CA1251386A - Selection using opine synthase genes - Google Patents

Selection using opine synthase genes

Info

Publication number
CA1251386A
CA1251386A CA000560626A CA560626A CA1251386A CA 1251386 A CA1251386 A CA 1251386A CA 000560626 A CA000560626 A CA 000560626A CA 560626 A CA560626 A CA 560626A CA 1251386 A CA1251386 A CA 1251386A
Authority
CA
Canada
Prior art keywords
dna
plant
gene
cells
fragment
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired
Application number
CA000560626A
Other languages
French (fr)
Inventor
Gary A. Dahl
Dennis W. Sutton
Richard F. Barker
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Agrigenetics Research Associates Ltd
Original Assignee
Agrigenetics Research Associates Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from CA 462886 external-priority patent/CA1251386C/en
Application filed by Agrigenetics Research Associates Ltd filed Critical Agrigenetics Research Associates Ltd
Priority to CA000560626A priority Critical patent/CA1251386A/en
Application granted granted Critical
Publication of CA1251386A publication Critical patent/CA1251386A/en
Expired legal-status Critical Current

Links

Abstract

Abstract of the Disclosure A method is described for selecting non-tumorous transformed plant cells expressing a gene coding for an opine synthase and containing a plant expressible heterologous gene from a mixture containing transformed plant cells and untransformed plant cells. The method involves plating the mixture on a growth medium containing an amino acid analog toxic to normal cells but metabolized by a plant cell expressing the opine synthase encoded by the gene, growing the mixture on the growth medium and selecting colonies of plant cells exhibiting greater growth rates.

Description

:~2~

Sl3I.ECTION USING
OPINE SYNTHASE GENES
During the past ten years, the ability to splice 5 DNA from a variety of sources into a recombinant DNA
molecule and then to transfer such DNA molecules into different species of prokaryotes and eukaryotes has led to the most exciting revolution in the history of biology. Most of this work has involved the use of bacteria, fungi and animals. Plants have been relatively neglected primarily because suitable vectors were not available. Recently a numher of possible vectors have become available but in their natural state they have not been efficient in the transfer and expression of genes from various sources to plant species. The present invention describes`some no~el discoveries which increase the usefulness of naturally occurring plant DNA vectors in the genetic engineering of plants.
This application is a division of copending Canadian patent application Serial No. 462,886 filed Septe~ber 11, 1984. The parent application describes a plasmid contaîning no tumor-forming genes and comprising a first DNA segment that is a replicon capable of replication in Agrobacterium and a second DNA segment comprising (a) at least one T-DNA repetitive sequence located at an end of the second DNA segment, (b) at least one opine synthase gene that is expressible in a plant, and (c) at least one heterologous gene that is expressible in a plant and which is not an ant.ibiotic resistance gene. Also described is a transformed plant cell containing the plasmid.
In this speci~ication, reference is made to the accompanying drawings, wherein:
Figure 1 is a graphical representation o~ the effect of homoarginine on the growth of two tumor lines;
Figure 2 shows the nucleotide sequences of a position of the Ti plasmid 15955 containing the ~-DNA
region;
Figure 3 is a restriction endo nuclease map of the sequenced region of Figure 2 for five different restriction enzymes;
Figure 4 is a graphical representation of the effect of canavanino or the growth of two tissue culture lines;
Figure 5 is a graphical representation of the effect of 2 amino-ethyl-cysteine on the growth of two tissue culture lines;
Figures 6 and 6a illustrate the isolation of fragments 2 and 2a from plasmid p233;
Figure 7 illustrates the isolation of fragment 1 from plasmid plO2;
Figure 8 illustrates cleavage of plasmid p SUP106 at the Hind III and Cla I sites;
Figure 9 illustrates ligation of fragments 1 and 2 ko produce fragment 3 and ligation of ~ra~ments 3 and 4 to produce fragment 5;
Figure 10 illustrates the isolation of fragment from plasmid p501;
Figure 11 illustrates the formation o ~ragment 6 from plasmid p403;
25Fiyure 12 illustrates the isolation of fragment 7 from plasmid p501 and ligation of fragments 1, 6 and 7 and further processing of the ligate to form fragment 8;
Figure 13 illustrates the isolation of fragment 9 from plasmid pBR322;
30Figure 14 illustrates the ligation of fragments 6, 7 and 9 to form fragment 10 and the further processing of fragment 10 to form fragment 11;
Figure 15 illustrates the ligation of ~ragments 6 and 7 to form fragment 12 and isolation of fragment 13 from plasmid p233;
Figure 16 illustrates the recombinant fragment 14 formed by ligation of fragments 1, 12 and 13; and Figure 17 illustrates ligation of fragments 2 and 13 and llgation o~ that ligate with fragments 1 and 1 to yield fragment 15.
The field of this invention involves the insertion of foreign DNA which is desired to be expressed in a plant, into a transformation vector, i.e., a plasmid capable of introducing the DNA into plant cells and then maintaining that DNA. Plant cells containing foreign genes which have been introduced by this transformation vector, are used to regenerate morphologically normal plants that carry foreign genes. In an ideal situation, these forsign genes should be carried through meiosis to subsequent generations oE plants.
A goal of plant genetic manipulation is to introduce desired genes into a plant in such a manner that these genes will be functional in the desired tissue at the correct time. The most promising vehicle for such plant genetic manipulations makes use of the Ti-plasmid of Aqrobacterium tumefa~iens and the Ri-plasmid of A. rhizoqenes. A number of investigators have used the soil organism A. rhizo~enes which causes hairy root disease ~Costantino, P. at al. (1980) Gene 11.79-87~. The transformed plant tissue contains plasmid-derived DNA sequences and the tissue synthesizes an opine resembling agropine (Chilton, M-D., et al.
25 (1982) Nature, London 295.432-434; White, F. F. et al.
~1982) Proc. Nat. Acad. Sci. U.S.A. 79:3193-3197). One advantage of A. rhizoqenes is that, unlike crown gall tumors, transformed tissue quite easily regenerates into plantlets containing high concentrations of opines.
In addition, specific foreign DNA fragments have been inserted into the T-DNA region of the Ti-plasmids of A. tumefaciens (Leemans, J. et al. tl981) J. Mol.
Appl. Genet. 1:149-164; Ooms, G. et al. (1982) Plasmid 7:15-29). Standard in vitro recombinant DNA technology was used to insert a chosen restriction ~ragment into a specific restriction site lying in a cloned portion of the T-region. A~ter the resulting plasmid was introduced into an A. tumefaciens strain carrying a wild type Ti plasmid, then homologous recombination between the two plasmids produced a Ti~plasmid carrying foreign DNA in the T-region. Tumors produced by infection of plants with A. tumefaciens containing this recombined Ti-plasmid contained the foreign DNA (Garfinkel, D. J.
et al. (1981~ Cell 27:143-153- Ooms, G. et al. (1981) Gene 14:33_50; Hernalsteens, J. P. et al. (1980) Nature, London 287:654-656).
Transformation v~ctors, e.g., Ti-plasmids, can carry Poreign genes and stably introduce them into plant cells by transfer of the T-DNA regions into the plant genome. These vectors should be able to transform single cells or protoplasts. Such a transformation has been achieved by (i) fusion of bacterial spheroplasts with protoplasts (Hasezawa, SO et al. (1981) Mol. Gen.
15 Genet. 182:206) (ii) the transformation of protoplasts with partially regenerated cell walls by intact bacteria (Martont L. et alO (1979) Nature, London 277:129-131) and ~iii) the delivery of intact Ti-plasmids into protoplasts either as free DNA in the presence of polyethylene glycol and calcium (Krans, F. A. et al.
(19~32) Nature, London 296:72~74), or encapsulated in liposomes (Draper, J. et al (1982) Plant Cell Physiol.
23:255) When any of thPse m2thods are used, a selectablP
marker should be available. One possibility is to use antibiotic resistance markers but it would be undesirable to spread such resistance genes in the commercial applications of genetic engineering.
Transformed cells which form tumors are generally to be avoided because it is sometimes difficult to regenerate plantlets from such tumor tissue. A possibility of avoiding tumor formation i5 to use crippled Ti~plasmids which do not cause the formation of tumors but still insert th~ir T--DNA into the plant genome (Leemans, J. et 35 alO (1982) EMBO. J. 1:147-152).
In summary, the field of the present invention involves the construction of plant transformation vectors which possess maximum efficiency in the transfer of ~oreign genes to a plant genome and which then can be amplified in the plant genome whenever desired. These plant transformation vectors should confer maximum expression of these foreign genes in the desired tissues and at the desired time.
crown gall disease of dicotyledonous plants results from an infection by the gram-negative soil bacterium Aqrobacterium tumefaciens. The ability of a strain of A. tumefaciens to trans~orm plant cells and to induce tumors can be correlated to the presence of a large single copy plasmid (the Ti plasmid, which ranges in size from 140 to 235 kilobases). The transformation of plant cells is the result of transferring genetic information, i.e., T-DNA, from the Ti plasmid to the nucleus of the plant cell. Once this transfer has been achieved, the bacterial cell is no longer needed to maintaln the transformation (Chilton, M-D., Drummond, ~.
H., Merlo, D. J., Sciaky, D., Montoya, A. L., Gordon, M.
P. and E. W. Nester (1977~ Cell 11:263-271).
The transferred T-DNA produces observable phenotypes in the host such as opine synthesis. The earliest opines to be identified resulted from the condensation of an amino acid and a keto acid (Goldman, A., Tempe, J. and G. Morel (1968) Compt. Rend. Acad.
Sci. ~Paris) 162:630-631). As noted above, plant tumors can be grown in the absence of the causative bactaria and yat they still synthesize opines. Octopine synthase talso named lysopine dehydrogenase) is a single polypeptide chain of molecular weight 38,000 to 39,000 and catalyses the condensation of pyruvats with arginine, histidine, lysine or ornithine (Hack, E. and J. D. Kemp (1980) Plant Physiol. 65:949-955) using NADH
or NADP~ as co-factor. A second group of opines is produced by nopaline synthase which is a tetramer of four identical polypeptide chains, each of which is approximately ~0,000 molecular weightO Again NADH or NADPH is used as the co-factor, but the keto acid is ~
ketoglutaric acid and the amino acid is arginine or ornithine (Kemp, J. D., Sutton, D. W. and E. Hack (1979~
Biochemistry 18:3755-3760). La~er, it was shown that a 3~;

