Transgenic Research 7, 265±271 (1998)
Coat protein-mediated resistance to pea enation mosaic virus in transgenic Pisum sativum L. G A N G A M M A M . C H OW R I R A 1 x , T I M OT H Y D. C AV I L E E R 2 x , S A N JAY K . G U P TA 1 , PAU L F. L U R Q U I N 1 and P H I L I P H . B E R G E R 2 1 2
Department of Genetics and Cell Biology, Washington State University, Pullman, WA 99164-4234 Department of Plant, Soil, and Entomological Sciences, University of Idaho, Moscow, ID 83844-2339
Received 1 December 1997; revised and accepted 16 March 1998
Pea (Pisum sativum L.) plants were transformed in planta by injection=electroporation of axillary meristems with a chimeric pea enation mosaic virus (PEMV) coat protein gene contruct. R1 progenies of these plants were shown to harbor the transgene by polymerase chain reaction (PCR) and genomic Southern analysis, while transgene expression was demonstrated by western blot analysis. Transgenic R2 , R3 and R4 plants displayed delayed or transient PEMV multiplication and attenuated symptoms as compated to control inoculated individuals. Keywords: coat protein-mediated virus resistance; PEMV, pea enation mosaic virus, electroporation; pea transformation; transgenic plants
Introduction Genetic engineering techniques have shown great promise in the development of pest resistant or tolerant crop plants. Coat protein-mediated virus resistance was demonstrated for the first time a decade ago (Powell-Abel et al., 1986). A number of virus resistant crops are in the process of being commercialized, and more are in development phases. In spite of this early success, there are no reports on the engineering of grain legumes for virus resistance. This is undoubtedly due to the special challenge presented by the genetic transformation of these crops. Indeed, it is well known that most legumes are refractory to in vitro culture techniques, thereby limiting the use of classical direct gene transfer methodologies. Further, an Agrobacterium tumefaciens-based gene transfer technique established for pea (Schroeder et al., 1993) has met with limited success, although it is possible that careful selection of target tissues will render this methodology more widely applicable (Grant et al., 1995). Nevertheless, some researchers consider that alternative DNA delivery To whom correspondence should be addressed at: rm 242, Ag. Sci, Dept. of Plant, Soil and Entomological Science, University of Idaho, Moscow, ID 83844-2339 (Fax: 208 885 6319; Email:
[email protected]). x Authors contributed equally.
0962±8819 # 1998 Chapman & Hall
techniques should be explored in order to bypass the shortcomings of the existing methods and possibly speed up the production of transgenic plants (Songstad et al., 1995). In that light, we recently demonstrated (Chowrira et al., 1995; 1996) that pea (Pisum sativum L.), lentil (Lens culnaris Medik), cowpea (Vigna unguiculata L.), and soybean (Glycine max L.) could be stably transformed by injection and subsequent electroporation of DNA into the axillary buds in planta. Here, we report the application of this method to produce transgenic pea plants expressing the coat protein of pea enation mosaic virus (PEMV). PEMV is currently the only member of the enamovirus group. This virus is characterized by a genome of two positive-strand ssRNAs encapsidated in separate, structurally distinct isometric particles (Demler et al., 1993). PEMV is transmitted in nature by several species of aphids, particularly the pea aphid (Acyrthosiphon pisum Harris), in a persistent manner. Aphid transmission of PEMV during short feeding periods tends to circumvent efforts to control pea enation mosaic disease by applying systemic aphicides to peas (Hampton, 1989). As such, PEMV causes periodic serious losses to crops of susceptible pea and lentil cultivars, particularly in the northeastern and northwestern pea-producing areas of the United States. In this article we describe resistance to PEMV infection exhibited by transgenic pea lines expressing a chimeric PEMV coat protein gene.
