Contribution of the Plasma Membrane and Central Vacuole in the

Plant Cell Physiol. 45(7): 951–957 (2004) JSPP © 2004

Short Communication

Contribution of the Plasma Membrane and Central Vacuole in the Formation of Autolysosomes in Cultured Tobacco Cells Kanako Yano 1, Sumiko Matsui 1, Tomohiro Tsuchiya 2, Masayoshi Maeshima 2, Natsumaro Kutsuna 3, Seiichiro Hasezawa 3 and Yuji Moriyasu 1, 4 1

School of Food and Nutritional Sciences, University of Shizuoka, Shizuoka, Japan Graduate School of Bioagricultural Sciences, Nagoya University, Nagoya, Japan 3 Department of Integrated Biosciences, Graduate School of Frontier Sciences, University of Tokyo, Kashiwa, Japan 2

;

using the endocytosis markers FM4-64 and Lucifer Yellow CH and three resident proteins of the central vacuole (AtVam3p, sporamin, and gamma-VM23), which were tagged with green fluorescent protein (GFP). Tobacco BY-2 cells were subcultured and subjected to sucrose starvation conditions as described previously (Moriyasu and Ohsumi 1996). The GFP(S65T)-AtVam3p/TDNA vector was kindly provided by Dr. M.H. Sato (Kyoto University) (Uemura et al. 2002). Plasmids harboring the 35S promoter-preprosporamin-GFP were constructed as follows. Genomic DNA was isolated from the tubers of sweet potato (Ipomoea batatas, cv. Kokei 14). The DNA region encoding one of the preprosporamins, which has no introns, was amplified by PCR using the isolated DNA as a template and the following primers: sense primer, 5′-cccctcgagatgaaagccctcacactcgcac-3′, which includes the XhoI site and 22 bases from the 5′-end of the coding region in the coding strand of preprosporamin DNA; antisense primer, 5′-gtgatcaaacctaccgttgtgtccatggcatg-3′, includes the NcoI site and 22 bases from the 5′-end of the coding region in the anticoding strand of preprosporamin DNA. The amplified DNA was digested with XhoI and NcoI, and was then ligated into the SalI–NcoI site of a GFP expression vector, 35Somega-sGFP(S65T) plasmid (Niwa et al. 1999), in order to obtain a plasmid expressing preprosporaminGFP fusion protein under the control of the cauliflower mosaic virus 35S promoter. This plasmid was designated pKY1. It was then cleaved with HindIII and EcoRI to obtain a DNA fragment that included the 35S promoter-preprosporamin DNA-synthetic GFP DNA-poly A signal of the nopaline synthase gene. The DNA region from the HindIII to EcoRI sites of the binary vector pSMAB701 (see Niwa et al. 1999) was replaced with this DNA fragment. The resulting plasmid was designated pKY101. Plasmids harboring the 35S promoter-gamma-VM23-GFP were constructed by amplifying the DNA region encoding gamma-VM23 by PCR using pBluescript containing gammaVM23 cDNA (Higuchi et al. 1998) as a template and the following primers: sense primer, 5′-tacctcgagatctcccacaactctccttatc-3′, which includes the XhoI and BglII sites and 18 bases

Autolysosomes accumulate in tobacco cells cultured under sucrose starvation conditions in the presence of a cysteine protease inhibitor. We characterized these plant autolysosomes using fluorescent dyes and green fluorescent protein (GFP). Observation using the endocytosis markers, FM4-64 and Lucifer Yellow CH, suggested that there is a membrane flow from the plasma membrane to autolysosomes. Using these dyes as well as GFP-AtVam3p, sporamin-GFP and gamma-VM23-GFP fusion proteins as markers of the central vacuole, we found transport of components of the central vacuole to autolysosomes. Thus endocytosis and the supply from the central vacuole may contribute to the formation of autolysosomes. Keywords: Autolysosome — Autophagy — Central vacuole — Endocytosis — FM4-64 — Green fluorescent protein. Abbreviation: GFP, green fluorescent protein.

