INTRODUCTION

Top Abstract Introduction Materials and Methods Results and Discussion References

The metabolism of 2,4-dichlorophenoxyacetate (2,4-D) has been well described for both the enzymatic steps (9) and their genetic basis (20, 23) as studied, for example, with Ralstonia eutropha (Alcaligenes eutrophus) JMP 134(pJP4). Although the first enzymatic step in this strain, encoded by the gene tfdA, was previously thought to be carried out by a monooxygenase-type reaction, it has since been shown to be catalyzed by an -ketoglutarate-dependent dioxygenase: oxidative decarboxylation of this electron acceptor results in the formation of succinate, whereas 2,4-D is oxidatively cleaved to 2,4-dichlorophenol (DCP) and glyoxylate (7). TfdA-like activities for the degradation of phenoxyalkanoates are widespread in proteobacteria of the -subgroup (8, 26, 28); often harbored on plasmids (5), they have also been found on the chromosome (10, 24). In addition, however, 2,4-D-utilizing bacteria have been isolated in which the tfdA sequence could not be identified (10, 11, 14) by applying the primers from the conserved region of this gene (28). Although this does not necessarily contradict the enzyme functioning according to the above-mentioned mechanism, the monooxygenase-catalyzed cleavage of 2,4,5-trichlorophenoxyacetate was demonstrated in the case of Burkholderia cepacia AC 1100 (30). In addition to phenoxyacetates, cleavage by an -ketoglutarate-dependent dioxygenase has also been described for other phenoxyalkanoates. In the case of Sphingomonas herbicidovorans (19, 31), Rhodoferax sp. strain P230 (6), and Comamonas acidovorans MC1 (18), this has been shown for the degradation of phenoxypropionates such as (RS)-2-(2,4-dichlorophenoxy)propionate (2,4-DP). The general equation for this oxidative etherolytic step is presented in Fig. 1.

View larger version (4K): [in this window] [in a new window]   FIG. 1.   Diagram of oxidative etherolytic step of phenoxyalkanoates. R1 = H or CH3; R2 = Cl or CH3; n = 0 or 2 in respective combinations for the usual herbicides.

The conversion of phenoxyalkanoic acids via an -ketoglutarate-dependent dioxygenase thus plays an important role in the degradation of various herbicides. The initiation of degradation via this step requires -ketoglutarate for continual substrate utilization. The consequences for substrate consumption and growth are investigated here in a theoretical study. Emphasis is placed on strains which only exhibit etherolytic activities by forming the phenolic intermediate as a dead-end product. Such strains are known (cf. references 15 and 27). The oxidized alkanoic acids are the only sources in these cases for both regeneration of the electron acceptor from succinate and growth. The efficiencies and limits of carbon conversion under these conditions are calculated and compared to the productive degradation of (chlorinated) phenolic compounds as described elsewhere (16).

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results and Discussion References

The calculations were performed on the basis of the Y_ATP concept (1, 22), taking into account a biomass synthesis efficiency of 10.5 g of bacterial dry mass/mol of ATP (3). Biomass synthesis was assumed to start from a central carbon precursor (29), which is considered to be 3-phosphoglycerate (PGA). Consequently, primary assimilatory pathways were formulated up to the formation of this metabolite. The average biomass "molecule" was taken as C₄H₈O₂N₁. Energy equivalents were synthesized with an efficiency of (P/O) ATP from NAD(P)H and with (P/O  1) ATP from FADH₂ (for further details, see references 2 and 3).

	RESULTS AND DISCUSSION

Top Abstract Introduction Materials and Methods Results and Discussion References