266 Materials and methods Plasmid construct The plasmid pPCP4-5 (Fig. 1), a plant expression vector, contained the PEMV coat protein (PEMV-CP) gene. The 595 bp PEMV-CP gene was amplified by PCR from clone pP221 (kindly provided by G. de Zoeten) and subcloned into the PCR cloning vector pCRII (Invitrogen) and subsequently into Xho I=Sac I digested pKYLX 71 S2 . pKYLX 71 S2 is a plant expression binary vector comprised of an enhanced CaMV 35S promoter (Benfey et al., 1990) and 39 rbcs terminator (Schardl et al., 1987). This vector was used because we had initially intended to use Agrobacterium-mediated plant transformation. The method actually used for pea transformation (below) did not, in fact, require a binary vector or a selection system. In planta injection=electroporation Pea transformation was done essentially as described previously (Chowrira et al., 1995; 1996). Briefly, the apical portions of the pea plants (var. `Sparkle') were decapitated close to the node of a fully expanded leaf and discarded, prior to electroporation. The stipule and adjacent petiole were removed to expose the most terminal axillary bud. All other axillary buds were excised. The buds thus retained were injected with 2 ìl pPCP4-5 DNA=lipofectin=MS salts solution (200 ìg DNA=ml solution) and then electroporated at 100 V using 2 square wave pulses. A few plants were injected=electroporated with pKYLX vector alone (not containing the PEMV-CP insert) and maintained as negative controls. After treatment, the plants were placed back in the greenhouse and allowed to grow. R1 progeny were raised from seeds obtained from shoots that developed from the buds subjected to electroporation in planta. Genomic DNA preparation and analysis by PCR Plant genomic DNA was isolated according to the CTAB method of Doyle and Doyle (1990) and further purified by phenol-chloroform extraction subsequent to RNase treatment (Maniatis et al., 1982). DNA preparations from R1 pea plants were analyzed using PCR with 28-mer oligonucleotide primers (59-CGACTCGAGAACAATGGCGACTAGATCG-39 and 59-GCAGAGCTCTTATCAGAGG GAGGCATTCA-39) designed to amplify the PEMV-CP coding sequence. Reactions (100 ìl) were run on a Techne PHC3 apparatus using one denaturing cycle at 95 8C for 5 min followed by 40 cycles of 1 min at 95 8C, 1 min at 56 8C, 1 min at 72 8C. Final extension was accomplished by holding the reactions at 72 8C for 10 min. PCR controls included non-transformed pea genomic DNA spiked and non-spiked with PEMV-CP containing plasmid, no DNA template control, and a PEMV positive control (pPCP4-5 using PEMV-CP primers). The PCR products were analyzed by 1% agarose gel electrophoresis, blotted and
Chowrira et al. hybridized to the PEMV-CP probe by Southern hybridization (see below). Genomic Southern blot analysis The Genius TM kit (Boehringer Mannheim) was used for Southern analysis of genomic DNA isolated from R1 plants. The PEMV-CP probe consisted of the gel-isolated 595 bp Xho I=Sac I fragment from pPCP4-5 (Fig. 1). This fragment contains the complete PEMV-CP coding sequence. DNA transfer, labeling of the probe and detection were according to the manufacturer's protocol (Genius System User's Guide for Memberane Hybridization Ver. 3.0, Mannheim Boehringer, Indianapolis, IN). Western blot analysis Total protein was extracted from frozen single leaves according to Vetten et al. (Vetten et al., 1992). Protein concentration in the supernatant was determined using the Bio-Rad Coomassie assay. Aliquots of leaf extracts corresponding to 75 ìg total protein were electrophoresed on 12% SDS-polyacrylamide gel and transferred by electroblotting to nitrocellulose membranes. The membranes were probed with an antibody raised against the PEMV-CP. Symptom assays of transgenic plants Seeds from selfed transgenic plants (Southern-positive for PEMV-CP) and control plants were sown in individual pots. The R2 , R3 , and R4 individuals were then challenged with PEMV by mechanical inoculation. The viral inoculum was prepared by grinding PEMV-infected