Introduction

Potato leafroll virus (PLRV) is the type species of genus Polerovirus, family Luteoviridae. Its virions are isometric, with icosahedral symmetry T = 3 and a diameter of 25 nm. PLRV capsids are composed of 180 subunits of major coat protein (CP, 22,328 Da) and several copies of the secondary structural read-through protein (76 kDa). The latter contains the C-terminal extension of the major CP and is involved in PLRV transmission by aphid vectors. In addition, a single molecule of the genome-linked protein VPg (3.7 kDa) is covalently linked to the 5′-end of viral RNA [1,2,3,4].

PLRV has a positive-sense ssRNA genome of approximately 5.9 kb. The PLRV genome contains eight open reading frames (ORFs, see Fig. 1). The P0 protein encoded by ORF0 is involved in symptom development and is a suppressor of post-transcriptional gene silencing [5]. Proteins P1, P2 and P8 are involved in viral RNA replication [3]. The functions of P6 and P7 are unknown. ORF3, ORF4 and ORF5 are expressed from subgenomic RNA. ORF4, which encodes the movement protein MP (17 kDa, P4), is completely enclosed in ORF3 encoding the major CP (23 kDa, P3). It was recently shown that protein P3a that is coded by sgRNA1 is required for long-distance movement [6].

Fig. 1
figure 1

Scheme for constructing the chimeric crTMV-based vector expressing CP PLRV. pBNUP110: pBIN19 binary vector containing cDNA of PLRV (Canadian strain, GenBank: D13954.1), 35S: CaMV transcription promoter, 35ST: CaMV terminator of transcription, coded proteins: P1, P2, P8: replication proteins, P0: anti-silencing protein, P3, P5: major and minor CPs, P4: movement protein, and P6 and P7: non-characterized proteins. Arrows show read-through (RT). pCambia-crTMV: pCambia binary vector containing cDNA of crTMV. Act2-Pr: actin transcription promoter of Arabidopsis, nosT: nopaline synthase terminator of transcription of Agrobacterium, RdRp: replicase, MP: movement protein gene, CP: coat protein gene, t-like: tRNA-like structure. pCambia-crTMV-CPPLRV: pCambia binary chimeric viral vector with PLRV CP (P3) gene substituted for TMV CP gene and expressed from TMV CP subgenomic promoter

Full size image

The initiation codon of ORF4 is a few nucleotides downstream from the initiation codon of ORF3 [7,8,9]. The ORF4-encoded 17-kDa protein is unable to facilitate viral transport from phloem to mesophyll tissue [10]. Therefore, PLRV is confined to phloem tissue and is mechanically non-transmissible.

It is known that PLRV infection symptoms include chlorosis, necrosis and leaf curling. Necrosis symptom is caused by the selective death and damage to cells in the vascular tissues of the plant.

Wild-type PLRV can only be transmitted by aphids, mainly the peach aphid Myzus persicae. PLRV is a phloem-limited and low-titer virus. The yield of purified virus depends on the host plant but is usually poor. The mean yields of wild-type (WT) PLRV in potato and Physalis are 400 and 1300 ng/g leaf [11, 12], respectively. Because of low virus yield, it is difficult to obtain large quantities of virus antigen coat protein and to raise virus-specific antibodies for mass diagnostics in practice.

Our goal is to develop an easy and inexpensive protocol to obtain large quantities of the practically important PLRV CP or PLRV particles to routinely reprepare antibodies because massive diagnosis campaigns use large amounts of PLRV antisera and antibodies.

To overcome the problem of the low virus yield, the tasks of this study were: (1) construction of an efficient agroinfiltrable viral vector allowing overcoming the phloem limitation of a Polerovirus and producing large quantities of the artificial chimeric virus tentatively called the chimeric PLRV (chPLRV) composed of the tobamovirus RNA and the PLRV CP; (2) analysis of the infectiousness and assembly of chimeric virions; (3) raising antiserum against the PLRV CP antigen and determining its efficacy in recognizing wild-type (WT) PLRV.

On the way to the goal, we obtained several infectious recombinant crTMV [13] clones encompassing the CP PLRV gene, which was replicated by the very productive tobamovirus polymerase.

Materials and Methods

WT PLRV and antibodies against PLRV were a gift by Dr. Yu. Varitzev (Lorch Potato Farming Institute of the Russian Academy of Agriculture).

Infection of Plants

We have tried to agroinoculate N. tabacum, N. excelsior, N. clevelandii, salad, dope plants with recombinant clones of crTMV. It turned out that N. benthamiana was one of the best.

