Biotechnol. Prog. 2008, 24, 372−376
372
Scale-Up of Agrobacterium-Mediated Transient Protein Expression in Bioreactor-Grown Nicotiana glutinosa Plant Cell Suspension Culture Kristin M. O’Neill, Jeffrey S. Larsen, and Wayne R. Curtis* Department of Chemical Engineering, The Pennsylvania State University, University Park, Pennsylvania 16802
The reporter gene β-glucuronidase was transiently expressed in a 51-L bioreactor-grown plant cell suspension culture of Nicotiana glutinosa at a yield of approximately 1.1 mg through coculture with an auxotrophic strain of Agrobacterium tumefaciens. The three order of magnitude scale-up involved the investigation of factors contributing to transient expression including the timing of Agrobacterium inoculation relative to the plant cell growth phase, plant tissue culture hormonal triggers and plant cell cycle synchronization. The co-culture process was simplified to facilitate implementation in a pilot-scale bioreactor. At the shake flask scale it was determined that elevated concentrations of oxygen in the headspace were detrimental to transient expression levels and the addition of acetosyringone to the co-culture had a negligible effect. The bacterial preparation process was also streamlined, permitting the direct transfer of the Agrobacterium culture from a bench-scale fermentor to the pilot-scale plant cell culture bioreactor. Increasing expression levels and overcoming batch-to-batch variability despite extensive procedure systemization remain the major technical hurdles.
Introduction Transient protein expression in plant cell culture is being developed as a rapid alternative to the production of heterologous protein in transgenic plants (1). Our approach has been to develop auxotrophs of Agrobacterium tumefaciens that are compatible with scalable suspensions of plant cells (2). We have found this approach to be comparable to leaf infiltration methods in terms of absolute expression levels (3). Since plant tissue culture can be successfully grown to tens of thousands of liters (4), the approach is potentially scalable for production beyond our initial objective of providing a means of rapid assessment of plant-expressed heterologous proteins to assist in the development of transgenic field-grown plants (5). The scalability and containment of plant cell culture make it appealing as a commercial protein expression platform for human therapeutics. Like mammalian cells, plants cells are capable of post-translational modifications such as glycosylation (6). However, unlike mammalian cells, plant cells are not carriers of human pathogens. Plant cell suspension cultures are also amenable to Agrobacterium-mediated transient expression, a rapid, scalable and low-cost gene delivery system. When cocultured with plant cell suspensions, recombinant Agrobacteria can efficiently deliver heterologous DNA to the nucleus of virtually every plant cell. Transient expression of this episomal DNA then results in a temporal peak in heterologous protein expression after 3-5 days of co-culture. This peak in protein expression is not only reached much sooner than with nuclear or plastid transformation, but is generally higher than nonselected stable expression levels (7). Several recently proposed large-scale Agrobacterium-mediated transient expression systems rely on Agro-infiltration of whole plants (8, 9). However, the inherent variability in yield and purity of heterologous protein derived from field or * To whom correspondence should be addressed. Ph: 814-863-4805. Fax: 814-865-7846. E-mail:
[email protected]. 10.1021/bp0703127 CCC: $40.75
greenhouse-grown biomass may limit these products to smallscale in Vitro applications and animal models. In contrast, the containment and rigorously controlled cultivation conditions of plant cell culture are more amenable to reliable commercial production of human therapeutics (10). In addition, the high accessibility of Agrobacterium to the plant cell surface and the general ability to optimize the physiological state may provide for avenues to reproducible, high-level heterologous protein expression in plant cell culture. Our approach to synchronizing tissue by nutrient limitation provides for control over cell physiology at all scales in suspension cultures. We have previously generated an auxotrophic strain of A. tumefaciens to facilitate scale-up (2). Use of this strain eliminates the need for antibiotic selection to prevent bacterial over-growth during co-culture. This work details the adaptation and scale-up of plant cell culture-based transient expression to production in a pilot-scale bioreactor.