Autophagy is the process by which cellular components are degraded in lysosomes/vacuoles through the rearrangement and/or synthesis of cell membranes. Autophagy can be induced in plant cells cultured under nutrient starvation conditions (as is the case in mammalian and yeast cells), and such cultured cells have been used to analyze the mechanisms of autophagy in plant cells (Aubert et al. 1996, Moriyasu and Hillmer 2000 for review). Lysosome-like structures accumulate in tobacco cells cultured under sucrose starvation conditions and in the presence of a cysteine protease inhibitor (Moriyasu and Ohsumi 1996). Several lines of evidence suggest that these structures correspond to the autolysosomes observed in mammalian autophagy (Moriyasu and Ohsumi 1996). This includes the acidic pH within the lumen and the presence of acid phosphatase and protease as well as partially degraded cytoplasmic components such as mitochondria. However, the mechanism of autolysosome formation has yet to be elucidated in plant cells. In this study, we characterized these plant “autolysosomes” 4

Corresponding author: E-mail, [email protected]; Fax, +81–54–264–5099. 951

952

Autolysosomes in tobacco cells

in the noncoding region of gamma-VM23 cDNA; antisense primer, 5′-tgaccatgggcagtcctccgtaatca-3′, which includes the NcoI site and sequence similar to 18 bases around the stop codon in the anticoding strand of gamma-VM23. Amplified DNA was digested with XhoI and NcoI, and was integrated into the SalI–NcoI site of the 35Somega-sGFP(S65T) plasmid (Niwa et al. 1999). This plasmid was cut with BglII and NotI, and the fragment containing gamma-VM23 cDNA-GFP DNA was integrated into the BglII–NotI site of the binary vector pMAT137 (provided by Ken Matsuoka, RIKEN) in order to obtain pMAT (gamma-VM23)-sGFP. The GFP(S65T)-AtVam3p/T-DNA vector and the plasmids pKY101 and pMAT (gamma-VM23)-sGFP were used for Agrobacterium-mediated transformation of tobacco cells (Matsuoka and Nakamura 1991) in order to prepare cells expressing GFP-AtVam3p, sporamin-GFP and gamma-VM23GFP fusion proteins, respectively. In mammalian cells, the autophagic and endocytic pathways are known to converge at the endosomes (Liou et al. 1997). It was also suggested that a similar process occurs in plant cells (Record and Griffing 1988, Herman and Lamb 1992), but, confirmative evidence appears to be lacking. We thus examined whether autolysosomes in tobacco cells are located on the endocytic pathway using a fluorescent tracer of endocytosis, FM4-64. Tobacco cells at 4 d in culture, i.e., in the logarithmic growth phase, were transferred to sucrose-free Murashige and Skoog culture medium and cultured for 1 d in the presence or absence of 10 µM of the cysteine protease inhibitor E-64c. As previously reported (Moriyasu and Ohsumi 1996), autolysosomes accumulated in cells cultured in the presence of E-64c, but not in control cells cultured without E-64c (data not shown). The plasma membranes of these two types of cells were stained by incubation in culture medium containing 100 µM FM4-64 at 0°C for 30–60 min. At this temperature, endocytosis does not occur (Vida and Emr 1995), and thus the plasma membrane was stained with FM4-64 in both cell types (Fig. 1A; E64, 0d and MeOH, 0d). Cells were washed to remove the excess dye and then cultured in fresh sucrose-free medium without the dye at 26°C for 1 d. FM4-64 fluorescence moved to the membranes of the central vacuoles in the control cells in which autolysosomes had not accumulated (Fig. 1A; MeOH, 1d). In contrast, in the cells in which autolysosomes had accumulated, the fluorescence moved from the plasma membrane to autolysosomes as well as to the membranes of the central vacuoles, with the majority of fluorescence located to the autolysosomes leaving weaker fluorescence on vacuolar membranes (Fig. 1A; E64, 1d). In this pulse-labeling experiment, it was difficult to maintain the time resolution of observations because we were unable to completely block the process of endocytosis during the washing step. We thus omitted the washing step and simply increased the incubation temperature to 26°C after pre-labeling at 0°C for 30 min. After 30 min at 26°C, numerous dotted structures with a strong FM4-64 fluorescence appeared in the