Agrobacterium tumefaciens strain GV-3101 was chemically (20 mM CaCl2) transformed with pCambia-crTMV-CPPLRV (see “Results” section) and other binary vectors using standard procedure [14]. Bacteria were grown in a 2YT medium containing antibiotics (rifampicin, gentamycin and kanamycin at 50, 25 and 100 μg/ml, respectively) at 28° C overnight. The cells were collected and resuspended in infiltration buffer (10 mM MgCl2 + 10 mM MES pH 6.5) to a density of 0.2 at OD600.

Leaves of N. benthamiana plants grown in a greenhouse were infiltrated with transformed bacteria on the underside of a leaf using a syringe without a needle. The injected mixture also contained the Tombusvirus p19 [15] gene, suppressor that causes RNA silencing. Fifty N. benthamiana plants at the seven-leaf stage were agroinfiltrated with pCambia-crTMV-CPPLRV in fully expanded lower leaves.

Chimeric Virus Isolation

Leaves and stems with fully developed symptoms were collected just before complete necrosis and were immediately ground in a blender (Waring) in 5 volumes of 0.1 M citrate buffer pH 6.5, containing 0.5% 2-mercaptoethanol and 1% Triton X-100. The pH of the extract was adjusted to 7.0 with 0.5 M Na2HPO4, and the mixture was incubated for 1 h on ice with shaking. This solution was then centrifuged at 10,000 rpm for 10 min in a JA-20 rotor (Beckman Coulter, Inc.). The resulting supernatant was emulsified with a chloroform/butanol (0.25:0.1 v/v) mixture and centrifuged at 10,000 rpm for 10 min. The aqueous phase was centrifuged at 100,000×g for 1.5 h at 4 °C through a 1.5 cm 20% sucrose cushion. We have determined experimentally sedimentation time of the chimeric virus at 100,000×g in the 10-cm tube (8 cm top solution, 1.5 cm sucrose cushion). 90-min centrifugation was found the best time, after 2-h centrifugation, some quantity of the host Rubisco protein appeared in the chimeric virus preparation. Purified chimeric virus was resuspended in 0.01 M Tris–HCl, pH 7.5.

Analyses of the Chimeric Virus

SDS-PAGE and Immunosorbent Electron Microscopy

Squares (6 × 6 mm) were cut from the infected leaves of the first, second and third tiers (starting from the top of the plant). The green material was triturated in 3 volumes of 10 mM Tris–HCl, pH 7.5 and analyzed using 12% SDS–polyacrylamide gel electrophoresis according to Laemmli [16] and by immunosorbent electron microscopy.

About 30 µl of antibodies against WT PLRV (0.2 mg/ml) was placed on the collodion film on the grid, incubated for 10 min, and the excess antibody was washed away with phosphate buffered saline (PBS). The chimeric virus suspension in PBS (0.05–0.5 ng/ml) was placed onto the film and incubated in a wet chamber overnight at room temperature. The excess reagents were washed away with PBS and water. Purified chimeric PLRV was contrasted with 2% uranyl acetate and examined with a Jeol JEM 1011 microscope.

Nanoparticle Tracking Analysis

Nanoparticle tracking analysis experiments were performed using a NanoSight NS500 system (Malvern Instruments Ltd.) [17]. Video images were analyzed with the Nanoparticle Tracking Analytical Software (ImageJ version). The measurements were performed at room temperature, and the capture rate of each video clip was 60 s.

RNA Isolation

Chimeric virus RNA and total RNAs from the infected and uninfected plants were isolated by the classical SDS–phenol/chloroform (1% and 10:1 v/v, respectively) procedure or by the phenol/bentonite [18] method and with guanidinium thiocyanate–phenol–chloroform extraction [19] using TRIZOL reagent (Thermo Fisher Scientific).

RNA samples were analyzed by electrophoretic separation in 1.4% agarose gel and TBE buffer (89 mM Tris, 89 mM boric acid, 2 mM EDTA, pH 8.0) and visualized using an ultraviolet light after staining with ethidium bromide solution.

Dot-Blot and Northern Blot Hybridization

The DNAPLRV probe labeled with biotin (biotin-11-dUTP) to detect the PLRV CP coding sequence was obtained by PCR with random primers of the cloned CP PLRV gene in a plasmid with the Biotin DecaLabel DNA Labeling Kit (Thermo Fisher Scientific) according to the manufacturer’s protocol.

