Oxidative Transformation of Fluoroquinolone Antibacterial Agents and

Yang Zhou , Yuan Gao , Su-Yan Pang , Jin Jiang , Yi Yang , Jun Ma , Yue Yang ... Shaofang Sun , Lihong Wang , Zhen Wang , Yuan Gao , Yi Yang , Jin Jia...
46 downloads 0 Views 263KB Size
Environ. Sci. Technol. 2005, 39, 4474-4483

Oxidative Transformation of Fluoroquinolone Antibacterial Agents and Structurally Related Amines by Manganese Oxide HUICHUN ZHANG AND CHING-HUA HUANG* School of Civil and Environmental Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332

Various members of the popular fluoroquinolone antibacterial agents (FQs) have been frequently detected in municipal wastewater and surface water bodies in recent years. This study was conducted to gain a better understanding of the fate of FQs in the sediment-water environment. Seven FQs were examined for adsorptive and oxidative interactions with δ-MnO2 under environmental conditions and exhibited reactivity in the order of ciprofloxacin ∼ enrofloxacin ∼ norfloxacin ∼ ofloxacin > lomefloxacin > pipemidic acid . flumequine. Four amines that are structurally related to the aniline and piperazine functional groups of FQs showed reactivity to oxidation by δ-MnO2 in the order of 1-phenylpiperazine > aniline > N-phenylmorpholine > 4-phenylpiperidine. Comparison among the above compounds clearly indicates that the piperazine moiety of FQs is the predominant adsorptive and oxidative site to MnO2. Product analyses showed that oxidation by MnO2 results in dealkylation and hydroxylation at the piperazine moiety of FQs, with the quinolone ring essentially intact. The reaction kinetics, reactivity comparison, and product characterization point to a surface reaction mechanism that likely begins with formation of a surface complex between FQ and the surface-bound MnIV, followed by oxidation at the aromatic N1 atom of FQ’s piperazine moiety to generate an anilinyl radical intermediate. The radical intermediates subsequently undergo N-dealkylation, C-hydroxylation, and possibly coupling to yield a range of products. Even though the quinolone ring appears to be stable with respect to MnO2, it affects the overall reactivity and potentially product distribution of FQs via substituent effects. Results of this study strongly suggest that manganese oxides commonly present in soils will likely play an important role in the abiotic degradation of fluoroquinolone antibacterial agents in the environment.

Introduction Fluoroquinolone antibacterial agents (FQs) are a group of potent synthetic antibiotics that are widely used in human and veterinary medicines (1). Because of their extensive usage, FQs may enter the environment via wastewater effluent and biosolids from sewage treatment plants and via manure and litters from food-producing animal husbandry. Several recent occurrence studies have demonstrated the potential omni* Corresponding author phone: 404-894-7694; fax: 404-894-8266; e-mail: [email protected]. 4474

9

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

presence of FQs in the environment. For example, ciprofloxacin and norfloxacin were detected in Switzerland at 40570 ng/L in domestic wastewaters and from nondetectable to 120 ng/L in surface waters that received wastewater discharge (2, 3). In the United States, ciprofloxacin and ofloxacin were repeatedly detected in various municipal wastewaters at 80 ng/L to 2 µg/L (4, 5), and four FQs (ciprofloxacin, norfloxacin, enrofloxacin, and sarafloxacin) were reported with mean concentrations from nondetectable to 0.12 µg/L in surface waters (6). The highest concentrations of FQs were reported in hospital wastewaters, with 0.7-124.5 µg/L of ciprofloxacin (7). The presence of antibiotic contaminants in the aquatic environment, albeit at low concentrations, may pose serious threats to the ecosystem and human health by inducing proliferation of bacterial drug resistance (8). FQsswhich interfere with bacterial DNA replication (1)shave been shown to contribute to a large proportion of measured bacterial genotoxicity in hospital sewage effluent (7, 9). To properly assess the risks of FQs, a better understanding of their environmental fate is imperative. As antibacterials, it is not surprising that FQs are rather resistant to microbial degradation (e.g., refs 10-12). On the other hand, studies have shown substantial degradation of ciprofloxacin and enrofloxacin by hydroxyl radical-mediated enzyme systems characteristic of brown-rot fungi (sp. Gloeophyllum striatum) (e.g., refs 13 and 14). Significant photolytic degradation of FQs by direct UV photolysis or radical-mediated photolysis has been known for some time (e.g., refs 15-19). Depending on the reaction conditions such as pH, cosolutes, etc., more than 10 photodegradation products may be generated via various pathways including dealkylation, defluorination, and hydroxylation (17-19). Sorption to soils, sediments, or dissolved organic matter was indicated to be another important environmental sink for FQs (20-22). For example, Nowara et al. (21) reported strong interactions of enrofloxacin with clay minerals and also found that greater than 90% of various FQs adsorbed to different soils with only small variations in the sorption coefficients Kd. The strong adsorption of FQs to soils and sediments may result in a lower concentration of freely dissolved species and thus reduce the photodegradation and biodegradation potential of FQs (23). Manganese oxides, commonly found in soils and sediments, are among the most important naturally occurring reactants or catalysts in facilitating organic pollutant transformation. δ-MnO2, with a reduction potential of 1.23 V, has been shown to be an effective oxidant for a wide range of pollutants including substituted phenols (e.g., refs 24 and 25) and anilines (e.g., refs 26 and 27). The recent work by Li et al. (28) demonstrated that soil manganese(III/IV) oxides play the most significant role, in addition to other soil components such as organic matter and iron oxides, in the irreversible oxidation and transformation of aromatic amines. On the basis of the strong affinity of FQs for soils and sediments (21, 22), there is a great possibility for FQs to react with mineral components in such systems. As will be shown later, the present investigation demonstrates that FQs are highly susceptible to manganese oxide-facilitated oxidation. This study examined a wide range of compounds including seven FQs [ciprofloxacin (CIP), enrofloxacin (ENR), norfloxacin (NOR), ofloxacin (OFL), lomefloxacin (LOM), pipemidic acid (PIP), and flumequine (FLU); structures shown in Figure 1] and four related aromatic and alicyclic amines [aniline, 1-phenylpiperazine (PP), N-phenylmorpholine (PM), and 4-phenylpiperidine (PD); structures shown in Figure 2] 10.1021/es048166d CCC: $30.25

 2005 American Chemical Society Published on Web 05/11/2005

FIGURE 1. (a) Structures of fluoroquinolones in this study. (b) Speciation pattern of CIP (Ka1 and Ka2 are macroscopic constants; Ka11, Ka21, Ka12, and Ka22 are microscopic constants).

TABLE 1. Properties, Initial Reaction Rate Constants, and Adsorption Extent of Fluoroquinolones in the Presence of MnO2a compound

FIGURE 2. Structures of related model amines in this study. to obtain a fundamental understanding of the chemical reactivity of FQs in reactions with manganese oxide. On the basis of the results of reaction kinetics, product characterization, and structure-activity assessment, schemes of reaction mechanisms are proposed. As shown in Figure 1b with CIP as an example, FQs (except for FLU) exhibit pH-dependent speciation in cationic, neutral, zwitterionic, or anionic forms. Because neutral and zwitterionic microspecies are often indistinguishable by simple potentiometric titration techniques, macroscopic constants Ka1 and Ka2 may be determined to describe the equilibria between cationic and neutral/zwitterionic FQ and between neutral/zwitterionic and anionic FQ, respectively. Although the macroscopic constant is not for a particular functional group, literature has linked the macroscopic Ka1 to the carboxylate group and the macroscopic Ka2 to the piperazinyl N4 atom of FQs because of pKa values similar to those of monofunctional analogues (29-32). The four microscopic constants (Ka11, Ka21, Ka12, and Ka22) describe protonation equilibria among the four microspecies. The reported (2932) and estimated (by the SPARC computer program; 33) macroscopic pKa values for the FQs are listed in Table 1.

Materials and Methods Chemical Reagents. Reagent-grade water was prepared by a Barnstead Nanopure water system. High-purity CIP, ENR,

ciprofloxacin (CIP) enrofloxacin (ENR) norfloxacin (NOR) ofloxacin (OFL) lomefloxacin (LOM) pipemidic acid (PIP) flumequine (FLU)

MW

pKa1 pKa2 refb

331.3459 5.46 6.2 359.3995 5.46 6.1 319.3349 5.46 6.22 361.3721 5.41 6.20 351.3522 5.38 5.49 303.32 5.20 5.42 261.2521 5.45 6.5

7.67 8.8 7.03 7.7 7.67 8.51 7.08 8.11 7.85 8.78 6.38 8.18

33 30 33 31 33 29 33 32 33 29 33 32 33 32

Kinit (h-1)

adsorptionc (%)

0.81 ( 0.10

76 ( 6

1.11 ( 0.44

62d

1.40 ( 0.36

69d

0.91 ( 0.26

68d

0.54 ( 0.04

44d

0.12 ( 0.04

39 ( 10

no reaction

10d

a Reaction conditions: [FQ] ) 1.5 µM, [MnO ] ) 100 µM, 0.01 M pH 0 2 0 6 MOPS buffer, 0.01 M NaCl, 22 °C. b Reference for pKa values. c Adsorption % ) C d Only one replicate was conducted. ads/Ctotal (%). Confidence levels (95%) were reported with rate constants and adsorption extents when applicable.

NOR, OFL, and FLU were purchased from ICN Biomedical, and LOM, PIP, and N-ethylformanilide were from Sigma. Aniline, PP, and PM from Aldrich and PD from Acros were at greater than 98% purity. Other employed chemical reagents were obtained from Fisher Scientific or Aldrich at greater than 98% purity (for solids) or of HPLC and GC grade (for solvents). All chemicals were used directly without further purification. Stocks of FQs were prepared in methanol/H2O mixture (10/90 v/v), whereas stocks of model amines were prepared in Nanopure water, at 100-120 mg/L. All stocks were protected from light, stored at PD (Table 2). Compared to PP, aniline, PM, and PD reacted with MnO2 about 3-, 28-, and 3350-fold more slowly. Experiments were also conducted with PP under reaction conditions identical to those used in the FQ experiments (i.e., [PP]0 ) 1.5 µM, [MnO2]0 ) 100 µM, pH 6, and 22 °C). In this case, PP yielded an average 3-fold higher initial reaction rate constant (kinit ) 3.08 ( 0.32 h-1) than the FQs (kinit ) 0.12∼1.40 h-1, Table 1). The adsorption to MnO2 followed the order of aniline (75%) > PP and PM (53-56%) > PD (19%) (Table 2). Oxidation Products of FQs. Reaction mixtures of FQs were stopped by centrifugation and the supernatants were analyzed by LC/MS. Analyses of CIP reaction mixtures indicated the presence of primarily six products with molecular ions of m/z 263, 306, 334a, 334b, 362, and 364 (two products had the same m/z ratio; see Figure 5 for proposed product structures and Table S1 in Supporting Information for spectral information). The molecular ion of CIP was m/z 332. The above molecular ions were identified on the basis of the difference of 22 in m/z ratios between the sodiated ([M + Na]+) and the molecular ([M + H]+) ions. The products were also short-written as M - 69, M - 26, M + 2a, M + 2b, M + 30, and M + 32 products, indicating the net mass loss or gain of the product from the parent CIP. All products exhibited fragmentation patterns quite similar to that of CIP, suggesting that they shared a common base structure with CIP, that is, the heterocyclic quinolone ring. For example, the [M + H - H2O]+ ion was almost always the base peak (i.e., the ion of the highest abundance) among the productssa fragmentation pattern corresponding to the loss of H2O from the carboxylate group of the quinolone ring. Additionally, losses of CO2, CO, HF, the three-membered propyl ring, and C2H5N of the piperazine ring (for those applicable) accounted for other less abundant fragments. The MS results obtained in this study agreed well with previous findings in the photolytic degradation products of FQs (e.g., ref 37). The M - 26 product was derived from partial dealkylation of the CIP piperazine ring, while the M - 69 product (with a final aniline functional group) was derived from full dealkylation of the piperazine ring; both products were previously identified in photodegradation of CIP (e.g., ref 37). The other four products (M + 30, M + 2a, M + 2b, and M + 32), in comparison to CIP, exhibited loss of more than one CO (-m/z 28) or one H2O (-m/z 18) in LC/MS fragmentation, suggesting that additional >CdO or -OH groups were present in these products (Table S1 and Figure 5). For example, the MS analyses showed that there could well be two additional >CdO groups in the M + 30 product (m/z 362) based on the presence of m/z 288 (i.e., 4478

9

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

after loss of 2 CO and 1 H2O) and m/z 217 (i.e., after loss of 3 CO, 1 H2O, and 1 C2H5N) daughter ions. One additional >CdO group in the M + 2b product (m/z 334b) was also quite plausible based on its m/z 217 daughter ion (i.e., after loss of 2 CO, 1 H2O, and 1 C2H5N). The M + 32 product (m/z 364) should contain two or more -OH groups to account for the loss of 2 H2O (plus other structural fragments) to yield the m/z 288 and 217 daughter ions. Furthermore, the fact that earlier studies on the oxidation of piperazine-containing compounds have reported the presence of >CdO group in the end products (17, 38) also supports the above conclusion. To evaluate the possible sources of additional >CdO or -OH groups in the reaction products, the following modifications to the experimental procedures were examined for their effects: (1) sparging solutions with N2 gas initially and excluding the involvement of O2 throughout the reaction, (2) preparing CIP stock solution in acetonitrile instead of in methanol, and (3) using HCl to adjust the pH of the reaction mixture instead of using acetic acid buffer. Reactions were monitored for 30 days and the same oxidation products were detected throughout the reaction period (4.5 h and 2, 11, and 30 days; data not shown) for all of the above experiments. These results showed that the additional >CdO and -OH groups in the products did not come from atmospheric oxygen, organic solvent, or acetic acid buffer. Oxidation products of ENR (m/z 360), NOR (m/z 320), OFL (m/z 362), LOM (m/z 352), and PIP (m/z 304) were also analyzed by LC/MS (see Tables S2-6 in Supporting Information for spectral information). The same types of products were identified for all of these FQs (i.e., M - 69, M - 26, M + 30, M + 32, and two M + 2 products), strongly suggesting that the FQs reacted with MnO2 via a similar mechanism. Note that not all of the fully dealkylated M - 69 products (i.e., with a final aniline functional group) have the same mass loss of 69 from the parent FQs. However, they are classified as M - 69 because of their structural similarity to the M - 69 product of CIP. Additionally, a trace amount (based on its small MS signal) of M + 16 product was also found in the oxidation of ENR, OFL, LOM, and PIP. The M + 16 product was not detected for CIP and NOR; however, it probably existed but at a concentration below the detection limit. On the basis of the literature that has characterized oxidation products of tertiary amines (39, 40), the M + 16 product is likely a N-oxide analogue of the FQ (Figure 5). Furthermore, there were several minor products that were not common for all FQs, including m/z 332 for ENR, m/z 348 for NOR, and m/z 314 and 298 for OFL. The m/z 332 product of ENR could be a CIP analogue as an oxidation intermediate of ENR reaction (37), though additional confirmation is needed. The other three products could not be identified from the spectra. The above results clearly demonstrated a complicated reaction mechanism for FQ oxidation by MnO2 with many possible reaction products; some may not be detectable by LC/MS. Analyses of oxidation products of FQs (at pH 5) over several hours showed that the MS peak area of the M - 26 product increased sharply at the beginning and then slowly decreased until below the detection limit, indicating that it is a reaction intermediate. All FQs exhibited quite similar product evolution patterns, with CIP shown as the example (Figure 6). The M + 16 and M + 32 products are also believed to be intermediates due to their transient presence, though these two products had much smaller MS signals and faster disappearing rates than the M - 26 product (data not shown). The abundance of the other four products (M + 2a, M + 2b, M + 30, and M - 69) was found to increase and then nearly plateau over time (Figure 6), indicating them to be the end products, although extended time studies up to several days revealed a small decrease in the abundance for the M - 69 product.

FIGURE 5. Proposed oxidation products of CIP analyzed by LC/MS. only the products in the centrifugation supernatants were analyzed and accounted for), (ii) the presence of other reaction products that could not be detected by the LC/MS techniques, and (iii) further reaction of the M - 69 products at longer reaction time.

FIGURE 6. Evolution of CIP oxidation products by MnO2 based on LC/MS detection. Note: A ) MS peak area of each compound in the reaction mixture at a given time; Amax ) MS peak area of CIP in the oxide-free control experiment at a given time. Attempts were made to evaluate if the distribution among the three types of oxidation end products (M + 2, M + 30, and M - 69) differed among the FQs. Due to the absence of authentic standards, a rudimentary quantification of the products was performed by using the crude assumption that the mass spectrometric response of the products is the same as the parent compounds. The abundance of each product in percentage relative to the total initial amount of the parent compound was calculated by their MS peak area ratio. At either 3.5 or 54 h of reaction time, the percentages of the M + 2 and M - 69 products did not vary significantly among CIP, ENR, NOR, OFL, and LOM, at -1.48 for LOM > -5.28 for PIP, estimated by the SPARC computer program (33). In other words, the electron-withdrawing substituents present in the aromatic ring of LOM and PIP (i.e., the two fluorine substituents in LOM and the two ring Ns in PIP) reduce the basicity of N1 via electronic effects and hence slow the reaction rate of LOM and PIP with MnO2. Such effects of electron-withdrawing substituents in reducing compounds’ susceptibility to oxidation have been reported in many previous studies (e.g., refs 24-26). Therefore as shown in Scheme 1, after forming a surface complex, MnO2 oxidizes FQ substrate at the N1 atom initially, in a manner similar to the oxidation of aniline-type compounds. The formed radical intermediate can be stabilized by resonance of the aromatic ring. The electron-transfer

reaction at the N1 atom is likely rate-limiting in this study; the surface complex formation step that occurs before the rate-limiting step also affects the overall reaction rate. Factors that affect either of the above two steps will influence the overall reaction rate. For instance, the fact that PP is about 3 times more reactive than CIP is likely due to the electronwithdrawing substituent effect in CIP that is absent in PP. The fact that the reaction rate of FQ oxidation by MnO2 slows as pH increases (Figure 4a) can also be explained by this proposed reaction scheme, in which increasing pH decreases adsorption of substrate to MnO2 (Figure 4b) as well as the oxidation power of MnO2 [i.e., pepH0 ) 20.8 - 2(pH), derived from ref 43]. Although it is clear now that the initial oxidation site of FQs is the piperazinyl N1 atom, the N4 atom also participates in the overall reaction on the basis of the results of product identification and a close examination of the observed compound reactivity. The experiments showed a 28-fold faster reaction rate for PP (kinit ) 6.70 ( 0.10 h-1) than PM (kinit ) 0.24 ( 0.07 h-1), although both compounds adsorbed similarly to MnO2 (53% for PP and 56% for PM, Table 2). PM differs from PP only by replacing N4 with an O atom. The faster rate of PP than PM reflects the reactivity of N4 (in addition to N1) versus the inertness of O atom to oxidation. However, on the basis of the very slow reaction rate of PD (kinit ) 2.0 × 10-3 h-1 for the N4 atom), the reactivity of the N4 atom alone cannot explain the 28-fold increase in reactivity from PM to PP. In other words, the reactivity of the N1 and N4 atoms is not solely additive. Interactions likely exist between the N1 and N4 atoms during the oxidation reaction and affect the overall reaction rate. As suggested in the work by Wang and Sayre (42), the oxygen atom present in PM may exert an electron-withdrawing effect on the N1 atom and reduce its reactivity for oxidation. Surface Reaction Scheme: Further Reactions of Radical Intermediates. As shown in Scheme 1, the formed radical intermediates may undergo several subsequent reactions to generate a range of products observed in this study. Pathway I is coupling of the radical intermediates. Earlier studies have demonstrated that anilinyl radicals most likely undergo coupling via the amino group (head) or the para position of the benzene ring (tail) to form head-to-head, head-to-tail, or tail-to-tail coupling products (e.g., refs 26 and 44). Coupling of FQ radicals, however, is likely difficult because of steric hindrance at the N1 position and the blockage at the para position in the aromatic ring. Coupling products of FQs were in fact not detected in this study, potentially due to low tendency of coupling reaction or strong adsorption of dimeric products, if formed, to manganese oxide surfaces and thus difficulty in being extracted for analyses. As will be discussed later, this radical coupling pathway is considered more likely for the further oxidation of the dealkylated products by excess manganese oxide. Pathway II is a hydroxylation process (Scheme 1), in which one electron is transferred from the radical N1 to MnIV/MnIII, yielding an iminium ion (39, 41, 44-46). The N4 atom in the piperazine ring then undergoes a similar two-electron transfer, generating another iminium ion at the same ethylenediamine edge as the N1 (42). This “double-iminium” ion is then quickly hydrolyzed to a 1,2-diol intermediate (i.e., M + 32) (40, 47). In the presence of MnO2, the 1,2-diol intermediate is readily oxidized to yield the dialdehyde product (i.e., M + 30) (48, 49) and possibly the two monoaldehyde products (i.e., M + 2). Pathway III is a N-dealkylation process (Scheme 1), resembling pathway II initially on the formation of an iminium ion. However, hydrolysis of the iminum ion follows immediately. Although hydrolysis of an iminium ion with the N atom connecting to an aromatic ring is a relatively slow reaction (40, 47), it may still kinetically compete with VOL. 39, NO. 12, 2005 / ENVIRONMENTAL SCIENCE & TECHNOLOGY

9

4481

the oxidation reaction at the N4 atom as described above. Afterward, similar oxidation and hydrolysis occurs at the N4 atom, leading to the production of the partially dealkylated M - 26 product. In a similar fashion, the M - 26 product is oxidized to its corresponding iminium ion at its inner N1 and finally hydrolyzed to the fully dealkylated M - 69 product (42). In the presence of an excess amount of MnO2, the M - 69 product can be further oxidized in a manner similar to oxidation of aniline, as shown by the slow decrease of this product over extended reaction time. On the basis of the earlier discussion on the reactivity of piperazinyl N atoms and the steric effects in radical coupling, the M - 69 product is expected to be less reactive to MnO2 but more likely to form dimeric products as aniline (26, 50) than its parent FQ. Because of the overall low mass balance in the product distribution analyses, the possibility of reaction pathways other than the three discussed above cannot be excluded (listed as pathway IV in Scheme 1). One example could be a pathway leading to the trace amount of N-oxide product (i.e., M + 16) detected for some of the FQs. Despite the high complexity of the surface radical reactions, one can expect that the likelihood of the N1 radical intermediate to undergo either of the aforementioned pathways will be affected by electronic and steric effects on the radical and result in different yields among the various products as indicated by the preliminary product distribution trends observed in this study. Among FQs, comparable steric effects are expected due to their highly similar structures and thus electronic effects likely exert the most influence. Evaluation of product distribution with analytical approaches that can account for most of the products will provide further insight to these effects. Environmental Implications. This study demonstrates high reactivity and fast reaction of FQs with manganese oxides. On the basis of the high affinity of FQs for soils and sediments (22), reactions with manganese oxides likely play an important role in the fate of these antibacterial agents in the soil-water environment. Reaction of FQs with manganese oxides yielded various N-dealkylated, hydroxylated, and possibly coupling oxidation products. Dealkylated products similar to those found in this study have also been identified in the FQ photodegradation mixtures (19, 37). A number of studies have examined the antimicrobial activity of FQ photodegradation products (15, 51, 52). Among those, no detectable antimicrobial activity was observed for the photodegradation products of CIP; in contrast, antimicrobial activity against Escherichia coli, Enterobacter cloacae, and Klebsiella oxytoca was statistically significant even after increased levels of photodegradation for OFL and levofloxacin (51). Fasani et al. (17) reported that OFL photodegrades much more slowly than CIP, ENR, and NOR due to its 5-oxygen electron-withdrawing substituent (versus 5-hydrogen substituent in the others) on the aromatic ring, and it undergoes primarily defluorination while the other three FQs undergo primarily dealkylation. Thus it may be postulated that the dealkylated products have much lower antimicrobial activity than the defluorinated products. If the above reasoning were indeed correct, the dealkylated oxidation products of FQs with manganese oxide would mean reduction in antimicrobial activity. Currently the antimicrobial activity of the hydroxylated products of FQs is not available and necessitates further investigation.

Acknowledgments This material is based upon work supported by the National Science Foundation under Grant 0229172. We thank Dr. Guofeng Li for assistance in FTIR analyses. 4482

9

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

Supporting Information Available Tables S1-6 and Tables S8-9 for LC/ESI-MS spectral information on the FQs, model amines, and their reaction products; Table S7 for distribution of oxidation products; Figure S1 for the effect of CIP and MnO2 loadings on the initial reaction rate; and Figure S2 for FTIR spectrum of PP oxidation products. This material is available free of charge via the Internet at http://pubs.acs.org.

Literature Cited (1) Walsh, C. Antibiotics: Actions, Origins, Resistance; ASM Press: Washington, DC, 2003. (2) Golet, E. M.; Alder, A. C.; Giger, W. Environmental exposure and risk assessment of fluoroquinolone antibacterial agents in wastewater and river water of the Glatt Valley Watershed, Switzerland. Environ. Sci. Technol. 2002, 36, 3645-3651. (3) Golet, E. M.; Alder, A. C.; Hartmann, A.; Ternes, T. A. G. W. Trace determination of fluoroquinolone antibacterial agents in urban wastewater by solid-phase extraction and liquid chromatography with fluorescence detection. Anal. Chem. 2001, 73, 36323638. (4) Renew, J. E. Novel analytical method development and fate assessment for fluoroquinolone, sulfonamide and trimethoprim antibiotics in engineered water treatment systems. M.S. Thesis, Department of Environmental Engineering, Georgia Institute of Technology, Atlanta, GA, 2003. (5) Renew, J. E.; Huang, C. H. Simultaneous analysis of fluoroquinolone, sulfonamide and trimethoprim antibiotics in wastewater using tandem solid-phase extraction and liquid chromatography electrospray mass spectrometry. J. Chromatogr. A 2004, 1042, 113-121. (6) Kolpin, D. W.; Furlong, E. T.; Meyer, M.; 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. (7) Hartmann, A.; Golet, E. M.; Gartiser, S.; Alder, A. C.; Koller, T.; Widmer, R. M. Primary DNA damage but not mutagenicity correlates with ciprofloxacin concentrations in German hospital wastewaters. Arch. Environ. Contam. Toxicol. 1999, 36, 115119. (8) Volmer, D. A.; Mansoori, B.; Locke, S. J. Study of 4-quinolone antibiotics in biological samples by short-column liquid chromatography coupled with electrospray ionization tandem mass spectrometry. Anal. Chem. 1997, 69, 9, 4143-4155. (9) Hartmann, A.; Alder, A. C.; Koller, T.; Widmer, R. M. Identification of fluoroquinolone antibiotics as the main source of umuC genotoxicity in native hospital wastewater. Environ. Toxicol. Chem. 1998, 17, 377-382. (10) Al-Almad, A.; Daschner, F. D.; Ku ¨ mmerer, K. Biodegradability of cefotiam, ciprofloxacin, meropenem, penicillin G, and sulfamethoxazole and inhibition of waste water bacteria. Arch. Environ. Contam. Toxicol. 1999, 37, 158-163. (11) Ku ¨ mmerer, K.; Al-Ahmad, A.; Mersch-Sundermann, V. Biodegradability of some antibiotics, elimination of the genotoxicity and affection of wastewater bacteria in a simple test. Chemosphere 2000, 40, 701-710. (12) Marengo, J. R.; Kok, R. A.; Burrows, L. A.; Velagaleti, R. R.; Stamm, J. M. Biodegradation of 14C-sarafloxacin hydrochloride, a fluoroquinolone antimicrobial by Phanerochaete chrysosporium. J. Sci. Ind. Res. 2001, 60, 121-130. (13) Parshikov, I. A.; Freeman, J. P.; Lay, J. O. J.; Beger, R. D.; Williams, A. J.; Sutherland, J. B. Regioselective transformation of ciprofloxacin to N-acetylciprofloxacin by the fungus Mucor ramannianus. FEMS Microbiol. Lett. 1999, 177, 131-135. (14) Wetzstein, H.-G.; Stadler, M.; Tichy, H.-V.; Dalhoff, A.; Karl, W. Degradation of ciprofloxacin by basidiomycetes and identification of metabolites generated by the brown rot fungus Gloeophyllum striatum. Appl. Environ. Microbiol. 1999, 65, 15561563. (15) Phillips, G.; Johnson, B. E.; Ferguson, J. The loss of antibiotic activity of ciprofloxacin by photodegradation. J. Antimicrob. Chemother. 1990, 26, 783-789. (16) Burhenne, J.; Ludwig, M.; Spiteller, M. Polar photodegradation products of quinolones determined by HPLC/MS/MS. Chemosphere 1999, 38, 1279-1286. (17) Fasani, E.; Rampi, M.; Albini, A. Photochemistry of some fluoroquinolones: effect of pH and chloride ion. J. Chem. Soc., Perkin Trans. 2 1999, 9, 1901-1907.

(18) Fasani, E.; Mella, M.; Monti, S.; Albini, A. Unexpected photoreactions of some 7-amino-6-fluoroquinolones in phosphate buffer. Eur. J. Org. Chem. 2001, 2, 391-397. (19) Mella, M.; Fasani, E.; Albini, A. Photochemistry of 1-cyclopropyl6-fluoro-1,4-dihydro-4-oxo-7-(piperazin-1-yl)quinoline-3-carboxylic acid (ciprofloxacin) in aqueous solutions. Helv. Chim. Acta 2001, 84, 2508-2519. (20) Lu ¨tzhøft, H.-C. H.; Vaes, W. H. J.; Freidig, A. P.; Halling-Sørensen, B.; Hermens, J. L. M. 1-Octanol/water distribution coefficient of oxolinic acid: influence of pH and its relation to the interaction with dissolved organic carbon. Chemosphere 2000, 40, 711-714. (21) Nowara, A.; Burhenne, J.; Spiteller, M. Binding of fluoroquinolone carboxylic acid derivatives to clay minerals. J. Agric. Food Chem. 1997, 45, 1459-1463. (22) Tolls, J. Sorption of veterinary pharmaceuticals in soils: A review. Environ. Sci. Technol. 2001, 35, 3397-3406. (23) Lu ¨tzhøft, H.-C. H.; Vaes, W. H. J.; Freidig, A. P.; Halling-Sørensen, B.; Hermens, J. L. M. Influence of pH and other modifying factors on the distribution behavior of 4-quinolones to solid phases and humic acids studied by “negligible-depletion” SPME-HPLC. Environ. Sci. Technol. 2000, 34, 4989-4994. (24) Stone, A. T. Reductive dissolution of manganese(III/IV) oxides by substituted phenols. Environ. Sci. Technol. 1987, 21, 979988. (25) Zhang, H.; Huang, C. H. Oxidative transformation of triclosan and chlorophene by manganese oxides. Environ. Sci. Technol. 2003, 37, 2421-2430. (26) Laha, S.; Luthy, R. G. Oxidation of aniline and other primary aromatic amines by manganese dioxide. Environ. Sci. Technol. 1990, 24, 363-373. (27) Pizzigallo, M. D. R.; Ruggiero, P.; Crecchio, C.; Mascolo, G. Oxidation of chloroanilines at metal oxide surfaces. J. Agric. Food Chem. 1998, 46, 2049-2054. (28) Li, H.; Lee, L. S.; Schulze, D. G.; Guest, C. A. Role of soil manganese in the oxidation of aromatic amines. Environ. Sci. Technol. 2003, 37, 2686-2693. (29) Taka´cs-Nova´k, K.; Nosza´l, B.; Hermecz, I.; Keresztu ´ri, G.; Poda´nyi, B.; Sza´sz, G., Protonation equilibria of quinolone antibacterials. J. Pharm. Sci. 1990, 79, 1023-1028. (30) Vzquez, J. L.; Berlanga, M.; Merino, S.; Domnech,.; Vias, M.; Montero, M. T.; Hernndez-Borrell, J., Determination by fluorimetric titration of the ionization constants of ciprofloxacin in solution and in the presence of liposomes. Photochem. Photobiol. 2001, 73, 14-19. (31) Barbosa, J.; Barrn, D.; Jimnez-Lozano, E.; Sanz-Nebot, V. Comparison between capillary electrophoresis, liquid chromatography, potentiometric, and spectrophotometric techniques for evaluation of pKa values of zwitterionic drugs in acetonitrile-water mixtures. Anal. Chim. Acta 2001, 437, 309321. (32) Barbosa, J.; Barro´n, D.; Cano, J.; Jime´nez-Lozano, E.; Sanz-Nebot, V.; Toro, I. Evaluation of electrophoretic method versus chromatographic, potentiometric, and absorptiometric methodologies for determining pKa values of quinolones in hydroorganic mixtures. J. Pharm. Biomed. Anal. 2001, 24, 1087-1098. (33) SPARC on-line calculator; U.S. EPA National Exposure Research Laboratory, Athens, GA; http://ibmlc2.chem.uga.edu/sparc/; accessed in 2004. (34) Shindo, H.; Huang, P. M. Role of manganese(IV) oxide in abiotic formation of humic substances in the environment. Nature 1982, 298, 363-365. (35) Murray, J. W. Surface chemistry of hydrous manganese dioxide. J. Colloid Interface Sci. 1974, 46, 357-371. (36) Nowack, B.; Stone, A. T. Homogeneous and heterogeneous oxidation of nitrilotrismethylenephosphonic acid (NTMP) in the presence of manganese(II,III) and molecular oxygen. J. Phys. Chem. B 2002, 106, 6227-6233.