Materials and Methods An Agrobacterium rhizogenes-transformed plant cell suspension culture of Nicotiana glutinosa was maintained and cocultured with an auxotrophic strain of A. tumefaciens transformed with the pGPTVK-GI binary vector in 125-mL shake flasks as previously described (7). The yield of GUS was quantified through a fluorometric assay with 4-methylumbelliferyl β-D-glucuronide (MUG) as the substrate (11). GUS activity was measured as the molar rate of formation of 4-methyl-7-hydroxycoumarin (MU) in 10 mg/L total soluble protein extract. Total soluble protein levels were determined using a modified Bradford assay with BSA as the protein standard (12). Enzyme activity measurements were converted into yield of GUS as a percent of total soluble protein (% TSP) based on assays of β-glucuronidase type VII-A from E. coli (Sigma) as a standard (11). Pilot-scale culture of plant tissue occurred in a modified 80-L Lee Industries reactor vessel that was designed for mammalian
© 2008 American Chemical Society and American Institute of Chemical Engineers Published on Web 03/12/2008
Biotechnol. Prog., 2008, Vol. 24, No. 2
cell culture. Aeration was carried out initially at 1 standard liter per minute (SLPM) of gas addition through a sintered metal sparger located near the bottom of the vessel with a headspace pressure of 5.0 psig. A dissolved oxygen set-point minimum of 50% relative to saturation with air at the operational temperature of 25 °C was maintained using the ML-6100 multi-loop Bioprocess Controller (New Brunswick Scientific) to mix compressed air with oxygen. During co-culture, when the gas mix reached 40% oxygen, the gas flow rate was increased to reduce oxygen requirement and avoid oxygen toxicity. The pH was monitored but adjustments were not necessary to maintain it at 5.5 due to the ability of plant tissue culture to regulate pH by utilizing both ammonia and nitrate as nitrogen sources. Agitation was achieved using a 23-cm low-shear “elephant ear” impeller (ABEC Inc., Bethlehem, PA). An agitation set-point of 50 RPM achieved an operating tip speed of 0.598 m/s, comparable with tip speeds used previously in plant tissue culture (13). Inoculum addition occurred by positive displacement from a 10-L Pyrex carboy agitated on a rotary shaker table through an inoculation port designed from sanitary fittings to permit aseptic addition of viscous plant tissue culture. A reduced reactor headspace pressure of 0.5 psig was maintained during inoculation. A New Brunswick Scientific BioFlo 3000 bench-scale fermentor was used for the large-scale culture of the A. tumefaciens auxotroph containing the pGPTVK-GI plasmid. Agitation in the BioFlo was achieved using two 7.6-cm Rushton impellers at 150 RPM with four 2.5-cm baffles. The temperature was maintained at 25 °C and compressed air was sparged at 0.5 vvm using a ring-style sparger. Agrobacteria inoculum was looped from solid LB media (14) containing 50 mg/L kanamycin and cultured for 12-14 h in 5 mL of LB media in culture tubes; 2 mL was then transferred from culture tubes to 250-mL baffled shake flasks containing 60 mL of LB media. Agitation of culture tubes and shake flasks was achieved using a 1.9 cm rotary shaker table set at 200 RPM. 80-100 mL of 12-14 h-old bacteria were then added to 3.5 L of LB media in the BioFlo 3000. Following steam-in-place reactor sterilization, the 60-L working volume bioreactor was charged with 27 L of filter-sterilized MSG media at one-third dilution. MSG media (25 g/L sucrose), a derivative of MS salts media (15), contains 1X B5 vitamins (16), 0.5 mg/L 2,4-dichlorophenoxyacetic acid, and 0.2 mg/L kinetin. Then, 4.5 L of inocula was added from cultures grown in 2.8-L Fernbach flasks for 10-12 days prior to inoculation. Following sucrose depletion, as determined by minimal change in the media refractive index, the approximately 30 L culture was maintained in stationary phase for 24 h. Eighteen liters of MS macronutrients and sucrose were then added, both at double the concentration used in MSG media, and 24 h following media addition the reactor was connected to the BioFlo 3000 via silicone tubing. Approximately 500 mL of bacteria, grown to an absorbance of 1 AU measured at 600 nm, was transferred from the BioFlo 3000 by positive displacement using compressed air. Smaller flasks were run in parallel with the reactor to control for the reactor inoculation procedure and contamination testing. One 2.8-L Fernbach flask and one 250-mL Erlenmyer flask were inoculated in series with the reactor by drawing tissue from the reactor immediately following its inoculation. Leftover inoculum from the 10-L Pyrex carboy was used to inoculate a second set of control flasks in parallel with the reactor. Bacterial culture was added to control flasks from the BioFlo 3000 immediately following reactor co-culture initiation; 10 and 0.75 mL of
373
bacterial culture were added to the 2.8-L and 250-mL flasks, respectively, to achieve a 1% (v/v) addition of bacteria.
