- Email: [email protected]
Integrated liquid chromatography method in enantioselective studies: Biodegradation of ofloxacin by an activated sludge consortium
Accepted Manuscript Title: Integrated liquid chromatography method in enantioselective studies: biodegradation of ofloxacin by an activated sludge consortium Author: Alexandra S. Maia Paula M.L. Castro Maria Elizabeth Tiritan PII: DOI: Reference:
S1570-0232(16)30413-5 http://dx.doi.org/doi:10.1016/j.jchromb.2016.06.026 CHROMB 20112
To appear in:
Journal of Chromatography B
Received date: Revised date: Accepted date:
9-4-2016 12-6-2016 15-6-2016
Please cite this article as: Alexandra S.Maia, Paula M.L.Castro, Maria Elizabeth Tiritan, Integrated liquid chromatography method in enantioselective studies: biodegradation of ofloxacin by an activated sludge consortium, Journal of Chromatography B http://dx.doi.org/10.1016/j.jchromb.2016.06.026 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Integrated liquid chromatography method in enantioselective studies: biodegradation of ofloxacin by an activated sludge consortium
Alexandra S. Maia1,2, Paula M. L. Castro2 and Maria Elizabeth Tiritan1,3,4 1
CESPU, Instituto de Investigação e Formação Avançada em Ciências e Tecnologias da Saúde, Rua Central de
Gandra, 1317, 4585‐116 Gandra PRD, Portugal 2
Universidade Católica Portuguesa, CBQF ‐ Centro de Biotecnologia e Química Fina – Laboratório Associado,
Escola Superior de Biotecnologia, Rua Arquiteto Lobão Vital, Apartado 2511, 4202‐401 Porto, Portugal 3
Laboratório de Química Orgânica e Farmacêutica, Departamento de Ciências Químicas, Faculdade de
Farmácia, Universidade do Porto, Rua de Jorge Viterbo Ferreira, 228, 4050‐313 Porto, Portugal. 4
Centro Interdisciplinar de Investigação Marinha e Ambiental (CIIMAR/CIMAR), Universidade do Porto, Rua
dos Bragas 289, 4050‐123 Porto, Portugal. Corresponding author: Maria Elizabeth Tiritan () CESPU, Instituto de Investigação e Formação Avançada em Ciências e Tecnologias da Saúde, Rua Central de Gandra, 1317, 4585‐116 Gandra PRD, Portugal E‐mail: [email protected]; [email protected] Telephone: +351 224 157 204 Fax: +351 224 157 100
Highlights
An integrated liquid chromatography method approach allowed the evaluation of the enantioselective biodegradation of ofloxacin and levofloxacin by an activated sludge consortium Enantiomerization was observed in the biodegradation of the isolated (S)‐enantiomer with formation of the (R) enantiomer. Enantioselectivity and enantiomerization during the biodegradation were confirmed by exact mass spectrometry with LTQ Orbitrap XL.
Abstract Ofloxacin is a chiral fluoroquinolone commercialized as racemate and as its enantiomerically pure form levofloxacin. This work presents an integrated liquid chromatography (LC) method with fluorescence detection (FD) and exact mass spectrometry (EMS) developed to assess the enantiomeric biodegradation of ofloxacin and levofloxacin in laboratory‐scale microcosms. The optimized enantioseparation conditions were achieved using a macrocyclic antibiotic ristocetin A‐ bonded CSP (150 x 2.1 mm i.d.; particle size 5 µm) under reversed‐phase elution mode. The method was validated using a mineral salts medium as matrix and presented selectivity and linearity over a concentration range from 5 µg L‐1 (quantification limit) to 350 µg L‐1 for each enantiomer. The method was successfully applied to evaluate biodegradation of ofloxacin enantiomers at 250 µg L‐1 by an activated sludge inoculum. Ofloxacin (racemic mixture) and (S)‐ enantiomer (levofloxacin) were degraded up to 58 and 52%, respectively. An additional degradable carbon source, acetate, enhanced biodegradation up to 23%. (S)‐enantiomer presented the highest extent of degradation (66.8%) when ofloxacin was supplied along with acetate. Results indicated slightly higher biodegradation extents for the (S)‐enantiomer when supplementation was done with ofloxacin. Degradation occurred faster in the first 3 days and proceeded slowly until the end of the assays. The chromatographic results from LC‐FD suggested the formation of the (R)‐enantiomer during levofloxacin biodegradation which was confirmed by LC‐MS with a LTQ Orbitrap XL. Keywords Ofloxacin; Levofloxacin; Macrocyclic antibiotic‐based stationary phases; Enantioselectivity; Biodegradation; Activated sludge
2
