CA1341470C - Genetically engineered plant cells resistant to glutamine synthetase inhibitors - Google Patents

Abstract

The invention relates to a DNA fragment containing a determined gene, the expression of which inhibits the antibiotic and herbicidal effects of Bialaphos and related products. It also relates to recombinant vectors, containing such DNA fragment, which enable this protective gene to be introduced and expressed into cells and plant cells.

Description

This application is a division of Canadian application serial number 534,044 (Division R) filed on Rpril 7, 1987.
GENETICALLY ENGINEERED PLANT CELLS RESISTANT TO GLUTAMINE
SYNTHETASE INHIBITORS.
S
The invention relates to a process for protecting plant cells and plants against the action of glutamine synthetase inhibitors.
It also relates to applications of such process, IO particularly to the development of herbicide resistance into determined plants.
It relates further to non-biologically transformed plant cells and plants displaying resistance to glutamine synthetase inhibitors as well as to suitable DNA fragments ~5 and recombinants containing nucleotide sequences encoding resistance to glutamine synthetase inhibitors.
Glutamine synthetase (hereafter simply designated by GS) constitutes in most plants one of the essential enzymes for the development and life of plant cells. It is known that GS converts glutamate into glutamine. GS is involved in an efficient pathway (the only one known nowadays) in most plants for the detoxification of ammonia released by nitrate reduction, aminoacid degradation or photorespiration. Therefore potent inhibitors of GS are very toxic to plant cells. A particular class of herbicides has been developped, based on the toxic effect due to inhibit inhibition of GS in plants.
These herbicides comprise as active ingredient a GS inhibitor.
There are at least two possible ways which might lead to plants resistant to the inhibitors of the action of glutamine synthetase ; (1) by changing the target. It can be envisaged that mutations in the GS enzyme can lead to insensitivity towards the herbicide ; (2) by inactiva tion of the herbicide. Breakdown or modification of the ,;.... herbicide inside the plant could lead to resistance.

Hialaphos and phosphinothricin (hereafter simply designated by PPT) are two such inhibitors of the action of GS, (ref. 16, 17) and have been shown to possess excellent herbicidal properties (see more particularly ref. 2 as concerns Bialaphos).
Bialaphos has the following formula (I) .
cH3 NH2 i H3 i H3 HO - ; - CH 2 - CH 2 - C - CONH - C - CONH - C - COOH
O H H H
PPT has the following formula (II) .

HO - P - CH 2 - CH 2 - iH
O COOH
Thus the structural difference between PPT and Hialaphos resides in the absence of two alanine aminoacids in the case of PPT.
These two herbicides are non selective. They inhi-bit growth of all the different species of plants present on the soil, accordingly cause their total destruction.
2n Bialaphos was first disclosed as having antibiotic properties, which enabled it to be used as a pesticide or a fungicide. Bialaphos can be produced according to the process disclosed in united-states patent n' 3 832 399, assigned to MEIJI SEIKA KAISHA LTD. It comprises cultivating Strectomvces hvurosconicus, such as the strain available at the American Type Culture Collection, under the ATCC
number 21,705, and recovering Hialaphos from its culture medium. However, other strains, such as Streptomvces viridochromoaenes, also produce this compound (ref. 1).
_'. 0 Other tripeptide antibiotics which contain a PPT
moiety are or might be discovered in nature as well, e.g.
phosalacin (ref. 15).
PPT is also obtained by chemical synthesis and is commercially distributed by the industrial Company 1~~+14~0 HOECHST.
A number of Streptomyces species have been disclo-sed which produce highly active antibiotics which are known to incapacitate procaryotic cell functions or enzy-mes. The Streptomyces species which produce these anti-biotics would themselves be destroyed if they had not a self defence mechanism against these antibiotics. This self defence mechanism has been found in several instances to comprise an enzyme capable of inhibiting the antibiotic effect, thus of avoiding autotoxicity for the Streptomyces species concerned. This modification is generally reversed when the molecule is exported from the cell.
The existence of a gene which encodes an enzyme able to modify the antibiotic so as to inhibit the anti biotic effect against the host has been demonstrated in several Streptomyces producing antibiotics, for example, in S. fradiae, S. azureus, S. vinaceus, S. ervthreus, pro ducing neomycin, thiostrepton, viomycin, and MLS (Macro lide Lincosamide Streptogramin) antibiotics respectively (ref. 4), (ref. 5), (ref. 6),(ref.l4 by CRATER et al., 1982 describes standard techniques which can be used for bringing these effects to light).
In accordance with the present invention, it has been found that Streptomvces hvgroscopicug ATCC 21,705, also possesses a gene encoding an enzyme responsible of the inactivation of the antibiotic properties of Hialaphos. Experiments carried out by the applicants have lead to the isolation of such a gene and its use in a process for controlling the action of GS inhibitors, based on PPT or derived products.
An object of the invention is to provide a new process for controlling the action in plant cells and plants of GS inhibitolrs.
Another object of the invention is to provide DNA

fragments and DNA recombinants, particularly modified vectors containing said DNA fragments, which DNA fragments contain nucleotide sequences capable, when incorporated in plant cells and plants, to protect them against the action of GS inhibitors.
A further object of the invention is to provide non-biologically transformed plant cells and plants capable of neutralizing or inactivating GS inhibitors.
A further object of the invention is to provide a process for selectively protecting plant species against herbicides of a GS inhibitor type.
More specifically an object of the invention is to provide a DNA fragment transferable to plant cells- and to whole plants- capable of protecting them against the herbicidal effects of Bialaphos and of structurally analogous herbicides.
A further object of the invention is to provide plant cells resistant to the products of the class examplified by Hialaphos, which products possess the PPT
unit in their structure.
The process according to the invention for controlling the action in plant cells and plants of a GS
inhibitor when contacted therewith, comprises providing said plants with a heterologous DNA fragment including a foreign nucleotide sequence, capable of being expressed in the form of a protein in said plant cells and plants, under condition such as to cause said heterologus DNA
fragment to be integrated stably through generations in the cells of said plants, and wherein said protein has an enzymatic activity capable of inactivating or neutra-lization of said glutamine synthetase inhibitor.
A preferred DNA fragment is one derived from an antibiotic-producing-Streptomyces strain (or a sequence comprising a nucleotide sequence encoding the same activity) and which encodes resistance to a said GS

, 134'40 inhibitors.
Preferred nucleotide sequences for use in this invention encode a protein which has acetyl tranferase activity with respect to said GS inhibitors.
A most preferred DNA fragment according to the invention comprises a nucleotide sequence coding for a polypeptide having a PPT acetyl transferase activity.
A particular DNA fragment according to the invention, for the subsequent transformation of plant cells, consists of a nucleotide sequence coding for at least part of a polypeptide having the following sequen-ce .
X SER pR0 GLUT

ARu ARu FRO ALA ASP ILE ARG ARG ALA THR GLU ALA ASP MET PRO
2~8 ALA VAL CYS THR ILE VAL ASN HIS TYR ILE GLU THR SER THR VAL

ASN PHE ARG THR GLU PRO GLN GLU PRO GLN GLU TRP THR ASP ASP
31a . ' LE'J VAL ARG LEU ARG GLU ARG TYR PRO TRP LEU VAL ALA GLU VAL

ASP GLY GLU '.'AL ALA GLY ILE ALA TYR ALA GLY PRO TRP LYS ALA
G~8 ARu ASN ALA T'rR ASP TRP THR ALA GLU SER THR VAL TYR VAL SCR
4 ~3 PRO ARG HIS GLN AR a THR GLY LEU GLY SER THR LcU TYR THR HIS
a 98 LEU LEU LYS SCR LEU GLU ALA GLN GLY PHE LYS SCR VALVAL ALA

J~3 JaA

LcU GLY TYR ALA PRO ARG GLY MET LEU ARG ALA ALA GLYPHE LYS

633 _ . . _ HIS GLY ASN TRP HI ASP VAL GLY PHE-TRP GLN LEU ASP'PHE SGR
S

LE'J PRO VAL PRO PRO ARG PRO VAL LEU PRO VAL THR GLU ILE
723 ~ _ in which X represents MET or VAL, which part of said po-lypeptide is of sufficient length to confer protection against Bialaphos to plant cells, when incorportated genetically and expressed therein, i.e. as termed hereafter "plant-protecting capability' against Hialaphos.
A preferred DNA fragment consists of the following nucleotide sequence .
I 83 . -~ OTO AGC CCA GAA
--10_ - ~ _ _ . ._ , ,.
_ CGA CGC CCG GCC GAC ATC CGCCGT GCC ACC GAO GCO GAC AT~ CCO

~28 GCG GTC TGC ACC ATC GTC AACCAC TAC ATC GAG ACA AGC ACG GTC

~73 AAC TTC CGT ACC GAG CCG CAGGAA CCG CAG GAG TGG ACG GAC GAC

318 _ CTC GTC CuT CTG CGG GAG CGCTAT CCC TGG CTC GTC GCC GAG GTO

r . _ _ _ GAC GGC GAG FTC GCC GGC ATCGCC TAC GCG GGC CCC TGG AAG GCA

roe CGC AAC GCC TAC GAC TGG ACGGCC GAG TCG ACC GTG TAC GTC TCC

CCC CGC CAC CAG CGG ACG GGACTG GGC TCC ACG CTC TAC ACC CAC

CTG CTG AAG TCC CTG GAG GCACAG GGC TTC AAG AGC GTG GTC GCT

GTC ATC GGG CTG CCC AAC GAC CCG AGC GTG CGC ATG CAC GAG GCO

CTC GGA TAT GCC CCC CGC GGC ATG CTG CGG GCG GCC GGC TTC AAG
b33 CTG CCG GTA CCG CCC CGT CCG GTC CTG CCC GTC ACC GAG ATC

or of a part thereof expressing a polypeptide having , 13~+147a plant-protecting capability against Bialaphos.
The invention also relates to any DNA fragment differing from the preferred one indicated hereabove by the replacement of any of its nucleotides by others, yet without modifying the genetic information of the preferred DNA sequence mentioned hereabove (normally within the meaning of the universal genetic code), and furthermore to any equivalent DNA sequence which would encode a poly peptide having the same properties,particularly a Bialaphos-resistance-activity.
It will be understood that the man skilled in the art should be capable of readily assessing those parts of the nucleotide sequences that could be removed from either side of any of the DNA fragments according to the invention, for instance by removing terminal parts on either side of said DNA fragment, such as by an exonucleolytic enzyme, for instance Ba131, by recloning the remaining fragment in a suitable plasmid and by assaying the capacity of the modified plasmid to transform appropriate cells and to protect it against the Bialaphos antibiotic or herbicide as disclosed later, whichever assay is appropriate.
For the easiness of language, these DNA fragments will be termed hereafter as "Bialaphos-resistance DNA". In a similar manner, the corresponding polypeptide will be termed as "Hialaphos-resistance enzyme".
While in the preceding discussion particular emphasis has been put on DNA fragments capable, when introduced into plant cells and plants, to confer on them protection against Bialaphos or PPT, it should be under-stood that the invention should in no way be deemed as limited thereto.
In a same manner, the invention pertains to DNA
fragments which, when introduced into such plant cells, would also confer on them a protection against other GS

inhibitors, for instance of intermediate products involved in the natural biosynthesis of phosphinotricin, such as the compounds designated by the abbreviations MP101 (III), MP102 (IV), the formula of which are indicated hereafter .
1t 1°

