Biodegradation of All Stereoisomers of the EDTA Substitute Iminodisuccinate by Agrobacterium tumefaciens BY6 Requires an Epimerase and a Stereoselective C-N Lyase

Abstract

Biodegradation tests according to Organization for Economic Cooperation and Development standard 301F (manometric respirometry test) with technical iminodisuccinate (IDS) revealed ready biodegradability for all stereoisomers of IDS. The IDS-degrading strain Agrobacterium tumefaciens BY6 was isolated from activated sludge. The strain was able to grow on each IDS isomer as well as on Fe²⁺-, Mg²⁺-, and Ca²⁺-IDS complexes as the sole carbon, nitrogen, and energy source. In contrast, biodegradation of and growth on Mn²⁺-IDS were rather scant and very slow on Cu²⁺-IDS. Growth and turnover experiments with A. tumefaciens BY6 indicated that the isomer R,S-IDS is the preferred substrate. The IDS-degrading enzyme system isolated from this organism consists of an IDS-epimerase and a C-N lyase. The C-N lyase is stereospecific for the cleavage of R,S-IDS, generating d-aspartic acid and fumaric acid. The decisive enzyme for S,S-IDS and R,R-IDS degradation is the epimerase. It transforms S,S-IDS and R,R-IDS into R,S-IDS. Both enzymes do not require any cofactors. The two enzymes were purified and characterized, and the N-termini were sequenced. The purified lyase and also the epimerase catalyzed the transformation of alkaline earth metal-IDS complexes, while heavy metal-IDS complexes were transformed rather slowly or not at all. The observed mechanism for the complete mineralization of all IDS isomers involving an epimerase offers an interesting possibility of funneling all stereoisomers into a catabolic pathway initiated by a stereoselective lyase.

Synthetic d,l-aspartic-N-(1,2-dicarboxyethyl)tetrasodiumsalt, also known as sodium iminodisuccinate (IDS), belongs to the group of aminopolycarboxylate chelating agents. IDS is a medium-strong chelator that is able to replace EDTA when rather moderate chelating agents are sufficient for masking alkaline earth or heavy metal ions. The most commonly used chelator, EDTA, is of major concern because of its persistence in conventional wastewater treatment system. Because of its versatile use, large amounts of EDTA are readily released into the aquatic environment (5). Accumulated EDTA in water has the potential to remobilize toxic heavy metals in aqueous sediments or infiltration areas (15, 16). Thus, remobilization could cause heavy metal contamination in drinking water. Successful biodegradation of EDTA was observed only under laboratory conditions by special strains (12, 14, 24). For Fe³⁺-EDTA, the only known natural attenuation is photodegradation (10).

IDS as a substitute for EDTA is used in a variety of applications, including detergent formulations, corrosion inhibitors, production of pulp and paper, textiles, ceramics, photochemical processes, and as trace nutrient fertilizers in agriculture. It is effective as a bleaching agent stabilizer (H₂O₂), water softener, and deposit remover, with a superior ecological profile (2).

Sodium iminodisuccinate is produced from maleic anhydride, water, sodium hydroxide, and ammonia. This route yields a mixture of stereoisomers consisting of 25% S,S-IDS, 25% R,R-IDS, and 50% of the meso-form R,S-IDS (20). It is a so-called pentadentate chelating ligand. The chelation involves the four carboxylate groups and the nitrogen atom. It forms an octahedral complex conditional upon occupation of the sixth coordination position by a water molecule (3).

In the present work, we describe a two-enzyme system from Agrobacterium tumefaciens BY6 that catalyzes the key reaction in the degradation of all IDS isomers.

MATERIALS AND METHODS

IDS test samples and chemicals.

The tetrasodium iminodisuccinate isomer mixture, or technical IDS (Baypure CX 100, solid), contained 50% R,S-IDS, 25% S,S-IDS, and 25% R,R-IDS. The amount of iminodisuccinic acid sodium salt was at minimum 65%, and the solid content (sum of sodium salts) was at minimum 85%.

The isomeric purity of the S,S-IDS test sample was 100%, whereas the R,R-IDS sample was 94.2% pure (with 4.1% R,S-IDS and 1.7% fumaric acid). The R,S-IDS sample was 94.9% pure (containing 4% H₂O and 1.1% R,R-IDS plus S,S-IDS). The aqueous IDS stock solutions were neutralized to pH 7.0 with either hydrochloric acid or sodium hydroxide.

R,S-IDS, S,S-IDS, and R,R-IDS as well as technical IDS were gifts from Bayer AG (Leverkusen, Germany). All other chemicals were of analytical grade and were obtained from Aldrich (Steinheim, Germany), Fluka (Neu-Ulm, Germany), Merck (Darmstadt, Germany), Roth (Karlsruhe, Germany), or Sigma (Deisenhofen, Germany).

Biodegradation test.