particular strain of A. tume~aciens, which induced tumors synthesizing, for example, octopine could use octopine as a sole source of carbon and nitrogen (Montoya, A. L., Chilton, M. D., Gordon, M. P. Sciaky, D. and E. W. Nester (1977) J. Bacteriol. 129 101-107)o Another opine has been identified (Firmin, J. L. and G.
R. Fenwick (1978) Nature, London 276:842~84~). This opine is agropine and is the result of a condensation reaction between an amino acid and a sugar (Coxon, D.
T., Davies, A. M. C., Fenwick, G. R., Self, R. Eirmin, J. L., Lipkin, D. and N. F. James (1980) Tetrahedron Letters 21:495-~98).
The T-DNA (the DNA tran~ferred ~rom the Ti plasmid to the plant nucleus) varies from one Ti plasmid group to another. The octopine A-type plasmids trans~er two pieces of plasmid DNA to plant cells sither together or separately tThomashow, M. F., Nutter, R., Montoya, A.
L., Gordon, M. P. and E. W. Nester (1980) Cell 19:729-739~. One piece of a ca. 8 x 106 daltons is always found in tumors with a frequency of one copy/cell whereas another piece o~ ca. 5 x 106 daltons i5 ~ometimes absent and sometimes present at a frequency of up to 30 copies/cell. The two pieces are co-linear on the Ti plasmid restriction endonuclease map but are integrated into the host plant nucleus as separat~
units. In contrast, the nopaline-type plasmids transfer a single larger piece of Ti plasmid DNA (ca. 10 x 106 daltons) and ~his i5 maintaihed at a frequency o~ 1-20 copies/tumor cell (Holsters, M. et 31. Plasmid 30 3:212 230).
The transferred T-DNA is integrated into the chromosomal DNA of the transformed plant cells (Thomashow, M. F. et al. (1980) supra) and is extensively transcribed into poly-A-containing mRNA
(Gurley, W. B., Kemp, J. D., Alb~rt, M. J., Sutton, D.
W. and J. Collins (1979) Proc. Nat. Acad. Sci. U.S.A.
76:1273-1277: McPherson, J. C. et al. ~1980) Proc. Nat.
Acad. Sci. U.S.A. 77:2666-2670). Part o~ this 3~

transcription codes for the opine synthases (KoeXman, B.
T. et al. (1979) Plasmid 2O347~357).
Tumor cells containing T-DNA can be maintained in culture indefinitely. Further, unlike normal plant cells, tumor cells can be grown on a chemically definsd medium which lacks added auxins and cytokinins ~plant growth hormones) (A. C. Braun (1958~ Proc. Nat. Acad.
Sci. U.S.A. 44:344). Mutants of Ti plasmids can be used to define the location of functional genes such as (i) loci in the T-DNA which determine tumor morphology and opine synthesis (ii) loci outside the T-DNA which are required for virulence and (iii) regions which have no apparent effect on tumorigenicity (Holsters, M. et al.
(1980) Plasmid 3:212-230; de Greve, H. et al. (1981) Plasmid 6:235-248; Ooms, G. et al . (1982) Plasmid 7:15-29).
The genetic organization of the T-DNA of a number of tumors promoted by strains of A. tumefaciens containing octopine Ti plasmids has been studied (Merlo, D. J. et al . (1980) Mol. Gen. Genet. 177:637-643: De Beuckeleer, M. et al. (1981) Mol. Gen. Genet.
183:283-288; Thomashow, M. F. et al. (1980) Cell 19:729_739). In some tumor lines, the T-DNA occurs as two segments. The left end of the T-DNA is called TL
and includes tms ~tumor morphology shoot), tmr ~tumor morphology root), tml (tumor morphology large) and, sometimes, ocs (octopine synthase) while the right end is called TR. Tumor maintenance requires TL but not TR.
Transformed cells induced by wild type octopine Ti plasmids have been selected in two ways (i) the tissue must be able to form tumors and ~ii) the tissue must be able to grow in vitro in the absence of phytohormones.
A serious problem associated with the use o~ wild type Ti plasmids is that it is very difficult to regenerate whole plants from tha tumor tissue.
In order to regenerate plants from transformed tissue culture, mutants can be obtained in the tms, tmr and tml loci. However, it then becomes difficult to distinguish transformed tissues from untransformed ~2~

tissue leading to a requirement for some form of selection. One form of selection has been by the insertion of antibiotic resistance genPs into the T-DNA
(Ursic, D. et al. (1981) Biochem. Biophys. Res~ Comms.
101:1031) (Jiminez, A. et al. (1980) Nature (London) 287:869) (Jorgensen, R. A. et al. (1979) Mol. Gen.
Genet. 177:65). In one instance resistance to the antibiotic G418, which is toxic to tobacco cells in culture, was incorporated into the T-DNA. In a second instance, the bacterial transposon Tn5, which confers resistance to kanamycin, was incorporated into the T-DNA. A disadvantage to the use of this type of selection to detect transformed plant tissue is that it leads to needless spread o~ antibiotic resistance genes.
Such a spread of antibiotic resistance genes would be of very large significancP if such selection were to be used in the agricultural commercial applications of gene transfer via Ti plasmid vectors. Some other method of differentiating transformed from untransformed plant tissue would be highly desirable especially if such selection were to utilize a normal component of Ti plasmids.
In any program involving the genetic al~-eration of plants, foreign DNA of interest must be inserted into a transformation vector, i.e., a plasmid, which in turn can stably introduce the DNA into plant cells. These plant cells containing the foreign genes introduced by way of the transformation vector can then be used to regenerate morphologically normal plants. In an ideal situation, the altered plants should transmit the foreign genes through their seeds.
The best available transformation vector at pr~sent is the T-DNA fragment which is transferred and integrated into the plant genome from the Ti-plasmid of Aqrobacterium tumefaciens following infection. However, if wild type A. tumefaciens is used, then the plant develops a crown gall and it becomes difficult to regenerate whole plants from such crown gall tissu~. If mutants in the tumor genes (tms, tmr or tml~ are used, ~25-l~E~

then re~eneration of plants becomes a practical proposition but selection of transformed tissue becomes di~ficult. 5election c~n be re-established by inserting an antibiotic resistance gene into the T-DNA but this method is u~desirable for reasons already discussed (see above).
After the work described in this specification was completed, Van Slogteren et al. (Van Slogteren, G. M. S.
(1982) Plant Mol. Biol. 1:133-142) reported that homoarginine was less toxic to transformed cells than to normal cells. Although homoarginine is a substrate for octopine synthase (Otten, L. A. B. (1979) Ph.D. Thesis, University Leyden, Netherlands; Petit, A. et al. (1966) Compt. Rend. Soc. Biol. (Paris) 160:1806-1807~, in independent experiments we did not obtain a selection for transformed cells by use of homoarginine. Little or no selectivity by homoarginine for octopine synthase containing tissues compared to other crown gall or normal tissues was observed (Fig. 1). Thes~ independent by us results made it seem unli}cely that the use of any amino acid analogs to select ~ransformad tissues and cells would be successfulO
In accordancs with the present invention, there is provided a method of selecting non-tumorous transformed ~5 plant cells expressing a gene coding for an opine synthase and containing a plant expressible heterologous gene from a mixture containing the transformed plant cells and untransformed plant cells comprising (a) plating the mixture on a suitable growth medium containing an amino acid analog toxic to normal cells and metabolized by a plant cell expressing the opine synthase encoded by the gene; (b) growing the mixture on the growth medium for a selected period of time to provide colonies of plant cells; and (c) selecting from the colonies those colonies ~xhi~iting greater growth rat~s.
This invention makes use of unique T-DNA
constructions f:rom the Ti- plasmid of Aqrobacterium tumefaciens to transfer foreign genes for expression in new plants and to recognize transformed cells carrying such foreign genes incorporated into their genomes by selection for an unalterad T-DNA opine synthase gene.
Such a selection for transformed cells carrying foreign genes can be done without the development of tumor tissue which makes plant regeneration difficulk and without spreading antibiotic resistance genes throughout the plant population.
These T-DNA constructions may contain only the direct nucleotide repeats which are involved in the incorporation of the T-DNA into the plant genome and one or more opine synthase genes which are used in the selection to detect and isolate transformed cells.
Preferably, these uni~ue T-DMA constructions contain a versatile restriction endonuclease site with sticky ends which are compatible to the sticky ends o~ a number of restriction endonucleases and which can therefore be used for the insertion of foreign DNA fragments obtained with the aid of different restriction endonucleases. In these T-DNA constructions, the tumor forming genes may be deletPd and no antibiotic resistance genes need be utilized.
In the presence of a toxic amino acid analog, normal, untrans~ormed cell~ ara unable to grow in culture media or only able ko grow extremely slowly, whereas transformed cells containing one or more opine synthase genes are able to detoxi~y the toxic amino acid analog and so are able to grow. The difference between the growth rate~ of normal and transformed cells is quite clear when a variety of toxic amino acid analogs are used to select for transformed cells or protoplasts.
In the absence of tumor inducing genes, it is easier to regenerate plantlets from tissue culture cells or protoplasts.
In addition, the T-DNA constructions used in this invention permit the recognition of plant cells or protoplasts which have incorporated only parts of the TL-DNA, only parts o~ TR-DNA or parts of both TL-DNA and TR-DNAo Since multiple copies of TR DNA are often ~ound 3~

