WO1992019719A1

WO1992019719A1 - Glyphosate degrading bacteria

Info

Publication number: WO1992019719A1
Application number: PCT/GB1992/000773
Authority: WO
Inventors: Rosemary Elaine Dick; John Quinn; Sarah Bronwen Rees; Wolfgang Walter Schuch
Original assignee: Zeneca Limited
Priority date: 1991-05-03
Filing date: 1992-04-27
Publication date: 1992-11-12
Also published as: AU1650092A

Abstract

Novel bacterial strains, SC9 (Alcaligenes xylosoxidans subspecies denitrificans) and SC11 (a strain of an uncommon Pseudomonas species) have been isolated. These strains can degrade the herbicide glyphosate, and may be used to produce glyphosate-degrading proteins. The bacteria are sources of genes coding for glyphosate-degrading proteins which may be used in the production of herbicide-resistant plants.

Description

G YPHOSATE DEGRADING BACTERIA

This invention relates to glyphosate-degrading bacteria and their use. More specifically, the invention relates to a novel strain of Alcaligenes xylosoxidans subspecies denitrificans and a novel strain of a Pseudomonas species.

Glyphosate (N-phosphonomethylglycine) is the active component of a herbicide sold under the trade name 'Round-up' by the Monsanto Corporation. Glyphosate acts on a highly specific target site: it is a potent inhibitor of 5-enolpyruvylshikimate 3-phosphate synthase (EPSP synthase; EC 2.5.1.19). It is readily translocated within plants and so has good systemic action, it shows limited persistence in the soil, and it has an excellent toxicological profile. In addition, it has proved difficult to synthesise chemical analogues which have the same specificity and efficacy as glyphosate. These factors have contributed greatly to the success and value of glyphosate as a broad spectrum herbicide.

As many crop plants are susceptible to glyphosate, the use of the herbicide is restricted to appropriate application times and suitable species. The development of glyphosate-tolerant crops would allow extended use of this benign, broad spectrum herbicide to replace the use of less advantageous chemicals.

In certain circumstances, such as at production or spill sites, excessive glyphosate may accumulate in the environment. Development of

SUBSTITUTESHEET a method to remove this unwanted herbicide more rapidly would be beneficial.

It has been clearly demonstrated that metabolism of glyphosate by soil microflora is the predominant route by which the herbicide is degraded in the environment (Rueppel et al, 1977, J Agric Food Che , 25, 517-528; Sprankle et al, 1975, Weed Sci, 23, 229-234; Torstensson and Aa isepp, 1977, Weed Res, 17, 209-212). The dominant metabolic process appears to be by co-metabolism; in other words the micro-organisms do not use glyphosate as a primary growth substrate (see also Nomura and Hilton, 1977, Weed

Res, 17, 113-121). Soil metabolism studies using glyphosate separately labelled with 14C in each of its carbons showed that, under a range of conditions, glyphosate degraded rapidly and the sole major metabolite was aminomethylphosphonic acid or AMPA (Rueppel et al, 1977, J Agric Food Chem, 25, 517-528). This compound subsequently underwent rapid degradation to yield 14C-labelled co₂.

Glyphosate has been shown to have no drastic effect on the populations of soil micro-organisms when used at the recommended rates. Although there are many potential pathways for the metabolism of glyphosate, only two pathways have been demonstrated unambiguously. These are (1) the sarcosine pathway, involving the cleavage of the C-P bond and (2) the AMPA pathway involving the cleavage of the C-N bond of the glycyl moiety of glyphosate (Malik et al, 1989, BioFactors, 2, 17-25).

Examples of bacteria utilising the sarcosine

SUBSTITUTE SHEET pathway for glyphosate metabolism include Pseudomonas sp. PG2982 (Moore et al, 1983, Appl Environ Microbiol, 46, 316-320). Subsequent studies have demonstrated that this organism can use a wide range of phosphonates as growth substrates (Shinabarger et al, 1984, Appl Environ Microbiol, 48, 1049-1050; Kishore and Jacob, 1987, J Biol Chem, 262, 12164-12168). Glyphosate degradation in Pseudomonas PG2982 is apparently analogous to the bacterial metabolism of alkyl and arylphosphonates involving the C-P lyase enzyme to yield first sarcosine and then glycine with the phosphonomethyl carbon released to Cl metabolism (Malik et al, 1989, BioFactors, 2, 17-25).

