当前位置:首页 >> 农学 >>

A disarmed binary vector from Agrobacterium tumefaciens functions in Agrobacterium rhizogenes


Plant Molecular Biology6: 403-415, 1986 ? Martinus Nijhoff Publishers, Dordrecht - Printed in the Netherlands

A disarmed binary vector from

Agrobacterium rhizogenes

Agrobacterium tumefaciens functions

in

Frequent co-transformation of two distinct T-DNAs
Robert B. Simpson, Albert Spielmann, Linda Margossian & Thomas D. McKnight 1

Molecular Biology Group, ARCO Plant Cell Research Institute, 6560 Trinity Court, Dublin, CA 94568, U.S.A. 1present Address: Biology Department, Texas A&M University, College Station, TX 77843, U.S.A.
Keywords: Agrobacterium tumefaciens, A. rhizogenes, hairy root, plant, transformation, vector

Summary
Binary Ti plasmid vector systems consist of two plasmids in Agrobacterium, where one plasmid contains the DNA that can be transferred to plant ceils and the other contains the virulence (vir) genes which are necessary for the DNA transfer but are not themselves stably transferred. We have constructed two nononcogenic vectors (pARC4 and pARC8) based on the binary Ti plasmid system of Agrobacterium tumefaciens for plant transformation. Each vector contains the left and right termini sequences from pTiT37. These sequences, which determine the extent of DNA transferred to plant cells, flank unique restriction enzyme sites and a marker gene that functions in the plant (nopaline synthase in pARC4 or neomycin phosphotransferase in pARC8). After construction in vitro, the vectors can be conjugatively transferred from E. coli to any of several Agrobacterium strains containing vir genes. Using A. rhizogenes strain A4 containing the resident Ri plasmid plus a vector with the nopaline synthase marker, we found that up to 50% of the hairy roots resulting from the infection of alfalfa or tomato synthesized nopaline. Thus, vector DNA encoding an unselected marker was frequently co-transferred with Ri plasmid DNA to an alfalfa or a tomato cell. In contrast, the frequency of co-transfer to soybean cells was difficult to estimate because we encountered a high background of non-transformed roots using this species. Up to five copies of the vector DNA between the termini sequences were faithfully transferred and maintained in most cases suggesting that the termini sequences and the vir genes from the Ri and Ti plasmids are functionally equivalent.

Introduction
Because of their natural ability to transfer DNA to plant cells, Agrobacterium tumefaciens and its Ti plasmids have been used as vectors to introduce foreign DNA into plants (as reviewed recently 27, 34, 54). Since the production of auxins and cytokinins by transformed cells is often incompatible with normal plant regeneration, it is frequently desirable to 'disarm' the plasmids by removing the oncogenes responsible for the synthesis of these growth regulators and to introduce selectable or screenable markers in their place. The large size of the Ti plasmids makes it necessary to use intermediate vectors. One intermediate vector method is

the 'co-integration' approach, whereby foreign DNA is inserted into a vector that cannot replicate in Agrobacterium, but can recombine with the Ti plasmid through a homologous portion of the vector, producing a co-integrate of the two plasmids (16, 21, 55). Another method is the binary vector approach whereby a foreign gene is inserted into a disarmed T-DNA which itself is joined to a broad host range replicon that can replicate in Agrobacterium (3, 6, 17, 24, 26, 29). The mechanism of T-DNA transfer from the bacteria to the plant is not known in detail, but at a minimum, transfer requires termini sequences and vir genes from the Ti plasmid in addition to bacterial chromosomal genes. The termini sequences

404 are imperfect 25 basepair direct repeats found flanking the T-DNA, at least one of which is required for the transfer (9). The vir genes are required for DNA transfer but are themselves not stably transferred to the plant (34). Although the vir genes and the T-DNA are normally part of the same bacterial replicon, a binary system is possible, in which these functions are on separate replicons (17, 24, 26). Since an oncogenic marker is not present in these disarmed vectors, other markers have to be used to identify genetically transformed plant cells. Some markers, such as enzymes which result in the production of opines, can be used to screen transformed tissue for the presence of opines (e.g., octopine, nopaline or agropine; 37). Other markers, such as enzymes which confer resistance to antibiotics, can be used to select transformed tissue which can grow in the presence of an antibiotic (e.g., kanamycin, chloramphenicol or methotrexate; 23). Agrobacterium rhizogenes (27) is considered a close relative of A. tumefaciens because of its similar mechanism of plant transformation based on DNA transfer to plants, the similar function of the vir genes, and the production of opines by transformed tissue. A. rhizogenes frequently produces transformed, hairy roots so the endogenous plasmid has been called the Ri ('root-inducing') plasmid. Hairy roots from several species have been regenerated into plants which contain T-DNA from the Ri plasmid (10, 12, 43, 44). In cases where plants can be regenerated from roots, the combination of vir genes from the Ri plasmid and a gene transfer vector derived from the T-DNA of A. tumefaciens may be the system of choice for gene transfer. We report here, the construction of binary vectors for use in A. tumefaciens or A. rhizogenes containing either, a selectable marker that confers kanamycin-resistance to transformed plant cells, or a marker that is easy to screen, nopaline synthase. Inoculation of several plant species with A. rhizogenes containing a vector resulted in hairy roots. With alfalfa and tomato, we could demonstrate frequent co-transfer of vector DNA and Ri plasmid DNA. Southern analysis of the roots has shown that in most cases the DNA was transferred and integrated faithfully into the plant genome.
Materials and methods

The procedure for transformation of Escherichia coli with plasmid DNAs is as described (1). Other manipulations of nucleic acids are essentially those described by Maniatis et al. (32), unless otherwise indicated. Restriction enzymes and pUC8 were obtained from Bethesda Research Laboratories. The BglII linkers were from New England Biolabs. The bacterial strains and plasmids used for these experiments are listed in Table 1.