in a transformed plant genome, this very desirable feature could be used when a high level of expression o~
foreign genes is required.
This invention involves the construction of recombinant plasmids derived from the T-DNA region of Ti plasmids of Aqrobacterium tumefaciens. As mentioned earlier, these plasmids are claimed in the parent application. These recombinant plasmids contain a left and a right direct repeat region normally involved in the transfer of T-DNA from the Ti-plasmid t~ the plant genome as well as the octopine synthase gene, which normally catalyses the condensation of an amino acid with pyruvate, and/or the agropine/mannopine synthase genes which normally catalyse the condensation of an amino acid with a carbohydrate. The preferred recombinant plasmid constructions described here specifically delete the gsnes controlling tumor formation and each construction contains a unique B~lII
site into which foreign DNA ~ragments encompassing one or more functional prokaryotic or eukaryotic genes can be inserted. Other constructions containing other restriction sites, either substituting for or in addition to the ~g~II site, may be incorporated, as will be understood by those of ordinary skill in the art.
Plant cells, which have become infected by A
tume~aciens carrying these constructed recombinant plasmids, may have incorporated the T-DNA section. When tumor-formation genes are deleted or rendered inoparative, the cells containing T-DNA do not display the altered morphology and growth habits that normally make transformed cells distinguishable from untransformed cells. In order to recognize which plant cells have incorporated the T-DNA sections of these constructed recombinant plasmids lacking tumor genes into their genomes, the plant cells are grown on certain toxic amino acid analogs disclosed herein. Those cells which carry one or more of certain opine synthase genes can metabolize the toxic amino acid analog to a non-toxic product and will therefore he able to continue growth while those c~lls which do not contain such opine synthase genes will incorporate the toxic amino acid analog into their proteins resulting in death of such cells.
In order to be able to precisely construct a variety of recombinant T-DNA plasmids with the characteristics described above, the complete nucleotide sequ~nce of T-DNA (22,874 nucleotides) and approximately soo nucleotides on each side of the T-DNA was obtained.
A number of T-DNA restriction fragments were sub-cloned into PBR322 and propagated in either E. coli. strains HB101 or GM33. The individual clones were then sequenced using the method of Maxam and Gilbert [(1977) Proc. Nat. Acad. Sci. U.S.A. 74:560] (see Example 1).
The nucleotide sequence of a portion of the Ti plasmid 15955 containing the T-DNA region is shown (Fig.
2). Only one strand of the DNA seguence is presented.
It is orientated from 5' to 3' and extends continuously for 24~595 bases, from a BamHI site on the left of fragment Bam8 to an EcoRI site on the right of fragment EcoD (Fig. 3). Both strands were sequenced for 90% of the DNA. The remaining 10% was sequenced on on~ strand but this was often duplicated by sequencing from different restriction sitesO
A restriction endonuclease map of the sequenced region is shown for five different restriction enæymes (Fig. 3). BamHI, EcoRI and HindIII were used to divide the T-DNA region into suitable fragments for subcloning into pBR322. The fragments, indicated by the shaded areas, were used in the construction of T-DNA
recombinants cloned into pSUP106 for transfer and subsequent replication in A. tumefaciens. The construction of thsse T-DNA recombinants was facilitated by a knowledge of the other restriction Pndonuclease sites (Table 1). Of the 73 enzymes analys~d, only sites for EcoK were not present in the Ti sequence. The site locations of enzymes which would digest the DNA more than 30 times are not given.

It has been reported that extended direct repeats of 21-25 bases occur at the borders of the T-DNA (Bevan, M. W., and Chilton, M-D. (1982) Ann. Rev. Genet.
16:357_384). In this work, these two repeats have bean shown to start at positions 909 and 23,783, respectively, and they are marked at positions A and D
(Fig. 3). These two repeats are referred to hereafter as RoTL(A) and RoTR(D). They are exact direct repeats for 12 bases but they can be extPnded into 2~ base imperEect repeats. Assuming repeats RoTL(A) and RoTR(D) set the outer limits, the total T-region length is 22,874 nucleotides. These border repeat sequences occur in both octopine and nopaline Ti plasmids and play a fundamental role in the transfer of the T-region to the plant genome. In the prasent study, similar repeated sequences were also found at two locations within the T-region at positions 14,060 (B) and 15,900 (C). These two repeats are referred to hereafter as RoTL(B~ and RoTR(C). The presence of thess internal rapeats is especially interesting because of the observations that the T-DNA in octopine crown gall cells has a complex organization involving one (TL-DNA) or two (TL-DNA and TR-DNA~ regions (De Beuckeleer, M. et al. (1981~ Mol.
Gen. GenetO 1~3:283-28~, Thomashow, M. F. t al. ~1980~
25 Cell 19:729-739). The TL-DNA contains the tumor inducing genes (tms, tmr and tml) and occu~s in all tumor lines so far examined whereas the TR~DNA occurs only in primary tumors and in some stable tumor lines.
The fact that these two T-DNA regions appear to be able to integrate independently and that both are bounded by the basic repeats provides additional evidence of their fundamental role in the transfer of the T-DNA from the Ti-plasmid to the plant genome. The nucleotide sequence of the four repeats is presented here.

~25~

Repeat Nucleotide Sequenc~
RoTL(A) G G C A G G A T A T A T T C A A T T G T A A A T

RoTL(B) G G C A G G A T A T A T A C C G T T G T A A T T

RoTL(C) G G C A G G A T A T A T C G A G G T G T A A A A

RoTL(D) G G C A G G A T A T A T G C G G T T G T A A T T

Within the total T region there are 26 open reading frames (Fig. 3) longer than 300 nucleotides which start with an ATG initiation codon. These possible transcripts would encode polypeptides ranging in size from 11.2 Xilodaltons to 83.8 kilodaltons. Fourteen of these open reading frames show possible eukaryotic promoter sequences with close homology to the Goldberg-Hogness box~ They also show close agreement to the possible eukaryotic ribosome binding site reported by Kozak (M. Kozak (1978) Cell 15:1109-1123; Kozak, M.
[(1979) Nature (London3 280:82-85] and contain possible poly(A) addition sitas at their 3~~ends which are thought to act as transcriptional termination signals (Brawerman, G. (1974) Ann. Rev. Biochem. 43:621-642;
(1975) Prog. Nucl. Acid Res. Mol. Biol. 17~117-148~.
Open raading frame 11 codes for octopine synthase while open reading frames 24, 25 and 26 code for mannopine and agropine synthases. It is noteworthy that all the possible eukaryotic transcripts occur within the TL and TR-DNA regions. The open reading frames between the direct repeats RoTL(B) and RoTR(C~ and also to the right of RoTR(D) and to the left of RoTL(A~ show possible Shine-Dalgarno ribosome binding sites and thus appear to be prokaryotic in origin.
The information made available by the prPsent invention as described above enables the construction of a variety of recombination plasmids eminently useful in the transfer of foreign genes and/or foreign DNA into the genome of recipient plant cells. Foreign gPnes and/or foreign DNA are here defined as DNA nucleotide,
3~
~5 seguences which are not normally present in the T-DNA
of a Ti-plasmid. Knowledge of the nucleotide sequence provides information on the locations of the promoter regions of the various genes and the boundaries of the open reading frames and therefore, as a corollary, also gives the amino acid sequence and molecular weight of the various encoded proteins. In addition, the nucleotide sequence has located all of the direct repeat regions which are of vital importance in the transfer of the T-DNA (including various foreign genes and/or foreign DNA) from the Ti-plasmid to the plant cell genome. Examples 2 - 7 give details of the construction of a number of useful recombinant T-DNA's which can be inserted into a wide host range plasmid. In each of these constructions, suitably located unique restriction sites are available for the insertion of foreign DNA
containing one or more genes. In addition, antibiotic resistance genes are incorporated in the vector region of the recombinant plasmid, i.e., outside the T DNA
repetltive regions which include the selective opine synthase gene and the foreign DNA and/or foreign ~enes.
A novel feature of these T-DNA recombinant constructions is that the tumor inducing genes (tms, tmr and tml) are preferably deleted or inactivated with the result that it becomes easier to regenerate intact plants from transformed cells. However, it also becomes more di~ficult to distinguish between transformed an~
normal cells, since they no longer form tumors and so a new method of selecting transformed cells must be devised. Van Slogteren et al. [(1982) Plant Mol. BiolO
1:133-142] transformed plant cells into octopine-synthesizing tumor cells after infection with a strain of A. tumefaciens carrying an octopine Ti-plasmid.
They claimed that, when small normal shoots and octopine synthase containing crown gall shoots were transferred to solidified agar media containing various levels of homoarginine (HA), growth of normal shoots was clearly inhibited after two to three weeks. In contrast, the crown gall shoots were not inhibited at 3~
1~
all at the lower concentrations oE homoarginine and only slightly inhibited at the higher concentrations.
Van Slogteren et al. ~(1982) ~E~] also tested the effect of homoarginine on reshly isolated normal and crown gall protoplasts. After three weeks they observed a clear di*ference (microscopically) in the growth o~
normal protoplasts compared to crown gall protoplasts.
In seven experiments crown gall protoplast showed no growth inhibition and hardly any difference was observed in the crown gall cultures in the presence or absence of H~, whereas in normal cultures in HA the survival rate was estimated to be 10% or less.
Independent experiments by us gave different results. Crown gall tumor lines 15955/1 and 15955/01 were used for these studies: these tissues are derived from single cell clones obtained ~rom tumors incited on N. tabacum cv."Xanthi" by A~ tumefaciens strain 15g55 (Gelvin, S. B. et al. (1982) Proc. Nat. Acad. Sci.
U.S.A. 79:76-80). Both lines grow on medium lacking cytokinins and auxins and both contain T-DNA. However, octopine is not found in line 15955/l, whereas it is found in line 15955/01 reflecting the fact that the octopine synthase gene is not present in the former line, whereas it is present in the latter line. The presence of octopine synthase in line 15955/01 and its absence in line 15955/1 was later zonfirmed. Thus, these two lines are ideal for testing whether or not the toxic analog homoarginine could be used to select octopine synthase containing genes as claimed by van Slogteren et al. (supra). As can be seen from Figure 1, homoarginine affected the growth of the two tumor lines about equally. Clearly, tissue from a line containing the octopine synthase gene was nok distinguishable from a line without the octopine synthase gene. The reason for the discrepancy between our results and those of van Slogteren et al. is at present unclear.
Surprisingly, therefore, further experimentation with a number of other amino acid analogs showed that salection of octopine synthase containing transformed plant cells was successful with some of the analogs but not with others. The relative differences in growth of tissues on medium that contain~d L-2, 4-diamino-n-butyric acid (an analog o~ L-ornithine) or 2-thiazolyl-alanine (an analog of histidine) were small and inconsistent from experiment to experiment and wor~
with these analogs was abandoned.
The same two tissue culture lines (i.e., 15955/1 and 15955/01) as were used to test the selective abilities of homoarginine, were also used to test the selective abilities of canavanine-S04 (also an analog of arginine) and 2 amino-ethyl-cystein~ (an analog of L-lysine~. In view of the failure to select Por cells with the ability to synthesize octopine synthase by growth on medium containing homoarginine, it was most surprising to find a considerable degree of selection for such cells when tests were conducted on canavanine-S04 (CS~ (Fig. 4~. Even more surprising was the powerful and absolute selection for transformed octopine ~ynthase producing cells when the selection was done on medium containing even very low concentrations of 2-amino-ethyl-cysteine (AEC) (Fig. 5). These two amino acid analogs (CS and AEC) can be used to select for transformed cells in tissue cultures and regenerating plants obtained from these tissue cultures or for transformed protoplasts.
Furthermore, the same approach to selection is useful with the agropine synthase genes. Agropine and mannopine appear to be derived from condensation of glutamine and a carbohydrate and results ~rom this invention have also clearly demonstrated that when agropine/mannopine synthase genes are present in transformed cells, then the use of a toxic amino acid analog distinguishes transformed cells from normal cells. In principle, nopaline synthase could be used to detoxify an appropriate toxic analog of its arginine substrate to achieve selection in the described manner.
An appropriate analog has not been identified to date.