More recently Lerbs et al (1990, Archives Microbiol, 153(2), 146-150) reported the different control processes that Alcaligenes sp. strain GL exhibits in the metabolism of glyphosate via the sarcosine pathway under different environmental conditions. Uptake and dephosphonation of glyphosate are regulated by phosphate starvation, the intensity of glyphosate degradation is controlled by the cellular ability to utilise the C-skeleton derived from glyphosate.

Another well-studied strain capable of using glyphosate as a sole phosphate source is Arthrobacter sp. GLP-1 and a recently isolated derivative of this strain has been shown to be capable of utilising glyphosate as a nitrogen source also (Pipke and Amrhein, 1988, Appl Environ Microbiol, 54, 2868-2870).

Glyphosate is metabolised to AMPA in Arthrobacter atrocyaneus (Pipke and Amrhein, 1988,

SUBSTITUTE SHEET Appl Environ Microbiol, 54, 1293-1296) and Pseudomonas sp. Br (Jacob et al, 1988, Appl Environ Microbiol, 54, 2953-2958). Very little is known about the enzymology and nature of the genes involved in the conversion of glyphosate to AMPA (Malik et al, 1989, BioFactors, 2, 17-25).

Resistance or tolerance to glyphosate occurs naturally in some plants. There have been various attempts to improve or extend this trait using molecular techniques. Three routes used to generate genetically-engineered plants resistant to glyphosate are detailed below.

An altered EPSP synthase enzyme has been isolated from selected bacterial strains, such as Salmonella typhimurium. The mutant form of EPSP synthase showed decreased affinity for glyphosate but its kinetic properties were maintained. The gene for the bacterial enzyme was transferred to tobacco and expressed within the plant. These transgenic plants were tolerant to increased levels of the herbicide both in greenhouse and field experiments.

Mutant forms of plant EPSP synthase have been produced through mutagenesis and selection in bacteria. This has led to further improvement in the levels of glyphosate tolerance.

EPSP synthase has also been isolated from plants which are naturally resistant to increased levels of glyphosate. It was found that this enzyme was normal, but its over expression led to glyphosate tolerance. Thus transgenic plants produced to give high levels of EPSP synthase expression should show increased glyphosate tolerance.

SUBSTITUTESHEET Although glyphosate-resistant plants have been produced, in several instances the level of resistance generated to date has not been sufficient for field rate applications of the herbicide.

In the above examples, resistance is achieved by manipulating the target site of the herbicide, either by alteration of the target enzyme or by its over-production to reduce the overall effect of inhibition. Herbicide resistance can also be achieved through detoxification by degrading or modifying the herbicide before it acts on its target site. This latter mechanism has been used to generate plants resistant to two other herbicides: bromoxynil and phosphinothricin (Basta). The identification and isolation of genes coding for proteins which can be used to detoxify the herbicide have played a critical part in these experiments. This was achieved through the isolation of bacterial strains containing enzymes capable of detoxifying the herbicide: Klebsiella sp. strains containing a gene encoding a nitrilase enzyme which degrades bromoxynil, and Streptomyces strains containing the bar gene encoding an enzyme which modifies phosphinothricin by acetylation. These genes can be used to generate transgenic plants expressing the detoxifying enzymes. Generation of transgenic plants capable of degrading or modifying a herbicide is thus a feasible and effective approach to obtaining herbicide resistance. This mechanism has not so far been applied to glyphosate resistance.

SUBSTITUTE SHEET We have now isolated novel bacterial strains capable of degrading glyphosate by various catabolic pathways including conversion to aminomethylphosphonic acid (AMPA) . An object of the present invention is to provide a source of genes which encode enzymes capable of degrading glyphosate. These genes may be inserted into plants to render them tolerant to the herbicide. A further object is to provide a useful agent for removing unwanted glyphosate from the environment, such as production or spill sites.

According to the present invention, we provide a glyphosate-degrading agent comprising a novel microorganism which can metabolise glyphosate via the AMPA pathway. A novel glyphosate-degrading strain, SC9, of Alcaligenes xylosoxidans subspecies denitrificans and a novel glyphosate-degrading strain, SCll, of a Pseudomonas species are provided. Strains SC9 and SCll were deposited under the terms of the Budapest Treaty at The National Collections of Industrial and Marine Bacteria- (NCIMB), 23 Machar Drive, Aberdeen, AB2 IRY, Scotland, UK, on 10 May 1991 under the accession numbers NCIMB 40418 and NCIMB 40419 respectively.