Plasmid constructions
The vector pARC4 was constructed as illustrated in Fig. 1A. The plasmid pBstEII 9, 14 (52) carries the 1.5 kb EcoRI fragment 29 derived from Ti plasmid pTiT37. This fragment contains the left terminus of the T-DNA region located approximately 50 basepairs from the right end of the fragment (51, 53). The plasmid pT37H23, a generous gift from Scott Stachel, carries the 3.2 kb HindIII fragment 23 from Ti plasmid pTiT37 (18). From left to right, this fragment contains the 5' portion of the DNA encoding transcript 6b, the entire nopaline synthase or NOS gene, the right terminus sequence and a portion of the Ti plasmid which is not transferred to plant cells. The chimeric gene ' N O S / N P P was constructed by placing the coding region for the neomycin phosphotransferase II gene (NPT) from the bacterial transposon Tn5 under control of the transcriptional regulatory signals of the nopaline synthase gene (NOS). The 'NOS/NP'I" gene is part of the plasmid pNEO105 whose structure is shown schematically in the bottom panel of Fig. 2. The NOS gene fragment was derived from the plasmid pT37H23 (18). Based on the numbering convention of Depicker et al. (18), pNEO105 contains the nopaline synthase promotor (from the BclI site at position -265 to position + 30) and polyadenylation site (from the SphI site at position + 1136 to the HindlII site at position + 1972). The NPT fragment was derived from the plasmid pNEO (P-L Biochemicals) which contains the neomycin phosphotransferase II gene from Tn5 cloned into pBR322. Based on the numbering convention of Beck et al. (4), pNEO105 contains a portion of the NPT gene (from position 1543 to the SmaI site at

405
Table 1.

Plasmids name pBstEII 9, 14 pUC8 pT37H23 pUC8-22-21 pRK290 pRK2013 pARC 1 pRKMI pARC3 pARC4 pNEO pNEO 105 pARC8 pRUD26 pRUD27 pFW94
Eschericia coli strains

Description pTiT37 BstEIl fragments 9 & 14 cloned in pMB9 AmpR AmpR; pTiT37 HindIII fragment 23 cloned into pBR322 AmpR; pTiT37 EcoRI fragment 29 cloned into pUC8 IncP; TetR, derivative of pRK2 IncP; KanR; can mobilize pRK290 and its derivatives AmpR; pTiT37 HindIII fragment 23 cloned into pUC8-22-21 pRK290 with EcoRI site eliminated Deletion derivative of pARCI pARC3 cloned into BgIII site of pRKM1 NPTII gene from Tn5 cloned into pBR322 Nos/NPT gene in pBR322 Replacement of Nos gene in pARC4 with NOS/NPT gene Chimeric soybean SS/NPT gene in pARC4 Chimeric soybean SS/NPT gene in pARC4 Left T-DNA of pRiA4

Reference Yang & Simpson (52) Vieira & Messing (47) Depicker et al. (18) This paper. Ditta et al. (19) Ditta et al. (19) This paper. This paper. This paper. This paper. P-L Biochemicals. This paper. This paper. McKnight & Simpson, in preparation. McKnight & Simpson, in preparation. Huffman et al. (28)

HBI01 JM83
Agrobacterium strains

F-, hsdS20(rB-mB-), recA 13, ara-14, proA2 lacY1, galK2, rpsL20 (SMR), xyl-5, mtl-1, supE44, lambda ara, del lac-pro, strA, thi, Phi80dlacZ del M15

Boyer and Roulland-Dussoix (8) Vieira & Messing (47)

A4

Agrobacterium rhizogenes; pRiA4

White and Nester (48)

p o s i t i o n 2516), which includes the entire c o d i n g region f r o m the N P T gene. T h e vector p A R C 8 was c o n s t r u c t e d f r o m pNEO105 a n d p A R C 4 as illust r a t e d in Fig. lB.
Introduction into A g r o b a c t e r i u m

G r o w t h o n A B m i n i m a l sucrose m e d i u m c o n t a i n ing b i o t i n (49), plus 5 / ~ g / m l tetracycline selects for Agrobacteria c o n t a i n i n g the vector. F u r t h e r m o r e , the use o f m i n i m a l m e d i u m w i t h o u t a d d e d a m i n o acids selects a g a i n s t E. coli a m i n o acid a u x o t r o p h s such as HB101.
Inoculation a n d establishment o f hairy root cultures

T h e vectors were transferred f r o m E. coli to A. rhizogenes A 4 by c o n j u g a t i o n , in the presence o f a third b a c t e r i u m , E. coli strain HB101 c o n t a i n i n g the p l a s m i d pRK2013 (19) to m o b i l i z e the vectors.

P l a n t material: Stems f r o m t o b a c c o (Nicotiana

406

A

Eco~RI cORI EcoRI29
' f
,~, H

~ f / ~

B

o,i,
pARC 4 Amp (pUt J) lgl II |?o RI
[ 1

+?~
Tilt J!:!:i:

~

/~/,:,o
Eco RI

"

p>
l"Hind III

E(OIll 4111 II

Hind III

HindIII

Hindnl lira HI Iglll

~
S linkers \ ~
H ~ i n d

E?oRI ~LLP~L,.IEcoIII
IN

Ec: RI

Amp (p|R31l) mimNO.~liliTU i m HindIll IIJI

.+j,_, ~ o ~ ,,( (o,~ P ~ ~,,,
oriT
pAmC e

?

~

Igl II S , ~

Igl a |CO Ri

f

I

I I I HindIII ~ I H I Itll.

E??
i I

+
oriT pARC 4 Amp (pUtS) Amp
I (puca)

pi:i:!:!:i:i:!:i:.~l~:.. I
HindIII Dim HI

leo RI Tit li:i:i:~

Igl II ECO RI

HindIn lira HI Igl II

f

f

I

C

pARC 4

,
I lkb I

p~

,

I

i

_

*

1

pARC 8 m E ~ I ~ N,..N "