3~

numbar of strains o f Nicotiana tabacum were tested for their ability to grow on toxic analogs of amino acids participating in the agropine mannopine biosynthetic pathway. Details of the strains used are 5 shown ~TablP 2 ) .
In all experiments 50mg pieces of tissue were used initially. Three pieces we.re placed on each petri dish and duplicate dishes were used. These tissue pieces were grown for 9 weeks before they were harvested, dried 10 and weighed. In one set of experiments two strains of Nicotiana tabacum cvO Xanthi were tested. One strain 159 No.-l was agropine/mannopine negative while the other strain 1590-1 was agropine/mannopine positive.
These two strains were tested for their ability to grow on various levels of glutamic-~ hydrazide (GH-Table 3), S-carbamyl-L-cysteine ( CC-Table 5 ~ and 6-diazo-5-oxo-L-norleucine (DON-Table 7). When grown on 20~g GH/ml medium, the growth rate of Strain 159 No.-1 (agropine/mannopine negative) had declined to 9.1% of 20 the control growth rate (T~ble 3) whereas Strain 15g0-1 (agropine/mannopine positive) had only declined to 86.4%
of the control rate. The differences between the growth rates of the two strains on this toxic amino acid analog were clear and striking. When these two strains were 25 grown on 25~g CC/ml medium, there was a claar drop in the growth rate o:E the agropine/mannopine negative strain to 6.9~ oE the control growth (Table 5) whereas there was essentially no drop in the growth rate of the agropine/mannopine positive strain. Finally, when the 30 agropine/mannopine negative strain was grown on 0 . 25,ug DON/ml medium, there was no growth whatsoever, whereas th2 agropine/mannopine positive strain had only declined to 24% of the control growth rate (Table 7).
In a second set of experiments, an 35 agropine/mannopine negative strain (WC 5~-Table 2 ) and an agropine/mannopine positive strain (W-B634) of Nicotiana tabacum cv Wisconsin 38 were tested on various lev~ls of GH, CC and DON. In the case of GH (Table 4), the agropine/mannopine negative strain WC58 was dead in the presence of 10~g GH/ml of medium, whereas the growt~
rate of the agropine/mannopine positive strain W-B634 was still 55% of the control growth rate. When the same two strains were tested on CC tTable 6), the agropine/mannopine negative strain was dead in the presence of 50~g CC/ml of medium while the positive strain showed no decrease at all in the growth rate.
Finally, the results were also quit~ clear cut when these strains were grown on DON (Table 8). In the presence of 0.12~g DON/ml of medium, the agropine/mannopine n~gative strain (W-C58) was dead while the agropine/mannopine positive strain (W-B63~) had grown to some extent (i.e., 12% of the control growth rate).
Thus, the results revealed in this invention teach a unique method of selecting for transformed tissues, cells and protoplasks. These results make use of the fact that untran~formed cells lacking the g~nes for opina synthesis cannot gxow in the presence of toxic amino acid analogs, whereas transformed cells containing opine synthases are able to srow. Furthermore, the selection can be used either when the octopine synthase genes are present or when tha agropine/mannopine synthase genes ara present.
The ability to select for transformed cells by selecting for the octopine synthase or the agropine/mannopine synthase genes permits a further refinement in the engineerin~ oE foreign genes which are desired to be expressed. Thus, the trans~ormed cells may be obtained by selecting for one of the opine synthase biosynthetic pathways and inserting the foreign gene(s) into the reconstructed T-DNA in such a manner khat expression of these genes comes under the control of the other opine synthase. For example, it would be possible to construct a T-DNA containing RoTL(A~, octopine synthase, RoTL(B), agropin~/mannopine synthase-RoTR(D). An analog of lysine, e.g., 2 amino-ethyl-cysteine could be used to select for transformed cells by selecting for octopine synthase while the uniqu~ MstII site (nucleotide 19471 within the agropine/mannopine genes) could be used to insert any foreign DNA.
A further advantage of such a T-DNA construction would be that the two repetitive sequence~ RoTL(A) and RoTL(B) as well as the octopine synthase gene used to select for transformed gene(s) which would still be expressed under the control of the agropine/mannopine synthase gene promoter region.
In summary, the use of reconstructed T-DNA plasmids containing only the direct repeats involved in incorporation of the T-DNA into the plant genome and one or more opine synthesizing genes has resulted in the following useful results: 1. The tumor inducing genas have been deleted resulting in greater success of plant regeneration from transformed tissue cultures or protoplasts. 2. The opine synthesizing genes have been used to select for those plant cells which have incorporated the T-DNA (and therefore in addition any foreign genes which have been inserted~. 3. The invention now permits the recognition of plant cells which have incorporated only parts of TL, only parts of TR or parts of both TL and TR. Since multiple copie~ of TR are found in a transform~d plant genome, this feature could result in a much hiyher level of expression of foreign genes incorporated into some of these T-DNA
constructions.
The invention is illustrated by the following Examples:
Example 1: Sequencin~ of the nucleotides of the T-DNA from the Ti plasmid pTi 15955 Fragments of T-DNA and the flanking regions obtained by use of restriction endonucleases were cloned into pBR322 and then propagated in either E~ coli strains HB101 or GM33. The îndividual clones were then sequenced using the method of A. M. Maxam and W. Gilbert [(1977) Proc. Nat. Acad. Sci. U.S.A. 74:560]. For sequencing lO~g of the cloned DNA's were cut with a suitable restriction enzyme and then treated for 30 ~5~