peripheral cytoplasm in both types of cells (Fig. 1B; arrowheads). These punctated structures have been reported to be endosomes (Ueda et al. 2001). Most of these putative endosomes emerged far away from the perinuclear region where autolysosomes were accumulating. While these putative endosomes possessed a strong fluorescence, the autolysosomes did not show fluorescence with comparable intensity (Fig. 1B; compare fluorescence and Nomarski images in E64, 30 min). Although we occasionally observed dotted structures that possessed FM4-64 fluorescence in the perinuclear region, the fluorescence images of the E-64c-treated cells were similar to those of control cells (Fig. 1B; compare E64, 30min and MeOH, 30min). These results suggest that autolysosomes exist downstream of endosomes on the endocytic pathway from the plasma membrane to the vacuolar membrane. The presence of autolysosomes on the pathway of endocytosis can also be demonstrated using another fluorescent marker, Lucifer Yellow CH, which is incorporated into the vacuoles in some plant cells not by endocytosis but through anion transporters. In our experiment, however, pretreatment of tobacco cells with 10 µM wortmannin for 2 h prevented the uptake of Lucifer Yellow into the central vacuoles (data not shown). This confirms that most of the Lucifer Yellow is taken up into the central vacuoles by endocytosis under the present experimental conditions. The same types of cells as used in the FM4-64 experiment were incubated in sucrose-free culture medium containing 2 mg ml–1 Lucifer Yellow (Fig. 1C). At 2 h, faint fluorescence was observed in the central vacuole (Fig. 1C; MeOH, 2h). In the E-64c-treated cells, autolysosomes were observed to have a stronger fluorescence in addition to the fluorescence in the central vacuole (Fig. 1C; E64, 2h). At 4 h, both the central vacuoles and autolysosomes accumulated Lucifer Yellow at higher concentrations (Fig. 1C; MeOH, 4h and E64, 4h). However, the intensity of fluorescence in autolysosomes seemed to be saturated before that of the central vacuole. This supports the observations obtained using FM4-64 and shows that autolysosomes are present upstream from the central vacuole on the endocytic pathway. Autolysosomes have also been reported to fuse with the central vacuole and pass undegraded cytoplasmic components to the central vacuole in E-64c-treated cells (Moriyasu and Ohsumi 1996). All these data support the notion that plant autolysosomes exist on the endocytic pathway. In the course of our experiments on endocytosis using FM4-64, we noticed the possibility that there is a flow of membrane materials from the vacuole to autolysosomes. To confirm this possibility, we stained the plasma membrane of 3-day-old cells and then cultured these cells for 1 d under nutrient-sufficient conditions. This resulted in the preparation of cells whose vacuolar membranes were stained with FM4-64, but whose plasma membranes were not (Fig. 2A; 0d). When these cells were cultured under sucrose starvation conditions in the presence of E-64c, autolysosomes that were stained with FM4-64 accumulated in the cells (Fig. 2A; E64, 1d). Concomitant with


953

Fig. 1 The localization of autolysosomes in cultured tobacco cells on the endocytic pathway by the use of the fluorescent dyes FM4-64 and Lucifer Yellow CH. (A) Tobacco cells (4-day-old) were cultured under sucrose starvation conditions for 1 d in the presence of 10 µM E-64c (E64, 0d and E64, 1d) or in the presence of 1% (v/v) methanol as a solvent control (MeOH, 0d and MeOH, 1d). The plasma membranes of these two kinds of cells were pulse-labeled with 100 µM FM4-64. The cells were immediately observed (E64, 0d and MeOH, 0d), or cultured for another 1 d and then observed (E64, 1d and MeOH, 1d) using a confocal laser microscope (MRC-1024, Bio-Rad) to obtain the image of FM4-64 fluorescence (red). Arrows indicate the accumulation of autolysosomes. n denotes the location of the nucleus. Bar represents 20 µm. (B) Two kinds of cells, prepared in an identical manner as A, were incubated in sucrosefree culture medium containing 100 µM FM4-64 at 0°C for 30 min and then for 30 min at 26°C. These cells (E64, 30min and MeOH, 30 min) were observed using a confocal laser microscope (LSM510, Zeiss) to obtain FM4-64 fluorescence (red, upper) and Nomarski (lower) images. Arrows indicate the accumulation of autolysosomes. Arrowheads indicate putative endosomes. n denotes the location of the nucleus. Bar represents 20 µm. (C) Two kinds of cells, prepared in an identical manner as A, were incubated in sucrose-free culture medium containing 2 mg ml–1 Lucifer Yellow CH. At 2 and 4 h, the cells were observed using a conventional epifluorescence microscope fitted with Nomarski optics (OptiPhoto, Nikon). For each treatment indicated in the figure, the images of Lucifer Yellow fluorescence are shown on the left; the Nomarski images on the right. Arrows indicate the accumulation of autolysosomes. n denotes the location of the nucleus. Bar represents 20 µm.

