localization was not observed with GFP–Chi (Supplementary Fig. S1B). Again, no specific higher fluorescence densities were detected at the tonoplast (Supplementary Fig. S1), indicating immediate processing of the NAtVI2–mRFP fusion protein in the acidic environment of the LVs. AtVI2 reaches the vacuole via a membrane anchor sorting mechanism: effect of wortmannin To determine the path followed by the NAtVI2–GFP fusion protein, time course localization experiments (12, 15 and 18 h post-transformation) were performed (Figs 2, 3) with RFP fusion proteins as markers for the ER (p24–RFP) and Golgi apparatus (KAM1ΔC–mRFP), respectively. At 12 h post-transformation, most of the NAtVI2 fusion protein appeared in the ER network, as confirmed by co-localization with p24–RFP (Fig. 2A). In the vicinity of the ER network, small punctate structures with membrane-localized fluorescence were observed (Fig. 2A, GFP panel indent). At 15 h post-transformation, the fusion protein was detected in the Golgi apparatus (Fig. 2B) as well as in the membranes of associated punctate structures (Fig. 2B2). As demonstrated before, most of the NAtVI2 fusion protein was localized in the CV (Fig. 1) at 18 h post-transformation. Fig. 2 View largeDownload slide AtVI2 travels through the ER and the Golgi via a membrane-anchored mechanism. (A) Co-expression of NAtVI2–GFP and the ER marker p24–RFP. Images were taken 12 h post-transformation. Single and merged versions are presented. Magnification shows a punctate structure with membrane-localized NAtVI2. A bright field (BF) image is shown for comparison. Scale bar, 10 µm. (B) Co-expression of NAtVI2–GFP and the Golgi marker KAM1ΔC–mRFP. Confocal images from two different Z stacks [B(1) and B(2)] of a single protoplast were taken at 15 h post-transformation. Single and merged versions are presented. Arrows indicate the punctate structures in stack B(1). Magnification in stack B(2) shows punctate structures with membrane-localized NAtVI2. A BF image is shown for comparison. Scale bar, 10 µm. Fig. 2 View largeDownload slide AtVI2 travels through the ER and the Golgi via a membrane-anchored mechanism. (A) Co-expression of NAtVI2–GFP and the ER marker p24–RFP. Images were taken 12 h post-transformation. Single and merged versions are presented. Magnification shows a punctate structure with membrane-localized NAtVI2. A bright field (BF) image is shown for comparison. Scale bar, 10 µm. (B) Co-expression of NAtVI2–GFP and the Golgi marker KAM1ΔC–mRFP. Confocal images from two different Z stacks [B(1) and B(2)] of a single protoplast were taken at 15 h post-transformation. Single and merged versions are presented. Arrows indicate the punctate structures in stack B(1). Magnification in stack B(2) shows punctate structures with membrane-localized NAtVI2. A BF image is shown for comparison. Scale bar, 10 µm. Fig. 3 View largeDownload slide AtVI2 reaches the vacuole and sorting is sensitive to wortmannin. (A) Expression of NAtVI2–GFP at 18 h post-transformation. The NAtVI2–GFP membrane localization in putative ‘small vacuoles’ was found in some of the protoplasts. A bright field (BF) image is shown for comparison. Scale bar, 10 µm. (b) Expression of NAtVI2–GFP in protoplasts treated with 20 µM wortmannin (+Wort). The left panel was taken at 14 h post-transformation, the right panel at 18 h post-transformation. Magnification shows a punctate structure with membrane-localized NAtVI2–GFP. A BF image is shown for comparison. Scale bar, 10 µm. (C) Expression of Aleu–GFP in wild-type protoplasts at 18 h post-transformation, in the absence (left panel) and presence (right panel) of 20 µM wortmannin (+Wort). A BF image is shown for comparison. Scale bar, 10 µm. Fig. 3 View largeDownload slide AtVI2 reaches the vacuole and sorting is sensitive to wortmannin. (A) Expression of NAtVI2–GFP at 18 h post-transformation. The NAtVI2–GFP membrane localization in putative ‘small vacuoles’ was found in some of the protoplasts. A bright field (BF) image is shown for comparison. Scale bar, 10 µm. (b) Expression of NAtVI2–GFP in protoplasts treated with 20 µM wortmannin (+Wort). The left panel was taken at 14 h post-transformation, the right panel at 18 h post-transformation. Magnification shows a punctate structure with membrane-localized NAtVI2–GFP. A BF image is shown for comparison. Scale bar, 10 µm. (C) Expression of Aleu–GFP in wild-type protoplasts at 18 h post-transformation, in the absence (left panel) and presence (right panel) of 20 µM wortmannin (+Wort). A BF image is shown for comparison. Scale bar, 10 µm. In some of the protoplasts at 18 h post-transformation, NAtVI2–GFP was also detected in the membranes of punctate structures with variable size (Fig. 3A). Although the exact nature of all these above-mentioned punctate structures remains unclear and requires further investigations, the major point is that the fluorescence is clearly detected in the membranes of these structures, visualizing the membrane-anchored transport of the GFP fusion protein. It can be speculated that some of the larger structures represent PVCs, late PVCs or fused PVCs, also termed ‘small vacuoles’, in accordance with the previously determined presence of AtVI2 in PVCs (Jung et al. 2011). Wortmannin is a well-known stimulator of the formation of such ‘small vacuoles’ (Wang et al. 2009). Wortmannin treatment of protoplasts transformed with NAtVI2–GFP showed the presence of punctate structures with fluorescence in their membranes (Fig. 3B). In contrast, such clear membrane fluorescence was not observed in wortmannin-treated protoplasts transformed with an Aleu–GFP construct (Fig. 3C). Overall, it can be concluded that AtVI2 remains intact and inserted into membranes via the TMD in its N-terminus (Tables 1, 2), until it reaches the tonoplast. Sorting of AtVI2 in vsr1/bp80 and AP3β loss-of-function protoplasts To examine whether NAtVI2 and aleurain