![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Department of Entomology and Plant Pathology, Oklahoma State University, Stillwater, Oklahoma 74078
Received 14 September 2006/ Accepted 26 November 2006
 |
ABSTRACT |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
INTRODUCTION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Poliovirus (PV), vaccinia virus, tobacco etch virus, cowpea mosaic virus, and brome mosaic virus are also examples of viruses which cause proliferation and invaginations of the ER for replication (8, 9, 12, 20, 41, 45, 48, 54). The membrane invaginations are containers for the viral replicase protecting the replication complexes from cellular degrading enzymes. Brome mosaic virus 1a is responsible for induction of vesicles, while the nature of the structures formed vary with the ratio of 1a:2a proteins (50, 51). When 2a is expressed at low levels, vesicles are induced. However, with increasing 2a concentrations, membranes accumulate in stacks rather than vesicles (51).
The tobacco mosaic virus (TMV) 126-kDa replicase protein induces membraneous bodies to form in the absence of other TMV proteins. The TMV movement protein associates with these bodies and transports them along the microfilament network toward the periphery of the cell and, possibly, across the plasmodesmata (25, 30). Thus, TMV is the first plant virus that has been reported to involve membrane-bound replication complexes in the cell-to-cell transport pathway.
Beyond TMV, many plant viruses involve the endomembrane system in virus intracellular and intercellular movement. Examples of other viruses encoding small hydrophobic movement proteins which associate with the endomembrane system include the carmovirus (p9 and p8 proteins); panicovirus (ORF2 and ORF3 proteins; 6.6 and 14.6 kDa); closterovirus (p6 protein); potex-, carla-, allexi-, fovea-, hordei-, pomo-, and benyviruses (TGBp2 and TGBp3 proteins); and sobemovirus (p4 protein; 15 kDa) (16, 17, 22, 35-37, 42, 44, 55-57, 62, 63). Recent studies of potex- and pomoviruses report that endosomal or ER-related vesicles contribute to virus cell-to-cell movement (18, 23, 38).
Research in our laboratory focuses on the potexvirus potato virus X (PVX). PVX encodes three movement proteins from three overlapping open reading frames, termed the triple gene block (TGB). The TGB is conserved among viruses belonging to the genera Potexvirus, Hordeivirus, Benyvirus, Carlavirus, Allexivirus, Foveavirus, and Pomovirus. These three proteins are named TGBp1, TGBp2, and TGBp3 and are required for virus movement.
The PVX TGBp1 protein induces plasmodesma gating, moves from cell to cell, binds viral RNA, has ATPase activity, and forms inclusion bodies in virus-infected cells (4, 11, 13, 19, 32, 33, 46, 59). PVX TGBp1 is also a suppressor of RNA silencing, and a recent study showed this activity is necessary for virus cell-to-cell movement (6, 58). The PVX TGBp2 and TGBp3 proteins are ER-associated proteins (23, 28, 38, 49, 53, 62). PVX TGBp2 has two transmembrane segments and a central domain that is conserved among TGB-containing viruses (38). Amino acid sequence analyses of the potexvirus TGBp2 proteins identified two transmembrane segments and a central domain containing conserved amino acids (38) (see Fig. 1A). In the same study, two mutations, named m1 and m3 (see Fig. 1A), were introduced into the TGBp2 coding sequence and disrupted the transmembrane domains (38). These mutations inhibited virus cell-to-cell movement, indicating that membrane association of TGBp2 is necessary for virus movement (38). Electron microscopic analysis showed that the PVX TGBp2 protein induces formation of ER-derived vesicles during virus infection (23). While some have suggested these are transport vesicles carrying virus from the site of replication to the plasmodesmata, the role of these vesicles in virus movement has not been characterized (34).
|
In this study, we asked the following: are the granular vesicles induced by TGBp2 specifically needed for virus cell-to-cell movement, or is general remodeling of the ER network important? Here we identified a deletion mutation (m2) in the central domain of the TGBp2 protein which causes TGBp2 to accumulate in enlarged vesicles and in the ER network (see Fig. 1A). The same mutation inhibits virus movement. Further amino acid substitution mutations were used to characterize a segment of the PVX TGBp2 protein which modulates the vesicle phenotype. Single-amino-acid substitution mutations were enough to cause TGBp2 to associate with enlarged vesicles. These data indicate that specific granular vesicles induced by PVX TGBp2 drive virus cell-to-cell movement.
|
MATERIALS AND METHODS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
, and XL10 Gold were used for transformation of all plasmids. J. Hasselof (Medical Research Council Laboratory of Molecular Biology, Cambridge, United Kingdom) (52) provided the pBIN-mGFP5-ER plasmid, used here to transform Agrobacterium tumefaciens strain LBA4404 to prepare transgenic BY-2 cultures. The parental and mutant pPVX-GFP, pPVX-GFP:TGBp2, and pPVX-GFP:TGBp2m2 plasmids contain the PVX genome beside a bacteriophage T7 promoter (5). The plasmids pPVX-GFP and pPVX-GFP:TGBp2 contain the green fluorescent protein (GFP) or GFP-TGBp2 fused genes inserted next to the duplicated coat protein subgenomic promoter, as described previously (23). Nucleotide (nt) positions 5170 and 5423 were deleted within the viral genome in both the pPVX-GFP:TGBp2 and pPVX-GFP:TGBp2m2 plasmids (23). This mutation removes most of the TGBp2 coding sequence within the triple gene block (Fig. 1). The pPVX-GFP:TGBp2m2 plasmid (Fig. 1) is similar to pPVX-GFP:TGBp2 but has 30 nt deleted from the central domain of TGBp2 within the GFP-TGBp2 fused sequences (Fig. 1). The GFP-TGBp2m2 coding sequence was PCR amplified from pRTL2-GFP:TGBp2m2 plasmids using primers containing added ClaI and SalI restriction sites (38). The GFP-TGBp2m2 coding sequence was inserted into the PVX genomic cDNA between ClaI and SalI sites next to the duplicated coat protein subgenomic promoter (23).
The TGBp2 protein has a conserved amino acid sequence: G52G53XY55XD57G58T59K60XI62XY64 (38). Substitution mutations were introduced into pPVX-GFP and pRTL2 plasmids using the Quick Change site-directed mutagenesis kit (Stratagene, La Jolla, CA). Three nucleotides encoding S (AGC) replaced Y55 (TAC); nucleotides encoding A (GCC) or E (GAC) replaced D57 (GAC); nucleotides encoding A (GCC) replaced T59 (ACC); nucleotides encoding A (GCC) replaced K60 (AAA); nucleotides encoding A (GCG) replaced I62 (ATC); and nucleotides encoding S (AGC) replaced Y64 (TAC). Mutagenic primers were extended using the prescribed temperature cycling regime. Sequences of the forward and reverse mutagenic primers are detailed in Table 1.
|
In vitro transcription. In vitro transcription was carried out using the pPVX-GFP, pPVX-GFP:TGBp2, and pPVX-GFP:TGBp2m2 plasmids and the mMESSAGE mMACHINE High Yield capped RNA transcription kit (Ambion, Inc., Austin, TX). Plasmids were linearized using the SpeI enzyme. Transcripts were directly used to inoculate protoplasts or plants as described previously (23).
Inoculation of plants and RT-PCR analysis of mutant viruses. Nicotiana benthamiana plants were used for studying virus cell-to-cell and systemic movement (23). Five microliters of parental and mutant PVX-GFP, PVX-GFP:TGBp2, or PVX-GFP:TGBp2m2 transcripts was inoculated into an N. benthamiana plant dusted with carborundum. A handheld UV lamp was used to monitor virus cell-to-cell and vascular movement. In some plants, mutant viruses spread systemically. Total RNA was isolated from upper leaves of systemically infected plants using Triazol reagent (Invitrogen, Carlsbad, CA) (3). Reverse transcriptase (RT)-PCR was conducted using the Superscript III One-Step PCR system with Platinum Taq (Invitrogen) to amplify the entire TGB coding sequence. The TGB cDNAs were ligated to pGEM-T Easy plasmids (Promega Corp., Madison, WI) and used to transform JM109 cells. Three colonies were selected, and DNA was isolated and sequenced. Sequencing results for all three colonies were identical for each mutant.
Preparation of transgenic BY-2 cells expressing mGFP5-ER. Transgenic BY-2 suspension cells expressing mGFP5-ER were used as a positive control for experiments studying the relationship of the PVX TGBp2 protein with the ER. Tobacco BY-2 suspension cells were transformed using Agrobacterium tumefaciens strain LBA4404 containing pBIN19-mGFP5-ER plasmids as described previously (21). Five milliliters of 3-day-old BY-2 suspension cells and 100 µl of Agrobacterium suspension were mixed and coincubated on petri plates containing BY-2 culture medium (4.3 g/liter Murashige and Skoog salts [Sigma, St. Louis, MO], 30 g/liter sucrose, 256 mg/liter KH2PO4, 100 mg/liter myoinositol, 1 mg/liter thiamine, and 0.2 mg/liter 2,4-dichlorophenoxyacetic acid, pH 5.6) plus 0.8% agar, for 2 days at 28°C in the dark. The tobacco BY-2 suspension cells were washed twice with 20 ml of BY-2 culture medium and then plated onto BY-2 selection medium (BY-2 culture medium plus 0.8% agar, 500 µg/ml carbenicillin, and 300 µg/ml kanamycin) and then maintained for 10 days at 28°C in the dark. The transgenic tobacco BY-2 suspension cells were then transferred to fresh BY-2 selection medium three times weekly. BY-2 suspension cells were examined with Olympus SZH-ILLK stereomicroscopy (Olympus Optical Co., Ltd., Japan) equipped with an excitation mercury lamp and GFP emission filter. Stably transformed BY-2 suspension cells were transferred to 250-ml Erlenmeyer flasks containing 50 ml of liquid BY-2 selection medium (BY-2 culture medium plus 200 µg/ml kanamycin).
Nontransgenic and transgenic BY-2 suspension cells (39) were maintained on a rotary shaker at 120 rpm in a growth chamber at 28°C in the dark. Cultures were transferred each week into 250-ml Erlenmeyer flasks containing 50 ml of either fresh BY-2 culture medium or liquid BY-2 selection medium.
BY-2 protoplast preparation and transfection.
Protoplasts were prepared from 3-day-old BY-2 suspension cells as described previously (15, 23, 43). To transfect protoplasts with infectious transcripts, 2 µl (roughly 30 µg) transcripts and 5 x 105 protoplasts (in 0.5 ml of solution 2) were mixed and transferred to a 0.4-cm-gap cuvette (Bio-Rad Laboratories, Hercules, CA) on ice. To transfect protoplasts with plasmids, 5 µg plasmids, 40 µg sonicated salmon sperm DNA, and 1 x 106 protoplasts (in 0.5 ml solution 2) were mixed and transferred to a 0.4-cm-gap cuvette on ice. Protoplasts were electroporated using a Gene Pulser (Bio-Rad Laboratories, Hercules, CA) at 0.25 kV, 100 
, and 125 µF (23).
After electroporation, protoplasts were immediately transferred into a new tube containing 1 ml solution 2 and incubated on ice for 30 min and then at room temperature for 5 min. Protoplasts were collected by centrifugation at 79 x g for 5 min, resuspended in 1.5 ml of BY-2 culture medium plus 0.45 M mannitol, and transferred to six-well cell culture plates (Corning, Corning, NY). The bottom of each well in the culture plates was coated with a solution of BY-2 culture medium plus 0.45 M mannitol and 1.0% agarose (pH 5.7). Transfected protoplasts were cultured at 26°C, collected at various times between 18 and 48 h by centrifugation at 79 x g for 5 min, and then examined using laser scanning confocal microscopy (23).
Fluorometric assays and cycloheximide treatment of BY-2 protoplasts. A VICTOR2D fluorometer (Perkin-Elmer, Boston, MA) was used to measure GFP expression in protoplasts transfected with the pRTL2-GFP, -GFP:TGBp2, or -GFP:TGBp2m2 plasmid (23). Samples of 1 x 106 protoplasts were harvested at 18, 24, 30, 36, and 48 h posttransfection and then transferred to liquid nitrogen. To measure protein turnover, the culture medium was removed from transfected protoplasts at 24 h posttransfection and replaced with fresh culture containing 500 µM cycloheximide (Sigma, St. Louis, MO). Samples of 1 x 106 protoplasts were harvested at 0, 4, 8, and 12 h following addition of the cycloheximide (23) and transferred to liquid nitrogen.
One hundred microliters of protein grinding buffer (10 mM Tris-HCl, pH 7.5, 100 mM NaCl, 1 mM MgCl2, and 10 mM dichlorodiphenyltrichloroethane) was added to each sample. Samples were vortexed for 1 min, sonicated for 10 min, frozen at –80°C for 10 min, and then thawed at room temperature. Sample vortexing, sonication, freezing, and thawing were repeated. Following centrifugation at 3,000 x g for 10 min, the supernatants of untreated and cycloheximide-treated extracts were used for fluorometric analyses. The average values from three samples at each time point were plotted using Microsoft Office Excel 2003 software (Microsoft Corp., Redmond, WA) (23).
Transient assays with tobacco leaves. N. benthamiana leaves were bombarded with pRTL2-GFP:TGBp2 and -GFP:TGBp2m2 transcripts using the PDS1000 He system (Bio-Rad) as described previously (30). Leaf segments were treated with FM4-64 dye and analyzed using confocal microscopy (19, 30).
Microscopy. A Leica TCS SP2 (Leica Microsystems, Bannockburn, IL) confocal laser scanning microscope was used to examine GFP, 4',6'-diamidino-2-phenylindole, and FM4-64 fluorescence (Molecular Probes Inc., Eugene, OR). The Leica TCS SP2 imaging system is attached to a Leica DMRE upright microscope with UV and Ar/Kr lasers. Images were compiled into figures using Adobe Photoshop CS software (Adobe Systems, Inc., San Jose, CA) (28, 29).
Statistical analysis. The SAS PROC CORR procedure using SAS 9.1 version (SAS Institute, Cary, NC) was used to test correlations between the percentages of protoplasts showing fluorescence in granular vesicles, the ER network, and enlarged vesicles (see Fig. 5). Tests compared granular vesicles and the ER network, enlarged vesicles and the ER network, or granular and enlarged vesicles for all mutant pRTL2-GFP:TGBp2 plasmids.
|
|
RESULTS |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
In PVX-GFP-inoculated leaves, fluorescence occurred in the cytoplasm and nucleus in all infected cells (Fig. 2A and B). Perinuclear inclusion bodies were also observed in PVX-GFP-infected cells (Fig. 2B and C). These are likely X-bodies, described in the early literature as masses containing virus particles, ER, and ribosomes (2, 26, 27). In previous reports describing the formation of X-bodies, they were often referred to as centers for virus replication, translation, and encapsidation (2, 14, 26, 27). The pattern of fluorescence seen in PVX-GFP-infected cells was similar at the center and leading edge of infection.
|
In a related study, we compared the accumulation of GFP-TGBp2 and GFP-TGBp3 in plasmid-transfected and transgenic tobacco leaves using confocal and electron microscopy (23). GFP-TGBp2 and GFP-TGBp3 were both reported in the ER; however, the granular vesicles were seen only in GFP-TGBp2-expressing samples (23). These granular vesicles contained ribosomes, and immunolabeling detected BiP (an ER resident protein) in these vesicles (23). We also reported that these granular vesicles accumulate in tobacco leaves infected with PVX-GFP:TGBp2. Thus, the granules reported in Fig. 2E and F are the same granular vesicles induced by TGBp2 that were described in a previous study (23).
The m2 mutation is a deletion mutation that eliminates 10 amino acids in the central domain of the TGBp2 protein (Fig. 1A) (38). This mutation overlaps a segment of highly conserved amino acids. While previous experiments showed that the m2 mutation had no effect on ER association of the TGBp2 protein (38), its impact on vesicle formation was unexamined. We introduced the m2 mutation into the PVX-GFP:TGBp2 infectious clone to study the mutation's effects on virus cell-to-cell movement. Plants inoculated with PVX-GFP:TGBp2m2 were monitored for 3 weeks, and fluorescence was restricted to single cells (Fig. 2E; also Table 2). Fluorescence in these single cells was mainly in the ER network and in bright fluorescent bodies (Fig. 2F, G, and H). The small granular vesicles were rare. When we analyzed the bright fluorescent bodies at highest magnification, we discovered they were comprised of enlarged vesicles (Fig. 2H). The X-bodies seen in PVX-GFP- or PVX-GFP:TGBp2-infected cells were also "bright fluorescent bodies," However, at the highest resolution these resembled amorphous masses. We never observe enlarged vesicles in the X-bodies. Thus, these data suggest that the enlarged vesicles seen in plants inoculated with PVX-GFP:TGBp2m2 transcripts result from the m2 mutation.
The diameters of 50 granular and 50 enlarged vesicles were measured. The average diameter of the granular vesicles was 0.5 µm ± 0.1 µm, and that of the enlarged vesicles was 1.4 µm ± 0.5 µm. The dimensions of the granular vesicles measured in this study were similar to the dimensions measured previously using electron micrographs (23).
Subcellular localization of fluorescent proteins during virus infection in protoplasts. To view early events in virus infection, protoplasts were transfected with PVX-GFP, PVX-GFP:TGBp2, or PVX-GFP:TGBp2m2 transcripts and then analyzed at 18, 24, 36, and 48 h postinoculation (hpi) using confocal microscopy. Protoplasts expressing mGFP5-ER were included for comparison (Fig. 3A).
|
Subcellular targeting of wild-type and mutant GFP-TGBp2 proteins in leaves. Tobacco leaves were bombarded with pRTL2-GFP:TGBp2 or pRTL2-GFP:TGBp2m2 plasmids and treated with FM4-64 dye. GFP-TGBp2 was seen in granular vesicles, and GFP-TGBp2m2 accumulated mainly in the ER and enlarged vesicles (Fig. 4A and D). Since previous studies indicated that the GFP-TGBp2-related vesicles were novel structures derived from the ER, we decided to conduct further tests to determine if the granular or enlarged vesicles are related to the plasma membrane or endocytic structures. Red fluorescence due to FM4-64 dye is initially seen in the plasma membrane and in endocytic vesicles budding from the plasma membrane. Following prolonged incubation, endocytic vesicles carry FM4-64 dye into the Golgi and vacuole. Thus, FM4-64 is used to trace the endocytic pathway from the plasma membrane to the Golgi and vacuole (7). FM4-64 dye does not label the ER or nuclear envelope (7).
|
Subcellular targeting of wild-type and mutant GFP-TGBp2 proteins in protoplasts. Five pRTL2 plasmids containing the GFP, GFP-TGBp2, GFP-TGBp2m1, GFP-TGBp2m2, or GFP-TGBp2m3 coding sequence were transfected into BY-2 protoplasts. Prior investigations showed that the m1 and m3 mutations lie in the transmembrane domains, while the m2 mutation lies in the central conserved domain (Fig. 1A) (38). Since m1, m2, and m3 lie in separate subdomains of TGBp2, these experiments were designed to determine if the central conserved domain of TGBp2 is functionally distinct from the transmembrane domains. The subcellular accumulation patterns seen here in protoplasts were also confirmed with bombarded leaves (Fig. 4A) (23, 38).
GFP-TGBp2 fluorescence was mainly associated with granular vesicles at 24 h posttransfection (Fig. 5A). Sometimes we detected fluorescent aggregates in pRTL2-GFP:TGBp2-transfected protoplasts which were not detected in PVX-GFP-TGBp2-infected protoplasts. These aggregates may be aggregates of granular vesicles or artifacts resulting from protein overexpression from a CaMV 35S promoter. GFP-TGBp2m1 and GFP-TGBp2m3 fluorescence accumulated in the cytoplasm and nucleus (Fig. 5B and E). The m1 and m3 mutations, which lie in the transmembrane domains of TGBp2, disrupted ER association of GFP-TGBp2, as reported previously (38). GFP-TGBp2m2 fluorescence associated with enlarged vesicles and the perinuclear and cortical ER (Fig. 5C and D) (38). Granular vesicles were rare in GFP-TGBp2m2-expressing cells. This pattern of fluorescence resembled observations with PVX-GFP:TGBp2m2-inoculated plants and protoplasts. Thus, the m2 mutation, disrupting the central conserved domain of TGBp2, caused the protein to accumulate in the ER and in enlarged vesicles.
To determine if the enlarged vesicles were rare or common occurrences in GFP-TGBp2- and GFP-TGBp2m2-expressing cells, the presence of fluorescence in the ER network and granular and enlarged vesicles was quantified (Fig. 6A and B). For comparison, protoplasts expressing GFP only or mGFP5-ER were also quantified (Fig. 6B). Plasmids containing mGFP5-ER encode a version of GFP that has an N-terminal basic chitinase signal peptide and a C-terminal HDEL sequence for ER targeting and retention. One hundred percent protoplasts (30/30) expressing GFP contained fluorescence in the cytosol and nucleus (data not shown). GFP fluorescence was not membrane associated (Fig. 6B). This is contrasted by mGFP5-ER, where 100% (15/15) of protoplasts contained fluorescence in the ER network. Granular and enlarged vesicles were absent from mGFP5-ER-expressing protoplasts (Fig. 6B).
|
Wild-type and mutant GFP-TGBp2 fusion proteins have similar half-lives. Clearly the m1, m2, and m3 mutations altered the subcellular distribution patterns of GFP-TGBp2 fluorescence. The best explanation is that these mutations disrupt targeting sequences in the PVX TGBp2 protein. However, it is also possible that the mutations alter the stability of TGBp2, causing the fusion proteins to be targeted to the cytosol for degradation. Fluorometric assays were used to quantify GFP expression over time as a measure of both protein accumulation and degradation in BY-2 protoplasts (23). Transfected protoplasts were chased at 24 h with cycloheximide to halt protein synthesis. Fluorometric values were measured at 0, 4, 8, and 12 h following cycloheximide treatment, and the protein half-lives were calculated. The fluorometric values were normalized to the measurement at 0 h (this value was set at 100%). The data were plotted, and linear regression was used to calculate protein half-lives. We predicted that if protein turnover was stimulated as the result of the mutations, then fluorescence would decline more rapidly than in pRTL2-GFP- or -GFP:TGBp2-transfected protoplasts.
Fluorescence values measured in protoplasts transfected with pRTL2-GFP were higher than in protoplasts expressing the fusion proteins (Fig. 7A). Between 18 h and 48 h posttransfection, GFP values were three- to fivefold greater than GFP-TGBp2 values. In general, fluorometric values for all proteins seemed to fluctuate across a plateau between 24 and 48 h posttransfection (Fig. 7A) (23). Following addition of cycloheximide, the rate of decrease in GFP fluorescence was greater than the rate of decrease for wild-type and mutant GFP-TGBp2 fluorescence (Fig. 7B). Our calculation determined the half-life of GFP to be 10.2 h. The GFP-TGBp2, GFP-TGBp2m1, GFP-TGBp2m2, and GFP-TGBp2m3 proteins had similar half-lives of 24.5, 23.7, 25.5, and 24.3 h, respectively. The wild-type and mutant GFP-TGBp2 proteins were more stable than the nonfused GFP protein (Fig. 7B). Thus, changes in the subcellular distribution of mutant GFP-TGBp2 proteins were a specific result of the m1, m2, and m3 mutations rather than changes in protein stability.
|
Seven mutations encoding amino acid substitutions were introduced into pRTL2-GFP:TGBp2 plasmids, replacing Tyr55 with Ser; Asp57 with Ala or Glu; Thr59 with Ala; Lys60 with Ala; Ile62 with Ala; and Tyr64 with Ser. Protoplasts were transfected with mutant pRTL2-GFP:TGBp2 plasmids and viewed at 24 h posttransfection. A random set of 30 protoplasts was scored for the presence of fluorescence in the ER, granular vesicles, and enlarged vesicles.
Fluorescence occurred mainly in the ER and enlarged vesicles in protoplasts expressing GFP-TGBp2-Y55S, -D57A, -D57E, -K60A, -I62A, or -Y64S (Fig. 6A and 8). Between 0 and 17% of protoplasts contained a few granular vesicles. The reduction in granular vesicles due to these mutations resembles the reduction due to GFP:TGBp2m2. These data suggest that Tyr55, Asp57, Lys60, Ile62, and Tyr64 modulate the nature of vesicles associated with GFP-TGBp2 fluorescence (Fig. 6A and 8).
|
In comparing GFP-TGBp2 with each of the mutant proteins in Fig. 6A, it seems that deletion and substitution mutations which reduce protein accumulation in granular vesicles also increase accumulation in the ER and enlarged vesicles. Statistical analysis revealed an inverse correlation between granular and enlarged vesicles (r = –0.96; P < 0.0001) and between granular vesicles and the ER (r = –0.66; P = 0.0551). Fluorescence in enlarged vesicles and the ER network positively correlates (r = 0.66; P = 0.0529). Thus, a decline in granular vesicles is responsible for an increase in fluorescence in the ER and enlarged vesicles. Since TGBp2 induces vesicles from the ER network, the statistical data indicate that the mutations either cause TGBp2 to be redirected from granular into enlarged vesicles or alter vesicle morphology, causing enlarged vesicles to accumulate.
Substitution mutations inhibit PVX cell-to-cell movement. Each substitution mutation was introduced into the PVX-GFP infectious clone to examine the effects of the single-amino-acid substitution mutations on virus cell-to-cell movement. Transcripts were prepared and inoculated into tobacco leaves to determine if the mutations had any effect on virus movement. GFP expression was used to monitor the spread of infection until 16 dpi. As mentioned previously, PVX-GFP infection was detected in inoculated leaves as early as 3 dpi and in upper leaves by 6 dpi. PVX-GFP-Y55S, -D57A, -D57E, -T59A, and -K60A were restricted to single cells in inoculated leaves (Table 2). These data indicate that mutations causing TGBp2 to accumulate in the ER and enlarged vesicles also inhibit virus cell-to-cell movement and that granular vesicles are important for virus cell-to-cell movement.
Between 8 and 18% of plants inoculated with PVX-GFP-I62A or -Y64S showed systemic virus accumulation. In these plants, GFP expression was detected in expanding infection foci on the inoculated leaves by 7 dpi and on the upper leaves by 9 dpi. Virus movement was delayed relative to that of PVX-GFP. For all plants that showed systemic virus accumulation, RNA was isolated and RT-PCR was conducted to amplify the TGB coding region. PCR products were sequenced, and the systemically accumulating virus in plants inoculated with PVX-GFP-I62A or -Y64S did not contain the same mutation. Thus, virus movement was seen only when the mutations reverted to the wild-type sequence. The same mutations causing GFP-TGBp2 to accumulate in enlarged vesicles also inhibited virus cell-to-cell movement (Table 2).
The T59A mutation caused GFP-TGBp2 to accumulate in enlarged vesicles and showed a high proportion of protoplasts containing granular vesicles. Since this mutation inhibited PVX cell-to-cell movement (Table 2), Thr59 likely contributes to other TGBp2 activities needed for virus movement, beyond regulating granular vesicle morphology.
|
DISCUSSION |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A study of potato mop top virus (PMTV) showed that TGB proteins associate with two types of vesicles: small granular vesicles and endocytic vesicles. The granular vesicles are described as transport vesicles carrying viral RNA to the plasmodesmata (18). In this model, these granular vesicles fuse with the plasma membrane (18). Prior studies showed that the PVX TGBp2-induced vesicles associate with the microfilament network (23). The PVX TGBp2-induced vesicles might traffic along microfilaments toward the plasmodesmata. For PMTV, evidence indicates that TGBp2 and TGBp3 proteins are recaptured from the plasma membrane by endocytic vesicles, which carry them back to the site of virus replication (18). While experiments using FM4-64 in this study did not detect PVX TGBp2 associating with the endosome, the results of mutational analysis indicate that the PVX TGBp2-induced granular vesicles are specifically required for virus cell-to-cell movement. Eight mutations were introduced into the central region of the PVX TGBp2 protein. One mutation deleted 10 amino acids, while 7 other mutations were substitutions replacing individual residues. Granular vesicles were absent from cells expressing the mutant GFP-TGBp2 proteins. Table 2 shows that these mutations also inhibited virus movement. These data suggest that PVX, similar to PMTV, requires these granular vesicles for virus movement.
TGBp2 has three domains: N-terminal and C-terminal transmembrane segments and a central sequence which lies in the ER lumen (61). The central domain of TGBp2 has a segment of highly conserved amino acid residues between positions 40 and 64. In this study, we introduced substitution mutations replacing residues between position 55 and 64. These mutations increased GFP-TGBp2 association with the ER and enlarged vesicles while decreasing its association with granular vesicles. These data suggest that Tyr55, Asp57, Lys60, Iso62, and Tyr64 represent a segment of conserved amino acids which determine the nature of vesicles containing GFP-TGBp2 proteins.
One explanation for the enlarged vesicles is that the mutations had altered the subcellular targeting of GFP-TGBp2, causing the proteins to accumulate in the endosome rather than in the granular vesicles. To test this idea, cells expressing the mutant GFP-TGBp2 proteins were treated with FM4-64. The results indicate that the enlarged vesicles were not related to the endosome. Further evidence that the enlarged vesicles are not endocytic vesicles was presented in Fig. 3. In PVX-GFP:TGBp2-infected protoplasts, we observed enlarged vesicles associating with the nuclear envelope. In some instances we observed these vesicles tethered to the cortical ER network (Fig. 3I). The protoplast images in Fig. 3G and I suggest that the GFP-TGBp2m2 proteins may cause bubbles to form from the perinuclear and cortical ER which become enlarged vesicles. It is worth speculating that the enlarged vesicles result from defects in vesicle formation triggered by TGBp2. However, we cannot be certain whether the enlarged vesicles are the result of mutations that alter the dimensions of the vesicles, cause a defect in the budding process, or cause TGBp2 to be directed into another type of cellular vesicle. Further analysis is needed to characterize the identity or origin of these enlarged vesicles.
Since the central domain of TGBp2 lies in the interior of the ER, this sequence may interact with ER lumenal factors, such as small GTPases (ARFs), which regulate vesicle formation. If TGBp2-induced vesicles function as transport vesicles, it is reasonable to consider that they may have features resembling plant transport vesicles. For example, formation of COPI, COPII, or post-Golgi vesicles typically involves ARFs, which are involved in the recruitment of coat proteins. These ARFs are crucial for vesicle budding and for fusion with target membranes (24, 40, 60). Perhaps TGBp2 associates with ARFs in the ER lumen to drive vesicle formation. Another possibility is that the central conserved domain of PVX TGBp2 is necessary for oligomerization of TGBp2 along the membrane surface. TGBp2 oligomers may cause deformation of ER membranes, leading to vesicle formation. Perhaps the mutations tested in this study functioned to weaken or inhibit TGBp2 oligomerization and thereby alter vesicle morphology. Further mutational analysis is needed to define the extent of the conserved sequence which may affect conversion of granular vesicles into enlarged vesicles and to determine if this sequence governs TGBp2 interactions with itself or cellular proteins.
Another explanation for the images in Fig. 3 showing enlarged vesicles tethered to the ER is that these vesicles are fusing with the ER. If the granular vesicles transport viral RNA to the cell surface, they could fuse with the cortical ER, releasing their contents into the plasmodesmata for transport into neighboring cells. In this scenario, the mutations may cause a defect in vesicles fusing with the ER, resulting in the enlarged bubbles that appear to be tethered to the ER network. This model requires granular vesicles are successfully produced, traffic to the periphery of the cell, associate with the ER network near plasmodesmata, and get stuck while fusing with the ER. This could explain loss of movement for PVX-GFP-T59A. In protoplasts expressing GFP-TGBp2-T59A, we saw that 83% contained granular vesicles and 37% contained enlarged vesicles. This mutation had little effect on granular vesicles while causing an increase in enlarged vesicles. Loss of virus movement could occur if the mutation causes vesicles to get stuck and form bubbles while fusing to the ER and if vesicle fusion with the ER is important for virus movement. Further research is needed to define the origin and target of granular and enlarged vesicles.
Many RNA viruses cause ER modifications to promote virus replication or maturation (1, 8, 9, 12, 20, 41, 45, 47, 48, 54). These modifications influence protein sorting, secretion, membrane permeability affecting host defense responses, ER stress, and apoptosis (31). Real-time imaging of TMV has shown that the viral MP binds to membrane-bound replication complexes and carries these toward the plasmodesmata for cell-to-cell transport (25, 30). We do not yet know if PVX TGBp2 associates with the PVX replicase or, like TMV, moves replication complexes across the plasmodesmata. While evidence of membrane-bound bodies contributing to plasmodesmata transport has been described recently for TMV and PVX, this study is the first to use mutational analysis to show that virus-induced vesicles are necessary for intercellular trafficking.
|
ACKNOWLEDGMENTS |
|---|
We thank J. Fletcher and S. Marek for use of their electroporation systems. We also thank Terry Colberg for training and assistance with the confocal microscope located at the Oklahoma State University Electron Microscopy Center.
|
FOOTNOTES |
|---|
Published ahead of print on 6 December 2006.
|
REFERENCES |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