pea plants (1:5, w=v) 10 d after inoculation, with 0.1 M sodium acetate, pH 6.0 and cellite 1:6 (w=v). Prepared inoculum was used at 1 ml per 5 newly germinated pea plants (at the 4 full leaf stage). Inoculum was gently rubbed on the leaves and washed with tap water 3 min post-inoculation. Samples were collected at 0 d (prior to inoculation), and then at 5, 10, 15, and 20 d post-inoculation. Visual symptoms were monitored and scored during the entire period of challenge cycle. Indirect ELISA tests were performed as described by Crowther (Crowther, 1995). Leaf samples were ground in modified (Mink et al., 1992) carbonate buffer (Clark et al., 1977), while antiserum (kindly provided by G. de Zoeten) and goat anti-rabbit alkaline phosphatase conjugate (Sigma Chem. Co.) were each used at dilutions of 1:2000 (V=V) in PEP buffer (phosphate buffered saline containing 2% polyvinylpyrrolidone and 0.2% ovalbumin). Additional modifications included a blocking step using 5% Carnation nonfat dry milk in PBS and crossabsorbtion of the PEMV antibody with a 1:10 dilution of sap extracted from healthy peas in PBS for 1 h at 37 8C prior to use in ELISA. A405 readings were taken 15, 30, and 60 min using a Microplate Autoreader EL 311, from
Transgenic Peas
267
Bio-Tek Instruments. Readings greater than 0.2 OD405 were considered positive. Results Integration and expression of PEMV-CP gene in pea plants Terminal axillary buds of pea plants were injected and electroporated in planta (Chowrira et al., 1995, 1996) with 200 ìg=ml pPCP4-5 plasmid DNA (Fig. 1). Up to 95% of the axillary buds survived the injection=electroporation treatment and developed into normal looking branches, which eventually produced flowers and pods with seeds. Seeds from electroporated R0 pea plants were used to obtain the R1 progeny. Genomic DNA was obtained from 23 individual R1 plants (progeny from 7 R0 plants) and analyzed by PCR using primers corresponding to the PEMV-CP coding sequence. Three plants showed an expected 595 bp product in agarose gels. The PCR products were confirmed to the PEMV-CP coding sequence by Southern hybridization (Fig. 2A) with a PEMV-CP probe. Genomic DNA from the PCR-positive R1 plants were subjected to genomic Southern analysis (Fig. 2B). The DIG-labeled PEMV-CP probe hybridized to a 595 bp fragment in all three samples, which corresponds to the intact PEMV-CP insert released from the genomic DNA when digested with Xho I=Sac I (lanes 2, 5, 7, Fig. 2B). The probe also hybridized to junction fragments (containing PEMV-CP gene and a fragment of plant genome, the size of which would depend upon where the DNA had integrated) of different sizes in the samples digested with Xho I alone (lanes 1, 4, 6, Fig. 2B). At least two copies of the transgene were integrated in the genome 1 2 3 4 5 6 7 8 P35S2
EcoR I
Trbcs
PEMV-CP
Sac I
Xho I
Cla I
595 bp PROBE pPCP4-5 (PEMV-cp:pKYLX71S2) 9
Fig. 1. Partial map of the chimeric PEMV-CP region in pPCP4-5 (PEMV-CP:pKYLX71S2 ). The 595 bp Xho I=Sac I fragment from pPCP4-5 containing the entire PEMV-CP coding sequence was labeled with DIG-11-dUTP, and used as the PEMV-CP probe in Southern blots.
Fig. 2. (A) Southern blot of PCR amplified products from R1 pea plants. Lane 1: positive control (amplified PEMV-CP insert from pPCP4-5); lanes 2 & 3, 7 & 8: PCR amplified negative controls; lanes 4, 5 & 6: PCR amplified products from pea lines P1A, P1 B, P3A. (B) Genomic Southern blot analysis of three transgenic R1 pea plants (lines P1A, P1 B, P3A) hybridized to DIG-labeled PEMVCP probe. Lanes 1, 4, 6: Xho I-digested DNA from P1A, P1 B & P3A, respectively; lanes 2, 5, 7: DNA from P1A, P1 B & P3A digested with Xho I=Sac I; lanes 10, 11, 12: undigested DNA from P1A, P1 B & P3A; lanes 3, 8, 9: Xho I-digested DNA from negative control plants. (C) Western blot analysis of transgenic R1 pea plants using anti-PEMV-CP polyclonal antibodies. Lane 1: purified PEMV-CP (positive control); lanes 2 & 4: negative control; lanes 3 & 5: lines P3A and P1 B, respectively.