RNA samples were denatured for 15 min at 55°C in MAE buffer (200 mM MOPS, 50 mM sodium acetate, 5 mM EDTA, pH 7.0) with 50% formamide and 6.7% formaldehyde, and analyzed by electrophoresis on a 1% agarose gel in MAE with 6.7% formaldehyde. The RNAs were electroblotted (semidry transfer) from the gel to Hybond N+ nylon membranes (GE Healthcare) and fixed by UV irradiation using the UV Stratalinker 1800 (Stratagene, Agilent Technologies). The membranes were stained with 0.04% methylene blue in 0.5 M sodium acetate, pH 4.4 and photographed. The dye was removed from membranes with 1% acetic acid and the membranes were then neutralized with 6xSSC buffer. Membranes with blotted RNAs were incubated for 1 h at 65°C in hybridization buffer (5xSSC, 2xDenhardt solution, 0.1% SDS) with 50 µg/ml heterologous sonicated and denatured calf thymus DNA. The blotted RNAs were hybridized with the biotinylated DNA probe to the CP gene of chPLRV RNA overnight at 65°C [20]. Chromogenic heteroduplex detection was carried out using the streptavidin–alkaline phosphatase conjugate and the chromogenic substrate of alkaline phosphatase BCIP/NBT with the Biotin Chromogenic Detection Kit (Thermo Fisher Scientific) according to the manufacturer’s protocol.

Immunization and Antibody Isolation

Two female rabbits (Chinchilla) were immunized nine times, three subcutaneous injections per cycle, with three chPLRV preparations at intervals of 10–15 days per cycle. Complete (20%) and incomplete (20%) Freund’s adjuvants were added for first and second immunizations, respectively.

In total, approximately 10 mg (rabbit 1) and 5 mg (rabbit 2) of the chPLRV preparations were utilized. Blood samples were collected, and antisera were then analyzed within 1 week after the third immunization in the cycle. Both rabbits showed very close titers of antisera against chPLRV after 6-9 immunizations. Before immunoglobulins isolation, antisera were combined and stored at −80 °C. The following study was carried out with antibodies isolated from this combined antiserum.

Isolation of antibodies to the chPLRV was performed according to the previously described method [21] with modifications. Immunoglobulins G were purified by the three-step procedure at 4 °C: (1) ammonium sulfate precipitation, dialysis; (2) anion-exchange chromatography on DEAE-Toyopearl 650 M TSK gel. Immunoglobulins G were eluted from DEAE-Toyopearl 650 M with 50 mM NaCl. The yield of immunoglobulin G eluted with 50 mM NaCl was 50 mg from 86 ml of the antiserum; (3) concentration on the Millipore Amicon Ultra 50-kDa centrifugal filters.

Laemmli SDS-PAGE Showed Typical Pure ~45 and ~22 K Subunits of Immunoglobulins

A double-antibody sandwich enzyme-linked immunosorbent assay was carried out as previously described [22]. Quantitative analysis was performed by admixing definite quantities of WT PLRV to the aliquot of the uninfected juice of the potato leaves.

Results

Construction of pCambia-crTMV-CPPLRV

Prior to constructing the chimeric vector, an alignment analysis was performed (data not shown) using published sequences of the CP gene from PLRV isolates originating from various countries. The published sequence data [23] had 98.7% similarity for all samples except the PLRV isolate SymlessLS10 [24]. Thus, we proposed that the CP PLRV gene we had (Canadian strain, GenBank: D13954.1) was suitable for raising universal polyvalent antiserum against wild-type PLRVs.

The crTMV RNA is composed of 6312 nucleotides and contains four ORFs encoding the proteins 122 K (ORF1), 178 K (ORF2), 29 K (ORF3) and 18 K (capsid protein, ORF4). ORF4 overlaps ORF3 by 74 nucleotides [13].

In the generated chimeric construct, the transcribed viral recombinant RNA was 6522 nt, whereas WT PLRV is 5882 nt long [3]. The crTMV-based vector, which produces hybrid virus particles of chimeric crTMV RNA encapsidated by PLRV CP, was engineered by crTMV CP gene substitution with the PLRV gene coding for 22.3 K CP (ORF3 product) and deletion of the intergenic region upstream of the PLRV CP gene. The resulting construct was named pCambia-crTMV-CPPLRV (Fig. 1, see “Discussion” section).

In pCambia-crTMV-CPPLRV, the PLRV transport gene p4, which is completely inside the p3 coat protein gene is expressed with the same subgenomic RNA, but is located in another translation frame (Fig. 1). P4 was inactivated by replacing in the PCR primers of T with C (underlined in Fig. 1), which eliminated two start ATG codons in p4, but did not change the amino acid composition of p3 (N, Asp, aat-aac).

The fragment of 627 bp containing ORF3 was amplified by polymerase chain reaction (PCR) using PLRV cDNA in pBNUP110 [25] as a template and the following oligonucleotides as primers:

5′-TTCTCGAGATGAGTACGGTCGTGGTTAAAGGAAACGTCAACGG-3′ (PLRV PΔ4-p) and 5′-TTGGTACCCTATTTGGGGTTTTGCAAAGCCA-3′ (PLRV CP-m).