Results and Discussion Cell cycle synchronization by sucrose depletion and subsequent nutrient addition prior to co-culture increases reporter gene expression on the basis of total soluble protein and batch yield. Experiments qualitatively demonstrated that tissue at an early point in the growth curve was more amenable to transient expression than older tissue (results not shown). While initiating co-culture 24 h following inoculation maximizes reporter gene expression in terms of GUS produced per unit of tissue, the relatively low cell density results in low overall batch yields. It is much more economical to operate a bioreactor at the cell density limits that can be supported in liquid cell culture. A synchronization procedure that provides for initial growth followed by nutrient exhaustion then fresh nutrient addition allows for transient expression to be carried out in a higher density culture of rapidly growing and dividing cells than initiating co-culture 24 h following the inoculation of plant cells. Relative to the non-synchronized treatments, synchronization enhanced GUS expression by almost 20-fold (t test, R ) 0.05; Figure 1A). For the synchronized treatments there was no significant difference in expression between 1X MSG, 2X MSG and minimal 2X MSG media additions (t test, R ) 0.05; Figure 1A). Consequently, the physiological state achieved by synchronization of the cells is more important than the composition of the nutrient spike used for synchronization. The addition of minimal 2X MSG is more cost-effective, especially in larger batches, since it contains only MS macronutrients and sucrose at 2X MSG concentrations. The hormones 2,4-dichlorophenoxyacetic acid (2,4-D) and kinetin are not triggers for the enhancement in expression associated with cell cycle synchronization. It was surprising that the addition of kinetin and 2,4-D (present in 1X MSG and 2X MSG) to sucrose-depleted cultures did not lead to significantly higher expression in the synchronized treatments (Figure 1A). Kinetin is a cytokinin responsible for regulating cell division while 2,4-D is an auxin with the ability to regulate cell elongation (17). The close relationship between DNA replication, cell division and the observed positive effects of cell synchronization motivated a more detailed investigation on the effect of kinetin on transient expression of episomal DNA. Despite the physiological importance of kinetin, treatments receiving 2X MSG supplemented with an additional 0.2 mg/L of kinetin produced less GUS relative to control treatments receiving unaltered 2X MSG or kinetin-free 2X MSG (t test, R ) 0.05; Figure 1B). Separate experiments involving treatments with modified concentrations of the auxin 2,4-D also failed to demonstrate this hormone as a trigger of enhanced transient expression (data not shown). These results support the use of hormone-free minimal 2X MSG as the nutrient spike for synchronization of cultures maintained on media containing exogenous hormones. Oxygen supplementation is detrimental to reporter gene expression and the addition of acetosyringone to the coculture has a negligible effect. Due to the excellent mixing in shake flasks, the liquid phase readily equilibrates with the gas phase in the headspace. However, at the 80-L scale the relatively low surface area to volume ratio necessitates sparging of the gas phase to meet the oxygen demand of the heterotrophic liquid culture. Concerns over increased oxygen demand during bacterial co-culture motivated an investigation into the effect of an oxidative environment on transient gene expression. An increase
374
Biotechnol. Prog., 2008, Vol. 24, No. 2
Figure 1. Factors affecting transient expression of GUS (expressed as the rate of production of fluorescent MU from MUG substrate) in 125-mL shake flasks. Error bars represent one standard deviation among triplicate treatments. Experiments were carried out as described in Materials and Methods with any deviations noted. (A) 5 mL of 2-week-old cell culture was inoculated into 30 mL of either 1X MSG or 1/3X MSG for the non-synchronized and synchronized treatments, respectively. 24 h following sucrose depletion in the synchronized treatment, both the synchronized and respective non-synchronized treatment each received a 20 mL media addition of either 1X MSG, 2X MSG or a minimal nutrients solution containing only sucrose and MS macronutrients at the concentrations found in 2X MSG. (B) The specified concentration of Kinetin was added to the cultures 24 h following sucrose depletion along with 20 mL of 2X MSG with respect to every other component. (C) Oxygen-supplemented treatments and controls (21% oxygen) were co-cultured with Agrobacteria grown in a modified YEB induction medium consisting of 5 g/L sucrose, beef extract, peptone, 1 g/L yeast extract, 0.493 g/L magnesium sulfate heptahydrate and 1.95 g/L MES. The pH was adjusted to 5.6 and acetosyringone was added to a final concentration of 20 µM. Elevated oxygen treatments were connected to an oxygen overlay after synchronization for the duration of the co-culture. (D) The 0 µM AS treatments were co-cultured as described in Materials and Methods except for the addition of AS which was omitted.