1. Introduction Quinolone antibiotics, discovered in the 1960s when nalidixic acid was developed, have been gaining great pharmaceutical interest [1, 2]. Between 1985 and 1990 ofloxacin (Fig. 1), which belongs to the second generation of fluoroquinolones, became commercially available in some European countries. Ofloxacin is nowadays largely used in human antimicrobial therapies and in veterinary medicine, the latter with both therapeutic and prophylactic value. Ofloxacin is a racemic mixture and the (S)‐enantiomer corresponds to levofloxacin. The enantiomerically pure form is up to fourfold more active against numerous bacterial strains than the racemic mixture and with up to 128 times more antibacterial activity when compared to the (R)‐enantiomer [3, 4]. Levofloxacin prevailed in time‐kill studies against methicillin‐susceptible Staphylococcus aureus, when compared to ofloxacin and ciprofloxacin [5]. Because of their fluorinated nature, fluoroquinolones exhibit higher recalcitrance and persistence in the environment [6] and have been labeled as emerging environmental micropollutants [7]. These characteristics make them potential targets for environmental risk assessment studies regarding occurrence of pharmaceutical compounds. In recent years, the biodegradability of fluoroquinolones in different conditions has been assessed in several works [8‐10]. Although it is known that biotransformation processes can be enantioselective [11, 12] only a few classes of pharmaceuticals have been included in enantioselective environmental studies [12‐14]. Monitoring the enantiomeric fraction (EF) of chiral pollutants by liquid chromatography in complex natural matrices is a powerful tool to study the enantioselective behavior of such molecules in the environment [11]. However, the difficulty to establish suitable conditions for liquid chromatography with mass spectrometry (LC‐MS) analysis is a drawback in enantioselective studies. The classic approach to select the right chiral stationary phase (CSPs) is based on trial and error studies and usually starts with CSPs with broad applicability such as polysaccharide and macrocyclic antibiotic‐based CSPs [15]. In recent years works regarding environmental analysis of chiral pharmaceuticals report the use of protein‐based, polysaccharide‐based and macrocyclic antibiotic‐based CSPs [12] and other derivatization methods [16‐18]. Methods for enantioseparation of chiral fluoroquinolones using other CSPs than crown ethers‐based and protein‐based are still scarce [19, 20] and none are compatible with LC‐MS. This work describes the development and optimization of a LC method for the resolution of ofloxacin enantiomers. The optimized analytical method was validated in accordance to 3
international criteria and applied to the quantification of the EF of ofloxacin and its (S)‐enantiomer levofloxacin during biodegradation assays by an activated sludge inoculum from a municipal wastewater treatment plant. LC‐MS with a LTQ Orbitrap XL exact mass spectrometer confirmed the identity of the enantiomers. To the best of our knowledge this is the first report on the enantioseparation of ofloxacin by Chirobiotic based CSPs and validation of the enantioselective method to follow the biodegradation of its enantiomers. 2. Material and methods 2.1. Chemicals and materials Standards of the fluoroquinolones ofloxacin and levofloxacin were purchased from Sigma‐Aldrich. The standards presented a purity degree > 98%. Chromatographic gradient grade solvents ethanol, methanol, isopropanol, and acetonitrile, were obtained from Fisher Scientific UK (Leicestershire, UK); hexane was obtained from Merck (Darmstadt, Germany). Triethylamine and diethylamine, both with ≥ 99% purity, were acquired from Sigma‐Aldrich (St. Louis, USA). Acetic acid, formic acid, and trifluoroacetic acid were purchased from VWR International (Fontenay‐sous‐Bois, France), Merck (Darmstadt, Germany) and Acros Organics (New Jersey, USA), respectively. Ammonium acetate and ammonium formate were both obtained from Sigma‐Aldrich (St. Louis, USA). Ultrapure water was supplied by a Milli‐Q water system. All chromatographic solvents were filtered prior to use with 0.45 µm glass microfiber filters from Whatman™. Ofloxacin 1000 mg L‐1 and levofloxacin ((S)‐ofloxacin) 500 mg L‐1 stock solutions were prepared by dissolving the standards in a mixture of water:acetic acid 10% (50/50, v/v). These solutions were stored in amber bottles at ‐20 °C. Working standard solutions were obtained by dilution of stock solutions in ultrapure water or in ethanol (depending on the elution mode in use) to 1 mg L‐1 and prepared weekly. 2.2. Instrumentation Chiral analysis was accomplished in a Shimadzu UFLC Prominence equipment, using two pumps LC‐20AD, an autosampler SIL‐20AC, a column oven CTO‐20AC, a degasser DGU‐20A5, a fluorescence detector RF‐10AXL, and a system controller CBM‐20A. The detection wavelengths were set at 290 and 460 nm for excitation and emission, respectively. Data acquisition was performed using LC Solution, Version 1.24 SP1 from Shimadzu. CSPs included: (S,S)‐Whelk‐O1 and L‐Phenylglycine (250 x 4.6 mm i.d., particle size 5 µm, pore size 100 Å) columns, both from Regis 4