The PCR fragment was double-digested with XhoI and Acc65I, ligated into the pCambia-crTMV plasmid [26] between XhoI and BsrGI restriction sites, and sequenced.

In the cDNA, which is under control of the plant Act2 promoter, there is a transcription terminator at the end of the protein gene of the CP PLRV (Fig. 1). Thus, when the plant is transfected with the cDNA that is transcribed as a recombinant viral RNA, which is capable of both systemic infection of the plant and the synthesis of the PLRV coat protein, the last forms an isometric chimeric viral particle with the recombinant RNA.

Hybrid PLRV-Like Particle Production by the crTMV-Based Vector Expressing PLRV CP

Wild-type crTMV induces severe symptoms similar to U1 TMV on benthamiana plants, maybe a little bit milder, and no necrosis of stalks was observed.

Five days after infiltration with constructed pCambia-crTMV-CPPLRV vector, the region of N. benthamiana that was inoculated became necrotic (Fig. 2a), and 2–3 days later, symptoms developed on the systemic uninoculated uppermost leaves. After approximately 2 weeks, viral symptoms of systemic severe infection in the form of leaf tissue deterioration and obsolescence developed on inoculated plants and necrosis of leaf veins and stalks (petioles) (Fig. 2b) stopped the virus from spreading.

Fig. 2
figure 2

Symptoms developed on N. benthamiana plant after agroinfiltration of pCambia-crTMV-CPPLRV . a 1—necrosis of infiltrated zone, 2—petiole necrosis. From the top: I—first tier leaf, II—second tier leaf, III—third tier leaf, IV—healthy lower leaf. b Arrow indicates the stop of chimeric virus spread along the leaf vein due to necrosis

Full size image

Using double-antibody sandwich enzyme-linked immunosorbent assay (DAS-ELISA), we compared the concentration of WT PLRV and chPLRV after agroinfiltration with pBNUP110 and pCambia-crTMV-CPPLRV in tiers of infected plants (Table 1).

Table 1 Dependence of PLRV coat protein accumulation in N. benthamiana leaves on inoculation method and sample type
Full size table

Study of N. benthamiana Infection with Chimeric Virus and Characterization of Purified pCambia-crTMV-CPPLRV Virions

Chimeric virus infection was verified by protein gel electrophoresis and immunosorbent electron microscopy (ISEM) of the infected cell extracts with antibodies against WT PLRV.

As a result of infection by pCambia-crTMV-CPPLRV, 23-kDa viral CP accumulated in the N. benthamiana leaves, as shown in Fig. 3a.

Fig. 3
figure 3

a Coomassie-stained SDS-PAGE of proteins isolated from I to IV tier leaves. The arrow on the right of the panel indicates the expected molecular weight of the CP PLRV protein. BSA was added as a control to approximate the CP quantity. b 12% SDS-PAGE of the chimerical virus preparation. Coomassie R-250 stained SDS-PAGE of the PLRV isolated and purified from 5 g of the infected N. benthamiana leaves and dissolved in 3.5 ml of PBS. Aliquots of this solution were applied onto gel: 1—20 µl (~20 µg of the virus), 2—10 µl, 3—5 µl. M—markers of molecular mass of proteins. The arrow on the left marks the supposed position of the putative CP PLRV; c Immunosorbent electron microscopy of the chimeric PLRV trapped from the upper leaf extracts by using commercially available antibodies to WT PLRV (staining with 2% uranyl acetate)

Full size image

Laemmli PAGE of the chimeric PLRV preparation showed a good purity of chimeric virus. pCambia-crTMV-CPPLRV virus capsid protein corresponded to the ~23 kDa protein, which was similar in size to the PLRV WT capsid protein. Only negligible contamination of the samples with high molecular weight proteins, apparently corresponding to cellular proteins, was revealed in the “hybrid” virus preparation; a small amount of proteins smaller than 23 kDa may be products of the partial proteolysis of the capsid protein during isolation of the chimeric virus (Fig. 3b). Upon comparing Coomassie staining intensity with the intensity of a defined quantity of BSA, the yield of the chimeric virus was determined to be much higher than the yield of WT PLRV [12]. ISEM of the cell extracts from the visually infected N. benthamiana (upper tier) leaves using antibodies (Agdia, USA) to PLRV revealed typical isometric PLRV-like spherical hybrid particles (Fig. 3c). The shape and size of these were similar to WT PLRV particles [28]. Mock-infected control plants and an experiment with preimmune serum showed no virus particles. Moreover, antibodies against this chimera recognize PLRV in the infected potato specifically as was shown by immunochromatography [27].