staining of the autolysosomal membranes was a drastic decrease in FM4-64 fluorescence from vacuolar membranes (Fig. 2A; E64, 1d). In contrast, when cells with FM4-64labelled vacuolar membranes were subjected to sucrose starvation in the absence of E-64c, movement of FM4-64 was not observed, although a slight decrease in the intensity of fluorescence on the vacuolar membrane was apparent (Fig. 2A; MeOH, 1d). This suggests that a net membrane flow occurs from the central vacuole to the autolysosomes that accumulate in the presence of E-64c. Because the vacuolar membrane remains at the original position even after a drastic decrease in FM4-64 fluorescence, some membrane material must be supplied to the vacuolar membrane compensating for the net transfer of vacuolar mem-

brane to the autolysosomes. We are not certain how such compensation occurs, but it is known that vacuoles have complicated intravacuolar membrane structures (Saito et al. 2002, Uemura et al. 2002). We suppose that these structures function as a membrane reservoir and compensate for the outflow of vacuolar membranes in this case. The flow of membrane from the central vacuole to the autolysosomes also occurs in the absence of E-64c. Hence, we believe that the autolysosomal membrane returns to the central vacuole during the process of autophagy in the absence of E-64c. In the presence of E-64c, autolysosomes cannot accomplish the degradation of enclosed cytoplasm and so may be unable to return to the vacuolar membrane resulting in abnormal accumulation of autolysosomes.

954


Fig. 2 Transport of the fluorescent dyes FM4-64 and Lucifer Yellow from the central vacuole to autolysosomes. (A) The plasma membrane of 3-day-old tobacco cells was pulse-labeled with 100 µM FM4-64 for 30 min at 0°C. The cells were further cultured at 26°C for 1 d (0d). These cells were transferred to sucrose-free culture medium and cultured for another 1 d in the presence of 10 µM E-64c (E64, 1d), or in the presence of 1% (v/v) methanol as a solvent control (MeOH, 1d). A confocal laser microscope (MRC-1024, Bio-Rad) was used to obtain the images of FM4-64 fluorescence (red). Arrows indicate the accumulation of autolysosomes. n denotes the location of the nucleus. Bar represents 20 µm. (B) Tobacco cells (3-day-old) were incubated in nutrient-sufficient medium containing 2 mg ml–1 Lucifer Yellow CH for 4 h. The dye was then washed out of the culture medium, and the cells were transferred to sucrose-free culture medium (0d). The cells were then cultured for another 1 d in the presence of 10 µM E-64c (E64, 1d), or in the presence of 1% (w/v) methanol as a solvent control (MeOH, 1d). A conventional epifluorescence microscope equipped with Nomarski optics was used to obtain fluorescence (upper) and Nomarski (lower) images. n denotes the location of the nucleus. Arrows indicate the accumulation of autolysosomes. Bar represents 20 µm.

Like the membrane marker FM4-64, the lumenal marker Lucifer Yellow CH also moved from the central vacuole to the autolysosomes. Tobacco cells were incubated in Lucifer Yellow for 4 h to prepare cells containing the dye in the central vacuoles (Fig. 2B, 0d). When such cells were kept in sucrosefree culture medium in the presence of E-64c, Lucifer Yellow accumulated in the newly formed autolysosomes (Fig. 2B; E64, 1d). The intensity of fluorescence in the autolysosomes was higher than that in the central vacuole. This may be due to the adsorption of Lucifer Yellow by inclusions in autolysosomes.