13414~~

HO- P- CH2- CH2- CH- COOH (III) H

HO- P- CH2- CH 2- CH- CO- Ala- Ala (IV) I
H
More generally, the invention has opened the route to the production of DNA fragments which, upon proper incorporation into plant cells and plants, can protect them against GS inhibitors when contacted therewith, as this will be shown in a detailed manner in relation to Bialaphos and PPT in the examples which will follow.
This having been established, it will be 2n appreciated that any fragment encoding an enzymatic activity which would protect plant cells and plants against said GS inhibitors, by inactivationg, should be viewed as an equivalent of the preferred fragments which have been disclosed hereabove. This would apply especially to any DNA fragments that would result from genetic screening of the genomic DNAs of strains, particularly of antibiotic-producing strains, likely to possess genes which, even- though structurally different, would encode similar activity with respect to Hialaphos or PPT, or even with respect to other GS inhibitors. This applies to any gene in other strains producing a PPT derivative.
Therefore, it should be understood that the language "Bialaphos-resistance DNA' or "Bialaphos resistance enzyme' used thereafter as a matter of convenience is intended to relate not only to the DNAs and enzymes specifically concerned with resistance to PPT or most directly related derivatives, but more generally with other DNAs and enzymes which would be capable, under the same circumstances, of inactivating the action in plants of GS inhibitors.
The invention also relates to DNA recombinants containing the above defined Bialaphos-resistance DNA
fragments recombined with heterologous DNA, said heterologous DNA containing regulation elements and said Bialaphos-resistance DNA being under the control of said regulation elements in such manner as to be expressible in a foreign cellular environment compatible with said regulation elements. Particularly the abovesaid Bialaphos-resistance-DNA fragments contained in said DNA
recombinants are devoid of any DNA region involved in the biosynthesis of Bialaphos, when said Bialaphos-resistance-DNA fragment originate themselves from Bialaphos-producing strains.
By "heterologous DNA" is meant a DNA of an other origin than that from which said Bialaphos-resistance-DNA
originated, e.g. is different from that of a Streptomvces hvgrosconicus or Streptomvces viridochromogenes or even more preferably a DNA foreign to Streptomyces DNA.
Particularly said regulation elements are those which are capable of controlling the transcription and translation of DNA sequences normally associated with them in said foreign environment. "Cellular" refers both to micro-organisms and to cell cultures.
This heterologous DNA may be a bacterial DNA, par-3 c7 ticularly when it is desired to produce a large amount of the recombinant DNA, such as for amplification purposes.
In that respect a preferred heterologous DNA consists of DNA of ~. coli or of DNA compatible with ~. coli. It may be DNA of the same origin as that of the cells concerned or other DNA, for instance viral or plasmidic DNA known as ~~414~0 capable of replicating in the cells concerned.
Preferred recombinant DNA contains heterologous DNA compatible with plant cells, particularly Ti-plasmid DNA.
S
Particularly preferred recombinants are those which contain GS inhibitor inactivating DNA under the control of a promoter recognized by plant cells, particu-larly those plant cells on which inactivation of GS
inhibitors is to be conferred.
Preferred recombinants according to the invention further relate to modified vectors, particularly plasmids, containing said GS-inhibitor-inactivating DNA so posi-tioned with respect to regulation elements, including particularly promoter elements, that they enable said GS
inhibitor-inactivating DNA to be transcribed and translated in the cellular environment which is compatible with said heterologous DNA. Advantageous vectors are those so engineered as to cause stable incorporation of said GS
inhibitor inactivating DNA in foreign cells, particularly in their genomic DNA. Preferred modified vectors are those which enable the stable transformation of plant cells and which confer to the corresponding cells, the capability of inactivating GS inhibitors.
It seems that, as described later, the initiation codon of the Hialaphos-resistance-gene of the Streptomvces hvaroscopicus strain used herein is a GTG codon. Hut in preferred recombinant DNAs or vectors, the sialaphos resistance-gene is modified by substitution of an ATG
initiation codon for the initiation codon GTG, which ATG
enables translation initiation in plant cells.
In the example which follows, the plant promoter sequence which has been used was constituted by a promoter of the 35 S cauliflower mosaic virus. Needless to say that the man skilled in the art will be capable of selecting other plant promoters, when more appropriate in relation ~34~~70 to the plant species concerned.
According to an other preferred embodiment of the invention, particularly when it is desired to achieve transport of the enzyme encoded by the Bialaphos resistance-DNA into the chloroplasts, the heterologous DNA
fragment is fused to a gene or DNA fragment encoding a transit peptide, said last mentioned fragment being then intercalated between the GS inhibitor inactivating gene and the plant promoter selected.
As concerns means capable of achieving such cons-tructions, reference can be made to the following European Patent Office Application no. 85 402586.2, filed on December 20, 1885, published August 6, 1986.
Reference can also be made to the scientific li-terature, particularly to the following articles - VAN DEN HROECR et al., 1985, Nature, 313, 358-363 ;
- SCHREIER and al'., Embo. J., vol. 4, n' 1, 25-32.
For the sake of the record, be it recalled here that under the expression 'transit peptide', one refers to a polypeptide fragment which is normally associated with a chloroplast protein or a chloroplast protein sub-unit in a precursor protein encoded by plant cell nuclear DNA. The transit peptide then separates from the chloroplast pro-tein or is proteolitically removed, during the transloca-tion process of the latter protein into the chloroplasts.