,,

J~, I ,
AMP

J

J

J ,,
NO6/NPT

,

,~

I

Jill
TET

407
Nopaline Synthase Gene (NOS)
H Sp Bc

Ba Sp Sp

H

tabacum, cv. Xanthi), tomato (Lycopersicum esculentum, cv. red cherry), soybean (Glycine max cv. Williams 82) and alfalfa (Medicago sativa cv.
CUF101) plants grown in a greenhouse were sterilized by soaking in 0.5°70 Ca(OC1)2 + 0.25°7o Tween 80 for 3 0 - 6 0 min and then washed 3 - 4 times with sterile distilled water. The stems were cut in sections ( 2 - 3 cm), inverted and transferred aseptically to solid hormone-free TM-1 medium (40) + 250 m g / l cefoxitin (Merck, Sharp and Dohme) or recently, cefotaxime (Calbiochem) in plastic boxes. Some of the Agrobacterium strains are resistant to ampicillin and to Carbenicillin but not to cefoxitin (11) or cefotaxime (23). Over a period of 6 months, a minim u m of 5 individual plants of each species were used. A bacterial suspension in stationary phase, taken from either solid or liquid selective medium containing tetracycline, was spread on the upper surface of the inverted stem section. The inoculated stem sections were incubated at 25 °C in low light with a 16 hr photoperiod. Roots emerging from the top of the stem section after 2 - 3 weeks were excised several days later when about 1 cm long and transferred to the same medium. They were grown in the dark at 25 °C and subcultured every 3 to 4 weeks. In order to select for kanamycin-resistant hairy roots, roots were transferred to hormone-free TM4- medium (40) containing cefotaxime (300 mg/1) and kanamycin ( 2 0 - 3 0 mg/1 for alfalfa, 3 0 - 50 m g / l for tomato, 25 - 30 mg/1 for tobacco, and 2 0 - 1 0 0 mg/1 for soybean). After 2 weeks roots which were still growing were considered to be kanamycin resistant. Higher levels of kanamycin in this assay slowed the growth of all roots, even those expressing high levels of NPT, while lower levels of

Neomycin Phosphotransferase Gene (NPT)
H Bg $m

Ba

Plasmid pNE0105 containing Chimeric Gene (NOSINPT) Bc Bg Ba
H E Bc

CC

T

C

G

NPT 2516 Linker NOS .1136 CATG CGCCCACCCC GGGATCCC CATrCATCAA

Fig. 2. Schematic drawings of the nopaline synthase gene (NOS; represented by the white bar) from the T37 Ti plasmid (first panel), the neomycin phosphotransferase gene (NPT; represented by striped bar) from Tn5 (second panel), and the plasmid pNEO105 containing the chimeric gene N O S / N P T (third panel) where pBR322 sequences are represented by the thin line. Plasmid pNEO105 is shown linearized at the Bcll site. The drawings are to indicate relative positions only and are not to scale. The following symbols are used for orientation: startsite of transcription (+ 1), startsite of translation (ATG), site of translational termination (UAA or UGA), and site of transcriptional termination (polyA). The 'atg' refers to a region of the N P T gene containing an ATG codon (other than the initiation ATG) that we did not include in the N O S / N P T gene. Bevan (6) and Fraley et al. (21) report that the elimination of this 'atg' in a chimeric gene construction similar to N O S / N P T results in a higher level of resistance to kanamycin. The symbols for restriction enzymes sites discussed in the text are the following: BarnHI (Ba), BclI (Bc), BgllI (Bg), EcoRI (E), HindlII (H), Smal (Sm), and Sphl (Sp). At the bottom are the DNA sequences in the vicinity of the BgllI and BamHI linkers in the N O S / N P T gene. The numbers refer to the numbering system of Depicker et al. (18) for the N O S gene and to the numbering convention of Beck et al. (4) for the N P T gene. Note that the N O S / N P T gene is drawn in the opposite orientation to that in Fig. 1.

Fig. 1. Schematic drawings of the steps in the construction of vectors pARC4 (panel A) and pARC8 (panel B). Fragments from pTiT37 containing the left terminus sequence (EcoRl 29) and the nopaline synthase gene plus right terminus sequence (HindlII 23) were sequentially cloned into pUC8 creating pUC8-22-21 and then pARC1. To eliminate a HindllI site and an EcoRI site in pARC1, it was digested with BstEII, the ends were filled in, a BgllI linker was inserted, and it was cut with BgllI and religated to form pARC3. After linearization with BgllI, pARC3 was inserted into the BgllI site of pRKM1 (a derivative of pRK290 without an EcoRI site) to create the binary vector pARC4. The binary vector pARC8 was created by the replacement of the 'small' EcoRIBamHl fragment from PARC4 with the 'large' EcoRI-BclI fragment of pNEO105. Note that the the N O S / N P T g e n e is drawn in the opposite orientation to that in Fig. 2. The symbols used for regions of DNA are the following: bacterial genes for resistance to tetracycline (Tet) and ampicillin (Amp); origin of vegetative DNA replication (oriV); origin of conjugative transfer (oriT); pBR322 or iaUC8 sequences (open bar); pTiT37 EcoRI fragment 29 (bar with large dots); pTiT37 Hind fragment 23 (bar with small dots); pRK290 sequences (thin line); right and left termini sequences (flags); pTiT37 nopaline synthase gene (NOS); N O S / N P T gene (striped bar). The location of restriction enzyme sites discussed in the text are indicated. Where the name has been crossed out, the restriction site has been eliminated. The vector pARC4 is about 26 kb including a transferred DNA of about 5 kb while pARC8 is about 28 kb including a transferred DNA of about 7 kb. Panel C, which is drawn to scale, contains detailed restriction enzyme maps of the region around the transferred DNA in pARC4 and pARC8.

408 kanamycin permitted some roots without a NPT gene to continue growing for 2 weeks. Kanamycinresistant roots all contained N P T II activity based on the native gel assay of Reiss et al. (39). To increase the quantity of tissue rapidly, we transferred the roots to the same medium without kanamycin. Prior to DNA isolation, roots were propagated without cefotaxime to verify that they were free of bacteria. In preparation for hypocotyl infections, seeds were sterilized by soaking in 70070 ethanol for 2 minutes, then in 20°7o of commercial bleach for 30 minutes and finally washed 3 - 4 times with sterile distilled water. The seeds were germinated on TM-1 medium in plastic boxes. Seedlings were infected with Agrobacterium after 2 - 3 weeks by wounding the hypocotyl with a needle tip covered with stationary-phase bacteria. The roots were then handled as described above.

Nopaline assay and DNA analysis
For the nopaline assay, tissue (20-100 mg) in a 1.5 ml Eppendorf tube was homogenized using a wooden applicator stick. The debris were pelleted for 5 min in a Brinkman Eppendorf microcentrifuge. Up to 50 tzl of supernatant was pipetted onto 5 mm filter paper discs (Whatman #3). The discs were air-dried and then placed at the origin prior to high voltage paper electrophoresis and staining essentially as described by Otten and Schilperoort (37). For DNA analysis, gels were prepared for Southern transfer and hybridization essentially as described in Thomashow et al. (45).