Lucifer Yellow fluorescence was not altered significantly when accumulation of autolysosomes did not occur in the absence of E-64c (Fig. 2B; MeOH, 1d). Further evidence for the membrane flow from vacuolar membranes to autolysosomes was obtained using transgenic tobacco cells expressing a GFP-AtVam3p fusion protein (Kutsuna and Hasezawa 2002). AtVam3p belongs to the syntaxin family of Arabidopsis and is localized in the vacuolar membrane (Sato et al. 1997). This fusion protein has been shown to be useful for observing the detailed structure of vacu-


955

Fig. 3 Transport of vacuolar resident proteins from the central vacuole to autolysosomes. (A) Four-day-old tobacco cells expressing a fusion protein of AtVam3 and GFP (0d) were cultured under sucrose starvation conditions for 1 d in the presence of 10 µM E-64c (E64, 1d) or in the presence of 1% (v/v) methanol as a solvent control (MeOH, 1d). A confocal laser microscope (LSM510, Zeiss) was used to obtain GFP fluorescence (green, left) and Nomarski (right) images. Arrows indicate the accumulation of autolysosomes. n denotes the location of the nucleus. Bar represents 20 µm. (B) Four-day-old tobacco cells expressing a fusion protein of sporamin and GFP (0d) were cultured under sucrose starvation conditions for 1 d in the presence of 10 µM E-64c resulting in the accumulation of autolysosomes (E64, 1d) or in the presence of 1% (v/v) methanol as a solvent control (MeOH, 1d). A confocal laser microscope (LSM510, Zeiss) was used to obtain the images of GFP fluorescence (green). Arrows indicate the accumulation of autolysosomes. Arrowheads indicate the Golgi apparatus and/or ER. n denotes the location of the nucleus. Bar represents 20 µm. (C) The plasma membrane of 3day-old tobacco cells expressing a fusion protein of gamma-VM23 and GFP was pulse-labeled with 100 µM FM4-64 for 30 min at 0°C. The cells were then cultured at 26°C for 1 d (0d). These cells were transferred to sucrose-free culture medium and cultured for another 1 d in the presence of 10 µM E-64c (E64, 1d), or in the presence of 1% (w/v) methanol as a solvent control (MeOH, 1d). A confocal laser microscope (MRC1024, Bio-Rad) was used to obtain GFP (gammaGFP; green, left) and FM4-64 (FM4-64; red, middle) fluorescence images. The two fluorescence images are combined (Merge; right). Arrows indicate the accumulation of autolysosomes. n denotes the location of the nucleus. Bar represents 20 µm.

olar membranes in tobacco cells and Arabidopsis plants (Kutsuna et al. 2003, Uemura et al. 2002). In the present study also, the membrane of the central vacuole, including membranes surrounding the transvacuolar strands, could be recognized in 4-day-old cells (Fig. 3A; 0d). The number of transvacuolar strands decreased in both E-64c-treated and methanol-treated control cells under sucrose starvation conditions (Fig. 3A; E64, 1d and MeOH, 1d). This is consistent with our previous finding (Moriyasu and Ohsumi 1996) and shows that the transgenic cells respond to sucrose starvation in a

manner similar to wild-type cells. Interestingly, autolysosomes that accumulated in the cells exhibited a strong GFP fluorescence (Fig. 3A; E64, 1d; arrows). In contrast, the morphology as determined by GFP fluorescence was hardly altered when the cells were cultured in the absence of E-64c (Fig. 3A; MeOH, 1d). Using a sporamin-GFP fusion protein as a vacuolar lumenal marker, we obtained results similar to those shown using the GFP-AtVam3p fusion protein (Fig. 3). Sporamin is a storage protein in tuberous roots of sweet potato (Maeshima et al.