268
Chowrira et al.
of one of the transgenic lines (lane 1, Fig. 2B), as shown by the production of two bands after digestion with Xho I. The probe hybridized to DNA at high molecular weight when undigested (lanes 10, 11, 12, Fig. 2B), but not to DNA from control plants (digested with Xho I) (lanes 3, 8, 9). Genomic Southern analysis confirmed that these plants were derived from three independent transformation events, since they showed three different hybridization patterns. The expression of the PEMV-CP gene, at the protein level, was studied in plants that tested positive by genomic Southern analysis. Analysis of crude leaf protein extract by western blotting, using anti-PEMV-CP antibody, revealed the presence of a 21 kDa polypeptide that co-migrated with the coat protein purified from PEMVinfected pea plants (Fig. 2C). From the western blot, the level of PEMV-CP expression in P3A appeared to be higher than in P1 B. Expression of coat protein was not detected in the transgenic line harboring two copies of the transgene (data not shown). Resistance to PEMV Disease resistance was assessed by PEMV inoculation of descendants of PEMV-CP-harboring R1 plants and control plants (without the CP gene), and comparing disease
1. PEMV-resistant phenotype in R2 plants. Seeds from R1 lines were sown along with seeds from control plants. Initially, transgenic parents expressing the PEMV-CP were mechanically inoculated with PEMV at the 4-leaf stage. Half of the control plants were maintained non-inoculated (Fig. 3A), and half were inoculated with PEMV. Inoculated control plants developed typical stunting, leaf distortion, and enation symptoms by 9 days post-inoculation, followed by severe plant distortion (Fig. 3C). In contrast, a few descendants of transgenic individuals did not exhibit any of the symptoms associated with PEMV infection up to 12 days post-inoculation. Other descendants started to develop mild symptoms by day 16, and the symptoms remained mild until harvest (Fig. 3B). Mild mosaic symptoms and a few chlorotic lesions began to appear by day 13 in a few plants, and these plants developed intermediate levels of disease symptoms (data not shown). While some transgenic plants developed typical PEMV symptoms by day 9, the new leaves produced subsequently developed progressively less symptoms, and by flowering, the newly formed leaves were
P1B10 P3A2 P1B11 NI sum I sum
2.00 E 1.50 Absorbance405
symptom development and viral replication in the two types of plants. R3 progeny from resistant R2 lines and subsequent R4 progeny were again subjected to virus challenge to monitor stability of the resistant phenotype.
1.00
0.50
0.00
0
6 12 18 Days Post Inoculation
24
Fig. 3. PEMV-CP resistant phenotypes in R2 plants. A: Noninoculated control; B: Transgenic line P3A2 , inoculated with PEMV; C: inoculated control plant; D: Line P1 B11 showing recovery from initial infection (arrow indicates area of recovery); E: ELISA readings in R2 plants (NI sum: Average of noninoculated controls; I sum: Average of inoculated controls). Photographs taken 30 d post-inoculation.
Transgenic Peas
269
almost symptomless (Fig. 3D). In terms of symptom expression, these plants appeared to recover from the initial PEMV infection. Out of a total of 933 progeny R2 seeds (from 30 independent transgenic lines), 78 seeds failed to germinate. After inoculation, 123 plants exhibited dealyed PEMV symptom development and 114 plants maintained an almost symptomless (mild) phenotype. The rest were susceptible to infection. None of the lines tested were completely symptomless. Table 1 summarizes the results of the challenge experiment performed on R2 plants. ELISA results, indicative of the extent of virus replication, followed a pattern similar to that of visual assesment of disease symptoms, with mosaic symptoms appearing 7±8 days after inoculation. Symptoms were Table 1. Summary of R2 transgenic peas (Pisum sativum `Sparkle') inoculated with pea enation mosaic virus
Line#
Total# challenged
P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P16 P17 P18 P19 P22 P23 P24 P25 P27 P28 P29 P30 P31 P32 P35 P36
47 18 14 24 58 2 38 6 7 21 43 15 54 31 23 7 46 48 66 39 18 5 27 22 19 29 13 27 49 26
#Resistant A mild
late
#Susceptible B
3 3 1 7 4 0 10 1 0 5 6 3 9 7 1 1 9 8 10 0 0 1 5 3 1 2 3 3 8 3
5 5 4 6 10 2 5 3 1 6 15 1 12 9 6 0 2 4 7 0 3 0 2 4 2 0 2 3 3 2
39 10 9 11 5 0 23 2 6 10 22 11 33 15 6 6 35 36 49 39 15 4 20 15 16 27 8 21 38 21
A Resistance was based on symptoms and categorized as mild (very few visual symptoms throughout the experiment) or late (symptoms were slow in developing, generally not showing up until after 14 days postinoculation). Most plants developing late symptoms and some with mild symptoms did not produce seed. B Susceptibility defined as observing typical PEMV symptoms by 7±10 days post-inoculation.
