| Journal of General Virology |
| SUMMARY | INTRO | METHODS | RESULTS | DISCUSSION | FOOTNOTES | REFS |
| First posted online 19 May 2000 | FULL-LENGTH ARTICLE |
| Rec 28 February 2000; Acc 26 April 2000 | DOI: 10.1099/vir.0.17007-0 |
Hiroshi Satoh, Hironori Matsuda, Takehiro Kawamura, Masamichi Isogai, Nobuyuki Yoshikawa and Tsuyoshi Takahashi
Plant Pathology Laboratory, Faculty of
Agriculture, Iwate University, Ueda 3-18-8, Morioka 020-8550, Japan
The 50 kDa protein (50KP) encoded by ORF2 of Apple chlorotic leaf spot virus (ACLSV) fused to green fluorescent protein (GFP) was expressed transiently in cells of Nicotiana occidentalis and Chenopodium quinoa leaves. Its intracellular distribution, cell-to-cell trafficking in leaf epidermis and tubule formation on the surface of protoplasts were analysed. The 50KPGFP fluorescence was distributed as small irregular spots or a fibrous network structure on the periphery of epidermal cells and protoplasts of both plant species. In leaf epidermis of N. occidentalis, the protein spread from the cells that produced it into neighbouring cells in both young and mature leaves and targetted plasmodesmata in these cells. In contrast, GFP was restricted to single cells in most cases in mature leaves. When 50KP and GFP were co-expressed in leaf epidermis of N. occidentalis, GFP spread more widely from the initial cells that produced it than when GFP was expressed alone, suggesting that 50KP facilitated the cell-to-cell trafficking of GFP. 50KPGFP was able to complement local spread of 50KP-deficient virus when expressed transiently in leaf epidermis of C. quinoa. Expression of 50KPGFP in protoplasts resulted in the production of tubular structures protruding from the surface. Mutational analyses showed that the C-terminal region (aa 287475) was not essential for localization to plasmodesmata, cell-to-cell trafficking, complementation of movement of 50KP-deficient virus or tubule formation on protoplasts. In contrast, deletions in the N-terminal region resulted in the complete disruption of all these activities.
Introduction |
Plant viruses move from cell to cell in plant
tissues through plasmodesmata, which connect the cytoplasm of neighbouring
cells (Carrington et al., 1996
; Lucas & Gilbertson, 1994
). Many plant viruses encode movement proteins (MPs) to
assist their spread (Koonin & Dolja, 1993
). At present, there are at least two distinct pathways for
the cell-to-cell movement of plant viruses. One is exemplified best by
Tobacco mosaic virus (TMV), in which the MP interacts with the
plasmodesmata at the infection front, allowing the movement of an
MPvirus RNA complex to neighbouring cells (McLean et al.,
1995
; Oparka et al., 1997
; Padgett et al., 1996
; Reichel et al., 1999
). In another type, represented by Cowpea
mosaic virus (CPMV), virus particles move from cell to cell through
tubular structures protruding from plasmodesmata (van Lent et al.,
1990
, 1991
).
Plant virus MPs are reported to be multifunctional.
Generally, the MP is localized to the plasmodesmata in infected and
transgenic plant cells and induces a significant increase in
plasmodesmatal permeability (Atkins et al., 1991
; Ding et al., 1992
; Tomenius et al., 1987
; Wolf et al., 1989
). The protein binds single-stranded nucleic acids and can
traffic from cell to cell itself (Citovsky et al., 1990
; Fujiwara et al., 1993
; Waigmann & Zambryski, 1995
). Recent studies with MP fused to green
fluorescent protein (GFP) showed that MP is associated with microtubules
and the endoplasmic reticulum (ER), suggesting the involvement of the
cytoskeleton and ER in the intracellular trafficking of MP from the site
of synthesis in the cytoplasm to the plasmodesmata (Heinlein et
al., 1995
, 1998
; McLean et al., 1995
; Huang & Zhang, 1999
). MPGFP fusions of Alfalfa mosaic virus
(AlMV) and Cucumber mosaic virus (CMV) are reported to be capable
of cell-to-cell trafficking in the leaf epidermis (Itaya et al.,
1997
; Huang & Zhang, 1999
).
