Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Risk‐managed production of bioactive recombinant proteins using a novel plant virus vector with a helper plant to complement viral systemic movement

Risk‐managed production of bioactive recombinant proteins using a novel plant virus vector with a... Introduction Plants represent an advantageous system for expressing recombinant proteins in terms of cost effectiveness and the ease of scaling up production ( Ma , 2003 ; Arntzen , 2005 ; Kaiser, 2008 ). However, the drawback for the genetic engineering of plants has always been the very long development period ( Hiatt and Pauly, 2006 ). The ideal expression system for recombinant proteins including vaccines and antibodies over short periods of time would be a viral vector system ( Hiatt and Pauly, 2006 ; Arntzen, 2008 ). Although the high potential of plant viral vectors for heterologous gene expression in plants has been well recognized ( Gleba , 2007 ; Lico , 2008 ), the safest course is to limit the use of viral vectors to when other methods (i.e. microorganisms, animal cells) to produce the target protein do not work because of the great concern of using recombinant viruses in open fields. Seeking a possible solution to this problem, we here constructed a novel Cucumber mosaic virus (CMV) vector as a viral confinement system (VCS) to produce a high‐quality recombinant protein in plants. CMV belongs to the genus Cucumovirus and has a very broad host range of more than 1200 species. CMV is divided into two subgroups I and II, and subgroup I is further divided into IA and IB ( Roossinck, 2002 ). CMV has tripartite, plus‐sense RNAs designated RNAs 1, 2 and 3 in decreasing order of molecular weight. RNAs 1 and 2 encode replicase proteins (1a and 2a, respectively), and RNA3 encodes the movement protein (3a), which allows CMV to move to neighbouring cells ( Suzuki , 1991 ; Ding , 1995a ). The coat protein (CP) is translated from a subgenomic RNA, RNA4, derived from RNA3, and has been found to be necessary for long‐distance movement ( Taliansky and García‐Arenal, 1995 ). RNA4A is a subgenomic RNA generated from the 3′ end of RNA2 and encodes a multifunctional protein, the 2b protein (2b), which has been shown to be an RNA silencing suppressor (RSS). 2b is not itself the movement protein but can enhance the viral cell‐to‐cell and long‐distance movement by increasing viral RNAs perhaps through its RSS ability ( Ding , 1995b ; Soards , 2002 ; Shi , 2003 ). We previously developed CMV as a gene expression vector to produce a human cytokine, acidic fibroblast growth factor (aFGF), in plants ( Matsuo , 2007 ). By modifying the original vector, here we tried to develop a novel vector that cannot move independently from cell‐to‐cell and can infect systemically only with the aid of a transgenic helper plant that expresses the 3a gene encoding the viral movement protein. We call this system the VCS. With cucumoviruses, mixed infection often results in RNA recombination between CMV isolates ( Chen , 2002 ; Lin , 2004 ; Bonnet , 2005 ), within a CMV isolate ( Canto ,2001 ), and between CMV and another cucumovirus, Tomato aspermy virus (TAV) ( Masuta ,1998 ; Aaziz and Tepfer,1999 ; de Wispelaere ,2005 ). In addition, recombination can occur to produce common 3′ ends within a pseudorecombinant virus containing RNAs 1 and 2 from CMV and RNA3 from TAV ( Fernández‐Cuartero ,1994 ; Shi ,2003 ). Towards the development of a more useful system, we sought to diminish the risk of CMV recombination. By using a combination of two kinds of the 3a genes, one from CMV subgroup I and the other from subgroup II ( Roossinck,2002 ), we could prevent RNA recombination in RNA3. Possible mechanisms underlying viral systemic movement and the recombination‐free nature of the VCS are also discussed. To demonstrate the expression of high‐quality recombinant proteins by our VCS, we here describe the expression of a functional single‐chain variable fragment against dioxins (DxscFv), which are persistent organic pollutants, and human interleukin‐1 receptor antagonist (IL1‐Ra), a rheumatoid arthritis agent. The protein expression efficiency was compared to previously published systems. Results Novel expression system to confine the CMV vector in the infected cells The CMV vector constructs used in this study are shown in Figure 1a . RNAs 1–3 are all necessary for creation of the virus and thus infection, although foreign genes are inserted in the vector clone of RNA2 (pCY2‐H1). When the virus moves in inoculated plants, expression of the 3a and the CP genes on RNA3 are necessary for cell‐to‐cell movement and long‐distant movement, respectively. The CP for encapsidation will be synthesized by RNA4 derived from RNA3 of the inoculated CMV vector. To destroy the 3a gene, a stop codon was inserted in the 3a gene of RNA3 at the 4th position to create pCY3‐YΔ3a (transcript, YΔ3a), which has the 3a sequence of subgroup I CMV‐Y ( Figure 1a ). We also engineered pCM3‐m2Δ3a (transcript, m2Δ3a), which encodes the subgroup II 3a protein containing two stop codons in the 3a gene at the 4th and 10th position ( Figure 1a ). Next, we created transgenic Nicotiana benthamiana plants expressing the 3a gene of CMV‐Y by using the conventional plant expression vector (pBE2113) ( Figure 1b ). As shown in Figure 1c , the entire scheme for VCS is based on complementation of the movement‐defective CMV vector to produce a recombinant protein with the helper plants, which supply the movement protein (3a) in trans to enable cell‐to‐cell movement of the virus. To demonstrate VCS, we selected DxscFv as a target protein. The DxscFv gene (837 bp) was then inserted into the pCY2‐H1 vector to create pCY2‐H1:DxscFv. As illustrated in Figure 1a , the viral inoculum consisted of the transcripts from pCY1 (R1), pCY2‐H1:DxscFv (H1Dx) and pCY3‐YΔ3a (YΔ3a) or pCM3‐m2Δ3a (m2Δ3a) to generate the recombinant virus, Δ3aH1Dx (YΔ3aH1Dx or m2Δ3aH1Dx). For another target IL1‐Ra, the DxscFv gene was simply replaced with the gene (480 bp) for the mature form of IL1‐Ra. 1 Viral confinement system (VCS) to produce recombinant proteins in helper plants. (a) Constructs of the movement‐defective Cucumber mosaic virus (CMV) vector to express the DxscFv gene. The 3a genes of CMV‐Y (subgroup I) and CMV‐m2 (subgroup II) have one and two stop codons immediately after the initiation codon, respectively. T7p, T7 promoter; 1a, helicase; 2a, RNA polymerase; 2b, silencing suppressor; 3a, movement protein; CP, coat protein; MCS, multiple cloning site; SP, signal peptide; VH, variable region of H chain; VL, variable region of L chain; 6His, 6 × his tag; KDEL, retention signal. (b) Constructs of a binary vector to produce helper plants expressing the 3a gene of CMV‐Y (subgroup I). RB and LB, binary right and left borders; Pnos, promoter of nopaline synthase; NPTII, neomycin phosphotransferase gene; Tnos, polyadenylation signal of nopline synthase; P35S, Cauliflower mosaic virus 35S promoter. (c) Scheme of VCS. The recombinant ‘movement‐defective CMV’ vector containing DxscFv (Δ3aH1Dx) multiplies only in the invading cell but should be confined within the cell. Non‐Tg, non‐transgenic plants. Selection of the best helper plant for VCS and evaluation of DxscFv produced by VCS The 3a protein was detected in 24 of 29 transgenic lines by Western blot analysis, and the expression levels varied in the T0 transformants (Table S1). From these transformants, we selected five T0 lines (No. 3, 7, 22, 24 and 33), which all produced high levels of the 3a protein. The self‐pollinated seedlings derived from line T0 were selected for kanamycin (Km) resistance to obtain T1 plants. The highest levels of the 3a protein were detected in No. 3 and No. 22 by Western blot analysis ( Figure 2a ). The segregation ratio of Km‐resistant to Km‐sensitive was 52 : 19 in the 3a‐transgenic (Tg) (T1) No. 3 plants and thus fit the theoretical ratio of 3 : 1; the χ 2 value was calculated to be 0.117 ( P < 0.05). For the 3aTg (T1) No. 22 plants, the segregation ratio of Km‐resistant and Km‐sensitive was 100 : 31, also fitting the theoretical ratio of 3 : 1; the χ 2 value was calculated to be 0.125 ( P < 0.05). These data suggest that No. 3 and No. 22 contained a single copy of the 3a gene. 2 Analysis of the helper plants (3aTg). (a) Western blot analysis of the 3a protein expressed in Nicotiana benthamiana leaves of kanamycin‐resistant T1 plants. N, non‐Tg; CMV, CMV‐infected plants. The 3a protein was detected in the 3aTg N. benthamiana plants (1.5 months old). The blots were treated with anti‐3a antibodies. (b) Accumulation of the CMV coat protein (CP) in the inoculated (I) and upper leaves (U) of the 3aTg (T1) plants. Δ3aCMV‐Y was used to inoculate five 3aTg lines (No. 3, 7, 22, 24 and 33) and detected 13 days post‐inoculation (dpi). The blots were treated with anti‐CP antibodies. H, healthy 3aTg plants; CMV, CMV‐infected plants. (c) Detection of DxscFv in the 3aTg (T1) No. 3, No. 22 and non‐Tg plants inoculated with YΔ3aH1Dx or Y3aH1Dx. (d) Western blot analysis of the 3a protein (first panel) and semiquantitative RT‐PCR analysis of the 3a transcript (second panel) in the 3aTg (T2) No. 3 healthy leaves ( n = 12). N, non‐Tg; C, the control plant expressing the 3a protein to identify the position of the 3a band. RT‐PCR of Ubi3 was included as a control (third panel). (e) Western blot analysis of DxscFv (first panel) and 3a (second panel) in the 3aTg (T2) No. 3 plants and non‐Tg inoculated with YΔ3aH1Dx. Semiquantitative RT‐PCR analysis shows the levels of the 3a transcript (third panel). RT‐PCR of Ubi3 (fourth panel) was as in (d). Proteins and RNAs were isolated from the 3aTg (T2) No. 3 leaves ( n = 12) inoculated with YΔ3aH1Dx. N, non‐Tg. To examine whether the 3a protein is functionally active, we inoculated the T1 plants (T0 line No. 3, 7, 22, 24 and 33) with the movement‐defective CMV‐Y (Δ3aCMV‐Y; R1 + R2 + YΔ3a). Among the T1 plants, line No. 3 was the earliest to develop systemic symptoms. Viral symptom appearance did not necessarily correlate with the levels of 3a protein expression. At 13 days post‐inoculation (dpi), the CMV CP was detected in the inoculated leaves of the T1 plants, 3aTg (T1) by Western blot analysis ( Figure 2b ). These results showed that the 3aTg plants were able to produce in trans the 3a protein that can mobilize Δ3aCMV‐Y. Then, the T1 plants of 3aTg No. 3 and No. 22 were tested for their ability to complement the systemic movement of YΔ3aH1Dx and to produce DxscFv. When the expression levels of DxscFv were monitored from 5 to 22 dpi using Western blot analysis, DxscFv was most efficiently produced by YΔ3aH1Dx in the No. 3 upper leaves at 13 dpi and produced to a much less extent (at non‐detectable levels) by YΔ3aH1Dx in the No. 22 upper leaves at 13 dpi ( Figure 2c ). On the other hand, DxscFv was not detected in the non‐transgenic (non‐Tg) plants inoculated with YΔ3aH1Dx ( Figure 2c ). For the non‐Tg plants inoculated with the virus expressing the 3a protein (Y3aH1Dx), DxscFv was detected at the highest level in the upper leaves at 5 dpi ( Figure 2c ). These results indicate that the 3aTg plants can complement the systemic movement of YΔ3aH1Dx and produce the recombinant antibody in a short period. Systemic infection of the movement‐deficient CMV vector in helper plants To investigate whether the ability of 3aTg No. 3 to complement YΔ3aH1Dx for the viral cell‐to‐cell movement is stable and remains high, we produced the next generation (T2) of 3aTg and confirmed 3a expression. Western blot analysis and RT‐PCR revealed that all the progeny 3aTg No. 3 plants expressed the 3a protein to a certain extent ( Figure 2d ). We then inoculated these 3aTg (T2) No. 3 plants with YΔ3aH1Dx. Western blot analysis showed that DxscFv was expressed in 94.9% of the inoculated plants (56 of 59 plants) at 13 dpi. However, DxscFv was detected at low levels only in 45% of the 3aTg (T2) No. 22 plants inoculated with YΔ3aH1Dx. Semiquantitative RT‐PCR analyses revealed that the 3a transgene was expressed at similar levels in all the 3aTg (T2) No. 3 individual plants ( Figure 2d ) and that the 3a transcripts were also found in all the YΔ3aH1Dx‐infected plants although the levels varied somewhat in the infected plants ( Figure 2e ). For the virus‐infected plants, the 3a protein was not well detected in the plants that accumulated DxscFv expressed through the CMV vector, but we detected 3a protein in the DxscFv‐non‐detected 3aTg (T2) No. 3 plants ( Figure 2d,e ). Therefore, there may be a negative correlation between DxscFv expression and 3a accumulation. Among these 3aTg plants, we eventually selected line No. 3 as the Tg helper (3aTg No. 3) for ideal VCS. To investigate whether DxscFv produced by the CMV vector was biochemically active, we then analysed the upper leaf sap from T1 plants (3aTg No. 3) inoculated with YΔ3aH1Dx. ELISA tests indicated that 3aTg No. 3 leaf sap containing DxscFv has an affinity with the dioxin antigen and that the virus‐derived DxscFv was indeed biochemically active (Figure S1a,b). 