Examples of suitable transit peptides are those associated with the small subunit of ribulase-1,5 biphosphate (RuHP) carboxylase or that associated with the chlorophyl a/b binding proteins.
There is thus provided DNA fragments and DNA
recombinants which are suitable for use in the process defined hereafter.
More particularly the invention also relates to a process, which can be generally defined as a process for producing plants and reproduction material of said plants including a heterologous genetic material stably integrated therein and capable of being expressed in said plants or reproduction material in the form of a protein capable of inactivating or neutralizing the activity of a glutamine synthetase-inhibitor, comprising the non biological steps of producing plants cells or plant tissue including said heterologous genetic material from starting plant cells or plant tissue not able to express that inhibiting or neutralizing activity, regenerating plants or reproduction material of said plants or both from said plant cells or plant tissue including said genetic material and, optionally, biologically replicating said last mentioned plants or reproduction material or both, wherein said non-biological steps of producing said plant cells or plant tissue including said heterologous genetic material, comprises transforming said starting plant cells or plant tissue with a DNA-recombinant containing a nucleotide sequence encoding said protein, as well as the regulatory elements selected among those which are capable of enabling the expression of said nucleotide sequence in said plant cells or plant tissue, and to cause the stable integration of said nucleotide sequence in said plant cells and tissue, as well as in the plant and reproduction material processed therefrom throughout generations.
The invention also relates to the cell cultures containing Bialaphos-resistance-DNA, or more generally said GS-inhibitor-inactivating DNA, which cell cultures have the property of being resistant to a composition containing a GS inhibitor, when cultured in a medium containing a such composition at dosages which would be destructive for non transformed cells.
The invention concerns more particularly those plant cells or cell cultures in which the Bialaphos-resistance DNA is stably integrated and which remains present over successive generations of said plant cells. Thus the resistance to a GS inhibitor, more particularly Bialaphos or PPT, can also be considered as a way of characterizing the plant cells of this invention.
Optionally one may also resort to hybridization experiments between the genomic DNA obtained from said plant cells with a probe containing a GS inhibitor inactivating DNA sequence, More generally the invention relates to plant cells, reproduction material, particularly seeds, as well 2n as plants containing a foreign or heterologous DNA
fragment stably integrated in their respective genomic DNAs, said fragments being transferred throughout generations of such plant cells, reproduction material, seeds and plants, wherein said DNA fragment encodes a protein inducing a non-variety-specific enzymatic activity capable of inactivating or neutralizing GS inhibitors, particularly Bialaphos and PPT, more particularly to confer an said plant cells, reproduction material, seeds and plants a corresponding non-variety-specific phenotype 3n of resistance to GS inhibitors.
"Non-variety-specific" enzymatic activity or phenotype aims at referring to the fact that they are not characteristic of specific plant genes or species as this will be illustrated in a non-limitative way by the examples which will follow. They are induced in said plant 1 s '~ ~ 4 1 4 '~ 0 materials by essentially non-biological processes applicable to plants belonging to species normally unrelated with one another and comprising the incorpora tion into said plant material of heterologous DNA, e.g.
bacterial DNA or chemically synthesized DNA, which does not normally occur in said plant material or which normally cannot be incorporated therein by natural breeding processes, and which yet confers a common phenotype (e. g. herbicide resistance) to them.
The invention is of particular advantageous use in processes for protecting field-cultivated plant species against weeds, Which processes comprise the step of treating the field with an herbicide, e.g. Bialaphos or PPT in a dosage effective to kill said weeds, wherein the cultivated plant species then contains in their genome a DNA fragment encoding a protein having an enzymatic activity capable of neutralizing or inactivating said GS
inhibitor.
By way of illustration only, effective doses for use in the abovesaid process range from about 0.4 to about 1.6 kg/Hectare of Hialaphos or PPT. .
There follows now a disclosure of how the preferred DNA fragment described hereabove was isolated starting from the Streptomvces hvaroscopicus strain available at the American Type Culture Collection under deposition number ATCC 21 705, by way of examplification only.
The following disclosure also provides the techni-que which can be applied to other strains producing compounds with a PPT moiety.
The disclosure will then be completed with the description of the insertion of a preferred DNA fragment conferring to the transformed cells the capability of inactivating Bialaphos and PPT. Thus the Bialaphos-inactivating-DNA fragment designated thereafter by Bialaphos-resistance gene or "sfr" gene, isolated by the above described technique into plasmids which can be used for transforming plant cells and conferring to them a resistance against Bialaphos, also merely by way of example for non-limitative illustration purposes.
The following disclosure is made with reference to the drawings in which .
- fig. 1 is a restriction map of a plasmid contai-ning a Strentomvces hv4roscopicus DNA fragment encoding Bialaphos-resistance, which plasmid, designated hereafter as pBG1 has been constructed according to the disclosure which follows ;
- fig. 2 shows the nucleotide sequence of a smal 15 ler fragment obtained from pBGI, subcloned into another plasmid (pBG39) and containing the resistance gene ;
- fig. 3 shows the construction of a series of plasmids given by way of example, which plasmids aim at providing suitable adaptation means for the insertion 2G therein of the Bialaphos-resistance gene or "sfr" gene ;
- fig. 4A and 4B show the construction of a series of plasmids given by way of example, which plasmids con-tain suitable plant cell promoter sequences able to ini-tiate transcription and expression of the foreign gene 25 inserted under their control into said plasmids ;
- fig. SA shows a determined fragment of the nu-cleotide sequence of the plasmid obtained in figure 3 ;
- fig. SB shows the reconstruction of the first codons of a Hialaphos-resistance gene, from a Fo~II/$q~II
3n fragment obtained from pBG39 and the substitution of an ATG initiation codon for the GTG initiation codon of the natural "sfr" gene ;
- fig. SC shows the reconstruction of the entire "sfr" gene, namely the last codons thereof, and its inser-tion into a plasmid obtained in figures 4A and 4H ;
- fig. 6A shows an expression vector containing ~3414~0 the 'sfr' gene placed under the control of a plant cell promoter ;
- fig. 6B shows another expression vector deriving from the one shown in fig. 6A, by the substitution of some nucleotides.
- fig. 7 shows the construction of a series of plasmids given by way of examples, to ultimately produce plasmids containing the promoter region and the transit peptide sequence of a determined plant cell gene, for the insertion of the "sfr' gene under the control of said pro-moter region and downstream of said transit peptide se-quence.
- fig. 8 to 11 will be referred to hereafter.
The following experiment was set up to isolate a eialaphos-resistance-gene from S. h~aroscopicus, according to standard techniques for cloning into Streptomvces.
2.5 ug of S. hvcroscopicus genomic DNA and 0.5 Ng of Strentomvces vector pIJ61 were cleaved with g,~I accor ding to the method described in ref. 6. The vector frag ments and genomic fragments were mixed and ligated (4 hours at 10'C followed by 72 hours at 4'C in ligation salts which contain 66 mM Tris-HC1~(pH 7.5), 1 mM EDTA, 10 mM MgCl2, 10 mM 2-mercaptoethanol and 0.1 mM ATP) at a total DNA concentration of 40 Ng ml 1 with T4 DNA lipase.
Ligation products were introduced into 3 x 109 S. lividans strain 66 protoplasts by a transformation procedure media-ted by polyethylene-glycol (PEG) as described hereafter.
These protoplasts gave rise to 5 x 107 colonies and 4 x 10 4 pocks after regeneration on 20 plates of R2 agar containing 0.5 't of Difco yeast extract (R2 YE). Prepara-tion and composition of the different mediums and buffers used in the disclosed experiments are described herein-after. When these lawns were replica-plated on minimal medium plates containing 50 ~g ml 1 Bialaphos, drug resis-tant colonies appeared at a frequency of 1 per 104 trans-formants. After purification of the drug resistant colo-nies, there plasmid DNA was isolated and used to retrans-form S. lividans protoplasts. Non selective regeneration followed by replication to Bialaphos-containing-medium demonstrated a 100 ~ correlation between pocks and Bialaphos resistant growth. The recombinant plasmids of several resistant clones all contained a 1.7 Kb PstI in Bert (see fig. 1).
,~ubclonina of the herbicide resistance uene The 1.7 Kb PstI insert was then subcloned into the high copy number streptomycete vector pIJ385 to generate plasmid pBG20. S. lividans strains which contained pHG20 were more than 500 times more resistant to Hialaphos.

5. lividan_s, growth is normally inhibited in minimal medium containing 1 Ng/ml Bialaphos ; growth of transformants containing pBG20 was not noticeably inhibited in a medium containing 500 Ng/ml Bialaphos. The PstI fragment was also subcloned in either orientation into the ~stI site of the plasmid pBR322, to produce plasmids pBG1 and pBG2, accor-ding to their orientation. A test on minimal M9 medium demonstrated that gs coli E8767 containing pBG1 or pHG2 was resistant to Bialaphos.
A ~ 1.65 Kb PstI - HamHI fragment was subcloned from pBG1 into the plasmid pUCl9 to produce the plasmid pBG39, and conferred Bialaphos resistance to E. coli, W3110, C600 and JM83.
Using an ~n vitro coupled transcription-translation system (ref. 5) from B. lividans extracts, the 1,65 Kb PstI - BamHI fragment in pBG39 was shown to direct the synthesis of a 22 Kd protein. In the following, this 1,65 Kb insert includes a fragment coding for a 22 Kd pro-tein and will be called "sfr" gene.

Fine mapnina and seQUencina of the Qene A 625 by af3A fragment was subcloned from pHG39 into pUCl9 and still conferred Bialaphos resistance to a E. coli W3110 host. The resulting clones were pHG93 and pBG94, according to the orientation.
The orientation of the gene in the ~3A fragment was indicated by experiments which have shown that Bialaphos resistance could be induced with IPTG from the pUCl9 lac promoter in pBG93. In the presence of IPTG
(0.5 mM) the resistance of pBG93/W3110 increased from 5 to 50 ug/ml on a M9 medium containing Bialaphos. The W3110 host devoid of pBG93, did not grow on M9 medium containing 5 ug/ml Hialaphos. These experiments demonstrated that the Sau3A fragment could be subcloned without loss of activi-ty. They also provided for the proper orientation as shown in the fig. 2, enclosed thereafter. The protein encoded by these clones was detected by using coupled transcription translation systems derived from extracts of S L lividans (ref. 7). Depending on the orientation of the Sau3A frag ment, translation products of different sizes were obser ved ; 22 Kd for pBG94 and t 28 Kd for pBG93. This indica ted that the S"~a 3A fragment did not contain the entire resistance gene and that a fusion protein was formed which included a polypeptide sequence resulting from the trans lation of a pUCl9 sequence.
In order to obtain large amounts of the protein, a 1.7 Kb g~ fragment from pBG1 was cloned into the high copy number Streptomycete replicon pIJ385. The obtained plasmid, pHG20, was used to transform S. hvcrrosconicus.
Transformants which contained this plasmid had more than 5 times as much PPT acetylating activity and also had increased amounts of a 22 kd protein on sodium dodecyl sulfate gels (SDS gels). Furthermore, both the acetyl transferase and the 22 kd protein appeared when the pro duction of Hialaphos begun. The correlation of the i yitr data, kinetics of synthesis, and amplified expres-sion associated with pBG20 transfarmants strongly implied that this 22 Kd band was the gene product.
The complete nucleotide sequence of the 625 by Sau3A fragment was determined as well as of flanking se-quences. Computer analysis revealed the presence of an open reading frame over the entire length of the Sau3A
fragment.
Character,j<zation of the sfr gene product A series of experiments were performed to deter mine that the open reading frame of the "sfr" gene indeed encoded the Bialaphos resistance enzyme. To determine the 5' end of the resistance gene, the NH2-terminal sequence of the enzyme was determined. As concerns more particu larly the technique used to determine the said sequence, reference is made to the technique developed by J. VANDEKERCKHOVE, Eur. J. Bioc. ~, p. 9-19, 1985, and to French patent applications~n' 85 14579 filed on October 1st, 1985 and published April 3, 1987, and no. 85 13046 filed on Se tember 2nd 1985 and p . published March 6, 1987.
This technique allows the immobilization on glass fibre sheets coated with the polyquaternary amine commer-cially available under the registered trademark POLYHRENE
of proteins and of nucleic acids previously separated on a sodium dodecylsulfate containing polyacrylamide gel. The transfer is carried out essentially as for the protein blotting on nitrocellulose membranes (ref. 8). This allows the determination of amino-acid composition and partial sequence of the immobilized proteins. The portion of the sheet carrying the immobilized 22 kd protein produced by S. hvar.~s~opicus pHG20 was cut out and the disc was mounted in the reaction chambre of a gas-phase sequenator to subject the glass-fibre bound 22 Kd protein to the Edman degradation procedure. The following amino-acid se quence was obtained .