most apparent at 10±12 days after inoculation. Virus titer, in plants exhibiting delayed and mild symptoms, were at background levels until day 18, after which there were incremental increases in titer until day 20 (Fig. 3E). In plants exhibiting recovery from initial infection, virus replication during the initial period after inoculation was high, but there was a dramatic decrease in virus titer after day 12 (Fig. 3E). Intermediate levels of virus titer were recorded for plants with intermediate symptoms (data not shown). 2. PEMV resistant phenotype in R3 plants. Twenty eight progeny plants of R2 parents, with interesting PEMVresistant phenotypes, were subjected to mechanical challenge with PEMV. When compared to the noninoculated and inoculated control plants (Fig. 4A & 4D), a few R3 progeny exhibited PEMV resistant phenotypes (Fig. 4B & 4C) very similar to that of their parents. The initial absence of visual symptoms was associated with a lack of virus accumulation in some individuals (Fig. 4E). By 20 d post-inoculation, however, virus titers in these plants were similar to those of inoculated control plants, although they manifested attenuated symptoms. Resistance to infection in these plants was exhibited as development of fewer chlorotic lesions and generally healthy plant growth (data not shown). A few R3 progeny of one of the lines, produced disease symptoms early following inoculation, but subsequently seemed to recover from disease. Again, the change in disease symptoms correlated well with levels of virus replication data (Fig. 4E). Most other R3 progeny plants exhibited intermediate phenotypes while a few were totally susceptible to PEMV infection (data not shown). 3. PEMV resistant phenotype in R4 plants. A small number of R4 progeny plants have been tested to far. These R4 progeny had a resistant phenotype similar to that of their respective parents. Again, as in the previous generations, some of the progeny were sensitive and some resistant, in every line challenged (data not shown). Discussion Transgenic plants expressing viral coat protein genes have been found to resist or tolerate crop diseases induced by the virus from which the coat protein gene was derived (for review, see Beachy et al., 1990). In this study we observed that R2 , R3 , and R4 generation pea plants transformed with the coat protein gene from PEMV exhibited resistance to infection by PEMV. While a few plants maintained a disease symptom-free phenotype, accompanied by a corresponding low level of virus accumulation, others recovered from an initial high level of virus accumulation and subsequent new growth was devoid of PEMV symptoms (not shown). Whether more than one mechanism is operating, as has been suggested
270
Chowrira et al.
P1B10-3 P1B11-1 P1B11-2 P3A2-1 NI sum I sum
2.00 E
Absorbance405
1.50
1.00
0.50
0.00 0
5 15 10 Days Post Inoculation
20
for replicase-mediated resistance (Hellwald et al., 1995), remains to be determined. Progeny of one susceptible line, while positive by Southern hybridization and by PCR analysis, produced no detectable amounts of coat protein by western analysis, suggesting that expression of coat protein is required for resistance to PEMV. We cannot comment with certainty if the presence of two copies of the transgene in its genome is in anyway responsible for this phenotype, although the possibility of transgene inactivation due to gene duplication cannot be ruled out (Meyer et al., 1996). The in planta method of gene transfer employed in this work to generate transgenic pea plants is different from the conventional methods of plant transformation in that the plants do not go through a selection process. The R0 materials developing after in planta electroporation are highly chimeric. As such, all the R1 progeny plants arising from a given R0 parent seed material are not identical, as evidenced by the molecular analyses of these plants. PCR and genomic Southern analysis revealed that only a few R1 plants from any given R0 line harbored the transgene. One of the other challenges of working with peas is
Fig. 4. PEMV-CP resistant phenotype in R3 plants. A: noninoculated control plant; B: transgenic line P1 B10 -3 inoculated with PEMV; C: transgenic line P3A2 -1 inoculated with PEMV; D: inoculated control plant; E: ELISA readings in R3 plants (NI sum: average of non-inoculated controls; I sum: average of inoculated controls). Photographs taken 28 d post-inoculation.
the low number of progeny seed produced by a single plant, making traditional genetic analysis difficult or impossible. Thus, it cannot be determined at this stage of the research whether the resistant phenotype is inherited in a predictable manner or if it is even stable. The fact that we have obtained R4 individuals with a resistant phenotype would suggest that the gene is stable, however. Further work is necessary to generate sufficient seed for agronomic testing as well as a source of germplasm for breeding programs. It is clear, however, that PEMV coat protein-mediated resistance can reduce virus replication, and that this resistance may provide economic levels of protection against this virus. Acknowledgments The authors thank Dr G. de Zoeten for providing the PEMV-CP antiserum and cDNA clone, Steve Wyatt for his helpful discussions, and Patrick Shiel for excellent technical assistance. This work was supported by grants from the Dry Pea & Lentil Commission of Washington, the USDA=Cool Season Food Legume Research Program, the Northwest Agricultural Research Foundation, and the
Transgenic Peas WSU Summer Graduate Research Assistantship awarded to G.M. Chowrira during Summer 1995.