Results

Construction of the binary vector pARC4
Figure 1A illustrates the steps in the construction of our first vector, pARC4. For the purpose of DNA manipulations in bacteria, the vector contains a wide host range bacterial replicon, a bacterial origin of transfer, a bacterial antibiotic resistance marker, and unique restriction enzyme sites. The wide host range origin of replication (oriV), derived from the wide host range plasmid RK2 (19), permits replication of the plasmid in both Es-

cherichia coli and in Agrobacterium. In contrast, pBR322 and ColE1 replicons are unable to replicate in Agrobacterium. The origin of transfer (oriT), also from plasmid RK2, permits the vector to be mobilized by a helper plasmid; mobilization is the most efficient means to introduce plasmids into Agrobacterium. The antibiotic marker, bacterial resistance to tetracycline (Tet) from plasmid RK2, allows selection of bacteria containing the vector. The unique restriction enzyme sites, EcoRI and HindlII, facilitate the insertion in vitro of 'foreign' DNA fragments into the vector. For the purpose of transferring DNA to plant cells, pARC4 contains the signals (termini sequences) which Agrobacterium uses to delimit the DNA it transfers to plant cells - the foreign DNA is placed between these two signals. The termini sequences, denoted by flags in Fig. 1, are both derived from pTiT37, the left from EcoRI fragment 29 and the right from Hind III fragment 23 (51, 53). Although, in some cases, a single terminus region has been shown to be all that is absolutely required for transfer (9), none of the resulting transfers have been characterized. Both termini sequences were therefore used to increase the likelihood of predictable and reliable transfer. To identify plant cells that contain vector DNA, pARC4 contains the nopaline synthase gene (NOS) from the T37Ti plasmid as part of the transferred DNA. A rapid assay of plant cells can identify those which synthesize nopaline and thus contain the transferred DNA. In addition to these essential components, the vector contains within the transferred DNA the bacterial ampicillin-resistance gene (Amp) and the narrow host range ColE1 origin of replication derived from pUC8. These could prove useful in rescuing the transferred DNA from the plant subsequent to transformation. Also, during some manipulations with the vector, such as the replacement of the small E c o R I / H i n d l I I fragment in pARC4 with a foreign gene, 'Ampscreen' (Bethesda Research Laboratories) has been used to screen for the loss of ampicillin resistance. Finally, the addition of the COLE1 origin of replication permits one to amplify the plasmid in E. coli (but not in Agrobacteria). This facilitates rapid, small-scale plasmid isolations and characterizations, as well as the initial preparation of the vector.

409

Construction of the selectable marker ' N O S / N P T ' and vector pARC8
Shown in Fig. 2 is the structure of the plasmid pNEO105 which contains the chimeric gene NOS/NP1], a selectable marker for transformed plant cells. The 'NOS/NPI" gene consists of the promotor and transcriptional termination site of the nopaline synthase gene (NOS) from pTiT37 and the coding region from the neomycin phosphotransferase II (NPT) from the bacterial transposon Tn5. As illustrated in Fig. 1B, the vector pARC8 is similar to pARC4 except that the NOS marker is replaced by the N O S / N P T marker, which functions in plant cells to produce the enzyme neomycin phosphotransferase II. Plant cells containing the enzyme are resistant to kanamycin. Further, the presence of the enzymatic activity can be assayed directly (39).

Table 2. Transfer of nopaline synthesis and kanamycin resistance markers into roots resulting from infection of 4 plant species with A. rhizogenes containing pARC4 type or pARC8 type vectors.
Plant system Proportion of independent roots~ Nopaline Positive (pARC4 Type) Alfalfa Stem Section Hypocotyl Tomato Hypocotyl Tobacco Stem Section Soybean Stem Section Hypocotyl Kanamycin Resistant (pARC8 Type)

63/115 (55070) la/43 (33070) 8/24 (33%) NT 1/58 (207o) 2/108 (2070)

NT* 2/40 (4%) 18/95 (19%) 16/90 (18%) 0/80 (0070) NT

Introduction into Agrobacterium
The vectors pARC4, pARC8 and their derivatives were transferred by conjugation to several Agrobacterium strains including A4 (48) and LBA4404 (36). In 2 out of 68 occasions, we found that a vector construction had a different structure in Agrobacterium than it had in E. coli. Klee et al. (29) have also observed an alteration in the structure of a binary vector after transfer from E. coil to Agrobac-

t As discussed in the text, the alfalfa, tomato and tobacco roots were hairy roots while the majority of the soybean roots were untransformed. * NT - Not Tested.

terium. Plant transformation
Using Agrobacterium rhizogenes strain A4, containing either the nopaline (pARC4) type vector or the N O S / N P T (pARC8) type vector, we inoculated inverted stems or hypocotyls of tobacco, tomato, alfalfa and soybean. The resulting roots were excised and transferred to hormone-free media and grown as separate root clones. Neither uninoculated tissue nor tissue inoculated with a disarmed A. tumefaciens strain (LBA4404, ref. 36) produced roots with tomato, tobacco or alfalfa. In contrast, such controls frequently produced roots with soybean. Table 2 summarizes the results of assays on the roots from each of the four species either for the presence of nopaline or for ability to grow in the presence of kanamycin. Since the type of insert

had no apparent effect on the results, we report the data grouped by type of vector. The results include pARC4, pARC4 containing 8 different inserts ranging in size from 3 kb to 7 kb, pARC8, and pARC8 containing a 4 kb insert or a 7 kb insert. Only pARC8 was used with all four species. Surprisingly, over 50% of the roots, derived from the infection of alfalfa stem sections with A4 containing pARC4 or a pARC4 derivative, synthesized nopaline. The percentage of nopaline positive roots in pARC4 experiments was higher than the percentage of kanamycin resistant roots in pARC8 experiments using either alfalfa hypocotyls, tomato hypocotyls or soybean stem sections. Since the nopaline assay was a screen and the kanamycin resistance was used as a selection, one explanation is that the level of kanamycin used for selection was too high for some transformed roots to survive. Kanamycin slowed the growth of hairy roots in our experiment, even those expressing high levels of NPE. Also, kanamycin promotes an alternative developmental pathway, shooting, in tobacco and carrot (38). Another explanation is that DNA transfer from pARC4 may be more efficient than DNA transfer from pARCS.

