Acta Biotechnol. 23 (2003) 1, 3 –17
`
`Purification and Characterisation of the Enantiospecific Dioxygenases
`from Delftia acidovorans MC1 Initiating the Degradation
`of Phenoxypropionate and Phenoxyacetate Herbicides
`
`WESTENDORF*, A., MÜLLER, R. H., BABEL, W.
`
`UFZ – Umweltforschungszentrum Leipzig – Halle GmbH
`Sektion Umweltmikrobiologie
`Permoserstraße 15
`04318 Leipzig, Germany
`
`* Corresponding author
`Phone: + 49 341 235 2418
`Fax: + 49 341 235 2247
`E-mail: annewest@umb.ufz.de
`
`Summary
`
`After cultivation on (R,S)-2-(2,4-dichlorophenoxy)propionate, two α-ketoglutarate-dependent dioxy-
`genases were isolated and purified from Delftia acidovorans MC1, catalysing the cleavage of the ether
`bond of various phenoxyalkanoate herbicides. One of these enzymes showed high specificity for the
`cleavage of the R-enantiomer of substituted phenoxypropionate derivatives: the Km values were 55 µM
`and 30 µM, the kcat values 55 min–1 and 34 min–1 with (R)-2-(2,4-dichlorophenoxy)propionate
`[(R)-2,4-DP] and (R)-2-(4-chloro-2-methylphenoxy)propionate, respectively. The other enzyme
`predominantly utilised the S-enantiomers with Km values of 49 µM and 22 µM, and kcat values of
`50 min–1 and 46 min–1 with (S)-2-(2,4-dichlorophenoxy)propionate [(S)-2,4-DP] and (S)-2-(4-chloro-
`2-methylphenoxy)propionate, respectively. In addition, it cleaved phenoxyacetate herbicides (i.e.
`2,4-dichlorophenoxyacetate: Km = 123 µM, kcat = 36 min–1) with significant activity. As the second
`substrate, only α-ketoglutarate served as an oxygen acceptor for both enzymes. The enzymes were
`characterised by excess substrate inhibition kinetics with apparent Ki values of 3 mM with (R)-2,4-DP
`and 1.5 mM with (S)-2,4-DP. The reaction was strictly dependent on the presence of Fe2+ and
`ascorbate; other divalent cations showed inhibitory effects to different extents. Activity was
`completely extinguished within 2 min in the presence of 100 µM diethylpyrocarbonate (DEPC).
`
`Introduction
`
`The degradation of chlorinated/methylated phenoxyalkanoate herbicides is usually initi-
`ated by the cleavage of the ether bond of these compounds leading to the formation of
`respective phenolic moieties and the oxidised alkanoates. The prevailing reaction is ca-
`talysed by an α-ketoglutarate-dependent dioxygenase as studied with the purified
`enzyme from Ralstonia eutropha JMP134; α-ketoglutarate is oxidatively decarboxylated
`
`© WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim, 2003 0138-4988/03/0101-0003 $ 17.50+.50/0
`
`Inari Ex. 1027
`Inari Agric. v. Corteva Agriscience
`PGR2023-00022
`Page 00001
`
`