References Beachy, R.N., Loesch-Fries, S. and Tumer, N.E. (1990) Coat protein-mediated resistance against virus infection. Annu. Rev. Phytopathol. 28, 451±74. Benfey, P.N., Ren, L. and Chua, N.-H. (1990) Combinational and synergistic properties of CaMV 35S enhancer subdomains. EMBO J. 9, 1685±96. Chowrira, G.M., Akella, V., Fuerst, P.E. and Lurquin, P.F. (1996) Transgenic grain legumes by in planta electroporationmediated gene transfer. Mol. Biotech. 5, 85±96. Chowrira, G.M., Akella, V. and Lurquin, P.F. (1995) Electroporation-mediated gene transfer into intact nodal meristems in planta. Mol. Biotech. 3, 17±23. Clark, M.F. and Adams, A.N. (1977) Characteristics of the microplate method of enzyme-linked immunosorbent assay for the detection of plant viruses. J. Gen. Virol. 34, 475±83. Crowther, J.R. (1995) ELISA: Theory and Practice. Totowna, NJ: Humana Press. Demler, S.A., Rucker, D.G. and de Zoeten, G.A. (1993) The chimeric nature of the genome of pea enation mosaic virus: The independent replication of RNA 2. J. Gen. Virol. 74, 1±14. Doyle, J.J. and Doyle, J.L. (1990) Isolation of plant DNA from fresh tissue. Focus 12, 13±15. Grant, J.E., Cooper, P.A., McAra, A.E. and Frew, T.J. (1995) Transformation of peas (Pisum sativum L) using immature cotyledons. Plant Cell Repts. 15, 254±8.
271 Hampton, R.O. (1989) Pea enation mosaic virus. In D.J. Hagedorn, ed., Compendium of Pea Diseases pp. 32±3. St. Paul, Minnesota: APS Press. Hellwald, K.H. and Palukaitis, P. (1995) Viral RNA as a potential target for two independent mechanisms of replicase-mediated resistance against cucumber mosaic virus. Cell 83, 937±46. Maniatis, T., Fritsch, E.F. and Sambrook, J. (1982) Molecular Cloning. A laboratory manual. New York: Cold Spring Harbor Laboratory Press. Meyer, P. and Saedler, H. (1996) Homology-dependent gene silencing in plants. Annu Rev. Plant Physiol. Plant Mol. Biol. 47, 23±48. Mink, G.I. and Silbernagel, M.J. (1992) Serological and biological relationships among viruses in the bean common mosaic virus subgroup. Arch. Virol. [Suppl. 5]. 397±406. Powell-Abel, P., Nelson, R.S., De, B., Hoffmann, N., Rogers, S.G., Fraley, R.T. and Beachy, R.N. (1986) Delay in disease development in transgenic plants that express the tobacco mosaic virus coat protein gene. Science 232, 738±43. Schardl, C.L., Byrd, A.D., Benzion, G., Altschuler, M.A., Hildebrand, D.F. and Hunt, A.G. (1987) Design and construction of a versatile system for the expression of foreign genes in plants. Gene 61, 1±11. Schroeder, H.E., Schotz, A.H., Wardley-Richardson, T., Spencer, D. and Higgins, T.J. (1993) Transformation and regeneration of two cultivars of pea (Pisum sativum L.). Plant Physiol. 101, 751±7. Songstad, D.D., Somers, D.A. and Griesbach, R.J. (1995) Advances in alternative DNA delivery techniques. Plant Cell Tissue Org. Culture 40, 1±15. Vetten, N.C., Lu, G. and Ferl, R.J. (1992) A magic protein associated with the G-box binding complex has homology to brain regulatory proteins. Plant Cell 4, 1295±1307.