Science of the Total Environment 470–471 (2014) 299–310

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Photodegradation of antibiotics under simulated solar radiation: Implications for their environmental fate Sudha Rani Batchu a,b, Venkata R. Panditi a,b, Kevin E. O'Shea a, Piero R. Gardinali a,b,⁎ a b

Department of Chemistry and Biochemistry, Florida International University, Miami, FL 33199, USA Southeast Environmental Research Center, Florida International University, Miami, FL 33199, USA

H I G H L I G H T S • • • • •

Photolysis was studied using environmentally relevant conditions. Photolysis of antibiotics was studied in natural water matrices. Integrative effects of matrix composition on the rate of photolysis were reported. First report showing conversion of roxithromycin to erythromycin at UV 254 nm 15 photo transformation products were identified for the studied antibiotics.

a r t i c l e

i n f o

Article history: Received 5 July 2013 Received in revised form 19 September 2013 Accepted 19 September 2013 Available online xxxx Editor: Damia Barcelo Keywords: Photodegradation Antibiotics Simulated solar radiation Intermediates Photolysis products High resolution mass spectrometry

a b s t r a c t Roxithromycin, erythromycin, ciprofloxacin and sulfamethoxazole are frequently detected antibiotics in environmental waters. Direct and indirect photolysis of these problematic antibiotics were investigated in pure and natural waters (fresh and salt water) under irradiation of different light sources. Fundamental photolysis parameters such as molar absorption coefficient, quantum yield and first order rate constants are reported and discussed. The antibiotics are degraded fastest under ultraviolet 254 nm, followed by 350 nm and simulated solar radiation. The composition of the matrix (pH, dissolved organic content, chloride ion concentration) played a significant role in the observed photodegradation. Under simulated solar radiation, ciprofloxacin and sulfamethoxazole degrade relatively quickly with half-lives of 0.5 and 1.5 h, respectively. However, roxithromycin and erythromycin, macrolides are persistent (half-life: 2.4–10 days) under solar simulation. The transformation products (15) of the targeted antibiotics produced under irradiation experiments were identified using high resolution mass spectrometry and degradation pathways were proposed. © 2013 Published by Elsevier B.V.

1. Introduction Antibiotics are chemotherapeutic agents used in human and veterinary medicine to prevent and treat microbial infections (Sarmah et al., 2006). Antibiotics enter the environment from many sources including household sewer systems, hospital effluents, agricultural runoffs, manufacturing process, livestock, and disposal of unused and expired products (Kummerer, 2001). Hence, they have been routinely detected in environmental waters at concentrations up to 1.3 μg/L (Batt and Aga, 2005; Brown et al., 2006; Watanabe et al., 2008; Zuccato et al., 2010). The United States Food and Drug Administration (USFDA) reported that 3.28 million kilograms of antibiotics was sold for human medicinal ⁎ Corresponding author at: 3000 NE 151st ST, FIU Biscayne Bay Campus, MSB-350, North Miami, Florida 33181, USA. Tel.: +1 305 348 6354; fax: +1 305 348 3772. E-mail addresses: sbatc001@fiu.edu (S.R. Batchu), vpand001@fiu.edu (V.R. Panditi), osheak@fiu.edu (K.E. O'Shea), gardinal@fiu.edu (P.R. Gardinali). 0048-9697/$ – see front matter © 2013 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.scitotenv.2013.09.057

use in 2010 (Pham, 2012). Prevalence and persistence of antibiotics in the aqueous environments are of particular ecological concern because of their biological activity and potential to induce bacterial resistance in natural populations (Wright, 2010). Antibiotic resistance is a serious problem and widespread in wastewaters, environmental waters, marine sediments and in fish (Andersen and Sandaa, 1994; Costanzo et al., 2005; Miranda and Zemelman, 2001; Witte, 1998). Antibiotic resistance to widely used medical and veterinary medications is a serious problem and poses a significant threat to the health humans and livestock infected with the resistant bacterial strains. Particularly alarming is the fact that resistant genes can be mobilized between various environmental compartments and transferred into the food chain (Chee-Sanford et al., 2001). Although the exposure periods required to induce bacterial resistance to antibiotics are largely unknown, the nature and environmental concentrations of antibiotics have a pronounced influence on the ability to induce such resistance. The environmental concentration and thus

300

S.R. Batchu et al. / Science of the Total Environment 470–471 (2014) 299–310

Table 1 Target antibiotics, structures, CAS and their classification. Name

Structure

Sulfamethoxazole

CAS

Classification

723-46-6

Sulfonamide antibiotic

Roxithromycin

80214-83-1

Macrolide antibiotic

Erythromycin

114-07-8

Macrolide antibiotic

Ciprofloxacin

85721-33-1

Fluoroquinolone antibiotic

the ability to induce resistance are strongly correlated to antibiotic stability and natural attenuation through hydrolysis, sorption, biotransformation and photodegradation. Unfortunately, the common processes associated with sewage treatment, hydrolysis, biodegradation and sorption are ineffective for the removal of many antibiotics (Gartiser et al., 2007) and thus photodegradation may be the predominant transformation pathway for antibiotics in the environment. In the recent years, the photodegradation of antibiotics was studied using different light sources including UV light (Pereira et al., 2007), solar simulation (Ge et al., 2010) and natural sunlight (Avisar et al., 2010) for a variety of water matrices including pure water, fresh water (from rivers and lakes) and salt water (Sirtori et al., 2010). The effectiveness of photodecomposition depends on the integrative effects of photon flux, the structure of the molecule and water-matrix composition. For example, the photodegradation of five membered sulfonamides in natural sunlight is solely due to pH dependent direct photolysis (faster) whereas in nitrate-enriched waters it is due to the combination of direct and indirect photolysis (slower) (Boreen et al., 2004). Sirtori et al. (2010) studied the photodegradation kinetics of trimethoprim in demineralized and simulated salt water in a solar simulator and reported that the degradation rate was hindered by the salt content of the water matrix. Moreover, they reported that the nature and abundance of photo-transformation products varied with water matrices (Sirtori et al., 2010). Avisar et al. (2010) reported that photodecomposition rates by UV photolysis decreased for sulfamethoxazole, but was enhanced for oxytetracycline and ciprofloxacin when the pH of the water was increased

Table 4 Optimized parameters for the detection of all analytes and internal standard in MS/MS SRM mode. Analyte

ESI parameter Precursor ion

Sulfamethoxazole Roxithromycin Erythromycin Ciprofloxacin Sulfathiazole

[M + [M + [M + [M + [M +

H]+ H]+ H]+ H]+ H]+

m/z SRM 1 SRM 2

Optimal collision Rt (min) energy

254 837 734 332 256

32 32 30 55 30

188 679 558 314 156

194 522 576 288 190

11.5 11.7 10.4 5.87 8.21

from 5 to 7. Stangroom et al. (1998) showed that dissolved organic matter in natural waters can inhibit or enhance the rate of photolysis of organic compounds. Clearly, past studies suggest that photochemical transformation of antibiotics is a complex process and not easily predicted. In fact, most of the photodegradation studies for antibiotics were performed in amended matrices enriched with humic acids, particulate organic matter, oxidants like titanium dioxide, metal ions like Fe3+ or in controlled pH solutions (Belden et al., 2007; Boreen et al., 2004; Paul et al., 2010; Vione et al., 2009; Yan et al., 2013). In contrast to previous studies, kinetic experiments were performed in natural water matrices here. The rate constants obtained in our studies are important for the fundamental understanding required to predict the environmental fate and concentration of these antibiotics in natural waters and drinking water sources. It is also important to understand the effects of water quality on the enhancement/attenuation of the degradation processes involved in the natural transformation of antibiotics. With this in mind, we studied the photolysis of four antibiotics frequently detected in surface waters. Sulfamethoxazole [SMX] is a sulfonamide antibiotic, commonly used in humans in synergistic combination with trimethoprim (under the trade name Bactrim) to treat urinary tract infections and in veterinary medicine as growth promoters (Crosby, 1991). Roxithromycin [ROX] and erythromycin [ERY] are macrolide antibiotics widely used to treat respiratory tract infections (Cals et al., 2008). Ciprofloxacin [CIP], is a fluoroquinolone antibiotic and is one of the most popular fluoroquinolone antibiotic used in human medicine (Giger et al., 2003). Detailed structural information on the selected antibiotics is shown in Table 1. Photolyses were conducted on the selected antibiotics in matrices modeled to natural water bodies using specific light sources. Reported herein is the photolysis of four problematic antibiotics in pure and natural water matrices under 254 nm, 350 nm and solar simulated light. Our results provide a better fundamental understanding of the photochemical conversion of antibiotics in aqueous media and provide insight about the potential application of photochemical water treatment of problematic antibiotics.

Table 2 Characteristics of canal water and salt water used in the experiment. Parameter

Canal water

Salt water

pH Dissolved organic carbon (mg-C/L) Electrical conductivity (μS/cm) Salinity (ppt)

8.1 10.4 544 0.2

7.9 1.37 88,000 36.0

Table 3 Summary of the gradient program for the chromatographic separation. Time (min)

% solvent A (methanol)

% solvent B (0.3% formic acid in water)

0 13 16 25

20 90 20 20

80 10 80 80

Fig. 1. SRM chromatograms of canal water spiked with target antibiotics and internal standard at a concentration of 100 μg/L.

S.R. Batchu et al. / Science of the Total Environment 470–471 (2014) 299–310

2. Experimental section 2.1. Chemicals Standards of sulfamethoxazole, roxithromycin, erythromycin and ciprofloxacin and sulfathiazole were purchased from Sigma-Aldrich (St. Louis, MO, USA). Sulfathiazole was used as an internal standard for quantification and was also purchased from Sigma-Aldrich (St. Louis, MO, USA). Standard stock solutions (100 μg/mL) of all the compounds were prepared in methanol and stored in the dark at b4 °C. Ultrapure water (H2O), methanol (CH3OH) (all of optima grade or equivalent) and Whatman GFF glass fiber filters were obtained from Fisher Scientific (Fair Lawn, NJ). Reverse osmosis deionized water (RODW) was produced in the laboratory using a Milli-Q RG water system from Millipore. 2.2. Sample collection Canal water (fresh water, CW) was collected from Tamiami Canal at its confluence with the Miami River in Miami, FL. Salt water (SW) was taken from the shore at Bill Baggs State Park at Key Biscayne, Miami, FL. Canal and salt water characteristics are shown in Table 2. Both natural waters were filtered using a 0.2 μm membrane fiber filter to remove any particles and microorganisms and stored in the dark at b 4 °C until the experimental solutions were prepared, typically within a month. 2.3. Comparison of light sources employed in the study Photochemical experiments were conducted using Rayonet UV photochemical reactors (Southern New England Ultraviolet Co., Branford, CT) and a Sun Test XLS Tabletop Xenon Exposure System (ATLAS Material Testing Technology LLC, Chicago, Illinois, USA).

301

The photochemical reactor can be operated with up to 16-mercury vapor lamps (UV 254 nm) or black light phosphor bulbs (UV 350 nm). UV 254 nm radiation is a commonly used light source to induce photolysis of a wide variety of organic compounds for wastewater treatment (Thomas, 2003) and hence included in the present study to assess the likelihood of the antibiotic to survive water treatment during advanced treatment process. UV 350 nm is comparable to the range of UVA region (315 nm–400 nm) of sunlight and hence commonly used to predict the UV initiated photodegradation of pharmaceutical compounds in environment (Lam et al., 2003; Radjenovic et al., 2009; Sturini et al., 2009). Sunlight plays an important role in the persistence and environmental fate of antibiotics. However, intensity of natural sunlight depends on latitude (Li et al., 2002), depth of the photic zone and overcast conditions. Because of this, experiments conducted with sunlight take longer and may pose a challenge in comparing the data among different studies. To overcome that experiment was conducted with a SunTest, which is a surrogate of natural sunlight. The SunTest XLS produces continuum of wavelengths from 300 nm to 800 nm by using a properly filtered Xenon lamp on the top of the exposure chamber. The xenon lamp was used at its maximum intensity (750 W/cm2) for macrolide antibiotics i.e. erythromycin and roxithromycin. Both ciprofloxacin and sulfamethoxazole degraded almost completely (in less than 2 h), to obtain at least 5 data points in the degradation curve, the experiment was repeated under a xenon lamp with its minimal abundance (250 W/cm2). A spectral comparison of light sources employed in this study can be found elsewhere (Batchu et al., 2013). 2.4. UV absorbance spectra versus type of water matrix The pH of water matrices used in this study is listed in Table 2 and ranged from 5.5 (RODW) to 8.1 (CW). The pH of the matrix can

Fig. 2. Absorbance spectra of 5 μg/mL solutions of sulfamethoxazole, roxithromycin, erythromycin and ciprofloxacin in various matrices.