The biodegradation test was performed as described in the Organization for Economic Cooperation and Development (OECD) guideline for testing chemicals by manometric respirometry test 301F (17). The test was carried out with the OxiTop Control system (WTW, Weilheim, Germany). The activated sludge inoculum was obtained from the Bürrig wastewater treatment plant (Leverkusen, Germany), which is a combined industrial-municipal plant. The chemical oxygen demand was determined with test system LCK 314 (Dr. Lange, Düsseldorf, Germany).

Bacterial strain and growth conditions.

A. tumefaciens BY6 was isolated from activated sludge of the Bürrig wastewater treatment plant and identified by 16S ribosomal DNA sequencing.

For isolation, growth, and batch experiments, a mineral salts medium without nitrogen was used, containing (per liter) 20 mg of Fe(III)-citrate, 1 g of MgSO₄ · 7H₂O, 50 mg of CaCl₂ · 2H₂O, and 0.1% (vol/vol) trace element solution (18). The mineral salts medium was buffered with 50 mM Na⁺/K⁺-phosphate buffer (pH 7.4) and contained 5 to 20 mM IDS as the sole source of carbon, nitrogen, and energy. Solid media were prepared by the addition of 1.5% (wt/vol) agar.

The cells were grown in baffled 3-liter Erlenmeyer flasks containing medium in volumes up to 30% of the flask volume on a rotary shaker at 125 rpm and room temperature (approximately 23°C). The cells were harvested by centrifugation immediately after complete consumption of IDS at an optical density of 1.5 to 2.5 at 546 nm. The pellet was washed twice with 0.05 M Tris-HCl (pH 8.0). Cells were frozen in liquid nitrogen and stored at −30°C until use.

Preparation of cell extract.

Frozen cell paste (15 to 30 g, wet weight) was thawed at room temperature and resuspended in 40 ml of 0.05 M Tris-HCl (pH 8.0). The cell suspension was treated with 30 mg of lysozyme (Fluka, Neu-Ulm, Germany) for 1 h at 23°C and thereafter overnight at 6°C. The cells were then disrupted by four passages through a French press (SLM Aminco, Urbana, Ill.) at 7 MPa. After the first passage, 10 mg of DNase I (ICN Biomedicals) was added. The suspension was kept on ice after each passage. Cell debris was removed by centrifugation at 100,000 × g and 4°C for 45 min. The resulting cell extract was filtered through a 0.22-μm-pore-size filter and finally used for enzyme assays and protein purification.

Purification of the carbon-nitrogen lyase (C-N lyase) and the IDS-epimerase.

The cell extract was dialyzed overnight at 6°C (dialysis tubing size 3; molecular weight cutoff, 12 to 14 kDa; Medicell Int. Ltd.) against 0.05 M Tris-HCl (pH 8.0) and then applied to an anion exchange column (Q Sepharose HR, 1.6 by 10 cm; Pharmacia, Uppsala, Sweden) preequilibrated with 0.05 M Tris-HCl buffer (pH 8.0) at a flow rate of 1.5 ml min⁻¹. In order to optimize the separation of the two enzymes on the anion exchange column, the purification was done by stepwise elution.

Purification of the C-N lyase.

A linear gradient (150 ml) of 0 to 0.75 M NaCl in 0.05 M Tris-HCl (pH 8.0) was used with a step at 0.325 M NaCl (8 ml). Fractions containing R,S-IDS-degrading activity were eluted at NaCl concentrations of 0.15 M to 0.3 M. The collected fractions were dialyzed as described above and loaded on a second anion exchange column (Mono Q HR, 1.6 by 10 cm; Pharmacia) preequilibrated with 0.05 M Tris-HCl buffer (pH 8.0) at a flow rate of 0.4 ml min⁻¹. The corresponding fractions were eluted from the column by a linear gradient (100 ml) of 0 to 1 M NaCl in 0.05 M Tris-HCl (pH 8.0) at NaCl concentrations of approximately 0.25 to 0.35 M. Ammonium sulfate was added to the combined fractions containing R,S-IDS-degrading activity to a final concentration of 1 M.

Precipitated proteins were removed by centrifugation (5,000 × g, 15 min), and the supernatant was applied to a Phenyl-Superose HR column (1.0 by 10 cm; Pharmacia) preequilibrated with 1 M ammonium sulfate in 0.025 M Tris-HCl (pH 8.5) at a flow rate of 0.75 ml min⁻¹. The enzyme was eluted by a linear gradient (80 ml) of 1 to 0 M ammonium sulfate in 0.025 M Tris-HCl (pH 8.5) at a concentration of 0.625 to 0.55 M. Fractions containing R,S-IDS-degrading activity were pooled and then concentrated with Vivaspin 2 and 6 concentrators (molecular weight cutoff, 10 kDa; Sartorius, Göttingen, Germany). Aliquots of 0.5 ml of the concentrated sample (4 ml) were applied on a gel filtration column (Superose 6 HR 1.0 by 30 cm; Pharmacia) preequilibrated with 2.5 mM K⁺/Na⁺-phosphate buffer (pH 7.5) at a flow rate of 1.0 ml min⁻¹. The collected samples were demineralized and thus prepared for loading on a hydroxyapatite column (Bio-Scale CHT-I, 5 ml; Bio-Rad Laboratories, Hercules, Calif.). Fractions containing R,S-IDS-degrading activity were eluted from the column by a linear gradient (50 ml) of 2.5 to 100 mM potassium phosphate buffer (pH 7.5) at salt concentrations of 50 mM.