Technologies, Inc. (Morton Grove, IL, USA); and macrocyclic antibiotic‐based CSPs. The latter were used to perform a sequential optimization of ofloxacin enantiomers separation, namely Chirobiotic V (vancomycin‐bonded CSP), Chirobiotic T (teicoplanin‐bonded CSP), Chirobiotic TAG (teicoplanin aglycone‐bonded CSP), and Chirobiotic R (ristocetin A‐bonded CSP). Vancomycin and teicoplanin CSPs with 150 x 4.6 mm i.d. and particle size 5 µm, and teicoplanin aglycone and ristocetin A CSPs with 150 x 2.1 mm i.d. and particle size 5 µm. All the macrocyclic antibiotic‐based CSPs were provided from SUPELCO analytical (Sigma‐Aldrich, Steinheim, Germany). An HPLC Accela with Accela PDA detector, Accela Autosampler and Accela 600 Pump (Thermo Fischer Scientific, Bremen, Germany) coupled to a LTQ Orbitrap XL mass spectrometer (Thermo Fischer Scientific, Bremen, Germany) was used to analyze biodegradation samples. Data acquisition software was LTQ Tune Plus 2.5.5 and Xcalibur 2.1.0. MS detection settings were as follows: capillary voltage of the electrospray ionization (ESI) 3100 V; capillary temperature 275 °C, sheath gas and auxiliary gas flow rates (nitrogen) set to 40 and 10, respectively (arbitrary unit as provided by the software seetings), capillary voltage 36 V, and tube lens voltage 110 V. The identification was performed in a positive ionization full scan mode ranging from 100 to 1000 m/z. 2.3. Optimization of chromatographic conditions Optimization experiments were performed in isocratic mode. Flow rate variations were between 0.2 and 1.0 mL min‐1. Sample injection volume was 10 µL. Mobile phase compositions were: hexane and an additional modifier, ethanol, or isopropanol in normal elution mode approach; and methanol, ethanol, isopropanol, and acetonitrile solo or in combinations, with different proportions of triethylamine and acetic acid (0.01 to 0.1%) in polar and polar ionic elution mode, respectively. More extensive experiments were done in reversed‐phase elution mode, combining different aqueous buffers, namely ammonium acetate and ammonium formate, concentrations ranging from 10 to 50 mM, and diethylamine and triethylamine solutions, 0.1 to 0.45% (v/v), at different pH values adjusted with acetic, formic or trifluoroacetic acids and organic solvents (methanol, ethanol and acetonitrile). The optimized conditions were attained with ristocetin A‐ bonded CSP under reversed‐phase mode with isocratic elution and a flow rate of 0.4 mL min‐1, using a mobile phase of 0.45% triethylamine aqueous solution pH 3.6 adjusted with acetic acid, and ethanol as the organic solvent (80/20, v/v). The column oven temperature was set to 32 °C. The elution order was recognized according to levofloxacin ((S)‐ofloxacin) retention time based on the injection of the enantiomer standard. The optimal chromatographic conditions for LC‐MS 5
analysis with LTQ Orbitrap consisted in ammonium formate in water (concentration 20 mM) pH 4.25 / ethanol (80/20, v/v) and elution was performed in isocratic mode with a flow rate of 0.2 mL min ‐1. The injection volume was 10 µL and the column oven temperature was set to 40 °C. 2.4. Method validation The optimized method conditions were subsequently validated according to the International Conference on Harmonization guidelines [21] and in accordance to our previously reports concerning quantification in biodegradation [8, 13, 14]. Selectivity, linearity and range of application, accuracy, intra‐ and inter‐day precision, recovery, and detection and quantification limits were evaluated. Selectivity was evaluated by comparison of chromatograms obtained by the injection of ofloxacin and levofloxacin standard solutions prepared in ultrapure water with those prepared in mineral salts medium (further details in section 2.5.) in the presence of the activated sludge inoculum. Linearity was assessed considering seven ofloxacin standard concentrations, each prepared in triplicate by dilution with mineral salts medium, ranging from 10 to 700 µg L−1. Calibration curves were calculated by linear regression, establishing the correspondence between peaks areas and the known prepared concentration. Intra‐ and inter‐day precisions, expressed as relative standard deviation percentages (%RSD), were estimated using three quality control standard solutions prepared at three different concentrations within the application range, all in triplicate. Accuracy was calculated as the