410

Analysis of DNA from transformed plant cells
DNA was isolated from roots, digested with one of several different restriction enzymes, fractionated by gel electrophoresis and transferred to nitrocellulose. The resulting Southern blots were probed with radiolabelled DNA corresponding to portions of the vectors. If there were faithful transfer of DNA to the plant cell, we would expect all of the DNA between the termini sequences to have been transferred. Examples of faithful and aberrant transferred copies are shown in Fig. 3. Constructs pRUD26 and pRUD27 contain two different versions of a chi-

Fig. 3. Southern blot analysis of DNA from soybean hairy
roots incited by A4(pRUD26) or A4(pRUD27). The DNA in each lane was digested with both EcoRI and Hindill, electrophoresed, blotted and hybridized with nicktranslated pK26 which contains 1.5 kb of soybean DNA representing the SRS 2.1 small subunit promotor (McKnight, T. D. and Simpson, R. B., unpublished). Lanes 1 and 2 show 5 copy- and 1 copy-per haploid genome reconstructions for pRUD26. The lower band in these lanes indicates the expected size (3.6 kb) for a faithful, full length transfer of this fragment to the plant genome. Lane 3 contains DNA from untransformed soybean leaves and shows the 1.5 kb band representing the endogenous soybean small subunit 10romotor fragment from which the probe was derived. Lanes 4 and 5 represent DNA from 2 independent soybean hairy root lines (A2 and M3) arising from infection with A4(pRUD26). Lanes 6 and 7 represent DNA from 2 independent soybean hairy root lines (K1 and Q1) arising from infection with A4(pRUD27). Lanes 8 and 9 show 5 copy- and 1 copy-per haploid genome reconstructions for pRUD27. The lower band in these lanes indicates the expected size (2.4 kb) for a faithful, full length transfer of this fragment to the plant genome.

meric gene inserted into pARC4 (McKnight and Simpson, in preparation). The chimeric genes are composed of portions of a gene from soybean (SRS 2.1 from ref. 5) encoding the small subunit of ribulose bisphosphate carboxylase (SS) and the coding region of the neomycin phosphotransferase gene (NP'/~ ref. 4). The probe and the hybrid SS/NPT genes were derived from a 1.5 kb soybean small subunit promotor fragment. Lane 3, which contains DNA from untransformed soybean leaves, has a 1.5 kb band representing the endogenous small subunit fragment. This band is also visible in the other lanes containing soybean DNA. The soybean hairy root lines A2 and M3 resulted from infection by A4(pRUD26). The DNA from line A2 (lane 4) shows a band of hybridization of about single copy intensity with the same mobility as the 3.6 kb band in the reconstructions (lanes 1 and 2). This suggests that there has been a faithful transfer of this fragment from pRUD26 to the soybean genome. DNA from the soybean hairy root line M3 in lane 5, shows several bands which hybridize to the probe. These bands are present at near single copy levels, but none are of the expected size suggesting that they are aberrant copies. Lanes 6 and 7, contain DNA from two independent soybean hairy root lines transformed by A4(pRUD27). In both lanes a single copy, full length band can be seen with a mobility of 2.4 kb. In addition to this faithfully transferred fragment, the DNA of line K1 in lane 6 has a fragment of higher molecular weight. One likely explanation for this larger size is that transfer of the T-DNA stopped short of the EcoRI site which is near the left termini sequence of pARC4 (see Fig. 1) and thus the copy is an aberrant one. This larger fragment presumably ends at an EcoRI site near the integration site in the soybean genome. Fig. 4 summarizes our conclusions about the copy number and structural integrity of the DNA in 12 root lines from the infection of soybean, alfalfa and tobacco based on Southern blot analyses. For each of the lines analyzed, the copy number was estimated by comparing the intensity of the hybridization signal to the reconstructions, in addition to the size and number of non-internal (border) fragments derived, for example, from restriction with EcoRI or HindlII alone (data not shown). The figure shows the number of root lines with the indicated approximate copy number.

411 data), transfers DNA to plants. As described in the Results section, the vectors contain several features which allow them to be easily manipulated. Also, for the identification of transformed plant cells, one vector (pARC4) contains the nopaline synthase (NOS) marker while the other (pARC8) contains a chimeric gene composed of regulator signals from the NOS gene driving the expression of the neomycin phosphotransferase marker (NOS/NPT). The use of binary vectors introduces additional flexibility to plant transformation approaches. Once a vector construction is complete, it can be used without modification in any of several different Agrobacterium strains including LBA4404 (unpublished work; 3, 6, 29), an Agrobacterium mutant containing the functional vir genes but no T-DNA (36). Our use of binary vectors in Agrobacterium rhizogenes permits the unusual regeneration potential of hairy roots (10, 12, 43, 44) to be exploited in conjunction with an efficient vector system. The choice of plant species inoculated made a considerable difference to the proportion of roots that were positive for the vector marker nopatine synthase (Table 2). It was technically easy to isolate a reasonable number of hairy roots that contain the appropriate marker using alfalfa, tomato or tobacco where up to half of the roots assayed w e r e positive. However, to find positive soybean roots perhaps ten times more roots must be assayed since only a few percent of the soybean roots arising from the infection site were hairy roots. Perhaps even a high percentage of soybean roots which were hairy roots, synthesized nopaline. This conjecture is based on the high background production of soybean roots in the absence of Agrobacteria and on the observation that nopaline-containing roots had more lateral roots and an increased growth rate compared to normal soybean roots. Each of the seven nopaline positive roots of alfalfa examined by Southern blot analysis contained both vector DNA and Ri plasmid DNA, confirming that the roots were in fact transformed, hairy roots. Thus, there was a high frequency of cotransfer to a plant cell of two distinct T-DNAs which originate on two separate bacterial replicons. Binary vectors have been derived from two different broad host range plasmids (pR772 and pRK2), the vectors contain termini sequences from either an octopine Ti plasmid (pTiAch5 or pTiA6) or a

8"

A - Affalfa S-

Soybean

Number 6 " of Lines
4"

Tb. Tobacco

2"

1

2

3

4

Copy Number of Vector DNA

Fig. 4. Copy number and structural integrity of vector DNA
found in transformed roots. Summary of the structural analysis of DNA from 12 root cultures initiated by Agrobacteriurn rhizogenes A4 containing the vectors pARC4, pARC8 or one of these vectors with an insert. Each culture is represented by a box where the species is as indicated. The approximate number of copies of the DNA transferred from the vector is based on Southern blot analyses. A box with hatching denotes that one or more of the copies was 'aberrant' as described in the text.