3aTg plants confine the movement‐defective CMV vector To confirm that 3aTg plants can confine the movement‐defective CMV vector, we first inoculated 3aTg plants with YΔ3aH1Dx or m2Δ3aH1Dx. When non‐Tg plants were inoculated with those viruses, we did not find any CMV infection ( Figure 3 ). By re‐inoculating non‐Tg plants with leaf sap from the upper leaves of the YΔ3aH1Dx‐infected plants, we confirmed that YΔ3aH1Dx and m2Δ3aH1Dx did not have the ability to systemically infect non‐Tg plants ( n = 120). As expected, m2Δ3aH1Dx could replicate in the 3aTg (T2) No. 3 plants and thus infect another 3aTg (T2) No. 3 plant ( Figure 3 ). 3 Western blot analysis of coat protein (CP) and DxscFv from non‐Tg plants inoculated with the recombinant virus‐infected tissues. Non‐Tg plants inoculated with YΔ3aH1Dx or m2Δ3aH1Dx were harvested 15 dpi and found not to be infected. As expected, CP and DxscFv were detected in the upper leaves of the 3aTg (T2) plants inoculated with m2Δ3aH1Dx. Y3aH1Dx‐infected non‐Tg tissues (5 dpi) were used as a positive control. I, inoculated leaves; U, upper leaves; H, healthy leaves. Tolerance to recombination in the 3a gene between CMV subgroups I and II Because RNA3 of the movement‐defective CMV vector (Δ3aCMV) still contains the 3a gene sequence even though it cannot produce the 3a protein because of the introduced stop codons ( Figure 1a ), there is a possibility of recombination between the RNA3 and the transgene transcript. However, we have never found any recombination in the Tg helper plants (3aTg No. 3) inoculated with the recombinant viruses (YΔ3aH1Dx or m2Δ3aH1Dx). By an RT‐PCR‐based method, the virus was monitored for recombination between the viral m2Δ3a and the Y3a sequence from the transgene to recapture any functional 3a gene. RNA from the upper leaves of all the 3aTg No. 3 plants infected with m2Δ3aH1Dx was first subjected to reverse transcription with primer a in Figure 4a that hybridizes downstream of the 3a gene in RNA3 to amplify only the viral 3a gene sequence. By subsequent PCRs using the primer pairs (primers b1 and c; b2 and c in Figure 4a ) that discriminate m2Δ3a and Y3a, we confirmed that the introduced stop codons were not replaced by the Y3a sequence to produce the intact 3a gene; a representative result is shown in Figure 4b . There was no amplification with primer pair b1 and c, but a 600‐bp fragment was amplified with primer pair b2 and c in all seven preparations. In addition, we frequently sequenced those PCR products and confirmed that there was no recombination between the Y3a transgene transcript and the m2Δ3a in viral RNA. For further confirmation, we co‐inoculated N. benthamiana with the Potato virus X vector expressing the Y3a gene (PVX‐Y3a) and either YΔ3aH1 (R1 + H1 + YΔ3a) or m2Δ3aH1 (R1 + H1 + m2Δ3a) and repeated serial passages (Figure S2). Sequencing of the PCR‐amplified 3a genes revealed that there was no recombination between m2Δ3aH1 and PVX‐Y3a even after the 10th passage (Table S2). We then conducted RT‐PCR analyses to detect any recombinant 3a genes in the upper leaves of the plants that were co‐inoculated with CMVs of subgroups I and II (Y and m2, respectively). Even in this mixed infections, the primers that were designed to amplify hybrid 3a genes were not able to amplify any recombinant molecules, suggesting that no recombination occurred in RNA3 in the mixed infection ( Figure 5 ). 4 Analysis of RNA recombination between the m2Δ3a sequence in viral RNA3 and the Y3a transgene transcript. (a) Strategy for RT‐PCR‐based method to monitor possible recombination between m2Δ3a and Y3a in the 3aTg No. 3 plants infected with m2Δ3aH1Dx. The first‐strand cDNA from RNA preparation was synthesized using primer a that hybridizes downstream of the 3a gene containing the stop codons (m2Δ3a) in CMV‐m2 RNA3. The subsequent PCR was carried out using either primer pair b1 and c for Y3a or b2 and c for m2Δ3a; primer c hybridizes with the 3′ conserved region between Y3a and m2‐3a. (b) Representative result of RT‐PCR. RT‐PCRs were performed using RNA isolated from the upper leaves of seven 3aTg No. 3 plants infected with m2Δ3aH1Dx. 5 PCR analysis to detect RNA3 recombination between CMV‐Y and CMV‐m2 in the mixed infection. (a) Positions of sequence‐specific primers. CMV‐Y and CMV‐m2 RNA3 are in yellow and orange, respectively. Boxes represent ORFs of the 3a and coat protein genes. Arrows indicate the primer position and orientation. (b) Agarose gel electrophoresis to analyse the RT‐PCR products. Nicotiana benthamiana plants were inoculated with CMV‐Y + CMV‐m2, CMV‐Y, or CMV‐m2 and RNAs were isolated from the upper, non‐inoculated leaves 14 dpi. Inocula and primer pairs were described in the Experimental procedures section. CMV, Cucumber mosaic virus. Expression levels of high‐quality recombinant proteins by VCS To confirm that our system actually works for production of a recombinant protein without observing any recombination, we then performed co‐inoculations with m2Δ3aH1Dx and PVX‐Y3a; we here switched the basic expression vector from YΔ3aH1 to m2Δ3aH1 to produce recombinant proteins without RNA recombination. The highest expression levels of DxscFv were found 7 dpi ( Figure 6a ). Sequencing analyses of the RT‐PCR products showed that m2Δ3aH1Dx did not capture an intact 3a by RNA recombination. When the expression levels of DxscFv in 3aTg (T2) No. 3 plants inoculated with m2Δ3aH1Dx were monitored from 5 to 17 dpi, the yield of DxscFv was the highest at 7 dpi and was estimated to be 1.9 mg/g total soluble protein (TSP) or 21 mg/kg fresh mass (FW) by Western blot using the twofold dilution series of control DxscFv ( Figure 6b ), which is comparable to the yield (30 mg/kg FW) of a scFv by the Tobacco mosaic virus (TMV) vector ( McCormick , 1999 ). The m2Δ3aH1Dx was faster and more efficiently spread than YΔ3aH1Dx ( Figure 6b ). Interestingly, the expression level of DxscFv was even higher in the 3aTg plants infected with m2Δ3aH1Dx than in the non‐transgenic plants infected with m2‐3aH1Dx ( Figure 6a,b ). To further demonstrate the use of this recombinant‐free VCS for pharmaceutical proteins, the encoding sequence of human IL1‐Ra was subcloned into the pCY2‐H1 vector (H1ILRa), and the in vitro transcript was mixed with R1 and m2Δ3a creating the virus m2Δ3aH1ILRa. When we inoculated 3aTg (T2) No. 3 plants with m2Δ3aH1ILRa, IL1‐Ra was detected in the inoculated leaves at 5 dpi ( Figure 6c ). By Western blot analysis using a twofold dilution series of control IL1‐Ra, the yield of IL1‐Ra was estimated to be 2.4 mg/g TSP. In addition to the authentic IL1‐Ra protein, a few more bands were detected; they may be modified in plant cells. 6 Analysis of the expression levels of the target protein in 3aTg plants infected with recombinant Cucumber mosaic virus (CMV) vectors. (a) Detection of DxscFv by Western blot analysis from the non‐Tg inoculated either with m2‐3aH1Dx or with m2Δ3aH1Dx + PVX‐Y3a. m2‐3aH1Dx, R1 + H1Dx + m2R3; m2Δ3aH1Dx, R1 + H1Dx + m2Δ3a. Leaf tissues were harvested at 7 dpi. Lane P contains 30 ng of DxscFv standard. I, inoculated leaves; U, upper leaves; H, healthy non‐Tg plants. The sample containing 60 μg of total soluble protein (TSP) was loaded on each lane. (b) Detection of DxscFv in the 3aTg (T2) No. 3 and the non‐Tg plants inoculated either with m2Δ3aH1Dx or with YΔ3aH1Dx (7 dpi). As a control, the m2‐3aH1Dx‐infected non‐Tg tissues (7 dpi) were used. YΔ3aH1Dx, R1 + H1Dx + YΔ3a. DxscFv standards (6.25, 12.5 and 25 ng) were included. TSP (40 μg) from leaf tissues was loaded on each lane. Lane I* contained 1/5 of lane I (8 μg). Based on the control, the expression level of the target protein was estimated to be 1.9 mg/g TSP. (c) Detection of IL1‐Ra in the 3aTg (T2) No. 3 and non‐Tg plants inoculated with m2Δ3aH1ILRa (5 dpi) and in the m2‐3aH1ILRa‐infected non‐Tg tissues (5 dpi). H1ILRa is the H1 vector containing the IL1‐Ra gene. m2Δ3aH1ILRa, R1 + H1ILRa + m2Δ3a. IL1‐Ra standards (6.25, 12.5 and 25 ng) with a molecular weight of about 17.8 kDa were included. Each lane contained 20 μg of TSP. Based on the control, the yield of IL1‐Ra was estimated to be 2.4 mg/g TSP. Cross‐protection in the 3aTg plants infected with the movement‐defective CMV vector against challenge inoculation with wild‐type CMV It should be noted that we produced recombinant proteins in the helper plants, which were already systemically infected by the vector virus, raising the possibility of cross‐protection. We thus hypothesized that although a wild‐type CMV is accidentally brought to vector‐infected helper plants, mixed infection will never be established because of cross‐protection. We then tested whether m2Δ3aH1Dx‐infected 3aTg No. 3 was cross‐protected against challenge inoculation with CMV‐m2 and whether RNA recombination in the 3a gene occurs if CMV‐m2, which contains almost the same 3a gene as that in m2Δ3aH1Dx, can systemically infect the vector‐infected helper plants. We first used m2Δ3aH1Dx to inoculate 3aTgNo. 3 plants, which produced DxscFv through the virus vector ( Figure 7a,b ). The same inoculated leaves were challenge inoculated with CMV‐m2 10 days after the m2Δ3aH1Dx inoculation. Seven days after the challenge inoculation, when the control plants had clear systemic symptoms, we harvested the upper leaves and used them as inoculum for sap‐inoculation of non‐Tg plants. We detected neither CMV CP by Western blots ( Figure 7c , upper panel) nor DxscFv RNA by RT‐PCR ( Figure 7c , lower panel) in the inoculated plants, suggesting that the intact 3a gene was not reconstructed in the infected leaves. Because m2Δ3aH1Dx consists of R1, H1Dx and m2Δ3a, if the intact 3a gene had been generated by recombination (or pseudorecombination) between m2Δ3aH1Dx and CMV‐m2, we would have detected DxscFv in the upper leaves. When challenged with CMV‐Y, a subgroup I CMV, instead of CMV‐m2, we also observed similar cross‐protection (data not shown), suggesting that the pCY2‐H1 vector lacking 2b is indeed effective against both subgroups. 