Purification of the IDS-epimerase.

The IDS-epimerase was obtained from an anion exchange column (Q Sepharose HR, 1.6 by 10 cm; Pharmacia) by a linear gradient (230 ml) of 0 to 1 M NaCl in 0.05 M Tris-HCl (pH 8.0) with a step at 0.3 M NaCl (30 ml). IDS-epimerase activity was measured in the fractions with 0.4 to 0.5 M NaCl. The combined fractions were set to a final concentration of 1 M ammonium sulfate, precipitated proteins were removed by centrifugation (5,000 × g, 15 min), and the supernatant was loaded on a Phenyl-Superose HR column (1.0 by 10 cm; Pharmacia). The column preequilibration and the gradient were performed as described above for the R,S-IDS-degrading enzyme. The epimerase was eluted with 0.7 to 0.55 M ammonium sulfate.

All purification procedures were carried out at 6°C. The protein concentration was estimated by the method of Bradford (4) with bovine serum albumin (Fluka) as the standard. The purification procedure was monitored by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). The molecular mass of the subunit and the subunit composition of the enzyme were determined with a discontinuous SDS-10% polyacrylamide gel following the protocol of Laemmli (11). The proteins of the low-molecular-weight marker kit (Pharmacia) were used as standards. The gels were stained with Coomassie brilliant blue G-250. The native weights of the two enzymes were determined by analytical gel filtration (Superose 6 HR, 1.0 by 30 cm; Pharmacia). The calibration curve was prepared with the gel filtration calibration kit (Pharmacia) for high- and low-molecular-weight proteins containing the following protein standards: ferritin (450 kDa), catalase (232 kDa), aldolase (158 kDa), and bovine serum albumin (68 kDa). The isoelectric point was analyzed with a PhastSystem (Pharmacia) and PhastGel IEF 3-9 (Pharmacia).

Enzyme assays.

C-N lyase and IDS-epimerase activities were assayed by adding 5 to 10 μl of enzyme to 1.0 ml containing 5 mM R,S-IDS in 0.05 M Tris-HCl (pH 8.0) or 5 mM R,R-IDS in 0.05 M Tris-HCl (pH 8.5), respectively. Aliquots of 100 μl were taken, diluted with 10 μl of copper sulfate (200 mM, pH 1.3, adjusted with phosphoric acid) to terminate the reaction. The activities of IDS-epimerase and the C-N lyase were determined by measuring the disappearance of the appropriate substrate by ion pair chromatography. Enzyme assays were performed at 23°C. For amino acid analysis from the lyase reaction, the enzyme was inactivated by heating (3 min, 100°C) instead of acidification.

Determination of pH, temperature optima, and kinetic parameters.

The enzyme assays were performed as described above. The specific activity was measured at various pHs within the range from 5 to 10. A reaction buffer containing 50 mM Tris buffer (pH 7.5 to 9.0), sodium phosphate buffer (pH 5 to 7), and sodium carbonate buffer (pH 9.0 to 10) was used for the assay. For C-N lyase activity, 5 mM R,S-IDS was used as the substrate, and epimerase activity was tested with 5 mM S,S-IDS. The temperature optima of the two enzymes were determined at pH 8.5.

The kinetic parameters were determined at the pH optimum with various substrate concentrations from 0.25 to 12.5 mM. Data were fitted with the Michaelis-Menten equation, and V_max was deduced from the initially linear conversion rates. The enzyme assays were performed with 10 μl of purified C-N lyase and 5 μl of IDS-epimerase.

Analytical methods.

For separation of the IDS diastereomers and analysis of the metal-IDS complexes, ion pair chromatography was performed with a Purospher RP18 endcapped column (250 by 4.0 mm; Merck) and an isocratic eluent of 12.5% (vol/vol) methanol in formate buffer (15 mM sodium formate, 5 mM formic acid, and 2 mM tetrabutylammonium hydrogen sulfate). IDS diastereomers were measured as Cu²⁺ complexes at 240 nm. The metabolite fumaric acid was identified by cochromatography with an authentic sample at 215 nm. Activity towards different metal-IDS complexes was measured after conversion to the respective copper complexes as described previously (7).