2, Pro-Glu-Arg-Arg-Pro-Ala-Asp-Ile-Arg-Arg This sequence matched an amino-acid sequence which was deduced from the open reading frame of the 625 by Sau3A fragment. It corresponded to the stretch from codon 3 to codon 12.
Thus, the NH2-terminus of the 22 Kd protein was upstream of this sequence. It was determined that transla-tion of the actual protein was likely to be initiated 2 amino-acids earlier at a GTG initiation codon. GTG is often used as initiator codon in Streptomyces and transla-ted as methionine. The protein translated from the GTG
initiation codon would be 183 amino-acids long and would have a molecular weight of 20 550. This was in good agree-went with the observed approximate molecular weight of 22 000.
Furthermore, the termination codon, TGA, was lo-cated just downstream of the $au3A site. Cloning of the 625 by Sau3A fragment in a BamHI site digested pUCl9 did not result in the reconstruction of the termination codon.
This explained the fusion proteins which were observed in the in vitro transcription-translation analysis.
Mechanism of PPT-resistance Having defined a first phenotype and some of the physical characteristics of the resistance gene and its gene product, a series of experiments was then carried out to understand the mechanism by which it confers resistan ce. As described hereabove, PPT is the portion of Hialaphos which inhibits glutamine synthetase (GS) and that N-acetyl PPT is not an inhibitor. Using a standard assay (ref. 9), S. h~rQroscopicus ATCC 21 705 derivates were shown to contain a PPT acetyl transferase which was not found in S. lividans. The activity does not acetylate the Bialaphos tripeptide. S. lividans carrying the re sistance gene cloned in pBG20 or pHGl6 (a plasmid contai ning the 625 by aS~y3A fragment cloned into another yitr data, kinetics of synthesis, 22 13414?0 streptomycete vector, pIJ680) also contained the activity which could acetylate PPT but not Bialaphos. The PPT deri-ved reaction product produced by extracts of pBG20/
lividans was isolated in order to confirm that it was indeed acetyl-PPT. Anal sis b mass s ectrosco Y Y P py showed that the molecular weight had increased relative to PPT by the equivalent of one acetyl group. It was thus concluded that the 625 by Sau3A fragment contained sequences which code for PPT acetyl transferase.
The experimental conditions and reagents used in the techniques disclosed hereabove were as follows .
Preparation end composition of the mediums and buffers ~~ove used 1' P medium . 10.3 g of sucrose, 0.025 g of K2SOQ, 0.203 g of MgC12.6H20 and 0.2 ml of a trace element solu-tion are dissolved in 80 ml of distilled water and auto-claved . Then in order, 1 ml of KH2 P04 (0. 5 % ) , 10 ml of CaCl2, 2H20 (3.68 %), and 10 ml of TES buffer (0.25 M), pH . 7.2) are added. Trace element solution (per litre) .
ZnCl2 , 40 mg ; FeCl3 . 6H2 O, 200 mg ; CuCl2 . 2H2 O, 10 mg ;
MnCl 2. 4H 20, 10 mg ; Na2 H4 O~ . 10H2 O, 10 mg ;
(NH4) 6Mo~024 .4H20, 10 mg.
2' g,~y~ . 10. 3 g of sucrose, 0.025 g of K2 S04 , 1.012 g of MgC12.6H20, 1 g of glucose, 0.01 g of Difco casamino acids, and 2.2 g of Difco agar are dissolved in 80 ml distilled water and autoclaved. 0.2 ml of trace element solution, 1 ml of KH 2P04 (0.5 %), 8.02 ml of CaC12.2H20 (3.68 %), 1.5 ml of L-proline (20 %), 10 ml of TES buffer (0.25 M) (gH . 7.2), 0.5 ml of (1 M) NaOH, 5 ml of yeast extract (10 %) are sequentially added.
3' ~ : 10 mM TRIS HCl, 1 mM EDTA, pH 8Ø
4' YEME . Difco yeast extract (0.3 %), Difco pep-tone (0.5 %), oxoid malt extract (0.3 %), glucose (1 ~t).
Transformation of S lividans protoptasta 1. A culture composed of 25 ml YEME, 34 % sucrose, 0.005 M

~o 13 ~4 MgCl2, 0.5 1: glycine, in a 250 ml baffled flask, is cen-trifuged during 30 to 36 hours.
2. The pellet is suspended in 10.3 se sucrose and centri-fuged. This washing is repeated once.
3. The mycelium is suspended in 4 ml lysozyme solution (1 mg/ml in P medium with CaCl2 and MgCl2 concentrations reduced to 0.0025 M) and incubated at 30'C for 15 to 60 minutes.
4. The solution is mixed by pipetting three times in a 5 ml pipette and incubated for further 15 minutes.
5. P medium (5 ml) is added and mixed by pipetting as in step 4.

6. The solution is filtered through cotton wool and pro toplasts are gently sedimented in a bench centrifuge at 800 x G during 7 minutes.

7. Protoplasts are suspended in 4 ml P medium and centri-fuged again.

8. Step 7 is repeated and protoplasts are suspended in the drop of P medium left after pouring off the supernatant (for transformation).

9. DNA is added in less than 20 N1 TE.

10. 0.5 ml PEG 1 000 solution (2.5 g PEG dissolved in 7 . 5 ml of 2. 5 ~ sucrose, 0.0014 IC2 S04 , 0. 1 M CaCl2 , 0. 05 M
TRIS-malefic acid, pH 8.0, plus trace elements) is immedia tely added and pipetted once to mix the components.

11. After 60 seconds, 5 ml of P medium are added and the protoplasts are sedimented by gentle centrifugation.

12. The pellet is suspended in P medium (1 ml).

13. 0.1 ml is plated out on R2YE plates (for transforma-tion dry plates to 85 ~t of their fresh weigh e. g. in a laminar flow cabinet).

14. Incubation at 30'C.
A - Construction of a 'sf~," uene cassette A 'sfr" gene cassette was constructed to allow subsequent cloning in plant expression vectors.

~~4~470 Isolation of a ~I-~q~,,II fragment from the plas-mid pBG39 containing a "sfr" gene fragment led to the loss of the first codons, including the initiation codon, and of the last codons, including the stop codon.
This fragment of the "sfr" gene could be recons-tructed ~,n vitro with synthetic oligonucleotides which encode appropriate amino-acids.
The complementary synthetic oligonucleotides were 5'-CATGAGCCCAGAAC and 3'-TCGGGTCTTGCTGC.

Hy using such synthetic oligonucleotides, the 5' end of the "sfr" gene could be reformed and the GTG ini-tiation codon substituted for a codon well translated by plant cells, particularly an ATG codon.
The DNA fragment containing the oligonucleotides linked to the "sfr" gene was then inserted into an appro-priate plasmid, which contained a determined nucleotide sequence thereafter designated by an "adapter" fragment.
This adapter fragment comprised .
- a TGA termination codon which enabled the last codons of the "sfr" gene to be reformed ;
- appropriate restriction sites which enabled the insertion of the fragment of the nucleotide sequence comprising the "sfr" gene partially reformed with the synthetic oligonucleotides ; this insertion resulted in the reconstruction of an intact "sfr" gene ;
- appropriate restriction sites for the isolation of the entire "sfr" gene.
The "sfr" gene was then inserted into another plasmid, which contained a suitable plant promoter sequen ce. The plant promoter sequence consisted of the cauli flower mosaic virus promoter sequence (p35S). Of course the invention is not limited to the use of this particular promoter. Other sequences could be chosen as promoters suitable in plants, for example the TR 1'-2' promoter region and the promoter fragment of a Rubisco small subunit gene from Arabidopsis thaliana hereafter described.
1' construction of the nlasmid nLK56.2 fia.31 The construction of plasmid pLK56.2 aimed at ob-taining a suitable adaptor including the following sequen ce of restriction sites , tS~I, a~HI, NcoI, ~I, III, 1M-I, ~amHI, HindIII and XbaI.
The starting plasmids used for this construction, 10 pLK56, pJB64 and pLK33 were those disclosed by BOTTERMAN
(ref. 11).
The DNA fragments hereafter described were iso-lated and separated from low melting point agarose (LGA).
The plasmid pLK56 was cleaved by the enzymes CHI
and NdeI. A NcoI-Nd_e_I fragment (referred to in the drawings by arc "a" in broken line) obtained from plasmid pJB64 was substituted in pLK56 for the BamHI-~gI fragment shown at "b". Ligation was possible after filling in the HamHI and ICI protruding ends with the DNA polymerase I
of E. coli (Klenow's fragment).
Particularly recircularization took place by means of a T4 DNA lipase. A new plasmid pLK56.3 was obtained.
This plasmid was cleaved by the enzymes ~I and PstI.
The ~a HI-p~I fragment of pLK33 (c) (on fig. 3) was substituted for the ~I-~stI fragment (d) of pLK56.3, of ter repairing of their respective ends by Klenow's fragment.
After recircularization by means of the T4 DNA
lipase, the obtained plasmid pLK56.2 contained a nucleo tide sequence which comprised the necessary restriction sites for the subsequent insertion of the "sfr" gene.
2' Construction of the plasm~id pGSH150 ffia. 4A) Parallel with the last discussed construction, ~34~47~

there was produced a plasmid containing a promoter sequen-ce recognized by the polymerases of plant cells and inclu-ding suitable restriction sites, downstream of said promo-ter sequence in the direction of transcription, which restriction sites are then intented to enable the accomo-dation of the "sfr" gene then obtainable from pLK56.2, under the control of said plant promoter.
Plasmid pGV825 is described in DEBLAERE et al.
(ref. 10). Plasmid pJB63 is from BOTTERMAN (ref. 11).
pGV825 was linearized with III and recircula-rized by the T4 DNA lipase, resulting in the deletion of an internal vuII fragment shown at (e), (plasmid pGV956).
pGV956 was then cleaved by arHI and $g~,II.
The amHI-B-~c.,II fragment (f) obtained from pJH63 was dephosphorylated with calf intestine phosphatase (CIP) and substituted fox the BamHI-III fragment of pGV956.
Plasmid pGV1500 was obtained after recirculariza-tion by means of T4 DNA lipase.
An EcoRI- indIII fragment obtained from plasmid pGSH50 was purified. The latter plasmid carried the dual TR 1'-2' promoter fragment described in VELTEN et al., (ref.l3). This fragment was inserted in pGV1500, digested with ~I and HindIII and yielded pGSH150.
This plasmid contains the promoter fragment in front of the 3' end of the T-DNA transcript 7 and a p~HI
and ~I sites for cloning.
3' construction of the plasmid n~SJ260 (fia. 4H) CP3 is a plasmid derived from pBR322 and which contains the 35S promoter region of cauliflower mosaic virus within a CHI fragment.
pGSH150 was cut by BamHI and ~g~,,II.
The BamHI-III fragment (h) of CP3, which con tained the nucleotide sequence of p35S promoter, was substituted for the HamHI-,~g3,II fragment (i) in pGSH150 to form plasmid pGSJ250. pGSJ250 was then opened at its III