Fig. 4 also illustrates which lines contained at least one 'aberrant copy' as indicated by the presence of bands on the Southern blot other than the expected internal fragments when an internal probe was used. The approximate copy number of the vector DNA in these lines varied from one to four. Of the 12 lines, 9 contained the expected portion of the vector DNA. The Ri plasmid from A. rhizogenes strain A4 contains two separate DNAs that can be transferred to plant cells. The presence of the left TDNA in 7 independent alfalfa root lines was examined by probing with pFW94 (28). In each case, transfer of the left T-DNA was indicated by the presence of bands with the mobilities expected of the 4.2, 3.4, 1.8 and 1.6 kb HindlII fragments internal to the left T-DNA (28), as well as other bands that probably represent border fragments.

Discussion We have constructed two binary, disarmed vectors based on the Ti plasmid of Agrobacterium tumefaciens. Each vector, in several different Agrobacterium strains (this work and unpublished

412 n o p a l i n e Ti p l a s m i d (pTiT37), a n d the vectors function in Agrobacteria c o n t a i n i n g any one o f four families o f T i / R i p l a s m i d s (3, 6, 17, 24, 25, 29). Indeed, T - D N A can be transferred to the p l a n t even when moved to the bacterial c h r o m o s o m e while the vir genes r e m a i n p l a s m i d b o r n e (26). Thus, it is unlikely t h a t the two p l a s m i d s m u s t be a co-integrate to transfer D N A f r o m the vector (17, 24). T h e t r a n s f o r m e d roots are o r g a n clones a n d also p r o b a bly cellular clones (12) suggesting t h a t p l a n t cells frequently can take up a second unlinked m a r k e r in a d d i t i o n to the first, selected marker. The d a t a at this p o i n t d o n o t show a significant difference in the n u m b e r o f copies o f vector TD N A per p l a n t cell using the two different vector a p p r o a c h e s . The range r e p o r t e d in this p a p e r (Fig. 4) for a b i n a r y vector, a p p r o x i m a t e l y 1 - 5 , is consistent with the range for t r a n s f o r m e d p l a n t s derived from a n o t h e r b i n a r y vector (29) a n d from c o - i n t e g r a t i o n vectors (13, 16, 20, 21, 55). However, 5 - 2 0 copies have been r e p o r t e d in p l a n t s derived f r o m a b i n a r y vector (6, 7, 29) a n d from a cointegration vector (14). T h e d a t a s u m m a r i z e d in Fig. 4 indicate that 9 o f 12 i n d e p e n d e n t l y isolated r o o t cultures c o n t a i n the p o r t i o n o f the vector D N A t h a t stretches from the left t e r m i n u s sequence to the right t e r m i n u s sequence. In each case, we ascertained the presence o f an E c o R I site which lies o n l y a b o u t 50 b a s e p a i r s inside the left t e r m i n u s sequence. It is thus a very sen-

Tumor Une
* ' ° Tobacco Tobacco Tobacco Tobacco Tobacco Tobacco TOINKco Tobacco Tobacco Alfalfa AHJa Sunflowc~r Ar~lopsis Tobacco Tobacco Tobacco Tobacco Tobacco TOINKco Alfeifa Petunia Tobacco Flax TObaCCO Tobacco

T-DNA
" A A A A A A A A A A A A A A A II AiB C C C C A&C A&B&C D E
! I

ll4MI0~9 Tu.4NSS3 WII*US3 SRI-I~S3 WE-A6 4013-1 4013-2 4001 42211 Vettus ACHS VemJs 116oct ? PSCG1SgS5 A-AchS *WMCM-I *liT3? UIA4013-4 Ilhl4 1S9SS/1 A2T//S Vertus lli octP-AchS 159S501 *FT37/1 4013-5 4013

J

A D

]

m

U

Fig. 5. Summary of T-DNA Structures in crown gall tumors. On the left is a list of crown gall tumors incited by wildtype Agrobacterium tumefaciens octopine strains with the exception of the three lines preceeded by an asterisk (*) which were incited by wildtype nopaline strains. The letter or letters under the column headed 'T-DNA' indicate the structure or structures of the T-DNA in that line which were established by Southern blot analysis. We did not include data from tumors incited by insertion mutants or the data concerning the right T-DNA of the octopine Ti plasmid. The letters indicate whether the data is consistent with the T-DNA ending at the termini sequences (A), not reaching the left terminus sequence (B), not reaching the right terminus sequence (C), not reaching either terminus sequence (D) or extending past the left terminus sequence (E). On the right are schematic illustrations of each type of T-DNA structure. They are not drawn to scale. The structures are derived from the following references: 45 (A6S/2, B6806/E9, 15955/1, A277/5); 15 (Tu-B6S3, WB-B6S3, SR1-B6S3, WB-A6, AAchS, P-AchS); 35 (4013-1,4013-2, 4001, 4229, 4013-5, 4013); 33 (Vertus ACHS, Vertus B6 oct+, Vertus B6 oct-); 46 (PSCG15955); 31 (W38C58-1, BT37); 50 (LBA4013-4); 2 (Bla4); 30 (1595501); 22 (FT37/l).

413 sitive indicator of transfer near the left terminus sequence. Published DNA sequence analysis of the extent of T-DNA transfer from wild type nopaline Ti plasmids showed that one T-DNA ended in the left terminus sequence and included this EcoRI site but three others ended about 20 to 30 basepairs to the right of this EcoRI site (51, 53). Our analysis did not include restriction enzyme sites closer than about 1.5 kb from the right terminus sequence. However, the roots did synthesize nopaline and Shaw et al. (41) have shown that the nopaline synthase promotor requires sequences including those less than 350 basepairs from the right terminus sequence. Also, all four of the nopaline T-DNAs, which were characterized by DNA sequence analysis, ended within one base of the right terminus sequence (53). The analysis of DNA from transformed roots indicates that the termini sequences of A. tumefaciens can function in A. rhizogenes. The termini sequences and other interacting factors appear to be functionally interchangable between octopine and nopaline wide-host-range Ti plasmids, octopine narrow-host-range Ti plasmids, and Ri plasmids (3, 17, 24, 25). Recent DNA sequence analysis confirms that Ri and Ti plasmids have similar termini sequences (42). At least 3 of the 12 root cultures that we examined contain a minimum of one 'aberrant' copy of the vector T-DNA. This frequency can be compared to the frequency of aberrant copies from wild type Ti plasmids which we have summarized from the literature in Fig. 5. The structure of T-DNA copies (based on Southern blot analysis) in 26 tumors is indicated by letters. Data consistent with a 'perfect' transfer (A) is found in 16 cases but in 9 cases, at least one copy of the T-DNA does not reach the left terminus sequence (B) or does not reach the right terminus (C) or does not reach either terminus (D). In one case the T-DNA stretched past the left terminus sequence (E). Thus, the frequency of faithful transfer from our vectors is similar to the frequency with wild type Ti plasmids. to publication, Jerry Slightom for his manuscript prior to publication, Karen Long for typing the manuscript and for help with the figures, Elias Shaheen for valuable discussions, Mayer Yashar for excellent technical support, Dale Cetlinski for artistic support, and Simon Bright, Jack Erion, Val Williamson, Phil Filner and Elias Shaheen for criticisms of the manuscript. For a portion of this work, A.S. was supported by the Swiss National Science Foundation.