`

`4
`
`Acta Biotechnol. 23 (2003) 1
`
`to succi nate in this step [1, 2]. The enzyme from the above strain was determined to be
`highly specific for the phenoxyacetate structure. Genes encoding this type of enzyme,
`i.e. tfdA, were found to be widespread in microbial communities [3–9]. By contrast,
`there are only a few examples of axenic strains that are able to productively utilise
`phenoxypropionate herbicides: Sphingomonas herbicidovorans (Flavobacterium sp.)
`MH [10, 11], Rhodoferax sp. P230 [12] and Delftia (Comamonas) acidovorans MC1
`[13]. Remarkably, these strains utilise a broader spectrum of phenoxy herbicides; in
`addition to their ability of degrading the racemic phenoxypropionates, they similarly
`attack phenoxyacetate derivatives. Again, this reaction proved to be catalysed by an
`α-ketoglutarate-dependent dioxygenase [14–16].
`Preliminary investigations revealed enantioselective properties of the enzymes
`catalysing the cleavage of the ether bond of the racemic phenoxypropionates as holds
`for S. herbicidovorans MH [14, 17] and D. acidovorans MC1 [15]. The enantioselectiv-
`ity of this reaction is also supported by the fact that a strain of Alcaligenes denitrificans
`is only able to utilise the R-enantiomer of 2-(4-chloro-2-methylphenoxy)propionate
`[18]. These enzymes have, however, not been studied in detail.
`The present investigation is aimed at elucidating the enzymatic basis of the broad herbi-
`cide consumption profile including the phenoxyacetate derivatives in D. acidovo-
`rans MC1. Enzymes carrying ether-cleaving activity with respect to the various phen-
`oxyalkanoate herbicides were purified and characterised with regard to their catalytic
`properties.
`
`Materials and Methods
`
`Cultivation
`Strain MC1 was inoculated from individual colonies grown on (R,S)-2,4-DP into PYE medium (3 g/l
`of each peptone and yeast extract and 1.8 g/l of fructose) and propagated in an overnight culture; then
`it was incubated for a further day in the presence of 100 mg/l of (R,S)-2,4-DP. One litre of this culture
`was used to inoculate a fermenter (INFORS, ALGU 503) operating at a working volume of 4 l.
`Cultivation proceeded under aerobic conditions at pH 8.0 and 30 °C. The medium was composed of
`[mg/l]: NH4Cl, 1520; KH2PO4, 680.5; K2HPO4, 870.9; CaCl2 × 6 H2O, 11; MgSO4 × 7 H2O, 142;
`FeSO4 × 7 H2O, 10; CuSO4 × 5 H2O, 1.57; MnSO4 × 4 H2O, 1.23; ZnSO4 × 7 H2O, 0.88; Na2MoO4
`× 2 H2O, 0.5. (R,S)-2,4-DP was supplied as the sole carbon and energy source in a fed batch regime by
`adapting the feed rate to the substrate consumption capacity of the culture. Excess substrate in the
`medium did not exceed 0.5 g/l. In addition, a trace element solution containing the respective salts in
`the concentration specified above was continuously fed at a rate of 0.77 ml/h. Biomass increase
`(OD700) and substrate consumption were monitored off-line. After the cell density had reached an
`OD700 = 2, about 50% of the suspension was harvested. The remaining culture in the fermenter was
`supplied with an adequate portion of the mineral salt solution and cultivation continued.
`Cells were harvested by centrifugation at 4 °C at 8875 × g. The pellet was washed and re-suspended in
`10 mM Tris/HCl buffer, pH 7.5 (buffer A). The suspension was stored at –20 °C pending further use.
`
`Purification
`Purification was performed by application of a GradiFrac System including pump P-1 and moni-
`tor UV-1 (AMERSHAM PHARMACIA BIOTECH). Unless otherwise indicated, all steps were car-
`ried out at 4 °C.
`
`PGR2023-00022 Page 00002
`
`

`

`WESTENDORF, A. et al., Enantiospecific Dioxygenases from D. acidovorans
`
`5
`
`Step 1: Cells were disintegrated using a French Pressure Cell (AMINCO, Silver Spring, USA) by three
`passages at 140 MPa. Particle-free supernatants were obtained by centrifugation at 4 °C for 25 min at
`20,000 × g.
`Step 2: Protamine sulphate (1%, pH 7.0) was added to the supernatant to give a final concentration of
`0.15%. The solution was stirred for 20 min and separated from the precipitate by centrifugation for
`20 min at 20,000 × g.
`Step 3: Ammonium sulphate was added by stirring the supernatant from step 2 to give a final
`concentration of 1.25 M. After stirring for 20 min, the precipitate was removed (20 min centrifugation
`at 20,000 × g). A further quantity of ammonium sulphate was added to the supernatant to give a final
`concentration of 2.5 M. After stirring for a further 20 min, the pellet was collected by centrifugation
`and used for further steps.
`Step 4: The pellet was solubilised in a small quantity of buffer A and applied to a Hi Prep 26/10
`desalting column (PHARMACIA, Sweden) filled with Sephadex G-25. The column was equilibrated
`with buffer A. Application of the sample and elution proceeded at a rate of 5 ml/min. The eluate was
`sampled in fractions of 10 ml, which were monitored for conductivity and protein content.
`Step 5: Ion exchange chromatography (IEX) was carried out on a Source30Q (PHARMACIA,
`Sweden) column (25/14). The column was equilibrated with buffer A. The sample was applied with a
`rate of 1 ml/min and further treated with buffer A to remove non-bound proteins. The proper elution
`was started at the same rate by linearly increasing NaCl gradients of varying steepness. The total
`volume of the gradient solutions amounted to 120–180 ml. Eluates were collected in fractions of 6 ml.
`Step 6: Hydrophobic interaction chromatography (HIC) was performed on a Butyl-Sepharose
`(PHARMACIA, Sweden) column with a bed volume of 20 ml. The column was equilibrated with
`buffer A containing different initial concentrations of ammonium sulphate. The protein samples
`derived from each purification step were supplied with ammonium sulphate to reach the desired initial
`concentration corresponding to the column equilibrium concentration. Desorption was performed by
`using a linearly decreasing ammonium sulphate gradient at a rate of 1 ml/min. The total volume of the
`gradient solutions amounted to 180 ml. Eluates were collected in fractions of 6 ml.
`Step 7: Gel filtration was performed on a Superdex200 prep grade (PHARMACIA, Sweden) column
`(100/10). It was equilibrated with buffer A containing 0.15 M NaCl. Before application, the protein
`solutions were concentrated by precipitation with 3.4 M ammonium sulphate, re-solubilised in a small
`volume and applied and eluted at a rate of 0.5 ml/min. The eluates were collected in fractions of 10 ml.
`
`Electrophoresis
`
`One-dimensional electrophoresis was carried out with a PowerEase 500 system (NOVEX, San Diego,
`USA) on 12% Tris glycine precast acrylamide gels (NOVEX). The samples contained 1–2 µg of
`protein. The running buffer was composed of 2.9 g/l Tris base, 14.4 g/l glycine and 1.0 g/l SDS. The
`sampling buffer contained 2 ml glycerol, 4 ml 10% SDS [w/v], 0.5 ml 1% bromophenol blue and
`2.5 ml 0.5 M Tris/HCl, pH 6.8, in 10 ml distilled water. The gels were treated by 10 mA. Mark
`12TM Wide Range Standard (NOVEX) was applied as a reference.
`Gels were stained by using the SilverXpress kit (NOVEX) following the instructions of the supplier, or
`by using Coomassie brilliant blue.
`
`Enzyme Measurement
`
`Ether-cleaving dioxygenase activity was routinely measured according to FUKUMORI and HAUSINGER
`[1] by determining the phenolic intermediates liberated in this enzymatic reaction after their reaction
`with 4-aminoantipyrine. The assay for the enzyme reaction contained 1 mM herbicide (sodium salt),
`1 mM α-ketoglutarate, 200 µM ascorbic acid, 200 µM ammonium iron(II) sulphate in 10 mM
`imidazole buffer (pH 6.75). The buffer was gassed with air at 30 °C for 30 min before adding the
`individual components. The reaction was performed at 30 °C, started by adding the enzyme at a
`
`PGR2023-00022 Page 00003
`
`