restriction site.
A aiHI fragment obtained from mGV2 (ref. 12) was inserted in pGSJ250 at the ~,g~II site to form plasmid pGSJ260.
However prior to inserting the "sfr" gene obtai-nable from pLK56.2 into plasmid pGSJ260, it was still de-sirable to further modify the first in order to permit insertion in a more practical manner. Thus pLK56.2 was further modified as discussed below to yield pGSRI.
Starting from plasmid pGSJ260, two plasmid cons-tructions for subsequent transformations of plant cells Were made .
- a first plasmid permitting the expression of the "sfr" gene in the cytoplasm of plant cells, and - a second plasmid so modified as to achieve transport of the Bialaphos-resistance enzymes to the chlo-roplasts of plant cells.
First case pl~smid enabling the expression of the "sfr"
gene in the cytoplasm of plant cells Cloning of the sfr gene cassette in a plant excression yector (pGSR2) (fig. 5) On figure SA, the nucleotide sequence of the adap-ter of pLK56.2 is shown. In particular, the locations of BamHI, ~I, $c,~II restriction sites are shown.
This adapter fragment was cleaved by the enzymes Ncol and ~q~,II.
Figure 5B shows the ~I-~q],II fragment (j) obtai-ned from pBG39. The locations of these two restriction sites are shown on figure 2.
Using synthetic oligonucleotides, the first codons of the "sfr" gene were reformed, particularly the 5' end of the gene in which a ATG initiation codon was substitu-ted for the initial GTG codon.
This ~g,$I-$c~II fragment completed with the syn-thetic oligonucleotides was then substituted in pLK56.2 ~~~~470 for the ~I-III fragment of the adapter. The 3' end of the gene was thus reformed too, after recircularization with T4 DNA ligased. The plasmid obtained, pGSRI, thus contained the entire "sfr" gene inserted in its adapter.
The plasmid pGSJ260 was then opened by BamHI
(fig. 5C) and the CHI fragment obtained from pGSRI, which contained the entire "sfr" gene, was inserted into pGSJ260.
The obtained plasmid, pGSR2 (see fig. 6A) contained a pBR322 replicon, a bacterial streptomycin resistance gene (SDM-SP-AD-transferase) and an engineered T-DNA consisting of .
- the border fragments of the T-DNA ;
- a chimeric kanamycin gene which provided a domi-nant selectable marker in plant cells ; and - a chimeric "sfr" gene.
The chimeric "sfr" gene consisting of .
- the promoter region of the cauliflower mosaic virus (p35S) ;
- the "sfr" gene cassette as described in fig. 5 ;
- the 3' untranslated region, including the poly-adenylation signal of T-DNA transcript 7.
pGSR2 was introduced into Avrobacterium ~u~efaciens recipient C58C1RifR (pGV2260) according to the procedure described by DEBLAERE et al. (ref. 10).
This strain was used to introduce the chimeric "sfr" gene in N. t~bacum SR1 plants.
Two variant plasmids deriving from pGSR2, namely pGSFR280 and pGSFR281, have been constructed. They differ in the untranslated sequence following the transcription initiation site. In pGSR2, this fragment consists of the following sequence .
GAGGACACGCTGAAATCACCAGTCTCGGATCC~ ;
while it consists of .
GAGGACACGCTGAAATCACCAGTCTCTCTACAAATCGATCCATG

29 ~ 3 4 in pGSR280 and of GAGGACACGCTGAAATCACCAGTCTCTCTACAAATCGATG
in pGSFR281, with an ATG codon being the initiation codon of the "sfr" gene. The "sfr" gene is also fused to the TR1'-2' promoter in the plasmid pGSH150 (fig. 4A) yielding pGSFR160 and pGSFR161 (fig. 68). These plasmids contain slight differences in the pTR2 "sfr" gene configuration .
the "sfr" gene is correctly fused to the endogenous gene 2' ATG in pGSFR161 (for sequences see ref. 13), whereas 4 extra base pairs (ATCC) are present just ahead of the ATG
codon in pGSFR160. Otherwise, plasmids p65FR161 and p65FR160 are completely identical.
All plasmids are introduced in Acrroba~_cterium by cointegration in the acceptor plamids pGV2260 yielding the respective plasmids pGSFR1280, pGSFR1281, pGSFR1160 and pGSFR1161.
Second case construction of a plasmid containinc th "sfr" gene downstrean~of a DNA seguence encodincr a transit a a t r t'o " " a a s' o chloroplasts In another set of experiments, the nucleotide sequence which contained the 'sfr" gene was fused to a DNA
sequence encoding a transit peptide so as to enable its transport into chloroplasts.
A fragment of the "sfr" gene was isolated from the adapter fragment above described and fused to a transit peptide. With synthetic oligonucleotides, the entire "sfr"
gene was reconstructed and fused to a transit peptide.
The plasmid (plasmid pATS3 mentioned below) which contained the nucleotide sequence encoding the transit peptide comprised also the promoter sequence thereof.
~onstruct~on of thEg plasmid pGSR4 which contains the "sfr"
crepe fused to a DNA seguence encodina t ansit peptide (fia. 7) ~341~70 Plasmid pLK57 is from HOTTERMAN, (ref.ll). Plasmid pATS3 is a pUCl9 clone which contains a 2 Kb EcoRI genomic DNA fragment from Arabidopsis thaliana comprising the pro moter region and the transit peptide nucleotide sequence of the gene, the expression thereof is the small subunit of ribulose biphosphate carboxylase (ssu). The ~. thaliana small subunit was isolated as a 1 500 by SRI-~I frag-ment. The SphI cleavage site exactly occurs at the site where the coding region of the mature ssu protein starts.
Plasmids pLK57 and pATS3 were opened with coRI
and SPItI. After recircularization by means of the T4 DNA
lipase, a recombinant plasmid pLKABI containing the se-quence encoding the transit peptide (Tp) and its promoter region (Pssu) was obtained.
In order to correctly fuse the "sfr" gene at the cleavage site of the signal peptide, the N-terminal gene sequence was first modified. Since it was observed that N-terminal gene fusions with the "sfr" gene retain their enzymatic activity, the second codon (AGC) was modified to a GAC, yielding an ~I site overlapping with the ATG
initiator site. A new plasmid, pGSSFR2 was obtained. It only differs from pGSR1 (fig. 5B), by that mutation. The NcoI-HamHI fragment obtained from pGSFR2 was fused at the ~I end of the transit peptide sequence. In parallel, the "sfr" gene fragment was fused correctly to the ATG initia-tor of the ssu gene (not shown in figures).
~ntxoduction of the "sfr" Qene into a different plant species The Bialaphos-resistance induced in plants by the expression of chimeric genes, when the latter have been transformed with appropriate vectors containing said chimeric genes, has been demonstrated as follows. The recombinant plasmids containing the "sfr" gene were intro-duced separately by mobilization into P~arobacterium strain C58C~ RifR (pGV2260) accordin to the g procedure described by DEBLAERE and al., Nucl. Acid. Res., 13, p. 1 477, 1985.
Recombinant strains containing hybrid Ti plasmides were formed. These strains were used to infect and transform leaf discs of different plant species, according to a method essentially as described by HORSH and al., 1985, Science, vol. 227. Transformation procedure of these different plant species given by way of example, is described thereafter.

1. Leaf disc transformation of ~~cotiana tabacum Used Media are described thereafter .
A1 MS salt/2 + 1% sucrose 0.8 ~. agar pH 5.7 A 10 BS-medium + 250 mg/1 NH4NO3 750 mg/1 CaCl2 2H20 .
0.5 g/1 2-(N-Morpholino)ethane-sulfonic acid (MES) pH 5.7 30 g/1 sucrose A ~~ H5-medium + 250 mg/1 NH4N03 0.5 g/1 MES pH 5.7 2 % glucose 0.8 % agar 40 mg/1 adenine + 1 mg/1 6-Benzylaminopurine (BAP) 0.1 mg/1 Indole-3-acetic acid (IAA) 500 mg/1 Claforan A 12 B5-medium + 250 mg/1 NH4N03 0.5 g/1 MES pH 5.7 2 % glucose 0.8 % agar 40 mg/1 adenine + 1 mg/1 BAP
200 mg/1 claforan , A13 MS-salt/2 + 3 % sucrose 0.5 MES g/1 pH 5.7 0.7 ~t agar 200 mg/1 claforan Bacterial medium = min A . (Miller 1972) 60 mM
K 2EIP0 Q, 3H 20, 33 mM KH2P04 ; 7.5 mM (NH4 )2S04 1.7 M trinatriumcitrat; 1 mM
MgS04 ;
2 g/1 glucose ; 50 mg/1 vita-mine B1 - Plant material .
Nicotiana tabacum cv. Petit Havana SR1 Plants are used 6 to 8 weeks after subculture on medium A1 - Infection .
- midribs and edges are removed from leaves.
- Remaining parts are cut into segments of about 0.25 cm2 and are placed in the infection medium A10 (about 12 segments in a 9 cm Petri dish containing 10 ml A10).
- Segments are then infected with 25 N1 per Petri dish of a late log culture of the ~arobacteriu~ strain grown in min A medium.
- Petri dish are incubated for 2 to 3 days at low light intensity.
- After 2 to 3 days medium is removed and replaced by 20 ml of medium A10 containing 500 mg/1 clarofan.
_ Selection and shoot induction - The leaf discs are placed on medium A11 contain-ing a selective agent .
100 mg/1 kanamycin and 10 to 100 mg/1 phosphinotricin.

~~41470 - Leaf discs are transferred to fresh medium week-ly.
- After 3 to 4 weeks regenating calli arise. They are sepa rated and placed on medium A~2 with the same con s centration of selective agent as used for the selection.
- Rootinc - After 2 to 3 weeks the calli are covered with shoots, which can be isolated and transferred to rooting medium A~3 (without selection).
- Rooting takes 1 to 2 weeks.
- After a few more weeks, these plants are propagated on medium A~.
2. Tuber disc infection of Solanum_ tuberosum (potato) Used media are described thereafter .
C~ H5-medium + 250 mg/1 NH4N03 300 mg / 1 ( CaCH 2P0 4 ) 2 0.5 g/1 MES pH 5.7 0.5 g/1 polyvinylpyrroli-done (PVP) 40 g/1 mannitol (=0.22M) 0.8 ~ agar 40 mg/1 adenine C 2 H5-medium + 250 mg/ 1 NH4 N03 400 mg/1 glutamine 0.5 g/1 MES pH 5.?