The sum of the IDS isomers was measured by a rapid photometric test (7). d- and l-aspartic acid were separated and identified by precolumn derivatization with the FLEC/ADAM system [(+)-1-(9-fluorenyl)ethyl chloroformate/1-aminoadamantane] according to Einarsson and Josefsson (8) on a Grom-Sil FLEC 1 (250 by 4.0 mm; Grom, Herrenberg, Germany). Ammonia concentrations were determined spectrophotometrically by use of a commercial test system (Microquant 14750; Merck, Darmstadt, Germany). The dissolved organic carbon was measured with the micro N/C analyzer (IDC, Langewiesen, Germany).

Quantification of S,S-IDS and R,R-IDS in equilibrium.

The equilibrium composition of the three IDS isomers was determined with the help of a Ralstonia sp. strain SLRS7 cell extract (7). Different concentrations of each IDS isomer (5 to 18.5 mM) were prepared as described above with 10 μl of purified epimerase. Aliquots were taken and measured until the equilibrium was established, approximately 18 to 24 h. Then the assay mixture (500 μl) was heated for 10 min at 100°C to inactivate the IDS-epimerase activity and incubated with the cell extract until R,R-IDS was completely consumed. The cell extract from Ralstonia sp. strain SLRS7 is able to transform the S,S-IDS and R,S-IDS isomers, whereas R,R-IDS remains intact. The S,S-IDS concentration in equilibrium was calculated from the difference between the starting concentration of R,R-IDS plus S,S-IDS and the R,R-IDS remaining after enzymatic treatment.

Conversion of IDS by whole cells.

A. tumefaciens BY6 was grown in mineral salts medium with technical IDS (5 mM) and harvested immediately after complete IDS consumption. The cells were resuspended in phosphate buffer (50 mM; pH 7.4) to an optical density of 3.5 at 546 nm and incubated with a mixture of 7.5 mM R,S-IDS and 7.1 mM S,S-IDS at 23°C on a rotary shaker (125 rpm). The cell suspension was divided into two samples when the maximum rate of R,R-IDS consumption was observed. One sample was incubated further until all IDS was consumed. The other sample was supplemented with R,S-IDS.

Analysis of protein sequences.

Purified proteins, approximately 500 pmol, were blotted onto a polyvinylidene difluoride membrane (ProSorb; Applied Biosystems, Weiterstadt, Germany) and subjected to automatic sequencing (476A Protein Sequencer; Applied Biosystems). Database searches were performed with BLAST (1).

Nucleotide sequence accession number.

The sequence will appear in the GenBank nucleotide sequence database under accession number AY576874.

RESULTS

Biodegradation tests.

A standardized OECD test (301F) was carried out to analyze the biodegradability of the three IDS isomers, technical-grade IDS, and the metal-IDS complexes. As shown in Fig. 1, biodegradation of all IDS isomers and technical IDS attained approximately the level of the reference compound sodium benzoate. Interestingly, for the R,R-IDS isomer, a significant degradation rate was not observed before day 3.

FIG. 1.

Manometric respirometry test for biodegradation of technical IDS and the three isomers R,R-IDS, S,S-IDS, and R,S-IDS with activated sludge inoculum from the wastewater treatment plant in Bürrig (Leverkusen, Germany) as inoculum. Sodium benzoate was used as reference compound. Data are expressed as a percentage of the theoretical O₂ consumption for complete oxidation.

The pass level for ready biodegradability is reached at 60% of the theoretical O₂ consumption in a 10-day window within the 28-day period of the test. The 10-day window begins when the degree of biodegradation has reached 10% of the theoretical O₂ consumption and must end before day 28 of the test. Therefore, all IDS isomers and technical IDS (Baypure CX 100) were characterized as readily biodegradable substrates.

For the biodegradation test with metal-IDS complexes, equimolar amounts of each metal and IDS were used. The test was monitored over a period of 42 days. Figure 2 shows that Fe²⁺-IDS and Ca²⁺-IDS were readily biodegradable, whereas Mn²⁺-IDS and Cu²⁺-IDS revealed 55 and 40% biodegradation after 28 days, respectively. During the test on IDS chelates, precipitation of the corresponding heavy metal hydroxide was observed, indicating liberation of free metal ions.

FIG. 2.

Manometric respirometry test for biodegradation of the metal-IDS complexes Ca²⁺-IDS, Fe²⁺-IDS, Cu²⁺-IDS, and Mn²⁺-IDS with activated sludge from the wastewater treatment plant in Bürrig (Leverkusen, Germany). Sodium benzoate was used as a reference compound. Data are expressed as a percentage of the theoretical O₂ consumption for complete oxidation.

Isolation and identification of microorganisms.