956


1985). It is localized in the vacuolar lumen (Hattori et al. 1988). Its precursor consists of a signal peptide, N-terminal propeptide which contains a vacuolar targeting signal, and the mature protein. When the precursor is expressed in tobacco cells, sporamin targets the central vacuole (Matsuoka et al. 1990). A fusion protein of a polypeptide consisting of a signal peptide and N-terminal propeptide similar to those of sporamin and GFP has recently been shown to target the central vacuole in Arabidopsis (Tamura et al. 2003). In 4-day-old transgenic cells expressing the fusion protein, GFP fluorescence was mainly observed in the lumen of central vacuoles (Fig. 3B, 0d) and also in other organelles such as ER and/or Golgi apparatus (Fig. 3B, arrowheads in all panels). When these transgenic cells were cultured under sucrose starvation conditions in the presence of E-64c, strong GFP fluorescence was observed on autolysosomes in addition to the central vacuole, ER, and Golgi apparatus (Fig. 3B; E64, 1d; arrows). These results suggest that GFP-AtVam3p and sporamin-GFP move from the central vacuole to autolysosomes in accordance with the incorporation of components of the central vacuole into autolysosomes. To examine whether the autolysosomes accumulating the GFP fusion proteins are the same as autolysosomes stained with FM4-64, we used tobacco cells expressing a gammaVM23-GFP fusion protein, another marker for the vacuolar membrane. VM23 is a major hydrophobic membrane protein in vacuoles of radish taproots (Maeshima 1992). The gammaVM23-GFP fusion protein mainly localized to the vacuolar membranes of 4-day-old transgenic cells (Fig. 3C; gammaVM23, 0d). The vacuolar membranes of these cells stained with FM4-64 (Fig. 3C; FM4-64, 0d) showed fluorescence colocalized with GFP fluorescence (Fig. 3C; Merge, 0d), while the plasma membranes and nuclear membranes showed GFP fluorescence but not FM4-64 fluorescence. Sucrose starvation of these cells in the presence of E-64c, resulted in accumulation of autolysosomes with both GFP and FM4-64 fluorescence (Fig. 3C; E64, 1d). This clearly shows that the autolysosomes having gamma-VM23-GFP correspond to the same population of autolysosomes stained with FM4-64. It should be noted that the contribution of de novo synthesis and transport of the GFP fusion proteins from the ER via the Golgi apparatus to autolysosomes cannot be excluded. The fluorescence of sporamin-GFP in the autolysosomes was indeed stronger than that in the central vacuole, suggesting this possibility. However, this is not the only explanation for this phenomenon; the intensity of GFP fluorescence may be affected by several factors such as the turnover rate and quantum yield of sporamin-GFP in each organelle. We were unable to design an experiment where the biosynthetic pathway was blocked by cycloheximide because this reagent inhibits the autophagic pathway in tobacco cells (Takatsuka et al. 2004). All the data obtained in this study strongly suggest that the components of the vacuolar membranes and sap moved to autolysosomes that are newly formed under sucrose starvation

conditions. We interpret these data as showing that the central vacuole contributes to autophagy by supplying components, such as membranes and hydrolytic enzymes, to the autophagosomes/autolysosomes. By acquiring these hydrolytic enzymes from the central vacuole, autophagosomes are believed to transform to autolysosomes. In the autophagic process of mammalian cells, an autophagosome fuses with a pre-existing lysosome to become an autolysosome, in which the degradation of the enclosed cytoplasm is completed. To date, the presence of lysosomes has not been described in mature plant cells, although it has been reported that the central vacuole itself fuses with autophagosomes in autophagy of cultured sycamore cells (Aubert et al. 1996). The present study, however, revealed that a sort of “retrograde” transport from the central vacuole to autophagosomes occurs in the autophagic pathway in tobacco cells and that plant cells possess a lytic compartment, autolysosomes, which is distinct from the central vacuole.

Acknowledgments We thank Dr. Masa H. Sato, Dr. Hiroaki Ichikawa, and Dr. Ken Matsuoka for providing the vectors, GFP(S65T)–AtVam3, pSMAB701, and pMAT137 respectively. We also thank Dr. Alison Hills for critical reading of the manuscript. This work was supported in part by ISAS Grant for Basic Biology Study oriented to utilization of Space station, and “Ground-based Research Announcement for Space Utilization” promoted by Japan Space Forum.