`

`6
`
`Acta Biotechnol. 23 (2003) 1
`
`concentration of 0.05–0.3 µM (with respect to the monomer). After respective times, usually within a
`period of 10 min, samples (up to 5) were taken and the enzyme reaction was stopped by adding 50 µl
`of 20 mM EDTA to 1 ml of the reaction mixture. The phenolic products were determined by adding
`100 µl of borate buffer, pH 10 (3.09 g H3BO4; 3.73 g KCl; 44 ml 1 N NaOH ad 1 l), 10 µl 2% 4-
`aminoantipyrin and 10 µl 8% potassium hexacyanoferrate III. After 5 min of incubation at 30 °C, the
`extinction of the samples was measured at 510 nm (U-2000, HITACHI, Tokyo, Japan). The progress
`curves proved almost linear under these conditions as followed from linear regression giving a
`measure of confidence of > 0.98 with the substrate concentrations tested. The standard deviation of
`the individual points derived from triplicates amounted to ≤ 8 %. A comparison of the colorimetric
`assay with a direct measurement of the reaction by following substrate disappearance and product
`accumulation via HPLC gave identical rates.
`
`Analytical Methods
`
`Phenoxyalkanoates were determined by HPLC according to [19]. The biomass concentration was
`measured on the basis of the optical density at 700 nm. An OD700 = 1 corresponds to a biomass
`concentration of about 0.5 g/l dry mass.
`
`Results
`
`Enzyme Purification
`
`Treatment of crude extract in the presence of 0.15% protamine sulphate and precipita-
`tion of proteins in the range of 1.25–2.5 M ammonium sulphate kept almost all of the
`ether-cleaving enzymes of the crude extract detected by measuring activities in the pres-
`ence of the individual enantiomers of 2,4-DP (Tab. 1) and of 2,4-D (not shown).
`Application of IEX chromatography separated two protein fractions from each other,
`which exhibited distinct activity to the various substrates. The fraction eluted at low salt
`concentrations of around 0.12 M exhibited pronounced activity toward (S)-2,4-DP and,
`in addition, activity with respect to 2,4-D which coincided with the peak at low NaCl
`concentration. A second fraction found at around 0.35 M NaCl exhibited selective
`activity by only converting the R-enantiomer. Pooled fractions carrying the respective
`activities were further treated by HIC chromatography. This resulted in a significant
`purification of the two enzymes (Tab. 1), which were consecutively named the R-spe-
`cific RdpA and the S-specific SdpA enzyme, respectively. Again, activity with respect
`to 2,4-D cleavage coincided with the activity profile for (S)-2,4-DP. Characterisation by
`SDS-PAGE revealed that this purification protocol resulted in a rather pure product
`with the S-specific enzyme (Fig. 1). In contrast, significant impurities were still
`observed with the R-specific enzyme (Fig. 2). Densidometric analysis revealed an en-
`richment of 80% of the R-specific enzyme at this preparation step. Use of gel filtration
`to remove these impurities was less successful as the activity was drastically diminished
`(Tab. 1). This step was therefore omitted in further preparations. These impurities did,
`however, not disturb consecutive kinetic investigations as it was proven that any reac-
`tion using the phenolic intermediate as the substrate (chlorophenol hydroxylase) was
`absent after IEX chromatography. This was verified by keeping product, i.e. dichloro-
`phenol, unutilised within 12 h of incubation in the presence of the respective enzyme
`
`PGR2023-00022 Page 00004
`
`