The IDS-degrading microorganism in this study was isolated from activated sludge. It was enriched in batch culture with buffered mineral salts medium supplemented with 5 mM R,R-IDS as the sole source of carbon, nitrogen, and energy. The enrichment cultures were incubated in a baffled Erlenmeyer flask on a rotary shaker at 125 rpm and room temperature. The cultures were subcultivated daily for 5 days. Subsequently, the cultures were repeatedly transferred on agar plates also containing 5 mM R,R-IDS until a pure culture (BY6) was isolated. Strain BY6 grew with R,R-IDS as the sole source of carbon, nitrogen, and energy. It was gram negative and oxidase positive and formed motile cocci of 1 to 1.5 μm. Furthermore, strain BY6 was identified by 16S ribosomal DNA sequencing. Sequence alignment with BLAST (1) achieved highest homologies (99%) with A. tumefaciens (accession number D14500) and A. tumefaciens strain CIP111-78 (AJ389897). Thus, the organism was named A. tumefaciens BY6.

Growth of A. tumefaciens BY6 on IDS.

A. tumefaciens BY6 grew in mineral salts medium with each of the three IDS isomers as the sole source of carbon, nitrogen, and energy. The doubling times on R,S-IDS, S,S-IDS, and R,R-IDS at 23°C were 3, 3.3, and 3.8 h, respectively. During growth on R,S-IDS, R,R-IDS, and S,S-IDS, the pH increased as a result of ammonia excretion by the cells. The amount of ammonia released into the medium after 5 days corresponded to approximately 30% ± 5% of the nitrogen contained in IDS (data shown for R,R-IDS in Fig. 3).

FIG. 3.

Growth of A. tumefaciens BY6 in 500-ml Erlenmeyer flasks with 100 ml of mineral salts medium containing 4.1 mM R,R-IDS (•) as the sole carbon, nitrogen, and energy source. The culture was inoculated with a 1% (vol/vol) preculture. The incubation temperature was 23°C on a rotary shaker at 125 rpm. Shown are optical density (○), pH (▵), and ammonia (⧫).

In addition, growth of A. tumefaciens BY6 was also possible on different metal-IDS chelates, namely, Mg²⁺-, Ca²⁺-, Fe²⁺-, Cu²⁺-, and Mn²⁺-IDS. The complexes with the metal ions were generated by adding their sulfate salts in equimolar amounts to technical IDS. The degradation of metal-IDS complexes was monitored by measuring dissolved organic carbon. Mg²⁺- and Ca²⁺-IDS complexes were degraded completely within 3 days, while only 75% of Fe²⁺-IDS was mineralized in this time span. Complete mineralization of Fe²⁺-IDS and about 75% mineralization of Mn²⁺-IDS were achieved within 10 days. In contrast, only about 25% of Cu²⁺-IDS was mineralized within 10 days (data not shown). Metal hydroxide precipitation was noted during growth on Fe²⁺-, Cu²⁺-, and Mn²⁺-IDS. These results correspond with the observations made during biodegradation tests with metal-IDS complexes.

In order to find out whether different isomers of IDS are simultaneously or sequentially used as growth substrates, A. tumefaciens BY6 was incubated with R,R-IDS plus R,S-IDS. As shown in Fig. 4, these two isomers were not simultaneously consumed during the exponential growth phase. Significant decrease of R,R-IDS did not start until almost all of the R,S-IDS was consumed.

FIG. 4.

Growth of A. tumefaciens BY6 in 500-ml Erlenmeyer flasks with 100 ml of mineral salts medium containing 4.7 mM R,R-IDS (•) plus 5.3 mM R,S-IDS (□) as the sole source of carbon, nitrogen, and energy. The culture was inoculated with a 1% (vol/vol) preculture grown on technical IDS and incubated on a rotary shaker (125 rpm) at 23°C. Shown are optical density (○) and pH (▵).

Turnover experiments with whole cells of A. tumefaciens BY6 incubated with a mixture of R,S-IDS and R,R-IDS showed that the rate of R,R-IDS consumption depended on the R,S-IDS concentration present (Fig. 5). If R,S-IDS was added during the maximum rate of R,R-IDS turnover (340 min), R,R-IDS turnover stopped immediately and started again after the R,S-IDS concentration decreased to less than 1 mM. The same results were observed with S,S-IDS instead of R,R-IDS (data not shown). The growth and turnover experiments indicate that R,S-IDS is preferred over R,R-IDS and S,S-IDS as a substrate of A. tumefaciens BY6.

FIG. 5.

Competitive consumption of R,S-IDS and R,R-IDS by whole cells of A. tumefaciens BY6. Whole cells were obtained by growth in mineral salts medium with technical IDS (5 mM). The cells were harvested, resuspended in phosphate buffer (optical density at 546 nm of 3.5), and incubated at 23°C with 7.5 mM R,S-IDS plus 7.15 mM R,R-IDS on a rotary shaker at 125 rpm. When the maximum rate of R,R-IDS degradation was observed, the cell suspension was split into two samples with the same volume. One sample was incubated further (dashed line). The other sample was amended with R,S-IDS.