Apple chlorotic leaf spot virus (ACLSV), the
type species of the genus Trichovirus, has very flexuous
filamentous particles, approximately 600700 nm in length, and
contains a polyadenylated, plus-sense ssRNA with a molecular mass of
2.48x106 Da and a single coat protein of 22 kDa (Yoshikawa
& Takahashi, 1988
). The genome of an apple
isolate of ACLSV (P-209) consists of 7552 nt and contains three open
reading frames (ORFs 1, 2 and 3) (Sato et al., 1993
). The 216 kDa protein (KP) encoded by ORF1 is a
replication-associated protein and a coat protein is encoded by ORF3. The
50 kDa protein (50KP) encoded by ORF2 is thought to be an MP, based on the
following evidence. (i) The amino acid sequence of 50KP has some
similarity to MPs of other plant viruses and the protein was detected in
the cell wall fraction from infected tissues (Sato et al., 1993
, 1995
). (ii) Immunoelectron microscopy with an antiserum against
50KP showed that the protein is localized to plasmodesmata in infected
Chenopodium quinoa cells (Yoshikawa et al., 1999
). (iii) In transgenic plants expressing 50KP
fused to GFP, the fluorescence was associated with plasmodesmata and
accumulated in sieve elements (Yoshikawa et al., 1999
). (iv) Transgenic Nicotiana occidentalis
plants producing 50KP can complement the systemic spread of
movement-defective ACLSV (Yoshikawa et al., 2000
).
In this study, we transiently expressed ACLSV 50KP fused to GFP in leaf epidermal cells and in leaf mesophyll protoplasts of N. occidentalis and C. quinoa and analysed its subcellular distribution, intercellular trafficking in epidermal cells and tubule formation on the surface of protoplasts. The results indicate that 50KPGFP is associated with a network, thought to be cortical ER, on the periphery of epidermal cells and protoplasts. 50KPGFP moved into adjacent cells from the cells that produced it in the leaf epidermis. The protein also induced formation of tubular structures on the surface of protoplasts. Mutational analysis suggested that these activities are related to each other.
Methods |
Construction of plasmids. For construction of
transient-expression vectors, the fragment between the HindIII and
EcoRI sites of pBE2113-GUS (Mitsuhara et al., 1996
) was ligated to the same restriction sites of
pUC18 and the resulting plasmid was designated pUC2113-GUS. Three
plasmids, p35S50KP, p35S50KPGFP and p35SGFP (Fig.
1), were constructed by replacing the GUS gene of pUC2113-GUS with
each gene, as described previously (Yoshikawa et al., 2000
).
The 50KP deletion mutants
A to
G (Fig. 1) were
constructed by the amplification of each DNA fragment by PCR followed by
ligation of each product to a restriction site indicated in Fig. 1. To construct mutants
A,
B,
D and
F, a DNA fragment was amplified by
using the forward primer 50K-Bam (+) (Yoshikawa et al., 2000
) and a reverse primer containing a SacII
(for
A), NruI (
B), NheI (
D) or HindIII (
G) site. The DNA product was
double-digested with BamHI and SacII, NruI,
NheI or HindIII and ligated to p35S50KPGFP
restricted with the same enzymes. For construction of
C,
E and
G, a DNA fragment was amplified by
using an oligonucleotide complementary to nt 67816798 of the viral
genome and containing a SacII site as a reverse primer and a
forward primer containing an NheI (
C), HindIII (
F) or StuI (
G) site. The product was restricted
with SacII and NheI, HindIII or StuI and
ligated to p35S50KPGFP as above. The positions of deleted amino
acids in 50KP are shown in Fig. 1.
Fig. 1. Schematic representation of DNA
constructs expressing 50KP, 50KPGFP and 50KP deletion mutants. El2,
double 5´ upstream sequence of CaMV 35S promoter (419 to
90); P, 5´ upstream sequence of 35S promoter (90 to
1);
, 5´-untranslated
sequence of TMV; Tnos, nopaline synthase terminator. Restriction enzymes
used for construction of deletion mutants are shown on p35S50KPGFP.
Numbers on the boxes indicate the amino acid positions deleted.
Particle bombardment. In transient GFP
expression experiments, leaves were detached from N. occidentalis
(12 true-leaf stage) or C. quinoa (8 true-leaf stage) plants and
placed in a Petri dish containing wet filter paper. The lower epidermis
was bombarded with microparticles coated with DNA constructs by using the
PDS-1000/He particle delivery system (Bio-Rad) as described before (Satoh
et al., 1999
). Leaves were kept under
moist conditions at 25 °C until used for observation.