follow an identical sorting pathway, NAtVI2–GFP and Aleu–GFP were transformed into protoplasts isolated from two vsr1/bp80 lines (SALK_123661 and SALK_150894) and wild type (WT) plants, respectively. While the NAtVI2 fusion protein was correctly targeted to the CV both in vsr1/bp80 (Fig. 4A) and in WT (Fig. 1A) protoplasts, Aleu–GFP did not reach the CV in vsr1/bp80 protoplasts (Fig. 4B) while in WT protoplasts the fluorescence was still detected there (Fig. 3C). However, it should be noted that only very few vsr1/bp80 protoplasts showed fluorescence of Aleu–GFP, suggesting that the majority of the protoplasts may have exported their fusion protein into the medium. Taken together with the wortmannin experiments (Fig. 3), these observations suggest that both AtVI2 and aleurain are targeted to the LV through the same classic pathway, but aleurain (a typical soluble protein lacking a TMD) trafficking depends on VSR1/BP80 while AtVI2 (a protein containing one functional TMD) trafficking does not (Fig. 5). Fig. 4 View largeDownload slide Sorting of AtVI2 and Aleu in bp80 protoplasts. Expression of NAtVI2–GFP (A) and Aleu–GFP (B) in vsr1/bp80 Arabidopsis mesophyll protoplasts. Images were taken at 18 h post-transformation. NAtVI2–GFP localizes in the central vacuole and Aleu–GFP does not. Bright field (BF) images are presented for comparison. Scale bar, 10 µm. Fig. 4 View largeDownload slide Sorting of AtVI2 and Aleu in bp80 protoplasts. Expression of NAtVI2–GFP (A) and Aleu–GFP (B) in vsr1/bp80 Arabidopsis mesophyll protoplasts. Images were taken at 18 h post-transformation. NAtVI2–GFP localizes in the central vacuole and Aleu–GFP does not. Bright field (BF) images are presented for comparison. Scale bar, 10 µm. Fig. 5 View largeDownload slide Hypothetical models for aleurain and AtVI2 sorting: a comparison. (A) In the ER, BP80 takes aleurain as cargo. The BP80–aleurain interaction most probably persists in the Golgi and Golgi-derived structures, but is broken by the lowered pH in the PVC (Saint Jean et al. 2010, this work). (B) In the ER, AtVI2 anchors in the membrane. It remains membrane bound during its passage through ER-derived vesicles, the Golgi and up to the PVC (Jung et al. 2011). N-terminal processing by vacuolar proteases is accomplished when AtVI2 reaches the acidic LV. Figures are adapted from Chrispeels and Barrieu (1999) and Saint Jean et al. (2010). Fig. 5 View largeDownload slide Hypothetical models for aleurain and AtVI2 sorting: a comparison. (A) In the ER, BP80 takes aleurain as cargo. The BP80–aleurain interaction most probably persists in the Golgi and Golgi-derived structures, but is broken by the lowered pH in the PVC (Saint Jean et al. 2010, this work). (B) In the ER, AtVI2 anchors in the membrane. It remains membrane bound during its passage through ER-derived vesicles, the Golgi and up to the PVC (Jung et al. 2011). N-terminal processing by vacuolar proteases is accomplished when AtVI2 reaches the acidic LV. Figures are adapted from Chrispeels and Barrieu (1999) and Saint Jean et al. (2010). Further examination was performed on the possibility that AtVI2 could also follow an alternative, AP3-dependent pathway (Feraru et al. 2010). It is known that the AP3µ subunit recognizes an YXXΦ motif, while the AP3β subunit recognizes a dileucine motif in mammals and in yeast (Bonifacino and Traub 2003). Similar AP3 subunits have been found in plants (Robinson et al. 2005, Bassham et al. 2008). Thus, NAtVI2 was transformed into protoplasts isolated from an ap3µ line (SALK_064486 and SALK_127431, Feraru et al. 2010), an ap3β line (pat2-1 and pat2-2, Niihama et al. 2009) and from WT plants. Aleu–GFP was used as a control. As expected, Aleu–GFP was correctly sorted into the CV in the mutant (Fig. 6) and WT (Fig. 3C) protoplasts, demonstrating that it travels through the classical, AP3-independent pathway. Intriguingly, NAtVI2 fusion proteins showed a variable behavior. NAtVI2 was correctly sorted to the CV in both WT (Fig. 1) and ap3µ protoplasts (Fig. 6A). However, in ap3β protoplasts, it was expressed in punctate structures (40%), in the CV (40%) or showed an apparent cytosolic localization (20%) (Fig. 6B). Fig. 6 View largeDownload slide Sorting of AtVI2 and Aleu in ap3β and ap3µ protoplasts. NAtVI2–GFP and Aleu–GFP are expressed in protoplasts isolated from ap3µ SALK lines (A) and ap3β SALK lines (B), respectively. All images were produced at 18 h post-transformation. Bright field (BF) images are presented for comparison. Scale bar, 10 µm. Fig. 6 View largeDownload slide Sorting of AtVI2 and Aleu in ap3β and ap3µ protoplasts. NAtVI2–GFP and Aleu–GFP are expressed in protoplasts isolated from ap3µ SALK lines (A) and ap3β SALK lines (B), respectively. All images were produced at 18 h post-transformation. Bright field (BF) images are presented for comparison. Scale bar, 10 µm. The dileucine motif and the basic region are crucial for correct sorting As type II transmembrane proteins, VIs are believed to take the Nin/Cout orientation (Ji et al. 2005. Jung et al. 2011. Rae et al. 2011), as predicted by the TMpred program (http://www.ch.embnet.org/software/TMPRED_form.html). To examine which NTPP amino acids are involved in sorting to the CV, we designed an extended series (>40) of mutants in NAtVI2 by site-directed mutagenesis (Table 2), and recorded their localization at 18 h post-transformation. In each case, 80 protoplasts were examined. As expected, the partial loss of the TMD resulted in a cytosolic localization (NAtVI2m28, Table 2). Deletion of the first 25 amino acids retained the TMD, and, as expected, led to localization in the ER (NAtVI2m1, Table 2). This 1–25 amino acid region was further explored by an array of specific deletion mutants as listed in Table 2. Deletions in the N-terminal 3SSDALLPIS11 region (NAtVI2m3–NAtVI2m12, Table 2) greatly affected protein sorting (Fig. 7A–E). Single deletions