References
1. Alexander DC, McKnight TD, Williams BG: A simplified and efficient vector-primer cDNA cloning system. Gene 31:79-89, 1984. 2. Amisino RM, Powell ALT, Gordon MP: Changes in TDNA methylation and expression are associated with phenotypic variation and plant regeneration in a crown gall tumor line. Mol Gen Genet 197:437-446, 1984. 3. An G, Watson BD, Stachel S, Gordon MP, Nester EW: New cloning vehicles for transformation of higher plants. EMBO J 4:277-284, 1985. 4. Beck E, Ludwig G, Auerswald EA, Reiss B, Schaller H: Nucleotide sequence and exact localization of the neomycin phosphotransferase gene from transposon Tn5. Gene 19:327-336, 1982. 5. Berry-Lowe SL, McKnight TD, Shah DM, Meagher RB: The nucleotide sequence, expression, and evolution of one member of a multigene family encoding the small subunit of ribulose-l,5-bisphosphatecarboxylase in soybean. J Mol Appl Gen 1:483-498, 1982. 6. Bevan, M. (1984). Binary Agrobacterium vectors for plant transformation. Nucleic Acids Res. 12:8711-8721, 1984. 7. Bevan MW, Mason SE, Goelet P: Expression of tobacco mosaic virus coat protein by a cauliflower mosaic virus promoter in plants transformed by Agrobacterium. EMBO 3 4:1921 - 1926, 1985. 8 Boyer HW, Roulland-Dussoix D: A complementation analysis of the restriction and modification of DNA in Escherichia coli. J Mol Biol 41:459-472, 1969. 9. Caplan AB, Van Montagu M, Schell J: Genetic analysis of integration mediated by single T-DNA borders. J Bacteriol 161:655 - 664, 1985. 10. Chilton M-D, Tepfer DA, Petit A, David C, Casse-Delbart F, Tempe J: Agrobacterium rhizogenes inserts T-DNA into the genomes of the host plant root cells. Nature (Lond) 295:432-434, 1982. 11. Christens AA, Kirkpatrick MA, Lurquin PF: Antibiotics and bacterial auxotrophic mutants in the co-cultivation of plant protoplasts and bacterial cells or speroplasts. Z Pflanzenphysiol Bd 113:213 - 221, 1984. 12. David C, Chilton M-D, Tempe J: Conservation of T-DNA in plants regenerated from hairy root cultures. Biotechnology Jan, 73-76, 1984. 13. De Block M, Herrera-Estrella L, Van Montagu M, Schell J, Zambryski P: Expression of foreign genes in regenerated plants and in their progeny. EMBO J 3:1681 - 1689, 1984.

Acknowledgements
We thank Scott Stachel for clone pT37H23, Arnie Horwitz for the construction of pNEO105, Frank White for pFW94 and his manuscript prior