`

`WESTENDORF, A. et al., Enantiospecific Dioxygenases from D. acidovorans
`
`7
`
`fractions. From the electrophoretic patterns (electropherograms), molecular weights of
`the (subunits of the) S-specific and the R-specific enzyme of about 32 kD and 36 kD,
`respectively, were derived.
`
`Tab. 1. Purification of 2,4-DP/α-KG dioxygenases from D. acidovorans MC1
`
`_____________________________________________________________________________________________________________________________________________________________________
`
` (R)-specific enzyme
`
`(S)-specific enzyme
`
`_____________________________________________________________________________________________________________________________________________________________________
`Purification
`Specific
`Purification
`Total
`Specific
`Purification
`Total
`step
`activity
`activity
`activity
`activity
`[-fold]
`[-fold]
`[mU/mg]
`[%]
`[mU/mg]
`[%]
`_____________________________________________________________________________________________________________________________________________________________________
`Crude
`7.2
`1
`100
`6.8
`1
`100
`extract
`PS-P
`AS-P
`Source 30Q
`Butyl-
`sepharose
`Gel filtration
`
`6.7
`13.8
`25.4
`172.8
`
`154
`
`0.93
`1.9
`3.5
`24
`
`21.4
`
`93
`92
`66
`45
`
`1
`
`5.8
`11.3
`101
`259
`
`205
`
`0.85
`1.7
`14.9
`38
`
`30.2
`
`95
`86
`65
`8
`
`0.8
`
`_____________________________________________________________________________________________________________________________________________________________________
`
`Fig. 1. Purification of (S)-2,4-DP/α-KG dioxygenase from D. acidovorans MC1
`Protein samples were analysed by gel electrophoresis on a 12% polyacrylamide gel and
`visualised by silver staining.
`1: Crude extract, 2: Supernatant after the treatment with 0.15% protamine sulphate,
`3: Fraction precipitated between 1.25–2.5 M ammonium sulphate, 5: Pooled fractions
`with (S)-2,4-DP-cleavage activity after IEX, 6+7: Fractions with (S)-2,4-DP-cleavage
`activity after HIC, 8+10: Pooled fractions with (S)-2,4-DP-cleavage activity after GF,
`4+9: Molecular weight marker.
`
`PGR2023-00022 Page 00005
`
`

`

`8
`
`Acta Biotechnol. 23 (2003) 1
`
`Fig. 2. Purification of (R)-2,4-DP/α-KG dioxygenase from D. acidovorans MC1
`Protein samples were analysed by gel electrophoresis on a 12% polyacrylamide gel and
`visualised by silver staining.
`1: Crude extract, 2: Supernatant after treatment with 0.15% protamine sulphate,
`3+4:Pooled fractions with (R)-2,4-DP-cleavage activity after IEX, 5+6: Pooled fractions
`with (R)-2,4-DP-cleavage activity after HIC, 7+8: Pooled fractions with (R)-2,4-DP-
`cleavage activity after GF, 9+10: Molecular weight marker.
`
`The R-specific enzyme was found to be stable and did not require supplements when
`stored at – 20° in buffer A. Storage at 4 °C resulted in a loss of activity within about
`20 days; the enzyme was significantly stabilised under these conditions by adding 1%
`BSA or 3.4 M ammonium sulphate (Fig. 3). A similar pattern with respect to stability
`was observed with the S-specific enzyme (not shown).
`
`Fig. 3. Stability of the (R)-2,4-DP/α-KG dioxygenase (after HIC) at different storage
`conditions
`( ) 4 °C without addition, () 4 °C after the addition of 50 µM ascorbic acid, ( ) 4 °C
`after the addition of 1% BSA, (∆) 4 °C after the addition of 3.4 M ammonium sulphate,
`(N) – 20 °C without addition.
`
`PGR2023-00022 Page 00006
`
`