RNA was extracted from the isolated chimeric virus particles by the classic SDS–phenol deproteinization method. Spot hybridization of the DNA probe to the PLRV CP gene confirmed the presence of the recombinant RNA in the chimera virus preparation and in the infected plant (Fig. 4a, line b and c).

Fig. 4
figure 4

a Dot-blot hybridization of the amplified PLRV CP gene products with the DNAPLRV probe. The DNA probe, labeled with biotin (biotin-11-dUTP), was obtained by PCR with random primers of the cloned CP gene in a plasmid. Lines: a—uninfected N. benthamiana; b—infected N. benthamiana; c—chimeric PLRV; d—TMV. Samples of RNA were applied from left to right: 500, 50, 5, 0.5 ng, respectively. b Agarose gel electrophoresis of the formaldehyde-denatured RNAs (ethidium bromide staining). Rows: 1—TMV RNA (1 µg was loaded onto the agarose gel); 2—1.1 and 2.1 kb RNA markers (2 µg, Thermo Scientific); 3—total RNA of the uninfected N. benthamiana (12 µg); 4—total RNA of N. benthamiana infected with chimeric PLRV (12 µg); 5—RNA extracted from the chimeric PLRV preparation (6 µg). c Northern blotting analysis with hybridization probe complementary to the PLRV CP gene. Samples in rows 3, 4, 5 are the same as in b. Double-headed arrows indicate proposed positions of the crTMV RNAs hybridizing with the CP PLRV DNA probe

Full size image

Agarose gel electrophoresis showed that RNA preparations obtained from the plant after early and late infection by the chimera, as well as from the chimeric virus preparation, were highly hydrolyzed, as shown in Fig. 4b, c (rows 4, 5). The same results were obtained using phenol–bentonite extraction of RNA from the chimeric virus [18] or with TRIZOL reagent [19] (not shown).

We further characterized the “hybrid” virus particles produced by the vector pCambia-crTMV-CPPLRV in the upper (non-inoculated) leaves of agroinfiltrated plants and comprising of chimeric crTMV RNA encapsidated by PLRV CP within the isometric PLRV-like capsid (Fig. 5b, c).

Fig. 5
figure 5

a Ultraviolet light absorption spectra of the chimeric PLRV preparation. Since the true absorption of viruses in the 320–350-nm range of the spectrum is practically zero, the OD at wavelengths between 350 and 320 nm was plotted against wavelength values on a log–log scale and the linear portion was extrapolated to 240 nm, as shown on the left (green line). Thus, the contribution of scattering into the optical absorption of virus suspension was defined. The log values obtained were converted into absorbance values using a table of logarithms and plotted (right panel). Subtraction of the light dispersion spectrum (blue curve) from observed spectrum (red curve) resulted in the true UV spectrum of the hybrid virus particles suspension (black curve). b Nanoparticle tracking analysis of WT PLRV. 1—“hybrid” pCambia-crTMV-CPPLRV virus; 2—WT PLRV virions. c Electron microscopy image of chimeric PLRV. Virus samples were stained with 2% uranyl acetate (Color figure online)

Full size image

chPLRV hybrid particles tend to aggregate, and aggregates give light scattering at the light length shorter than 300 nm. Therefore, the viral concentration according to UV light absorption appears to be higher than the actual concentration because of the addition of some light absorption due to the light scattering.

Thus, to measure the true spectrum, we used the so-called extrapolation approach [29] as follows from the Rayleigh scattering law (Escatt ~ 1/λα) for large particles, and plots of log OD versus log λ were generally straight in the whole range of the UV wavelengths of the spectrum of nucleoproteins.

The black curve presented in Fig. 5a shows the typical spectrum of a nucleoprotein. Taking into consideration spectral and protein electrophoresis data, we calculated the chimeric virus concentration using Gibbs’ formula [30]. In accordance with these calculations, the yield of chPLRV was 18-20 mg from 25 g of infected leaves (average yield from three isolations), almost three orders of magnitude higher than the yield of WT PLRV. The calculations showed good agreement among the results obtained for the chimeric virus by ELISA, electrophoresis in polyacrylamide gel and UV spectroscopy. Thus, it became much easier to produce PLRV antigen.

The electron micrograph of the hybrid pCambia-crTMV-CPPLRV virions presented in Fig. 5c showed isometric particles.

Under low-magnification (large field) electron microscopy of the chimeric virus preparation shows (see Fig. 6) small percentage of the stain permeable virus-like particles that filled with uranyl acetate, presumably those are empties or, in other words, virus-like particles (VLPs).

Fig. 6
figure 6

Low-magnification electron microscopy image of chimeric PLRV preparation. Virus samples were stained with 2% uranyl acetate

Full size image

Native WT PLRV virions and PLRV-like hybrid particles were compared using nanoparticle tracking analysis, which allowed us to concurrently visualize and determine the size of native virions and “hybrid” virions in a liquid. The results presented in Fig. 5b, c show that the size of the native virions and chimeric virus particles were identical, measuring 24–26 nm.