of D5, I10 and S11 and removal of the dileucine motif L7L8 and the diserine motif S3S4 blocked the protein in the ER network (Table 2, Fig. 7A–E). Deletion of only one L or S in the dileucine and diserine motif resulted in partial obstruction, as did changing P9 into an alanine (Table 2). Importantly, deletion of A5 and A12 did not affect sorting to the CV (Table 2). The correct sorting of the NAtVI2m7 deletion mutant (SSDLLPIS) suggests that the presence of SSD and LLP motifs is more important than their exact location relative to each other. In conclusion, the explored 3SSDALLPIS11 sequence contains features that partially resemble the well-known DXXLL dileucine-based sorting signals, a conserved sequence derived from protein sorting studies in mammals and yeast (Bonifacino and Traub 2003), in which the requirement for the D and LL were found to be strict, and the presence of S was not essential. Of particular importance, deletion of a YXXΦ like-motif (YTRL, Table 2), which was first considered to be crucial for sorting to the vacuole (Sanderfoot et al. 1998, Happel et al. 2004, Foresti and Denecke 2008, Rae et al. 2011), did not lead to disrupted sorting of NAtVI2 (NAtVI2m17, Fig. 7F). Fig. 7 View largeDownload slide The role of the dileucine motif during vacuolar sorting of the NAtVI2–GFP fusion protein (Table 2). Localization of a series of NAtVI2–GFP mutants, affected in the dileucine motif 3SSDALLPIS11 and the 21YTRL25 motif, in Arabidopsis mesophyll protoplasts. (A) NAtVI2m4, S3S4 deletion mutant; (B) NAtVI2m5, D5 deletion mutant; (C) NAtVI2m6, 6ALLP9 deletion mutant; (D) NAtVI2m11, I10 deletion mutant; (E) NAtVI2m12, S11 deletion mutant; (F) NAtVI2m17, 21YTRL25 deletion mutant. All images were collected at 18 h post-transformation. Bright field (BF) images are shown for comparison. Scale bar, 10 µm. Fig. 7 View largeDownload slide The role of the dileucine motif during vacuolar sorting of the NAtVI2–GFP fusion protein (Table 2). Localization of a series of NAtVI2–GFP mutants, affected in the dileucine motif 3SSDALLPIS11 and the 21YTRL25 motif, in Arabidopsis mesophyll protoplasts. (A) NAtVI2m4, S3S4 deletion mutant; (B) NAtVI2m5, D5 deletion mutant; (C) NAtVI2m6, 6ALLP9 deletion mutant; (D) NAtVI2m11, I10 deletion mutant; (E) NAtVI2m12, S11 deletion mutant; (F) NAtVI2m17, 21YTRL25 deletion mutant. All images were collected at 18 h post-transformation. Bright field (BF) images are shown for comparison. Scale bar, 10 µm. A chain of positively charged amino acids termed the BR, immediately preceding the TMD, is needed for correct delivery of alkaline phosphatase to the yeast vacuole (Ji et al. 2005). A similar BR (PRRRRP) is observed in the NTPP of AtVI2 (Table 2). Deletion of all four arginines (NAtVI2m21, Table 2) led to accumulation of fusion proteins in the ER in the majority (>80%) of protoplasts (Fig. 8A, left panel). Deletion of two arginines (NAtVI2m22) led to a delay in protein sorting, with fluorescence present in punctate structures in approximately 50% of the protoplasts (Fig. 8B, left panel). Mutagenesis of two arginines to alanines (NAtVI2m23) resulted in partial ER localization (Fig. 8C). Similar effects were found when P36, P41 or both were mutated into alanine (NAtVI2m24–NAtVI2m26, Fig. 8D–F). Mutation of both prolines (NAtVI2m26) showed an ER localization in the majority (>80%) of the protoplasts (Fig. 8F, left panel), while others showed a correct vacuolar localization (Fig. 8F, right panel). It can be concluded that at least two arginines are needed for sorting to the vacuole, and the presence of two neighboring prolines seems to contribute to the sorting efficiency. Fig. 8 View largeDownload slide Localization of NAtVI2–GFP versions mutated in the basic region (BR) (Table 2). Localization of a series of NAtVI2–GFP mutants, affected in the BR 36PRRRRP41, in Arabidopsis mesophyll protoplasts. (A) In NAtVI2m21, a four arginine (RRRR) deletion mutant, >80% of the protoplasts showed an ER localization (left panel), while others localized in the CV (right panel). (B) In NAtVI2m22, a two arginine deletion mutant, around 50% of the protoplasts showed an ER localization (left) and the other 50% showed punctate structures and CV localization (right). (C) In NAtVI2m23, R37R38 is changed to A37A38; approximately 50% of the protoplasts showed ER localization (left) and the other 50% showed punctate structures and CV localization (right). (D) In NAtVI2m24, a P36 to A36 mutant, approximately 40% of protoplasts showed an ER localization (left) and the other 60% localized in the CV (right). (E) In NAtVI2m25, a P41 to A41 mutant, approximately 40% of protoplasts showed an ER localization (left) and the other 60% localized in the CV (right). (F) In NAtVI2m26, both prolines are converted to alanine; >80% of the protoplasts showed an ER localization (left) and the others showed a CV localization (right). All images were collected at 18 h post-transformation. Bright field (BF) images are presented for comparison. Scale bar, 10 µm. Fig. 8 View largeDownload slide Localization of NAtVI2–GFP versions mutated in the basic region (BR) (Table 2). Localization of a series of NAtVI2–GFP mutants, affected in the BR 36PRRRRP41, in Arabidopsis mesophyll protoplasts. (A) In NAtVI2m21, a four arginine (RRRR) deletion mutant, >80% of the protoplasts showed an ER localization (left panel), while others localized in the CV (right panel). (B) In NAtVI2m22, a two arginine deletion mutant, around 50% of the protoplasts showed an ER localization (left) and the other 50% showed punctate structures and CV localization (right). (C) In NAtVI2m23, R37R38 is changed to A37A38; approximately 50% of the protoplasts showed ER localization (left) and the other 50% showed punctate structures and CV localization (right). (D) In NAtVI2m24, a P36 to A36 mutant, approximately 40% of protoplasts showed an ER localization (left) and the other 