Transformation of
Antirrhinum majus
L. by a
rol
-type
multi-auto-transformation (MAT) vector system
Cui Minlong
a, Kenji Takayanagi
a, Hiroshi Kamada
b, Shigeo Nishimura
a,
Takashi Handa
a,*
aInstitute of Agriculture and Forestry,Uni
6ersity of Tsukuba,Tsukuba,Ibaraki305-8572,Japan bInstitute of Biological Sciences,Uni
6ersity of Tsukuba,Tsukuba,Ibaraki305-8572,Japan
Received 3 May 2000; received in revised form 17 July 2000; accepted 24 July 2000
Abstract
A total of 11 independentb-glucuronidase (GUS) positive hairy roots were induced following co-cultivation of leaf explants of Antirrhinum majusL. withAgrobacterium tumefaciensstrain GV2260 containingrol-type multi-auto-transformation (MAT) vector pNPI702. A total of 326 adventitious shoots were regenerated from the hairy root lines on 1/2 MS medium without plant growth regulators at 25°C under a 16 h/day photoperiod condition 4 months after infection of theA.tumefaciensGV2260. The absence of the rol genes in five plants was verified by polymerase chain reaction (PCR) and Southern blot analysis. Acclimatized transformants exhibited normal phenotypes in height and in the morphology of leaves and flowers. Furthermore, the GUS gene was strongly expressed in the leaves, inflorescence of the transformed plant, and the progeny. This result demonstrates that the rol-type MAT vector can be used to study gene functions controlling the morphogenesis ofAntirrhinum majus plants. © 2000 Elsevier Science Ireland Ltd. All rights reserved.
Keywords:Hairy roots;Antirrhinum majusL.;Agrobacterium tumefaciensstrain GV2260; Multi-auto-transformation (MAT) vector; Polymerase chain reaction (PCR)
www.elsevier.com/locate/plantsci
1. Introduction
Antibiotics and herbicides are widely used as selectable markers in plant transformation [1 – 3]. These agents generally have a negative effect on proliferation and differentiation of the plant cell and may retard the regeneration of adventitious shoots during the transformation process. Some plant species, however, are insensitive to the selec-tive agents, so it is difficult to separate the trans-formed and non-transtrans-formed cells or tissues [4,5].
The ability of the rol gene of the Ri plasmid to
induce hairy roots has been utilized for the selec-tion of transformed cells, and this method has an advantage over other methods, for example when
analyzing an A. tumefaciens-mediated
transforma-tion because no marker genes are necessary [6].
Furthermore, the binary vector system using A.
rhizogenes have been transformed cultivated tomato plant species and phenotypically normal transgenic plants were obtained regenerated shoots from hairy roots [7].
The multi-auto-transformation (MAT) vector system is a novel transformation system using morphological changes caused by oncogenes or
rhizogenes of Agrobacterium (the ipt gene or the
rol gene) as the selection markers. The system is
designed to remove the oncogenes from transgenic plants after a transformation by inserting the oncogenes into removal elements [8,9]. This system does not need any antibiotic or herbicide resis-tance genes as selection markers. Researchers us-ing this system have described transformation in tobacco and hybrid aspen [8 – 10].
* Corresponding author. Tel.: +81-298-534796; fax: + 81-298-536617.
E-mail address:[email protected] (T. Handa).
Antirrhinum majus L., a typical ornamental plant, has been used for molecular studies of transposon, flower pigmentation, flower develop-ment, and Myb transcription factors [11 – 15].
Re-cently, transformation of A. majus was reported
using Agrobactrium rhizogenes [16 – 18] and A.
tumefaciens [19]; however, these transformation systems have several limitations or disadvantages. Moreover, frequencies of regeneration and trans-formation in both systems remain low.
In this paper, we describe the transformation of A. majus with a rol-type MAT vector system and the characteristics of marker-free transformed plants.