`

`9
`
`WESTENDORF, A. et al., Enantiospecific Dioxygenases from D. acidovorans
`Kinetic Properties
`Preliminary investigations were performed in order to elucidate the optimum conditions
`for the enzyme reaction. The activity of the R- and S-specific enzyme depended on the
`presence of both Fe2+ and ascorbic acid, as is shown in Figs. 4 and 5. The concentration-
`dependent activity profiles were characterised by a complex shape (see Discussion).
`Consecutive kinetic investigations were performed at a Fe2+ and an ascorbic acid con-
`centration of 200 µM. With regard to the physical parameters, the temperature optimum
`was determined as 30 °C with the R-specific and 25 °C with the S-specific enzyme; the
`optimum pH value was around 6 with both enzymes. The activity patterns depending on
`the latter parameters were of similar complexity as indicated above.
`
`Fig. 4. Effect of ascorbic acid on the activity of 2,4-DP/α-KG dioxygenases
`() S-specific enzyme, (∆) R-specific enzyme.
`
`Fig. 5. Activation of 2,4-DP/α-KG dioxygenases by ferrous ions
`() S-specific enzyme, (∆) R-specific enzyme.
`The enzyme activity was determined using standard assay conditions.
`
`PGR2023-00022 Page 00007
`
`

`

`10
`
`Acta Biotechnol. 23 (2003) 1
`
`Tab. 2. Substrate specificity and kinetic parameters of (R)-2,4-DP/α-KG dioxygenase
`Km [µM]
`kkat [min–1]
`kkat/Km
`
`_____________________________________________________________________________________________________________________________________________________________________
`Substrate (concentration range studied)
`[%]
`_____________________________________________________________________________________________________________________________________________________________________
`(20–200 µM)
`54.9 ± 4.4
`55.2
`1
`/ (100)
`(100–5000 µM)
`–
`–
`–
`(10–200 µM)
`30 ± 1.2
`34.4
`1.15
`(20–600 µM)
`261 ± 40
`3.8
`0.02
`(20–2000 µM)
`1305 ± 120
`7.6
`0.006
`(100–2000 µM)
`–
`–
`–
`(100–2000 µM)
`–
`–
`–
`
`/ (115)
`/
`(2)
`/
`(0.6)
`
`(R)-2,4-DP
`(S)-2,4-DP
`(R)-2,4-MCPP
`(S)-2,4-MCPP
`2,4-D
`2,4-DB
`3-Phenoxypro-
`pionic acid
`(R,S)-2-(2,4,5-Trichlorophenoxy)propionic acid
`(10-200 µM)
`(R,S)-2-(m-Chlorophenoxy)propionic acid
`(20-400 µM)
`(R,S)-2-(4-Chlorophenoxy)propionic acid
`(10-200 µM)
`
`130.8 ± 9.3
`176.2 ± 4.2
`
`131 ± 0.4
`
`50
`34.4
`
`65.2
`
`0.382
`0.195
`
`0.498
`
`/
`/
`
`/
`
`(38)
`(19)
`
`(50)
`
`_____________________________________________________________________________________________________________________________________________________________________
`
`(2.5–50 µM)
`α-Ketoglutarate
`(10–5000 µM)
`α-Ketobutyrate
`α-Ketoadipate
`(10–5000 µM)
`α-Ketoisovalerate (10–5000 µM)
`α-Ketovalerate
`(10–5000 µM)
`Pyruvate
`(10–10000 µM)
`
`27.8 ± 3.8
`–
`–
`–
`–
`–
`
`20
`–
`–
`–
`–
`–
`
`/ (100)
`
`0.719
`–
`–
`–
`–
`–
`
`_____________________________________________________________________________________________________________________________________________________________________
`
`(– = no activity detected)
`All experiments were performed at 30 °C in 10 mM imidazol buffer (pH 6) containing 200 µM
`ascorbate, 200 µM (NH4)3Fe(SO4)2, the respective substrates with the concentration ranges indicated,
`1 mM of the second substrate [α-KG at variable herbicides; (R)-DP at variable keto acids] and
`enzyme.
`
`Kinetic investigations were aimed at revealing the enzymatic basis of substrate speci-
`ficity and diversity with respect to the herbicides. The results obtained with the R-spe-
`cific enzyme are summarised in Tab. 2. As expected, the highest activity was obtained
`with the R-enantiomers of the dichlorinated and chloro/methyl-substituted phenoxypro-
`pionates taking into account kcat and Km and the quotient of both parameters. By
`contrast, very low or no activity was detected in the presence of the respective S-enan-
`tiomers; 2,4-D was not cleaved at all. This is reflected by both kcat and Km. Considerable
`activity was also found with other derivatives of 2-phenoxypropionates including the
`trichlorinated compound. Although these were applied as the racemates, it is most likely
`the R-enantiomers that were predominantly attacked. (Some effects exerted by the
`presence of the respective S-enantiomers cannot, however, be ruled out under these
`
`PGR2023-00022 Page 00008
`
`