Testing of the Polyclonal Antibodies for the PLRV Infection Diagnostics in Potato

Testing Immunosera by Dot Immunoassay

Results of detecting WT PLRV infection in potato with antiserum against chPLRV after six immunizations by dot immunoassay are shown in Fig. 7a. A positive signal was found in the clarified potato leaf juice of an infected potato plant at a 1:100 dilution, and a negative signal was found for the healthy plants at a 1:10 dilution, indicating an appropriate sensitivity and specificity of the obtained antiserum.

Fig. 7
figure 7

Immunoassays to test specificity of the antiserum against the viral chimera. a Dot assay. NC membrane with dotted extracts from a potato plant infected with wild-type PLRV (3, 4) and a healthy plant (1, 2). Clarified leaf juices were diluted 1:10 for 1 and 3, and 1:100 for 2 and 4. b Double-antibody sandwich enzyme-linked immunosorbent assay of the chimeric PLRV antibodies. Antigens: chimeric PLRV (solid line), wild-type PLRV (dashed line). Negative control (uninfected potato juice) at 450 nm was 0.08–0.12 OD. Each curve obtained was based on the results of two independent experiments

Full size image

Thus, the antiserum obtained against the chimeric PLRV recognizes natural PLRV isolates in infected potato plants.

Testing Immunoglobulins Against the Chimeric Virus by DAS-ELISA

A double-antibody sandwich enzyme-linked immunosorbent assay using antibodies against the chimeric PLRV at a concentration of 1 µg/ml reliably identified 30 ng/ml of the WT PLRV in clarified leaf juice from an infected potato plant (Fig. 7b).

Discussion

Fraenkel-Conrat and Williams [31] originally reconstituted in vitro rod-like infectious hybrid particles of TMV from viral CP and RNA. Later, numerous reports described mixed reassembly of TMV CP with heterologous RNAs both in vitro and in vivo. Under mixed infection conditions, heterologous CP and RNA may be assembled into so-called hybrid viruses. Various isometric viruses have also been reconstituted in vitro from purified CP and homologous or heterologous RNAs [32,33,34,35]. The phenomenon of in vivo genomic encapsidating in plants doubly infected with different TMV strains or unrelated rod-shaped viruses has been reported. Several helical plant virus vectors have been shown to express the CP of a foreign isometric plant virus, which can then assemble into infectious isometric virions containing chimeric genome [36]. There is evidence that CP of PLRV and BWYV can encapsidate foreign RNA of umbravirus CMotV so it become aphids transmittable [37].

Barker et al. [38] constructed a vector on the base of the TMV U1, in which its normal CP gene was replaced by the CP gene from PLRV (TMVΔCP-PLRV-CP). When N. benthamiana was inoculated with the chimeric virus TMV∆CP-PLRV-CP, virus multiplied profusely in the inoculated leaves; infected areas were visible by 4 days post-infection which spread to about 80% of the leaf area by 8 days post-infection. This vector showed limited accumulation in Nicotiana. It did not move systemically, probably because the CP is essential for TMV long-distance movement [39] and PLRV CP is unable to complement this function [37].

In their study, Barker et al. [38] examined the PLRV infection on transformed N. benthamiana with Agrobacterium tumefaciens containing plasmid pBNUP110 or infected the plant rubbing the PLRV transcripts onto carborundum dusted leaves. The PLRV yield was no more than 1600 ng/g of the transgenic plant, even when mixed with PVY infection.

In this work, we figured out to simplify the procedure for the production of the hard to access PLRV antigen in large quantities, which is appropriate for biotechnology and raise specific antibodies.

We used genome of another strain of TMV, crucifer infecting virus for the binary vector construction (Fig. 1) that was isolated earlier and its genome sequenced [13].

One of the peculiarities of the constructed crTMV-based vectors was that the crTMV CP gene was substituted with the PLRV CP (ORF3). As was mentioned, CP from PLRV (23 kDa) is completely contained the MP from PLRV (17 kDa). Thus, one construct contained two functional movement proteins: crTMV MP and PLRV MP. The 17-kDa MP from PLRV is responsible for controlling the short-distance transport between sieve elements and companion cells [8]. As mentioned above, this protein has biochemical properties similar to the properties of the 30 K MP of TMV [40, 41].