Complete degradation. The balances of carbon metabolism, including primary assimilation, carbon precursor, and the biomass synthesis of various phenoxyalkanoic acids, taking into account the -ketoglutarate-dependent mechanism of cleavage of the ether bond of the substrates, are listed in Tables 1 to 3. The degradation of 2,4-D by a monooxygenase-based type of reaction is shown for comparison (for other phenoxyalkanoates, see reference 15). Accordingly, assimilation of 2,4-D via -ketoglutarate-dependent dioxygenase has pronounced effects on carbon assimilation. In the primary assimilatory route, as much as 50% of carbon was lost; 1 CO₂ resulted from the cleavage of -ketoglutarate (-KG), and three more were derived from the regeneration of the acceptor via glyxoylate (Glyox) and citrate cycle (Fig. 2). A further CO₂ was liberated on the way to the precursor synthesis, meaning that a total of 62.5% of the carbon of the substrate appeared as CO₂ in the assimilatory branch and only 37.5% remained for biomass synthesis (Fig. 2 and Table 4). By contrast, in the case of assimilation via a monooxygenase, 75% of the carbon was available for assimilatory purposes. 4-(2,4-dichlorophenoxy)butyrate (2,4-DB) (Fig. 3) and 2,4-DP assimilated via an -ketoglutarate-dependent mechanism also resulted in a loss of carbon, although the relative amounts were lower and reached 33 to 40% after the synthesis of PGA (Table 4). In comparison, a monooxygenase-catalyzed initial step of cleavage would preserve 75 to 83% of the carbon up to this metabolite. Since no further loss of carbon was taken into account in biomass synthesis (3), the carbon conversion efficiency in the precursor synthesis corresponds to the maximum carbon conversion efficiency which can possibly be attained on these substrates (Table 4). The increase in oxidative decarboxylation is accompanied by a rise in the formation of energy equivalents (Table 2), reducing the portion of substrate dissimilated only for the purposes of energy generation. This contrasts with the pattern of a monooxygenase-mediated substrate cleavage, which is characterized by the pronounced expenditure of energy equivalents in the primary assimilation and precursor synthesis, increasing the dissimilatory portion of the substrate (Tables 1 and 2), which is satisfied by increased dissimilation. In an overall balance of the mass and energy flows via the assimilatory and the dissimilatory routes, the yield coefficients consequently became similar. The figures are lower by only 2 to 13% in the case of the -ketoglutarate-dependent version in comparison to a monooxygenase-catalyzed reaction (16). The liberation of HCl was not shown in the balance equations.

View larger version (18K): [in this window] [in a new window]   FIG. 2.   Scheme of the synthesis of PGA during complete degradation of 2,4-D by including the regeneration of -KG. Abbreviations: AcCoA, acetyl-CoA; OAA, oxaloacetic acid.

View this table: [in this window] [in a new window]   TABLE 4.   Carbon conversion efficiency (CCE) with monooxygenase (M)- and dioxygenase (-KG)-type reactionsa

View larger version (19K): [in this window] [in a new window]   FIG. 3.   Scheme of the synthesis of PGA during complete degradation of 2,4-DB by including the regeneration of -KG. Abbreviation: AcCoA, acetyl-CoA.

Incomplete degradation. Microbial strains which are characterized by incomplete metabolism for herbicide degradation, i.e., which harbor cleaving activity (TfdA or TfdA analogs) but lack at least TfdB ([chloro]phenolhydroxylase) are faced with more drastic consequences. The phenolic intermediate, e.g., 2,4-dichlorophenol, which via the formation of succinate (Succ) and acetyl coenzyme A (acetyl-CoA) would otherwise be a potential source for the regeneration of -ketoglutarate (Fig. 2 and 3), cannot be metabolized, i.e., it remains dead-end product. Strains of the case considered only dispose of glyoxylate (phenoxyacetates), pyruvate (phenoxypropionates), or succinate semialdehyde (putatively from phenoxybutyrates), in addition to succinate resulting from the cleavage of -ketoglutarate, for the regeneration of this acceptor and as a source of carbon and energy for growth. This strongly restricts metabolic flexibility, as can be seen in Table 5. In the case of 2,4-D, the regeneration of -ketoglutarate from succinate and glyoxylate, most likely via glyoxylate carboligase, is only possible at all if the loss of carbon in this step is compensated for by tapping additional carbon via the carboxylation of C₃ compounds (Fig. 4). Regeneration via this route includes the cleavage of malate resulting in the formation of acetyl-CoA and glyoxylate, which could be catalyzed by malate thiokinase-malyl-CoA lyase. From a stoichiometric point of view, the cleavage of 2,4-D could thus at best be maintained, assuming that carboxylation reactions of C₃ compounds work adequately. From a physiological point of view, quantitative regeneration is rather improbable since other reactions make use of the pool of these intermediates and thus waste carbon by forming CO₂. Under no circumstances, however, will growth be possible under these conditions. 2,4-D would prove inert despite the presence of TfdA.