in the oxygen uptake rate of the culture can be balanced by either increasing the transport rate (kLa) or increasing the driving force for transport by mixing compressed air with pure oxygen. Increasing the partial pressure of oxygen is preferable because it allows for a constant gas hold-up volume fraction and minimizes the production of foam. Flasks exposed to 29% and 53% oxygen in the headspace for the duration of bacterial coculture produced approximately 2.5% and 5% of the GUS in their respective un-supplemented controls (Figure 1C). Due to the sensitivity of transient gene expression in plant tissue to elevated oxygen partial pressures, on the pilot scale the dissolved oxygen set-point was maintained by gradually increasing total gas flow rate when the gas mix exceeded 40% oxygen during co-culture. Acetosyringone (AS) is a phenolic compound that can attract A. tumefaciens to wounded plant tissue by chemotaxis as well as induce the Vir genes to initiate T-DNA transfer (18-20). Although AS is commonly included in Agrobacterium culture media to enhance transient expression, its impact has been variable. At AS concentrations exceeding 0.45 mM, Wydro et al. reported a 2-fold enhancement in transient expression of gfp in N. benthamiana leaves only when co-infiltrated with the suppressor of silencing, HC-Pro (21). Transient expression of GUS was not enhanced by the addition of 0.1 mM AS to the co-culture relative to the control treatments (t test, R ) 0.05; Figure 1D). The Vir genes must be sufficiently induced by the sucrose and acidic pH of the co-culture medium or a compound naturally exuded by the de-differentiated plant cell suspensions. At larger scales the economic benefit of omitting even low concentrations of AS from the co-culture medium is significant.
At a final concentration of 0.5 mM, for instance, the addition of AS can increase the cost per liter of co-culture media by approximately 140%. Agrobacteria can be cultured in liquid media for up to 48 h in the absence of antibiotic selection pressure and directly transferred to plant tissue culture without intermediate processing. As described previously, pGPTVK-GI has a kanamycin selectable marker (7). Kanamycin, however, is toxic to plant tissue and when combined with LB to culture Agrobacteria it must be removed prior to co-culture. This was previously accomplished by washing the bacterial pellet in M9 media (14) before a final resuspension in M9 media supplemented with cysteine to support limited growth of the cysteine-dependent auxotroph of A. tumefaciens during co-culture. On larger scales it is desirable and economical to eliminate the need for antibiotics. Eliminating these processing steps also allows for the direct transfer of the Agrobacteria from a bench-scale fermentor to the 80-L bioreactor for co-culture. Relative to control treatments supplemented with 50 mg/L kanamycin, treatments cultured for 24 or 48 h without kanamycin led to statistically equivalent transient expression levels upon coculturing with plant tissue (t test, R ) 0.05; Figure 2). In this same figure, Agrobacteria cultured in kanamycin-free LB media for 24 h were transferred directly into the co-culture without compromising its transient expression abilities (t test, R ) 0.05). The sucrose present in plant tissue culture media and the free cysteine present in LB media must be sufficient to induce the auxotrophic Agrobacteria and meet its nutrient requirements. Even more importantly, Agrobacteria grown in the 3-L BioFlo fermentor are capable of yielding transient expression levels at
Biotechnol. Prog., 2008, Vol. 24, No. 2
375
Figure 2. Effect of Agrobacteria preparation and culture time on transient expression of GUS in N. glutinosa shake flask suspension cultures. Kanamycin-supplemented treatments (light-gray bars) included 50 mg/L kanamycin. Treatments subjected to “Prep.” were cultured in 5 mL of LB media on a 1.9 cm stroke rotary shaker, at 25 °C. The bacteria were then centrifuged at 3000 RCF for 7 min, washed, and resuspended in cysteine-supplemented M9 media to an absorbance of 1.0 AU at 600 nm before being added to plant tissue culture at 1% of the working co-culture volume. The asterisk (/) treatment refers to Agrobacteria grown in a 3-L BioFlo fermentor. Treatments subjected to “No Prep.” were transferred directly to co-culture with the omission of centrifugation, wash, and resuspension steps.