2. Materials and methods
2.1. Plant material
Seeds of A. majus cv. Floral Carpet Orchid and
Mdm. Butterfly Yellow (Sakata Seed Corp.) were
surface sterilized by a brief rinse in 70% (v/v)
ethanol and then in a 10% (v/v) solution of
sodium hypochlorite for 6 min, followed by wash-ing five times with sterile distilled water. The seeds were germinated and cultured on MS medium [20]
containing 2.5 g/l gellan gum and 25 g/l sucrose, at
25°C under a 16-h light photoperiod in a culture room.
2.2. Bacterial strains and 6ector plasmid
A disarmed A. tumefaciens strain, GV2260 [21],
was used for transformation.A. tumefaciensstrain
GV2260/pNPI702 harbored a rol-type MAT
vec-tor pNPI702 (Fig. 1A). The pNPI702 plasmid
contains the b-glucuronidase (GUS) gene under
CaMV 35S promoter, a ‘hit and run’ cassette in
which the rol A, B, C, andDgenes of Ri T-DNA
and the recombinase gene (R) of a recombination
gene with a 35S promoter, are located between directly oriented recombination site (RS)
se-quences [9]. A wild A. rhizogenes strain A4 was
used as a control.
2.3. Inoculation and root culture
Freshly grown A. tumefaciens strain GV2260/
pNPI702 in 3 ml LB medium containing 50 ml/l
rifampicin and 100 ml/l kanamycin at 28°C for 24
h was used for inoculation. The bacterial
suspen-sion culture was diluted 1/50 with liquid MS
medium. The leaf pieces of 6-week-old plants after germination were cut and inoculated with diluted bacterial suspension for 10 min, followed by trans-fer to solidified MS co-cultivation medium
supple-mented with 25 g/l sucrose, 100 mM
acetosyringone, and 2.5 g/l gellan gum (pH 5.8).
After 5 days of co-cultivation, the leaf explants
were transferred to solid MS medium with 25 g/l
sucrose and 250 mg/l cefotaxime. Root tips (about
1 cm) of induced adventitious roots including hairy roots were excised and transferred to the same medium. The culture was kept at 25°C under a 16 h photoperiod with fluorescent light (5000 lux). Axenic root cultures were established after two subcultures by transfers with 2-week intervals.
2.4. Plant regeneration
Transformed hairy roots identified by histo-chemical GUS analysis were maintained on solid
1/2 MS medium containing 25 g/l sucrose without
plant growth regulators. These cultures were trans-ferred to the same medium every 4 or 8 weeks at 25°C under 16 h photoperiod conditions. Regener-ated shoots excised from hairy roots were placed
on solid 1/2 MS medium at 25°C under 16 h
photoperiod conditions.
2.5. Polymerase chain reaction (PCR) analysis of GUS genes and rol B genes
Genomic DNA was extracted from leaf tissues according to the CTAB method of Rogers and
Bendich [22]. For the amplification of the gus-gene
homologous sequences androl Bgene homologous
sequences, the following primers were used: GUS
(+) primer, GGTGGGAAAGCGCGTTACAAG
and GUS (−) primer, CGGTGATACATA
TCCAGCCAT, and those for therol B gene were
rol B (+) primer, CCTCTAGAGTA
ACTATC-CAACTCACATCACAAG and B (−) primer,
TTGAATTCGTGGCTGGCGG TCTTCGATT-CATTTC. Polymerase chain reaction (PCR) was
performed in 20 ml reaction mixtures containing
100 ng of plant genome DNA, 100 mM of each
dNTP, 200 nM of each primer, and 1 unit of Taq
polymerase (TAKARA Co. Ltd.). Reactions were started with a denaturalization at 94°C for 3 min,
parameters: 93°C for 1 min, 55°C for 2 min, and 72°C for 3 min. The program was terminated by an extension at 72°C for 10 min. Amplified DNA bands were analyzed by 1% agarose gel elec-trophoresis at 100 V for 30 min followed by staining with ethidium bromide.
2.6. Southern hybridization analysis
Total DNA was isolated from 0.8 – 1.0 g leaves by the CTAB extraction method described by Rogers and Bendich [22]. A total of 10 mg of
DNA digested withEcoRI orEcoRI/HindIII were
subjected to electrophoresis on 1% agarose gel at
25 V for 16 h, transferred to Hybond-N+
(Amer-sham) nylon membranes, and hybridized with 32
P-labeled GUS gene fragment from plasmid pBI121. Final washing of the membranes was performed in
0.1% SDS, 2×SSC at 65°C for 15 min.