In further aspects, the invention comprises glyphosate-degrading proteins capable of being isolated from strains SC9 and SCll. The invention also comprises DNA sequences coding for proteins according to the invention. The DNA may be cloned or transformed into a biological system allowing expression of the encoded protein.

SUBSTITUTE SHEET The invention also comprises plants transformed with recombinant DNA encoding a glyphosate-degrading protein according to the invention. The invention also comprises a process of degrading glyphosate whereby the herbicide is exposed to the strains or proteins according to the invention.

The bacterial strains of this invention were isolated from soil. The strains can degrade glyphosate in the course of aerobic growth on minimal medium.

Formal identification and characterisation of strains SC9 and SCll has been undertaken by the National Collection of Industrial and Marine

Bacteria (NCIMB), Aberdeen, UK. NCIMB conclude that SC9 is a strain of Alcaligenes xylosoxidans subspecies denitrificans. NCIMB conclude that SCll is a strain of a Pseudomonas species but are unable to match it exactly with any of the species described in Bergey's Manual.

The bacterial strains may be used to degrade glyphosate in the environment.-

The strains may be further characterised by biochemical methods to indicate the enzyme system responsible for the metabolism of glyphosate.

The glyphosate-degrading proteins (enzymes) may be isolated: from their primary structure, the sequence or partial sequence of the DNA encoding them may be determined. This DNA may be manufactured using a standard nucleic acid synthesiser, or suitable probes (derived from the known sequence) may be used to isolate the actual gene(s) and control sequences from the bacterial

SUBSTITUTESHEET genome. This genetic material may then be cloned into a biological system which allows expression of the proteins.

The DNA may be inserted into a suitable micro-organism. The DNA may also be transformed by known methods into any plant species, so that the glyphosate-degrading proteins are expressed within the plant. Suitable plant species include tobacco, maize and sugarbeet. Transformed plants should show increased tolerance to glyphosate. The invention may be further understood by reference to the drawings, in which:

Figure 1 shows a soil percolation column for establishment of glyphosate-degrading cultures; Figure 2 is a series of time-course graphs showing glyphosate degradation by strains SC9 and SCll.

To illustrate the invention, there now follows a description of the bacterial strains, their isolation and screening for glyphosate-degrading activity.

SUBSTITUTESHEET CHARACTERISATION OF THE MICRO-ORGANISMS The morphological and biochemical characteristics of strains SC9 and SCll have been determined by the National Collection of Industrial and Marine Bacteria (NCIMB), Aberdeen, UK, and and their findings are summarised in Tables 1-4 below.

NCIMB identified SC9 as a strain of Alcaligenes xylosoxidans subspecies denitrificans. NCIMB identified SCll as a strain of a

Pseudomonas species but are unable to match it exactly with any of the species described in Bergey's Manual. The API profile suggests that strain SCll may be a Pseudomonas chlororaphis, but the following reactions were atypical of this species: gelatin hydrolysis; levan production; lecithinase; Tween 80 hydrolysis; denitrification; arginine dihydrolase; gluconate utilisation. Also, production of chlororaphin was not observed in either glycerol or King's A medium. In addition, the isolate does not appear to belong to the fluorescent Pseudomonas group (P_ cepacia, P_ stutzeri, P acidovorans or P alcaligenes, J? pseudoalcaligenes) .

SUBSTITUTESHEET TABLE 1 Morphology (growth on Lab M nutrient agar)

Strain SC9 SCll

°C incubation 30°C 30°C Gram stain - (KOH test +) - (KOH test +) Spores Motility + + Colony buff; buff; morphology semi-translucent; semi-translucent; round; regular; round; regular; entire; shiny; entire; shiny; smooth; smooth; low convex. low convex,

3 days approx. 1.0 mm approx. 0.5 mm diameter diameter. Accumulation of storage product with age.