View this table: [in this window] [in a new window]   TABLE 5.   Carbon balance after regeneration of -KG in the cases of incomplete degradation of herbicides (with DCP as dead-end product)a

View larger version (19K): [in this window] [in a new window]   FIG. 4.   Carbon flow during incomplete degradation of 2,4-D by forming DCP as a dead-end product. Abbreviations: AcCoA, acetyl CoA; OAA, oxaloacetic acid.

View larger version (15K): [in this window] [in a new window]   FIG. 5.   Carbon flow during incomplete degradation of 2,4-DB by forming DCP as a dead-end product (malic enzyme [C3-type] version). Abbreviations: AcCoA, acetyl-CoA.

View larger version (20K): [in this window] [in a new window]   FIG. 6.   Carbon flow during incomplete degradation of 2,4-DB by forming DCP as a dead-end product (malate thiokinase-malyl-CoA [C4-type]). Abbreviation: AcCoA, acetyl-CoA.

Examples. That these restrictions indeed occur became evident from various examples. For instance, this effect can be unequivocally attributed to a derivative of the strain A. eutrophus (R. eutropha) JMP 228, which has been described as only exhibiting TfdA activity but did not further metabolize the chlorophenolic intermediates (27). It became apparent from the data presented here that cleavage of 2,4-D proceeded only as long as pyruvate was present in the growth medium; after pyruvate exhaustion, 2,4-D did not undergo further degradation. Other indications of a mechanism as proposed here were obtained from phenomena observed with noninduced cells of Rhodoferax sp. strain P230 (6) and Comamomas acidovorans MC1 (18). These strains exhibit -ketoglutarate-dependent activities for the degradation both of the chiral phenoxypropionate herbicides dichlorprop (2,4-DP) and mecoprop [(RS)-2-(4-chloro-2-methylphenoxy)propionate] and of phenoxyacetate herbicides 2,4-D and 4-chloro-2-methylphenoxyacetate. The initial cleavage reaction was constitutively expressed with respect to the R and S enantiomers of 2,4-DP in the case of strain MC1. Incubation in the presence of streptomycin resulted in a partial to complete degradation of the enantiomers with the pure and the racemic substrates. By contrast, 2,4-D remained almost unutilized under these conditions. This was, however, not caused by an enzymatic deficit. The application of an external (auxiliary) source of carbon, e.g., yeast extract, -ketoglutarate, fructose, or lactate, resulted in the immediate degradation of these compounds accompanied by the liberation of DCP in approximately equimolar concentrations, a finding which is indicative of a shortage of metabolites required for the regeneration of -ketoglutarate. Similar effects were observed with Rhodoferax sp. strain P230 with respect to the R enantiomer. Such metabolic background is also discussed in terms of the properties of strains of Aureobacterium sp. and Rhodococcus erythropolis which were only able to cleave phenoxybutyrate herbicides but metabolized the resulting chlorophenolic intermediates either very slowly (Rhodococcus) or not at all (Aureobacterium) (15). It was found that these strains could not be continuously cultivated on 2,4-DB, even in the presence of a second strain productively eliminating the toxic compound DCP (Ochrobactrum anthropi K2-14 [17] ) (H. Mertingk, personal communication), since the complete loss of carbon from C₄ moieties can be easily explained. However, since the enzymatic mechanism of this cleavage reaction in these strains has not yet been elucidated (activity is completely lost after the disintegration of the cells), this lack of growth on these substrates cannot be unambiguously attributed to a mechanism as presently suggested. It should be mentioned in this context that the application of primers derived from the conserved region of the tfdA gene (28) did not result in respective amplification products in PCR (S. Kleinsteuber and D. Hoffmann, personal communication).

Conclusion. The mechanisms discussed here might be responsible for an apparent recalcitrance of phenoxyalkanoic acid herbicides. This especially holds for bacterial consortia in which the essential degradative steps were associated with different strains. The latter was shown in a global survey on 2,4-D degrading activity to apply to most biotopes (25) and has most often been found to be the case with phenoxypropionate herbicides as well (12, 13, 21). To conclude, these results demonstrate that any evidence of nondegradative potentials in given biotopes and under respective assay conditions need not necessarily be an indication of enzymatic deficits but may well be due to general metabolic reasons.