Hy-bridization signals were detected by a BAS-5000 image analyzer (Fuji Film).
2.7. Morphological characteristics
Marker-free transgenic plantlets were trans-ferred to pots with a mixture of vermiculite and
perlite (1:1 v/v) and grown in a closed greenhouse
for transgenic plants. Several morphological char-acteristics, including leaf shape, flower shape, and plant height, were observed in non-transformed
Fig. 1. Molecular analysis ofrol-type MAT vector pNPI702 in transformedAntirrhinum majusplants. (A) Map of the T-DNA region of rol-type MAT vector pNPI702 (RB, right boder of T-DNA; LB, left boder of T-DNA; R, recombinase gene; RS, recombination site; E,EcoRI, H,HindIII). The small arrows are PCR primers to amplify therolBandgusgenes. The expected size of PCR products is also shown. The probe ofgusencoding regien was labeled with32p-CTP. (B, C) Detection of the presence
ofgusandrolgenes in transformedA.majusplants by PCR analysis. M, size maker (l/HindIII); P, pNPI702; C, non-transformed plant; A4, a plant transformed with wild type A. rhizogenes A4; T0 yellow and T0 orchid, transformed plants of cv. Mdm.
Butterfly yellow and Floral Carpet Orchid byrol-type MAT vector pNPI702. (D) Southern hybridization ofEcoRI-digested DNA (10mg) extracted fromroltype MAT vector pNPI702 transformedA.majusplants of 6a and 7a of T0yellow, 1a of T0orchid and
Table 1
Shoots regeneration from hairy roots after 2 months culture on 1/2 MS medium
Cultivars
Bacterial strains No. of GUS No. of shoots regenerated No. of shoots GUS+/rol- by
PCR from GUS positive roots
positive roots (plasmid)
Floral carpet
GV2260 (pNPI702) 3 72 4
orchid
Mdm. butterfly 8 254 1
yellow
Floral carpet
Controla 0 0 0
orchid
Mdm. butterfly 0 0 0
yellow
aNon-infection.
plants (c), transformed wild-type A. rhizogenes
strain A4 plants (A4), and transformed rol-type
MAT vector pNPI702 plants (MAT).
2.8. Analysis of GUS expression in transgenetic plants and progeny
The plant shoots, leaves, and inflorescence were assayed for expression of the GUS gene following the histochemical staining procedure described by Jefferson [23]. After overnight staining (14 – 16 h), chlorophyll was excluded by soaking the tissues for several hours in 70% ethanol.
2.9. E6aluation of kanamycin resistance of T1
progeny
Progeny (T1) seeds were produced by crossing
non-transformed plants with marker-free
trans-formed plants. Mature T1seeds were
surface-steril-ized and germinated on solid MS medium
containing 50 mg/l kanamycin. On the selective
medium resistant progenies were green, whereas sensitive progenies were white. GUS expression in ten of the 6-week-old green seedlings was tested using histochemical staining.
3. Results
3.1. Induction of hairy root and shoot regeneration
After 3 weeks the infection of theA. tumefaciens
strain GV2260/pNPI702, the first adventitious
roots appeared from the wounded leaf pieces on
the MS medium added to 250 mg/l cefotaxime
without plant growth regulators. Eleven indepen-dent hairy roots emerged after 6 weeks of
infec-tion, showing fast growth on solid 1/2 MS
medium containing 250 mg/l cefotaxime and 25 g/l
sucrose without plant growth regulators, and they expressed the GUS gene. Adventitious shoots had
differentiated from hairy roots on the 1/2 MS
medium 4 weeks later. A total of 326 adventitious shoots were regenerated from 11 independent hairy roots, and they were transferred to the same medium (Table 1). Most of the regenerated shoots showed Ri syndrome symptoms such as dwarfism, wrinkled leaves, and an over abundance of roots. After 3 weeks of culturing, ten phenotypically normal shoots were obtained from seven indepen-dent hairy root lines (6a and 6b from one hairy
root line of T0yellow and 1a, 1b, and 1c from one
hairy root line of T0 orchid) within 4 months of
infection. These phenotypically normal shoots were transferred to the same fresh medium and rooted.