7 Cross‐protection against a subgroup II Cucumber mosaic virus (CMV) (CMV‐m2) in the 3aTg plants infected with m2Δ3aH1Dx. (a) Scheme of cross‐protection experiments. (b) Western blot analysis to detect the accumulation of m2Δ3aH1Dx in upper leaves of the 3aTg (T2) No. 3 plants ( n = 4). The first three plants (plant no. 1–3) were challenge inoculated with CMV‐m2. H, healthy plants. (c) Western blot analysis (upper panel) of the CMV coat protein (CP), and RT‐PCR analysis (lower panel) of the DxscFv sequence in the non‐Tg plants inoculated with the leaf sap prepared as illustrated in (a). Ten days after inoculation with m2Δ3aH1Dx, the 3aTg plants No. 1–3, which had been infected with m2Δ3aH1Dx, were inoculated with CMV‐m2. Seven days after the challenge inoculation with CMV‐m2, the sap of the upper leaves of the 3aTg (T2) No. 3 plants was used to inoculate non‐Tg plants ( n = 3 for each inoculum plant). As a control (lane C), the accumulation of CP and DxscFv RNA was assessed in the m2‐3aH1Dx‐infected plants (17 dpi) by Western blot and RT‐PCR, respectively. The red arrow indicates the position for the intact DxscFv PCR product; the black arrow indicates a truncated insert because a long‐term infection sometimes causes internal deletion of the insert. Discussion Although the CP gene is expressed from RNA3 together with 3a, it is clearly involved in viral long‐distance movement but not absolutely necessary for cell‐to‐cell movement ( Palukaitis and García‐Arenal, 2003 ). This involvement suggests that the lack of 3a from RNA3 sufficiently disables cell‐to‐cell movement of the recombinant CMV. The 3a gene was destroyed in RNA3 by inserting stop codon(s) to prevent systemic movement of the viral vector. Meanwhile, helper Tg plants were produced to complement viral movement upon inoculation with the vector. The 3a protein provided in trans from a Tg plant has been reported to complement cell‐to‐cell movement of several CMV mutants containing defective 3a genes ( Kaplan , 1995 ). However, efficient systemic movement of the CMV mutants was not observed, as we described in the present report. On the other hand, we could find by intensive screening the 3aTg helper plants that can complement efficient systemic movement of the Δ3aCMV. In addition, as for the helper plant, our results suggest that it is very important to select the best 3aTg plant for vector movement. We initially speculated that the expression level of 3a would correlate with the rate of systemic movement of the vector. However, we found that a high level of 3a expression was not necessarily correlated with viral spread in the 3aTg plant. After the vector systemically spread in the 3aTg plant, the 3a protein suddenly decreased to nearly undetectable levels ( Figure 2e ), suggesting that the 3a protein might be degraded in the virus‐infected tissues. This result actually agrees with the observation that 3a accumulation rapidly changed in the CMV‐infected plants with certain stages of leaf development ( Itaya , 1998 ). When we compared No. 3 and No. 22, which had similar levels of 3a expression, the Δ3aCMV vector replicated much more efficiently and with more systemic movement in No. 3 than in No. 22, indicating that No. 3 was the best helper plant for the vector system. Our success can be attributed to intensive selection of helper plants that complement vector movement. We found a possible negative correlation between the 3a expression level and vector movement, which can be a hallmark for selecting ‘good’ helper plants expressing 3a. Therefore, the CMV vector is now dependent on the helper Tg plant for its movement. It will be interesting to analyse the turnover of the 3a protein in such 3aTg plants. We did not detect any recombination between the inoculated virus and the transgene in the helper plant. However, we cannot completely deny the potential for such recombination because the recombinant virus actually retains the 3a gene sequence except that it has the introduced stop codons. In the past, several intensive studies using natural CMV populations have never found any RNA3 recombination between two discrete CMV groups (subgroups I and II) ( Fraile , 1997 ; Lin , 2004 ; Bonnet , 2005 ). Learning from such observations, we decided to apply two kinds of the 3a genes to prevention of RNA recombination in our VCS and successfully to demonstrate prevention that no recombination indeed occurred between the two 3a genes even in mixed infections ( Figure 5 ). This lack of recombination is perhaps because of the low sequence homology (∼70%) between the two 3a genes. In addition, the two CMVs are localized separately in the mixed‐inoculated tissues ( Takeshita , 2004 ), suggesting that they perhaps repel each other. There must be a mechanism that the CMVs acquired during their evolutionary histories to discriminate the two subgroups so that no chimeric viruses would be generated between the two subgroups. We evidently copied this natural secret to create this unique viral vector system. Although we found that no RNA recombination occurred in RNA 3 between the two subgroups, there is still a concern about pseudorecombination of viral RNAs in mixed infection. What if m2Δ3aH1Dx acquired an intact RNA 3 molecule from a challenged CMV? Because a previous report showed that a CMV mutant lacking 2b provided strong cross‐protection against wild‐type CMVs (both subgroups) ( Ziebell , 2007 ), we tested whether the helper plant that is systemically infected with m2Δ3aH1Dx is cross‐protected against challenge inoculation with a wild‐type CMV. We observed complete cross‐protection in the helper plants infected with the vector lacking 2b, agreeing with the results of Ziebell (2007) . These results suggest that mixed infection to generate recombination (pseudorecombination) is not established in our system. Taken together, we conclude that our system has a very low probability for viral RNA recombination and pseudorecombination, even when the helper plant is accidentally challenged by a wild‐type CMV. Yields of virus‐produced proteins were not necessarily consistent between inoculated and non‐inoculated leaves. Perhaps protein yields in the inoculated leaves depend mainly upon the inoculation efficiency, while viral movement and localization largely affect protein production in non‐inoculated upper leaves. As for the expression levels of target protein, our system is not superior at this moment but is on a level comparable to those by the old‐type TMV vector ( McCormick , 1999 ). At present, Agrobacterium ‐mediated delivery of the TMV vector, so‐called ICON system, is the most rapid path from gene to recombinant protein and provides the highest levels of protein production ( Marillonnet , 2005 ; Giritch , 2006 ). However, in this work, we mostly focused on managing potential risks of viral vectors, sacrificing efficient expression of target protein to some extent; thus, improvement of the protein yield in our system will certainly be the next step in our research. Our first strategy for improving yield is to use synergistic effects between CMV and a potyvirus; it is well under way for VCS. VCS has the advantage of speed, stability, cost effectiveness and product safety. Thus, this plant vector system has the potential to produce recombinant proteins that cannot be efficiently produced by any other method. Our plant‐made antibodies can be stably stored, easily distributed and instantly prepared for diagnosis in an ELISA directly using the plant tissues. The system may be applicable for producing many other high‐quality, useful recombinant proteins such as a vaccine against an influenza virus for quick, large‐scale production to prevent a pandemic. Our system enables the CMV vector to produce recombinant proteins in plants without viral escape and recombination. Given that tobacco seedlings can grow in a short time in the greenhouse, this system is convenient to produce high‐quality recombinant proteins like DxscFv and IL1‐Ra. Our results thus demonstrate that there are certainly ways to diminish the risk of viral vectors and to take advantage of a viral vector to produce recombinant proteins in plants. Experimental procedures Plant material Plants had been grown at 23 °C under continuous illumination. Seeds of N. benthamiana plants were surface‐sterilized and sown on nutrient agar plates containing half‐strength Murashige‐Skoog (MS) salts, 1% sucrose, half‐strength B5 vitamin and 1% agar. After germination, the seedlings were placed in peat pots (Jiffy‐7), grown at 22 °C in a greenhouse and allowed to self‐pollinate. Virus strains and movement‐defective CMV vector construction The movement‐defective CMV vector was constructed based on the CMV‐based vector that was created from CMV‐Y (subgroup I) ( Matsuo , 2007 ). pCY3‐YΔ3a was constructed to introduce a stop codon in the 3a gene by subcloning in the original vector a PCR fragment amplified by a recombinant PCR using primer pairs containing appropriate restriction sites: Y3‐T7‐5Bm‐F (5′‐cgggatccattaatacgactcactataggtaatctaaccacctgt‐3′) and 3a‐Stop‐3‐R (5′‐ctggtaccttagaaagccat‐3′) 3a‐Stop‐5‐F (5′‐atggctttctaaggtaccag‐3′) and Y3‐3Hind‐R (5′‐aacaagcttcttatcatattcc‐3′). Similarly, pCM3‐m2Δ3a was constructed from RNA3 of CMV‐m2 (subgroup II) ( Takeshita , 2004 ) with two stop codons by subcloning a PCR fragment amplified by a recombinant PCR using primer pairs: m2‐5‐23 (5′‐cgctggaggtaatcttaccactttctttttc‐3′) and m2‐3‐stp3 (5′‐cgtcctactggtaccttagaaagccat‐3′) m2‐stp‐5 (5′‐aggtaccagtaggacgtaaactcaaca‐3′) and m2‐3‐Xb‐1000 (5′‐gactctagactcacatgtattt‐3′). The PCR‐amplified fragments were then exchanged with the corresponding sequences in the original construct to generate plasmids, pCY3‐YΔ3a and pCM3‐m2Δ3a. In vitro RNA transcripts and the viral inoculation were prepared as described previously ( Matsuo , 2007 ). The PVX vector was obtained from Dr Baulcombe ( Baulcombe , 1995 ). Cloning of the anti‐dioxin IgG gene to create the scFv construct in the viral vector A hybridoma cell line was produced from BALB/c mice after intraperitoneal injection of a phenoxatine derivative‐Bovine serum albumin (BSA) conjugate. ELISA was performed to identify positive hybridoma (Dx02‐I) cell lines producing monoclonal antibodies (IgG) to dioxin (2,3,4,7,8‐PeCDF). RNA was isolated from the hybridoma (Dx02‐I) cells, and the full‐length clone for DxscFv was obtained by PCR, according to the protocol of the mouse scFv module/recombinant phage antibody system (GE Healthcare, Piscataway, NJ, USA). The anti‐dioxin scFv gene (GenBank accession AB474005 ) was amplified from pCANTAB 5E‐Dx02‐I using PCR with the primer pairs DxscFv‐F1 (5′‐cgaggcctagaatgtacttgggactgagc‐3′) and DxscFv‐R1 (5′‐gcgacgcgttcaaagttcatccttatgatg‐3′) containing Stu I and Mlu I sites, respectively. This 837‐bp DxscFv gene was then cloned between the Stu I and Mlu I sites in pCY2‐H1, resulting in the pCY2‐H1:DxscFv plasmid. Cloning of the IL1‐Ra gene in the viral vector The cDNA for the mature form of the IL1‐Ra gene (GenBank accession NM000573 ) was amplified from the human cDNA library (TaKaRa Bio, Ohtsu, Japan) by PCR and cloned between the Stu I and Mlu I sites in the pCY2‐H1 vector. The in planta expression of IL1‐Ra by the viral vector was confirmed by Western blot analysis using commercially available antibodies against IL1‐Ra (R&D Systems, Minneapolis, MN, USA). Plant transformation and selection The full‐length 3a gene (GenBank accession D12499 ) was amplified from pCY3 by PCR, using the 3a‐specific primer pair Y‐3a‐F (5′‐gcactagtatggctttccaaggtaccag‐3′) and Y‐3a‐R (5′‐cgagctcctaaagaccgttaaccacct‐3′) containing a Spe I and Sac I site, respectively. This 855‐bp fragment was cloned into pGEM‐T (Promega, Madison, WI, USA). The Y3a gene was then subcloned into the binary vector pBE2113 ( Mitsuhara , 1996 ). The recombinant binary plasmid (pBE2113:Y3a) was transformed into Agrobacterium tumefaciens LBA4404, and Tg N. benthamiana plants were generated by the leaf disk method ( Horsch , 1986 ). Regenerated plantlets (T0) were allowed to self‐pollinate, and seeds (T1) were collected and tested for resistance to 250 mg Km/mL. Virus inoculation Carborundum‐dusted leaves of non‐Tg and 3aTg N. benthamiana seedlings (1.5 months old) were rub‐inoculated either with the synthesized RNA transcripts or sap‐containing virus. The sap inoculum was prepared by grinding leaf tissues from young systemically seedlings, grown in a greenhouse at 22 °C in 0.1 m potassium phosphate buffer (pH 8.0). SDS‐PAGE and Western blot analysis Leaf samples were homogenized in 3 volumes (mass/v) of phosphate‐buffered saline (PBS) containing 0.1% Triton X‐100 and a proteinase inhibitor. Crude leaf extracts were centrifuged at 6000 g for 10 min, and the supernatant was separated by electrophoresis on SDS‐12% polyacrylamide gels (TEFCO, Tokyo, Japan) under reducing conditions. For Western blot analysis, proteins from the SDS‐PAGE were electrophoretically transferred to Hybond P membranes (GE Healthcare). Membranes were blocked with 3% skim milk powder (mass/v) in PBS and probed with the anti‐3a polyclonal antibodies. Positive signals were detected using the ECL plus detection reagent (GE Healthcare). RNA extraction and semiquantitative RT‐PCR Total RNA from N. benthamiana leaves was isolated using TRIzol (Invitrogen, Carlsbad, CA, USA) and treated with RQ1 RNase‐free DNase (Promega) to remove any contaminating DNA. Prepared RNA (10 ng) was used to synthesize cDNAs using the SuperScript III One‐step RT‐PCR System with Platinum Taq High Fidelity (Invitrogen). Transcript levels of target genes relative to those of the ubiquitin gene ( Ubi3 ) as an internal control were used to normalize the data. The primer pair designed for CMV‐Y‐3a was CMV3a‐F1 (5′‐atggctttccaaggtaccag‐3′) and CMV3a‐840R1 (5′‐ctaaagaccgttaaccacctgc‐3′), and the pair for Ubi3 was Ubi3‐F (5′‐gattggtggtattggaactgtcc‐3′) and Ubi3‐R (5′‐gagcttcgtggtgcatctc‐3′). Purification of DxscFv The leaf extracts were homogenized in PBS buffer containing 0.1% Triton X‐100 and a proteinase inhibitor (Complete, EDTA‐free; Roche, Mannheim, Germany). The insoluble fraction was removed by centrifuging at 10000 g for 15 min, and the supernatants were then applied to mini‐columns prepacked with 1.5 mL nickel nitrilotriacetic acid gel (Ni‐NTA Superflow Agarose; QIAGEN, Valencia, CA, USA) and equilibrated with binding buffer [25 m m PB, 340 m m NaCl, 20% glycerol (pH 7.4)]. After loading the lysates, the columns were washed with a fivefold volume of washing buffer [25 m m PB, 340 m m NaCl, 20% glycerol, 20 m m imidazole (pH 7.4)] and eluted with elution buffer [25 m m PB, 340 m m NaCl, 20% glycerol, 300 m m imidazole (pH 7.4)]. The eluted protein solution was then dialysed against the binding buffer. The amount of purified protein was determined with a protein assay kit (Bio‐Rad, Hercules, CA, USA), using a normal mouse IgG Fab fragment as a standard. We used the DxscFv standard (KEM, Kyoto, Japan) produced by Escherichia coli and transgenic tobacco. RT‐PCR to monitor recombination between m2Δ3a and Y3a Total RNA was isolated from the upper leaves of 3aTg (T2) No. 3 inoculated with m2Δ3aH1Dx using TRIzol regent (Invitrogen) and then with RQ1 RNase‐Free DNase (Promega) to remove any contaminating DNA. Purified RNA (50 ng) was reverse‐transcribed with the primer that hybridizes downstream of the 3a gene containing the stop codons (m2Δ3a) in CMV‐m2 RNA3 but not the Y3a transgene transcript in a 10 μL of reaction volume with Reverse Transcriptase XL (AMV) (TaKaRa Bio) at 50 °C for 15 min. The synthesized first‐strand cDNA (1 μL) was used for the subsequent PCR amplification, which was carried out in a total volume of 50 μL of reaction mixture with Ex Taq‐HS for hot start (TaKaRa Bio) with the following cycling parameters: 25 cycles of 94 °C, 30 s; 65 °C, 30 s; 72 °C, 1 min. Five microlitres of the PCR product was analysed in an agarose gel. As shown in Figure 4b , the reverse primer (c: 5′‐ctaaagaccgttaaccacctg‐3′) hybridizes both Y3a and m2Δ3a. The forward primer that can specifically recognize the introduced stop codons in m2‐3a was b2 (5′‐aaggtaccagtaggacGtA‐3′), and the corresponding forward primer for Y3a was b1 (5′‐aaggtaccagtaggacTtT‐3′); two different bases (in capitals) discriminate Y3a and m2Δ3a. Detection of recombination between PVX‐3a and the 3a gene in RNA3 of CMV Non‐Tg N. benthamiana plants were co‐inoculated with the transcripts of PVX‐3a and either YΔ3aH1 or m2Δ3aH1. Serial viral passages were performed every 2 weeks. To confirm the sequence of the 3a genes in the CMV RNA3, RT‐PCR was performed to amplify the viral 3a gene using a 3a‐intervening primer pair that does not exist in the PVX‐Y3a sequence. RT‐PCR to detect RNA3 recombination between CMV‐Y and CMV‐m2 Nicotiana benthamiana plants were co‐inoculated with CMV‐Y and CMV‐m2. Two weeks after inoculation, total RNA was extracted from the systemically infected leaves. RT‐PCR was performed using a one step RNA PCR kit (AMV) (TaKaRa Bio) according to the supplier’s instructions. The RT reaction was performed at 50 °C for 30 min and terminated with a 2‐min incubation at 94 °C. PCR amplification (30 cycles) was carried out in programmed steps of 94 °C for 30 s, 55 °C for 30 s and 72 °C for 60 s. The primer pair for CMV‐Y is Y‐3R‐5 (5′‐atttatttcgttgtaca‐3′) and Y‐3a‐3 (5′‐aaaaccagatgtgttcc‐3′); the primer pair for CMV‐m2 is m2‐3R‐5 (5′‐tgtgtgttagttagtgt‐3′) and m2‐3a‐3 (5′‐aaccccagatgggaaat‐3′). The PCR products were then electrophoretically separated on 1% agarose gels. Acknowledgements This work was supported in part by grants from the Ministry of Economy, Trade and Industry (METI). We thank Drs D. Baulcombe and M. Takeshita for providing the PVX vector and CMV‐m2, respectively. References Aaziz , R. and Tepfer , M. ( 1999 ) Recombination between genomic RNAs of two Cucumoviruses under conditions of minimal selection pressure . Virology , 263 , 282 – 289 . Arntzen , C.J. ( 2008 ) Using tobacco to treat cancer . Science , 321 , 1052 – 1053 . Arntzen , C. , Plotkin , S. and Dodet , B. ( 2005 ) Plant‐derived vaccines and antibodies: potential and limitations . Vaccine , 23 , 1753 – 1756 . Baulcombe , D.C. , Chapman , S. and Santa Cruz , S. ( 1995 ) Jellyfish green fluorescent protein as a reporter for virus infections . Plant J. 7 , 1045 – 1053 . Bonnet , J. , Fraile , A. , Sacristán , S. , Malpica , J.M. and García‐Arenal , F. ( 2005 ) Role of recombination in the evolution of natural populations of Cucumber mosaic virus , a tripartite RNA plant virus . Virology , 332 , 359 – 368 . Canto , T. , Choi , S.K. and Palukaitis , P. ( 2001 ) A subpopulation of RNA 1 of Cucumber mosaic virus contains 3′ termini originating from RNAs 2 or 3 . J. Gen. Virol. 82 , 941 – 945 . Chen , Y.‐K. , Goldbach , R. and Prins , M. ( 2002 ) Inter‐ and intramolecular recombinations in the Cucumber mosaic virus genome related to adaptation to alstroemeria . J. Virol. 76 , 4119 – 4124 . Ding , B. , Li , Q. , Nguyen , L. , Palukaitis , P. and Lucas , W.J. ( 1995a ) Cucumber mosaic virus 3a protein potentiates cell‐to‐cell trafficking of CMV RNA in tobacco plants . Virology , 207 , 345 – 353 . Ding , S.‐W. , Li , W.‐X. and Symons , R.H. ( 1995b ) A novel naturally occurring hybrid gene encoded by a plant RNA virus facilitates long distance virus movement . EMBO J. 14 , 5762 – 5772 . Fernández‐Cuartero , B. , Burgyán , J. , Aranda , M.A. , Salánki , K. , Moriones , E. and García‐Arenal , F. ( 1994 ) Increase in the relative fitness of a plant virus RNA associated with its recombinant nature . Virology , 203 , 373 – 377 . Fraile , A. , Alonso‐Prados , J.L. , Aranda , M.A. , Bernal , J.J. , Malpica , J.M. and García‐Arenal , F. ( 1997 ) Genetic exchange by recombination or reassortment is infrequent in natural populations of a tripartite RNA plant virus . J. Virol. 71 , 934 – 940 . Giritch , A. , Marillonnet , S. , Engler , C. , Van Eldik , G. , Botterman , J. , Klimyuk , V. and Gleba , Y. ( 2006 ) Rapid high‐yield expression of full‐size IgG antibodies in plants coinfected with noncompeting viral vectors . Proc. Natl Acad. Sci. U.S.A. 103 , 14701 – 14706 . Gleba , Y. , Klimyuk , V. and Marillonnet , S. ( 2007 ) Viral vectors for the expression of proteins in plants . Curr. Opin. Biotechnol. 18 , 134 – 141 . Hiatt , A. and Pauly , M. ( 2006 ) Monoclonal antibodies from plants: a new speed record . Proc. Natl Acad. Sci. U.S.A. 103 , 14645 – 14646 . Horsch , R.B. , Klee , H.J. , Stachel , S. , Winans , S.C. , Nester , E.W. , Rogers , S.G. and Fraley , R.T. ( 1986 ) Analysis of Agrobacterium tumefaciens virulence mutants in leaf discs . Proc. Natl Acad. Sci. U.S.A. 83 , 2571 – 2575 . Itaya , A. , Woo , Y.M. , Masuta , C. , Bao , Y. , Nelson , R.S. and Ding , B. ( 1998 ) Developmental regulation of intercellular protein trafficking through plasmodesmata in tobacco leaf epidermis . Plant Physiol. 118 , 373 – 385 . Kaiser , J. ( 2008 ) Is the drought over for pharming? Science , 320 , 473 – 475 . Kaplan , I.B. , Shintaku , M.H. , Li , Q. , Zhang , L. , Marsh , L.E. and Palukaitis , P. ( 1995 ) Complementation of virus movement in transgenic tobacco expressing the Cucumber mosaic virus 3a gene . Virology , 209 , 188 – 199 . Lico , C. , Chen , Q. and Santi , L. ( 2008 ) Viral vectors for production of recombinant proteins in plants . J. Cell. Physiol. 216 , 366 – 377 . Lin , H.‐X. , Rubio , L. , Smythe , A.B. and Falk , B.W. ( 2004 ) Molecular population genetics of Cucumber mosaic virus in California: evidence for founder effects and reassortment . J. Virol. 78 , 6666 – 6675 . Ma , J.K.‐C. , Drake , P.M.W. and Christou , P. ( 2003 ) The production of recombinant pharmaceutical proteins in plants . Nat. Rev. Genet. 4 , 794 – 805 . Marillonnet , S. , Thoeringer , C. , Kandzia , R. , Klimyuk , V. and Gleba , Y. ( 2005 ) Systemic Agrobacterium tumefaciens ‐mediated transfection of viral replicons for efficient transient expression in plants . Nat. Biotechnol. 23 , 718 – 723 . Masuta , C. , Ueda , S. , Suzuki , M. and Uyeda , I. ( 1998 ) Evolution of a quadripartite hybrid virus by interspecific exchange and recombination between replicase components of two related tripartite RNA viruses . Proc. Natl Acad. Sci. U.S.A. 95 , 10487 – 10492 . Matsuo , K. , Hong , J.‐S. , Tabayashi , N. , Ito , A. , Masuta , C. and Matsumura , T. ( 2007 ) Development of Cucumber mosaic virus as a vector modifiable for different host species to produce therapeutic proteins . Planta , 225 , 277 – 286 . McCormick , A.A. , Kumagai , M.H. , Hanley , K. , Turpen , T.H. , Hakim , I. , Grill , L.K. , Tusé , D. , Levy , S. and Levy , R. ( 1999 ) Rapid production of specific vaccines for lymphoma by expression of the tumor‐derived single‐chain Fv epitopes in tobacco plants . Proc. Natl Acad. Sci. U.S.A. 96 , 703 – 708 . Mitsuhara , I. , Ugaki , M. , Hirochika , H. , Ohshima , M. , Murakami , T. , Gotoh , Y. , Katayose , Y. , Nakamura , S. , Honkura , R. , Nishimiya , S. , Ueno , K. , Mochizuki , A. , Tanimoto , H. , Tsugawa , H. , Otsuki , Y. and Ohashi , Y. ( 1996 ) Efficient promoter cassettes for enhanced expression of foreign genes in dicotyledonous and monocotyledonous plants . Plant Cell Physiol. 37 , 49 – 59 . Palukaitis , P. and García‐Arenal , F. ( 2003 ) Cucumoviruses . Adv. Virus Res. 62 , 241 – 323 . Roossinck , M.J. ( 2002 ) Evolutionary history of Cucumber mosaic virus deduced by phylogenetic analyses . J. Virol. 76 , 3382 – 3387 . Shi , B.‐J. , Miller , J. , Symons , R.H. and Palukaitis , P. ( 2003 ) The 2b protein of cucumoviruses has a role in promoting the cell‐to‐cell movement of pseudorecombinant viruses . Mol. Plant Microbe Interact. 16 , 261 – 267 . Soards , A.J. , Murphy , A.M. , Palukaitis , P. and Carr , J.P. ( 2002 ) Virulence and differential local and systemic spread of Cucumber mosaic virus in tobacco are affected by the CMV 2b protein . Mol. Plant Microbe Interact. 15 , 647 – 653 . Suzuki , M. , Kuwata , S. , Kataoka , J. , Masuta , C. , Nitta , N. and Takanami , Y. ( 1991 ) Functional analysis of deletion mutants of Cucumber mosaic virus RNA3 using an in vitro transcription system . Virology , 183 , 106 – 113 . Takeshita , M. , Shigemune , N. , Kikuhara , K. , Furuya , N. and Takanami , Y. ( 2004 ) Spatial analysis for exclusive interactions between subgroups I and II of Cucumber mosaic virus in cowpea . Virology , 328 , 45 – 51 . Taliansky , M.E. and García‐Arenal , F. ( 1995 ) Role of cucumovirus capsid protein in long‐distance movement within the infected plant . J. Virol. 69 , 916 – 922 . De Wispelaere , M. , Gaubert , S. , Trouilloud , S. , Belin , C. and Tepfer , M. ( 2005 ) A map of the diversity of RNA3 recombinants appearing in plants infected with Cucumber mosaic virus and Tomato aspermy virus . Virology , 331 , 117 – 127 . Ziebell , H. , Payne , T. , Berry , J.O. , Walsh , J.A. and Carr , J.P. ( 2007 ) A cucumber mosaic virus mutant lacking the 2b counter‐defence protein gene provides protection against wild‐type strains . J. Gen. Virol. 88 , 2862 – 2871 . http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Plant Biotechnology Journal Wiley

Risk‐managed production of bioactive recombinant proteins using a novel plant virus vector with a helper plant to complement viral systemic movement

Loading next page...
 
/lp/wiley/risk-managed-production-of-bioactive-recombinant-proteins-using-a-k0Z1eqQIwH

References (36)

Publisher
Wiley
Copyright
"Copyright © 2011 Wiley Subscription Services, Inc., A Wiley Company"
eISSN
1467-7652
DOI
10.1111/j.1467-7652.2010.00529.x
pmid
20492549
Publisher site
See Article on Publisher Site

Abstract

Introduction Plants represent an advantageous system for expressing recombinant proteins in terms of cost effectiveness and the ease of scaling up production ( Ma , 2003 ; Arntzen , 2005 ; Kaiser, 2008 ). However, the drawback for the genetic engineering of plants has always been the very long development period ( Hiatt and Pauly, 2006 ). The ideal expression system for recombinant proteins including vaccines and antibodies over short periods of time would be a viral vector system ( Hiatt and Pauly, 2006 ; Arntzen, 2008 ). Although the high potential of plant viral vectors for heterologous gene expression in plants has been well recognized ( Gleba , 2007 ; Lico , 2008 ), the safest course is to limit the use of viral vectors to when other methods (i.e. microorganisms, animal cells) to produce the target protein do not work because of the great concern of using recombinant viruses in open fields. Seeking a possible solution to this problem, we here constructed a novel Cucumber mosaic virus (CMV) vector as a viral confinement system (VCS) to produce a high‐quality recombinant protein in plants. CMV belongs to the genus Cucumovirus and has a very broad host range of more than 1200 species. CMV is divided into two subgroups I and II, and subgroup I is further divided into IA and IB ( Roossinck, 2002 ). CMV has tripartite, plus‐sense RNAs designated RNAs 1, 2 and 3 in decreasing order of molecular weight. RNAs 1 and 2 encode replicase proteins (1a and 2a, respectively), and RNA3 encodes the movement protein (3a), which allows CMV to move to neighbouring cells ( Suzuki , 1991 ; Ding , 1995a ). The coat protein (CP) is translated from a subgenomic RNA, RNA4, derived from RNA3, and has been found to be necessary for long‐distance movement ( Taliansky and García‐Arenal, 1995 ). RNA4A is a subgenomic RNA generated from the 3′ end of RNA2 and encodes a multifunctional protein, the 2b protein (2b), which has been shown to be an RNA silencing suppressor (RSS). 2b is not itself the movement protein but can enhance the viral cell‐to‐cell and long‐distance movement by increasing viral RNAs perhaps through its RSS ability ( Ding , 1995b ; Soards , 2002 ; Shi , 2003 ). We previously developed CMV as a gene expression vector to produce a human cytokine, acidic fibroblast growth factor (aFGF), in plants ( Matsuo , 2007 ). By modifying the original vector, here we tried to develop a novel vector that cannot move independently from cell‐to‐cell and can infect systemically only with the aid of a transgenic helper plant that expresses the 3a gene encoding the viral movement protein. We call this system the VCS. With cucumoviruses, mixed infection often results in RNA recombination between CMV isolates ( Chen , 2002 ; Lin , 2004 ; Bonnet , 2005 ), within a CMV isolate ( Canto ,2001 ), and between CMV and another cucumovirus, Tomato aspermy virus (TAV) ( Masuta ,1998 ; Aaziz and Tepfer,1999 ; de Wispelaere ,2005 ). In addition, recombination can occur to produce common 3′ ends within a pseudorecombinant virus containing RNAs 1 and 2 from CMV and RNA3 from TAV ( Fernández‐Cuartero ,1994 ; Shi ,2003 ). Towards the development of a more useful system, we sought to diminish the risk of CMV recombination. By using a combination of two kinds of the 3a genes, one from CMV subgroup I and the other from subgroup II ( Roossinck,2002 ), we could prevent RNA recombination in RNA3. Possible mechanisms underlying viral systemic movement and the recombination‐free nature of the VCS are also discussed. To demonstrate the expression of high‐quality recombinant proteins by our VCS, we here describe the expression of a functional single‐chain variable fragment against dioxins (DxscFv), which are persistent organic pollutants, and human interleukin‐1 receptor antagonist (IL1‐Ra), a rheumatoid arthritis agent. The protein expression efficiency was compared to previously published systems. Results Novel expression system to confine the CMV vector in the infected cells The CMV vector constructs used in this study are shown in Figure 1a . RNAs 1–3 are all necessary for creation of the virus and thus infection, although foreign genes are inserted in the vector clone of RNA2 (pCY2‐H1). When the virus moves in inoculated plants, expression of the 3a and the CP genes on RNA3 are necessary for cell‐to‐cell movement and long‐distant movement, respectively. The CP for encapsidation will be synthesized by RNA4 derived from RNA3 of the inoculated CMV vector. To destroy the 3a gene, a stop codon was inserted in the 3a gene of RNA3 at the 4th position to create pCY3‐YΔ3a (transcript, YΔ3a), which has the 3a sequence of subgroup I CMV‐Y ( Figure 1a ). We also engineered pCM3‐m2Δ3a (transcript, m2Δ3a), which encodes the subgroup II 3a protein containing two stop codons in the 3a gene at the 4th and 10th position ( Figure 1a ). Next, we created transgenic Nicotiana benthamiana plants expressing the 3a gene of CMV‐Y by using the conventional plant expression vector (pBE2113) ( Figure 1b ). As shown in Figure 1c , the entire scheme for VCS is based on complementation of the movement‐defective CMV vector to produce a recombinant protein with the helper plants, which supply the movement protein (3a) in trans to enable cell‐to‐cell movement of the virus. To demonstrate VCS, we selected DxscFv as a target protein. The DxscFv gene (837 bp) was then inserted into the pCY2‐H1 vector to create pCY2‐H1:DxscFv. As illustrated in Figure 1a , the viral inoculum consisted of the transcripts from pCY1 (R1), pCY2‐H1:DxscFv (H1Dx) and pCY3‐YΔ3a (YΔ3a) or pCM3‐m2Δ3a (m2Δ3a) to generate the recombinant virus, Δ3aH1Dx (YΔ3aH1Dx or m2Δ3aH1Dx). For another target IL1‐Ra, the DxscFv gene was simply replaced with the gene (480 bp) for the mature form of IL1‐Ra. 1 Viral confinement system (VCS) to produce recombinant proteins in helper plants. (a) Constructs of the movement‐defective Cucumber mosaic virus (CMV) vector to express the DxscFv gene. The 3a genes of CMV‐Y (subgroup I) and CMV‐m2 (subgroup II) have one and two stop codons immediately after the initiation codon, respectively. T7p, T7 promoter; 1a, helicase; 2a, RNA polymerase; 2b, silencing suppressor; 3a, movement protein; CP, coat protein; MCS, multiple cloning site; SP, signal peptide; VH, variable region of H chain; VL, variable region of L chain; 6His, 6 × his tag; KDEL, retention signal. (b) Constructs of a binary vector to produce helper plants expressing the 3a gene of CMV‐Y (subgroup I). RB and LB, binary right and left borders; Pnos, promoter of nopaline synthase; NPTII, neomycin phosphotransferase gene; Tnos, polyadenylation signal of nopline synthase; P35S, Cauliflower mosaic virus 35S promoter. (c) Scheme of VCS. The recombinant ‘movement‐defective CMV’ vector containing DxscFv (Δ3aH1Dx) multiplies only in the invading cell but should be confined within the cell. Non‐Tg, non‐transgenic plants. Selection of the best helper plant for VCS and evaluation of DxscFv produced by VCS The 3a protein was detected in 24 of 29 transgenic lines by Western blot analysis, and the expression levels varied in the T0 transformants (Table S1). From these transformants, we selected five T0 lines (No. 3, 7, 22, 24 and 33), which all produced high levels of the 3a protein. The self‐pollinated seedlings derived from line T0 were selected for kanamycin (Km) resistance to obtain T1 plants. The highest levels of the 3a protein were detected in No. 3 and No. 22 by Western blot analysis ( Figure 2a ). The segregation ratio of Km‐resistant to Km‐sensitive was 52 : 19 in the 3a‐transgenic (Tg) (T1) No. 3 plants and thus fit the theoretical ratio of 3 : 1; the χ 2 value was calculated to be 0.117 ( P < 0.05). For the 3aTg (T1) No. 22 plants, the segregation ratio of Km‐resistant and Km‐sensitive was 100 : 31, also fitting the theoretical ratio of 3 : 1; the χ 2 value was calculated to be 0.125 ( P < 0.05). These data suggest that No. 3 and No. 22 contained a single copy of the 3a gene. 2 Analysis of the helper plants (3aTg). (a) Western blot analysis of the 3a protein expressed in Nicotiana benthamiana leaves of kanamycin‐resistant T1 plants. N, non‐Tg; CMV, CMV‐infected plants. The 3a protein was detected in the 3aTg N. benthamiana plants (1.5 months old). The blots were treated with anti‐3a antibodies. (b) Accumulation of the CMV coat protein (CP) in the inoculated (I) and upper leaves (U) of the 3aTg (T1) plants. Δ3aCMV‐Y was used to inoculate five 3aTg lines (No. 3, 7, 22, 24 and 33) and detected 13 days post‐inoculation (dpi). The blots were treated with anti‐CP antibodies. H, healthy 3aTg plants; CMV, CMV‐infected plants. (c) Detection of DxscFv in the 3aTg (T1) No. 3, No. 22 and non‐Tg plants inoculated with YΔ3aH1Dx or Y3aH1Dx. (d) Western blot analysis of the 3a protein (first panel) and semiquantitative RT‐PCR analysis of the 3a transcript (second panel) in the 3aTg (T2) No. 3 healthy leaves ( n = 12). N, non‐Tg; C, the control plant expressing the 3a protein to identify the position of the 3a band. RT‐PCR of Ubi3 was included as a control (third panel). (e) Western blot analysis of DxscFv (first panel) and 3a (second panel) in the 3aTg (T2) No. 3 plants and non‐Tg inoculated with YΔ3aH1Dx. Semiquantitative RT‐PCR analysis shows the levels of the 3a transcript (third panel). RT‐PCR of Ubi3 (fourth panel) was as in (d). Proteins and RNAs were isolated from the 3aTg (T2) No. 3 leaves ( n = 12) inoculated with YΔ3aH1Dx. N, non‐Tg. To examine whether the 3a protein is functionally active, we inoculated the T1 plants (T0 line No. 3, 7, 22, 24 and 33) with the movement‐defective CMV‐Y (Δ3aCMV‐Y; R1 + R2 + YΔ3a). Among the T1 plants, line No. 3 was the earliest to develop systemic symptoms. Viral symptom appearance did not necessarily correlate with the levels of 3a protein expression. At 13 days post‐inoculation (dpi), the CMV CP was detected in the inoculated leaves of the T1 plants, 3aTg (T1) by Western blot analysis ( Figure 2b ). These results showed that the 3aTg plants were able to produce in trans the 3a protein that can mobilize Δ3aCMV‐Y. Then, the T1 plants of 3aTg No. 3 and No. 22 were tested for their ability to complement the systemic movement of YΔ3aH1Dx and to produce DxscFv. When the expression levels of DxscFv were monitored from 5 to 22 dpi using Western blot analysis, DxscFv was most efficiently produced by YΔ3aH1Dx in the No. 3 upper leaves at 13 dpi and produced to a much less extent (at non‐detectable levels) by YΔ3aH1Dx in the No. 22 upper leaves at 13 dpi ( Figure 2c ). On the other hand, DxscFv was not detected in the non‐transgenic (non‐Tg) plants inoculated with YΔ3aH1Dx ( Figure 2c ). For the non‐Tg plants inoculated with the virus expressing the 3a protein (Y3aH1Dx), DxscFv was detected at the highest level in the upper leaves at 5 dpi ( Figure 2c ). These results indicate that the 3aTg plants can complement the systemic movement of YΔ3aH1Dx and produce the recombinant antibody in a short period. Systemic infection of the movement‐deficient CMV vector in helper plants To investigate whether the ability of 3aTg No. 3 to complement YΔ3aH1Dx for the viral cell‐to‐cell movement is stable and remains high, we produced the next generation (T2) of 3aTg and confirmed 3a expression. Western blot analysis and RT‐PCR revealed that all the progeny 3aTg No. 3 plants expressed the 3a protein to a certain extent ( Figure 2d ). We then inoculated these 3aTg (T2) No. 3 plants with YΔ3aH1Dx. Western blot analysis showed that DxscFv was expressed in 94.9% of the inoculated plants (56 of 59 plants) at 13 dpi. However, DxscFv was detected at low levels only in 45% of the 3aTg (T2) No. 22 plants inoculated with YΔ3aH1Dx. Semiquantitative RT‐PCR analyses revealed that the 3a transgene was expressed at similar levels in all the 3aTg (T2) No. 3 individual plants ( Figure 2d ) and that the 3a transcripts were also found in all the YΔ3aH1Dx‐infected plants although the levels varied somewhat in the infected plants ( Figure 2e ). For the virus‐infected plants, the 3a protein was not well detected in the plants that accumulated DxscFv expressed through the CMV vector, but we detected 3a protein in the DxscFv‐non‐detected 3aTg (T2) No. 3 plants ( Figure 2d,e ). Therefore, there may be a negative correlation between DxscFv expression and 3a accumulation. Among these 3aTg plants, we eventually selected line No. 3 as the Tg helper (3aTg No. 3) for ideal VCS. To investigate whether DxscFv produced by the CMV vector was biochemically active, we then analysed the upper leaf sap from T1 plants (3aTg No. 3) inoculated with YΔ3aH1Dx. ELISA tests indicated that 3aTg No. 3 leaf sap containing DxscFv has an affinity with the dioxin antigen and that the virus‐derived DxscFv was indeed biochemically active (Figure S1a,b). 3aTg plants confine the movement‐defective CMV vector To confirm that 3aTg plants can confine the movement‐defective CMV vector, we first inoculated 3aTg plants with YΔ3aH1Dx or m2Δ3aH1Dx. When non‐Tg plants were inoculated with those viruses, we did not find any CMV infection ( Figure 3 ). By re‐inoculating non‐Tg plants with leaf sap from the upper leaves of the YΔ3aH1Dx‐infected plants, we confirmed that YΔ3aH1Dx and m2Δ3aH1Dx did not have the ability to systemically infect non‐Tg plants ( n = 120). As expected, m2Δ3aH1Dx could replicate in the 3aTg (T2) No. 3 plants and thus infect another 3aTg (T2) No. 3 plant ( Figure 3 ). 3 Western blot analysis of coat protein (CP) and DxscFv from non‐Tg plants inoculated with the recombinant virus‐infected tissues. Non‐Tg plants inoculated with YΔ3aH1Dx or m2Δ3aH1Dx were harvested 15 dpi and found not to be infected. As expected, CP and DxscFv were detected in the upper leaves of the 3aTg (T2) plants inoculated with m2Δ3aH1Dx. Y3aH1Dx‐infected non‐Tg tissues (5 dpi) were used as a positive control. I, inoculated leaves; U, upper leaves; H, healthy leaves. Tolerance to recombination in the 3a gene between CMV subgroups I and II Because RNA3 of the movement‐defective CMV vector (Δ3aCMV) still contains the 3a gene sequence even though it cannot produce the 3a protein because of the introduced stop codons ( Figure 1a ), there is a possibility of recombination between the RNA3 and the transgene transcript. However, we have never found any recombination in the Tg helper plants (3aTg No. 3) inoculated with the recombinant viruses (YΔ3aH1Dx or m2Δ3aH1Dx). By an RT‐PCR‐based method, the virus was monitored for recombination between the viral m2Δ3a and the Y3a sequence from the transgene to recapture any functional 3a gene. RNA from the upper leaves of all the 3aTg No. 3 plants infected with m2Δ3aH1Dx was first subjected to reverse transcription with primer a in Figure 4a that hybridizes downstream of the 3a gene in RNA3 to amplify only the viral 3a gene sequence. By subsequent PCRs using the primer pairs (primers b1 and c; b2 and c in Figure 4a ) that discriminate m2Δ3a and Y3a, we confirmed that the introduced stop codons were not replaced by the Y3a sequence to produce the intact 3a gene; a representative result is shown in Figure 4b . There was no amplification with primer pair b1 and c, but a 600‐bp fragment was amplified with primer pair b2 and c in all seven preparations. In addition, we frequently sequenced those PCR products and confirmed that there was no recombination between the Y3a transgene transcript and the m2Δ3a in viral RNA. For further confirmation, we co‐inoculated N. benthamiana with the Potato virus X vector expressing the Y3a gene (PVX‐Y3a) and either YΔ3aH1 (R1 + H1 + YΔ3a) or m2Δ3aH1 (R1 + H1 + m2Δ3a) and repeated serial passages (Figure S2). Sequencing of the PCR‐amplified 3a genes revealed that there was no recombination between m2Δ3aH1 and PVX‐Y3a even after the 10th passage (Table S2). We then conducted RT‐PCR analyses to detect any recombinant 3a genes in the upper leaves of the plants that were co‐inoculated with CMVs of subgroups I and II (Y and m2, respectively). Even in this mixed infections, the primers that were designed to amplify hybrid 3a genes were not able to amplify any recombinant molecules, suggesting that no recombination occurred in RNA3 in the mixed infection ( Figure 5 ). 4 Analysis of RNA recombination between the m2Δ3a sequence in viral RNA3 and the Y3a transgene transcript. (a) Strategy for RT‐PCR‐based method to monitor possible recombination between m2Δ3a and Y3a in the 3aTg No. 3 plants infected with m2Δ3aH1Dx. The first‐strand cDNA from RNA preparation was synthesized using primer a that hybridizes downstream of the 3a gene containing the stop codons (m2Δ3a) in CMV‐m2 RNA3. The subsequent PCR was carried out using either primer pair b1 and c for Y3a or b2 and c for m2Δ3a; primer c hybridizes with the 3′ conserved region between Y3a and m2‐3a. (b) Representative result of RT‐PCR. RT‐PCRs were performed using RNA isolated from the upper leaves of seven 3aTg No. 3 plants infected with m2Δ3aH1Dx. 5 PCR analysis to detect RNA3 recombination between CMV‐Y and CMV‐m2 in the mixed infection. (a) Positions of sequence‐specific primers. CMV‐Y and CMV‐m2 RNA3 are in yellow and orange, respectively. Boxes represent ORFs of the 3a and coat protein genes. Arrows indicate the primer position and orientation. (b) Agarose gel electrophoresis to analyse the RT‐PCR products. Nicotiana benthamiana plants were inoculated with CMV‐Y + CMV‐m2, CMV‐Y, or CMV‐m2 and RNAs were isolated from the upper, non‐inoculated leaves 14 dpi. Inocula and primer pairs were described in the Experimental procedures section. CMV, Cucumber mosaic virus. Expression levels of high‐quality recombinant proteins by VCS To confirm that our system actually works for production of a recombinant protein without observing any recombination, we then performed co‐inoculations with m2Δ3aH1Dx and PVX‐Y3a; we here switched the basic expression vector from YΔ3aH1 to m2Δ3aH1 to produce recombinant proteins without RNA recombination. The highest expression levels of DxscFv were found 7 dpi ( Figure 6a ). Sequencing analyses of the RT‐PCR products showed that m2Δ3aH1Dx did not capture an intact 3a by RNA recombination. When the expression levels of DxscFv in 3aTg (T2) No. 3 plants inoculated with m2Δ3aH1Dx were monitored from 5 to 17 dpi, the yield of DxscFv was the highest at 7 dpi and was estimated to be 1.9 mg/g total soluble protein (TSP) or 21 mg/kg fresh mass (FW) by Western blot using the twofold dilution series of control DxscFv ( Figure 6b ), which is comparable to the yield (30 mg/kg FW) of a scFv by the Tobacco mosaic virus (TMV) vector ( McCormick , 1999 ). The m2Δ3aH1Dx was faster and more efficiently spread than YΔ3aH1Dx ( Figure 6b ). Interestingly, the expression level of DxscFv was even higher in the 3aTg plants infected with m2Δ3aH1Dx than in the non‐transgenic plants infected with m2‐3aH1Dx ( Figure 6a,b ). To further demonstrate the use of this recombinant‐free VCS for pharmaceutical proteins, the encoding sequence of human IL1‐Ra was subcloned into the pCY2‐H1 vector (H1ILRa), and the in vitro transcript was mixed with R1 and m2Δ3a creating the virus m2Δ3aH1ILRa. When we inoculated 3aTg (T2) No. 3 plants with m2Δ3aH1ILRa, IL1‐Ra was detected in the inoculated leaves at 5 dpi ( Figure 6c ). By Western blot analysis using a twofold dilution series of control IL1‐Ra, the yield of IL1‐Ra was estimated to be 2.4 mg/g TSP. In addition to the authentic IL1‐Ra protein, a few more bands were detected; they may be modified in plant cells. 6 Analysis of the expression levels of the target protein in 3aTg plants infected with recombinant Cucumber mosaic virus (CMV) vectors. (a) Detection of DxscFv by Western blot analysis from the non‐Tg inoculated either with m2‐3aH1Dx or with m2Δ3aH1Dx + PVX‐Y3a. m2‐3aH1Dx, R1 + H1Dx + m2R3; m2Δ3aH1Dx, R1 + H1Dx + m2Δ3a. Leaf tissues were harvested at 7 dpi. Lane P contains 30 ng of DxscFv standard. I, inoculated leaves; U, upper leaves; H, healthy non‐Tg plants. The sample containing 60 μg of total soluble protein (TSP) was loaded on each lane. (b) Detection of DxscFv in the 3aTg (T2) No. 3 and the non‐Tg plants inoculated either with m2Δ3aH1Dx or with YΔ3aH1Dx (7 dpi). As a control, the m2‐3aH1Dx‐infected non‐Tg tissues (7 dpi) were used. YΔ3aH1Dx, R1 + H1Dx + YΔ3a. DxscFv standards (6.25, 12.5 and 25 ng) were included. TSP (40 μg) from leaf tissues was loaded on each lane. Lane I* contained 1/5 of lane I (8 μg). Based on the control, the expression level of the target protein was estimated to be 1.9 mg/g TSP. (c) Detection of IL1‐Ra in the 3aTg (T2) No. 3 and non‐Tg plants inoculated with m2Δ3aH1ILRa (5 dpi) and in the m2‐3aH1ILRa‐infected non‐Tg tissues (5 dpi). H1ILRa is the H1 vector containing the IL1‐Ra gene. m2Δ3aH1ILRa, R1 + H1ILRa + m2Δ3a. IL1‐Ra standards (6.25, 12.5 and 25 ng) with a molecular weight of about 17.8 kDa were included. Each lane contained 20 μg of TSP. Based on the control, the yield of IL1‐Ra was estimated to be 2.4 mg/g TSP. Cross‐protection in the 3aTg plants infected with the movement‐defective CMV vector against challenge inoculation with wild‐type CMV It should be noted that we produced recombinant proteins in the helper plants, which were already systemically infected by the vector virus, raising the possibility of cross‐protection. We thus hypothesized that although a wild‐type CMV is accidentally brought to vector‐infected helper plants, mixed infection will never be established because of cross‐protection. We then tested whether m2Δ3aH1Dx‐infected 3aTg No. 3 was cross‐protected against challenge inoculation with CMV‐m2 and whether RNA recombination in the 3a gene occurs if CMV‐m2, which contains almost the same 3a gene as that in m2Δ3aH1Dx, can systemically infect the vector‐infected helper plants. We first used m2Δ3aH1Dx to inoculate 3aTgNo. 3 plants, which produced DxscFv through the virus vector ( Figure 7a,b ). The same inoculated leaves were challenge inoculated with CMV‐m2 10 days after the m2Δ3aH1Dx inoculation. Seven days after the challenge inoculation, when the control plants had clear systemic symptoms, we harvested the upper leaves and used them as inoculum for sap‐inoculation of non‐Tg plants. We detected neither CMV CP by Western blots ( Figure 7c , upper panel) nor DxscFv RNA by RT‐PCR ( Figure 7c , lower panel) in the inoculated plants, suggesting that the intact 3a gene was not reconstructed in the infected leaves. Because m2Δ3aH1Dx consists of R1, H1Dx and m2Δ3a, if the intact 3a gene had been generated by recombination (or pseudorecombination) between m2Δ3aH1Dx and CMV‐m2, we would have detected DxscFv in the upper leaves. When challenged with CMV‐Y, a subgroup I CMV, instead of CMV‐m2, we also observed similar cross‐protection (data not shown), suggesting that the pCY2‐H1 vector lacking 2b is indeed effective against both subgroups. 7 Cross‐protection against a subgroup II Cucumber mosaic virus (CMV) (CMV‐m2) in the 3aTg plants infected with m2Δ3aH1Dx. (a) Scheme of cross‐protection experiments. (b) Western blot analysis to detect the accumulation of m2Δ3aH1Dx in upper leaves of the 3aTg (T2) No. 3 plants ( n = 4). The first three plants (plant no. 1–3) were challenge inoculated with CMV‐m2. H, healthy plants. (c) Western blot analysis (upper panel) of the CMV coat protein (CP), and RT‐PCR analysis (lower panel) of the DxscFv sequence in the non‐Tg plants inoculated with the leaf sap prepared as illustrated in (a). Ten days after inoculation with m2Δ3aH1Dx, the 3aTg plants No. 1–3, which had been infected with m2Δ3aH1Dx, were inoculated with CMV‐m2. Seven days after the challenge inoculation with CMV‐m2, the sap of the upper leaves of the 3aTg (T2) No. 3 plants was used to inoculate non‐Tg plants ( n = 3 for each inoculum plant). As a control (lane C), the accumulation of CP and DxscFv RNA was assessed in the m2‐3aH1Dx‐infected plants (17 dpi) by Western blot and RT‐PCR, respectively. The red arrow indicates the position for the intact DxscFv PCR product; the black arrow indicates a truncated insert because a long‐term infection sometimes causes internal deletion of the insert. Discussion Although the CP gene is expressed from RNA3 together with 3a, it is clearly involved in viral long‐distance movement but not absolutely necessary for cell‐to‐cell movement ( Palukaitis and García‐Arenal, 2003 ). This involvement suggests that the lack of 3a from RNA3 sufficiently disables cell‐to‐cell movement of the recombinant CMV. The 3a gene was destroyed in RNA3 by inserting stop codon(s) to prevent systemic movement of the viral vector. Meanwhile, helper Tg plants were produced to complement viral movement upon inoculation with the vector. The 3a protein provided in trans from a Tg plant has been reported to complement cell‐to‐cell movement of several CMV mutants containing defective 3a genes ( Kaplan , 1995 ). However, efficient systemic movement of the CMV mutants was not observed, as we described in the present report. On the other hand, we could find by intensive screening the 3aTg helper plants that can complement efficient systemic movement of the Δ3aCMV. In addition, as for the helper plant, our results suggest that it is very important to select the best 3aTg plant for vector movement. We initially speculated that the expression level of 3a would correlate with the rate of systemic movement of the vector. However, we found that a high level of 3a expression was not necessarily correlated with viral spread in the 3aTg plant. After the vector systemically spread in the 3aTg plant, the 3a protein suddenly decreased to nearly undetectable levels ( Figure 2e ), suggesting that the 3a protein might be degraded in the virus‐infected tissues. This result actually agrees with the observation that 3a accumulation rapidly changed in the CMV‐infected plants with certain stages of leaf development ( Itaya , 1998 ). When we compared No. 3 and No. 22, which had similar levels of 3a expression, the Δ3aCMV vector replicated much more efficiently and with more systemic movement in No. 3 than in No. 22, indicating that No. 3 was the best helper plant for the vector system. Our success can be attributed to intensive selection of helper plants that complement vector movement. We found a possible negative correlation between the 3a expression level and vector movement, which can be a hallmark for selecting ‘good’ helper plants expressing 3a. Therefore, the CMV vector is now dependent on the helper Tg plant for its movement. It will be interesting to analyse the turnover of the 3a protein in such 3aTg plants. We did not detect any recombination between the inoculated virus and the transgene in the helper plant. However, we cannot completely deny the potential for such recombination because the recombinant virus actually retains the 3a gene sequence except that it has the introduced stop codons. In the past, several intensive studies using natural CMV populations have never found any RNA3 recombination between two discrete CMV groups (subgroups I and II) ( Fraile , 1997 ; Lin , 2004 ; Bonnet , 2005 ). Learning from such observations, we decided to apply two kinds of the 3a genes to prevention of RNA recombination in our VCS and successfully to demonstrate prevention that no recombination indeed occurred between the two 3a genes even in mixed infections ( Figure 5 ). This lack of recombination is perhaps because of the low sequence homology (∼70%) between the two 3a genes. In addition, the two CMVs are localized separately in the mixed‐inoculated tissues ( Takeshita , 2004 ), suggesting that they perhaps repel each other. There must be a mechanism that the CMVs acquired during their evolutionary histories to discriminate the two subgroups so that no chimeric viruses would be generated between the two subgroups. We evidently copied this natural secret to create this unique viral vector system. Although we found that no RNA recombination occurred in RNA 3 between the two subgroups, there is still a concern about pseudorecombination of viral RNAs in mixed infection. What if m2Δ3aH1Dx acquired an intact RNA 3 molecule from a challenged CMV? Because a previous report showed that a CMV mutant lacking 2b provided strong cross‐protection against wild‐type CMVs (both subgroups) ( Ziebell , 2007 ), we tested whether the helper plant that is systemically infected with m2Δ3aH1Dx is cross‐protected against challenge inoculation with a wild‐type CMV. We observed complete cross‐protection in the helper plants infected with the vector lacking 2b, agreeing with the results of Ziebell (2007) . These results suggest that mixed infection to generate recombination (pseudorecombination) is not established in our system. Taken together, we conclude that our system has a very low probability for viral RNA recombination and pseudorecombination, even when the helper plant is accidentally challenged by a wild‐type CMV. Yields of virus‐produced proteins were not necessarily consistent between inoculated and non‐inoculated leaves. Perhaps protein yields in the inoculated leaves depend mainly upon the inoculation efficiency, while viral movement and localization largely affect protein production in non‐inoculated upper leaves. As for the expression levels of target protein, our system is not superior at this moment but is on a level comparable to those by the old‐type TMV vector ( McCormick , 1999 ). At present, Agrobacterium ‐mediated delivery of the TMV vector, so‐called ICON system, is the most rapid path from gene to recombinant protein and provides the highest levels of protein production ( Marillonnet , 2005 ; Giritch , 2006 ). However, in this work, we mostly focused on managing potential risks of viral vectors, sacrificing efficient expression of target protein to some extent; thus, improvement of the protein yield in our system will certainly be the next step in our research. Our first strategy for improving yield is to use synergistic effects between CMV and a potyvirus; it is well under way for VCS. VCS has the advantage of speed, stability, cost effectiveness and product safety. Thus, this plant vector system has the potential to produce recombinant proteins that cannot be efficiently produced by any other method. Our plant‐made antibodies can be stably stored, easily distributed and instantly prepared for diagnosis in an ELISA directly using the plant tissues. The system may be applicable for producing many other high‐quality, useful recombinant proteins such as a vaccine against an influenza virus for quick, large‐scale production to prevent a pandemic. Our system enables the CMV vector to produce recombinant proteins in plants without viral escape and recombination. Given that tobacco seedlings can grow in a short time in the greenhouse, this system is convenient to produce high‐quality recombinant proteins like DxscFv and IL1‐Ra. Our results thus demonstrate that there are certainly ways to diminish the risk of viral vectors and to take advantage of a viral vector to produce recombinant proteins in plants. Experimental procedures Plant material Plants had been grown at 23 °C under continuous illumination. Seeds of N. benthamiana plants were surface‐sterilized and sown on nutrient agar plates containing half‐strength Murashige‐Skoog (MS) salts, 1% sucrose, half‐strength B5 vitamin and 1% agar. After germination, the seedlings were placed in peat pots (Jiffy‐7), grown at 22 °C in a greenhouse and allowed to self‐pollinate. Virus strains and movement‐defective CMV vector construction The movement‐defective CMV vector was constructed based on the CMV‐based vector that was created from CMV‐Y (subgroup I) ( Matsuo , 2007 ). pCY3‐YΔ3a was constructed to introduce a stop codon in the 3a gene by subcloning in the original vector a PCR fragment amplified by a recombinant PCR using primer pairs containing appropriate restriction sites: Y3‐T7‐5Bm‐F (5′‐cgggatccattaatacgactcactataggtaatctaaccacctgt‐3′) and 3a‐Stop‐3‐R (5′‐ctggtaccttagaaagccat‐3′) 3a‐Stop‐5‐F (5′‐atggctttctaaggtaccag‐3′) and Y3‐3Hind‐R (5′‐aacaagcttcttatcatattcc‐3′). Similarly, pCM3‐m2Δ3a was constructed from RNA3 of CMV‐m2 (subgroup II) ( Takeshita , 2004 ) with two stop codons by subcloning a PCR fragment amplified by a recombinant PCR using primer pairs: m2‐5‐23 (5′‐cgctggaggtaatcttaccactttctttttc‐3′) and m2‐3‐stp3 (5′‐cgtcctactggtaccttagaaagccat‐3′) m2‐stp‐5 (5′‐aggtaccagtaggacgtaaactcaaca‐3′) and m2‐3‐Xb‐1000 (5′‐gactctagactcacatgtattt‐3′). The PCR‐amplified fragments were then exchanged with the corresponding sequences in the original construct to generate plasmids, pCY3‐YΔ3a and pCM3‐m2Δ3a. In vitro RNA transcripts and the viral inoculation were prepared as described previously ( Matsuo , 2007 ). The PVX vector was obtained from Dr Baulcombe ( Baulcombe , 1995 ). Cloning of the anti‐dioxin IgG gene to create the scFv construct in the viral vector A hybridoma cell line was produced from BALB/c mice after intraperitoneal injection of a phenoxatine derivative‐Bovine serum albumin (BSA) conjugate. ELISA was performed to identify positive hybridoma (Dx02‐I) cell lines producing monoclonal antibodies (IgG) to dioxin (2,3,4,7,8‐PeCDF). RNA was isolated from the hybridoma (Dx02‐I) cells, and the full‐length clone for DxscFv was obtained by PCR, according to the protocol of the mouse scFv module/recombinant phage antibody system (GE Healthcare, Piscataway, NJ, USA). The anti‐dioxin scFv gene (GenBank accession AB474005 ) was amplified from pCANTAB 5E‐Dx02‐I using PCR with the primer pairs DxscFv‐F1 (5′‐cgaggcctagaatgtacttgggactgagc‐3′) and DxscFv‐R1 (5′‐gcgacgcgttcaaagttcatccttatgatg‐3′) containing Stu I and Mlu I sites, respectively. This 837‐bp DxscFv gene was then cloned between the Stu I and Mlu I sites in pCY2‐H1, resulting in the pCY2‐H1:DxscFv plasmid. Cloning of the IL1‐Ra gene in the viral vector The cDNA for the mature form of the IL1‐Ra gene (GenBank accession NM000573 ) was amplified from the human cDNA library (TaKaRa Bio, Ohtsu, Japan) by PCR and cloned between the Stu I and Mlu I sites in the pCY2‐H1 vector. The in planta expression of IL1‐Ra by the viral vector was confirmed by Western blot analysis using commercially available antibodies against IL1‐Ra (R&D Systems, Minneapolis, MN, USA). Plant transformation and selection The full‐length 3a gene (GenBank accession D12499 ) was amplified from pCY3 by PCR, using the 3a‐specific primer pair Y‐3a‐F (5′‐gcactagtatggctttccaaggtaccag‐3′) and Y‐3a‐R (5′‐cgagctcctaaagaccgttaaccacct‐3′) containing a Spe I and Sac I site, respectively. This 855‐bp fragment was cloned into pGEM‐T (Promega, Madison, WI, USA). The Y3a gene was then subcloned into the binary vector pBE2113 ( Mitsuhara , 1996 ). The recombinant binary plasmid (pBE2113:Y3a) was transformed into Agrobacterium tumefaciens LBA4404, and Tg N. benthamiana plants were generated by the leaf disk method ( Horsch , 1986 ). Regenerated plantlets (T0) were allowed to self‐pollinate, and seeds (T1) were collected and tested for resistance to 250 mg Km/mL. Virus inoculation Carborundum‐dusted leaves of non‐Tg and 3aTg N. benthamiana seedlings (1.5 months old) were rub‐inoculated either with the synthesized RNA transcripts or sap‐containing virus. The sap inoculum was prepared by grinding leaf tissues from young systemically seedlings, grown in a greenhouse at 22 °C in 0.1 m potassium phosphate buffer (pH 8.0). SDS‐PAGE and Western blot analysis Leaf samples were homogenized in 3 volumes (mass/v) of phosphate‐buffered saline (PBS) containing 0.1% Triton X‐100 and a proteinase inhibitor. Crude leaf extracts were centrifuged at 6000 g for 10 min, and the supernatant was separated by electrophoresis on SDS‐12% polyacrylamide gels (TEFCO, Tokyo, Japan) under reducing conditions. For Western blot analysis, proteins from the SDS‐PAGE were electrophoretically transferred to Hybond P membranes (GE Healthcare). Membranes were blocked with 3% skim milk powder (mass/v) in PBS and probed with the anti‐3a polyclonal antibodies. Positive signals were detected using the ECL plus detection reagent (GE Healthcare). RNA extraction and semiquantitative RT‐PCR Total RNA from N. benthamiana leaves was isolated using TRIzol (Invitrogen, Carlsbad, CA, USA) and treated with RQ1 RNase‐free DNase (Promega) to remove any contaminating DNA. Prepared RNA (10 ng) was used to synthesize cDNAs using the SuperScript III One‐step RT‐PCR System with Platinum Taq High Fidelity (Invitrogen). Transcript levels of target genes relative to those of the ubiquitin gene ( Ubi3 ) as an internal control were used to normalize the data. The primer pair designed for CMV‐Y‐3a was CMV3a‐F1 (5′‐atggctttccaaggtaccag‐3′) and CMV3a‐840R1 (5′‐ctaaagaccgttaaccacctgc‐3′), and the pair for Ubi3 was Ubi3‐F (5′‐gattggtggtattggaactgtcc‐3′) and Ubi3‐R (5′‐gagcttcgtggtgcatctc‐3′). Purification of DxscFv The leaf extracts were homogenized in PBS buffer containing 0.1% Triton X‐100 and a proteinase inhibitor (Complete, EDTA‐free; Roche, Mannheim, Germany). The insoluble fraction was removed by centrifuging at 10000 g for 15 min, and the supernatants were then applied to mini‐columns prepacked with 1.5 mL nickel nitrilotriacetic acid gel (Ni‐NTA Superflow Agarose; QIAGEN, Valencia, CA, USA) and equilibrated with binding buffer [25 m m PB, 340 m m NaCl, 20% glycerol (pH 7.4)]. After loading the lysates, the columns were washed with a fivefold volume of washing buffer [25 m m PB, 340 m m NaCl, 20% glycerol, 20 m m imidazole (pH 7.4)] and eluted with elution buffer [25 m m PB, 340 m m NaCl, 20% glycerol, 300 m m imidazole (pH 7.4)]. The eluted protein solution was then dialysed against the binding buffer. The amount of purified protein was determined with a protein assay kit (Bio‐Rad, Hercules, CA, USA), using a normal mouse IgG Fab fragment as a standard. We used the DxscFv standard (KEM, Kyoto, Japan) produced by Escherichia coli and transgenic tobacco. RT‐PCR to monitor recombination between m2Δ3a and Y3a Total RNA was isolated from the upper leaves of 3aTg (T2) No. 3 inoculated with m2Δ3aH1Dx using TRIzol regent (Invitrogen) and then with RQ1 RNase‐Free DNase (Promega) to remove any contaminating DNA. Purified RNA (50 ng) was reverse‐transcribed with the primer that hybridizes downstream of the 3a gene containing the stop codons (m2Δ3a) in CMV‐m2 RNA3 but not the Y3a transgene transcript in a 10 μL of reaction volume with Reverse Transcriptase XL (AMV) (TaKaRa Bio) at 50 °C for 15 min. The synthesized first‐strand cDNA (1 μL) was used for the subsequent PCR amplification, which was carried out in a total volume of 50 μL of reaction mixture with Ex Taq‐HS for hot start (TaKaRa Bio) with the following cycling parameters: 25 cycles of 94 °C, 30 s; 65 °C, 30 s; 72 °C, 1 min. Five microlitres of the PCR product was analysed in an agarose gel. As shown in Figure 4b , the reverse primer (c: 5′‐ctaaagaccgttaaccacctg‐3′) hybridizes both Y3a and m2Δ3a. The forward primer that can specifically recognize the introduced stop codons in m2‐3a was b2 (5′‐aaggtaccagtaggacGtA‐3′), and the corresponding forward primer for Y3a was b1 (5′‐aaggtaccagtaggacTtT‐3′); two different bases (in capitals) discriminate Y3a and m2Δ3a. Detection of recombination between PVX‐3a and the 3a gene in RNA3 of CMV Non‐Tg N. benthamiana plants were co‐inoculated with the transcripts of PVX‐3a and either YΔ3aH1 or m2Δ3aH1. Serial viral passages were performed every 2 weeks. To confirm the sequence of the 3a genes in the CMV RNA3, RT‐PCR was performed to amplify the viral 3a gene using a 3a‐intervening primer pair that does not exist in the PVX‐Y3a sequence. RT‐PCR to detect RNA3 recombination between CMV‐Y and CMV‐m2 Nicotiana benthamiana plants were co‐inoculated with CMV‐Y and CMV‐m2. Two weeks after inoculation, total RNA was extracted from the systemically infected leaves. RT‐PCR was performed using a one step RNA PCR kit (AMV) (TaKaRa Bio) according to the supplier’s instructions. The RT reaction was performed at 50 °C for 30 min and terminated with a 2‐min incubation at 94 °C. PCR amplification (30 cycles) was carried out in programmed steps of 94 °C for 30 s, 55 °C for 30 s and 72 °C for 60 s. The primer pair for CMV‐Y is Y‐3R‐5 (5′‐atttatttcgttgtaca‐3′) and Y‐3a‐3 (5′‐aaaaccagatgtgttcc‐3′); the primer pair for CMV‐m2 is m2‐3R‐5 (5′‐tgtgtgttagttagtgt‐3′) and m2‐3a‐3 (5′‐aaccccagatgggaaat‐3′). The PCR products were then electrophoretically separated on 1% agarose gels. Acknowledgements This work was supported in part by grants from the Ministry of Economy, Trade and Industry (METI). We thank Drs D. Baulcombe and M. Takeshita for providing the PVX vector and CMV‐m2, respectively. References Aaziz , R. and Tepfer , M. ( 1999 ) Recombination between genomic RNAs of two Cucumoviruses under conditions of minimal selection pressure . Virology , 263 , 282 – 289 . Arntzen , C.J. ( 2008 ) Using tobacco to treat cancer . Science , 321 , 1052 – 1053 . Arntzen , C. , Plotkin , S. and Dodet , B. ( 2005 ) Plant‐derived vaccines and antibodies: potential and limitations . Vaccine , 23 , 1753 – 1756 . Baulcombe , D.C. , Chapman , S. and Santa Cruz , S. ( 1995 ) Jellyfish green fluorescent protein as a reporter for virus infections . Plant J. 7 , 1045 – 1053 . Bonnet , J. , Fraile , A. , Sacristán , S. , Malpica , J.M. and García‐Arenal , F. ( 2005 ) Role of recombination in the evolution of natural populations of Cucumber mosaic virus , a tripartite RNA plant virus . Virology , 332 , 359 – 368 . Canto , T. , Choi , S.K. and Palukaitis , P. ( 2001 ) A subpopulation of RNA 1 of Cucumber mosaic virus contains 3′ termini originating from RNAs 2 or 3 . J. Gen. Virol. 82 , 941 – 945 . Chen , Y.‐K. , Goldbach , R. and Prins , M. ( 2002 ) Inter‐ and intramolecular recombinations in the Cucumber mosaic virus genome related to adaptation to alstroemeria . J. Virol. 76 , 4119 – 4124 . Ding , B. , Li , Q. , Nguyen , L. , Palukaitis , P. and Lucas , W.J. ( 1995a ) Cucumber mosaic virus 3a protein potentiates cell‐to‐cell trafficking of CMV RNA in tobacco plants . Virology , 207 , 345 – 353 . Ding , S.‐W. , Li , W.‐X. and Symons , R.H. ( 1995b ) A novel naturally occurring hybrid gene encoded by a plant RNA virus facilitates long distance virus movement . EMBO J. 14 , 5762 – 5772 . Fernández‐Cuartero , B. , Burgyán , J. , Aranda , M.A. , Salánki , K. , Moriones , E. and García‐Arenal , F. ( 1994 ) Increase in the relative fitness of a plant virus RNA associated with its recombinant nature . Virology , 203 , 373 – 377 . Fraile , A. , Alonso‐Prados , J.L. , Aranda , M.A. , Bernal , J.J. , Malpica , J.M. and García‐Arenal , F. ( 1997 ) Genetic exchange by recombination or reassortment is infrequent in natural populations of a tripartite RNA plant virus . J. Virol. 71 , 934 – 940 . Giritch , A. , Marillonnet , S. , Engler , C. , Van Eldik , G. , Botterman , J. , Klimyuk , V. and Gleba , Y. ( 2006 ) Rapid high‐yield expression of full‐size IgG antibodies in plants coinfected with noncompeting viral vectors . Proc. Natl Acad. Sci. U.S.A. 103 , 14701 – 14706 . Gleba , Y. , Klimyuk , V. and Marillonnet , S. ( 2007 ) Viral vectors for the expression of proteins in plants . Curr. Opin. Biotechnol. 18 , 134 – 141 . Hiatt , A. and Pauly , M. ( 2006 ) Monoclonal antibodies from plants: a new speed record . Proc. Natl Acad. Sci. U.S.A. 103 , 14645 – 14646 . Horsch , R.B. , Klee , H.J. , Stachel , S. , Winans , S.C. , Nester , E.W. , Rogers , S.G. and Fraley , R.T. ( 1986 ) Analysis of Agrobacterium tumefaciens virulence mutants in leaf discs . Proc. Natl Acad. Sci. U.S.A. 83 , 2571 – 2575 . Itaya , A. , Woo , Y.M. , Masuta , C. , Bao , Y. , Nelson , R.S. and Ding , B. ( 1998 ) Developmental regulation of intercellular protein trafficking through plasmodesmata in tobacco leaf epidermis . Plant Physiol. 118 , 373 – 385 . Kaiser , J. ( 2008 ) Is the drought over for pharming? Science , 320 , 473 – 475 . Kaplan , I.B. , Shintaku , M.H. , Li , Q. , Zhang , L. , Marsh , L.E. and Palukaitis , P. ( 1995 ) Complementation of virus movement in transgenic tobacco expressing the Cucumber mosaic virus 3a gene . Virology , 209 , 188 – 199 . Lico , C. , Chen , Q. and Santi , L. ( 2008 ) Viral vectors for production of recombinant proteins in plants . J. Cell. Physiol. 216 , 366 – 377 . Lin , H.‐X. , Rubio , L. , Smythe , A.B. and Falk , B.W. ( 2004 ) Molecular population genetics of Cucumber mosaic virus in California: evidence for founder effects and reassortment . J. Virol. 78 , 6666 – 6675 . Ma , J.K.‐C. , Drake , P.M.W. and Christou , P. ( 2003 ) The production of recombinant pharmaceutical proteins in plants . Nat. Rev. Genet. 4 , 794 – 805 . Marillonnet , S. , Thoeringer , C. , Kandzia , R. , Klimyuk , V. and Gleba , Y. ( 2005 ) Systemic Agrobacterium tumefaciens ‐mediated transfection of viral replicons for efficient transient expression in plants . Nat. Biotechnol. 23 , 718 – 723 . Masuta , C. , Ueda , S. , Suzuki , M. and Uyeda , I. ( 1998 ) Evolution of a quadripartite hybrid virus by interspecific exchange and recombination between replicase components of two related tripartite RNA viruses . Proc. Natl Acad. Sci. U.S.A. 95 , 10487 – 10492 . Matsuo , K. , Hong , J.‐S. , Tabayashi , N. , Ito , A. , Masuta , C. and Matsumura , T. ( 2007 ) Development of Cucumber mosaic virus as a vector modifiable for different host species to produce therapeutic proteins . Planta , 225 , 277 – 286 . McCormick , A.A. , Kumagai , M.H. , Hanley , K. , Turpen , T.H. , Hakim , I. , Grill , L.K. , Tusé , D. , Levy , S. and Levy , R. ( 1999 ) Rapid production of specific vaccines for lymphoma by expression of the tumor‐derived single‐chain Fv epitopes in tobacco plants . Proc. Natl Acad. Sci. U.S.A. 96 , 703 – 708 . Mitsuhara , I. , Ugaki , M. , Hirochika , H. , Ohshima , M. , Murakami , T. , Gotoh , Y. , Katayose , Y. , Nakamura , S. , Honkura , R. , Nishimiya , S. , Ueno , K. , Mochizuki , A. , Tanimoto , H. , Tsugawa , H. , Otsuki , Y. and Ohashi , Y. ( 1996 ) Efficient promoter cassettes for enhanced expression of foreign genes in dicotyledonous and monocotyledonous plants . Plant Cell Physiol. 37 , 49 – 59 . Palukaitis , P. and García‐Arenal , F. ( 2003 ) Cucumoviruses . Adv. Virus Res. 62 , 241 – 323 . Roossinck , M.J. ( 2002 ) Evolutionary history of Cucumber mosaic virus deduced by phylogenetic analyses . J. Virol. 76 , 3382 – 3387 . Shi , B.‐J. , Miller , J. , Symons , R.H. and Palukaitis , P. ( 2003 ) The 2b protein of cucumoviruses has a role in promoting the cell‐to‐cell movement of pseudorecombinant viruses . Mol. Plant Microbe Interact. 16 , 261 – 267 . Soards , A.J. , Murphy , A.M. , Palukaitis , P. and Carr , J.P. ( 2002 ) Virulence and differential local and systemic spread of Cucumber mosaic virus in tobacco are affected by the CMV 2b protein . Mol. Plant Microbe Interact. 15 , 647 – 653 . Suzuki , M. , Kuwata , S. , Kataoka , J. , Masuta , C. , Nitta , N. and Takanami , Y. ( 1991 ) Functional analysis of deletion mutants of Cucumber mosaic virus RNA3 using an in vitro transcription system . Virology , 183 , 106 – 113 . Takeshita , M. , Shigemune , N. , Kikuhara , K. , Furuya , N. and Takanami , Y. ( 2004 ) Spatial analysis for exclusive interactions between subgroups I and II of Cucumber mosaic virus in cowpea . Virology , 328 , 45 – 51 . Taliansky , M.E. and García‐Arenal , F. ( 1995 ) Role of cucumovirus capsid protein in long‐distance movement within the infected plant . J. Virol. 69 , 916 – 922 . De Wispelaere , M. , Gaubert , S. , Trouilloud , S. , Belin , C. and Tepfer , M. ( 2005 ) A map of the diversity of RNA3 recombinants appearing in plants infected with Cucumber mosaic virus and Tomato aspermy virus . Virology , 331 , 117 – 127 . Ziebell , H. , Payne , T. , Berry , J.O. , Walsh , J.A. and Carr , J.P. ( 2007 ) A cucumber mosaic virus mutant lacking the 2b counter‐defence protein gene provides protection against wild‐type strains . J. Gen. Virol. 88 , 2862 – 2871 .

Journal

Plant Biotechnology JournalWiley

Published: Jan 1, 2011

Keywords: ; ; ; ;

There are no references for this article.