+ (+)

SUBSTITUTE SHEET TABLE 2 Rapid Test (API) (48 hours, 30°C)

SC9 SCll

Nitrate reduction +

Indole production

Acid from glucose

Arginine dehydrolase - Urease

Aesculin hydrolysis

Gelatin hydrolysis β-Galactosidase

Glucose assimilation - + Arabinose assimilation - +

Mannose assimilation - +

Mannitol assimilation - +

N-acetylglucosamine assimilation +

Maltose assimilation Gluconate assimilation +

Caprate assimilation -

Adipate assimilation +

Malate assimilation + +

Citrate assimilation + + Phenylacetate assimilation +

Cytochrome oxidase +

SUBSTITUTE SHEET TABLE 3 Additional Testing of SC9 (30°C/8 days)

Acid from xylose - Utilisation of xylose

Utilisation of fructose -

Hydrolysis of gelatin -

Alkalisation of allantoin -

Alkalisation of tartrate Nitrate to nitrogen +

TABLE 4 Additional Testing of SCll (30°C/8 days)

Kings A Medium (pigment production) -

Kings B medium (pigment production) -

Acid from sucrose - Acid from maltose Acid from glycerol -

Growth in 5% NaCl +

Utilisation of trehalose (+)

Utilisation of inositol (+)

Utilisation of benzylamine - Production of arginine dihydrolaεe -

Hydrolysis of gelatin -

Hydrolysis of Tween 80 -

Hydrolysis of starch -

Egg yolk agar capacity - Levan from sucrose - Nitrate to nitrite Nitrate to nitrogen

Residual nitrate + Experiment 1

ESTABLISHMENT OF GLYPHOSATE-DEGRADING CULTURES IN SOIL PERCOLATION COLUMNS. Method Duplicate percolations (shown in Figure 1) were set up with inocula from each of glyphosate treated or untreated soil. Each column was filled with washed gravel of mean diameter 3mm. Both glassware and gravel were sterilised before use to exclude laboratory contaminants at the outset. Fresh soil was collected from a garden in Ballyronan, N.Ireland. This soil type is a very sandy loam, pH approximately 5.9. One gram of soil was vortex mixed with 10ml 0.08% NaCl for 2 minutes. 2ml of the resulting suspension was added as an inoculum. Glyphosate was added to a final concentration of 2.5mM. The total volume in each column was 20ml.

The columns were incubated at ambient temperature, in darkness to prevent algal growth, for a total of 280 days. The volume was readjusted weekly to 20ml by the addition of sterile distilled water. Further aliquots of sterile glyphosate solution were added as appropriate to achieve a gradual increase in concentration. At intervals, 0.5ml aliquots were removed and examined for change in pH, chemical and microbial composition.

The disappearance of glyphosate and/or the presence of intermediates of degradation was investigated by thin layer chromatography. Microbial numbers and diversity were estimated using half-strength nutrient agar and soil extract agar by the "Most Probable Number" method of

TESHEET^" Harris and Sommers ^'(1968, Appl Microbiol, 16, 330-334). Phosphate levels were also determined as described in Experiment 4.

At the end of the 280 day incubation period, the volume of each column was adjusted to 20ml, the liquor was drawn off, tested for glyphosate and phosphate composition. The remaining liquor was stored at 4°C. Results During the 280 days of incubation of the soil columns, the pH rose from an initial value of 6.2 to 6.5. Microbial numbers showed an initial rise, then decreased to the original level after 21 days. Subsequently, numbers appeared to increase in response to the addition of higher concentrations of glyphosate.

Up to approximately 100 days after the initiation of the soil columns, reductions in the added concentration of glyphosate could be qualitatively observed by thin layer chromatographic analysis. At no time were putative intermediates of glyphosate degradation observed. On termination of the experiment, the glyphosate concentration in each column was estimated to be in excess of 50mM. The phosphate level was estimated to be in excess of lOmM in both columns. Less than O.lmM phosphate had been detected in the columns at the outset of the experiment. Up to 10-fold greater numbers of colony forming units were observed on soil extract agar along with much greater diversity compared with half strength nutrient agar. Microbial diversity in the columns remained high with 200 colony types

SUBSTITUTESHEET per ml still apparent after 100 days of incubation. After this time, the diversity decreased until fewer than 50 colony types per ml could be distinguished at the termination of the experiment. At this stage some 10 colony types were dominant in each column.