3.2. PCR and Southern blot analysis
Ten phenotypically normal plants of 326 regen-erated shoots were subjected to PCR analysis. The
predicted 1.5-kb gus fragment and 1.3-kb rolB
fragment were amplified with the primer sets GUS
(+) — GUS (−) and rolB (+) — rolB (−),
respectively (Fig. 1A). If the cassette was excised,
the predicted 1.3-kb rolB fragment could not be
amplified with the primer set rolB (+) — rolB
(−). In ten normal plants, the predicted 1.5-kb
1.3-kb rolB fragments were not amplified in one
transformant of ‘Butterfly yellow’ (T0yellow) and
four transformants of ‘Floral Carpet Orchid’ (T0
orchid) (Fig. 1C).
In addition, each transformant of T0yellow (7a)
and T0 orchid (1a) that did not show a rolB
fragment and one transformant of T0 yellow (6a)
that showed a rolB fragment were subjected to
Southern blot analysis. If cassette had been excised in plant genomes, the predicted 3.4 kb band could
not be shown in Southern blot analysis with gus
fragment as a probe (Fig. 1A). One rolB positive
line appeared 3.4 kb band but 2rolBnegative lines
lacked 3.4 kb bands (Fig. 1D). These results indi-cate that two plants are transgenic plants without
maker genes (rol genes).
3.3. Characteristic of marker-free transgenic plants
The regenerated plant from a hairy root
trans-formed by A. rhizogenes wild-type A4 (the A4
plant) exhibited an abundant root system and showed reduced apical dominance with short in-ternodes (Fig. 2A). The leaves of the A4 plant were elliptical in shape and were smaller than those of a non-transformed plant (Fig. 3A). More-over, the flower shape of the A4 plant was greatly altered from that of a non-transformed plant. The corolla of the A4 plant was shorter than that of the non-transformed plant (Fig. 2B). In contrast, a
marker-free transgenic plant (1a of the T0 orchid)
showed morphology similar to that of a non-trans-formed plant in height, leaf shape, and flower shape (Figs. 2 and 3A).
3.4. Expression of the GUS gene in transformed plants
We tested the leaves, flowers, and inflorescence of the progenies of marker-free transgenic plants for histochemical GUS analysis. Endogenous GUS-like activity was not detected in the leaves of either non-transformed plants or plants
trans-formed by A. rhizogenes wild-type A4 (Fig. 3A).
In contrast, a high level of GUS expression was observed in the shoots, leaves, and inflorescence of the progenies that showed kanamycin resistance on the selective medium (Fig. 3).
In an evaluation of kanamycin resistance, 54 out of 100 seedlings obtained from crossing
non-transformed plants and marker-free non-transformed plants showed kanamycin resistance. Furthermore, the rol B gene of T1 plants with normal morphol-ogy was not amplified by PCR (data not shown).
4. Discussion
In this experiment, rol-type MAT vector
pNPI702 (Fig. 1A) was used and 11 independent transformed roots (hairy roots) were obtained without any antibiotics. In PCR analysis of the
gus gene and therolBgene, five out of ten normal
phenotype plants indicated the integration of the
gusgene and the excision ofrolgenes (Fig. 2B, C).
High expression of GUS activity was recognized in marker-free transgenic plants and their progenies (Fig. 3), and these progenies showed kanamycin resistance.
Fig. 2. Morphology of height and flowers of wild type A. rhizogenesA4 transformed plant (A4), non-transformed plant (C) and maker-free transformed plant (1a of T0 orchid)
Fig. 3. Histochemical staining forb-glucuronidase (GUS) in the leaves, flower and inflorescence from maker-free transformedA. majusplant (1a of T0orchid plant) and the progeny. (A) Leaves (A4, a plant transformed with wild type A.rhizogenesA4; C,
non-transformed plant; MAT, 1a of T0orchid); (B) flower of 1a of T0orchid plant; (C) inflorescence of 1a of T0orchid plant;
and (D) T1progeny of 1a of T0orchid plant.