Figure 3. Time-course of transient expression of GUS in N. glutinosa in the 80-L bioreactor relative to simultaneously inoculated 0.25-L shake flask and 2.8-L Fernbach controls. Treatments inoculated in series with the reactor (designated “-s”) contained tissue drawn from the reactor immediately following its inoculation, while leftover reactor inoculum was used to inoculate treatments in parallel with the reactor (designated “-p”).
least as high as those cultured in test tubes on a gyratory shaker (t test, R ) 0.05; Figure 2). Consequently, scale-up of the Agrobacteria culture can be accomplished without compromising their ability to transfer T-DNA to host plant cells. It is also worth noting that Agrobacteria (which can display substantial variations in its 2 h doubling time) can be grown under typical bacterial fermentor conditions of agitation and air sparging without detrimental impact on its maximum specific growth rate (data not shown). Agrobacterium-mediated transient expression of GUS in plant cell suspensions readily scaled-up to a 51-L working volume with no loss in productivity. GUS expression in a pilotscale bioreactor produced concentrations at least as high as in shake flask controls run in parallel (Figure 3), with levels being 5-10 times higher for the majority of the time course. Controls inoculated in series with the reactor performed as well as those inoculated in parallel, suggesting that the potentially stressful inoculation procedure did not harm the plant cells’ ability to transiently express heterologous protein. Refractive index measurements from the reactor confirmed that the growth of plant tissue culture was comparable to that observed in controls; therefore, the enhanced expression was due to greater productivity per unit of plant biomass. The enhanced expression within the bioreactor relative to controls indicates that the superior control of the culture conditions achievable in a bioreactor is beneficial to transient protein expression. GUS concentration peaked in the reactor at 4 days following co-culture initiation, giving 23 µg GUS/L (0.007% of the total soluble protein). Expression levels in this experiment were low relative to previously published work by our group for both the reactor and control treatments. Although we have systematically improved reproducibility via rigor in culture maintenance and synchronization, observed expression levels for controls can still vary by an order of magnitude between identical experiments inoculated at different times (3). Future work will require identification and control of the unidentified physical and physiological conditions that are required for optimal, reproducible heterologous gene transfer and expression in this complex co-culture process.