414 14. De Block M, Schell J, Van Montagu M: Chloroplast transformation by Agrobacterium tumefaciens. EMBO J 4:1367- 1372, 1985. 15. De Beuckeleer M, Lemmers M, DeVos G, Willmitzer L, Van Montagu M, Schell J: Further insight on the transferred-DNA of octopine crown gall. Mol Gen Genet 183:283- 288, 1981. 16. Deblaere R, Bytebier B, De Greve H, Deboeck F, Schell J, Van Montagu M, Leemans J: Efficient octopine Ti plasmid-derived vectors for Agrobacterium-mediated gene transfer to plants. Nucleic Acids Res 13:4777-4788, 1985. 17. de Framond A, Barton KA, Chilton M-D: Mini-Ti: a new vector strategy for plant genetic engineering. Biotechnology 1:262-269, 1983. 18. Depicker A, Stachel S, Dhaese P, Zambryski P, Goodman HM: Nopaline synthase: transcript mapping and DNA sequence. J Mol Appl Genet 1:561- 573, 1982. 19. Ditta G, Stanfield S, Corbin D, Helinski DR: Broad host range DNA cloning system for gram-negative bacteria: construction of gene bank of Rhizobium meliloti. Proc Natl Acad Sci USA 77:7347-7351, 1980. 20. Fraley RT, Horsch RB, Matzke A, Chilton M-D, Chilton WS, Sanders PR: In vitro transformation of petunia cells by an improved method of co-cultivation with A. tumefaciens strains. Plant Mol Biol 3:371- 378, 1984. 21. Fraley RT, Rogers SG, Horsch RB, Eichholtz DA, Flick JS, Fink CL, Hoffmann NL, Sanders PR: The SEV system: a new disarmed Ti plasmid vector system for plant transformation. Biotechnology 3:629- 635, 1985. 22. Hepburn AG, Clarke LE, Blundy KS, White J: Nopaiine Ti plasmid, pTiT37, T-DNA insertions into a flax genome. J Mol Appl Genet 2:211- 224, 1983. 23. Herrera-Estrella L, De Block M, Messens E, Hernalsteens J-P, Van Montagu M, Schell J: Chimeric genes as dominant selectable markers in plant cells. EMBO J 2:987 - 995, 1983. 24. Hoekema A, Hirsch PR, Hooykaas P J J, Schilperoort RA: A binary plant vector strategy based on separation of virand T-region of the Agrobacterium tumefaciens Tiplasmid. Nature (Lond) 303:179-180, 1983. 25. Hoekema A, Hooykaas P J, Schilperoort RA: Transfer of the octopine T-DNA segment to plant cells mediated by different types of Agrobacterium tumor- or root-inducing plasmids: generality of virulence systems. J Bacteriol 158:383- 385, 1984. 26. Hoekema A, Roelvink PW, Hooykaas P J J, Schilperoort RA: Delivery of T-DNA from the Agrobacterium tumefaciens chromosome into plant cells. EMBO J 3:2485- 2490, 1984. 27. Hooykaas P J J, Schilperoort RA: The molecular genetics of crown gall tumorigenesis. Advances in Genetics 22:209-283, 1984. 28. Huffman GA, White FF, Gordon MP, Nester EW: Rootinducing plasmid: physical map and homology to tumorinducing plasmids. J Bacteriol 157:269- 276, 1984. 29. Klee H J, Yanofsky MF, Nester EW: Vectors for transformation of higher plants. Biotechnology 3:637- 642, 1985. 30. Kwok WW, Nester EW, Gordon MP: Unusual plasmid DNA organization in an octopine crown gall tumor. Nucleic Acids Res 13:459- 471, 1985. 31. Lemmers M, De Beuckeleer M, Holsters M, Zambryski P, Depicker A, Hernalsteens JP, Van Montagu M, Schell J: Internal organization, boundaries and integration of Tiplasmid DNA in nopaline crown gall tumours. J Mol Biol 144:353 - 376, 1980. 32. Maniatis T, Fritsch EF, Sambrook J: Molecular cloning, a laboratory manual. Cold Spring Harbor Laboratory. 545 pages, 1982. 33. Mariotti D, Davey MR, Draper J, Freeman JP, Cocking EC: Crown gall tumorigenesis in the forage legume Medicago sativa L. Plant Cell Physiol 25:473- 482, 1984. 34. Nester EW, Gordon MP, Amasino RM, Yanofsky MF: Crown gall: a molecular and physiological analysis. Annu Rev Plant Physiol 35:387-413, 1984. 35. Ooms G, Bakker A, Molendijk L, Wullems G J, Gordon MP, Nester EW, Schilperoort RA: T-DNA organization in homogeneous and heterogeneous octopine-type crown gall tissues of Nicotiana tabacum. Cell 30:589- 597, 1982. 36. Ooms G, Hooykaas P J J, Van Veer RJM, Van Beelen P, Regensburg-Tuink TJG, Schilperoort RA: Octopine Tiplasmid deletion mutants of Agrobacterium tumefaciens with emphasis on the right side of the T-region. Plasmid 7:15- 29, 1982. 37. Otten LABM, Schilperoort RA: A rapid micro scale method for the detection of lysopine and nopaline dehydrogenase activities. Biochim Biophys Acta 527:497-500, 1978. 38. Owens LD: Kanamycin promotes morphogenesis of plant tissues. Plant Sci Lett 16:225-230, 1979. 39. Reiss B, Sprengel R, Will H, Schaller H: A new sensitive method for qualitative and quantitative assay of neomycin phosphotransferase in crude cell extracts. Gene 30:211-218, 1984. 40. Shahin EA: Totipotency of tomato protoplasts. Theor Appl Genet 69:235 - 240, 1985. 41. Shaw CH, Carter GH, Watson MD, Shaw CH: A functional map of the nopaline synthase promoter. Nucleic Acids Res 12:7831 - 7846, 1984. 42. Slightom JL, Jouanin L, Leach F, Drong RF, Tepfer D: Isolation and identification of TL-DNA/plant junctions in Convolvulus arvensis transformed by Agrobacterium rhizogenes strain A4. The EMBO J 4:3069-3077, 1985. 43. Spano L, Costantino P: Regeneration of plants from callus cultures of roots induced by Agrobacterium rhizogenes on tobacco. Z Pflanzenphysiol 106:87-92, 1982. 44. Tepfer D: Transformation of several species of higher plants by Agrobacterium rhizogenes: sexual transmission of the transformed genotype and phenotype. Cell 37:959- 967, 1984. 45. Thomashow MF, Nutter R, Montoya AL, Gordon MP, Nester EW: Integration and organization of Ti plasmid sequences in crown gall tumors. Cell 19:729-739, 1980. 46. Ursic D, Slightom JL, Kemp JD: Agrobacterium tumefaciens T-DNA integrates into multiple sites of the sunflower crown gall genome. Mol Gen Genet 190:494-503, 1983. 47. Viera J, Messing J: The pUC plasmids, an M13mp7-derived system for insertion, mutagenesis and sequencing with synthetic primers. Gene 19:259-268, 1982. 48. White FF, Nester EW: Relationship of plasmids responsible for hairy root and crown gall tumorigenicity. J Bacteriol 144:710-720, 1980.

415 49. White FF, Nester EW: Hairy root: plasmid encodes virulence traits in Agrobacterium rhizogenes. J Bacteriol 141:1134- 1141, 1980. 50. Wostemeyer A, Otten LABM, Schell JS: Sexual transmission of T-DNA in abnormal tobacco regenerants transformed by octopine and nopaline strains of Agrobacterium tumefaciens. Mol Gen Genet 194:500-507, 1984. 51. Yadav NS, Vanderleyden J, Bennett DR, Barnes WM, Chilton M-D: Short direct repeats flank the T-DNA on a nopaline Ti plasmid. Proc Natl Acad Sci USA 79:6322- 6326, 1982. 52. Yang F, Simpson RB: Revertant seedlings from crown gall tumors retain a portion of the bacterial Ti plasmid DNA sequences. Proc Natl Acad Sci USA 78:4151-4155, 1981. 53. Zambryski P, Depicker A, Kruger K, Goodman HM: Tumor induction by Agrobacterium tumefaciens: analysis of the boundaries of T-DNA. J Mol Appl Genet 1:361- 370, 1982. 54. Zambryski P, Herrera-Estrella L, De Block M, Van Montagu M, Schell J: The use of the Ti plasmid of Agrobacterium to study the transfer and expression of foreign DNA in plant cells: new vectors and methods. In Genetic Engineering, principles and methods. JK Setlow and A Hollaender, ed. Plenum Press: New York. pp. 253-278, 1984. 55. Zambryski P, Joos H, Genetello C, Leemans J, Van Montagu, Schell J: Ti plasmid vector for the introduction of DNA into plant ceils without alteration of their normal regeneration capacity. EMBO J 2:2143-2150, 1983. Received 11 November 1985; in revised form on March 1986; accepted 18 March 1986.


相关文章:
更多相关标签: