Cross-Coupling of Sulfonamide Antimicrobial Agents with Model

(23) Advanced Chemistry Development (ACD) Software Solaris V4.67;. ACD/Labs: Toronto, Canada, 1994-2004. (24) Shah, N. H.; Lazarus, J. H.; Sheth, P. R...
0 downloads 0 Views 648KB Size
Download PDF
Environ. Sci. Technol. 2005, 39, 4463-4473

Cross-Coupling of Sulfonamide Antimicrobial Agents with Model Humic Constituents HEIDI M. BIALK,† ANDRE ´ J. SIMPSON,‡ AND J O E L A . P E D E R S E N * ,†,§,| Molecular and Environmental Toxicology Center, University of Wisconsin, Madison, Wisconsin 53706, Department of Chemistry, University of Toronto at Scarborough, Toronto M1C MA4, Canada, Environmental Chemistry and Technology Program, University of Wisconsin, Madison, Wisconsin 53708, and Department of Soil Science, University of Wisconsin, Madison, Wisconsin 53706

The oxidative cross-coupling of sulfonamide antimicrobials to constituents of natural organic matter was investigated. Sulfonamide antimicrobials were incubated with surrogate humic constituents in the absence and presence of phenoloxidases (viz., peroxidase, laccase, and tyrosinase) or acid birnessite. Substituted phenols were chosen as simple model constituents to determine the structures in humic substances important for cross-coupling reactions. The extent of sulfonamide transformation was evaluated by the disappearance of the parent compound from solution. Incubation with phenoloxidases in the absence of substituted phenols resulted in little or no sulfonamide transformation. In contrast to this, direct oxidation of sulfonamides by acid birnessite was significant. Inclusion of o-diphenols and 2,6-dimethoxyphenols in reaction mixtures resulted in significant phenoloxidase-mediated transformation of sulfonamides and enhanced antimicrobial transformation in the presence of acid birnessite. Phenolic compounds with other substitution patterns were less effective in promoting sulfonamide transformation. Nuclear magnetic resonance spectroscopy experiments provided direct evidence of peroxidase-mediated covalent crosscoupling of sulfamethazine with syringic and protocatechuic acids. Our results indicate that sulfonamide antimicrobials may be chemically incorporated into humic substances. This may result in their diminished mobility, bioavailability, and biological activity.

Introduction Sulfonamide antimicrobials are synthetic, primarily bacteriostatic agents finding use in both human therapy and animal husbandry. Sulfonamides rank among the top 200 prescribed human drugs in the U.S. (1). Large quantities of these antimicrobials are used for the prevention and treatment of disease in livestock production (2). Sulfonamide antimicro* Corresponding author phone: (608) 263-4971; fax: (608) 2652595; e-mail: [email protected]. † Molecular and Environmental Toxicology Center, University of Wisconsin. ‡ Department of Chemistry, University of Toronto at Scarborough. § Environmental Chemistry and Technology Program, University of Wisconsin. | Department of Soil Science, University of Wisconsin. 10.1021/es0500916 CCC: $30.25 Published on Web 05/13/2005

 2005 American Chemical Society

bials are excreted as the unaltered parent compound or an acetylated metabolite (3) which can be reactivated by bacterial cleavage of the acetyl moiety (4). Veterinary antimicrobials introduced into soils by manure amendments and at animal feeding operations may enter surface waters and groundwater through runoff and infiltration. Sulfonamides used in human therapy are released into the environment through wastewater discharges and have been identified in municipal sewage, treated wastewater treatment plant effluent, groundwater, and soils irrigated with reclaimed water (5-7). Sulfonamides were among the most frequently detected antibiotics in a recent survey of organic microcontaminants in U.S. streams (8). The primary concern with the introduction of these agents into soil and water environments is the possible spread of antimicrobial resistance in response to increased selective pressure, potentially leading to the proliferation of resistant pathogens. Whether the low levels of antimicrobials in many environments are sufficient to exert selective pressure remains unclear. Sulfonamide antimicrobials are not readily biodegraded (9) and persist in soils (5, 10). While the low sorption coefficients obtained in short-term batch experiments indicate minimal association of sulfonamides with clays and humic substances (11, 12), some reports in the literature suggest the formation of nonextractable residues (4, 13). The extent and mechanisms of nonextractable sulfonamide residue formation have not been elucidated. We hypothesize that chemical incorporation of sulfonamides into natural organic matter (NOM) may contribute to bound residue formation. The aromatic amine common to all sulfonamides (see Table 1) represents a moiety likely to engage in covalent bond formation with NOM. Previous work with other aromatic amines (e.g., aniline, reduced TNT metabolites, chloroanilines, and fungicides) demonstrated that such compounds covalently cross-couple with NOM (14-16). Phenoloxidases (e.g., peroxidase, laccase, and tyrosinase) and metal oxides (especially manganese oxides) present in soils and sediments (17, 18) serve as important mediators of aromatic amine incorporation into NOM (19). Oxidative cross-coupling of aromatic amines results from radicalradical coupling with reactive species present in NOM or nucleophilic addition to electron-poor sites in humic substances (14, 20). Natural organic matter in soils and aquatic environments contains abundant phenolic moieties, including numerous substituted phenols (e.g., caffeic acid, catechol, ferulic acid, protocatechuic acid, and syringaldehyde) (21). Phenolic compounds serve as substrates for phenoloxidases and are transformed by manganese oxides. Oxidation of phenolic compounds by peroxidases, laccase, and manganese oxides produces phenoxy radicals that may self-couple to form oligomers and/or polymers or cross-couple with aromatic amines (14, 20, 22). Products of tyrosinase oxidation may undergo o-quinone self-coupling (22). Oxidation of phenolic compounds may generate electron-poor sites that can be attacked by nucleophilic aromatic amines (14, 20). The primary objectives of this study were to determine the degree to which sulfonamide antimicrobials engage in cross-coupling reactions with model NOM constituents in the presence of manganese oxide or phenoloxidases (viz., peroxidase, tyrosinase, and laccase), and to determine the nature of selected cross-coupled products. Substituted phenols were chosen as simple model constituents to determine the structures in humic substances important for cross-coupling reactions. A secondary objective was to VOL. 39, NO. 12, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4463

TABLE 1. Physicochemical Properties of Sulfonamidesa

a Abbreviations: M , molecular mass; C sat, aqueous solubility; K , i w ow 1-octanol-water partition coefficient. b Reference 23. c Reference 24. d Reference 25. e Reference 26. f Reference 27. g Reference 28. h Reference 29. i Reference 30.

examine the degree to which differences in the structure of sulfonamide antimicrobials affect their propensity to engage in cross-coupling reactions. To achieve these objectives, we incubated individual sulfonamide antimicrobial agents with surrogate humic constituents in the absence and presence of enzymes or manganese oxide. Sulfonamide transformation was evaluated by loss of the parent compound from solution. Direct evidence of covalent bond formation was obtained by nuclear magnetic resonance (NMR) spectroscopy. To our knowledge, this is the first study demonstrating covalent bond formation of sulfonamides with NOM constituents, a process that may result in their immobilization and detoxification.

Materials and Methods Chemicals. The chemicals used, suppliers, and purities are described in the Supporting Information. The structures and abbreviations of the sulfonamide antimicrobials and model humic constituents employed are presented in Tables 1 and 2. Phenoloxidases. The phenoloxidases employed, activities, purities, and suppliers are presented in the Supporting Information. Manganese Oxide Preparation. We prepared MnO2 by the method of McKenzie (31) (see the Supporting Information). The literature uses inconsistent and sometimes contradictory nomenclature for layer-type manganese(III,IV) oxides. In this paper, we adopt the terminology of Villalobos et al. (32) and refer to the manganese oxide synthesized by the McKenzie method (31) as “acid birnessite”. Transformation Reactions in the Presence of Phenoloxidases. We examined the extent of sulfonamide antimicrobial transformation in incubations with model humic constituents in the absence and presence of phenoloxidases. Acid-washed, silanized, 50 mL amber glass vials with Teflonlined screw caps were employed as reaction vessels. Each 10 mL reaction mixture contained 0.3 mM sulfonamide antimicrobial in 0.2 M sodium acetate adjusted to pH 5.6. Model humic constituents were added to achieve an initial concentration of 0.3 mM. Several types of control incubations were run in parallel. Controls containing only a sulfonamide antimicrobial demonstrated that losses to vessel walls and uncatalyzed reactions were negligible (p > 0.05). Controls consisting of a sulfonamide and individual substituted phenols showed that detectable uncatalyzed cross-coupling did not occur under our reaction conditions. Incubations of 4464

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 12, 2005

individual sulfonamides with phenoloxidases were used to control for any enzymatic transformation of the antimicrobials. Enzymatic reactions were quenched by addition of 23 mL of methanol (70% of the final volume). Experiments verifying the efficacy of this procedure as well as those demonstrating negligible sulfonamide sorption to enzymes and polymeric products are described in the Supporting Information. Controls and reaction mixtures were analyzed for changes in sulfonamide concentration by high-performance liquid chromatography with UV detection (HPLCUV; see the Supporting Information). All experiments were conducted in triplicate. Selected sets of experiments were repeated to verify reproducibility. Reaction mixtures in peroxidase experiments included 0.8 U‚mL-1 Arthromyces ramosus peroxidase (ARP) or horseradish peroxidase (HRP) and 0.5 mM H2O2 as an electron acceptor. Peroxidase-catalyzed reactions were allowed to proceed for 2 h. The effect of H2O2 on sulfonamide transformation was investigated using parallel controls containing a sulfonamide and H2O2 in the absence and presence of each model humic constituent. Negligible sulfonamide transformation (i.e., 0.05). Similar results were obtained in incubations with HRP (see the Supporting Information). Incubations of SMX and SPD with model humic constituents in the presence of ARP and H2O2 produced transformation patterns similar to those observed for SMZ, although a few exceptions were apparent. While the magnitude of SMZ and SMX transformation (∼80%) was comparable for reaction mixtures including syr, SPD exhibited substantially less transformation (∼40%). Incubations with fer did not impact SMZ levels, but resulted in small, but significant, decreases in SPD and SMX concentrations (p < 0.015). Incubations with p-hyd led to a statistically significant (p < 0.015) decrease in SMZ concentration, but not in SMX and SPD levels. In contrast to SMZ, incubations of SPD and SMX with gua brought about significant transformation (13%, 13%; p < 0.015). Transformation of Sulfonamide Antimicrobials in the Presence of Tyrosinase. Figure 2 displays the results of incubations of model humic constituents and sulfonamide antimicrobials with tyrosinase. No significant decreases in sulfonamide concentration were observed during 24 h incubations in controls containing tyrosinase alone (p > 0.05). As in peroxidase incubations, tyrosinase-mediated reactions with syr, pro, caf, cat, and syald produced large decreases (∼70-90% transformation) in sulfonamide concentrations (p < 0.001) (Figure 2). Relative to the controls, gua, res, orc, VOL. 39, NO. 12, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4465

FIGURE 1. (a) SMZ, (b) SMX, and (c) SPD remaining in solution after reaction with model humic constituents in the presence of ARP and H2O2. Error bars indicate 1 standard deviation. Significance of difference relative to incubations lacking model humic constituents: one asterisk, p < 0.015; two asterisks, p < 0.001. Key: hydroxybenzoic acids (gray bars), hydroxycinnamic acids (dark gray bars), hydroxybenzenes (left slanted hatched bars), and hydroxybenzaldehyde (right slanted hatched bars). See Table 2 for abbreviations. and p-hyd did not affect the parent compound concentrations (p > 0.05). While incubations with gal and fer led to significant decreases in all antimicrobial concentrations (p < 0.015), van resulted in a statistically significant decrease in concentration for SMZ only. Both SMZ and SPD underwent significant transformation when incubated with hyd in the presence of tyrosinase (p < 0.015); SMX remained unchanged. Comparable magnitudes of transformation were observed for each sulfonamide. Experimental conditions during the incubations (viz., enzyme activity, reaction duration, and electron acceptor concentration) differed for each class of enzyme, preventing us from making useful statistical comparisons among them. We therefore restrict ourselves to comparisons of sulfonamide 4466

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 12, 2005

FIGURE 2. (a) SMZ, (b) SMX, and (c) SPD remaining in solution after reaction with humic constituents in the presence of tyrosinase. Error bars indicate 1 standard deviation. Significance of difference relative to incubations lacking model humic constituents: one asterisk, p < 0.015; two asterisks, p < 0.001. Key: hydroxybenzoic acids (gray bars), hydroxycinnamic acids (dark gray bars), hydroxybenzenes (left slanted hatched bars), and hydroxybenzaldehyde (right slanted hatched bars). See Table 2 for abbreviations. transformation patterns among the phenoloxidases. With the exception of p-hyd and hyd, trends in SMZ transformation in the presence of tyrosinase and peroxidases appeared similar. Considerable declines in SMZ levels were observed in incubations with hyd in the presence of tyrosinase (∼20% transformation), but no changes were apparent in analogous reaction mixtures containing peroxidases. Incubations with p-hyd led to a decrease in SMZ concentration in incubations with peroxidases but not with tyrosinase. Sulfamethazine concentrations decreased significantly in incubations with gua, res, and van when HRP was present (Figure S-1 in the Supporting Information), but not in analogous reactions with tyrosinase (Figure 2a). Transformation of SMZ in the Presence of Laccase. Only minor SMZ transformation (∼5%) was evident in 24 h incubations with laccase alone (Figure 3). Only gal and orc

FIGURE 3. Sulfamethazine remaining in solution after reaction with humic constituents in the presence of laccase. Error bars indicate 1 standard deviation. Significance of difference relative to incubations lacking model humic constituents: one asterisk, p < 0.015; two asterisks, p < 0.001. Key: hydroxybenzoic acids (gray bars), hydroxycinnamic acids (dark gray bars), hydroxybenzenes (left slanted hatched bars), and hydroxybenzaldehyde (right slanted hatched bars). See Table 2 for abbreviations. failed to enhance SMZ transformation in the presence of laccase (p > 0.05). Large declines in SMZ concentration were observed in incubations with syr, cat, syald, pro, and caf (∼70%, 80%, 70%, 67%, 55%; p < 0.001) (Figure 3). Similar to the peroxidases and tyrosinase incubations, reactions with cat effected the largest SMZ transformation (∼80%). Major differences in the extent of laccase-mediated SMZ transformation relative to analogous reactions in the presence of peroxidases and tyrosinase included (1) significant SMZ transformation in reactions with p-hyd, gua, and res mediated by only HRP and laccase, (2) significant enzyme-mediated SMZ transformation in the presence of syr except in incubations with HRP, (3) significant laccase and tyrosinasemediated SMZ transformation in the presence of hyd, and (4) no significant HRP- or laccase-mediated SMZ transformation in the presence of gal. Discussion of Sulfonamide Transformation in Enzymatic Reactions. Major differences were not noted among the sulfonamides in enzyme-mediated transformations. Sulfonamide antimicrobials are apparently poor substrates for the enzymes employed. Sulfonamide transformation in controls containing phenoloxidases but lacking surrogate NOM constituents was relatively minor ( cat ≈ syald > caf (Figure 4a). Products from reactions of these model NOM constituents with SMZ differed in color from those resulting from oxidation of substituted phenols alone, and sulfonamide transformation was accompanied by reductions in the concentration of the each surrogate humic constituent. Incubations with the 4468

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 12, 2005

FIGURE 4. SMZ (a), SMX (b), and SPD (c) remaining in the supernatant after reaction with humic constituents in the presence of acid birnessite. Error bars indicate 1 standard deviation. Significance of difference relative to reaction mixtures lacking model humic constituents: one asterisk, p < 0.015; two asterisks, p < 0.001. Key: hydroxybenzoic acids (gray bars), hydroxycinnamic acids (dark gray bars), hydroxybenzenes (left slanted hatched bars), and hydroxybenzaldehyde (right slanted hatched bars). See Table 2 for abbreviations. remaining model NOM constituents did not enhance SMZ transformation beyond that observed in controls containing acid birnessite. Trends in SMX and SPD transformation in the presence of substituted phenols and birnessite were similar to that of SMZ with a few exceptions. In contrast to SMZ, SMX underwent significant transformation when reacted with hyd (∼60%; Figure 4b). Interestingly, inclusion of most substituted phenols in reaction mixtures inhibited SPD transformation (Figure 4c); syald, cat, syr, and pro were exceptions. Inclusion of these four model NOM constituents resulted in significant SPD transformation (∼85-95%; p < 0.001), with syald producing the largest effect. Our data indicate that syr, pro, cat, and syald enhanced acid birnessite-

FIGURE 5. 1H-13C HSQC spectra of [phenyl-13C6]-SMZ incubated with syr in the (a) absence and (b) presence of ARP. The appearance of new cross-peaks (circled) in spectrum b indicates a change in chemical environment due to a fraction of SMZ molecules engaging in covalent bond formation. The 13C label couples strongly with its attached protons and causes strong 13C satellites. (Compounds with 13C at natural abundance would not show these intense satellites.) In (a) the satellites of SMZ are highlighted with arrows; in (b) the satellites of the covalent product are highlighted with arrows. The presence of the satellites indicates the novel cross-peaks result from some reaction with the labeled SMZ and cannot simply be interference from background (unlabeled) constituents. Note: the carbon axes are internal projections from the two-dimensional datasets and are displayed for qualitative reference only. mediated transformation of all three sulfonamides; caf increased transformation of only SMZ and SMX. Similar to enzymatic reactions, o-diphenols (viz., caf, pro, and cat) and 2,6-dimethoxyphenols (viz., syr and syald) were effective in enhancing sulfonamide transformation. Products from the reaction of acid birnessite with these substituted phenols possess sites for nucleophilic attack by aromatic amines. For example, birnessite oxidizes cat and pro to electrophilic quinoid polymers (54) and syr to phenoxy radicals which upon subsequent demethoxylation yield o-benzoquinones (22, 54). These model humic constituents enhanced the manganese oxide-mediated transformation of some chloroanilines (40). Enhanced sulfonamide transformation may have also resulted from coupling of radical sulfonamide species with phenoxy radicals. Both radicalradical coupling and nucleophilic addition have been described as mechanisms for the covalent cross-coupling of anilines with humic substances (14, 20). Stone and Morgan (55) reported that, among the phenolic compounds they investigated, cat, pro, and syr were the most rapidly oxidized by manganese oxide. The model NOM constituents most effective in reductively dissolving the oxide also produced the largest enhancement of sulfonamide transformation by acid birnessite in the present study (Figure 4). The rapid formation of reactive products (55) likely contributed to the enhanced transformation of sulfonamides. Inclusion of van, gal, p-hyd, fer, hyd, gua, res, or orc in reaction mixtures containing acid birnessite did not impact the degree of sulfonamide reaction (SMZ and SMX; Figure 4a,b) or inhibit antimicrobial transformation (SPD; Figure 4c) relative to controls lacking substituted phenols. Factors that may have contributed to this outcome include (1) products resulting from oxidation of specific model humic constituents lacking sites favorable for cross-coupling with sulfonamides, (2) the relative rates of phenoxy radical selfcoupling and cross-coupling with sulfonamides, and (3) adsorption of oxidation products of substituted phenols onto the oxide surface (38, 55). Products resulting from the acid birnessite-mediated oxidation of phenolic compounds are similar to those formed enzymatically. For example, gua self-couples to form extended quinoid dimers in the presence of birnessite (22).

Incubations with acid birnessite and o-methoxyphenols did not enhance the transformation of sulfonamides relative to that of the controls. Manganese oxide oxidizes the p-diphenol hyd to o- and p-benzoquinones (54, 55). This p-diphenol led to the enhanced transformation of SMX only (Figure 4b). The m-diphenol res undergoes birnessite-mediated oxidation to generate coupled and ring cleavage products (55) but did not increase sulfonamide transformation over that observed in the controls. We find it interesting that van, gal, p-hyd, fer, res, and orc inhibited the transformation of SPD, but not that of the other two sulfonamides. This may have been due to decreased SPD reactivity toward oxidation products of these constituents. Alternatively, if acid birnessite oxidized SPD more slowly than the other two sulfonamides, active sites on the mineral surface may have become blocked by adsorption of substituted phenol oxidation products before substantial SPD transformation occurred. We note that, in substituted phenolfree controls, SPD transformation by acid birnessite was more extensive than that of SMZ or SMX. Solution-State NMR Experiments. We obtained direct spectroscopic evidence for covalent bond formation between SMZ and the model humic constituents syr and pro in the presence of ARP. These substituted phenols exhibited pronounced reactivity with sulfonamides in the presence of this peroxidase. HSQC experiments detect heteronuclei coupled across single bonds, thereby identifying directly connected nuclei. Figure 5 displays 1H-13C HSQC spectra of 13C-labeled SMZ incubated with syr in the absence and presence of ARP. The cross-peak at 112 ppm on the F1 axis corresponds to the two equivalent aromatic carbons ortho to the anilinic nitrogen; that at 129 ppm represents the equivalent carbons meta to the same substituent (Figure 5a). The 13C-labeled positions lacking attached protons did not give rise to cross-peaks in the HSQC experiment. Following incubation of SMZ with syr in the presence of ARP, two new cross-peaks appeared (Figure 5b). The new cross-peaks resulted from reactions of the 13C-labeled SMZ molecule as indicated by the 13C satellites that result from directly attached protons coupling strongly to labeled 13C nuclei (see Figure 5b). Changes in chemical shift were larger for carbon nuclei ortho to the aniline nitrogen than those meta to the same VOL. 39, NO. 12, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4469

FIGURE 6. (a) HMBC long-range 1H-15N correlations demonstrate that linkage of SMZ and syr was through the anilinic nitrogen. The equivalence of protons 15 and 17 and nonequivalence of protons 21 and 25 allow assignment of the position of the double bond between carbon 20 and nitrogen 19. The nitrogen spectrum was referenced to NH4Cl. (b) DPFGSE-NOESY shows that proton 15/17 was in close proximity to proton 25, but not proton 21. This indicates a rigid bond connects SMZ to the syringyl ring. moiety. This is consistent with the participation of the anilinic nitrogen in covalent bond formation, perhaps by nucleophilic addition to benzoquinones resulting from peroxidase-mediated oxidation of syr (14). Similar results were obtained from 1H-13C HSQC experiments examining the reaction product(s) resulting from incubation of [13C]-SMZ with pro in the absence and presence of ARP (Figure S-3 in the Supporting Information). To determine the nature of the product and the covalent linkage between SMZ and syr, TOCSY, 1H-13C HSQC, 1H13C HMBC, 1H-15N HMBC, and DPFGSE-NOESY experiments were performed on reaction products isolated by solid-phase extraction. We note that additional products may have been formed that were not retained by the solid-phase extraction cartridge or eluted with methanol. TOCSY, 1H-13C HSQC, and 1H-13C HMBC experiments were used to assign the H-C framework of the product (data not shown due to space limitations). HMBC identified long-range through-bond couplings between the 1H and 15N nuclei, while DPFGSENOESY established through-space correlations of protons proximal to one another. Figure 6 shows results from these latter two experiments along with the proposed structure of the reaction product. Protons attached to carbons 21, 25, 17, and 15 display couplings with the nitrogen at position 19 as demonstrated by three cross-peaks in Figure 6a. A single cross-peak was assigned to protons at positions 15 and 17 due to free rotation about the single bond between nitrogen 19 and carbon 16 resulting in 1H chemical equivalency. In contrast, two different cross-peaks were associated with protons 21 and 25. The rigidity of the double bond between carbon 20 and nitrogen 19 prevents free rotation and therefore results in chemical nonequivalence. The fact that protons 15 and 17 were identical while protons 21 and 25 were not confirms the position of the double bond. All relevant chemical shifts for the proposed product structure shown in Figure 6 were 4470

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 12, 2005

predicted using a combination of Advanced Chemistry Development (ACD) Laboratories Spec Manager, Nitrogen Predictor, and 2D Predictor (Figure S-4 in the Supporting Information). Small differences between the predicted and observed chemical shifts for the nitrogen at position 19 were noted and resulted from the lack of aromatic structures identical to that in Figure 6 in the ACD Laboratories internal database (containing ∼9000 nitrogen-containing chemicals used for chemical shift predictions). Simulation results are further discussed in the Supporting Information. We note that good chemical shift references for quinone imines are difficult to obtain. We searched logged NMR shifts (over 4 million chemical shifts) and found only 96 containing quinone imines as part of larger natural products. All of these were for protonated quinone imines; in no case was a 15N shift reported for an unprotonated quinone imine (i.e., that necessary for assigning the chemical shift of the quinone imine in the proposed structure). Despite the lack of relevant chemical shift references for unprotonated quinone imines, our simulated spectra argue strongly for the structure proposed in Figure 6. Through-space interactions were determined to refine the structure of the reaction product. A DPFGSE-NOESY experiment was used to selectively excite a particular resonance to identify protons in close proximity in space to the selected proton. This was carried out for all the protons on the aniline ring in SMZ in addition to protons on the syringyl ring. The key interaction between proton 15/17 on SMZ and proton 25 on the syringyl ring are shown in Figure 6b. No interaction was observed between protons 15/17 and 21. This observation not only provides information on the conformation of the molecule in solution, but also verifies the position of the nitrogen double bond. On the basis of the proposed structure, the distance between protons 21 and 15/17 is ∼5 Å, while that between protons 25 and 15/17 is shorter, estimated to be ∼1 Å. Additional through-space

SCHEME 2. ARP-Mediated Cross-Coupling of Sulfamethazine with Syringic Acida

a

Adapted from ref 14.

interactions indicated that protons 25 and 30 and protons 21 and 27 were in close proximity to one another (data not shown). All the resonances in these experiments and the mass spectrum of the main recovered product (m/z 427.6; Figure S-2) were consistent with the structure presented in Figure 6b. Scheme 2 outlines the proposed ARP-mediated formation of the product in Figure 6. Peroxidases mediate the oxidative decarboxylation of syr to form a 2,6-dimethoxy-p-quinone (14, 19). The anilinic nitrogen of SMZ subsequently engages in 1,2-nucleophilic addition to the keto group, resulting in the formation of the unprotonated imine quinone. Environmental Implications. This research was prompted by reports of nonextractable sulfonamide residues in soils (4, 13). Our data suggest that covalent cross-coupling of sulfonamide antimicrobials to soil organic matter may represent a plausible explanation for these observations. In our experiments, substituted phenols were used as surrogates for humic substances. In the presence of specific surrogate humic constituents, phenoloxidases and manganese dioxide mediated the transformation of sulfonamide antimicrobials, resulting in the formation of cross-coupled products. Soil organic matter represents an exceptionally complex mixture of heterogeneous organic molecules including aliphatic and aromatic components with carboxylic acid, hydroxyl, ester, and ether moieties (21). The extent that sulfonamide antimicrobials engage in covalent cross-coupling in a particular soil environment depends on the abundance and availability of phenoloxidases and reactive manganese oxide surfaces, as well as the nature of the NOM. Because humic substances contain free radicals and sites appropriate for the nucleophilic addition of aromatic amines (14), covalent cross-coupling of sulfonamide antimicrobials with NOM may occur in the absence of appreciable levels of phenoloxidases or manganese oxides, albeit at slower rates. Although we conducted our experiments with pure acid birnessite and enzymes in simplified solutions, valuable information was obtained on the types of transformations that may occur in the environment. For reactions mediated by phenoloxidases, our data suggest that nucleophilic addition of sulfonamides may be more important than radical-radical coupling. Although transformation was observed in incubations with acid birnessite, soil environments are unlikely to contain concentrations of sulfonamides sufficient for this to occur; cross-coupling with humic substances would be expected. Covalent cross-coupling of sulfonamide antimicrobials with humic substances would limit their mobility and bioavailability. The anilinic nitrogen common to all sulfonamide antimicrobials is essential for the competitive inhibition of dihydropteroate synthase, the mechanism by which these compounds exert their antimicrobial effect (4). A stable covalent linkage formed via the anilinic nitrogen would be expected to eliminate the bioactivity of these compounds following binding. Phenoloxidases and manganese oxides are being considered for use in soil remediation and treatment of

contaminated wastewater. Manganese oxides have also been suggested as ingredients for permeable reactive barriers to treat groundwater (56). Most previous studies have focused on phenols, anilines, and pesticides (19, 56, 57). Our data suggest that such approaches may also be effective for immobilizing and inactivating sulfonamide antimicrobials.

Acknowledgments We appreciate the helpful advice of Charlie Fry and Monika Ivancic (University of WisconsinsMadison NMR facility). The University of WisconsinsMadison NMR facility is partly supported by a grant from the National Science Foundation (NSF CHE-9629688). We thank Walt Zeltner for performing the BET analysis and Juan Gao and Alex Castillo for laboratory assistance. The constructive comments of three anonymous reviewers are gratefully acknowledged. This research was funded by USDA Cooperative State Research, Education and Extension Service Project WIS04621 and the Wisconsin Small System Waste Management Project. This paper was presented at the 228th ACS National Meeting, Aug 22-26, 2004, in Philadelphia, PA.

Supporting Information Available Chemical purity and supplier information, enzyme purity, activity, and supplier information, preparation and characterization of acid birnessite, description of experiments verifying efficacy of methanol precipitation for quenching enzymatic reactions, HPLC methods, product isolation, instrumental parameters for NMR experiments, mass spectrometry method, results from HRP-mediated reactions, mass spectrum of products from ARP-mediated reaction of syr with SMZ, 1H-13C HSQC spectra of SMZ incubated with pro in the absence and presence of ARP, and and simulated 1H15N HMBC spectra. This information is available free of charge via the Internet at http://www.pubs.acs.org.

Literature Cited (1) RxList: The internet drug index. http://www.rxlist.com/ top200.htm (accessed August 2003). (2) National Research Council. The Use of Drugs in Food Animals: Benefits and Risks; National Academy Press: Washington, DC, 1999. (3) Petri, W. A., Jr. Antimicrobial Agents: Sulfonamides, Trimethoprim-Sulfamethoxazole, Quinolones, and Agents for Urinary Tract Infections. In Goodman & Gilman’s The Pharmacological Basis of Therapeutics, 10th ed.; Hardman, J. G., Limbird, L. E., Gilman, A. G., Eds.; McGraw-Hill: New York, 2001; pp 11711179. (4) Berger, K. V.; Petersen, B.; Bunning-Pfaue, H. Persistenz von Gu ¨lle-Arzneistoffe in der Nahrungskette. Arch. Lebensmittelhyg. 1986, 37, 85-108. (5) Ku ¨ mmerer, K. Significance of antibiotics in the environment. J. Antimicrob. Chemother. 2003, 52, 5-7. (6) Miao, X.; Bishay, F.; Chen, M.; Metcalfe, C. D. Occurrence of antimicrobials in the final effluents of wastewater treatment plants in Canada. Environ. Sci. Technol. 2004, 38, 3533-3541. (7) Kinney, C.; Furlong, E.; Cahill, J.; Werner, S. Occurrence and behavior of wastewater derived pharmaceuticals in soils irrigated VOL. 39, NO. 12, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4471

(8)

(9)

(10)

(11)

(12)

(13)

(14)

(15)

(16)

(17)

(18)

(19)

(20)

(21) (22)

(23) (24)

(25)

(26)

(27)

(28)

with reclaimed water. Abstracts of Papers, Part 1, 227th National Meeting of the American Chemical Society, March 28 to April 1, 2004; American Chemical Society: Washington, DC, 2004; 112-ENVR. Kolpin, D. W.; Furlong, E. T.; Meyer, M. T.; Thurman, E. M.; Zaugg, S. D.; Barber, L. B.; Buxton, H. T. Pharmaceuticals, hormones, and other organic wastewater contaminants in U.S. streams, 1999-2000: A national reconnaissance. Environ. Sci. Technol. 2002, 36, 1202-1211. Ingerslev, F.; Halling-Sørenson, B. Biodegradability properties of sulfonamides in activated sludge. Environ. Toxicol. Chem. 2000, 19, 2467-2473. Christian, T.; Schneider, R. J.; Farber, H. A.; Skutlarek, D.; Meyer, M. T.; Goldbach, H. E. Determination of antibiotic residues in manure, soil, and surface waters. Acta Hydrochim. Hydrobiol. 2003, 31, 36-44. Gao, J.; Bialk, H. M.; Pedersen, J. A. Sorption of sulfonamide antimicrobials to soil constituents. Abstracts of Papers, Part 2, 226th National Meeting of the American Chemical Society, Sept 7-11, 2003; American Chemical Society: Washington, DC, 2003; 159-ENVR. Boxall, A. B. A.; Kolpin, D. W.; Halling-Sørensen, B.; Tolls, J. Are veterinary medicines causing environmental risks? Environ. Sci. Technol. 2003, 37, 286A-291A. Langhammer, J.-P.; Fu ¨ hr, F.; Bu ¨ nning-Pfaue, H. Verbleib von Sulfonamid-Ru ¨ cksta¨nden aus der Gu ¨ lle im Boden und Nutzpflanzen. Lebensmittelchemie 1990, 44, 93. Thorn, K. A.; Goldenberg, W. S.; Younger, S. J.; Weber, E. J. Covalent binding of aniline to humic substances. Comparison of nucleophilic addition, enzyme-, and metal-catalyzed reactions by 15N NMR. In Humic and Fulvic Acids: Isolation, Structure and Environmental Role; Gaffney, J. S., Marley, N. A., Clark, S. B., Eds.; ACS Symposium Series 651; American Chemical Society: Washington, DC, 1996; pp 299-326. Li, H.; Lee, L. S.; Jafvert, C. T.; Graveel, J. G. Effect of substitution on irreversible binding and transformation of aromatic amines with soils in aqueous systems. Environ. Sci. Technol. 2000, 34, 3674-3680. Weber, E. J.; Colon, D.; Baughman, G. L. Sediment-associated reactions of aromatic amines. 1. Elucidation of sorption mechanisms. Environ. Sci. Technol. 2001, 35, 2470-2475. Post, J. E. Manganese oxide minerals: Crystal structures and economic and environmental significance. Proc. Natl. Acad. Sci. U.S.A. 1999, 96, 3447-3454. Gramss, G., Voigt, K.-D., Kirsche, B. Oxidoreductase enzymes liberated by plant roots and their effects on soil humic material. Chemosphere 1999, 38, 1481-1494. Bollag, J.-M. Decontaminating soil with enzymes: An in situ method using phenolic and anilinic compounds. Environ. Sci. Technol. 1992, 26, 1876-1881. Thorn, K. A.; Kennedy, K. R. 15N NMR investigation of the covalent binding of reduced TNT amines to soil humic acid, model compounds, and lignocellulose. Environ. Sci. Technol. 2002, 36, 3787-3796. Stevenson, F. J. Humus Chemistry: Genesis, Composition, Reaction; John Wiley & Sons: New York, 1994. Dec, J.; Haider, K.; Bollag, J.-M. Decarboxylation and demethoxylation of naturally occurring phenols during coupling reactions and polymerization. Soil Sci. 2001, 166, 660-671. Advanced Chemistry Development (ACD) Software Solaris V4.67; ACD/Labs: Toronto, Canada, 1994-2004. Shah, N. H.; Lazarus, J. H.; Sheth, P. R.; Jarowski, C. I. Carboxymethylcelluloseseffect of degree of polymerization and substitution on tablet disintegration and dissolution. J. Pharm. Sci. 1981, 70, 611-613. Lin, C. E.; Chang, C. C.; Lin, W. C. Migration behavior and separation of sulfonamides in capillary zone electrophoresis III. Citrate buffer as a background electrolyte. J. Chromatogr., A 1997, 768, 105-112. Nasipuri, R. N.; Khalil, S. A. H. Adsorption-dissolution relationship in sulfamethazine benzoic acid system. J. Pharm. Sci. 1973, 62, 473-475. Clark, W. G.; Strakosch, E. A.; Levitan, N. I. Solubility and pH data of the commonly used sulfonamides. J. Lab. Clin. Med. 1942, 28, 188-189. Lin, C. E.; Chang, C. C.; Lin, W. C. Migration behavior and separation of sulfonamides in capillary zone electrophoresis. 2. Positively charged species at low pH. J. Chromatogr., A 1997, 759, 203-209.