`

`WESTENDORF, A. et al., Enantiospecific Dioxygenases from D. acidovorans
`
`11
`
`conditions. Hence these values are more indicative of the consumption of these
`compounds rather than usable kinetic constants.) As the second substrate, only α-keto-
`glutarate served as an oxygen acceptor; alternative substrates were of no effect in a wide
`range of concentrations tested. It should be noted that by applying variable substrate
`concentrations similarly complex characteristics were observed as shown in Figs. 4 and
`5, the reasons for which are discussed later. Nevertheless, the LINEVEAVER-BURK, HANES
`and EADIE-HOFSTEE plots exhibited fairly linear dependencies with confidence intervals
`in most cases of ≥ 95%. The kinetic constants derived from such complex characteris-
`tics should, however, only be considered as apparent (overall) constants.
`The picture was rather different with the S-specific enzyme. As expected, pronounced
`activities were found with the S-configuration of 2,4-DP and MCPP. Due to the Km
`value, (S)-MCPP was the preferred substrate. Most important, however, is the fact that
`this enzyme is also able to cleave 2,4-D with significant activity (Tab. 3). Again, only
`α-ketoglutarate served as the second substrate.
`
`Tab. 3. Substrate specificity and kinetic parameters of (S)-2,4-DP/α-KG dioxygenase
`Km [µM]
`kkat [min–1]
`kkat/Km
`
`_____________________________________________________________________________________________________________________________________________________________________
`Substrate (concentration range studied)
`[%]
`
`_____________________________________________________________________________________________________________________________________________________________________
`
`(50–500 µM)
`(20–200 µM)
`(20–1000 µM)
`(10–100 µM)
`(50–200 µM)
`(100–2000 µM)
`(100–2000 µM)
`
`(R)-2,4-DP
`(S)-2,4-DP
`(R)-2,4-MCPP
`(S)-2,4-MCPP
`2,4-D
`2,4-DB
`3-Phenoxypro-
`pionic acid
`(R,S)-2-(2,4,5-Trichlorophenoxy)propionic acid
`(100–2000 µM)
`(R,S)-2-(m-Chlorophenoxy)propionic acid
`(20–200 µM)
`(R,S)-2-(4-Chlorophenoxy)propionic acid
`(10–100 µM)
`
`–
`49 ± 9.1
`–
`21.8 ± 5
`122.8 ± 2
`–
`–
`
`–
`11.5 ± 6.4
`
`68.3 ± 9.8
`
`–
`50
`–
`46
`36
`–
`–
`
`–
`15
`
`17
`
`–
`1.02
`–
`2.11
`0.29
`–
`–
`
`–
`0.13
`
`0.25
`
`/ (100)
`
`/ (207)
`/
`(29)
`
`/
`
`/
`
`(13)
`
`(25)
`
`_____________________________________________________________________________________________________________________________________________________________________
`
`(2.5–50 µM)
`α-Ketoglutarate
`(10–5000 µM)
`α-Ketobutyrate
`α-Ketoadipate
`(10–5000 µM)
`α-Ketoisovalerate (10–5000 µM)
`α-Ketovalerate
`(10–5000 µM)
`Pyruvate
`(10–10000 µM)
`
`24.1 ± 7.6
`–
`–
`–
`–
`–
`
`1.7
`–
`–
`–
`–
`–
`
`/ (100)
`
`34
`–
`–
`–
`–
`–
`
`_____________________________________________________________________________________________________________________________________________________________________
`
`(– = no activity detected)
`All experiments were performed at 25 °C in 10 mM imidazol buffer (pH 6) containing 200 µM
`ascorbate, 200 µM (NH4)3Fe(SO4)2, the respective substrates with the concentration ranges indicated,
`1 mM of the second substrate [α-KG at variable herbicides; (R)-DP at variable keto acids] and
`enzyme.
`
`PGR2023-00022 Page 00009
`
`

`

`12
`
`Acta Biotechnol. 23 (2003) 1
`
`Fig. 6 a. Effect of (R)-2,4-DP on the activity of (R)-2,4-DP/α-KG dioxygenase
`The assay contained 0.055 µM enzyme.
`
`Fig. 6 b. Effect of (S)-2,4-DP on the activity of (S)-2,4-DP/α-KG dioxygenase
`The assay contained 0.055 µM enzyme.
`
`Both enzymes exhibited excess substrate inhibition. According to the data in Figs. 6a
`and b, Ki values of about 3 mM with the R-specific and of about 1.5 mM with the
`S-specific enzyme can approximately be taken from these data. However, this is only
`formal and does not reflect the real inhibition pattern, again caused by the complex de-
`pendencies on substrate concentration.
`Fe2+ is essential for the enzyme reaction (Fig. 5), other divalent cations tested such as
`Cu2+, Mn2+, Mg2+, Zn2+, Ni2+ and Co2+ could not substitute this ion.
`
`PGR2023-00022 Page 00010
`
`