To examine the complementation of cell-to-cell
movement of 50KP-deficient virus by 50KPGFP, the fifth true leaf of
a plant of C. quinoa (7 true-leaf stage) was bombarded with a
mixture (1:1) of p
StuNhe and
p35S50KPGFP. The plants were maintained in a glass chamber for 5
days. Total RNA was extracted from bombarded leaves and then subjected to
Northern hybridization analysis with an RNA probe, as described previously
(Yoshikawa et al., 2000
).
Isolation and transfection of protoplasts. Leaf mesophyll protoplasts were isolated from C. quinoa and N. occidentalis leaves as follows. Leaves were sliced into strips and then soaked for 3 h in an enzyme solution containing 2 % cellulase 'Onozuka' R-10 (Yakult Pharmaceutical), 0.1 % pectolyase Y-23 (Seishin Pharmaceutical), 10 mM CaCl2 and 0.5 M mannitol, pH 5.6. The resulting protoplast suspension was washed twice with MC solution (0.5 M mannitol, 10 mM CaCl2) and centrifuged in MC solution containing 20 % sucrose for 3 min at 700 r.p.m. Protoplasts were recovered from the middle layer and washed in MC solution.
To the protoplasts (about 3x105 cells), 20 µg plasmid DNA and 500 µl inoculation buffer (10 mM MES, 40 mM CaCl2, 0.5 M mannitol, pH 5.8) were added and then the suspension was mixed gently. Next, 900 µl PEG solution (40 % PEG 4000, 40 mM CaCl2, 0.5 M mannitol) was added, followed by incubation on ice for 30 min. After washing with 50 mM glycine, 50 mM CaCl2, 0.5 M mannitol, pH 8.5, 10 ml inoculation buffer was added to the protoplasts, which were then incubated on ice for 30 min. The protoplasts were then suspended in a medium containing 0.5 M mannitol, 0.2 mM KH2PO4, 1 mM KNO3, 1 mM MgSO4, 10 mM CaCl2, 1 µM KI, 0.01 µM CuSO4, pH 6.5, and incubated at 25 °C.
Fluorescence and confocal laser scanning
microscopy (CLSM). Observation of GFP fluorescence in epidermal cells
and protoplasts was conducted by using a Leica DMLB fluorescence
microscope as described previously (Yoshikawa et al., 1999
). A laser scanning microscope (Leica DMIRB
equipped with Yokogawa CLSM unit CSU10) was also used with excitation at
488 nm and emission at 516700 nm. Digital images were acquired with
a Yokogawa DNS10 CCD camera and processed by IPLab
(Scananalytics).
Results |
Intracellular distribution of 50KPGFP in leaf epidermis
Plasmids p35SGFP and
p35S50KPGFP, which express GFP or 50KPGFP transiently under
the control of the 35S promoter, were constructed, as well as a series of
p35S50KP deletion mutants (
A to
G) (Fig. 1;
Table 1). The plasmids were delivered into
leaf epidermis of N. occidentalis and C. quinoa by particle
bombardment. Fluorescence became visible under a fluorescence microscope
in the epidermal cells of both plant species 2 h after bombardment.
Observation by CLSM showed that the 50KPGFP fluorescence was
observed mainly on the cell periphery and on the cell wall of the
epidermal cells 1224 h after bombardment. This was clear when the
focus was brought to the upper, centre and lower planes of the cells (Fig. 2 A, B). 50KPGFP was present as small,
irregular spots (Fig. 2 B) or as a fibrous,
structure-forming network (Fig. 2 A, C) in the
outermost periphery of the cells. This was in contrast with the cells
expressing GFP only, in which bright fluorescence was found in the nucleus
and cytoplasm (Fig. 2 F). There was no difference
between N. occidentalis and C. quinoa in the
fluorescence-distribution patterns in epidermal cells.