60% localized in the CV (right). (E) In NAtVI2m25, a P41 to A41 mutant, approximately 40% of protoplasts showed an ER localization (left) and the other 60% localized in the CV (right). (F) In NAtVI2m26, both prolines are converted to alanine; >80% of the protoplasts showed an ER localization (left) and the others showed a CV localization (right). All images were collected at 18 h post-transformation. Bright field (BF) images are presented for comparison. Scale bar, 10 µm. The motifs NDEG, TITS, RARL and LWKL downstream of the TMD (NAtVI2m32, NAtVI2m34, NAtVI2m35, NAtVI2m40, Table 2) were also found to be involved in vacuolar sorting. Deletion of these motifs partially slowed down the delivery of proteins to the CV, with about 40% of the protoplasts showing fluorescence in the ER (Supplementary Fig. S2A). Other amino acid regions have been investigated as well (Table 2), but it was found that they are not involved in the sorting process (Table 2). However, one noticeable exception is the intriguing glutamate stretch (13REEEP17 motif) upstream of the TMD. Two extra glutamate residues were inserted in this motif (REEEEEP, NAtVI2m15, Table 2), resulting in an apparent delay of delivering punctate structures to the CV (Table 2, Supplementary Fig. S2B). However, it cannot be excluded that this result could arise from altered protein stability, and this requires further investigations. Discussion The NTPPs of VIs are more complex and longer than those of cell wall invertases. Therefore, more than a decade ago they were predicted to contain vacuolar targeting information (Sturm 1999), and were suggested to enter the ER without subsequent removal of their TMD, resulting in membrane-anchored sorting (Chrispeels and Barrieu 1999, Ji et al. 2005). In accordance with recent data on the NTTP of sugar cane VI (Rae et al. 2011), our results show that the NTPPs of AtVI1 and AtVI2 (Fig. 1) are sufficient to lead GFP fusion proteins to the vacuole, while this is not possible with GFP-only constructs (Rae et al. 2011). The absence of specific fluorescence at the tonoplast (Fig. 1) reconfirmed the recently established view that AtVI2 fusion proteins are rapidly processed by vacuolar proteases, immediately after their arrival at the tonoplast (Jung et al. 2011, Rae et al. 2011). Taken together with the data of Jung et al. (2011), our data (Figs. 2, 3) confirm that AtVI2 follows the classic ER→Golgi→ PVC pathway to the LV (Supplementary Fig. S1), and not to the PSV (Supplementary Fig. S1) as expected (Hunter et al. 2007). During the process, it was visualized here that AtVI2 remains intact and inserted in membranes (Figs. 2, 3), through the TMD in its N-terminus (Tables 1, 2). In contrast, aleurain, lacking such a TMD, needs to be recognized as cargo of BP80 (Figs. 3C, 4). Taken all together, these data lead to a model explaining how aleurain (Fig. 5A) and VIs (Fig. 5B) reach the LV through the classic pathway. The results of Saint Jean et al. (2010) indicated that aleurain is already released from BP80 in the more acidic environment of the PVC. Accordingly, all the punctate structures (some of them probably representing PVCs or small vacuoles) in Fig. 3C (Aleu–GFP) showed a uniform fluorescence, while those observed for AtVI2–GFP (Fig. 3A, B) showed fluorescence in their membranes. This is in accordance with the results of Jung et al. (2011), demonstrating the presence of AtVI2 in isolated PVC membranes. Rae et al. (2011) already predicted that VIs reach the tonoplast intact, after which rapid processing occurs by vacuolar proteases. Taken together, the decreasing pH in the PVCs seems to be able to disrupt the interaction between aleurain and BP80, releasing aleurain (Fig. 5A). In contrast, although proteases might be present in the PVC (Rae et al., 2011), it seems that the PVC does not harbor active proteases that are able to process VIs. The pH might not be low enough to activate them fully, a process that probably only occurs when they reach the LV (Fig. 5B). Vacuolar processing enzyme γ is such an example of a proteolytic enzyme that is promptly activated when it is released in the acidic vacuolar environment, and it is known to catalyze AtVI2 degradation during aging (Rojo et al. 2003). It can be hypothesized that vacuolar degradation of VIs is carefully regulated, in order to delineate a time window for sucrose degradation before these proteins are subjected to subsequent degradation. It has been demonstrated both in mammals and in yeast that an alternative AP3-mediated pathway is involved in lysosomal trafficking (Bonifacino and Traub 2003). Recent research showed that AP3 is also essential for plant vacuolar function (Braulke and Bonifacino 2009, Feraru et al. 2010, Zwiewka et al. 2011). Here, it is shown that AtVI2 sorting is partially affected in AP3β (recognizing dileucine-based motifs) but not in AP3µ knock-out protoplasts (Fig. 6). Therefore, a similar route probably exists in plants as well. Taken together with the recently obtained data on the transport of other transmembrane proteins to the tonoplast (Wolfenstetter et al. 2012), it can be concluded that plants probably use two different pathways, a classical and an alternative AP3-dependent pathway, to transport transmembrane proteins to the tonoplast/vacuole, as recently proposed in fig. 1B of Xiang et al. (2013). Furthermore, in Arabidopsis roots, AtVI2 is sorted into ER bodies (Rojo et al. 2003) before it is finally transported to the LV (Xiang et al. 2013). It can be concluded that an array of different sorting possibilities provides plants with an enormous flexibility to adapt themselves under environmental changes (Xiang et al. 2013, and references therein). This further confirms the general notion that sorting processes and their speed differ at the organ, tissue or cellular level and can depend on changing environmental conditions (Rose and Lee 2010). As