We have recently reported the use of DNA viral vectors based on the bean yellow dwarf virus using a complementation strategy where the viral replication protein gene is co-delivered in trans on a separate vector which recognizes the cis-acting elements on the vector containing the 35S-driven reporter gene. Dual Agrobacterium-mediated delivery of the two vectors to a cell culture of N. glutinosa resulted in a 2-fold enhancement of transient expression to 0.2% TSP (7). These enhanced transient expression levels are comparable to constitutive expression from stably transformed and selected cell lines, but are modest compared to the achievements other groups have reported with RNA-based viral vectors. While we have successfully generated transgenic tissues expressing Gemini viral replicase from inducible promoters for the in trans complementation of replication-competent sub-genomic viral vectors (22), we have not yet successfully generated cell lines from these transgenics that are amenable to transient expression. In fact, we have generally found that there are very dramatic cell-line dependent differences in expression in addition to plant genus and species (23). Also noteworthy is the fact that our group and several others (personal communication) have failed in attempts to regenerate fertile transgenic N. glutinosa (24). This has significantly slowed the progress of developing a cell culture-based transient expression platform because of difficulties in making complementary transgenics to reduce viral vector size and improve biocontainment. A promising route to achieving high-level transient expression involves engineering amplified expression cassettes from plant RNA viruses. Nearly a decade ago, we initiated a collaboration with Monsanto toward the development of cell culture based Agrobacterium-mediated transient expression that included the use of RNA viral vectors. During that time, Large Scale Biology Corporation tried unsuccessfully to commercialize RNA viral vectors for plant-expressed proteins. Meanwhile, our laboratory diverted efforts to developing DNA viral vectors (collaborating with Boyce Thompson Institute, Cornell) due to concerns of RNA virus instability and transgene size constraints (25). We hope that improved T-DNA expression cassettes derived from viral vectors carrying cell cycle and protein expression control
Conclusion
Biotechnol. Prog., 2008, Vol. 24, No. 2
376
mechanisms (26-28) can overcome these limitations to provide a rapid, scaleable heterologous protein expression platform for plant tissues.
Acknowledgment The authors acknowledge Calvin Leiter for his extensive voluntary assistance in rebuilding the 80-L bioreactor, as well as Benjamin T. Smith, Amalie Tuerk, Robert J. Hesser, Anne Marie Schilthuis (Niehaus) and Kurt Niehaus, who dedicated long hours assisting with the work. Jason Collens is also acknowledged specifically for his contributions toward identification of the importance of culture synchronization. This material is based upon work supported by the National Science Foundation under grant no. BCS-0003926 and GOALI Program. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.
References and Notes (1) Curtis, W. R. Protein production in transgenic plants. In Encyclopedia of Chemical Processing; Lee, S., Ed.; Taylor & Francis: New York, 2006; pp 2489-2500. (2) Collens, J. I.; Lee, D. R.; Seeman, A. M.; Curtis, W. R. Development of auxotrophic Agrobacterium tumefaciens for gene transfer in plant tissue culture. Biotechnol. Prog. 2004, 20 (3), 890896. (3) Andrews, L. B.; Curtis, W. R. Comparison of transient protein expression in tobacco leaves and plant suspension culture. Biotechnol. Prog. 2005, 21 (3), 946-952. (4) Smith, M. A. L. Large scale production of secondary metabolites. In Current Issues in Plant Molecular and Cellular Biology; Terzi, M., Cella, R., Falavigna, A., Eds.; Kluwer Academic Publishers: Dordrecht, The Netherlands, 1995; pp 66-74. (5) Curtis, W. R. Quantitative transient protein expression in plant tissue culture. U.S. Patent 6,740,526; May 25, 2004. (6) Gomord, V.; Faye, L. Posttranslational modification of therapeutic proteins in plants. Curr. Opin. Plant Biol. 2004, 7 (2), 171-181. (7) Collens, J.; Mason, H.; Curtis, W. Agrobacterium-mediated viral vector-amplified transient gene expression in Nicotiana glutinosa plant tissue culture. Biotechnol. Prog. 2007, 23 (3), 570-576. (8) Azhakanandam, K.; Weissinger, S.; Nicholson, J.; Qu, R.; Weissinger, A. Amplicon-plus targeting technology (APTT) for rapid production of a highly unstable vaccine protein in tobacco plants. Plant Mol. Biol. 2007, 63, 393-404. (9) Gleba, Y.; Klimyuk, V.; Marillonnet, S. Magnifection-a new platform for expressing recombinant vaccines in plants. Vaccine 2005, 23, 2042-2048. (10) Miele, L. Plants as bioreactors for biopharmaceuticals: regulatory considerations. Trends Biotechnol. 1997, 15, 45-50. (11) Rao, A. G.; Flynn, P. A microtiter plate-based assay for β-Dglucuronidase: A quantitative approach. In GUS Protocols: Using the GUS Gene as a Reporter of Gene Expression; Gallagher, S. R. Ed.; Academic Press: San Diego, 1992; pp 89-99. (12) Zor, T.; Selinger, Z. Linearization of the Bradford protein assay increases its sensitivity: theoretical and experimental studies. Anal. Biochem. 1996, 236 (2), 302-308.