Table 1. Functional activities of proteins expressed transiently in epidermal cells and protoplasts
|
Plasmid |
Protein expressed |
Distribution on cell periphery, localization to plasmodesmata |
Cell-to-cell trafficking |
Complementation of local movement of 50KP-deficient virus |
Tubule formation on surface of protoplasts |
|
p35S50KP |
50KP |
nt |
nt |
+ |
nt |
|
p35S50KPGFP |
50KPGFP |
+ |
+ |
+ |
+ |
|
|
|
+ |
+ |
+ |
+ |
|
|
|
+ |
+ |
+ |
+ |
|
|
|
+ |
+ |
+ |
+ |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
p35SGFP |
GFP |
|
(+) |
|
|
Fig. 2. Detection by CLSM
of fluorescence of 50KPGFP expressed transiently in epidermal cells
and protoplasts. (A) 50KPGFP fluorescence of upper, middle and lower
planes of an epidermal cell, showing fibrous structure on the cell
periphery. (B)(C) 50KPGFP detected as small spots (right in B)
and a fibrous structure forming a network (C) on the periphery of
epidermal cells. (D) Fluorescence of
D, showing large aggregates in an
epidermal cell. (E) Fluorescent spots on the cell wall of an epidermal
cell expressing 50KPGFP. (F)(H) Fluorescence of GFP (F),
50KPGFP (G) and
G (H)
expressed in leaf epidermis of mature leaves following bombardment. (I)
GFP distribution in a protoplast showing strong fluorescence in the
nucleus. (J)(L) 50KPGFP fluorescence showing spots (J) and
fibrous structures (K, L) on the periphery of protoplasts. (M)(N)
GFP fluorescence in leaf epidermis (7th leaves) bombarded with a mixture
of p35SGFP and p35S50KP (M) or p35SGFP alone (N). (O)(Q) Tubular
structures protruding from the surface of protoplasts expressing
50KPGFP (O, P) or
C (Q). (B)
and (C) are images of C. quinoa and the others are of N.
occidentalis. Fluorescence in epidermal cells (AH, M, N) was
photographed 24 h after bombardment. Fluorescence in protoplasts was
photographed 4 (J, K), 18 (I, L) or 24 h (OQ) after bombardment.
Bars represent 10 (AE), 50 (FH), 20 (IL), 100 (M, N) or
20 (OQ) µm.
Targetting to plasmodesmata and cell-to-cell trafficking of 50KPGFP in leaf epidermis
In N. occidentalis leaves 2448 h after
bombardment with p35S50KPGFP, fluorescence was observed on the cell
wall of the cells neighbouring most of the cells that originally produced
50KPGFP (Fig. 2 G), suggesting that
50KPGFP moves from the cells that produce it into the neighbouring
cells. To demonstrate cell-to-cell trafficking of 50KPGFP further,
DNA of the plasmids p35S50KPGFP or p35SGFP was bombarded into each
half of split N. occidentalis leaves and the number of single cells
and cell clusters showing fluorescence was counted 24 h after bombardment.
50KPGFP spread from single cells that produced it into neighbouring
cells in both young and mature leaves (Fig.
3), in contrast to GFP, which was restricted to single cells in
more than 80 % of cases in mature (3rd and 5th) leaves (Figs 2 F and 3). At higher magnification,
fluorescence in cells that adjoined cells initially bombarded with
p35S50KPGFP was detected as spots or strands passing through the
cell wall (Fig. 2 E), indicating that fluorescence
was located at plasmodesmata, as reported for transgenic N.
occidentalis plants expressing 50KPGFP (Yoshikawa et al.,
1999
). Small, irregular spots or thick,
fibrous structures, as described above, were not observed in these
neighbouring cells.
Fig. 3. Cell-to-cell
trafficking of 50KPGFP and GFP in leaf epidermis. The plasmids
p35S50KPGFP and p35SGFP (Fig. 1) were each
bombarded into half-leaves of split N. occidentalis leaves
at different growth stages. The numbers of single fluorescing cells and
clusters of fluorescing cells were counted 24 h after bombardment. The
experiments were repeated two times and the data are the percentages of
the sum of two experiments.
When the p35S50KP deletion mutants (Fig. 1) were bombarded into leaf epidermis of N.
occidentalis plants, the fluorescence derived from
A,
B or
C was observed as spots on the cell
wall of the neighbouring cells, suggesting that these deleted proteins
retained the same cell-to-cell trafficking ability as wild-type
50KPGFP (Table 1). In contrast, the
fluorescence from
D,
E,
F and
G was present as aggregates in the
cytoplasm and was restricted to single cells in both young and mature
leaves (Fig. 2 D, H), indicating that these proteins
had lost their ability to move from cell to cell.