C-N lyase from A. tumefaciens BY6 that cleaves R,S-IDS.

The enzyme was purified in order to analyze the initial transformation of IDS. From the whole-cell experiments, it was assumed that IDS degradation was based on a single enzyme with stereoselective preference for R,S-IDS. During the protein purification assays, it turned out that initial S,S-IDS and R,R-IDS degradation required two distinct enzymes, whereas for R,S-IDS the lyase activity was necessary. In the first step, an R,S-IDS-degrading enzyme was purified. The purification procedure is summarized in Table 1. The overall yield during purification was 30.5% and resulted in a 14.1-fold increase in specific activity. The purified enzyme did not require any cofactors and transformed R,S-IDS to fumaric acid and d-aspartic acid. l-Aspartic acid was not detected. These metabolites indicated that the purified R,S-IDS-degrading enzyme is a C-N lyase. It catalyzes a nonhydrolytic cleavage of the C-N bond, generating succinate plus aspartate.

TABLE 1.

Purification of the R,S-IDS-cleaving C-N lyase from A. tumefaciens BY6

Purification step	Vol (ml)	Total amt of protein (mg)	Sp acta (μmol of R,S-IDS min−1 [mg of protein]−1)	Yield (%)	Purification (fold)
Crude extract	45	572	1.15	100	1
Ultracentrifugation	35	221	1.54	51.7	1.3
Q-Sepharose HR 16/10	28	26.6	7.4	29.9	6.4
Phenyl-Superose HR 10/10	11.5	3.7	16.2	30.5	14.1

^aThe specific activity was determined by measuring the disappearance of R,S-IDS.

The C-N lyase also catalyzed the reverse reaction. The equilibrium constant K_eq was 90 × 10⁻³ for R,S-IDS (data not shown). The equilibrium constant was determined according to the equation K_eq = c(fumaric acid)²/c(R,S-IDS), where c is concentration. Hence, the equilibrium for the reaction is clearly in favor of R,S-IDS cleavage.

Surprisingly, the C-N lyase showed no activity against the R,R-IDS isomer. Only minor activity with S,S-IDS was observed (Table 2), generating l-aspartic acid and fumaric acid as products.

TABLE 2.

Activity of the C-N lyase with various substrates^a

Substrate	Activity (μmol of S,S-IDS/min/mg)	Relative activity
R,S-IDS	10.6	100
S,S-IDS	0.49	2.5
R,R-IDS	0	0
Ca2+-R,R-IDS	4.2	40
Mg2+-R,S-IDS	1.85	17
Mn2+-R,S-IDS	0.81	2.9
Fe3+-R,S-IDS	0	No transformation
Fe2+-R,S-IDS	0	No transformation
Zn2+-R,S-IDS	0	No transformation
Cu2+-R,S-IDS	0	No transformation
S,S-EDDS	0	No transformation
Technical EDDS	0	No transformation
dl-Aspartic acid	0	No transformation
Nopaline	0	No transformation

^aEnzyme assays were performed with 5 mM as described in Materials and Methods. The specific activities were determined by measuring the disappearance of the substrate.

Epimerase from A. tumefaciens BY6.

Further investigations were performed to find an R,R-IDS-degrading enzyme activity. Obviously, the key enzyme for R,R-IDS degradation is an epimerase. It was purified by applying several chromatographic steps, as reported in Table 3. The enzyme was purified 12.7-fold, with a yield of 22.1%.

TABLE 3.

Purification of the IDS-epimerase from A. tumefaciens BY6

Purification step	Vol (ml)	Total amt of protein (mg)	Sp acta (μmol of R,R-IDS min−1 [mg of protein]−1)	Yield (%)	Purification (fold)
Crude extract	46	1139
Ultracentrifugation	35	812
Q-Sepharose HR 16/10	29.5	284.4	2.0	100.0	1.0
Mono Q HR 16/10	11.5	89.7	5.0	79.4	2.5
Phenyl-Superose HR 10/10	20	17	13	39.4	6.6
Superdex 200 HiLoad 16/60	4	9.6	19	32.6	9.6
Hydroxyapatite cHT5-I	5	4.95	25	22.1	12.7

^aThe specific activity was determined by measuring the disappearance of R,R-IDS.

The purified IDS-epimerase did not require any cofactors to transform R,S-IDS into S,S-IDS and R,R-IDS and vice versa. The percentage of each IDS isomer in equilibrium at pH 8.5 was 55.5% R,S-IDS, 27.8% S,S-IDS, and 16.7% R,R-IDS.

Cell extracts of A. tumefaciens BY6 cells grown with nutrient broth medium as the sole source of carbon, nitrogen, and energy exhibited no IDS-epimerase or C-N lyase activity at all, indicating the inducibility of these enzymes.