Agroinfiltration of N. benthamiana with the binary vector that had both crTMV and PLRV movement proteins was examined. This vector induced local severe infection symptoms and slow systemic infection. We concluded that the similarity between the crTMV and the PLRV MPs manifested itself in interference, decreasing the power of crTMV MP and, hence, accumulating the “hybrid” PLRV. Therefore, we speculate that tobamovirus MP and MP PLRV compete with each other in plasmodesma. To our knowledge, this is the first direct study of the influence of two different MPs (the 30S MP of crTMV and P17 MP of PLRV) on chimeric virus multiplication.

One of the possible ways to clarify the 17-kDa PLRV MP function was to eliminate this gene from the chimeric virus to compare the levels of expression of crTMV(ΔCP)–PLRV(ΔMP)-CP and crTMV(ΔCP)–PLRV-MP-CP. The PLRV MP expression was blocked by eliminating the two start codons from the 5′-end of ORF4. In addition to elimination of the 17-kDa gene, the 5′-untranslated region upstream of ORF3 and the 3′-proximal ORF5 encoding the C-proximal region of the 76-kDa read-through protein production was removed (Fig. 1). As shown in Table 1, eliminating the 17-kDa PLRV MP resulted in a sharp increase in “hybrid” PLRV accumulation in the upper leaves of N. benthamiana agroinfiltrated with pCambia-crTMV-CPPLRV into expanded leaves. Thus, a chimeric virus was obtained that induced severe infection and necrosis of the leaf stalks and veins (Fig. 2). The data in Table 1 indicate that the PLRV antigen encoded by the virus vector accumulated in the small upper leaves of the agroinoculated plant in greater amounts than the amounts observed in the primary infected cells, but there was a relatively low yield of the isometric particles in the immediate area of agroinfiltration. This result agrees with the finding that cell-to-cell movement of WT PLRV is restricted to phloem, and the virus is unable to move efficiently between mesophyll cells [42].

Another chimeric vector included the untranslated long intergenic region of 208 nucleotides located upstream of the PLRV CP gene (unpublished). However, such chimeric virus did not infect N. benthamiana leaves except the uppermost leaf, where it accumulates in large quantities, whereas the vascular system of the stem and petioles was affected within two to three weeks only. Presumably, this PLRV infection spreads slowly because it is blocked by callose deposition at the plasmodesmata [43].

Earlier it was demonstrated that deletion of the intergenic region of 208 nucleotides located upstream of the PLRV CP gene [44] gave rise to multiple increasing of the MP PLRV synthesis in vitro. We presume that level of expression of 17K protein in infected leaves is comparable with that one of the CPs. Upon considering the influence of the upstream regulatory elements located within the read-through domain of CP on its translation we removed this intergenic region. This vector infected leaves from the first to the third tier but did not go further down to older leaves. It is interesting that the PLRV protein 3a [6] is inside of this intergenic region. Nevertheless, despite the p3a deletion, we had long-distance infection by the chimeric virus.

Compared with the previous study [38], in this work several changes were made: first, crTMV instead of TMV U1. It should be underlined that transfer protein of crTMV, but not MP TMV, has nuclear localization signal (NLS). Recently was shown that NLS of the tobamovirus MP enhances virus infection [45]; secondly, primers that differed from those used previously for PCR amplification of the CP PLRV gene; then, the intergenic region was removed, more efficient way of the plant infection upon agroinfiltration; and P19 silencer instead of P1/HC-Pro were employed. The sum of these modifications has resulted in the fast and long-distance infection of N. benthamiana and high chimeric virus titer.

Based on the data in Fig. 3 and Table 1, the chimeric PLRV was isolated from the infected N. benthamiana leaves of the upper tiers (systemically infected leaves) using the method described in “Materials and Methods” section. These PLRV-like hybrid particles were capable of cell-to-cell movement and long-distance transport infection of the plant, and we found that the coat protein of PLRV was essential for the long-distance movement. We suggest that PLRV CP is directly involved in transporting the chimera within the plant. Similar vectors in which the CP gene was substituted with GFP also infected N. benthamiana, but at much lower speed.

It should be mentioned that shape (icosahedra), size and symmetry (T3) of PLRV, CMV (Cucumber Mosaic Virus) and BMV (Brome Mosaic Virus) are similar. CP CMV and CP BMV have jelly-roll tertiary structure. As was shown, TMV RNA is packed by CP CMV or CP BMV in spherical particles (35 and references therein). Therefore, we assumed that CP PLRV is able to pack crTMV RNA. Finally, we see efficient, fast and long-distance infection and multiplication of the nucleoprotein particle, i.e., virus.

The serious issue of the chimeric virus purification is a high tendency of the virus particles to aggregate in a merely wide range of concentrations. During aggregation, virus particles might grab nearby surrounding substances. These aggregates were stable in a wide range of pH and in the presence of non-ionic detergents.

In our study, purity criteria of the chimeric virus preparation were based on data of the UV spectroscopy, protein electrophoresis, ISEM with commercial antibodies against WT PLRV, electron microscopy and ELISA.

SDS-PAGE of the purified virus (Fig. 3c) did not reveal much of the host admixtures (Fig. 3). All of them, the UV spectroscopy, chimeric virus preparation has a nucleoprotein spectrum with the maximal absorption at 260 nm (Fig. 5), dot-blot hybridization and Northern blotting analysis of isolated virus particles (Fig. 4), electron microscopy (Figs. 3, 5) confirmed that we deal with the chimeric virus. The low percentage of VLPs, we speculate, was also presented in the chimeric virus preparation (Fig. 6).

Even single freezing and thawing of the chimera result in the release of RNA from the chimeric virus particles. It appears that the chimeric virion is less structurally ordered (firmed) than the WT PLRV, presumably because of lack of P5 read-through protein. We knowingly did not express the P5 PLRV protein, because the expression of proteins above 70 kD in size by tobamovirus vector is not effective. We suppose that the chimeric virion is permeable for the host nucleases (see Fig. 4b, c).

According to dot-blot hybridization of the amplified PLRV CP gene products and Northern blotting analysis with a hybridization probe complementary to the PLRV CP gene, it can be concluded that the replicated recombinant RNA contains the PLRV CP encoding gene (Fig. 4), and both cellular and chimeric RNAs in infected cells (Fig. 4b) and virion RNAs (Fig. 4c) are highly hydrolyzed, while in uninfected cells, ribosomal 25S (3.7 kb) and 18S (1.9 kb) RNAs appeared to be normal. Moreover, the same results were obtained using both phenol–bentonite [15] or TRIZOL [16] RNA isolation procedures.

As mentioned above, recombinant crTMV RNA consisted of 6522 nucleotides, and subgenomic RNAs consist of sgRNA I1 (3128 nucleotides), I2 sgRNA encoding the capsid and transport proteins (1649 nucleotides) and LMC sgRNA encoding the capsid protein (875 nucleotides). Identification of chimeric virus-specific transcripts by the Northern blotting method confirmed the expression of genes from promoters of the transport protein and TMV capsid protein (Fig. 4c, rows 4, 5). Northern blot assay with the PLRV CP DNA probe showed that PLRV I2 and LMC transcripts, as well as some higher molecular weight RNA, which presented as a smear in Fig. 4c, row 5, were packed in the capsid by the CP of the chimeric virus.

We were surprised by the host ribosomal RNA hydrolysis in the chimeric virus-infected cells.

Even at the early time of infection (4 days), RNA isolated from the chimeric virus was hydrolyzed. These results can be explained by observation obtained from the thin sections light and electron microscopy of the infected cells. We found that the chimeric virus first accumulates in the nucleolus where a content of the last seems “dissolved.” The virus accumulates in the granular layer of the nucleolus where ribosomes mature. At this step, precursors of the ribosomal RNAs can be hydrolyzed (degraded). Then the chimeric virus fills nucleus, chromatin is destroyed and virus exits to the cytoplasm through the broken nucleus membrane (manuscript in preparation).

The mechanism of cell-to-cell movement of hybrid crTMV–PLRV CP particles from the primary agroinfected leaf is not clear. It cannot be ruled out that viral hybrid particles are entirely limited to phloem and move along this pathway to the upper non-inoculated leaves.

The morphology of the chimeric virion was found to be the same as the morphology of the WT virus (Fig. 3b) and Fig. 5b, c. Because commercial antibodies against WT PLRV CP recognized the chimeric virus (Fig. 3b), stereometric characteristics of the chimera antigenic determinants should be similar to the stereometric characteristics of natural isolates of the virus since they are determined by the general mechanism of viral particle assembly. This is a prerequisite for the use of chimeric virus particles as the target antigen for immunization of laboratory animals. Comparative cross DAS-ELISA of the chimeric and WT viruses with antibodies against WT and chimeric PLRV showed that capsid of the chimeric virus is similar to WT CP PLRV one (Fig. 7). In addition, we used antibodies against chimeric PLRV for diagnostic analysis of the PLRV-infected potatoes by the lateral flow immunochromatographic assay [28]. Thus, the quality of the chimeric virus was satisfactory for the task, and its calculated yield was 800–1000-fold higher than WT PLRV.

PLRV can be responsible for individual potato yield losses of over 50%. Viral diseases of agricultural plants are not yet curable; therefore, for their effective management, a sensitive diagnosis is extremely important to identify and eradicate infected plants.

The developed approach for the difficult to access virus antigens production might be useful for practical mass immune diagnostics of viral plant infections at the molecular level such as lateral flow immunochromatographic assay, which is indispensable for the large-scale express diagnostics, for example, in epiphytoty.