In observations of leaf epidermis of C. quinoa plants bombarded with p35SGFP, non-specific trafficking of GFP was found in young, developing leaves, similar to the situation in N. occidentalis. However, in epidermal cells of both young and mature leaves of C. quinoa bombarded with p35S50KPGFP, fluorescent spots in cell walls were detected in only two or three cells around an originally transfected cell, rather fewer than in the case of N. occidentalis (data not shown).
Increased cell-to-cell trafficking of GFP co-expressed with 50KP in leaf epidermis
Plant virus MPs are reported to modify the plasmodesmata and
to increase their size-exclusion limit (Derrick et al., 1992
; Vaquero et al., 1994
; Poirson et al., 1993
; Wolf et al., 1989
). As mentioned above, GFP was restricted to single cells
in most cases in mature leaves (Fig. 3). In order to
investigate whether GFP can spread from cell to cell in mature leaves when
50KP is co-expressed in the same cells, p35SGFP was bombarded into one
half-leaf and p35S50KP plus p35SGFP were bombarded into the other
half-leaf of N. occidentalis. When GFP was co-expressed with 50KP,
the fluorescence spread more widely from the cells that initially produced
it than when GFP was expressed alone (Table 2;
Fig. 2 M, N). The result suggests that the 50KP
expressed in cells may modify the plasmodesmata and facilitate
cell-to-cell trafficking of GFP that is expressed in the same
cells.
Table 2. Cell-to-cell trafficking of GFP co-expressed with ACLSV 50KP in leaf epidermis of N. occidentalis
|
Cells showing GFP fluorescence (%) |
|||||
|
Cell clusters |
|||||
|
Expressed protein* |
Single cells |
25 cells |
610 cells |
1115 cells |
>15 cells |
|
3rd leaf |
|||||
|
GFP |
98 |
2 |
0 |
0 |
0 |
|
GFP+50KP |
64 |
25 |
10 |
1 |
0 |
|
5th leaf |
|||||
|
GFP |
69 |
31 |
0 |
0 |
0 |
|
GFP+50KP |
22 |
26 |
39 |
13 |
0 |
|
7th leaf |
|||||
|
GFP |
49 |
30 |
19 |
2 |
0 |
|
GFP+50KP |
0 |
4 |
20 |
44 |
32 |
Complementation of local cell-to-cell spread of 50KP-deficient ACLSV with 50KP and 50KPGFP expressed transiently in leaf epidermis
Two 50KP-deficient
mutants (pStuStop and p
StuNhe) of
an infectious ACLSV cDNA clone (pCLSF) (Satoh et al., 1999
) were found to replicate in protoplasts from
C. quinoa leaves (data not shown). These clones were shown
previously to give systemic infection of transgenic N. occidentalis
plants that constitutively express 50KP (Yoshikawa et al., 2000
). To examine whether 50KP expressed transiently
in leaf epidermis can complement the spread of 50KP-deficient virus, p
StuNhe was co-bombarded with p35S50KP
or p35S50KPGFP into leaf epidermis of C. quinoa plants.
Northern hybridization analysis of leaves 3 days after bombardment showed
that progeny viral RNA was detected in leaves when p
StuNhe was co-bombarded with p35S50KP
or p35S50KPGFP (Fig. 4 A, lanes
69). In contrast, no viral RNA was found in samples bombarded with
p
StuNhe only (Fig. 4 A, lanes 35). Compared with the band of progeny
genomic RNA from leaves inoculated with pCLSF (Fig.
4 A, lane 2), the signals for viral RNA from co-bombarded leaves were
weak, suggesting that 50KP-deficient virus may spread to a limited area
around the cells that originally produce 50KP or 50KPGFP. These
results indicate that 50KP and 50KPGFP expressed transiently in
epidermal cells are functional and can complement local spread of
50KP-deficient virus. Similar complementation of 50KP-deficient virus
(p
StuNhe) was found in leaves
co-bombarded with
A,
B or
C (Fig. 4 B,
lanes 13), but not
D,
E,
F or
G (Fig. 4 B,
lanes 47).
Fig. 4. Northern blot
analysis of leaves bombarded with p
StuNhe, p
StuNhe plus p35S50KPGFP or p
StuNhe plus p35S50KP deletion mutants.