Experiment 2 ISOLATION OF GLYPHOSATE DEGRADING MICRO-ORGANISMS FROM SOIL COLUMNS. Method

For both the columns a 1% inoculum of column liquor was inoculated in 60ml medium in sterile 250ml Erlenmeyer flasks. The medium comprised:

0.8% NaCl solution containing ImM phosphate buffer pH 7.0, 20mM iminodiacetate (IDA) and lOmM glyphosate as carbon and nitrogen sources. 3ml was removed from each culture and C-glyphosate was added to a final concentration of 1 Ci ml^" in sterile 30ml universal containers. Control cultures were set up without glyphosate.

Both sets of cultures were incubated at 29°C for 12 days. At intervals, after the addition of sterile distilled water to replace loss through evaporation, samples were taken for analysis.

Non-radioactive cultures were examined for microbial numbers and diversity using half-strength soil extract agar (see Experiment 5) and lOmM glyphosate by the method of Harris and

Sommers (1968, Appl Microbiol, 16, 330-334). Degradation and/or disappearance of 14C-glypl in the radioactive cultures was monitored by scintillation counting and by TLC/autoradiography (see Experiment 6). Results Microbial numbers in the populations (subcultured from the soil columns) growing on IDA as sole nitrogen source from the treated column grew better in the presence of glyphosate than controls. Neither population released phosphate from glyphosate during the 12-day duration of the experiment.

Sub-cultured populations from both the treated and untreated columns were plated out on to equivalent solid medium from which micro-organisms SC9 and SCll were isolated from the untreated and treated soil columns respectively.

SUBSTITUTE SHEET Experiment 3

TEST TO DETERMINE THE ABILITY OF SC9 AND SCll TO METABOLISE GLYPHOSATE AND THE IDENTITY OF ANY DEGRADATION PRODUCTS.

a) Test for the degradation of 14C-glyphosate in soil extract broth.

SC9 and SCll were grown on soil extract agar (see Experiment 5) for 5 days after which a lO l loop of culture was removed from each plate and resuspended in soil extract broth containing l Ci ml^" 3- 14C-glyphosate and "cold" glyphosate to a final concentration of 100 M. lOO l samples were taken at intervals, centrifuged and the supernatants stored at -20°C. Analysis was by tlc/autoradiography and scintillation counting (see Experiment 6).

The cultures were incubated for 17 days. TLC/autoradiograms of the culture supernatants at day 17 showed release of glyphosate degradation products by SC9 and SCll. These were not present in the uninoculated control. Analysis of samples taken over the 17-day period, confirm the production of the intermediates. Their identification as AMPA and sarcosine was confirmed by co-chromatography with authentic cold sarcosine and AMPA.

b) Test for the degradation of 14C-glyphosate in half strength soil extract broth. The above experiment was repeated

tT using half-strength soil extract broth (see Experiment 5) containing 0.6gl^~ glucose. The results confirmed the ability of SC9 and SCll to degrade glyphosate with AMPA as an intermediate, although sarcosine was not produced on this occasion.

Test to investigate the ability of SC9 and SCll to degrade 14C-glyphosate in Arthrobacter culture medium.

SC9 and SCll cultures were maintained on agar slants of half-strength soil extract medium.

5ml of soil extract broth was added to each slant and incubated for a further 7 days. On day 7 the liquid phase from 4 similar slants

•was combined, centrifuged and the resultant pellets were suspended in 2.5ml 0.85% saline solution. 300 1 aliquots of saline solution were used to inoculate 2.5ml volumes of a range of Arthrobacter culture media (see

Experiment 7): (1) neat, (2) with a 50% soil extract broth supplement and (3) with phosphate free soil extract broth supplement

(phosphate was extracted from the soil extract using 10% w/v magnesium oxide continuously stirred for 4 hours at ambient temperature). All samples were incubated with

¹ C-glyphosate (100_//M, 1/c/Ci/ml) in 20ml tubes, at 25°C, shaking at 80rpm. 50/t/l samples were taken at days 0, 5, 10 and 15. The samples were centrifuged and 30^1 aliquots of supernatant were subjected to TLC analysis

(see Experiment 6). Glyphosate degradation was demonstrated by both strains in all three media types. Production of AMPA and sarcosine was observed, as shown in Figure 2.

Test for the presence of C-P lyase.