minutes at 55C with 2.5 units of calf intestinal alkaline phosphatase after the pH was adjusted by adding one-tenth Yolume of 1.0 M Tris/HCl pH 8.4 to the reaction tube. The alkaline phosphatase was removed by three phenol extractions followed by two ethanol precipitations. The dephosphorylated DNA was then dried and taken up in 15~1 water and 15~1 denaturation buffer consisting of 20mM Tris/HCl, pH 9.5, lmM spermidine and O.1 mM EDTA. This mixtur~ was incubated at 70C for 5 minlltes and then immediately put into iced water. After chilling, 4~1 of kinase buffer consisting of 500 mM
Tris/HCl pH9.5, 100 mM MgC12, 50 mM dithiothreitol, 50%
(v/v) glycerol, 100~ Ci of [~-32P] ATP and 2.0 units of polynucleotide kinase were added and the reaction mixturP was incubated at 37C for 30 minutes. The reaction was stopped by ethanol precipitation and the sample dried. The double end-labelled DNA was digested with a suitable restriction enzyme to produce single end-labelled fragments which were then separated on and eluted from a polyacrylamide gel (procedures 4, 5a, 7 and 9 o~ Maxam and Gilbert)~ The DNA ~equencing reactions were then carried out as described (Maxam, A.
M. and W. Gilbert (1977) ~upra) with the ~ollowing modi~ications. The limiting G ~ A reaction was carried 25 out by adding 30~1 of 88% formic acid to the reaction mix and incubating at 20C ~or 3 minutes. The reaction was stopped by the addition of 400~1 0.3M Na-acetate (hydrazine stop). The G reaction time was reduced to 20 seconds and incubated at 20~C. The C + T and C
reactions were reduced to three minutes at 20C and stopped by the addition of 400~1 hydrazine stop. All the reactions were then continued as described (Maxam, Ao M. and W. Gilbert (1977) supra). Long sequencing gels 20cm wide, lOcm length and 0.2mm thick were used to separate the oligonucleotides. The gel plates were treated with a silane (Garoff, H. and Ansorge, W.
(1981) Anal. Biochem. 115:450-457) to bind the acrylamide chemically to one ~ace plate. The other supporting plate is a thermostating plate which maintains the gel at 50 throughout electrophoresis.
Samples were separated on 4%, 6% and 16% acrylamide gels. Differential time loadings were avoided by applying aach sample to each of the three gels simultaneously. Gels were run for 14 hours at 3,000 volts to provide adequate ~ross-over o~ the sequencing ladders from gel to gel. After electrophoresis, the gel (bonded to the face plat~) was ~ixed in 10~ acetic acid for 15 minutes, then rinsed in water. The gel dried directly onto the face plate, shrinking to a thickness of approximately OOOlmm. Consequently, ~-ray film could be placed in direct contact wi~h the dried gel resulting in increased band intensity and resolution. Auto-radiography was carried out at room temperature without the use of intensifying screens. Using these techniques, it was possible to routinely sequence 500 base pairs per fragment. Therefore, by applying 5 fragments to each set o 3 gels, 2500 bases o~ saquence could be obtained. Analysis of the DNA and protein sequences were done by computer. The programs used were those of Dr. 0~ 5mithers and Dr. F. Blattner (University of Wisconsin, Madison).
Example 2~ Construction of a micro-Ti plasmid includinq RoTL~A), ~oTL~B~ and the octopine synthase qene Clone p233 consists of the Ba~HI fragment 17 and EcoRI fragment E (Figs. ~ and 6a) cloned into the vector pBR322 (Bethesda Research ~aboratories, Inc., PØ Box 577, Gaithersburg, MD 20760)~ The clone was cleaved with the restriction endonuclease SmaI (Fig.
6a), the blunt ended SmaI restriction site was converted to a BqlII site by the use of B~lII linkers and the DNA
was ligated and then transformed into E. coli K802. The resulting recombinant plasmid was puri~ied and then transformed into E. coll GM33 which is Dam- and therefore incapable of methylating adenine residues. In this strain the ClaI sits at nucleokide 14686 was not methylated and could be cleaved. Following ClaI and B~lII digestion, the procedure yielded a T-DNA fragment (fragment 2a from nucleotide 11207 to nucleotide 14686) containing the complete octopine synthase gene and the direct repeat at the right hand border of TL (i.e., RoTL(B~). Alternatively, a similar sized fragment (fragment 2) can be obtained by use of the restriction endonuclease BclI in place of ClaI (Fig. 6).
Next the HindIII clone plO2 (covering T-DNA
nucleotides 602-3390) previously cloned into the vector pBR322 and containing a BqlII site (nucleotide 1617) was cleaved with BglII and HindIII (Fig. 7) to yield a T-DNA
fragment (fraqment 1) containing the direct repeat at the left hand border of TL (i.e., RoTL(A)~.
Finally, the wide host range plasmid pSUP106 which will replicate in Aarobacterium umefaciens was clea~ed with HindIII and ClaI (Fig. 8). These three fragments have the following restriction sites at their borders:
Fragment 1 has BqlII and HindIII; fra~ment 2 has BqlII
and ClaI; and pSUP106 has ClaHI and HindIII. Since the restriction sites bounding the three fragments are never common to more than two fragments, ligation can only occur in one arrangement (A-ocs-B) ~Fig. 8). The ~
sit~ between ~ragments 1 and 2 is unique site and can be used to insert any foreign DNA fragment as desired.
This restriction site is very vsrsatile since foreign DNA sPgments which have been cleaved by BqlII, BamHI, BclI, MhoI and Sau3A, among others, can be inserted.
Restriction anzymes and ligase were all used according to the recommendations of the supplier.
Example 3: Construction of ~a m_cro-Ti plasmid including RoTL(A~, octopine synthase RoTL(B) RoTR(D~
As described in Example 2, the HindIII clone plO2 (spanning T-DNA nucleotides 602-3390) was used to yield a T-DNA fragm2nt 1 (Fig. 7) and clone p233 was used to produce fragment 2 (Fig. 6~. These two fragments were ligated to produce fragment 3 (Fig. 9) with a HindIII
site a-t c~e end and a BclI site at the other end.
~ragment 3 co:ntains the nucleotide sequences for RoTL(A), octopine synthase and RoTL(B~. It should be noted that RoTL(B~ could be omitted from the reconstructed T-DNA plasmid by re~triction cleavage with BamHI at the BamHI restriction site at nucl20tide 13774, i.e., 142 nucleotides upstream from the initiation codon of octopine synthase. When this site was used, it was found that the octopine synthase gsne expression was prevented. When the BclI site (nucleotide 14711) was used to isolate a fragment containing the octopine synthase gene, then the gene was actively expressed. It is possible that a di~ferent DNA fracJment could be substituted for the region between BclI and BamHI to restore expression of the octopine synthase gene, e.g., any of the repetitive sequences, i.e., RoTLtA), RoTLtB), RoTRtC) or RoTRtD).
The next step is to amplify the EcoRI clone p501 (Fig. 10) previously cloned into pBR322 and to purify the plasmid. The purified plasmid is then cleaved with restriction endonucleases StuI and K~I to yield a T-DNA
frayment covering nucleotide~ 21673 to 24337 and including the repetitive sequence RoTR(D) (fra~ment 4).
The cleavage produced by StuI is blunt ended, whereas -that produced by KpnI has a 3'-overhang. The blunt-ended StuI site is converted to a BamHI site by the use of BamHI linkers. After Ba~HI digestion, fra~ments 3 and 4 are ligatecl together to yi~ld fragment 5 (Fig. 9) which has a HindIII site at one e~d and a KpnI site at the other endO It should be noted that the endonucleases BclI and BamHI produce compatible cohesive ends.
Then the 3'-overhang produced by cleavage of fragment 4 with KpnI is made blunt ended using the 3'-exonuclease activity of bacteriophage T4 DNA
polymerase (Maniatis, T. et al. (1982) In Molecular Cloning p.l~0 Cvld Spring Harbor). Following inactivation of the polymerase by phenol extraction and precipitation of the plasmid in cold ethanol, the blunt end is convertecl to a ~IindIII site by the use of HinclIII
synthetic linkers so that fraqment 5 now has HindIII

3~6 sites at both ends. Fragment 5 is then digested with the en~yme HindIII.
Finally the wide host range plasmid pSUP106 is linearized with HindIII and treated with bacterial alkaline phosphatase. Fragment 5 is then ligated into the linearized pSUP106 to yield a recombinant plasmid in which the T-DNA section contains in sequence RoTL(A), octopine synthase, RoTL(B), and RoTR(D) (A-oc~-B-D) (Fig. 9). It should be noted that the versatile BglII
site at the junction of ~ragments 1 and 2 can be used to insert any foreign DNA fragment carrying one or more genes that are desired to be expressed.
Example 4: Construction of a_ micro-Ti plasmid including RoTL(A), RoTR(D~_ and the aqro~ine/mannopine synthase The agropine mannopine synthase genes include open reading frames 24, 25, and 26 (Fig. 3) but it is not yet known which of these open reading frames corresponds to agropine synthase and which corresponds to mannopine synthase. Reading frames 24-26 are therefore referred to as agropine/mannopine synthase genes (ags/mas). The EcoRI clone p403 (T-DN~ nucleotides 16202 to 216313 and the EcoRI clone p501 (T-DNA nucleotides 21632 to 24590) were both cloned into pBR322 and separately transformed into E coli HB101. Following amplification and purification, clone p403 is cleaved with SstI at the T-DNA nucleotide 18,472 (see Fig. 11). The overhanging 3~-end is mada blunt ended using the 3'-exonuclease activity of bacteriophage T4 DNA polymerase (Maniatis, T. et al (1982) In Molecular Cloning p.l40 Cold Spring Harbor). Following inactivation of the polymerase by phenol extraction and precipitation of the plasmid in cold ethanol, the blunt ends are converted to BqlII
sites by the use of BqlII synthetic linkers. Then the DNA is cleaved with ~coRI and ~lII to produce a fragment from the T-DNA covering nucleotides 18,~72 to 21,631. This fragment (fragment ~) contains some of the agropine/mannopine synthase genes and is bounded by a BglII site and an EcoRI site.

3~

Next EcoRI clone p501 (T-DNA nucleotides 21632-24590), pr~viously cloned into pBR322, is cleaved with restriction endonucleases EcoRI and KpnI. The resulting fragment 7 (Fig. 12) contains the remainder of the agropine/mannopine synthase genes (not included in fragment 6) and, in addition carries th~ repetitive sequence RoTR(D).
Fragment 1 (Fig. 7), fragment 6 (Fig. 11) and fragment 7 (Fig. 12) are mixed together and ligated.
Since no more than two of the six ends have compatible cohesive ends, the three ragments can only ligate in one orientation to give fragment 8 (Fig. 12) (A-ags/mas-D) with a HindIII site at one end and a KpnI
site at the other end. Finally, the 3'-overhang at the KPnI site is removed by T4 DNA polymerase and converted to a HindIII site by use of synthetic linkers. (Note:
this will convert both ends to bIunt ends and add HindIII linkers to the HindIII end. This adds 4bp to that end.) This fragment therefore has HindIII sites at both ends and can be inserted into pSUP106 after this wide host range vector has been linearized by restriction endonuclease HindIII.
Exam~le 5: Construction of a micro-Ti plasmid includin~ RoTL~B)I _ RoTR(D~ and the a~ropine synthase genes.
Clone p233 which spans the BamHI fragment 17 and EcoRI fragment E (Fig. 13) was cloned into the vector pBR322. The resulting recombinant plasmid was then transformed into E. coli GM33 which is Dam~ and was therefore incapable of methylation. In this strain the BclI site at nucleotide 14711 was not methylated and could be cleaved. Following amplification, the recom~inant plasmid is purified and cleaved with the restriction endonuclease H~aI (Fig. 13). The blunt ended ~E~I restriction site is converted to a B~lII site by the use of B~lII linkers. Then the fragment is cleaved with the restriction endonucleases ~qlII and BclI. This procedure yields a T-DNA fragment (fragment 9 from nucleotide 13,~00 to nucleotide 14711) which contains the direct repeat at the right hand border of TL ~i.e., RoTL(B)l.
A second fragment (fragment 10) which includes the agropine/mannopine synthase genes and the direct repeat at the right hand border of TR (i.e., RoTR(D)) was constructed by mixing fragments 6 and 7 (Fig. 12) with fragment 9 (Fig. 13) and ligating them together to give fragment 10 (Fiy. 14). The 3'- overhang at the ~E~I
restriction site is then blunt ended by use of T4 DNA
polymerase and converted to a HindIII site by use of synthetic linkers. This procedure will convert both ends to blunt ends and add HindIII linkers to the HindIII end. Subsequent digestion with HindIII and BclI
will produce a fragment with a BclI site at one end (compatible with a BamHI site~ and a HindIII site at the other end (~ragment 11-B-ags/mas-D) (Fig. 143.
The wide host range vector pSUP106 which can replicate in A~robacterium tume~aciens is then linearized by digestion with Bam~II and HindIII and ra~ment 11 is inserted into the linearized vector.
Ex mple 6: Construction of a _micro-Ti ~lasmid includlnq RoTL(A), tha aqropine/mannopine synthase_ qenes. RoTR(D~, _the_ octopine synthase ~ene and RoTL(B) Fragment 1 was obtained as describ~d in Example 2 (Fig. 7). This fragment has a HindIII site at one end and a BqlII site at the other end. Fragment 12 was obtained by ligation o~ fragments 6 and 7 (Fig. 12~ and has a BqlI1 site at one end and a KpnI site at the other end~
Clone p233 which spans the BamHI fragment 17 and EcoRI fragment E (Figs. 6 and 13) was cloned into the vector pBR322. The resulting recombinant plasmid was then transformed into E. coli GM33 which is Dam and was therefore incapable of methylation. In this strain the BclI site at nucleotide 14711 was not methylated and could be cleaved with the enzyme BclI. Following ampli~ication, the recombinant plasmid is purified and cleaved with the restriction endonucleases BclI and KpnI