agreement percentage between method quantitative results of the three quality control standards and the real amount of compound added. Blank samples composed by mineral salts medium with activated sludge were spiked with ofloxacin at the three quality control concentrations and centrifuged at 14,000 rpm for 10 min, after two hours of shaking at 110 rpm. The supernatant aliquots collected were used to calculate the recovery percentage of those fortified samples when compared with the same ofloxacin concentrations prepared in ultrapure water. Detection and quantification limits were calculated by the signal/noise ratio. The ratios were determined by comparing the measurable signals from samples with known low concentrations of the enantiomers with those of blank samples and establishing the minimum concentration at which each enantiomer could be detected and quantified. Signal‐ to‐noise ratios of 3:1 and 10:1 were considered for estimating the detection and quantification limits of each enantiomer, respectively [21]. 2.5. Biodegradation assays 6
The activated sludge inoculum used in the biodegradation assays was collected from the secondary treatment aerated tanks of a municipal wastewater treatment plant (Parada, Maia, Portugal), and preserved in amber glass flasks at 4 °C until usage. The inoculum was washed with mineral salts medium for three cycles prior to its use. The mineral salts medium used followed the composition per liter: Na2HPO42H2O, 2.67 g; KH2PO4, 1.40 g; MgSO47H2O, 0.20 g; (NH4)2SO4, 0.5 g and 10 mL of a trace elements solution with the following composition per liter: NaOH, 2.0 g; Na2EDTA22H2O, 12.0 g; FeSO47H2O 2.0 g; CaCl2, 1.0 g; Na2SO4, 10.0 g; ZnSO4, 0.4 g; MnSO44H2O, 0.4 g; CuSO45H2O, 0.1 g; Na2MoO42H2O, 0.1 g; H2SO4 98%, 0.5 mL. The biodegradation assays were performed in batch mode using 100 mL flasks containing 25 mL of mineral salts medium inoculated with the activated sludge inoculum, previously washed, in order to obtain an optical density of ca. 0.3 at λ = 600 nm according to work published elsewhere [13]. Ofloxacin and levofloxacin standard solutions were added to the flasks to obtain initial concentrations of 500 µg L‐1 and 250 µg L‐1 of the racemic mixture and the (S)‐enantiomer, respectively. Biodegradation was assessed under light and dark conditions, and with and without the presence of acetate as an extra carbon source at an initial concentration of 200 mg L‐1 (as sodium acetate). All experiments were prepared in duplicate, using glass flasks with a volume capacity of fourfold the medium volume used to assure the aeration of the cultures, and conducted aerobically at 25 °C with constant shaking (110 rpm). Control assays without inoculation, under light and dark conditions, were also included. The assays were monitored for 46 days, and the enantiomers were quantified using the validated LC‐FD method by injecting 10 µL aliquots of the supernatant obtained after centrifuging 1 mL samples at 14,000 rpm for 10 min. 3. Results and discussion 3.1. Optimization of ofloxacin enantioseparation Several CSPs were selected and assessed to achieve enantioseparation of ofloxacin. Evaluation started with two Pirkle‐type columns, namely L‐Phenylglycine and (S,S)‐Whelk‐O1. Each CSP was evaluated under normal, polar organic, polar ionic and reversed elution modes, using isocratic conditions. L‐Phenylglycine was not able to resolve ofloxacin enantiomers in many trials and was therefore abandoned. (S,S)‐Whelk‐O1 CSP demonstrated enantioselectivity and resolution power under normal elution mode. Enantioselectivity and resolution were = 1.15 and Rs = 1.24, respectively, (Table 1) at mobile phase composed by hexane/ethanol/CH3COOH (25/75/0.01, v/v). However, even with an augmented polarity of this mobile phase (hexane:ethanol below 50:50, v/v 7
) the retention time was too high and Rs was lower than 1.2. To bypass this situation addition of triethylamine (0.01%, v/v) to the mobile phase was evaluated. Although this modification led to a decrease in the retention time, it showed no advantage in the enantioresolution and additionally it contributed to a decline of the detector signal strength. (S,S)‐Whelk‐O1 was also evaluated under polar ionic mode with different organic eluents, namely methanol, ethanol, isopropanol, and acetonitrile with triethylamine and acetic acid as additives (0.01%, v/v). However, the additives did not improve the chromatographic parameters. Assessment of reversed elution mode using the same organic solvents referred above, and water with triethylamine and acetic acid (pH 3 and 5) also did not result in suitable resolutions. Regarding the versatility