SC9 and SCll were subcultured from half-strength nutrient agar and tested for their ability to use glyphosate and/or AMPA as sole source of phosphorus. 50 μl of bacterial suspension was used to inoculate 5ml of phosphate-free minimal medium, M3-P (see Experiment 8). The cultures were incubated at 29°C for 5 days to starve the cells of any phosphate reserves. This process was repeated by sub-culture to ensure phosphate depletion. 100 1 aliquots were then taken to inoculate 10ml of each of the following media (a-d) and incubated for 7 days at 29°C.

a. M3-P b. M3-P + ImM phosphate c. M3-P + ImM glyphosate as sole phosphate source d. M3-P + ImM AMPA as sole phosphate source

The organisms were also tested for their ability to use 5mM methylphosphonate as sole source of phosphorus. The supernatants of the cultures grown in the presence of glyphosate, AMPA and methylphosphonate were tested for phosphate release after 7 days incubation, by which time growth was complete even in the more slow growing cultures.

SUBSTITUTE SHEET The results are described in Table 5, and show that strains SC9 and SCll can use glyphosate and AMPA as sole phosphorous source.

TABLE 5 Phosphonate utilisation under conditions of phosphorus limitation

KEY: m minimal growth (OD <<1.0) due to carry-over of nutrients + growth (OD <1.0) ++ good growth (OD >1.0 )

Experiment 4

DETECTION OF PHOSPHATE AND TOTAL PHOSPHORUS. Phosphate was detected by the method of Fiske and SubbaRow (1925, J Biol Chem, 66, 375-381). The solution under test was added to a phosphorus-free test tube containing 1.0ml acid molybdate solution (ammonium molybdate.4H₂0 1.25g/100ml in 2.5N H₂S0₄) and 0.25ml Fiske and SubbaRow reducer solution (l-amino-2-naphthol- 4-sulphonic acid 0.8%, sodium sulphite and sodium bisulphite, commercial preparations from Sigma). Double-distilled water was added if necessary. The volume of solution tested and the volume of

SUBSTITUTESHEET water added was varied according to the expected phosphate content of the sample. The assay was left for one hour to allow colour to develop before reading the absorbance at 600nm. A set of standards was prepared for each assay.

Glyphosate and other phosphonates were detected as total phosphorus by a modification of the method of Ames (1966, Methods Enzymol, 8, 115-118). 100/t/l of a solution of 10% Mg(N0₃)₂ in 95% ethanol in a phosphorus-free pyrex boiling tube was heated over a strong bunsen flame to dryness and the organic matter ashed until the evolution of brown fumes had ceased. 1.0ml of 0.5N HCl was added to the residue and the tube was then covered with parafilm and placed in a boiling waterbath for 15 minutes. The resulting solution was tested for phosphorus as previously described.

Experiment 5 SOIL EXTRACT MEDIUM.

Soil extract medium was prepared by the method of James (1958, Can J Microbiol, 4, 336), using fresh Ballyronan garden soil. It was filter-sterilised before use to ensure removal of heat-resistant spores.

Glyphosate was added to a final concentration of lOmM.

Half strength soil extract was also used and supplemented with 0.6mg 1^~ glucose plus glyphosate to a final concentration of 10mM.

For solid medium 1.5% (w/v) Bactoagar (Difco) was used.

SUBSTITUTE SHEET Experiment 6 CHROMATOGRAPHY.

The following solvent systems were evaluated for the separation of glyphosate and its metabolites by thin layer chromatography.

(1) EtOH:H₂0:17N NH₄OH:TCA:15N acetic acid, 55:35:2.5:3.5:2, vol/vol/vol/wt/vol (Shinabarger and Braymer, 1986, J Bacteriol, 168, 702-707); (2) n-butanol:acetic acid:water, 60:20:20

(Pavoni, 1978, Report of Bologna Chemical Laboratory, 157-161).

Compounds containing primary and secondary amine groups were visualised by spraying plates with 0.2% ninhydrin in ethanol and heating to 110°C on a hotplate. Phosphorus-containing compounds were detected by the method of Maile et al (1977, J Chromatog, 132, 366-368). For the detection of radio-labelled compounds, tic plates were either sprayed with Amplify spray (Amersham International, pic) and stored with X-ray film at -70°C for up to 3 weeks or were analysed by a tic plate scanner.

Solvent system (1) was found to give the best separation of glyphosate and its metabolites, but the presence of ammonium resulted in a dark background after development with ninhydrin, reducing sensitivity of visualisation. Consequently, this system was mainly used for the separation of radiolabelled compounds. Solvent system (2) was used for the separation of compounds from non-radioactive cultures.

SUBSTITUTESHEET

pH 7.2

Experiment 8 PHOSPHOROUS-FREE MINERAL SALTS BASE (M3-P).

In order to study bacterial cultures for their ability to support growth on glyphosate as phosphate source, bacterial cultures were grown on phosphorus-free medium.

To 1 litre water: Trizma base (Sigma) 6.0g NH₄C1 5.0g

MgSO ₄.7H₂0 0.16g

CaCl₂-2H₂0 0.08g

Trace elements solution 1.0ml (see below) FeEDTA solution 1.0ml

(FeS0₄.7H₂0 EDTA disodiu salt l.Og l -1) pH 7.0

SUBSTITUTESHEET The solution is heat sterilised at 121°C for 15 minutes. On cooling to 55°C the following supplements were added as filter sterilised solutions (per litre): 5.0g potassium gluconate 5.0g sodium pyruvate 1.0ml vitamin solution.

Where glucose or glycerol replaced gluconate and pyruvate as carbon sources, they were autoclaved separately as concentrated solutions. Filter-sterilised solutions of phosphate, glyphosate or other phosphonates are added as phosphorus sources. Phosphate is added as freshly-prepared buffered solution, pH 7.0, filter-sterilised and diluted accordingly. Glyphosate and other phosphonates were prepared as concentrated solutions, adjusted to pH 7.0 by the addition of dilute NaOH, and filter-sterilised.

SUBSTITUTE SHEET

Claims

1. A glyphosate-degrading agent comprising a micoorganism of the species Alcaligenes that metabolises glyphosate via the AMPA pathway.

2. A glyphosate-degrading agent as claimed in claim 1 where the microorganism is Alcaligenes xylosoxidans subspecies denitrificans, strain SC9, a culture of which was deposited on 10 May 1991 under the teims of the Budapest Treaty with the National Collection of Industrial and Marine Bacteria Limited, Aberdeeen, United Kingdom, under the Accession Number NCIMB 40418.

3. A glyphosate-degrading agent that metabolises glyphosate via the AMPA pathway comprising a Pseudomonas species, strain SCll, a culture of which was deposited on 10 May 1991 under the terms of the Budapest Treaty with the National Collection of Industrial and Marine Bacteria Limited, Aberdeeen, United Kingdom, under the Accession Number NCIMB 40419.

4. Glyphosate-degrading proteins isolated from microorganisms claimed in claim 1.

5. Glyphosate-degrading proteins isolated from strain SC9 as claimed in claim 2.

6. Glyphosate-degrading proteins isolated from strain SCll as claimed in claim 3.

7. Pure protein having the amino acid sequence of proteins as claimed in any of claims 4—6.

8. Protein as claimed in claim 7 which is synthetic.

9. A recombinant DNA sequence coding for a protein as claimed in any of claims 4-8.

10. A vector containing a DNA sequence as claimed in claim 9.

11. A biological system including DNA as claimed in claim 9 which allows expression of the encoded protein.

12. A biological system as claimed in claim 11 which is a micro-organism.

13. A biological system as claimed in claim 11 which is a plant.

14. Plants transformed with recombinant DNA as claimed in claim 9.

15. Protein derived from expression of the DNA as claimed in claim 9.

16. Glyphosate-degrading protein produced by expression of recombinant DNA within plants as claimed in claim 14.

17. Plants tolerant to the herbicide glyphosate in consequence of having been transformed by a gene expressing a protein that metabolises glyphosate via the AMPA pathway.

SUBSTITUTE SHEET

18. A glyphosate-degrading composition containing as active ingredient an agent as claimed in any of claims 1-3 in admixture with a carrier composition acceptable in agricultural or waste-disposal practice.

19. A glyphosate-degrading composition containing as active ingredient one or more of the proteins as claimed in any of claims 4-8 in admixture with a carrier composition acceptable in agricultural or waste-disposal practice.

20. A process of degrading glyphosate which comprises exposing it to the agents, proteins or compositions as claimed in any of claims 1-8 or claims 15-16 or claims 18-19.

SUBSTITUTESHEET