References Aubert, S., Gout, E., Bligny, R., Marty-Marzars, D., Barrieu, F., Alabouvette, J., Marty, F. and Douce, R. (1996) Ultrastructural and biochemical characterization of autophagy in higher plant cells subjected to carbon deprivation: Control by the supply of mitochondria with respiratory substrates. J. Cell Biol. 133: 1251–1263. Hattori, T., Matsuoka, K. and Nakamura, K. (1988) Subcellular localization of the sweet potato tuberous root storage protein. Agric. Biol. Chem. 52: 1057– 1059. Herman, E.M. and Lamb, C.J. (1992) Arabinogalactan-rich glycoproteins are localized on the cell surface and in multivesicular bodies. Plant Physiol. 98: 264–272. Higuchi, T., Suga, S., Tsuchiya, T., Hisada, H., Morishima, S., Okada, Y. and Maeshima, M. (1998) Molecular cloning, water channel activity and tissue specific expression of two isoforms of radish vacuolar aquaporin. Plant Cell Physiol. 39: 905–913. Kutsuna, N. and Hasezawa, S. (2002) Dynamic organization of vacuolar and microtubule structures during cell cycle progression in synchronized tobacco BY-2 cells. Plant Cell Physiol. 43: 965–973. Kutsuna, N., Kumagai, F., Sato, M.H. and Hasezawa, S. (2003) Three-dimensional reconstruction of tubular structure of vacuolar membrane throughout mitosis in living tobacco cells. Plant Cell Physiol. 44: 1045–1054. Liou, W., Geuze, H.J., Geelen, M.J.H. and Slot, J.W. (1997) The autophagic and endocytic pathways converge at the nascent autophagic vacuoles. J. Cell Biol. 136: 61–70. Maeshima, M. (1992) Characterization of the major integral protein of vacuolar membrane. Plant Physiol. 98: 1248–1254. Maeshima, M., Sasaki, T. and Asahi, T. (1985) Characterization of major proteins in sweet potato tuberous roots. Phytochemistry 24: 1899–1902. Matsuoka, K., Matsumoto, S., Hattori, T., Machida, Y. and Nakamura, K. (1990) Vacuolar targeting and posttranslational processing of the precursor to the sweet potato tuberous root storage protein in heterologous plant cells. J. Biol. Chem. 265: 19750–19757.

Autolysosomes in tobacco cells Matsuoka, K. and Nakamura, K. (1991) Propeptide of a precursor to a plant vacuolar protein required for vacuolar targeting. Proc. Natl Acad. Sci. USA 88: 834–838. Moriyasu, Y. and Hillmer, S. (2000) Autophagy and vacuole formation. In: Vacuolar Compartments. Edited by Robinson, D.G. and Rogers, J.C. pp. 71–89. Sheffield Academic Press, Sheffield. Moriyasu, Y. and Ohsumi, Y. (1996) Autophagy in tobacco suspension-cultured cells in response to sucrose starvation. Plant Physiol. 111: 1233–1241. Niwa, Y., Hirano, T., Yoshimoto, K., Shimizu, M. and Kobayashi, H. (1999) Non-invasive quantitative detection and applications of non-toxic, S65T-type green fluorescent protein in living plants. Plant J. 18: 455–463. Record, R. and Griffing, L. (1988) Convergence of the endocytic and lysosomal pathways in soybean protoplasts. Planta 176: 425–432. Saito, C., Ueda, T., Abe, H., Wada, Y., Kuroiwa, T., Hisada, A., Furuya, M. and Nakano, A. (2002) A complex and mobile structure forms a distinct subregion within the continuous vacuolar membrane in young cotyledons of Arabidopsis. Plant J. 29: 245–255. Sato, M.H., Nakamura, N., Ohsumi, Y., Kouchi, H., Kondo, M., HaraNishimura, I., Nishimura, M. and Wada, Y. (1997) The AtVAM3 encodes a

957

syntaxin-related molecule implicated in the vacuolar assembly in Arabidopsis thaliana. J. Biol. Chem. 272: 24530–24535. Takatsuka, C., Inoue Y., Matsuoka, K. and Moriyasu, Y. (2004) 3-Methyladenine inhibits autophagy in tobacco culture cells under sucrose starvation conditions. Plant Cell Physiol. 45: 265–274. Tamura, K., Shimada, T., Ono, E., Tanaka, Y., Nagatani, A., Higashi, S., Watanabe, M., Nishimura, M. and Hara-Nishimura I. (2003) Why green fluorescent fusion proteins have not been observed in the vacuoles of higher plants. Plant J. 35: 545–555. Ueda, T., Yamaguchi, M., Uchimiya, H. and Nakano, A. (2001) Ara6, a plantunique novel type Rab GTPase, functions in the endocytic pathway of Arabidopsis thaliana. EMBO J. 20: 4730–4741. Uemura, T., Yoshimura, S.H., Takeyasu, K. and Sato, M.H. (2002) Vacuolar membrane dynamics revealed by GFP-AtVam3 fusion protein. Genes Cells 7: 743–753. Vida, T.A. and Emr, S.D. (1995) A new vital stain for visualizing vacuolar membrane dynamics and endocytosis in yeast. J. Cell Biol. 128: 779–792.

(Received October 29, 2003; Accepted April 24, 2004)