0.5 g/1 PVP
40 g/1 mannitol 40 mg/1 adenine 0.8 =. agar 35 1 3 4'i 4 70 + 0.5 mg/1 transzeatine 0.~ mg/1 IAA
500 mg/1 clarofan C 5 MS salt/2 + 3 ~S sucrose 0 . 7 : agar pH 5.7 C 7 B5-medium + 250 mg/1 NH4 N03 400 mg/1 glutamine 0.5 g/1 MES pH 5.7 0.5 g/1 PVP
20 g/1 mannitol ~5 20 g/1 glucose 40 mg/1 adenine 0.6 °b agarose + 0.5 mg/1 transzeatine 20 0~'1 mg/1 IAA
500 mg/1 clarofan C 8 H5-medium + 250 mg/ 1 NH4 N03 400 mg/1 glutamine 25 0.5 g/1 MES pH 5.7 0.5 g/1 PVP
20 g/1 mannitol 20 g/1 glucose 40 mg/1 adenine 0.6 ~ agarose + 200 mg/1 clarofan 1 mg/1 transzeatine C 9 B5-medium + 250 mg/1 NH4 N03 400 mg/1 glutamine 0.5 g/1 MES pH 5.7 0.5 g/1 PVP
20 g/1 mannitol 20 g/1 glucose 40 mg/1 adenine 0.6 ~ agarose + 1 mg/1 transzeatine 0.01 mg/1 Gibberellic acid A3 (GA3) 100 mg/1 clarofan C 11 MS salt/2 + 6 °~ sucrose 0 . ? 1; agar Bacterial medium = min A . (Miller 1972 60 mM K2HP04.3H20;
33 mM KH ZP04; 7 . 5 mM (NH4 )2 S04 ;
1.7 trinatriumcitrat; 1 mM
MgS04 ;
2 g/1 glucose; 50 mg/1 vitami-ne 81 .
- PLant mare Tubers of Solanu~~berosum c.v Berolina c.v D~sir~e - Infec ion - Potatoes are peeled and washed with water.
- Then they are washed with concentrated commer-cial bleach for 20 minutes, and - rinsed 3 to 5 times with sterile water.
- The outer layer is removed (1 to 1.5 cm) - The central part is cut into discs of about 1 cm2 and 2 to 3 mm thick.
- Discs are placed on medium C1 (4 pieces in a 9 .
'341470 cm Petri dish).
- 10 N1 of a late log culture of an Aarobacterium strain grown in min A medium is applied on each disc.
- Discs are incubated for 2 days at low light intensity.
- ,S~:lection and shoot induction - Discs are dried on a filter paper and transfer-red to medium C2 with 100 mg/1 kanamycin.
- After one month small calli are removed from the discs and transferred to medium C.~ containing 50 mg/1 kanamycin.
- After a few more weeks, the calli are transfer-red to medium C8 containing 50 mg/1 kanamycin.
- If little shoots start to develop, the calli are transferred to elongation medium C9 containing 50 mg/1 Kanamycin.
- B~otina - Elongated shoots are separated and transferred to rooting medium C11.
- Rooted shoots are propagated on medium C5.
3. Leaf disc infection of Lvcope~~i~u~n esculentum (tomato) Used media are described thereafter A1 MS salt/2 + 1 ~ sucrose 0.8 ~t agar pH 5.7 B ~ H5-medium + 250 mg/1 NH4N03 0.5 gJl MES pH 5.7 0.5 g/1 PVP
300 mg/ 1 Ca ( H2 P04 )2 2 ~: glucose mg/1 adenine 40 g/1 mannitol B 2 B5-medium + 250 mg/1 NH4 N03 0.5 g/1 MES pH 5.7 0.5 g/1 PVP
400 mg/1 glutamine 2 a glucose 0.6 a agarose 40 mg/1 adenine 40 g/1 mannitol + 0.5 mg/1 transzeatine 0.01 mg/1 IAA
500 mg/1 claforan B3 B5-medium + 250 mg/1 NH4N03 0.5 g/1 MES pH 5.7 0.5 g/1 PVP
400 mg/1 glutamine 2 s. glucose 0.6 ~ agarose 40 mg/1 adenine g/1 mannitol + 0.5 mg/1 transzeatine 0.01 mg/1 IAA
500 mg/1 clarofan H4 B5-medium + 250 mg/1 NH4N03 0.5 g/1 MES pH 5.7 0.5 g/1 PVP
400 mg/1 glutamine 2 ~ glucose 0.6 ~ agarose mg/1 adenine 35 20 g/1 mannitol + 0.5 mg/1 transzeatine 0.01 mg/1 IAA
500 mg/1 clarofan B5 B5-medium + 250 mg/1 NH4N03 0.5 g/1 MES pH 5.7 0.5 g/1 PVP
400 mg/1 glutamine 2 % glucose 0.6 % agarose 40 mg/1 adenine 10 g/1 mannitol + 0.5 mg/1 transzeatine 0.01 mg/1 IAA
500 mg/1 clarofan B 6 B5-medium + 250 mg/1 NH4 N03 0.5 g/1 MES pH 5.7 0.5 g/1 PVP
400 mg/1 glutamine 2 se glucose 0.6 % agarose 40 mg/1 adenine + 0.5 mg/1 transzeatine 0.01 mg/1 IAA
200 mg/1 clarofan B 7 H5-medium + 250 mg/1 NH4 N03 0.5 g/1 MES pH 5.7 0.5 g/1 PVP
400 mg/1 glutamine 2 % glucose 0.6 % agarose 40 mg/1 adenine + 1 mg/1 transzeatine 200 mg/1 clarofan B8 MS salt/2 + 2 % sucrose 0.5 g/1 MES pH 5.7 0.7 % agar 10 g 9 g5-medium + 250 mg/1 NH4 N03 0.5 g/1 MES pH 5.7 0.5 g/1 PVP
2 % glucose 0.6 % agarose 40 mg/1 adenine + 1 mg/1 transzeatine O . 01 mg/ 1 GA3 Bacterial medium = min A . (Miller 1972) 60 mM
K2 HP04 . 3H2 O ;
33 mM KH2P04; 7. 5 mM (NH4 )2 S04 ;
1.7 M trinatriumcitrat; 1 mM
MgS04 ;
2 g/1 glucose; 50 mg/1 vitami-ne B1 - Plank m~te~ial COl~grsicum esculentum cv. Lucullus.
Plants are used 6 weeks after subculture on medium A1.
- In~~ction - Midrib is removed from the leaves.
- Leaves are cut in segments of about 0.25 to 1 cm2 (the edges of the leaves are not wounded, so that only maximum 3 sides of the leaf pieces is wounded).
- Segments are placed in infection medium B1 1341+74 (upside down), about 10 segments in a 9 cm Petri dish.
- Segments are then infected wiht 20 ul per Petri dish of a late log culture of the Acxrobacterium strain grown in min A medium.
- Petri dishes incubate for 2 days at low light intensity.
- Medium is removed after 2 days and replaced by 20 ml of medium B1 containing 500 mg/1 clarofan.
- Selection and shoot induction - The leaf discs are placed in medium B2 + 50 or 100 mg/1 kanamycin.
- Each 5 days the osmotic pressure of the medium is lowered by decreasing the mannitol concentration, transfers are done consecutively in medium B3 , H4 , H5 , and B 6.
- After one month calli with meristems are separa-ted from the leaf discs and placed on medium B~ with 50 or 100 mg/1 kanamycin.
- Once little shoots have formed, calli are transferred to elongation medium B9 with 50 or 100 mg/1 kanamycin.
- Rooting - Elongated shoots are separated and transferred to medium B8 for rooting.
- Plants are propagated on medium A1.
Greenhouse tests for herbicide resistance Material and method In this experiment, two herbicides comprising phosphinotricin as active ingredient, are used.
These compounds are those commercially available under the registered trademarks BASTAR and MEI,3I
HERHIACER.
These products are diluted to 2 ~ with tap water.
Spraying is carried out on a square metre area from the four corners. Temperature of the greenhouse is about 22'C
for tobaccos and tomato, and above 10'C to 15'C for potato.
Results - Tobacco spraytest a) Nicotiana tab~cum cv. Petit Havana SR1 plants transformed with the chimeric "sfr" genes as present in pGSFR1161 or pGSFR1281, as well as unstrans-formed control plants (from 10 cm to 50 cm high) are treated with 20 1 BASTAR/ha. Control SR1 plants die after 6 days, while transformed plants are fully resistant to 20 1 BASTAR/ha and continue growing undistinguishable from untreated plants. No visible damage is detected, also the treatment is repeated every two weeks. The treatment has no effect in subsequent flowering. The recommended dose of BASTAR herbicide in agriculture is 2.5-7.5 1/ha.
b) A similar experiment is performed using 8 1/ha MEIJI HERBIACER. The transformed plants (the same as above) are fully resistant and continue growing undistin guishable from untreated plants. No visible damage is detectable.
- Potato spraytest Untransformed and transformed potato plants (Solanum to erosum cv. Berolina) (20 cm high) with the chimeric "sfr" gene as present in pGSFR1161 or pGSFR1281 are treated with 20 1 BASTAR/ha. Control plants die after 6 days while the transformed plants do not show any visible damage. They grow undistiguishable from untreated plants.
- tomato spraytest Untransformed and transformed tomato plants (lvcoug~r,,g~u~t esculentum c.v. luculus) (25 cm high) with the chimeric "sfr" gene as present in pGSFR1161 and pGSFR1281 are treated with 20 1 HASTAR/ha. control plants die after six days while transformed plants are fully resistant. They do not show any visible damage and grow undistiguishable from untreated plants.
- Growth control of phytopathogenic fungi with transformed plants In another set of experiments, potato plants ex pressing chimeric "sfr" genes as present in pGSFR1161 or pGSFR1281 are gxown in a greenhouse compartment at 20'C
under high humidity. Plants are innoculated by spraying 1 ml of a suspension of 106 Phvtov,~t~ra infestans spores per ml. Plants grow in growth chambers (20'c, 95 % humidity, 4 000 lux) until fungal disease symptoms are visible (one week). One set of the plants are at that moment sprayed with Bialaphos at a dose of 8 1/ha. Two weeks later, untreated plants are completely ingested by the fungus.
The growth of the fungus is stopped on the Bialaphos treated plants and no further disease symptoms evolve. The plants are effectively protected by the fungicide action of Hialaphos.
- Transmission of the PPT resistance through seeds Transformed tobacco plants expressing the chimeric "sfr" gene present in pGSFR1161 and pGSFR1281 are brought to flowering in the greenhouse. They show a normal fertility.
About 500 F1 seeds of each plant are sown in soil, F1 designating seeds of the first generation, i.e directly issued from the originally transformed plants. When seedlings are 2-3 cm high, they are sprayed with 8 1 HASTAR/ha. 7 days later, healthy and damaged plants can be distinguished in a ratio of approximately 3 to 1. this shows that PPT resistance is inherited as a dominant marker encoded by a single locus.
10 resistant F1 seedlings are grown to maturity and seeds are harvested. F2 seedlings are grown as described above and tested fox PPT-resistance by spraying s BASTA R at a dose of 8 1/ha. Some of the F1 plants produce F2 seedlings which are all PPT-resistant showing that these plants are homozygous for the resistance gene. The invention also concerns plant cells and plants non essentially-biologically-transformed with a GS inhibitor-inactivating-gene according to the invention.
In a preferred embodiment of the invention, plant cells and plants are non-biologically-transformed with the "sfr" gene hereabove described.
Such plant cells and plants possess, stably integrated in their genome, a non-variety-specific character which render them able to produce detectable amounts of phosphinotricin-acetyl transferase.
This character confers to the transformed plant cells and plants a non-variety-specific enzymatic activity capable of inactivating or neutralizing GS inhibitors like Bialaphos and PPT.
Accordingly, plant cells and plants transformed according to the invention are rendered resistant against the herbicidal effects of Bialaphos and related compounds.
Since Bialaphos was first described as a fungicide, transformed plants can also be protected against fungal diseases by spraying with the compound several times.
In a preferred embodiment, Bialaphos or related compounds is applied several times, particularly at time intervals of about 20 to 100 days.
The invention also concerns a new process for selectively protecting a plant species against fungal diseases and selectively destroying weeds in a field comprising the steps of treating a field with an herbicide, wherein the plant species contain in their genome a DNA fragment encoding a protein having an enzymatic activity capable of neutralizing or inactivating GS inhibitors and wherein the used herbicide comprises as 45 ~ 3 4 1 4 7 0 active ingredient a GS inhibitor.
It comes without saying that the process according to the invention can be employed with the same efficiency, either to only destroy weeds in a field, if plants are not infected with fungi , either to only stop the development of fungi if the latter appears after destruction of weeds.
In a preferred embodiment of the process according to the invention, plant species are transformed with a DNA
fragment comprising the "sfr" gene as described hereabove, and the used herbicide is PPT or a related compound.
Accordingly, a solution of PPT or related compound is applied over the field, for example by spraying, several times after emergence of the plant species to be cultivated, until early and late germinating weeds are destroyed.
It is quite evident that before emergence of plant species to be cultivated, the field can be treated With an herbicidal composition to destroy weeds.
On the same hand, fields can be treated even before the plant species to be cultivated are sowed.
Before emergence of the desired plant species, fields can be treated with any available herbicide, including Bialaphos-type herbicides.
After emergence of the desired plant species, Bialaphos or related compound is applied several times.
In a preferred embodiment, the herbicide is applied at time intervals of about from 20 to 100 days.
Since plants to be cultivated are transformed in such a way as to resist to the herbicidal effects of Hialaphos-type herbicides, fields can be treated even after emergence of the cultivated plants.
This is particularly useful to totally destroy early and late germinating weeds, without any effect on the plants to be produced.
Preferably, Bialaphos or related compoud is ~~~~~~o applied at a dose ranging from about 0.4 to about 1.6 kg/ha, and diluted in a liquid carrier at a concentration such as to enable its application to the field at a rate ranging from about 2 to about 8 1/ha.
There follows examples, given by way of illustra-tion, of some embodiments of the process with different plant species.
- Sucrarbeets The North European sugarbeet is planted from March 15 up to April 15, depending upon the weather condition and more precisely on the precipitation and average temperature. the weeds problems are more or less the same in each country and can cause difficulties until the crop closes its canopy around mid-July.
Weed problems can be separated in three situa-Lions .
- early germination of the grassy weeds, - early germinating broadleaved weeds, - late germinating broadleaved weeds.
Up to now, pre-emergence herbicides have been succesfully used. Such compounds are for example those commercially available under the registered trademarks .
PYRAMIN-, GOLTI~ and VENZAit-. However, the susceptibility to dry weather conditions of these products as well as the lack of residual activity to control late germinating weeds have led~the farmer to use post-emergence products in addition to pre-emergence ones.
Table (I) thereafter indicates the active ingre dients contained in the herbicidal compositions cited in the following examples.

, TABLE (I) Commercial Name Active Ingredient Formulation AVADEX R Diallate EC 400 g/1 AVADEX BW R Triallate EC 400 g/1 GOLTIXR Metamitron WP 70 %

RONEET R Cycloate EC 718 g/1 TRAMAT R Ethofumerate EC 200 g/1 FERVINAL R Alloxydime-sodium SP 75 %

BASTA R Phosphinotricin 200 g/1 PYRAMIN FL R Chloridazon SC 430 g/1 According to the invention, post-emergence herbicides consist of Bialaphos or related compounds, which offer a good level of growth control of annual grasses ( ro , Ayena spp., Alopecurus, POA) and broadleaves ( aliu , Polvaonum, Senecio, Bolanum, Mercurialis).
Post-emergence herbicides can be applied at different moments of the growth of sugarbeet ; at a cotyledon level, two-leave level ox at a four-leave level.
Table (II) thereafter represents possible systems of field-treatment, given by way of example.
In those examples, the post-emergence herbicide of the class of Hialaphos used is BASTAR, in combination with different pre-emergence herbicides. Concentrations are indicated in 1/ha or kg/ha.

, 1~4147Q

m oc Y

x x oo ~c eo tW E ,~

C -: J
N

C ~ O p N

V V

~ ~ x ~

Lu r- 1 ~ 1 1 < 1 < v 1 1 a 1 -V ~ cr '~

D i N
_ O < < 1'1 N

C3 V. m m Z

D_ O

C

O

V

a y 2 a OC

~ r ,~ ...
C

E ~ E o0 oo N IC '- eC .Y
~ .Y

1. a H v- ,~~Ny-v \ \ \

4. IC cC fL oG n: L x ~C

O G < < < < ~ 00< < y .~ a J ~

(r 1 1 ~.- c., r. 1 N ,y c. -a r r .~

VJ O V: Cn N VJ C V! H
1 ~

H a < < < < a _ <-, r, 1.

E- ~ ~, ~ ~ m H ~

a v r m z=

,~O
\

m < .~-Y r a .
L t ~

H I
n L

C
O

J V: L

20 ""

N C

v '~ z ac < < <

L~ y 1 1 / t ~ .J
V

Vi 1 f r v y s '"' _ _ J I

a t, a N C
~' a ( a 5 ~
-_ . ac cc z oc a in j ~ I x x x x G ~c ~ ~

'" V ~ ' r. J
'-- CC 00 1 ~ ~ r 1 V 1 1 1 .Y ~ J1 L

' Y

r v G p ~
v V1 N N ,. W

~' ~ - .. v a z~ ~ -CC i L

? ~
c j 30 _ L = Y = Y _ _ _ ' ..1 Z r ~
~

- C X r x ~ L~ '~ ~ < -'~ r y ~ ,J
' , ~ Vf .~ !L ~~ C ~ .
V7 ~ .~ _ V7 1 ~ ~'~1= ~ Ls7 .~ 1 1 < ~ u1 -1 If1 N C' < < Z 2 < _ ~(1 .
N ~? ~T ~1 _ , <
O i ~ :~1 7 O ',~ tr W t'1 < < 2' LY (r l y N t"7 C t!f t'~ 1~ b C1 13414~~

- potatoes Potatoes are grown in Europe on about 8.106 Ha.
The major products used for weed control are Linuronjmonolinuron or the compound commercially available under the denomination METRABUZIN
These products perform well against most weedspecies.
However, weeds such as Galium and Solanum plus late germinating Chenopodium and ~olvcronum are not always effectively controlled, while control of the annual grasses is also sometime erratic.
Once again, late germinating broadleaved weeds are only controllable by post-emergence applications of herbicides such as BASTA R.
Table (III) thereafter represents some examples given by way of example of field-treatment in the case of potatoes.

TABLE (III) Weeds control systems in potatoes, based on the use of BASTAR, providing potatoes are rendered resistant to BASTAR.
Linuron + monolinuron (375 g + 375 g/ha)~prior to emergen-ce HASTAR 3-4 lt/ha after emergence (5-15 cm) BASTAR/fluazifop-butyl 3-4 lt/ha + 2 lt/ha after emergence (5-15 cm) Linuron WP 50 ~ (AFALONR) Monolinuron WP 47.5 ~ (ARESSINR) fluazifop-butyl EL 250 g/1 (FUSILADER) The strains pGSJ260 and p8G39 used hereabove have , been deposited on December l2nd, 1985, at the "German Collection of Micro-organisms" (DEUTSCHE SAMMLUNG VON
MIKR00RGANISMEN) at Gottingen, Germany. They received the deposition numbers DSM 3 606 and DSM 3 607 respectively.

Further embodiments of the invention are described hereafter with reference to the figures in which .
- fig. 8 shows the restriction map of a plasmid pJS1 containing another Bialaphos-resistance-gene ;
- fig. 9 shows the nucleotide sequence of the "sfrsv" gene containing the resistance gene ;
- fig. 10 shows the amino acid homology of "sfrsv"
gene and "sfr" gene, - fig. 11 shows the construction of a plasmid, given by way of example, which contains the "sfrsv" gene and suitable for the transformation of plant cells.
Another Bialaphos-resistance-gene has been isolated form another Bialaphos-producing-strains, i.e.
streptomvces viridochromouenes. This second resistance-gene is thereafter designated by "sfrsv" gene.
This second preferred DNA fragment according to the invention, for the subsequent transformation of glant cells, consists of a nucleotide sequence coding for at least part of a polypeptide having the following sequen-ce .

i~

V S P S R R P V E I R P A T A A D M
A A V C D I V N H Y I E T S T V N F
R T E P Q T P R, E W I D D L E R. L Q
1p D R Y P W L V A E V E G V V A G I A
Y A G P W K A R N A Y D W T V E S T
15 V Y V S H R H ø R L G L G S T L Y T
H L L K S M E A Q G F K S V V A V I
2p C L P N D P S V R L H E A L G Y T A

25 ' T Q I

which part of said length polypeptide is of to sufficient confer protection rotecting-against Bialaphos-"plant-p capability"-, to plant cells, when in corporated genetically and expressed therein. Reference ll also wi be made here-after to the "plant-protecting-capability" against Bialaphos of the abovesaid nucleotide sequence.

Meaning of the designation of amino acids by a single letter given therafter.
is Alanine A Leucine L

Arginine R Lysine K

Asparaqine N Methionine M

Aspartic Acid D Phenylalanine F

CYsteine C Proline P

Cystine C Serine S

Glycine G Threonine T

Glutamic Acid E Tryptophan W

Glutamine Q Tyrosine Y

Histidine H Valine V

Isoleucine I

,.

This second preferred DNA fragment consists of the following nucleotide sequence .
TAAAGAGGTGCCCGCCACCCGCTTTCGGAGAACACCGAAGGAGACCACAC
=GdT~G_AGCCCAGAACGACGCCCGGTCGAGATCCGTCCCGCCACCGCCGCCGA
CATGGCGGCGGTCTGCGACATCGTCAATCACTACATCGAGACGAGCACGG
TCAACTTCCGTACGGAGCCGCAGACTCCGCAGGAGTGGATCGACGACCTG
GAGCGCCTCCAGGACCGCTACCCCTGGCTCGTCGCCGAGGTGGAGGGCGT
CGTCGCCGGCATCGCCTACGCCGGCCCCTGGAAGGCCCGCAACGCCTACG
ACTGGACCGTCGAGTCGACGGTGTACGTCTCCCACCGGCACCAGCGGCTC
GGACTGGGCTCCACCCTCTACACCCACCTGCTGAAGTCCATGGAGGCCCA
GGGCTTCAAGAGCGTGGTCGCCGTCATCGGACTGCCCAACGACCCGAGCG
TGCGCCTGCACGAGGCGCTCGGATACACCGCGCGCGGGACGCTGCGGGCA
GCCGGCTACAAGCACGGGGGCTGGCACGACGTGGGGTTCTGGCAGCGCGA
CTTGGAGCTGCCGGCCCCGCCCCGGCCCGTCCGGCCCGTCACACAGAT
GAGCGGAGAGCGCATGGC

or of a part thereof expressing a polypeptide having plant-protecting capability against Bialaphos ;
There follows hereafter the description of experiments carried out for the isolation of the "sfrsv"
resistance gene, the construction of expression vectors which contain the resistance gene and which allow the subsequent transformation of plant cells, in order to render them resistant to GS inhibitors.
Clonincr of the bialaphos-resistance-"sfrsv" uene from Streptomvces viridochromoQenes The strain yStrentomvces viridochromocrenes DSM

, 40736 (ref 1) was grown and total DNA of this strain was prepared according to standard techniques. DNA samples were digested respectively with _~tI, ~I and Sa_u_3AI in three different reactions and separated on an agarose gel, together with plasmid DNA from pGSR1 (fig. 5B) digested with BamHI. In a Southern analysis the DNA was blotted on a nitrocellulose filter and hybridized with the labbeled BamHI fragment from pGSR1 containing the "sfr" gene. In all four lanes of the gel, a restriction fragment was showing strong homology with the probe . a stI fragment of about 3 kb, a SmaI fragment of about 1.2 kb and Sau3AI
fragment of 0.5 kb. In order to clone this gene, ~s I
restriction fragments were directly cloned in the Escherichia coli vector pUCB. 3000 colonies obtained after transformation were transferred to nitrocellulose filters, and hybridized with the "sfr" probe. Positive candidates were further tested for their growth on minimal medium plates containing 300 Ng/ml PPT. One transformant that grew on PPT-containing-medium was further analysed. The plasmid map and relevant restriction sites of this plasmid pJS1 are represented in fig. 8. The strain MC1061 (pJS1) has been deposited on March 06, 1987 at the DEUTSCHE
SAMMLUNG VON MIKROORGANISMEN (DSM) under deposition number DSM 4023. The clone restriction fragment has been sequenced according to the Maxam and Gilbert method and the coding region of the gene could be identified through homology. The sequence of the "sfrsv" gene is represented in fig.9 and the homology on the nucleotide and amino acid sequence level with "sfr" gene is shown in fig. 10.
Expression of the "sfrsv" aene A "sfrsv gene cassette" was also constructed to allow subsequent cloning in plant expression vectors. A
HanII-III fragment containing the "sfrsv" coding region without the initiation codon GTG was isolated from pJSl.

~34Z47~

This fragment was ligated in the vector pLK56-2 digested with NcoI and Bg~II, together with a synthetic oligonucleotide 5'-CATGAGCC-3', similar with the one described for "sfr" gene and shown in fig. 5. The construction of pGSRISV is schematically shown in fig. 11.
Since similar cassettes of bath genes are present in respectively pGSR1 and pGSR1SV, previous described constructions for the expression of the "sfr" gene in plants can be repeated.
Enzymatic analysis of crude extracts from E. coli strains carrying plasmid pGSRISV demonstrated the synthesis of an acetylase which could acetylate PPT. This was shown by thin layer chromotography of the reaction Porducts.
The "sfrsv" gene was then inserted into the plasmid vector pGSJ260 (fig. 4B) under the control of the CaMV 35s promoter, to yield a plasmid pGS2SV, similar to pGSR2 (fig. 6A) except that the "sfrsv" gene is substitu ted for the "sfr" gene.
It is clear that herbicide resistance genes of the above type may be obtained from many other microorganisms that produce PPT or PPT derivatives. Herbicide resistance gene can then be incorporated in plant cells with a view of protecting them against the action of such Glutamine Synthetase inhibitors. For instance, a Hialaphos-resistance-gene is obtained from Kitasotosporia (ref. 15).
Transformed plant cells and plants which contain the "sfrsv" resistance gene can be obtained with plasmid PGSR2SV, using the same Agrobacterium-mediated-trans formation system as hereabove described for the transfor-mation of different plant species with the "sfr" gene.

i~
134147~

Plants are regenerated and tested for their resistance with similar spraying tests as described hereabove. All plants behaved similarly and show resist-s ance against herbicides consisting of glutamine synthetase inhibitors.
Finally, the inventors also pertains to the combination of the plants resistant to an inhibitor of glutamine synthetase as defined above with the corres-ponding inhibitor of glutamine synthetase for use in the production of cultures of said plants free form weeds.

58 1 3 4 1 ~ i 0 REFERENCES
1. BAYER et al., HELVETICA CHEMICA ACTA, 1972 2. WAKABAYASHI K. and MATSUNAKA S., Proc. 1982, British Crop Protection Conference, 439-450 3. M. MASON et al., PHYTOCHEMISTRY, 1982, vol. 21, n' 4, p. 855-857.
4. C. J. THOMPSON et al., NATURE, July 31, 1980, vol. 286, n' S 772, p. 525-527 5. C. J. THOMPSON et al,, JOURNAL OF BACTERIOLOGY, August 1982, p. 678-685 6. C. J. THOMPSON et al., GENE 20, 1982, p. 51-62 7. C. J. THOMPSON et al., MOL. GEN. GENET., 1984, 195, p. 39-43 8. TOWBIN et al., PROC. NATL. ACAD. SCI. USA, 1979, ~, p. 4 350-4 354 9. METHODS OF ENZYMOLOGY, V.XLIII, p. 737-755 10. DEBLAERE H. et al., 1985, Nucl. Acid. Res., 13, 1 477 11. BOTTERMAN J., February 1986, Ph. D. Thesis, State University of Ghent 12. DEBLAERE R., february 1986, Ph. D Thesis, Free University of Brussel, Belgium 13. VELTEN et al, EMBO J. 1984, vol. 3, n'12, p. 2 723-14. CHATER et al, Gene cloning in Streptomyces. Curr. Top.
Microbiol. Immunol., 1982, 96, p. 69-75 15. OMURA et al, J. of Antibiotics, Vol. 37, 8, 939-940, 16. MURAKAMI et al, Mol. Gen. Genet., 205, 42-50, 1986 17. MANDERSCHEID and WILD, J. Plant Phys., 123, 135-142,

Claims (8)

1. A plant cell having a recombinant DNA stably integrated into its genome; said recombinant DNA comprising a heterologous DNA encoding a protein having an acetyl transferase activity with respect to a glutamine synthetase inhibitor, under the control of a promoter recognized by the polymerases of said plant cell, with the proviso that said protein does not have phosphinothricin acetyltransferase activity.

2. The plant cell according to claim 1, wherein said heterologous DNA is obtainable from the genome of a Streptomyces species.

3. The plant cell according to claim 1, in which said promoter is selected from the group consisting of a 35S
promoter of Cauliflower Mosaic Virus, a TR1' promoter, a TR2' promoter and a promoter of the gene encoding the Rubisco small subunit.

4. The plant cell according to claim 1, in which said recombinant DNA further comprises a second DNA encoding a chloroplast transit peptide between said promoter and said heterologous DNA.

5. The plant cell according to claim 4, wherein said second DNA encodes a transit peptide of the precursors of ribulose 1,5 biphosphate carboxylase or chlorophyl a/b binding protein.

6. The plant cell according to claim 1, in which said recombinant DNA further comprises a 3' untranslated end including a polyadenylation signal.

7. The plant cell according to claim 6, in which said untranslated end is from a T-DNA gene of an Agrobacterium tumefaciens.

8. A plant cell culture, which consists of the cells of any one of claims 1 to 7.