to yield fraqment 13 spanning T-DNA nucleotides from the KpnI site (nucleotide 983~) to the BclI site (nucleotide 14711). Ligation of a mixture of fragment 1 (Fig. 7), fragment 12 (Fig. 12~ and fragment 13 (Fig. 15) yields a recombinant fragment 14 (Fig. 16~ with a _IindIII site at one end and a BclI site at the other end (A ags/mas-D-ocs-B~.
The wide host range plasmid pSUP106 is finally linearized with the restriction endonucleases BamHI and _ dIII. After linearization, fragment 14 is inserted by ligation.
This construction has the advantage that, if desired, the foreign genes can be inssrted into one of the opine synthase genes and the promoter of that gene then used for the expression of the foreign genes while the second opine synthase gene can be used for the selection of trans~ormed cells.
Example 7: Construction of a m_cro-Ti plasmid includin~_ RoTLfA~. duPlicate octopine synthase genes and RoTL~B~ and_RoTR(D~
The methods of obtaining the fragments used in the present T-DNA construction have been described in the previous examples. Fra~mants 1 and 2 were obtainPd as described in ~xample 2 (Figs. S and 73. Fragment 4 is obtained as describ~d in Example 3 (Fig. 10) and fragment 13 is obtained as described in Example 6 ~Fig.
15)-In the present example, ~ragments 2 and 13 are ligated together to give fragment 14 containing two octopine synthase genes and two repetitive RoTL(B) sequences in opposite ori~ntations (Fig. 17). Following repurification of the ligated fra~ment, the correct orientation i5 checked by a restriction endonuclease map and the purified, ligated fragment 14 is then mixed with 35 fragments 1 and 4 to yield Eragment 15 (Fig. 17) containing all the required elements. ~ragment 15 is inserted into thP wide host range mutant pSUP106 following linearization of the ve~tor with the restriction endonucleases HindIII and BamHI. The ~9 presence of two octopin~ synthase genes available for selection will increase the ability to select transformed cells or protoplasts from a mixture of transformed and normal cells or protoplasts when they are grown in the presence of a toxic amino acid such as canavanine or 2-amino-ethyl-cysteine.
Example 8: Preparation o~ a Bacillus thurinqiensis crystal protein ~ene pES1 (H. E. Schnepf et al., Eur. Pat. Appl. 63,949, 10 ATCC 31995) is cut with PstI, then mixed with and ligated to PstI-linearized mWB2344 (W. M. Barnes, et al. (1933) Nucleic Acids Res. 11:349-368). The resultant mixture is transformed into E. coli JM103, and tetracycline resistant transformants are selected.
Double-stranded RFs (replicative form) are isolated from the transformants and characterized by restriction mapping. Two types of transformants are found: cells harboring mWB2344-ESl-A; and cells harboring mWB2344-ES1-S. These are M13 vectors which when in single-stranded viral form carry the antisense and sens~
strands of the crystal protein gene of pES1~
The sequence of the crystal protein gene (see below: b) (H. C. Wong et al. (1983) J. Biol. Chem.
258:1960-1967), if changed at three base-pairs, will have a ClaI site (5'...AT*CGAT..~3') immediately ahsad of the ATG translational start site. The o 1 i g o n u c 1 e o t i d e ( s e e b e 1 o w : a ) 5'ATGGAGGTAATCGATGGATAACA3' i9 synthesized by standard methods, is hybridized to the single strand viral ~orm of mWB2344-ESl-A, and is used to prime synthesis of a second strand o~ DNA by the Klenow fragment of E. coli DNA polymerase 1. After ligation and selection oE
covalently closed circular DNA, the mixture is transformed into JM103. RF DNAs are isolated from the infected cells and characterized by restriction enzyme analysis. A clone descend~d from and containing the mutant sequence (a) is identi~ied by the presence oE a novel ClaI site which maps at the 5' end o~ the crystal protein gene and is labeled mW~2344-ESl-A(Cla)0 The effect of the mutation can be seen by comparing the sequences of the oligonucleotide primer (a) with the 5' end of the crystal protein gene (b~:
a~ 5'ATGGAGGTAAT*CQ ATG GAT AAC ~3' b) 5'...AGAGATGGAGGTAAC TT/AT5/GAT~AAC/AA~CC...3' Met ~sp ~sn Asn Pro...
No~e that only three out of 23 base pairs have been changed (the underlined nucleotides of (a)), thereby assuring good hybridization propertiesO
The cryskal protein gene is removed from mWB2344-ES1-A(Cla) by digestion with ClaI and XhoI. The XhoI
sticky ends are converted to ClaI sticky ends by ligation to a linker, synthesized by standard methods, having a structure as follows:
XhoI ClaI
5'TCGAGCCCAT3' 3'CGGGTACG5' Excess linkers are trimmed off of the crystal protein gene-bearing ~ragment by digestion with ClaI. 0 Example 9~ Construction and modification o~ a promoter vehicle p~Slll~ which is a pRK290 clone corxesponding to ths T-DNA clone p403 (Fig. 3) which encodes a gene covering 1.6kb (C. F. ~ink (1982) ~.S. thesis, University of Wisconsin~Madison), or p403 itself, is digested with ClaI and then religated. The ligation mix is trans~ormed into Eo coli K802 ~W. B. Wood (1966) J.
Mol. Biol. 16-118) and selected for tetracycline resistance. Plasmids are isolated by doing "minipreps"
(plasmid preparations from small volume cell cultures) and restriction maps are obtained to prove the structure. The new vehicle, pKS-proI (see Canadian Patent Application Serial No. 451,767 ~iled April 11, 1984, can be lin~arized by ClaI.
The above manipulations are done with the following rationale^ The T~DNA gene in pKSlll is shown below in summarized form as follows:

-ClaI 960 kp 250 bp ClaI 60 ~p 50 kp 5'.. ...TAC~CA~I*OG/~IG/G~C~AIG/... /I~.... AI~CGAT.... AAAI~A... AA~I~A... 3 pro~oter Met Asp M~t~.stop polyadenylation si~s "1.6'~kbp g2n~
By removing the ClaI fragment, the promot~r reyion of the "1.6" gene is brought next to the 3'-downstream region of the gene. This 3' region includes polyadenylation signals. The resulting structure is summarized as follows-E~oe90 ClaI 60 bp 50 bp 5~.~.AIAcAccAAAl*cG~IAGT.. ~........ AA~rAA... A~AIA~A~... 3~ (p~sE~roI) pr~moter p3lyadenylation si~E~s Example 10: Insertion of the roI promoter into the A-ocs-B plasmid (Fiq. 8) pKS-~E~I is digested with EcoRI and mixed with and ligated to EcoRI/BamHI linkers, synthesized by standard procedures having the following structure:
~coRI BamHI
5'AATTCCCCG3' 3'GGGGCCTAG5~
The linkers are trimmed by digestion with Bam~II. After the T-DNA promoter fragment is purified by gel electrophoresis, it is mixed with and ligated to BqlII-linearized A-ocs-B (Example 2, Fig. 8). The mixture is transformed into E. coli GM33 and transformants xesistant to chloramphenicol are selected.
Plasmids isolated from such transformants are isolated and characterized by restriction enzyme analysis. A
plasmid containing the promoter fragment is labeled pA-ocs-B-proI.
Example 11: Insertion of the crystal protein gene into the vector p~-ocs-B-~E~I is linearized with ClaI, and is then mixed with and ligated to the crystal protein gene-bearing ~ragment terminated by ClaI sticky ends constructed above. The resulting plasmids are transformed into GM33. Plasmids isolated from transformants resistant to chloramphenicol are 5~

characterized by restriction analysis. A colony is chosen which contains a plasmid, pA-ocs-B~ I~ESl, having present a single copy of the crystal protein gene oriented in the same polarity as the ~E~I promoter.
pA-ocs-B-E~I-ES1 is transferred to an appropriat~
Agrobacterium host and used to transform tobacco cells as described in Example 12.
Example 12: Selection of plant cells transformed by trans~er of reconstructed T-DNA
recombinant plasmids from Agrobacterium tumefaciens The purpose of this example is to demonstrate a procedure to select transformed cells from mixtures of transformed and non-transformed cells. Generally, transformed cells are selected for their hormone autonomous growth. However, when Ti plasmids that have been mutated in tms, tmr or tml are used, khen transformed cells are not autonomous and a selectable marker becomes desirableO Kanamycin or G418 resistance is a possibility, but it requires engineering a resistance gene. AnothPr possibility is the use of octopine synthase a~ an enzyme to detoxify exogenously added toxic amino acid analogs, e.g., ~-aminoethyl-cysteine (2A~C). In this invention, it has been demonstrated that non-transformed tissues were killed by low levals of AEC (Fig. 5) whereas crown gall tissues expressing octopine synthase were not killed.
There~ore, the present example uses Ti-plasmids that are mutated in tms, tmr or tml but contain a non-mutated ockopine synthase gene. Plants are first stem inoculated as previously described (K. A. Barton et al. (1983) Cell 32:1033-1043). After 10-12 days, fresh growth is r~moved and shaken in liquid culture until a number of cells have sloughed off. The culture is then passed through a filker and the small clumps of cells are collected. These clumps are plated on a filter paper placed on top of a feeder culture containing hormones.