(13) Singh, G.; Curtis W. R. Reactor design for plant cell suspension culture. In Biotechnological Applications of Plant Cultures; Shargool, P., Ngo, T., Eds.; CRC Press: Boca Raton, FL, 1994; p 158. (14) Davis, R. W.; Botstein, D.; Roth, J. R. AdVanced Bacterial Genetics. A Manual for Genetic Engineering; Cold Spring Harbor Laboratory: Cold Spring Harbor, NY, 1980; pp 201-207, 209210, 253. (15) Murashige, T.; Skoog, F. A revised medium for rapid growth and bio assays with tobacco tissue cultures. Physiol. Plant. 1962, 15, 473-497. (16) Gamborg, O. L.; Miller, R. A.; Ojima, K. Nutrient requirements of suspension of soybean root cells. Exp. Cell Res. 1968, 50, 151158. (17) Moore, T. C. Biochemistry and Physiology of Plant Hormones, 2nd ed.; Springer-Verlag: New York, 1989; pp 33-38, 184-185. (18) Hiei, Y.; Ohta, S.; Komari, T.; Kumashiro, T. Efficient transformation of rice (Oryza sativa L.) mediated by Agrobacterium and sequence analysis of the boundaries of the T-DNA. Plant J. 1994, 6, 271-282. (19) Turk, S.; Melchers, L.; den Dulk-Ras, H.; Regensburg-Tuink, A.; Hooykaas, P. Environmental conditions differentially affect Vir gene induction in different Agrobacterium strains: Role of the VirA sensor protein. Plant Mol. Biol. 1991, 16, 1051-1059. Turk-199. (20) Godwin, I.; Todd, G.; Ford-Lloyd B.; Newbury H. The effects of acetosyringone and pH on Agrobacterium-mediated transformation vary according to plant species. Plant Cell Rep. 1991, 9, 671-675. (21) Wydro, M.; Kozubek, E.; Przemyslaw, L. Optimization of transient Agrobacterium-mediated gene expression in leaves of Nicotiana benthamiana. Acta Biochim. Pol. 2006, 53 (9), 289-298. (22) Curtis, W. R.; Tuerk, A. L. Oxygen Transport in Plant Tissue Culture Systems. In Plant Tissue Culture Engineering, Focus on Biotechnology Vol. 6; Gupta, D. S., Ibaraki, Y., Eds.; Springer: Dordrecht, 2006; pp 173-186. (23) Collins, J. I. Agrobacterium-mediated transient protein expression in plant cell suspension and root cultures. Ph.D. Thesis, The Pennsylvania State University, 2006. (24) Sailer, L. J. Therapeutic protein expressions system in transgenic N. glutinosa tissue culture. B.S. Thesis, The Pennsylvania State University, 2003. (25) Pogue, G. P.; Lindbo, J. A.; Garger, S. J.; Fitzmaurice, W. P. Making an ally from an enemy. Plant virology and the new agriculture. Annu. ReV. Phytopathol. 2002, 40, 45-74. (26) Voinnet, O.; Rivas, S.; Mestre, P.; Baulcombe, D. An enhanced transient expression system in plants based on suppression of gene silencing by the pl9 protein of tomato bush stunt virus. Plant J. 2003, 33, 949-956. (27) Mallory, A.; Parks, G.; Endres, M.; Baulcombe, D.; Bowman, L.; Pruss, G.; Vance, V. The amplicon-plus system for high-level expression of transgenes in plants. Nat. Biotechnol. 2002, 20, 622625. (28) Marillonnet, S.; Thoeringer, C.; Kandzia, R.; Klimyuk, V.; Gleba, Y. Systemic Agrobacterium tumefaciens-mediated transfection of viral replicons for efficient transient expression in plants. Nat. Biotechnol. 2005, 23 (6). Received September 11, 2007. Accepted January 30, 2008. BP0703127