Although shoot regeneration from hairy root
cultures ofA.majushas been reported [16 – 18], the
frequency of shoot regeneration was relatively low and 5 – 6 months were necessary for regeneration. The A. tumefaciens transformation system needs more than half a year with several media changes for the regeneration of shoots [19]. In this study, a
total of 326 shoots (average 29.6 shoots/hairy root
line) were regenerated from 11 independent hairy
roots transformed with rol-type MAT vector
pNPI702 on 1/2 MS medium consisting of 2.5 g/l
gellan gum and 25 g/l sucrose without plant
growth regulator within 8 weeks of culturing [24]. Five out of ten phenotypically normal shoots di-rectly developed from hairy roots possessed only
the gus gene, as confirmed by PCR analysis and
Southern blot analysis. Recently, Ebinuma et al.
[8 – 10] reported marker-free transgenic plants
ob-tained from the T0 generation of transformed
to-bacco by ipt-type MAT vector or rol-type MAT
vector. This result also indicated that the use of a
rol-type MAT vector system enables the
genera-tion of marker-free transgenic plants without
sex-ual crossing in A. majus. Also for the purpose of
large number of marker-free transgenic plant needs increase independent hairy root lines.
Hairy root syndrome is a consequence of the
morphogenic action of rolgenes of the Ri plasmid
reduced production of pollens and seeds. Further, such phenotypes are stably passed to the progeny
in tobacco and A. majus [16,17,25].
In this study, five marker-free transgenic plants obtained from hairy roots, and their progenies, did not exhibit symptoms of Ri syndrome (Fig. 2A, B, Fig. 3A). These results suggest that excised recom-bination sequences were not integrated into
differ-ent loci on chromosomes of marker-free
transformants.
This study proved that the chimeric rol genes
can be used as a selection marker forAgrobacter
-inum-mediated transformation of A. majus, with
the rol-type MAT vector can be used to study
genes controlling the morphogenesis of A. majus
plants.
Acknowledgements
We are grateful to Dr H. Ebinuma (Nippon Paper Industries, Co. Ltd.) for kindly supplying the MAT vector pNPI702. We also thank to Dr H. Ichikawa (Natl. Inst. Agrobiol. Res. Japan), Dr K. Sage-Ono, S. Ko, S. Tan and Mr C. Eun (Institute of Biological Sciences, University of Tsukuba) for excellent technical assistance and valuable discussion.
References
[1] M.W. Bevan, R.B. Flavell, M.D. Chilton, A chimeric antibiotic resistance gene as a selectable marker for plant cell transformation, Nature 304 (1983) 184 – 187. [2] C. Waldron, E.B. Murphy, J.L. Roberts, G.D.
Gustafson, S.L. Armour, S.K. Malcolm, Resistance to hygromycin B: a new marker for plant transformation studies, Plant Mol. Biol. 5 (1984) 103 – 108.
[3] K. Akama, H. Puchta, B. Hohn, Efficient Agrobac-terium-mediated transformation of Arabidopsis thaliana using the bar gene as selectable marker, Plant Cell Rep. 14 (1995) 450 – 454.
[4] R.J. Mathias, C. Mukasa, The effect of cefotaxime on the growth and regeneration of callus from four varieties of barley (Hordeum6ulgareL.), Plant Cell Rep. 6 (1987)
454 – 457.
[5] D.W. Catlin, The effect of antibiotics on the inhibition of callus induction and plant regeneration from cotyle-dons of sugarbeet (Beta 6ulgarisL.), Plant Cell Rep. 9
(1990) 285 – 288.
[6] R.G.F. Visser, E. Jacobsen, B. Witholt, W.J. Feenstra, Efficient transformation of potato (Solanum tuberosum L.) using a binary vector inAgrobacterinum rhizogenes, Theor. Appl. Genet. 78 (1989) 594 – 600.