(A) A control sample without bombardment (lane 1) and samples from leaves
bombarded with pCLSF (lane 2), p
StuNhe alone (lanes 35), p
StuNhe plus p35S50KP (lanes 6 and 7) or
p
StuNhe plus p35S50KPGFP
(lanes 8 and 9). (B) Samples from leaves bombarded with p
StuNhe plus p35S50KP deletion mutants
A (lane 1),
B (2),
C (3),
D (4),
E (5),
F (6) or
G (7). The arrow and the arrowhead
respectively indicate the positions of ACLSV RNA and RNA derived from
p
StuNhe.
The pattern of accumulation of 50KPGFP and its induction of tubules in protoplasts
Leaf mesophyll protoplasts isolated from N. occidentalis and C. quinoa plants were transfected with p35SGFP, p35S50KPGFP or p35S50KP deletion mutants and the distribution of fluorescence was observed at 4, 12, 24 and 48 h after transfection. In protoplasts transfected with p35SGFP, the fluorescence was found in nuclei and cytoplasm (Fig. 2 I), consistent with its distribution in epidermal cells. The GFP distribution pattern did not change until 48 h after transfection. In contrast, 50KPGFP was present as small, irregular spots (Fig. 2 J) and fibrous structures forming networks (Fig. 2 K, L) at the periphery of protoplasts, as observed in leaf epidermal cells, indicating that fibrous structures were formed without requiring the presence of a cell wall. The time-course of the distribution of 50KPGFP showed that the fluorescence was present as small, irregular spots in 96 % of sampled protoplasts (158/165) and as fibrous structures in 4 % of sampled protoplasts (7/165) 4 h after transfection. The percentage of protoplasts showing fibrous network structures (Fig. 2 K and L) increased to 37 and 39 % (54/145 and 42/107) in two independent experiments 12 h after transfection. The ratio did not change after that. These results suggest that 50KPGFP may accumulate initially as small, irregular spots in the periphery of the protoplast but then the distribution changes to a fibrous network structure.
Tubular structures protruding from the surface of
the protoplasts were also observed in a few protoplasts expressing
50KPGFP (Fig. 2 OQ). The tubules were of
variable lengths and appeared to be very fragile and to fragment easily
during observation. The percentages of protoplasts expressing
50KPGFP with tubules were 0.6 (1/165), 2 (3/145) and 3 % (5/178) at
4, 12 and 24 h after transfection. The percentage did not increase at 48 h
after transfection. The tubules were formed in protoplasts transfected
with mutants
A,
B or
C but not
D,
E,
F or
G (Table 1).
Interestingly, small, irregular spots and fibrous network structures were
never found in protoplasts containing tubular structures.
Discussion |
A TMV MPGFP fusion has been shown to
co-localize with cortical ER and microtubules (Heinlein et al.,
1995
, 1998
; McLean et al., 1995
). It has been reported that AlMV MPGFP expressed in
epidermal cells is also associated with the ER and that the association
may be important for intracellular and intercellular movement of AlMV MP
(Huang & Zhang, 1999
). In order to clarify whether 50KPGFP was
co-localized with the ER and/or microtubules, we conducted several
double-labelling experiments with a monoclonal antibody against tubulin
and rhodamine-labelled secondary antibody (Heinlein et al., 1995
) or using the hexyl ester of rhodamine B (Huang
& Zhang, 1999
). However, we could not
obtain definitive evidence of co-localization of 50KPGFP and ER
and/or microtubules (data not shown).
Cell-to-cell trafficking of MPGFP expressed
transiently in leaf epidermis has been reported in CMV and AlMV, in which
the protein moves into neighbouring cells from an originally transfected
cell and, in contrast, free GFP remains in single cells (Itaya et
al., 1997
; Huang & Zhang,
1999
). Recently, Oparka et al.
(1999
) reported that the capacity of
plasmodesmata to traffic macromolecules depends on the physiological
conditions of the leaves; in sink leaves, proteins of up to 50 kDa could
move freely through plasmodesmata. Our results presented here show that
free GFP (27 kDa) could move from cell to cell in young leaves of N.
occidentalis and C. quinoa, but was mostly restricted to single
cells in mature leaves (Fig. 3). This result is in
good agreement with the reported non-specific trafficking of GFP in
developing tobacco leaves (Oparka et al., 1999
). In contrast, 50KPGFP (77 kDa) spread into
neighbouring cells from cells that produced it even in mature leaves,
indicating that 50KP has a specific activity for cell-to-cell
trafficking.