Characterization of the C-N lyase.

The purified C-N lyase gave a single band by SDS-PAGE and had a molecular mass of approximately 57 kDa. By analytical gel filtration, the molecular mass of the native enzyme was determined to be 260 kDa. Therefore, it can be assumed that the C-N lyase is a homotetramer. N-terminal amino acid sequencing by automated Edman degradation revealed the sequence MRERLSASPNELIVKHLIGPRLFGNLDRDFLEM(X/S)KVN. A BLAST (1) search showed no significant similarity to any known protein sequences. The purified enzyme exhibited a typical protein absorption spectrum with a maximum at 280 nm. The pH optimum for lyase activity was found to be pH 8.0 to 8.5. No activity was left at pH 5.5 (Na⁺-phosphate buffer) and pH 9.5 (Na⁺-carbonate and Tris buffer). The enzyme showed maximum activity between 35 and 40°C, whereas no activity was observed at 60°C. The isoelectric point was found to be 5.95. The kinetic data for R,S-IDS at pH 8.5 were K_m, 1.0 ± 0.2 mM; V_max, 170 μmol/min; k_cat, 228 s⁻¹; and k_cat/K_m, 2.3 × 10⁶.

The C-N lyase activity with S,S-IDS as the substrate was only 2.5% of the activity with R,S-IDS (Table 2). Therefore, the enzyme kinetic parameters for S,S-IDS were not determined.

Some metal-R,S-IDS complexes were transformed by the C-N lyase: Ca²⁺-R,S-IDS was transformed with the highest specific activity, followed by Mg²⁺-R,S-IDS and Mn²⁺-R,S-IDS, whereas Fe²⁺-, Fe³⁺-, Zn²⁺-, and Cu²⁺-R,S-IDS were not transformed at all.

Other substrates such as aspartic acid racemate, the aminopolycarboxylate ethylenediaminedisuccinate (EDDS), and the opine N,N-(1,3-dicarboxypropyl)-l-arginine (nopaline) were not transformed by the C-N lyase (Table 2).

Characterization of the IDS-epimerase.

The molecular mass of native IDS-epimerase calculated by gel filtration was 170 kDa. The SDS-PAGE analysis gave a single protein band corresponding to a molecular mass of 48 kDa. The purified enzyme showed a typical protein absorption spectrum with a maximum at 280 nm. The N-terminal amino acid sequence of IDS-epimerase was determined to be MFTTKLAEKVVSAWKAKISQPALKAAQD. By protein sequence alignment with BLAST, no significant similarity to any known protein sequences was found.

The optimal pH for IDS-epimerase activity was 8.5, with a remaining activity of 87, 44, and 8% at pH 8.0, 7.5, and 9.0, respectively. The temperature optimum was 35°C, whereas no significant activity was observed at temperatures above 50°C. The pI was 6.3. The kinetic parameters are presented in Table 4.

TABLE 4.

Kinetic parameters of the IDS-epimerase^a

Substrate	Km (mM)	Vmax (μmol/min)	kcat (s−1)	kcat/Km (s−1 M−1)
S,S-IDS	1.5 (0.3)	250 (20)	139	93
R,R-IDS	1.3 (0.3)	130 (10)	72	55
R,S-IDS	1.1 (0.2)	30 (5)	17	15

^aThe values in parentheses are standard deviations, and the pH was 8.5.

Comparing the IDS isomers, S,S-IDS was transformed with the highest specific activity, twice as high as that of R,R-IDS and 10 times faster than that of R,S-IDS. Only the alkaline earth metal Ca²⁺-, Ba²⁺-, and Mg²⁺-R,R-IDS complexes showed moderate activity, with 52, 44, and 25% of the activity with S,S-IDS, respectively. The epimerase transformed Ca²⁺-R,R-IDS and R,R-IDS with similar activities. The other metal-IDS complexes were transformed at very slow rates or not at all (Table 5). EDDS isomers were not epimerized by the enzyme.

TABLE 5.

Activity of the IDS-epimerase with IDS, different metal-IDS complexes, EDDS, and aspartic acid

Substrate	Activitya (μmol of S,S-IDS/min/mg)	Relative activity
S,S-IDS	46	100
R,S-IDS	4.88	9
R,R-IDS	24.5	53
Ca2+-R,R-IDS	24	52
Ba2+-R,R-IDS	20.1	44
Mg2+-R,R-IDS	11.3	25
Mn2+-R,R-IDS	1 × 10−4	Negligible
Fe3+-R,R-IDS	3 × 10−4	Negligible
Fe2+-R,R-IDS	4.2 × 10−4	Negligible
Zn2+-R,R-IDS	0	No transformation
Cu2+-R,R-IDS	0	No transformation
S,S-EDDS	0	No transformation
Technical EDDS	0	No transformation
l-Aspartic acid	0	No transformation

^aEnzyme assays were performed with 5 mM substrate as described in Materials and Methods. The specific activities were determined by measuring the disappearance of the substrate.

DISCUSSION

The IDS isomers of the technical mixture are readily biodegradable, according to OECD guideline 301F. Even some metal-IDS complexes fulfill the criteria of ready biodegradability, such as Ca²⁺- and Fe²⁺-IDS. In contrast, Mn²⁺- and Cu²⁺-IDS revealed only 55 and 40% biodegradation after 28 days, respectively. Interestingly, the Mn²⁺- and Cu²⁺-IDS degradation rates were initially similar to these of Ca²⁺- and Fe²⁺-IDS (Fig. 2). We assume that prolonged exposure to the released Cu²⁺ and Mn²⁺ ions led to inactivation of the microorganisms.

A correlation between complex formation constants (pK) and degradation rates was not observed. The pK values of Ca²⁺-, Mn²⁺-, Fe²⁺-, and Cu²⁺-IDS were 6.7, 7.3, 8.2, and 14.3, respectively (3). We presume that metal-IDS complexes were not transported into the cells, since heavy metal precipitation was observed in the medium during growth and biodegradation tests. Interestingly, the C-N lyase from A. tumefaciens BY6 showed no activity against Fe²⁺-R,S-IDS, although strain BY6 degraded Fe²⁺-R,S-IDS. This supports the assumption that metal-IDS complexes were not taken up by the cells. Witschel et al. (26) reported EDTA uptake by strain DSM 9103. Only free EDTA and complexes with a low complex formation constant, such as the alkaline earth metal chelates, were taken up, whereas heavy metal EDTA complexes remained outside the cells.

The initial transformations of the three IDS isomers by A. tumefaciens BY6 are summarized in Fig. 6. A. tumefaciens BY6 caused a pH increase during growth on IDS. The metabolite ammonia was detected and should be a deamination product of aspartic acid, probably cleaved by an l-aspartate ammonia-lyase. d-Aspartate ammonia-lyases are still unknown, and therefore d-aspartic acid as a product of R,S-IDS degradation must be transformed into l-aspartic acid, probably by an aspartate racemase, as described by Rahmanian et al. (19).

FIG. 6.

Initial transformation of IDS isomers by A. tumefaciens BY6 involving an IDS-isomerase and a C-N lyase. Further metabolism by an aspartate racemase and an l-aspartase is assumed from the known metabolism of aspartate. The dashed arrow documents side activity of the C-N lyase.

The isolated R,S-IDS-degrading enzyme, classified as a C-N lyase (EC 4.3.3), had a clear preference for R,S-IDS and only minor activity for S,S-IDS (Table 2). The C-N lyase cleaved only the S configuration of IDS and required the R configuration at the other asymmetric carbon atom of IDS.

A similar IDS-cleaving C-N lyase from Ralstonia sp. strain SLRS7 has been described previously (7). This enzyme also cleaves only the S configuration of IDS but has equal specific activity towards S,S-IDS and R,S-IDS. The C-N lyases from Ralstonia sp. strain SLSR7 and A. tumefaciens BY6 resemble each other in molecular mass, subunit organization, pH and temperature optima, K_m, equilibrium constant K_eq, and activity towards different metal-IDS complexes. The R,R-IDS isomer was not cleaved by either C-N lyase. Remarkably the S,S isomer of EDDS, another important substitute of EDTA, was only transformed by the Ralstonia sp. strain SLRS7 C-N lyase. Since the R,R-IDS isomer can be degraded and used by A. tumefaciens BY6, an additional activity, an IDS-epimerase, is necessary. This activity transforms R,R-IDS and S,S-IDS into the C-N lyase substrate R,S-IDS.

Many epimerases, especially racemases, require pyridoxal 5′-phosphate (PLP) as a cofactor when racemization proceeds through the formation of aldimine Schiff base between the substrate amino acid and PLP (22). Since the spectrum of the purified IDS-epimerase shows no significant absorption in the range of 350 to 450 nm, there is no indication of PLP as a cofactor. Many other amino acid epimerases or racemases such as diaminopimelic acid epimerase (23), 4-hydroxyproline epimerase (9), proline racemase (6, 21), aspartate racemase (27), and glutamate racemase (13) also require no cofactor. We assume that enzyme-catalyzed epimerization of IDS is accomplished by a proton abstraction and subsequent return by a two-base mechanism, as proposed for PLP-independent epimerases and racemases (21, 23, 27). In a two-base mechanism, one enzyme base removes the proton from the substrate, and the conjugate base delivers a proton in the opposite direction (6).

The observed mechanism for the complete mineralization of all IDS isomers involving an epimerase offers an interesting possibility of circumventing an initial stereoselective catabolic step, as described for C-N lyases involved in IDS and EDDS degradation (7, 25).