4472

9

ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 39, NO. 12, 2005

(29) Thiele-Bruhn, S.; Seibicke, T.; Schulten, H.-R.; Leinweber, P. Sorption of sulfonamide pharmaceutical antibiotics on whole soils and particle-size fractions. J. Environ. Qual. 2004, 33, 13311342. (30) Kan, C. A.; Petz, M. Residues of veterinary drugs in eggs and their distribution between yolk and white. J. Agric. Food Chem. 2000, 48, 6397-6403. (31) McKenzie, R. M. The synthesis of birnessite, cryptomelane, and some other oxides and hydroxides of manganese. Mineral. Mag. 1971, 38, 493-502. (32) Villalobos, M.; Toner, B.; Bargar, J.; Sposito, G. Characterization of the manganese oxide produced by Pseudomonas putida strain MnB1. Geochim. Cosmochim. Acta 2003, 67, 2649-2662. (33) Kortum, G.; Vogel, W.; Andrussow, K. Dissociation Constants of Organic Acids in Aqueous Solution; Butterworth: London, 1961. (34) Bykova L. N., Petrov S. I.; Blagodatskava Z. G. Relative acidity of phenol and its derivatives in a medium of nonaqueous solvents. Zh. Obshch. Khim. 1970, 40, 2295-3000. (35) Perrin, D. D.; Serjeant, E. P.; Dempsey, B. pKa Prediction for Organic Acids and Bases; Chapman and Hall: New York, 1981. (36) Park, J.-W.; Dec, J.; Kim, J.-E.; Bollag, J.-M. Effect of humic constituents on the transformation of chlorinated phenols and anilines in the presence of oxidoreductive enzymes or birnessite. Environ. Sci. Technol. 1999, 33, 2028-2034. (37) Chorover, J.; Majcher, E. H.; Bollag, J.-M.; Huang, P. M. Evolution of CO2 during birnessite-induced oxidation of 14C-labelled catechol. Soil Sci. Am. J. 2000, 64, 157-163. (38) Klaussen, J.; Haderlain, S. B.; Schwarzenbach, R. P. Oxidation of substituted anilines by aqueous MnO2: Effect of co-solutes on initial and quasi-steady-state kinetics. Environ. Sci. Technol. 1997, 31, 2642-2649. (39) Bollag, J.-M.; Strynar, M.-Y. A.; Dec, J. Characterization of enzymatic or abiotic immobilization of xenobiotics in soil. In Developments of Soil Science, Vol. 28B; Violante, A., Huang, P. M., Bollag, J.-M., Gianfreda, L., Eds.; Elsevier Science: Amsterdam, 2002; pp 289-299. (40) Wang, C.-J.; Thiele, S.; Bollag, J.-M. Interaction of 2,4,6trinitrotoluene (TNT) and 4-amino-2,6-dinitrotoluene with humic monomers in the presence of oxidative enzymes. Arch. Environ. Contam. Toxicol. 2002, 42, 1-8. (41) Berry, D. F.; Boyd, S. A. Reaction rates of phenolic humus constituents and anilines during cross-coupling. Soil Biol. Biochem. 1985, 17, 631-636. (42) Figueroa-Espinoza, M. C.; Rouau, X. Effect of cysteinyl caffeic acid, caffeic acid, and L-Dopa on the oxidative cross-linking of feruloylated arabinoxylans by a fungal laccase. J. Agric. Food Chem. 1999, 47, 497-503. (43) Parris, G. E. Covalent binding of aromatic amines to humates. 1. Reactions with carbonyls and quinones. Environ. Sci. Technol. 1980, 14, 1099-1106. (44) Hsu, T.-S.; Bartha, R. Interaction of pesticide-derived chloroaniline residues with soil organic matter. Soil Sci. 1974, 116, 444-452. (45) Weber, E. J.; Spidle, D. L.; Thorn, K. A. Covalent binding of aniline to humic substances. 1. Kinetic studies. Environ. Sci. Technol. 1996, 30, 2755-2763. (46) Colon, D.; Weber, E. J.; Baughman, G. L. Sediment-associated reactions of aromatic amines. 2. QSAR development. Environ. Sci. Technol. 2002, 36, 2443-2450. (47) Bollag, J.-M.; Dec, J. Phenoloxidase-mediated interactions of phenols and anilines with humic materials. J. Environ. Qual. 2000, 29, 665-676. (48) Kobayashi, A.; Koguchi, Y.; Kanzaki, H.; Kajiyama, S.; Kawazu, K. A new type of antimicrobial phenolics produced by plant peroxidases and its possible role in the chemical defense systems against plant pathogens. Z. Naturforsch. 1994, 7-8, 411-414. (49) Rescigno. A.; Sollai, F.; Pisu, B.; Rinaldi, A.; Sanjust, E. Tyrosinase inhibition: General and applied aspects. J. Enzyme Inhib. Med. Chem. 2002, 17, 207-218. (50) Tatsumi, K. A.; Freyer, R. D. Minard.; Bollag, J.-M. Enzymemediated coupling of 3,4-dichlooraniline and ferulic acid: A model for pollutant binding to humic materials. Environ. Sci. Technol. 1994, 28, 210-215. (51) Bollag, J.-M.; Liu, S.-Y.; Minard, R. D. Cross-coupling of phenolic humus constituents and 2,4-dichlorophenol. Soil Sci. Soc. Am. J. 1980, 44, 52-56. (52) Prota, G. Melanins and Melanogenesis; Academic Press: New York, 1992. (53) Laha, S.; Luthy, R. Oxidation of aniline and other primary aromatic amines by manganese oxide. Environ. Sci. Technol. 1990, 24, 363-373.

(54) McBride, M. B. Adsorption and oxidation of phenolic compounds by iron and manganese oxides. Soil Sci. Soc. Am. J. 1987, 51, 1466-1472. (55) Stone, A. T.; Morgan, J. J. Reduction and dissolution of manganese (III) and manganese (IV) oxides by organics: 2. Survey of the reactivity of organics. Environ. Sci. Technol. 1984, 18, 617-624. (56) Kang, Ki-H.; Dec, J.; Park, H.; Bollag, J.-M. Effect of phenolic mediators and humic acid on cyprodinil transformation in presence of birnessite. Water Res. 2004, 38, 2738-2745.

(57) Dura´n, N.; Esposito, E. Potential applications of oxidative enzymes and phenoloxidase-like compounds in wastewater and soil treatment: a review. Appl. Catal., B 2000, 28, 83-99.

Received for review January 14, 2005. Revised manuscript received March 29, 2005. Accepted March 31, 2005. ES0500916

VOL. 39, NO. 12, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4473