explained above, YXXΦ-like motifs often play an important role in sorting processes. For instance, BP80 is recognized by AP1 through the (Nterm) YMPL (Cterm) motif (Fig. 9) in the C-terminal part of BP80 (Fig. 9). Would such a motif in the N-terminal part of VIs be directly recognized by AP1? Since AtVI2 is topologically in the reverse orientation as compared with BP80 (Fig. 9; von Heijne 1992), (Nterm) ΦXXY (Cterm)-like motifs would be candidates. However, VIs do not harbor such motifs in their N-terminal part (Table 1). As expected in this view, it was demonstrated here that the (Nterm) YTRL (Cterm) motif in AtVI2 (Table 2) did not affect the sorting process (NAtVI2m17, Fig. 7F, Table 2) since it has the wrong orientation. Instead, it was found that any change in the dileucine core motif 3SSDALLPIS11 markedly affected the sorting process (Table 2, Fig. 7A–E). Similar to AtVI1 and AtVI2, the C-terminal part of the tonoplastic inositol transporter AtINT1 (Wolfenstetter et al. 2012) and the N-terminal part of the tonoplastic glucose transporter AtESL1 (Yamada et al. 2010) also contain such dileucine-type motifs that fulfill a crucial role during sorting (Yamada et al. 2010, Wolfenstetter et al. 2012). Next to the essential dileucine core, leading serines and the acidic amino acid D5 are also often observed in mammalian-derived dileucine-based signals for lysosomal targeting (Bonifacino and Traub 2003, Storch et al. 2004). We conclude that key features for correct sorting of AtVI2 are embedded at the N-terminus, and require (i) a dileucine-like motif; (ii) an intact N-terminal BR motif adjacent to the TMD; (iii) a TMD with a certain length and hydrophobicity, and, most probably, (iv) an as yet unidentified motif to be specifically recognized by vacuolar proteases, activating VIs when entering the CV. Fig. 9 View largeDownload slide Putative interactions during AP1- and AP3-mediated sorting: a comparison of aleurain and AtVI2. (A) BP80- and clathrin-dependent sorting of proaleurain through the classical pathway. The N-terminal part of the type I transmembrane receptor BP80/VSR1 (green) recognizes the vacuolar sorting determinant (VSD) NPIRL in the C-terminal part of proaleurain (red), and takes it as cargo. The C-terminal part of BP80 interacts with AP1 (purple) via the YMPL motif. Subsequently, AP1 binds to clathrin-coated proteins. The KYRIR basic region, probably assisting in correct insertion of BP80 in the membrane, is also indicated. (B) BP80-independent sorting of proAtVI2 through the classical pathway. The proAtVI2 (yellow) inserts as a type II transmembrane protein. Its basic region PRRRRP, probably assisting in correct insertion in the membrane, is also indicated. Perhaps a VSD (the SSDALLPIS motif is a likely candidate) in its N-terminal part is directly recognized by AP1. (C) The alternative, AP3-mediated sorting pathway. A VSD of AtVI2 is possibly recognized by AP3 (pink), in a clathrin-independent pathway. Fig. 9 View largeDownload slide Putative interactions during AP1- and AP3-mediated sorting: a comparison of aleurain and AtVI2. (A) BP80- and clathrin-dependent sorting of proaleurain through the classical pathway. The N-terminal part of the type I transmembrane receptor BP80/VSR1 (green) recognizes the vacuolar sorting determinant (VSD) NPIRL in the C-terminal part of proaleurain (red), and takes it as cargo. The C-terminal part of BP80 interacts with AP1 (purple) via the YMPL motif. Subsequently, AP1 binds to clathrin-coated proteins. The KYRIR basic region, probably assisting in correct insertion of BP80 in the membrane, is also indicated. (B) BP80-independent sorting of proAtVI2 through the classical pathway. The proAtVI2 (yellow) inserts as a type II transmembrane protein. Its basic region PRRRRP, probably assisting in correct insertion in the membrane, is also indicated. Perhaps a VSD (the SSDALLPIS motif is a likely candidate) in its N-terminal part is directly recognized by AP1. (C) The alternative, AP3-mediated sorting pathway. A VSD of AtVI2 is possibly recognized by AP3 (pink), in a clathrin-independent pathway. Fig. 9 represents an overview of the possible interactions that may occur. Proaleurain (Fig. 9A) is taken up as cargo by BP80, and follows clathrin-dependent sorting. The N-terminal part of BP80 binds the NPIRL motif of proaleurain, while the C-terminal part of BP80 interacts with AP1 through the YMPL motif. Subsequently, AP1 binds to clathrin-coated proteins. In contrast, AtVI2 travels in a BP80-independent fashion (Fig. 9B) via the classical pathway. Probably, the SSDALPIS motif in its N-terminus is directly recognized by AP1, but this needs to be further corroborated. Finally, we speculate that AtVI2 might also travel via an alternative, AP3-dependent pathway, although the nature of the VSD putatively interacting with AP3 remains to be determined. In general, it can be concluded that more in-depth research is necessary to confirm the proposed interactions further. However, fewer players and interactions are needed during the membrane-anchored traveling by VIs (Fig. 9B, C) as compared with the BP80-dependent system (Fig. 9A). Therefore, it can be speculated that the membrane-anchored traveling mechanism of VIs evolved as a relatively simple and energetically favorable mechanism to carry VIs to the CV. AtVI2 was recently proposed as a crucial player in biomass production (Nägele et al. 2010). Taking a central position in cellular networks (Nägele et al. 2010), plant VIs as well as other Suc-splitting enzymes are emerging as important biotechnological targets to increase plant biomass production. Increasing VI activities at the whole-plant level might be a useful strategy to stimulate cell expansion and/or counteract stresses, which is especially important at the seedling stage. Thanks to deeper insights into structure–function relationships in the active sites of these enzymes, the