302

S.R. Batchu et al. / Science of the Total Environment 470–471 (2014) 299–310

influence speciation of the antibiotic, absorbance and photolytic degradation at specific wavelengths. Thus pH of the water sample and speciation (governed by the pKa) of the antibiotic can influence absorbance of light. In addition to pH, other two major differences between matrices are dissolved organic content (DOC, CW being rich) and ionic strength (SW being the highest). Organic matter present in the natural water matrices may act as a photosensitizer, light attenuator or quencher and therefore have a pronounced affect on the photolytic transformation of a variety of substrates (Stangroom et al., 1998). As the rate of photolysis of a compound in natural waters is likely be influenced by the combination of DOC, pH and ions in the medium, to study their integrative effects on light absorption, UV–visible spectra were recorded using standard solutions of antibiotics (at 5 μg/mL) in RODW, CW and SW matrices. UV–visible measurements were carried out with 1 cm quartz at room temperature (20 °C), using a Shimadzu UV-2101PC

UV–visible double beam spectrophotometer. MilliQ water was used as a reference. 2.5. Quantum yields The solution pH influences speciation and thus light absorbance properties of the antibiotics. Therefore quantum yields (QYs) were experimentally calculated for the photolysis of the studied antibiotics using 254 and 350 nm light sources in RODW and natural waters using the procedure reported by Pereira et al. (2007). Molar absorption coefficient of the target antibiotic was calculated by dividing its absorbance at a specific wavelength by its molar concentration and optical path length. QYs were calculated as the ratio of first order rate constant (s−1) to the specific rate of light absorption by the compound (Eqs. (1) & (2)). The irradiance intensity of 16-mercury vapor lamps (UV 254 nm) and

Fig. 3. Photolysis decay curves of sulfamethoxazole at a1) 254 nm a2) 350 nm a3) solar simulation; roxithromycin at b1) 254 nm b2) 350 nm b3) solar simulation. RODW = reverse osmosis deionized water CW = canal water SW = salt water.

S.R. Batchu et al. / Science of the Total Environment 470–471 (2014) 299–310

black light phosphor bulbs (UV 350 nm) is 233 and 844 μW/cm2/nm, respectively. 0

ϕ254=350

nm ½antiobiotic




KS λ254=350

 nm

¼

¼

kd ½antiobiotic
 ks λ254=350 nm


0 Ep λ254=350

nm

ð1Þ

 h −a λ ελ254=350 nm Þ 1−10 ð 254=350
 a λ254=350 nm z

nm

Þz

i

ð2Þ where k'd represents the time-based pseudo-first-order rate constant for the individual antibiotic at specific wavelength in a specific matrix. Ks (λ) represents the specific rate of light absorption by the compound, E0p (λ) the incident photon irradiance, ε(λ) the molar absorption coefficient which measures the probability that a compound will absorb light at a certain wavelength, a(λ) the solution absorbance, z the optical path length of the solution (1 cm), and ϕ the quantum yield. The incident photon irradiance was measured with an Ocean Optics spectrophotometer equipped with cosine corrector. 2.6. Sample irradiation The experimental working solutions of all antibiotics were prepared by diluting each stock solution to 1 mg/L with the three types of water: a) RODW b) CW and c) SW. Seven 30 mL-quartz tubes (Southern New England Ultraviolet Co., Branford, CT) were used for experiments in the photochemical reactor and one tube filled with 30 mL of RODW was used as blank. Dark controls (3 tubes) were totally covered with aluminum foil to prevent light exposure. The remaining three tubes were exposed to light. All the seven tubes were closed with quartz stoppers throughout the experiment. The reaction vessels were placed on a merry-go-round to ensure uniform irradiation in the Rayonet UV photochemical reactor chamber. For experiments employing the SunTest light source, 25 mL of the solution was placed in six UV transparent polyethylene bags (Nasco WHIRL-PAK 2 OZ) as described in previous studies. Three reaction bags covered with aluminum foil were used as dark control experiments, three bags uncovered and one bag filled with RODW as a blank. All bags were then placed in a water bath with circulating water to maintain a constant temperature (25 °C) throughout the photolysis. At specified time intervals, 500 μL aliquots were transferred into 2 mL amber glass vials, sulfathiazole (500 ng/mL) was added and samples were analyzed subsequently by liquid chromatography tandem mass spectrometry (LC–MS/MS).

303

2.7. Liquid chromatography/mass spectrometry A Thermo LCQ advantage max Ion trap Mass spec (IT-MS) equipped with an electrospray ionization source (ESI) and connected to surveyor LC system (Thermo Finnigan, San Jose, CA) was used for sample analysis. Separations were achieved on a Luna C18 column (150 cm × 4.6 mm × 5 μm particle size) equipped with a Luna C18 guard column, both purchased from Phenomenex (Torrance, CA). The pump was operated at a flow rate of 0.5 mL/min. The column oven temperature was 30 °C, and the full loop injection volume was 20 μL. The separation was performed using a simple binary gradient mobile phase consisting of methanol (A) and water with 0.3% formic acid (B) (Table 3). The ESI-IT-MS/MS was operated in positive ion mode. The capillary temperature was 315 °C, and the spray voltage was 4.5 kV. Nitrogen was used as a sheath gas and as an auxiliary gas at a flow rate of 59 and 20 arbitrary units, respectively. Collision energy is an important factor in the collision induced dissociation (CID) process by which fragment ions are produced. Optimized values of collision energy and fragment ions monitored for each antibiotic are shown in Table 4. Two SRM transitions were monitored for the identification and quantitation of target analytes. The SRM chromatograms obtained of canal water fortified at a concentration of 100 μg/L are shown in Fig. 1.

2.8. Identification of metabolites Product analysis was run following the photolysis of 1 μg/mL solutions of roxithromycin, erythromycin, ciprofloxacin at 254 nm and sulfamethoxazole at 350 nm in RODW. Blanks and dark controls described earlier were also analyzed. Samples were analyzed by direct injection into the LC/MS system. The analytes were separated on a Hypersil Gold column (Thermo Scientific, San Jose, CA, USA) using a gradient mobile phase consisting of acetonitrile and 0.1% formic acid in water. Data was first acquired in full scan mode at 35,000 resolution on a Thermo Scientific Q Exactive orbitrap mass spectrometer using a heated electrospray ionization source operating in positive ionization mode. The operating conditions of the mass spectrometer include: sheath gas flow: 35 arbitrary units; auxiliary gas flow: 10 arbitrary units; spray voltage: 3.1 kV; capillary temperature: 320 °C; S-lens RF level: 50; and heater temperature: 300 °C. Photolysis products identified in full scan were further investigated using targeted MS/MS experiments. Precursor ions were isolated and analyzed under the following conditions: resolution: 35,000; AGC target: 2E4; maximum IT: 50 ms; isolation width: 2.0; and NCE: 20–50.