(37) Burhenne, J.; Ludwig, M.; Nikoloudis, P.; Spiteller, M. Photolytic degradation of fluoroquinolone carboxylic acids in aqueous solution. Part I. Primary photoproducts and half-lives. Environ. Sci. Pollut. R. 1997, 4, 10-15. (38) Feng, W.; Liu, H.; Chen, G.; Malchow, R.; Bennet, F.; Lin, E.; Pramanik, B.; Chan, T.-M. Structural characterization of the oxidative degradation products of an antifungal agent SCH 56592 by LC-NMR and LC-MS. J. Pharm. Biomed. Anal. 2001, 25, 545557. (39) Seto, Y.; Guengerich, F. P. Partitioning between N-dealkylation and N-oxygenation in the oxidation of N,N-dialkylarylamines catalyzed by cytochrome P450 2B1. J. Biol. Chem. 1993, 268, 9986-9997. (40) Smith, M. B.; March, J. March’s Advanced Organic Chemistry, 5th ed.; John Wiley & Sons: New York, 2001. (41) Iley, J.; Tolando, R. The oxidative dealkylation of tertiary amides: mechanistic aspects. J. Chem. Soc., Perkin Trans. 2 2000, 11, 2328-2336. (42) Wang, F.; Sayre, L. M. Kinetics and mechanism of aliphatic amine oxidation by aqueous (batho)2CuII. J. Am. Chem. Soc. 1992, 114, 248-255. (43) Benjamin, M. M. Water Chemistry; McGraw-Hill: New York, 2002, p 488. (44) Bietti, M.; Cuppoletti, A.; Dagostin, C.; Florea, C.; Galli, C.; Gentili, P.; Petride, H.; Russo Caia, C. Stereoelectronic inhibition of deprotonation in the radical cation of N-benzylpiperidine. A contribution to the debate on the mechanism of N-dealkylation of tertiary amines. Eur. J. Org. Chem. 1998, 11, 2425-2429. (45) Smith, J. R. L.; Mortimer, D. N. Model systems for cytochrome P450-dependent monooxygenases. Part 5. Amine oxidation. Part 17. Oxidative N-dealkylation of tertiary amines by metalloporphyrin-catalyzed model systems for cytochrome P450 monooxygenases. J. Chem. Soc., Perkin Trans. 2 1986, 11, 17431749. (46) Cuppoletti, A.; Dagostin, C.; Florea, C.; Galli, C.; Gentili, P.; Lanzalunga, O.; Petride, A.; Petride, H. The oxidation of N-benzylaziridine catalyzed by iron porphyrin: radical versus electron-transfer mechanism. Chem. Eur. J. 1999, 5, 2993-2999. (47) Carey, F. A.; Sundberg, R. J. Advanced Organic Chemistry. Part A. Structure and Mechanisms, 4th ed.; Kluwer Academic/Plenum Publishers: New York, 2000. (48) Adler, E.; Becker, H.-D. Zur selektiven oxidation von benzylalkoholen. Acta Chem. Scand. 1961, 15, 849-852. (49) Ohloff, G.; Giersch, W. Conversion of vicinal diols into dicarbonyl compounds by manganese dioxide. Angew. Chem., Int. Ed. Engl. 1973, 12, 401-402. (50) Stassen, I.; Hambitzer, G. Anodic oxidation of aniline and N-alkylanilines in aqueous sulfuric acid studied by electrochemical thermospray mass spectrometry. J. Electroanal. Chem. 1997, 440, 219-228. (51) Sunderland, J.; Tobin, C. M.; Hedges, A. J.; MacGowan, A. P.; White, L. O. Antimicrobial activity of fluoroquinolone photodegradation products determined by parallel-line bioassay and high performance liquid chromatography. J. Antimicrob. Chemother. 2001, 47, 271-275. (52) Sunderland, J.; Tobin, C. M.; White, L. O.; MacGowan, A. P.; Hedges, A. J. Ofloxacin photodegradation products possess antimicrobial activity. Drugs 1999, 58, 171-172. (53) Perrin, D. D. IUPAC. Dissociation constants of Organic Bases in Aqueous Solution: Supplement 1972; Crane, Russak: New York, 1972.

Received for review November 22, 2004. Revised manuscript received March 25, 2005. Accepted April 11, 2005. ES048166D

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

9

4483