creation of superior VIs with a higher substrate affinity becomes a realistic option (Van den Ende et al. 2009, and references therein), but the production of mutated forms that are less susceptible to inhibition by invertase inhibitors or by Hex also holds great promise (Hothorn et al. 2010). Other approaches include suppression of VI degradation (Rojo et al. 2003) and VI inhibitors (Brummell et al. 2011). In the case of post-harvest quality preservation, however, suppression of VI (or overexpression of VI inhibitors) might be the better approach (Bhaskar et al. 2010). The proposed VI sorting mechanisms presented here should be further investigated and confirmed in other species. In the longer term, this knowledge can be exploited, for instance, to speed up or delay the delivery of VIs to the CV where they are subjected to vacuolar post-translational degradation processes. Manipulation of these processes might contribute to delaying leaf senescence and increasing biomass production in crop plants. Materials and Methods Plant material For transient expression experiments, Arabidopsis thaliana plants were grown in soil in 12 h/12 h growth conditions (75 µmol s−1 m−2), 21°C and 50–60% humidity. Knock-out seeds of vsr1/bp80 (At3g52850, SALK_123661 and SALK_150894) and ap3µ (SALK_064486, SALK_127431) were acquired from the Nottingham Arabidopsis Stock Center. Knock-out seeds of ap3β (pat2-1 and pat2-2) were kindly donated by Dr. J. Friml. Amplification of target clones Total RNA was extracted from a maximum of 100 mg of 30-day-old Arabidopsis leaves by using an Rneasy Plant Minikit (Qiagen). Reverse transcription–PCR (RT–PCR) was carried out according to the Access RT-PCR System kit (Promega). Transient expression of GFP fusion proteins in Arabidopsis Full-length (AtVI1, AtVI2) and NTPP (NAtVI1, NAtVI2) versions were subsequently ligated in-frame with the GFP tag into the expression vector HBT35S::GFP::NOSter (Sheen 1993). NAtVI1 consists of 103 amino acids (MASTE … NTILS), and NAtVI2 consists of 118 amino acids (MASSD … NSMLS). Similarly, NAtVI2–mRFP was cloned into vector pBI221 (donated by Professor Ikuko Hara-Nishimura). The constructs were transformed into Arabidopsis protoplasts as described (Yoo et al. 2007). The protoplasts were examined after different incubation times in the dark. p24–RFP was used as an ER marker (kindly donated by Professor Peter Pimple), Kam1ΔC–mRFP was used as a Golgi marker (kindly donated by Professor Ikuko Hara-Nishimura), and Aleu–GFP and GFP–Chi are used as markers of LVs and PSVs (kindly donated by Dr. Gian Pietro Di Sansebastiano). PVC markers were not used since Jung et al. (2011) already convincingly demonstrated the presence of AtVI2 in isolated PVC membranes. GFP images of transformed protoplasts were captured by confocal microscopy (Olympus). Site-directed mutagenesis NAtVI2 constructs served as template for two oligonucleotide primers containing the desired mutation (see Table 2). Mutations were generated as described (Schroeven et al. 2008). Funding This study was supported by FWO Vlaanderen [grants to W.V.d.E. and L.X.]. Acknowledgments The authors thank Filip Rolland, Johan Thevelein, Marta Rubia, Rudy Vergauwen, Katrien Le Roy, Willem Lammens, Yi Li, Jing Wen and Ruben Ghillebert for their support. Abbreviations Abbreviations AP adaptor protein BP80 binding protein of 80 kDa BR basic region Cterm C-terminal part CV central vacuole COP coat protein complex ER endoplasmic reticulum GFP green fluorescent protein Hex hexose LV lytic vacuole MVB multivesicular body Nterm N-terminal part NTPP N-terminal propeptide region PSV protein storage vacuole PVC pre-vacuolar compartment RFP red fluorescence protein Suc sucrose TMD transmembrane domain VI vacuolar invertase VSD vacuolar sorting determinant VSR vacuolar sorting receptor WT wild type. References Bassham DC,  Brandizzi F,  Otegui MS,  Sanderfoot AA.  The secretory system of Arabidopsis,  The Arabidopsis Book ,  2008, vol.  65 pg.  1  Bassham DC,  Raikhel NV.  Plant cells are not just green yeast,  Plant Physiol. ,  2000, vol.  122 (pg.  999- 1001) Google Scholar CrossRef Search ADS PubMed  Bhaskar PB,  Wu L,  Busse JS,  Whitty BR,  Hamernik AJ,  Jansky SH, et al.  Suppression of the vacuolar invertase gene prevents cold-induced sweetening in potato,  Plant Physiol. ,  2010, vol.  154 (pg.  939- 948) Google Scholar CrossRef Search ADS PubMed  Bonfig KB,  Berger S,  Fatima T,  Gonzalez MC,  Roitsch T.  Metabolic control of seedling development by invertases,  Funct. Plant Biol. ,  2007, vol.  34 (pg.  508- 516) Google Scholar CrossRef Search ADS   Bonifacino JS,  Traub LM.  Signals for sorting of transmembrane protein to endosomes and lysosomes,  Annu. Rev. Biochem. ,  2003, vol.  72 (pg.  395- 447) Google Scholar CrossRef Search ADS PubMed  Braulke T,  Bonifacino JS.  Sorting of lysosomal proteins,  Biochim. Biophys. Acta ,  2009, vol.  1793 (pg.  605- 614) Google Scholar CrossRef Search ADS PubMed  Brummell DA,  Chen R-KY,  Harris JC,  Zhang HB,  Hamiaux C,  Kralicek AV, et al.  Induction of vacuolar invertase inhibitor mRNA in potato tubers contributes to cold-induced sweetening resistance and includes spliced hybrid mRNA variant,  J. Exp. Bot. ,  2011, vol.  62 (pg.  3519- 3534) Google Scholar CrossRef Search ADS PubMed  Carter C,  Pan S,  Zouhar J,  Avila EL,  Girke T,  Raikhel NV.  The vegetative vacuole proteome of Arabidopsis thaliana reveals predicted and unpredicted proteins,  Plant Cell ,  2004, vol.  16 (pg.  3285- 3303) Google Scholar CrossRef Search ADS PubMed  Chrispeels MJ,  Barrieu F.  Delivery of a secreted soluble protein to the vacuole via a membrane anchor,  Plant Physiol. ,  1999, vol.  120 (pg.  961- 968) Google Scholar CrossRef Search ADS PubMed  Dalchau N,  Baek SJ,  Briggs HM,  Robertson FC,  Dodd AN,  Gardner MJ, et al.  