[7] E.A. Shahin, K. Sukhapinda, R.B. Simpson, R. Spivey, Transformation of cultivated tomato by a binary vector in Agrobacterinum rhizogenes: transgenic plants with normal phenotypes harbor binary vector T-DNA, but no Ri-plasmid T-DNA, Theor. Appl. Genet. 72 (1986) 770 – 777.
[8] H. Ebinuma, K. Sugita, E. Matsunaga, M. Yamakado, Selection of marker-free transgenic plants using the isopentenyl transferase gene, Proc. Natl Acad. Sci. USA 99 (1997) 2117 – 2121.
[9] H. Ebinuma, K. Sugita, E. Matsunaga, M. Yamakado, Principle of MAT vector system, Plant Biotech. 14 (1997) 133 – 139.
[10] K. Sugita, E. Matsunaga, H. Ebinuma, Effective selec-tion system for generating maker-free transgenic plants independent of sexual crossing, Plant Cell Rep. 18 (1999) 941 – 947.
[11] H. Sommer, U. Bonas, H. Saedler, Transposon induced alterations in the promoter region affect transcription of the chalcone synthase gene of Antirrhinum majus, Mol. Gen. Genet. 211 (1988) 49 – 55.
[12] C. Lister, C. Martin, Molecular analysis of a transposon induced deletion of the nivea locus inAntirrhinum majus, Genetics 123 (1989) 417 – 425.
[13] R. Carpenter, E.S. Coen, Floral homeotic mutations produced by transposon-mutagenesis inAntirrhinum ma-jus, Genes Dev. 4 (1990) 1483 – 1493.
[14] E.S. Coen, J.M. Romero, S. Doyle, R. Elliott, G. Mur-phy, R. Carpenter, Floricaula: a homeotic gene required for flower development in Antirrhinum majus, Cell 63 (1990) 1311 – 1322.
[15] L. Tamagnone, A. Merida, A. Parr, S. Mackay, M.F. Culianez, K. Roberts, C. Martin, The AmMYB308 and AmMYB330 transcription factors fromAntirrhinum reg-ulate phenylpropanoid and lignin biosynthesis in trans-genic tobacco, Plant Cell. 10 (1998) 135 – 154.
[16] T. Handa, Genetic transformation ofAntirrhinum majus L. and inheritance of altered phenotype induced by Ri T-DNA, Plant Sci. 81 (1992) 199 – 206.
[17] I. Senior, P. Holford, R.N. Cooley, H.J. Newbury, Transformation of Antirrhinum majus using Agrobac-terium rhizogenes, J. Exp. Bot. 46 (1995) 1233 – 1239. [18] Y. Hoshino, M. Mii, Bialaphos stimulates shoot
regener-ation from hairy roots of snapdragon (Antirrhinum ma-jus L.) transformed by Agrobacterium rhizogenes, Plant Cell Rep. 17 (1998) 256 – 261.
[19] H. Iris, E. Nadia, S. Heinz, Z.S. Sommer, A protocol for transformation and regeneration of Antirrhinum majus, Plant J. 13 (1998) 723 – 728.
[20] T. Murashige, F. Skoog, A revised medium for rapid growth and bioassay with tobacco tissue cultures, Phys-iol. Plant. 15 (1962) 473 – 497.
[21] R. Deblaere, B. Bytebier, H. De Greve, F. Deboeck, J. Schell, M. Van Montagu, J. Leemans, Efficient octopine Ti plasmid derived vectors for Agrobacterium-mediated gene transfer in plants, Nucleic Acids Res. 13 (1985) 4777 – 4785.
[23] R.A. Jefferson, T.A. Kavanagh, M.W. Bevan, GUS fusions: beta-glucuronidase as a sensitive and versatile gene fusion marker in higher plants, EMBO J. 6 (1987) 3901 – 3907.
[24] M. Cui, K. Takayanagi, H. Kamada, S. Nishimura, T. Handa, Efficient shoot regeneration from hairy roots of
Antirrhinum majus L. transformed by rol type MAT vector system, ( submitted).
[25] D. Tepfer, Transformation of several species of higher plants byAgrobacterium rhizogenes: sexual transmission of the transformed genotype and phenotype, Cell 37 (1984) 959 – 967.