Once the cslls have started to grow, the entire filter is transferred to hormone media containing 2AEC.
The transformed cells will be expressing octopine synthase that will detoxiEy the amino acid analog thus allowing them to grow whereas the untransformed cells will be killed.
Colonies that grow on 2AEC are picked and tested for octopine synthaseO These cells can then be regenerated and can be shown to carry T-DNA containing the octopine synthase gene.
Example 13: Tests for the _relative toxicities of amino acid analo~s usinq strains 15955/1 and 15955/01 ~rom Nicotia a tabacum (tobacc~L _v. "Xanthil' Aqueous solutions of the amino acid analogs were adjusted to pH 5~6-5.8 and sterilized by filtration (0.45 micron Millipore filter). Serial dilutions of each analog were add~d to cytokinin- and auxin-free agar medium (Linsmaier, E. M. and F. Skoog (1965) Physiol.
20 Plant 8:100-127) that had been autoclaved and allowed to cool to about 60~C. A number of other plant strains were us~d ~or initial screening of the analogs. Three 100mg pieces of 4-6 week old tissue were planted on analog-containing medium in 55mm Petri dishes using tissue lines from Helianthus annuus ~sunflower) cv.
"Russian Mammoth" ~Kemp, J. D. (1982) In Kahl, G. and J.
S. Schell (eds.) Academic Press, New York, London, Paris pp~461-474); E228 and Bo5~2, from N. tabacum cv.
"Samsun" (Sacristan, M. D. and G. Melchers tl977) Mol.
30 Gen. Genet. 152~ 117); Braun's teratoma from N.
tabacum cv. "Havanal' (Braun, A. C. and H. N. Wood.
(1976) Proc. Nat. Acad. Sci. U.S.A. 73:496-500); A66 isolated from N tabacum cv. "White Burley" (Finsin, J.
L. and G~ R. Fenwick ~19 7 8) Nature, London 2 7 6:842-844);
35 and W-A6, W-~63~, W-C58 and W-T37 isolated by J.
Tournew, CNRA, Versaillas, from tumors incited on N.
tabacum cvO "Wisconsin 38" by octopine- type A.
tumefaciens strains A6 or B634, or nopaline-type strains C58 or T37, respectively. The results with all o~ the above strains were essentially the sam~ as those described for N. tabacum cv. "Xanthil' strains 15955/1 and 15955/01. For the toxicity studies with the 15955/1 and 15955/01 tumor lines, three 50mg pieces of 4-6 week old tissue were planted on medium in 90mm dishes. In all cases, duplicate or triplicate dishe~ were used for each tissue and each concentration of an analog. The tissues were weighed after 5-7 weeks o~ growth at 25C
in the dark. Growth was expressed as a percentage of the fresh weight increase of the same tissue on medium that did not contain any analog or amino acid. The experiments were repeated after initial results delimited the range of concentrations to use for a given analog. Experiments with the 15955/1 and 15955/01 lines were repeated at least twice with the appropriate concentration rangesO
Example 14: Methgd of assaying_for opine synthases Octopine synthase was assayed using som~
modifications of previous methods tBirnberg, P. et al.
20 (1977) Phytochemistry 16 647-650; Goldmann, A. (~977) Plant Sci. Lett. 10:49-58; Hack, E. and J. D. Xemp (1977~ Biochem. Biophys. Res. Comms. 78:785-791;
Lejeune, B. (1967) C. R. Acad. Sci. Ser. Do 265:1753-1755; Otten, L. A. B. M. and R. A. Schilperoort ~5 ~197~) Biochim. Biophys. Acta. 527:497-500). For each mg o~ tissue in a 1.5ml Eppendorf tube, 3~1 of a buf~ered (0.15M potassium phosphate, pH 6.9) reaction mixture was added. This reaction mixture contained 20 ~M L-arginine, 50 mM pyruvate and 13.5 mM NADH. The tissue was carefully macerated in the reaction mixture with a glass rod and incubated at 25C ~or one hour.
The samples were then clarified by centrifugation and 2~1 o~ supernate wa spotted on Whatman~ 3MM paper. Up to 30~1 of sample was used to verify a negative assay result. The Whatman 3MM paper with the samples, as well as l-10~g o~ octopine (or other opine) as an authentic standard and Orange G as a migration marker, was care~ully wetted with electrophoresis buffer (formic acid/acetic acid/water; 3/6/91, v/v/v~ and electrophoresed at 50-75 voltsJcm (Gibson High Voltage Electrophoresis Apparatus, Model D) for 10-20 minutes.
The electrophor~tograms were dried in a current of hot air and stain~d ~or octopine ~or other opines) with either a phenanthrene quinon~ reagent (Yamada, S. and H.
A. Itano (1966) Biochim. Biophys. ActaO 130:538-540) or with Sakaguchi -.eagent (Easley, C. W. (1965) Biochim.
Biophys. Acta. 107:386- 388). The Sakaguchi reagent is several-fold less sensitive but was more specific for guanidinyl compounds. When the above conditions were used, it was also shown that tissue extraction buffers (Hack, E. and J~ D. Kemp (1977) su~ra; Otten, L. A. B.
M. and R. A. Schilperoort (1978) supra) did not improve assay results and, in ~act, retarded electrophoretic migration when larger sample volumes wPre spott~d.
Electrophoretic mobilities were measured from tha origin ~0) relative to Orange G (1.0).

RESTRICTION ENZYME SITES OF TH~ T DNA REGION OF pTI 15955 Enzyme Sit~s L~c~tions Apa I 111,930 Mat II 119,471 Kba I ~18,089 Mlu I 28,93912,943 Sal I 26,77823,292 Tth I 217,04324,288 Hpa I 37,2579,442 13,800 Kpn I 3625 9,838 24,337 Pst I 39~21110,06922,456 Sa~ I 32,61014,0~918,472 Sat II 314,99618,46223,123 Xho I 36,72715,20821,476 Xma III 3411 11,983 22,663 Aat II 44,51111,76314,6S5 15,140 Bal I 44,3195,456 6,253 21,~18 BatE II 411,76811,97622,865 24,501 Eco I 412,45217,04120,160 21,516 Rru I 416,51517,14418,88~ 24,213 Sma I 4155 2,212 4,850 11,207 Stu I 44,2176,938 14,675 21,673 Xor II ~327 670 1,206 23,033 Bam RI 5 1 7,602 8,0~2 9,062 13,776 Nar I 513,53617,15819,170 20,027 24,098 Bcl I 610,05814,71114,973 15,938 21,540 24,~06 Bgl XI 61,S174,254 5,033 6,023 7,720 22,930 TABLE 1 (Continued~
Enzyme Sites Locati~ns Nru I 614,27614,47516,42017,97321,416 2~,294 Sph I 63 ~ 24113,22013,28917,60119,295 21,562 BaaH II 7 667 9,41012,071 19,334 22,273 23,32124,069 Hind III 7 602 3,390 5,512 5,933 6,631 19,23919,953 Bgl I 8158 848 3,506 4,216 5,066 5,34212,15019,056 Eco RI 84,4945,54512,82313,026 13,362 16,20221,63124,590 Nac I 8511 5,197 6,27610,475 12,077 20,80622,35324,096 Nde I 82,1747,282 7,475 8,360 19,084 19,71521,73124,586 Aha III 9 752 2,679 2,726 2,799 3,799 9,66512,22113,68516,306 BetX I 9587 1,589 5,862 6,150 8,002 10,25913,75120,13222,741 Eco RV 92,7074,888 7,354 9,292 12,797 12,99618,02721,52233,041 Nco I 92,9215,28613,37815,421 15,562 lB,37221,08021,71024,065 Xan I 92,8065,793 6,567 6,839 6,992 10,10313,51217,67921,343 Mat I 101,4084,462 9,85511,632 15,017 15,07715,57017,60219,92820,494 BVU I 112,6105,022 6,96911,930 12,574 14,0891~,04918,47~22,31023,517 24,547 Ava I 12153 2,210 4,848 5,114 6,019 6,72711,20511,96~15,208 18,678 21,~7621,803 Cla I 121,2062,gl5 4,154 9,282 9,292 14,68615,67218,74~18,89020,128 21,43224, ~39 Pvu II122,8343,061 4,682 5,138 6,031 6,8319,975 11,83412,541 14,615 22,61624,091 Acc I 141,1612,687 6,587 6,779 6,794 11,48211,56013,99115,1161.9,942 23,29323,41723,67724,028 HgiA I14812 1,868 2,610 5,134 6,228 7,62812,48012,73414,089 14,583 18,18318,47220,86621,093 Rinc II 14 1,369 5,721 6,780 7,257 9,442 11,32~13,15613,80017,07519,393 21,47221,72722,44023,294 HgiC I17621 3,586 4,960 5,119 6,153 7,4439,834 12,01013,535 16,015 17,15719,16920,02622,70126,097 24,32~24,333 HgiD I211,3762,503 4,508 6,803 8,335 11,76012,51613,53614,66215,137 TABLE 1 (Continued) Enzyme Sites Locations 15,23115,801 16,470 17,158 19,170 19,38919,6~8 20,027 20,24~ 24,098 24,455 BstN I 24309 377 1,423 2,538 4,210 5,023 6,976 7,056 7,583 10,151 10,86511,868 12,146 12,602 13,553 14,67216,947 19,313 19,346 19,422 19,59019,677 20,790 22,830 Hae II 28539 2,206 2,331 3,327 5,196 5,210 5,309 5,981 6,539 9,789 10,47412,269 13,539 13,845 14,335 14,70715,731 15,872 16,412 17,161 17,98018,509 19,173 1~,576 20,030 22,3522~,101 24,398 Hph I 37 Ava II 38 Fok I 39 Nci I 40 Rsa I 40 Tth I 44 Hga I 45 Hin~ I ~7 SfaN I 47 Mbs II 61 ScrF I 64 Dde I 66 Tac I 67 Sau 96 69 Ha~ III 91 Hha I 98 Alu I 99 Hpa II 102 Fnu 6~ 103 Taq I 111 Sau 116 Mnl 158 ~E 2 A lis~ of the tissues used for h~ing ability to grow in khe pr~ of various toxic alTuno acid analogs strain No. a~ltivar l~ ormed ~y: M~pine P~pine 159 No. 1 Nicotiana t~cum Al~c~ium tumefaciens - -cv. Xanthi strain 15955 1590-1 do do + +
W C58 N. tabacum Ao t~nefaciens cv~ Wisc. 38 s~ain C589 W-B634 N. tab~cum A. t~nefaciens cv. Wisc. 38 strain B634 + +

Dry weights of transformed tissues containing agropine/mannopine synthase genes compared to dry w~aights of transformed tissues without agropine 5 mannopine synthase genes when grown in the presence of various levels of the toxic amino acid analog glutamic-y-hydrazide (GH).
Nic~tiana tabac~n cv. ~anthi(a) ~g/ml medium Sb~Ln 159 No. l(b) S*rain 1590-1(C) Wt. in mg. % o~ control Wt. in mg. % o~ con~rol o 7230 1250 2 . ~ 5690 78 . 7 880 70 . 4 2900 40 . 1 1180 9~ . 4 2900 16. 6 1200 96. 0 660 9 . 1 1080 86. 4 100 1 . 4 650 ~2 . 0 a. ~ril[~ts w~e done in duplicate dis~es with 3 pieoes tissue/dish.
b. Agrapine/m~nr~ine r~tive.
c. ~mpine/ma~ine positive.