The circadian oscillator gene GIGANTEA mediates a long-term response of the Arabidopsis thaliana circadian clock to sucrose,  Proc. Natl Acad. Sci. USA ,  2011, vol.  108 (pg.  5104- 5109) Google Scholar CrossRef Search ADS   Dell’Angelica EC.  AP-3-dependent trafficking and disease: the first decade,  Curr. Opin. Cell Biol. ,  2009, vol.  21 (pg.  552- 559) Google Scholar CrossRef Search ADS PubMed  Di Sansebastiano GP,  Paris N,  Marc-Martin S,  Neuhaus JM.  Regeneration of a lytic central vacuole and of neutral peripheral vacuoles can be visualized by green fluorescent proteins targeted to either type of vacuole,  Plant Physiol. ,  2001, vol.  126 (pg.  78- 86) Google Scholar CrossRef Search ADS PubMed  Feraru E,  Paciorek T,  Feraru MI,  Zwiewka M,  De Groodt R,  De Rycke R, et al.  The AP-3 beta adaptin mediates the biogenesis and function of lytic vacuoles in Arabidopsis,  Plant Cell ,  2010, vol.  22 (pg.  2812- 2824) Google Scholar CrossRef Search ADS PubMed  Foresti O,  Denecke J.  Intermediate organelles of the secretory pathway: identity and function,  Traffic ,  2008, vol.  9 (pg.  1599- 1612) Google Scholar CrossRef Search ADS PubMed  Happel N,  Honing S,  Neuhaus JM,  Paris N,  Robinson DG,  Holstein SE.  Arabidopsis mu A-adaptin interacts with the tyrosine motif of the vacuolar sorting receptor VSR-PS1,  Plant J. ,  2004, vol.  37 (pg.  678- 693) Google Scholar CrossRef Search ADS PubMed  Hothorn M,  Van den Ende W,  Lammens W,  Rybin V,  Scheffzek K.  Structural insights into the pH-controlled targeting of plant cell-wall invertase by a specific inhibitor protein,  Proc. Natl Acad. Sci. USA ,  2010, vol.  107 (pg.  17427- 17432) Google Scholar CrossRef Search ADS   Hunter PR,  Craddock CP,  Di Benedetto S,  Roberts LM,  Frigerio L.  Fluorescent reporter proteins for the tonoplast and the vacuolar lumen identify a single vacuolar compartment in Arabidopsis cells,  Plant Physiol. ,  2007, vol.  145 (pg.  1371- 1382) Google Scholar CrossRef Search ADS PubMed  Ji X,  Van den Ende W,  Schroeven L,  Clerens S,  Geuten K,  Cheng SH, et al.  The rice genome encodes two vacuolar invertases with fructan exohydrolase activity but lacks the related fructan biosynthesis genes of the Pooideae,  New Phytol. ,  2007, vol.  173 (pg.  50- 62) Google Scholar CrossRef Search ADS PubMed  Ji X,  Van den Ende W,  Van Laere A,  Cheng S,  Bennett J.  Structure, evolution, and expression of the two invertase gene families of rice,  J. Mol. Evol. ,  2005, vol.  60 (pg.  615- 634) Google Scholar CrossRef Search ADS PubMed  Jung C,  Lee GJ,  Jang M,  Lee M,  Lee J,  Kang H, et al.  Identification of sorting motifs of AtβFruct4 for trafficking from the ER to the vacuole through the Golgi and PVC,  Traffic ,  2011, vol.  12 (pg.  1774- 1792) Google Scholar CrossRef Search ADS PubMed  Kim J-Y,  Mahé A,  Brangeon J,  Prioul J-L.  A maize vacuolar invertase, IVR2, is induced by water stress organ/tissue specificity and diurnal modulatin of expression,  Plant Physiol. ,  2000, vol.  124 (pg.  71- 84) Google Scholar CrossRef Search ADS PubMed  Nägele T,  Henkel S,  Hörmiller I,  Sauter T,  Sawodny O,  Ederer M, et al.  Mathematical modeling of the central carbohydrate metabolism in Arabidopsis reveals a substantial regulatory influence of vacuolar invertase on whole plant carbon metabolism,  Plant Physiol. ,  2010, vol.  153 (pg.  260- 272) Google Scholar CrossRef Search ADS PubMed  Neuhaus J-M,  Paris N.  Plant vacuoles: from biogenesis to function,  Plant Cell Monogr. ,  2005, vol.  1 (pg.  63- 82) Nie PX,  Wang XY,  Hu LP,  Zhang HY,  Zhang JX,  Zhang ZX, et al.  The predominance of the apoplasmic phloem-unloading pathway is interrupted by a symplasmic pathway during Chinese jujube fruit development,  Plant Cell Physiol. ,  2010, vol.  51 (pg.  1007- 1008) Google Scholar CrossRef Search ADS PubMed  Niihama M,  Takemoto N,  Hashiguchi Y,  Tasaka M,  Morita MT.  ZIP genes encode proteins involved in membrane trafficking of the TGN-PVC/vacuoles,  Plant Cell Physiol. ,  2009, vol.  50 (pg.  2057- 2068) Google Scholar CrossRef Search ADS PubMed  Paris N,  Neuhaus J-M.  BP-80 as a vacuolar sorting receptor,  Plant Mol. Biol. ,  2002, vol.  50 (pg.  903- 914) Google Scholar CrossRef Search ADS PubMed  Rae AL,  Casu RE,  Perroux JM,  Jackson MA,  Grof CP.  A soluble acid invertase is directed to the vacuole by a signal anchor mechanism,  J. Plant Physiol. ,  2011, vol.  168 (pg.  983- 989) Google Scholar CrossRef Search ADS PubMed  Reinders A,  Panshyshyn JA,  Ward JM.  Regulation of vacuolar invertase by abscisic acid or glucose in leaves and roots from maize plantlets,  J. Biol. Chem. ,  2005, vol.  280 (pg.  1594- 1602) Google Scholar CrossRef Search ADS PubMed  Robinson DG,  Oliviusson P,  Hinz G.  Protein sorting to the storage vacuoles of plants: a critical appraisal,  Traffic ,  2005, vol.  6 (pg.  615- 625) Google Scholar CrossRef Search ADS PubMed  Rogers JC.  Multiple vacuoles in plant cells—response,  Plant Physiol. ,  2008, vol.  146 (pg.  1024- 1025) Google Scholar CrossRef Search ADS PubMed  Rojo E,  Zouhar J,  Carter C,  Kovaleva V,  Raikhel NV.  A unique mechanism for protein processing and degradation in Arabidopsis,  Proc. Natl Acad. Sci. USA ,  2003, vol.  100 (pg.  7389- 7394) Google Scholar CrossRef Search ADS   Rose JKC,  Lee S-J.  Straying off the highway: trafficking of secreted plant proteins and complexity in the plant cell wall proteome,  Plant Physiol. ,  2010, vol.  153 (pg.  433- 436) Google Scholar CrossRef Search ADS PubMed  Ruan YL,  Jin Y,  Yang YJ,  Li GJ,  Boyer JS.  Sugar input, metabolism, and signaling mediated by invertase: roles in development, yield potential, and response to drought and heat,  Mol. Plant ,  2010, vol.  3 (pg.  942- 955) Google Scholar CrossRef Search ADS PubMed  Sanderfoot AA,  Ahmed SU,  Marty-Mazars D,  Rapoport I,  Kirchhausen T,  Marty F, et al.  A putative vacuolar cargo receptor partially colocalizes with AtPEP12p on a prevacuolar compartment in Arabidopsis roots,  Proc. Natl Acad. Sci. USA ,  1998, vol.  95 (pg.  9920- 9925) Google Scholar CrossRef Search ADS   Saint Jean B,  Seveno Carpentier E,  Alcon C,  Neuhaus J,  Paris N.  The cytosolic tail dipeptide Ile-Met of the pea receptor BP80 is required for recycling from the prevacuole and for endocytosis,  Plant Cell ,  2010, vol.  22 (pg.  2825- 2837) Google Scholar CrossRef Search ADS PubMed  Schroeven L,  Lammens W,  Van Laere A,  Van den Ende W.  Transforming wheat vacuolar invertase into a high affinity sucrose:sucrose 1-fructosyltransferase,  New Phytol. ,  2008, vol.  180 (pg.  822- 831) Google Scholar CrossRef Search ADS PubMed  Seaman MNJ,  Marcusson EG,  Cereghino JL,  Emr SD.  Endosome to Golgi retrieval of the vacuolar protein sorting receptor, Vps10p, requires the function of the VPS29, VPS30, and VPS35 gene products,  J. Cell Biol. ,  1997, vol.  137 (pg.  79- 92) Google Scholar CrossRef Search ADS PubMed  Seo PJ,  Ryu J,  Kang SK,  Park CM.  Modulation of sugar metabolism by an INDETERMINATE DOMAIN transcription factor contributes to photoperiodic flowering in Arabidopsis,  Plant J. ,  2011, vol.  65 (pg.  418- 429) Google Scholar CrossRef Search ADS PubMed  Sergeeva LI,  Keurentjes J-JB,  Bentsink L,  Vonk J,  van der Plas L-HW,  Koornneef M, et al.  Vacuolar invertase regulates elongation of Arabidopsis thaliana roots as revealed by QTL and mutant analysis,  Proc. Natl Acad. Sci. USA ,  2006, vol.  103 (pg.  2994- 2999) Google Scholar CrossRef Search ADS   Sheen J.  Protein phosphatase activity is required for light-inducible gene expression in maize,  EMBO J. ,  1993, vol.  12 (pg.  3497- 3505) Google Scholar PubMed  Storch S,  Pohl S,  Braulke T.  A dileucine motif and a cluster of acidic amino acids in the second cytoplasmic domain of the batten disease-related CLN3 protein are required for efficient lysosomal targeting,  J. Biol. Chem. ,  2004, vol.  279 (pg.  53625- 53634) Google Scholar CrossRef Search ADS PubMed  Sturm A.  Invertases: primary structures, functions, and roles in plant development and sucrose partitioning,  Plant Physiol. ,  1999, vol.  121 (pg.  1- 7) Google Scholar CrossRef Search ADS PubMed  Van den Ende W,  Lammens W,  Van Laere A,  Schroeven L,  Le Roy K.  Donor and acceptor substrate selectivity among plant glycoside hydrolase family 32 enzymes,  FEBS J. ,  2009, vol.  276 (pg.  5788- 5798) Google Scholar CrossRef Search ADS PubMed  Vargas WA,  Salerno GL.  The cinderella story of sucrose hydrolysis: alkaline/neutral invertases, from cyanobacteria to unforeseen roles in plant cytosol and organelles,  Plant Sci. ,  2010, vol.  178 (pg.  1- 8) Google Scholar CrossRef Search ADS   von Heijne G.  Membrane protein structure prediction. Hydrophobicity analysis and the positive-inside rule,  J. Mol. Biol. ,  1992, vol.  225 (pg.  487- 494) Google Scholar CrossRef Search ADS PubMed  Wang H,  Rogers JC,  Jiang L.  Plant RMR proteins: unique vacuolar sorting receptors that couple ligand sorting with membrane internalization,  FEBS J. ,  2011, vol.  278 (pg.  59- 68) Google Scholar CrossRef Search ADS PubMed  Wang J,  Cai Y,  Miao Y,  Lam SK,  Jiang L.  Wortmannin induces homotypic fusion of plant prevacuolar compartments,  J. Exp. Bot. ,  2009, vol.  60 (pg.  3075- 3083) Google Scholar CrossRef Search ADS PubMed  Wang L,  Li XR,  Lian H,  Ni DA,  He YK,  Chen XY, et al.  Evidence that high activity of vacuolar invertase is required for cotton fiber and arabidopsisi root elongation through osmotic dependent and independent pathways respectively,  Plant Physiol. ,  2010, vol.  154 (pg.  744- 756) Google Scholar CrossRef Search ADS PubMed  Wolfenstetter S,  Wirsching P,  Dotzauer D,  Schneider S,  Sauer N.  Routes to the tonoplast: the sorting of tonoplast transporters in Arabidopsis mesophyll protoplasts,  Plant Cell ,  2012, vol.  24 (pg.  215- 232) Google Scholar CrossRef Search ADS PubMed  Xiang L,  Le Roy K,  Bolouri-Moghaddam M,  Vanhaecke M,  Lammens W,  Rolland F, et al.  Exploring the neutral invertase-oxidative stress defense connections in Arabidopsis thaliana,  J. Exp. Bot. ,  2011, vol.  62 (pg.  3849- 3862) Google Scholar CrossRef Search ADS PubMed  Xiang L,  Etxeberria E,  Van den Ende W.  Vacuolar protein sorting mechanisms in plants,  FEBS J. ,  2013, vol.  280 (pg.  979- 993) Google Scholar CrossRef Search ADS PubMed  Yamada K,  Osakabe Y,  Mizoi J,  Nakashima K,  Fujita Y,  Shinozaki K, et al.  Functional analysis of an Arabidopsis thaliana abiotic stress-inducible facilitated diffusion transporter for monosaccharides,  J. Biol. Chem. ,  2010, vol.  285 (pg.  1138- 1146) Google Scholar CrossRef Search ADS PubMed  Yoo SD,  Cho YH,  Sheen J.  Arabidopsis mesophyll protoplasts: a versatile cell system for transient gene expression analysis,  Nat. Protoc. ,  2007, vol.  2 (pg.  1565- 1572) Google Scholar CrossRef Search ADS PubMed  Zouhar J,  Rojo E.  Plant vacuoles: where did they come from and where are they heading?,  Curr. Op. Plant Biol. ,  2009, vol.  12 (pg.  677- 684) Google Scholar CrossRef Search ADS   Zwiewka M,  Feraru E,  Möller B,  Hwang I,  Feraru MI,  Kleine-Vehn J, et al.  The AP-3 adaptor complex is required for vacuolar function in Arabidopsis,  Cell Res. ,  2011, vol.  21 (pg.  1711- 1722) Google Scholar CrossRef Search ADS PubMed  © The Author 2013. Published by Oxford University Press on behalf of Japanese Society of Plant Physiologists. All rights reserved. For permissions, please email: journals.permissions@oup.com
TI - Trafficking of Plant Vacuolar Invertases: From a Membrane-Anchored to a Soluble Status. Understanding Sorting Information in Their Complex N-Terminal Motifs
JO - Plant and Cell Physiology
DO - 10.1093/pcp/pct075
DA - 2013-07-08
UR - https://www.deepdyve.com/lp/oxford-university-press/trafficking-of-plant-vacuolar-invertases-from-a-membrane-anchored-to-a-7FtN56QElI
SP - 1263
EP - 1277
VL - 54
IS - 8
DP - DeepDyve
ER -