`

`WESTENDORF, A. et al., Enantiospecific Dioxygenases from D. acidovorans
`
`13
`
`Fig. 7 a. Metal ion-dependent inactivation of (R)-2,4-DP/α-KG dioxygenase
`The enzyme activity was determined by using standard assay condition in the presence
`of various metal ions.
`(∆) Mg2+, ( ) Ni2+, ( ) Mn2+, (G) Co2+, () Zn2+, (N) Cu2+.
`
`Fig. 7 b. Metal ion-dependent inactivation of (S)-2,4-DP/α-KG dioxygenase
`The enzyme activity was determined by using standard assay condition in the presence of various
`metal ions.
`(∆) Mg2+, ( ) Ni2+, ( ) Mn2+, (G) Co2+, () Zn2+, (N) Cu2+.
`
`Moreover, they even behaved in an inhibitory manner. This was most significantly
`observed with copper and nickel, both of which exhibited, according to the properties
`shown in Figs. 7a and b, Ki values of around 50 µM with the R-specific but far less than
`50 µM with the S-specific enzyme. The other ions tested exerted weaker effects. In
`addition, the enzyme was inhibited by DEPC. The application of 100 µM led to an
`almost complete loss of activity within 2 min with both enzymes; lower concentrations
`behaved more gradually (Figs. 8a,b).
`
`PGR2023-00022 Page 00011
`
`

`

`14
`
`Acta Biotechnol. 23 (2003) 1
`
`Fig. 8 a. Inactivation of (R)-2,4-DP/α-KG dioxygenase with DEPC
`( ) 100 µM DEPC, (N) 50 µM DEPC, () 20 µM DEPC, (∆) 0 µM DEPC.
`
`Fig. 8 b. Inactivation of (S)-2,4-DP/α-KG dioxygenase with DEPC
`( ) 100 µM DEPC, (N) 50 µM DEPC, () 20 µM DEPC, (∆) 0 µM DEPC.
`
`Discussion
`
`The present investigation has clearly elucidated the enzymatic basis for the degradation
`of enantiomeric phenoxypropionate and the phenoxyacetate herbicides. Two different
`enzymes with different activity profiles were isolated. One of the enzymes is very spe-
`cific to the R-configuration of the phenoxypropionates, it neither cleaved the S-enantio-
`
`PGR2023-00022 Page 00012
`
`

`

`WESTENDORF, A. et al., Enantiospecific Dioxygenases from D. acidovorans
`
`15
`
`mers to a significant extent nor did it utilise phenoxyacetate derivatives (2,4-D). The
`other enzyme isolated has predominant activity towards the S-configuration of the phe-
`noxypropionates; in addition, it exhibits significant activities against 2,4-D. Due to their
`activities and the predominant substrates utilised, these enzymes should be called
`(R)-phenoxypropionate/α-ketoglutarate-dioxygenase (RdpA) and (S)-phenoxypropiona-
`te/α-ketoglutarate-dioxygenase (SdpA), respectively. The reaction type proved similar
`to the 2,4-D/α-ketoglutarate dioxygenase from Ralstonia eutropha JMP134(pJP4) [1,
`2]: the activity was strongly dependent on Fe2+ and ascorbate [16] as was the general
`case with α-ketoglutarate-dependent dioxygenases [20, 21]. It was inhibited by DEPC, a
`compound known to be reactive to histidine [22]. Two histidine residues have been
`shown to be essential components in the active centre of TfdA [24].
`Besides the enzymes of strain MC1, there are strong indications of enantioselective en-
`zymes with respect to 2-phenoxypropionates in the case of S. herbicidovorans MH. The
`separation of crude extracts of this strain by SDS electrophoresis revealed distinct pro-
`tein bands of 32 kD and 34 kD after growth on (S)-MCPP or (R)-MCPP, respectively,
`and of both of these bands after growth on the racemate [14]. This pattern tallies with
`the molecular weights found for the respective ether-cleaving proteins (subunits) of
`strain MC1. Moreover, a molecular weight of 32 kD is found with the canonical TfdA
`protein from R. eutropha JMP134 [2], cleaving 2,4-D with high specificity. The appli-
`cation of primers, derived from conserved regions of tfdA genes [7], did not result in a
`specific amplification product in PCR with the genome of strain MC1 [12, 16] and
`strain MH as the template. This might be expected in agreement with the different sub-
`strate specificity found with the various strains (i.e. JMP134 vs. MC1). But by
`comparing strain MC1 and MH, significant differences became obvious, too: whereas
`strain MH prefers 2,4-D as derived from the maximum growth rate on the various
`herbicides [17], strain MC1 has a four-fold higher rate on (RS)-2,4-DP in comparison to
`2,4-D [16]. This means that a similar herbicide consumption profile in these strains
`must be provoked by a divergent genetic background. This deviates from the experience
`with tfdA: this gene determining 2,4-D degradative specificity is distributed in the
`microbial world in a highly conserved manner [7]. To complete this picture, a third
`strain, Rhodoferax sp. P230, has the same profile of herbicide utilisation as strains MH
`and MC1. This strain carries TfdA (as followed from sequencing of a respective PCR
`product), but is nevertheless unable to productively degrade 2,4-D [12, 16]. Very
`recently, two further strains were described [24] as carrying phenoxypropionate cleav-
`age activity and also giving positive response to tfdA probes.
`Investigations are in progress to elucidate the molecular structure of both enzymes from
`strain MC1 and the genes they derive from. It is furthermore apparent from the present
`results that both enzymes are characterised by complex kinetic behaviour, indicating
`isoforms of the enzyme. Proteolytic modifications should be ruled out as the cause of
`this pattern as this was also observed in enzyme preparations carried out in the presence
`of protease inhibitors. However, two-dimensional electrophoretic separation of the
`respective enzyme preparations or applying crude extracts revealed protein spots which
`are indicative of isoforms differing in charge rather than molecular weight [25]. This
`has to be verified in order to elucidate the structure and catalytic properties of these
`isoenzymes.
`
`PGR2023-00022 Page 00013
`
`