Table 5 Molar absorption coefficient and quantum yields of sulfamethoxazole (SMX), roxithromycin (ROX), erythromycin (ERY) and ciprofloxacin (CIP) in RODW and natural water matrices measured under a254 nm and b350 nm. pKa values were obtained from SRC database (accessed on 1/18/2013) and (Qiang and Adams, 2004). Pseudo first order rate constants measured at c254 nm, d 350 nm and eSunTest. RODW = reverse osmosis deionized water SW = salt water CW = canal water. Compound

pKa

Matrix

Molar absorption coefficienta (M−1 cm−1)

Molar absorption coefficientb (M−1 cm−1)

Quantum yielda (mol einstein−1)

Quantum yieldb (mol einstein−1)

Rate constant (h−1)c

Rate constant (h−1)d

Rate constant (h−1)e

SMX

1.6, 5.8

ROX

9

ERY

8.7

CIP

5.9, 8.9

RODW SW CW RODW SW CW RODW SW CW RODW SW CW

24,018 19,958 23,909 24,139 9599 27,153 21,717 11,778 16,582 14,911 17,650 17,729

3834 2372 3537 11,175 4973 15,999 10,218 6611 8181 4074 2446 1684

1.23 1.37 1.24 0.00120 0.0492 0.0109 0.000300 0.00300 0.00450 0.159 0.832 0.297

1.08 0.737 0.375 0.000200 0.000700 0.000400 0.000200 0.000400 0.000700 0.157 1.55 4.02

73.8 73.8 73.6 0.103 1.01 1.81 0.0217 0.272 0.133 7.58 16.1 45.0

15.5 5.02 6.79 0.00730 0.0231 0.0146 0.00640 0.0231 0.00904 2.44 27.0 14.9

3.05 0.209 0.477 0.00570 0.0117 0.0212 0.0123 0.00990 0.00290 0.415 1.60 3.03

304

S.R. Batchu et al. / Science of the Total Environment 470–471 (2014) 299–310

Fig. 4. Photolysis decay curves of erythromycin at c1) 254 nm c2) 350 nm c3) solar simulation; ciprofloxacin at d1) 254 nm d2) 350 nm d3) solar simulation. RODW = reverse osmosis deionized water CW = canal water SW = salt water.

3. Results and discussion 3.1. UV absorbance spectra versus type of water matrix The absorption spectra for the selected antibiotics are shown in Fig. 2. The compounds absorb UV light between 200 and 350 nm. However, at equivalent concentrations (5 μg/mL) the intensity and the wavelength of the absorption maximum vary among the antibiotics. The absorbance maximum (λmax) for sulfamethoxazole is 255 nm, ciprofloxacin has λmax at 275 and 325 nm in RODW, while both macrolides do not exhibit distinct absorbance maxima under the experimental conditions. The solution pH can influence the absorbance spectrum i.e., ciprofloxacin with at λmax at 275 nm in RODW (pH 5.5) undergoes a hypsochromic shift to 270 nm in CW (pH 8.1). An analogous trend was reported by Avisar et al. (2010).

Fig. 5. Kinetic profiles of roxithromycin and its degradation products.

S.R. Batchu et al. / Science of the Total Environment 470–471 (2014) 299–310

305

Table 6 Elemental composition and mass accuracy of all studied antibiotics and their degradation products. Compound Metabolite Elemental formula ROX R1 R2 CIP C1 C2 C3 C4 C5 C6 C7 C8 ERY E1 E2 E3 E4 SMX S1

Fig. 6. HRMS spectra of a) roxithromycin at T0 b) roxithromycin after 20 h of irradiation at 254 nm c) MS/MS of the R1 (erythromycin, m/z 734.46).

3.2. Photolysis rate constants, nature of matrix and quantum yield The photodegradation of the different compounds were conducted by monitoring the concentration of the parent compound as a function of irradiation time. Photolysis decay curves of the antibiotics using different light sources were constructed by plotting Ln (Ct/C0) versus time of irradiation, where Ct is the concentration of substrate at any given point of time and C0 is the initial substrate concentration. Sigmaplot v11.0 software was used to plot the degradation curves and each data point in Fig. 3 is an average of the triplicate measurements using standard deviation as error bars. The substrate decay is nicely fit to pseudo-first-order kinetics and the slope used to obtain the pseudo-first-order rate constant (k). No degradation was observed in dark controls. The pH of the natural waters typically falls between 6 and 9, which is within the range of the pKa's of target analytes and thus can affect speciation of the compounds. The pKa values (SRC, 2013) and the pseudo-first-order rate constants of target antibiotics

C41H77N2O15 C37H68NO13 C29H54NO10 C17H19FN3O3 C17H19FN3O4 C17H20N3O5 C17H20N3O4 C16H18N3O4 C15H18N3O2 C15H18N3O3 C17H20N3O3 C13H13N2O3 C37H68NO13 C34H64NO12 C29H54NO12 C29H52NO11 C23H41O7 C10H12N3O3S C10H14N3O

Mass (m/z) Theor.

Exp.

837.5318 734.4685 576.3742 332.1405 348.1354 346.1397 330.1448 316.1292 272.1394 288.1343 314.1499 245.0921 734.4685 678.4423 608.3641 590.3535 429.2847 254.0594 192.1131

837.5324 734.4691 576.3748 332.1405 348.1354 346.1394 330.1444 316.1287 272.1391 288.134 314.1499 245.0917 734.4687 678.4429 608.3647 590.3542 429.2825 254.0593 192.1131

Error DBE Rt, min [ppm] 0.7 0.8 1.0 0.0 0.0 −0.9 −1.2 −1.6 −1.1 −1.0 0.0 −1.6 0.3 0.9 1.0 1.2 −5.1 −0.4 0.0

4.5 4.5 3.5 9.5 9.5 9.5 9.5 9.5 8.5 8.5 9.5 8.5 4.5 3.5 3.5 4.5 3.5 6.5 5.5

6.9 6.6 6.2 5.1 6.4 4.4 6.1 5.7 1.6 2.2 3.5 5.7 6.6 6.4–6.9 6.1 6.2 8.2 6.0 1.1

are listed in Table 5. In natural waters employed for our studies the target antibiotics each exist predominantly in the cationic form except for SMX which is anionic. The photolysis of SMX with 254 nm light results in the complete degradation within 1 min. Even with the reduction of the 254 nm light intensity by 75% the degradation was fast. The half-life of SMX in all water matrices is less than 2 min and thus under our experimental conditions too rapid to distinguish the difference in rates among the three matrices (Fig. 3a). The rate constants for SMX shown in Table 5 were corrected for 16 lamps (4 × assuming a linear relationship between k and intensity). The kinetic data for the photolysis under 350 nm and SunTest irradiation demonstrates that SMX is more stable at natural water pH in an anionic form. This observation is in good agreement with other studies (Avisar et al., 2010) (Fig. 3a2–a3), suggesting that photolysis rate is dependent on speciation in the given water sample. The molar absorption coefficient obtained at 254 nm and pH 7.9 (SW) in this study was 19,958 M− 1 cm− 1, similar to literature value for solution of pH 7.85 (16,580 M− 1 cm− 1) (Baeza and Knappe, 2011). The single wavelength quantum yields measured at 350 nm for sulfamethoxazole decrease with increasing solution pH, similar trends were observed by Boreen et al. (2004). In spite of minimal UV absorbance between 200 and 350 nm, macrolides showed moderate degradation under 254 nm irradiation (Figs. 3b1, 4c1). The degradation rates at 254 nm were up to 125 fold higher compared to those at observed under 350 nm and SunTest irradiation (Table 5). Photolysis of the roxithromycin in natural waters revealed significant enhancement in the rates in natural waters relative to that observed in RODW as illustrated in Fig. 3b1–b3, indicating the existence of important indirect or photosensitized photochemical process in the natural waters. The fresh water matrix (CW) has high dissolved organic carbon that can absorb a wide range of wavelengths and act as a

Fig. 7. Possible photo transformation pathways of roxithromycin in water under 254 nm irradiation.

306

S.R. Batchu et al. / Science of the Total Environment 470–471 (2014) 299–310

photosensitizer in the transformation of roxithromycin through energy transfer or by generation of reactive oxygen species, like hydroxyl radical (Burns et al., 2012). Rapid oxidation of roxithromycin by hydroxyl radical during ozonation process has been previously documented (Dodd et al., 2009). An analogous macrolide, azithromycin also undergoes degradation in the presence of a photosensitizer (humic acid) under simulated and natural solar light (Tong et al., 2011). The increased rates observed in salt water (SW) may possibly be attributed to abundant chloride ions, which in the presence of hydroxyl radical (OH) can form Cl− 2 (Vione et al., 2005), a strong oxidant for the destruction of antibiotics in surface waters (Chamberlain and Adams, 2006). The exact mechanism (3DOM⁎ or OH or Cl− 2 ) responsible for indirect photolysis process in natural waters can be complex and is beyond the scope of the current study. Because erythromycin and roxithromycin have similar functional groups as macrolide antibiotics, similar photochemical behavior may be expected. However, kinetic data shows that the rates of photolysis of erythromycin in natural waters relative to RODW depend on the light source. The rates were enhanced in natural waters relative to RODW for roxithromycin for all light sources. Variation in the behavior between roxithromycin and erythromycin can be attributed to the structural difference at the C-9 substituent position (Table 1), which is also responsible for the observed differential effects on growth inhibition in Haemophilus influenzae (Mabe et al., 2004). Under our experimental conditions the macrolides were degraded with half-lives of 3–10 days in natural waters under solar simulation in agreement with previous studies reporting half-lives in the range of 1–40 days (Tong et al., 2011; Vione et al., 2009). In the natural environment, sunlight intensity is greatly reduced with increasing water depth and hence these macrolides distributed in the water column would be expected to persist much longer than what is observed in our study.

Ciprofloxacin also showed enhanced degradation (2–11 fold higher) in natural waters compared to RODW (Fig. 4d1–d3). The increase in the photolysis in natural waters could be attributed to the [Cip-Fe3+] complex formation resulting in oxidative degradation of the antibiotic, commonly observed for fluoroquinolone antibiotics in adsorption and oxidation studies with goethite (Zhang and Huang, 2007). Ciprofloxacin is degraded very quickly with a half-life of about 14 min in CW in the SunTest. Under very similar conditions a structurally similar fluoroquinolone antibiotic, enrofloxacin was estimated to degrade with a half-life of 7 min (Sturini et al., 2009). Lam et al. (2003) reported a half-life of 11–15 min in deionized water when using xenon lamp with maximum intensity (765 W/cm2) (however the study also reported that the filter is not very efficient in removing wavelengths b 290 nm and hence there may be a systematic error in the results). These reported data correlates favorably with our results (100 min at 765 W/cm2). Molar absorption coefficients obtained in this study at 254 nm in RODW (14,911 M−1 cm−1), were within ±15% of the same reported by Pereira et al. (2007) under the same conditions (254 nm, laboratorygrade water, 12,400 M−1 cm−1). In a recent study we reported the analysis of antibiotics in reclaimed and surface waters in South Florida, sulfamethoxazole and erythromycin were detected frequently (80–100%) and at rather higher concentrations (up to 400 ng/L) (Panditi et al., 2013). Sulfamethoxazole higher frequency of detection despite of its low photo-stability may be explained based on its high consumption rates (Pham, 2012). Macrolides are of high ecological concern because of their high resistance to bio- (Gartiser et al., 2007) and photo-degradation (this study). Their high toxicity towards blue green algae (EC50 = 0.002 mg/L for erythromycin in Pseudokirchneriella subcapitata (Isidori et al., 2005) and their ability to enzymatically inhibit the cytochrome P-450 leading to adverse

Fig. 8. HPLC chromatograms of photodegradation products of ciprofloxacin.

S.R. Batchu et al. / Science of the Total Environment 470–471 (2014) 299–310

307

drug reactions in humans is a serious concern (Sagir et al., 2003). A positive correlation between macrolides use and penicillin nonsusceptible and macrolide resistant pneumococcal isolates was reported by Albrich et al. (2004). Although both macrolides were equally photostable, roxithromycin is rarely detected in surface waters likely due to the fact that it was only recently approved for use in United States (NIH, 2013).

3.3. Metabolite identification and degradation pathway of studied antibiotics Antibiotics are specially designed to cause a physiological effect. They have specific effects not only on humans and/or animals, but also on non-target aquatic organisms. In the environment they will be subjected to different attenuation processes including photolysis. During the process of degradation some antibiotics may form stable byproducts which could maintain biological activity (Bonvin et al., 2013) and/or increase the sensitivity of bacterial strains (Palmer et al., 2010). Because of this, an attempt was made to identify the by-products of the target antibiotics produced during the irradiation experiments using high resolution mass spectrometry. For roxithromycin, two new compounds were detected in the irradiated samples with m/z 734 (major) and m/z 576 (minor). The kinetic profiles of formation of the degradation products are shown in Fig. 5. Accurate mass measurements showed spectra corresponding to masses m/z 734.4691 (R1) and m/z 576.3748 (R2) (Fig. 6). The calculated mass error (ppm) for the phototransformation products is also shown in the figure. Roxithromycin produced a sodium adduct ion observed at m/z 859.5133. The proposed pathway for the formation of degradation products is shown in Fig. 7. In the presence of water, hydrolysis of the imine nitrogen on C-9 substituent position of roxithromycin results in the formation of carbonyl group in erythromycin. The formation of erythromycin (m/z 734.4691, C37H68NO+ 13) from roxithromycin by UV photolysis has yet to be reported. The R2 (m/z 576.3742, C29H54NO+ 10) was obtained by the loss of cladinose sugar moiety which had been reported in the literature (Kim et al., 2004). Elemental composition and mass error of all degradation products are shown in Table 6. It should be noted that in both ROX degradation products the lactone ring and desosamine sugar moiety that are responsible for the macrolide antimicrobial activity through the formation of hydrogen bonds with rRNA, remain intact (Mabe et al., 2004). The irradiation of the fluoroquinolone antibiotic, ciprofloxacin produced 8 degradation products (C1–C8) and the ion chromatograms are shown in Fig. 8. The relative peak areas of all products (except for m/z 272) exhibited similar time trends i.e. they increase rapidly to the highest intensity in the first 1.5 h and then decrease with increased time (Fig. 9a and b). The accurate mass, experimental observed mass, mass error (ppm) and retention time of the parent and its photolytic products are summarized in Table 6. The proposed degradation pathway of ciprofloxacin is shown in Fig. 10. The intermediate with m/z 348.1354 (C1) is a monohydroxylated product of the parent molecule. The same product was also identified in advanced oxidation studies using hydroxyl and azide free radicals (An et al., 2010). C2, with m/z 346.1397 is a dihydroxylated product of ciprofloxacin. This observation is in good agreement with a previous study where ciprofloxacin was biodegraded by fungi (Wetzstein et al., 1999). It is produced by the substitution of F atom with a hydroxyl group in C1. Among the eight identified products, C3 (m/z 330.1448) and C4 (m/z 316.1292) are the major ones. C3, with 2 mass units lower than the parent molecule is produced by the photosubstitution of the F atom by a hydroxyl group. The later intermediate, C4 is product of the cleavage of piperazinyl ring and formation of amide functionality. Structures proposed for both the products match those reported in earlier photolysis studies (Paul et al., 2010). By comparing the intensities of these products obtained after 1.5 h of irradiation with that of the parent, m/z 316.1292

Fig. 9. Evolution and decay curves at UV 254 nm for a) ciprofloxacin b) m/z 316, 330, 245, 272, 346 c) m/z 259, 288, 314, 348.

and 330.1448 are assigned to 65 and 48%, respectively. C4 loses carboxylic acid function to form m/z 272.1394 (C5). The intermediate with observed m/z 288 (C6) (Rt: 2.24 min) is reported in the literature as [MH-CO2]+, corresponding to molecular formula C16H19FN3O+ and had an exact mass of 288.1507 (Rodriguez et al., 2008). In the present study, the mass spectra showed the mass of the ion as 288.1343 and the structure proposed in the previous studies would produce a mass error −58 ppm indicated an incorrect structure. Razuc et al. proposed a structure, which lost an ethyl group from the piperazine ring which agrees within −1 ppm accuracy (Razuc et al., 2013) with our current study. However, major MS/MS fragments observed in targeted MS/MS i.e. m/z 217.0972 and 271.1077 could not be explained based on the proposed structure. Therefore, an alternate structure was proposed in here and is produced by the loss of carboxylic function, cleavage and oxidation of nitrogen containing ring. The structure of both MS/MS fragments proposed from this parent molecule has mass error less than 0.1 ppm. C7 (m/z 314.1499) is a dehalogenated product of ciprofloxacin and been used as one of the product ions in the SRM transitions (Hoof et al., 2005). The intermediate with observed m/z 245 (C8) (Rt;

308

S.R. Batchu et al. / Science of the Total Environment 470–471 (2014) 299–310

Fig. 10. Proposed revised photo transformation pathway for ciprofloxacin at 254 nm. *Denotes revised scheme compared to previous studies (*Rodriguez et al., 2008, Razuc et al., 2013; **Golet et al., 2001; Lee et al., 2007).

Fig. 11. Proposed phototransformation pathway for erythromycin at 254 nm.

S.R. Batchu et al. / Science of the Total Environment 470–471 (2014) 299–310

5.6 min) is reported in the literature as [MH–CO2–C2H5N]+ (Golet et al., 2001; Lee et al., 2007) and had an exact mass of 245.1085. In these studies, data was acquired on a MS system with unit mass resolution. However, when the data was acquired at high resolution during our study the accurate mass for the ion is 245.0917 and the structure proposed in the previous studies would have a mass error of 69 ppm. Therefore another alternate structure was proposed and is included in Fig. 10. A detailed reaction pathway is also included in the Supplementary information. The irradiation of erythromycin yields four primary degradation products with m/z 679 (E1), 608 (E2), 590 (E3) and 475 (E4). Its proposed degradation pathway is shown in Fig. 11. The cleavage of cladinose ring leads to the formation of E1, which forms both E2 and E3. The mass error (ppm) for the proposed structures is less than 1.3, except for E4 (−5.1 ppm). E4 was detected as a formate adduct. The relative abundance of the products at different time points is shown in Supplementary information (Fig. S9). None of these products were previously reported in the literature. The irradiation of sulfamethoxazole at 350 nm showed only one major photolysis product with m/z 192.1131 (0.0 ppm) consistent with the loss of sulfur dioxide from the molecule in good agreement with previous studies (Boreen et al., 2005). 4. Conclusion Antibiotics are highly water soluble and generally resistant to biodegradation (Gartiser et al., 2007). The presence of antibiotics in natural water bodies is a serious problem because of their potent biological activity. Our studies on the photolysis of target antibiotics demonstrate that photochemical degradation plays an important role in the environmental fate of antibiotics. The extent of photodegradation was largely dependent on the solution pH, the presence of dissolved organic content and chloride ions and the irradiation source. The photodegradation of the selected antibiotics in the natural environment will vary depending on structural characteristics as we observed the degradation of sulfamethoxazole and ciprofloxacin are complete within hours while macrolides require days for removal. The photolysis rates reported in this work were measured under constant irradiation in isolated systems and thus represent upper limit for the degradation in the environment. The degradation rates can decrease significantly depending on overcast conditions, water quality, light adsorption properties, the water matrix and the depth of the photic zone. The primary products from the photodegradation of macrolides retain lactone ring and desosamine sugar functionalities critical to biological activity. Given the heavy use and widespread presence of antibiotics in natural waters and drinking water sources it is critical to include the monitoring of the degradation by-products to better assess the risks associated with antibiotics released into the environment. The rapid photodegradation of the target antibiotics under 254 nm irradiation suggests that photolytic disinfection may also be effective for degradation of the antibiotics. Acknowledgments The Environmental Analysis Research Laboratory acknowledges the support from the Thermo Scientific Corporation in the development of this work. Sudha Rani Batchu would like to thank Florida International University graduate school for supporting her through Doctoral Evidence Acquisition Fellowship. This is SERC contribution # 641. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.scitotenv.2013.09.057.

309

References Albrich WC, Monnet DL, Harbarth S. Antibiotic selection pressure and resistance in Streptococcus pneumoniae and Streptococcus pyogenes. Emerg Infect Dis 2004;10: 514–7. An T, Yang H, Li G, Song W, Cooper WJ, Nie X. Kinetics and mechanism of advanced oxidation processes (AOPs) in degradation of ciprofloxacin in water. Appl Catal B-Environ 2010;94:288–94. Andersen SR, Sandaa RA. Distribution of tetracycline resistance determinants among gram-negative bacteria isolated from polluted and unpolluted marine-sediments. Appl Environ Microbiol 1994;60:908–12. Avisar D, Lester Y, Mamane H. pH induced polychromatic UV treatment for the removal of a mixture of SMX, OTC and CIP from water. J Hazard Mater 2010;175:1068–74. Baeza C, Knappe DRU. Transformation kinetics of biochemically active compounds in low-pressure UV photolysis and UV/H2O2 advanced oxidation processes. Water Res 2011;45:4531–43. Batchu SR, Quinete N, Panditi VR, Gardinali PR. Online solid phase extraction liquid chromatography tandem mass spectrometry (SPE-LC–MS/MS) method for the determination of sucralose in reclaimed and drinking waters and its photo degradation in natural waters from South Florida. Chem Cent J 2013;7:141. Batt AL, Aga DS. Simultaneous analysis of multiple classes of antibiotics by ion trap LC/MS/MS for assessing surface water and groundwater contamination. Anal Chem 2005:77. Belden JB, Maul JD, Lydy MJ. Partitioning and photo degradation of ciprofloxacin in aqueous systems in the presence of organic matter. Chemosphere 2007;66:1390–5. Bonvin F, Omlin J, Rutler R, Schweizer WB, Alaimo PJ, Strathmann TJ, et al. Direct photolysis of human metabolites of the antibiotic sulfamethoxazole: evidence for abiotic back-transformation. Environ Sci Technol 2013;47:6746–55. Boreen AL, Arnold WA, McNeill K. Photochemical fate of sulfa drugs in the aquatic environment: sulfa drugs containing five-membered heterocyclic groups. Environ Sci Technol 2004;38:3933–40. Boreen AL, Arnold WA, McNeill K. Triplet-sensitized photodegradation of sulfa drugs containing six-membered heterocyclic groups: identification of an SO2 extrusion photoproduct. Environ Sci Technol 2005;39:3630–8. Brown KD, Kulis J, Thomson B, Chapman TH, Mawhinney DB. Occurrence of antibiotics in hospital, residential, and dairy, effluent, municipal wastewater, and the Rio Grande in New Mexico. Sci Total Environ 2006;366:772–83. Burns JM, Cooper WJ, Ferry JL, King DW, DiMento BP, McNeill K, et al. Methods for reactive oxygen species (ROS) detection in aqueous environments. Aquat Sci 2012;74:683–734. Cals JWL, Hopstaken RM, Le Douxa PHA, Driessen GA, Nelemans PJ, Dinant GJ. Dose timing and patient compliance with two antibiotic treatment regimens for lower respiratory tract infections in primary care. Int J Antimicrob Agents 2008;31:531–6. Chamberlain E, Adams C. Oxidation of sulfonamides, macrolides, and carbadox with free chlorine and monochloramine. Water Res 2006;40:2517–26. Chee-Sanford JC, Aminov RI, Krapac IJ, Garrigues-Jeanjean N, Mackie RI. Occurrence and diversity of tetracycline resistance genes in lagoons and groundwater underlying two swine production facilities. Appl Environ Microbiol 2001;67:1494–502. Costanzo SD, Murby J, Bates J. Ecosystem response to antibiotics entering the aquatic environment. Mar Pollut Bull 2005;51:218–23. Crosby NT. Determination of veterinary residues in food. England: Woodhead Publishing Limited, Abington, Cambridge, 1991. Dodd MC, Kohler HPE, Von Gunten U. Oxidation of antibacterial compounds by ozone and hydroxyl radical: elimination of biological activity during aqueous ozonation processes. Environ Sci Technol 2009;43:2498–504. Gartiser S, Urich E, Alexy R, Kummerer K. Ultimate biodegradation and elimination of antibiotics in inherent tests. Chemosphere 2007;67:604–13. Ge LK, Chen JW, Wei XX, Zhang SY, Qiao XL, Cai XY, et al. Aquatic photochemistry of fluoroquinolone antibiotics: kinetics, pathways, and multivariate effects of main water constituents. Environ Sci Technol 2010;44:2400–5. Giger W, Alder AC, Golet EM, Kohler HPE, McArdell CS, Molnar E, et al. Occurrence and fate of antibiotics as trace contaminants in wastewaters, sewage sludges, and surface waters. Chimia 2003;57:485–91. Golet EM, Alder AC, Hartmann A, Ternes TA, Giger W. Trace determination of fluoroquinolone antibacterial agents in solid-phase extraction urban wastewater by and liquid chromatography with fluorescence detection. Anal Chem 2001;73:3632–8. Hoof NV, Wasch KD, Okerman L, Reybroeck W, Poelmans S, Noppe H, et al. Validation of a liquid chromatography–tandem mass spectrometric method for the quantification of eight quinolones in bovine muscle, milk and aquacultured products. Anal Chim Acta 2005;529:265–72. Isidori M, Lavorgna M, Nardelli A, Pascarella L, Parrella A. Toxic and genotoxic evaluation of six antibiotics on non-target organisms. Sci Total Environ 2005;346:87–98. Kim YH, Heinze TM, Beger R, Pothuluri JV, Cerniglia CE. A kinetic study on the degradation of erythromycin A in aqueous solution. Int J Pharm 2004;271:63–76. Kummerer K. Pharmaceuticals in the environment. Sources, fate, effects and risks. Springer-Verlag; 2001. Lam MW, Tantuco K, Mabury SA. PhotoFate: a new approach in accounting for the contribution of indirect photolysis of pesticides and pharmaceuticals in surface waters. Environ Sci Technol 2003;37:899–907. Lee H-B, Peart TE, Svoboda ML. Determination of ofloxacin, norfloxacin, and ciprofloxacin in sewage by selective solid-phase extraction, liquid chromatography with fluorescence detection, and liquid chromatography–tandem mass spectrometry. J Chromatogr A 2007;1139:45–52. Li DHW, Lam JC, Lau CCS. A study of solar radiation daylight illuminance and sky luminance data measurements for Hong Kong. Archit Sci Rev 2002;45:21–30. Mabe S, Eller J, Champney WS. Structure–activity relationships for three macrolide antibiotics in Haemophilus influenza. Curr Microbiol 2004;49:248–54.

310

S.R. Batchu et al. / Science of the Total Environment 470–471 (2014) 299–310

Miranda CD, Zemelman R. Antibiotic resistant bacteria in fish from the Concepcion Bay, Chile. Mar Pollut Bull 2001;42:1096–102. NIH. The effect of 0.5% Roxithromycin Lotion for Androgenetic Alopecia. http://www. clinicaltrials.gov/ct/show/NCT00197379?order=8, 2013. [Accessed on 1/10/2013]. Palmer AC, Angelino E, Kishony R. Chemical decay of an antibiotic inverts selection for resistance. Nat Chem Biol 2010;6:105–7. Panditi VR, Batchu SR, Gardinali PR. Online solid phase extraction-liquid chromatographyelectrospray-tandem mass spectrometric determination of multiple classes of antibiotics in environmental and treated waters. Anal Bioanal Chem 2013;405(18):5953–64. Paul T, Dodd MC, Strathmann TJ. Photolytic and photocatalytic decomposition of aqueous ciprofloxacin: transformation products and residual antibacterial activity. Water Res 2010;44:3121–32. Pereira VJ, Weinberg HS, Linden KG, Singer PC. UV degradation kinetics and modeling of pharmaceutical compounds in laboratory grade and surface water via direct and indirect photolysis at 254 nm. Environ Sci Technol 2007;41:1682–8. Pham T. Drug use review. http://www.fda.gov/downloads/Drugs/DrugSafety/ InformationbyDrugClass/UCM319435.pdf, 2012. [Accessed on:4/1/2013]. Qiang Z, Adams C. Potentiometric determination of acid dissociation constants (pKa) for human and veterinary antibiotics. Water Res 2004;38:2874–90. Radjenovic J, Godehardt M, Petrovic M, Hein A, Farre M, Jekel M, et al. Evidencing generation of persistent ozonation products of antibiotics roxithromycin and trimethoprim. Environ Sci Technol 2009;43:6808–15. Razuc M, Garrido M, Caro YS, Teglia CM, Goicoechea HC, Band BSF. Hybrid hard- and soft-modeling of spectrophotometric data for monitoring of ciprofloxacin and its main photodegradation products at different pH values. Spectrochim Acta A-Mol Biomol Spectrosc 2013;106:146–54. Rodriguez E, Moreno-Bondi MC, Marazuela MD. Development and validation of a solid-phase extraction method coupled to liquid chromatography with fluorescence detection for the determination of fluoroquinolone residues in powdered infant formulae application to the analysis of samples from the Spanish and Latin American market. J Chromatogr A 2008;1209:136–44. Sagir A, Schmitt M, Dilger K, Haussinger D. Inhibition of cytochrome P450 3A: relevant drug interactions in gastroenterology. Digestion 2003;68:41–8. Sarmah AK, Meyer MT, Boxall ABA. A global perspective on the use, sales, exposure pathways, occurrence, fate and effects of veterinary antibiotics (VAs) in the environment. Chemosphere 2006;65:725–59.

Sirtori C, Aguera A, Gernjak W, Malato S. Effect of water-matrix composition on Trimethoprim solar photodegradation kinetics and pathways. Water Res 2010;44: 2735–44. SRC. Interactive PhyProp Database Demo. http://www.syrres.com/what-we-do/ databaseforms.aspx?id=386, 2013. [Accessed on 1/10/2013]. Stangroom SJ, Macleod CL, Lester JN. Photosensitized transformation of the herbicide 4-chloro-2-methylphenoxy acetic acid (MCPA) in water. Water Res 1998;32:623–32. Sturini M, Speltini A, Pretali L, Fasani E, Profumo A. Solid-phase extraction and HPLC determination of fluoroquinolones in surface waters. J Sep Sci 2009;32:3020–8. Thomas O. Photochemical purification of water and air: advanced oxidation processes (AOPs): principles, reaction mechanisms, reactor concepts. New York: Wiley-VCH; 2003. Tong L, Eichhorn P, Perez S, Wang YX, Barcelo D. Photodegradation of azithromycin in various aqueous systems under simulated and natural solar radiation: kinetics and identification of photoproducts. Chemosphere 2011;83:340–8. Vione D, Maurino V, Minero C, Calza P, Pelizzetti E. Phenol chlorination and photochlorination in the presence of chloride ions in homogeneous aqueous solution. Environ Sci Technol 2005;39:5066–75. Vione D, Feitosa-Felizzola J, Minero C, Chiron S. Phototransformation of selected human-used macrolides in surface water: kinetics, model predictions and degradation pathways. Water Res 2009;43:1959–67. Watanabe N, Harter TH, Bergamaschi BA. Environmental occurrence and shallow ground water detection of the antibiotic monensin from dairy farms. J Environ Qual 2008;37: S78–85. Wetzstein HG, Stadler M, Tichy HV, 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;65:1556–63. Witte W. Medical consequences of antibiotic use in agriculture. Science 1998;279:996–7. Wright GD. Q&A: antibiotic resistance: where does it come from and what can we do about it? BMC Biol 2010:8. Yan W, Zhang JF, Jing CY. Adsorption of enrofloxacin on montmorillonite: two-dimensional correlation ATR/FTIR spectroscopy study. J Colloid Interface Sci 2013;390:196–203. Zhang H, Huang C-H. Adsorption and oxidation of fluoroquinolone antibacterial agents and structurally related amines with goethite. Chemosphere 2007;66:1502–12. Zuccato E, Castiglioni S, Bagnati R, Melis M, Fanelli R. Source, occurrence and fate of antibiotics in the Italian aquatic environment. J Hazard Mater 2010;179:1042–8.