I~E 4 ~y weights of trans~or~d tissues contai~ agrc~ine/~ar~ine synthas~
genes campar~ to dry weights of transfor~3d tissues withalt agrcpine/n~cpine synthase genes ~el~ gmwn in ~e presenoe of varic~us levels of the t~ic am~no acid analog glutami~,u-hy~razide (GH).
Nics:~tiana tabac~n cv. Wisc~nsin 38(a) ~g/ml medium Strain W{ 58(b) Strain W-B634(C) Wt. in mg.% of control ~t. in mg.~ of oontrol 0 1100 - 1670 ~
2.5 610 55.5 920 55.1 200 18.2 1300 77.8 Dead 920 55.1 Dead 620 37.1 Dead 430 25.7 Dead 580 34.7 a. EXperiments were done Ln duplicate dishes with 3 pieces tissue/dish.
b Agrcpine~manncpine negative.
c A~ropLne/mannopine poeitive.

~5~

I~LE 5 ~y weights OI transformed tissues contai~ agm~ine/marn~ine synthase gen~s c~ar~d to dry wei~h~s of trans~onn~d ti~sues wi~:h~ a~r~p~ne/mar~op~ne synthase gen~s ~n grc~n in ~e pres~nce o~ ; of the toxic am~no acid analog S car~l-lrcyst~ne (aC).
Nicotiana tabac~n cv. ~nthi(~) l~g/ml m~li~n Stram 159 No. l(b) Stxain 1590-1~C) Wt. in m~.~6 of con~rol Wt. in mg. % of cont~ol 0 7230 -- 1250 ~
6 7800 107 . 9 1180 9d~ . 4 12 6720 92 . 9 1640 131 . 2 500 ~ . 9 1160 92 . 8 500 6. 9 1400 112 ~ 0 100 400 5.5 1710 136.8 . . _ .
a. Experimellts w~re done in duplicate di~hes with 3 pieces tissue/dish.
b A~r~e/mar~ine ne~ative.
C. ~ine/manr~ir~ p~;itive.

I~E 6 Dry wei~ts of trarE;fonned tissu~ contai~ agm~i~e/mann~ne ~yn~ase genes cc~mparsd t~ dry wei~h~s of trar sform~d tissues with~ agra~ine/mannop~ne synth~æ gene~; when grown in the pres~oe of various levels o~ ~he tox~c am~no acid analcg S carb~[yl~I,cysteine (CC~.
Nicatiana ~bac~n cv. Wisconsin 38(a) ~g/ml medi~n St~ain ~ C58(b) Strain W-B634(C) Wt. in ~g.% of cont;rolWt. in m~.% of control 0 1100 -- 1670 ~
6 270 24 . 5 1890 113 . 2 12 1040 94 . 5 1440 86 . 2 220 20. 0 1720 103 . 0 50 Dead -- 2000 119 . 8 100 80 7 . 3 2580 154 ~ 5 aO Expe~nen~s ~re done ~ :~uplicate di~hes wit;h 3 piec~; tissu~3/dish.
b. Agr~pine/ma~ine ru~a~ive.
c. A~pine/mar~cpine positive~

I~E 7 ~y wei~hts of t~ansfonned tissues cont~ agr~pine/mar~ins synkhase genes synthasa gene~ ~en gr~ in the p~ of vari~ le~,7els of the toxic aTino acid analog 6~iaz~5 ox~I~norleucine (DCN).
Nic~tiana 'cabacmn cv. Xanthi(a) ~g/ml r~di~n Strain 159 No. l(b) Stra.~n 1590-1(C) Wt. in mg. % of controlWt. in mg. % of control 0 7230 ~ 1250 0.015 5260 72.8 1230 g8.4 0.03 2470 34.2 1330 106.4 0.06 2880 39.8 1180 94.4 0.12 90 1.2 600 48.0 0.25 60 0.8 30~ 2~ .0 O . 50 Dead - 140 11.2 1. O Dead ~ 100 8.0 2.0 Dead 300 24.0 .... _ _ a. EXperiments ~re done in duplicate dishe~ with 3 pieces tissue/dish.
b Agropine/mannopine negative.
c. Agropine/nannopine positive.

I~LE 8 Dry wei~hts of transformed tissues conta~ agrc~ineJmar~Lne ~r~ genes campa~ to dry weights of trans~orm~ tissu~ wi~t agr~pine/ma~}opine syn~hase g~nes w~en gra~ in ~e presence of various le~els of th~ t~xic ~nino acid anal~g 6~iaz~5-oxo-~rnorleu~ne (D~
Nicotiana tabac~m cvO Wiscons~n 38(a) medi~n Strain W{~58(b) Strain W-B634(c~
Wt. in m~.% of controlWt. in mg.% of cont~ol 0 1100 ~ 1670 0 . 015 ~50 22 . 7 ~70 28 . 1 0 . 03540 49 . 1 270 16 . 2 0 . 06120 10. 9 360 21. 6 0. 12Dea~ -- 200 12 . 0 0. 25 ~ad -- 120 7 . 2 a. Experim~ts were done in duplica~e di~hes with 3 pieces tissue/dish.
bo Ag~p~e/maT~ine r~a~ive~
c. A3r~ine/ma~ine positi~e.

Claims (7)

THE EMBODIMENTS OF THE INVENTION IN WHICH AN EXCLUSIVE
PROPERTY OR PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:
1. A method of selecting non-tumorous transformed plant cells expressing a gene coding for an opine synthase and containing a plant expressible heterologous gene from a mixture containing said transformed plant cells and untransformed plant cells comprising:
(a) plating said mixture on a suitable growth medium containing an amino acid analog toxic to normal cells and metabolized by a plant cell expressing the opine synthase encoded by said gene;
(b) growing said mixture on said growth medium for a selected period of time to provide colonies of plant cells; and (c) selecting from said colonies those colonies exhibiting greater growth rates.
2. The method of claim 1 wherein said transformed plant cells comprise a DNA segment comprising:
(a) at least one T-DNA repetitive sequence located at an end of said DNA segment;
(b) at least one opine synthase gene that is expressible in a plant; and (c) at least one heterologous gene that is expressible in a plant, said heterologous gene (5) not being an antibiotic resistance gene(s), wherein said DNA segment contains no tumor-forming genes.
3. The method of claim 1 wherein said amino acid analog is canavanine.
4. The method of claim 1 wherein said amino acid analog is 2-amino-ethyl-cysteine.
5. The method of claim 1 wherein said amino acid analog is glutamic-a-hydrazide.
6. The method of claim 1 wherein said amino acid analog is S-carbamyl-L-cysteine.
7. The method of claim 1 wherein said amino acid analog is 6-diazo-5-oxo-1-norleucine.
CA000560626A 1983-09-14 1988-03-04 Selection using opine synthase genes Expired CA1251386A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CA000560626A CA1251386A (en) 1983-09-14 1988-03-04 Selection using opine synthase genes

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US53228083A 1983-09-14 1983-09-14
US532,280 1983-09-14
CA 462886 CA1251386C (en) 1983-09-14 1984-09-11 Selection using opine synthase genes
CA000560626A CA1251386A (en) 1983-09-14 1988-03-04 Selection using opine synthase genes

Related Parent Applications (1)

Application Number Title Priority Date Filing Date
CA000560626A Division CA1251386A (en) 1983-09-14 1988-03-04 Selection using opine synthase genes

Related Child Applications (1)

Application Number Title Priority Date Filing Date
CA000560626A Division CA1251386A (en) 1983-09-14 1988-03-04 Selection using opine synthase genes

Publications (1)

Publication Number Publication Date
CA1251386A true CA1251386A (en) 1989-03-21

Family

ID=25670488

Family Applications (1)

Application Number Title Priority Date Filing Date
CA000560626A Expired CA1251386A (en) 1983-09-14 1988-03-04 Selection using opine synthase genes

Country Status (1)

Country Link
CA (1) CA1251386A (en)

Similar Documents

Publication Publication Date Title
De Bruijn et al. Rhizobium meliloti 1021 has three differentially regulated loci involved in glutamine biosynthesis, none of which is essential for symbiotic nitrogen fixation
EP0122791B1 (en) Plant gene expression
Joos et al. Genetic analysis of T-DNA transcripts in nopaline crown galls
JP3083531B2 (en) Methods for inducing lysine overproduction in plants
US5102796A (en) Plant structural gene expression
CA1251386C (en) Selection using opine synthase genes
US5750871A (en) Transformation and foreign gene expression in Brassica species
JP2993964B2 (en) Phosphinothricin resistance gene active in plants and uses thereof
US5668298A (en) Selectable marker for development of vectors and transformation systems in plants
US8273954B1 (en) Genetically transformed plants
US6037526A (en) Method of inserting viral DNA into plant material
AU6036898A (en) Methods for (agrobacterium)-mediated transformation
JPH0258917B2 (en)
NL8300699A (en) METHOD FOR BUILDING FOREIGN DNA INTO THE NAME OF DIABIC LOBAL PLANTS; METHOD FOR PRODUCING AGROBACTERIUM TUMEFACIENS BACTERIEN; STABLE COINTEGRATE PLASMIDS; PLANTS AND PLANT CELLS WITH CHANGED GENETIC PROPERTIES; PROCESS FOR PREPARING CHEMICAL AND / OR PHARMACEUTICAL PRODUCTS.
JPH0740941B2 (en) Method for introducing exogenous DNA into plant genome
EP1781082A2 (en) Biological gene transfer system for eukaryotic cells
US6048730A (en) Selectable marker for development of vectors and transformation systems in plants
Lundquist et al. Genetic complementation of Agrobacterium tumefaciens Ti plasmid mutants in the virulence region
Baldes et al. Transformation of soybean protoplasts from permanent suspension cultures by cocultivation with cells of Agrobacterium tumefaciens
JP3164731B2 (en) Method for promoting plant transcription using octopine T-DNA promoter
EP0126546B2 (en) Plant structural gene expression
Jen et al. Activity of T-DNA borders in plant cell transformation by mini-T plasmids
CA1251386A (en) Selection using opine synthase genes
JPS6181793A (en) Dismantled t-dna
JP2574130B2 (en) Genetic modification of plant cells

Legal Events

Date Code Title Description
MKEX Expiry