`

`Acta Biotechnol. 23 (2003) 1
`
`Bovine serum albumin
`2,4-Dichlorophenoxyacetate
`4-(2,4-Dichlorophenoxy)butyrate
`2-(2,4-Dichlorophenoxy)propionate
`Diethylpyrocarbonate
`Hydrophobic interaction chromatography
`Ion exchange chromatography
`2-(4-Chloro-2-methylphenoxy)propionate
`Gel filtration
`
`16
`
`Abbreviations
`
`BSA
`2,4-D
`2,4-DB
`2,4-DP
`DEPC
`HIC
`IEX
`MCPP
`GF
`
`Received 1 February 2002
`Received in revised form 4 June 2002
`Accepted 6 June 2002
`
`References
`
`[1] FUKUMORI, F., HAUSINGER, R. P.: Alcaligenes eutrophus JMP134 “2,4-dichlorophenoxyacetate
`monooxygenase” is an α-ketoglutarate-dependent dioxygenase. J. Bacteriol. 175 (1993),
`2083–2086.
`[2] FUKUMORI, F., HAUSINGER, R. P.: Purification and characterization of 2,4-dichlorophenoxyace-
`tate/α-ketoglutarate dioxygenase. J. Biol. Chem. 268 (1993), 24311–24317.
`[3] NEILSON, J. W., JOSEPHSON, K. L., PEPPER, I. L., ARNOLD, R. B., DIGIOVANNI, G. D., SIN-
`CLAIR, N. A.: Frequency of horizontal gene transfer of a large catabolic plasmid (pJP4) in soil.
`Appl. Environ. Microbiol. 60 (1994), 4053–4058.
`[4] HONG, S. M., AHN, Y. J., PARK, Y. K., MIN, K. H., KIM, C. K., KA, J. O.: Effects of genetically
`different 2,4-D-degradative plasmids on degradation phenotype and competitiveness of soil
`microorganisms. J. Microbiol. 33 (1995), 208–214.
`[5] FULTHORPE, R. R., RHODES, A. N., TIEDJE, J. M.: Pristine soils mineralise 3-chlorobenzoate and
`2,4-dichlorophenoxyacetate via different microbial populations. Appl. Environ. Microbiol.
`62 (1996), 1159–1166.
`[6] HOGAN, D. A., BUCKLEY, D. H., NAKATSU, C. H., SCHMIDT, T. M., HAUSINGER, R. P.:
`Distribution of the tfdA gene in soil bacteria that do not degrade 2,4-dichlorophenoxyacetic acid
`(2,4-D). Microb. Ecol. 34 (1997), 90–96.
`[7] VALLAEYS, T., FULTHORPE, R. R., WRIGHT, A. M., SOULAS, G.: The metabolism of 2,4-dichlo-
`rophenoxyacetic acid degradation involves different families of tfdA and tfdB genes according to
`PCR-RFLP analysis. FEMS Microbiol. Ecol. 20 (1996), 163–172.
`[8] VALLAEYS, T., COURDE, L.