It was unexpected that cell-to-cell trafficking of
50KPGFP was restricted to only a few cells in leaf epidermis of
C. quinoa plants. There may be structural and/or functional
differences between the plasmodesmata that interconnect epidermal cells of
N. occidentalis and C. quinoa plants. 50KPGFP
expressed in epidermal cells of C. quinoa leaves was able to
complement local spread of the movement-defective virus (Fig. 4). Thus, 50KPGFP may move from the epidermal cells
to underlying mesophyll cells in C. quinoa leaves. It has been
reported that a CMV mutant (M8) 3aGFP fusion protein was unable to
traffic through plasmodesmata that interconnect epidermal cells, as the
wild-type 3aGFP did, in tobacco (Nicotiana tabacum) and
Nicotiana benthamiana (Canto & Palukaitis, 1999
). Because M8 CMV infects tobacco systemically,
the virus spread via plasmodesmata to and within mesophyll tissue (Canto
& Palukaitis, 1999
).
In addition to the cell-to-cell trafficking of 50KP itself, we showed that 50KP can facilitate the cell-to-cell trafficking of GFP when both proteins are co-expressed transiently in epidermal cells of N. occidentalis leaves (Table 2). However, there was no facilitation of cell-to-cell movement of GFP when the protein was expressed in leaf epidermis of transgenic N. occidentalis constitutively expressing a functional 50KP (data not shown).
Transgenic N. occidentalis plants expressing
50KP are known to complement 50KP-deficient ACLSV for movement (Yoshikawa
et. al., 2000). As shown in Fig. 3, 50KP
expressed transiently in leaf epidermis also complemented local spread of
50KP-deficient virus. It has been reported that transgenic plants
expressing CMV 3a protein could complement 3a-deficient CMV, but plants
expressing 3a protein fused to GFP could not, showing that CMV 3a protein
fused to GFP is not biologically functional (Canto & Palukaitis,
1999
; Kaplan et al., 1995
). We also found that there was no
complementation of the movement of 50KP-deficient virus in N.
occidentalis expressing 50KPGFP (unpublished results). However,
50KPGFP could complement local movement of a 50KP-deficient virus
when pStuStop was co-bombarded with p35S50KPGFP. These results
suggest that complementation could occur when both p35S50KPGFP and
pStuStop were introduced in cells at the same time. On the other hand,
50KP expressed transiently in cells may be functionally different from
that in transgenic plant cells.
We do not know the biological significance of the
tubules on the surface of protoplasts induced by 50KPGFP. No tubular
structures spanning cell walls were found in ultrathin sections of tissues
infected with ACLSV (Yoshikawa et al., 1997
, 1999
) and the tubules were
observed in only a small portion of protoplasts that expressed
50KPGFP. This is similar to what has been described for TMV, where
fluorescent protrusions were formed in a small percentage of protoplasts
infected with TMV MPGFP (Heinlein et al., 1998
). In contrast, CMV 3aGFP generated
tubules on the surface of 1485 % of infected protoplasts (Canto
& Palukaitis, 1999
). So far, tubule
formation on protoplasts has been reported for spherical viruses in the
Comoviridae, Bromoviridae and Caulimoviridae and for
Tomato spotted wilt virus but not for TMV (Kasteel et al.,
1996
, 1997
; Storms et al., 1995
; van Lent et al., 1991
; Zheng et al., 1997
).
As summarized in Table 1,
mutational analysis has shown that the C-terminal region (between aa 287
and 475) of 50KP is not essential for localization to plasmodesmata,
cell-to-cell trafficking through plasmodesmata, complementation of local
movement of 50KP-deficient virus or tubule formation on the surface of
protoplasts. In contrast, deletions in the N-terminal region of 50KP
resulted in the complete disruption of all these activities, suggesting
that there must be a close correlation between these functions. The
expressed proteins (
D to
G) all formed large aggregates in the
cytoplasm. It is probable that these proteins are not folded properly and
that this makes the protein incapable of interacting with a subcellular
structure(s) or undergoing intracellular and intercellular
trafficking.
We thank Drs H. Taira and T. Yamashita for helpful advice. We also thank Dr R. H. Converse (USDA, ARS, HCRL) for critical reading of the manuscript. This study was supported in part by research grant no. 10660041 from the Ministry of Education, Science, Sports and Culture of Japan.
References |
© 2000 SGM
This article is now available in the August 2000 print issue of JGV (vol. 81, 20852093). The complete issue of the journal may be seen in electronic form on JGV Online.