of Chirobiotic™ columns [22] and the robustness demonstrated before with Chirobiotic V in monitoring biodegradation [13, 14] and wastewater effluents [23, 24], vancomycin, teicoplanin aglycone, teicoplanin, and ristocetin A‐bonded CSPs were exhaustively evaluated in order to achieve suitable enantioseparation of ofloxacin. Taking into account the good performance verified in previous works [13, 14, 23], the trials started with the vancomycin‐ bonded CSP in polar ionic mode of elution with a mobile phase composed of ethanol/methanol/triethylamine (50/50/0.075, v/v/v) at pH 6.7 adjusted with acetic acid, according to that proposed by Ribeiro et al. [14] for the enantioseparation of three basic pharmaceuticals. Subsequently, several pH conditions were assessed within this mobile phase, modifying the acetic acid content. Chromatographic results were unsatisfactory with total absence of resolution for the ofloxacin enantiomers under these conditions (data not shown). Reversed elution mode was also considered, and a mobile phase of ammonium acetate/methanol mixture, with different aqueous buffer concentrations and pH values, was used as the starting point. Methanol and ethanol proportions between 20 and 90% were evaluated. Reversed elution mode also did not show acceptable results (data not shown). High retention was observed, with some chromatographic runs lasting up to 90 minutes. Considering teicoplanin aglycone, teicoplanin, and ristocetin A‐bonded CSPs, the first attempt was carried out under polar ionic mode to evaluate the influence of different organic solvents and the basic and acid modifiers. Mobile phases constituted by ethanol or methanol with triethylamine/acetic acid (0.1/0.1, v/v) solo or in combination of different volume ratios (between 25:75 to 80:20 of ethanol:methanol) were evaluated under isocratic conditions with flow rates ranging from 0.8 to 1.0 mL min‐1.Teicoplanin and ristocetin A CSPs presented high retentions in all the conditions tested. Teicoplanin aglycone‐bonded CSP was able to partially resolve ofloxacin 8
enantiomers within ca. 25 minutes, when ethanol/methanol (30/70, v/v) with triethylamine/acetic acid (0.1/0.1, v/v) was used as mobile phase, but low resolution (Rs < 0.5) was observed (Table 1). Teicoplanin aglycone, teicoplanin and ristocetin A‐bonded CSPs were also subjected to extensive experiments under reversed elution mode. The influence of the organic modifier was studied for ethanol and methanol with several concentrations of aqueous buffer and pH values, the most important aspects assessed in the optimization. Triethylamine, diethylamine, ammonium acetate and ammonium formate were tried in aqueous solvents, using different concentrations. Triethylamine and ammonium acetate showed the most promising results (Table 2). Different pH values were assessed, ranging from 3.5 to 6.0, respecting the proposals reported in the Chirobiotic™ columns guidelines. As in polar ionic mode, the acid and basic modifiers play an important role in reversed elution mode adjusting ionization states and thus interfering in the resolution. Also in the reversed elution mode, the hydrophobic, charge‐charge, and steric interactions are the prevailing mechanisms [25]. Acidification of the aqueous buffers with acetic, formic, and trifluoroacetic acids was evaluated. Acetic acid revealed to be the most advantageous concerning the compromise between the binomial resolution/selectivity and maintenance of the column integrity, avoiding the employment of halogenated acids. Generally, better results were achieved when the aqueous buffer was prepared at lower pH values (Table 2). The influence of triethylamine concentration in the enantioresolution was also exhaustively evaluated. Triethylamine content in the aqueous solvent varied between 0.1 and 0.45% (v/v) and chromatographic results showed increased resolution, better peak shapes and lower tailing effect as the proportion of triethylamine was augmented. Table 2 shows the chromatographic results obtained in reversed elution mode and the parallelisms observed between the results achieved with teicoplanin aglycone, teicoplanin, and ristocetin A‐bonded CSPs. The mobile phase composed of ammonium acetate 20 mM buffer with acetic acid (pH adjusted to 4.0) in combination with ethanol (50/50, v/v), using a 1.0 mL min‐1 flow rate, resulted in the best achieved resolution value (Rs = 2.04) for the teicoplanin CSP. However, despite the baseline resolution accomplished, the high retention (k1 = 9.63 for the first enantiomer) resulted in a long chromatographic run (over 77 minutes) that along with a low detector response was incompatible for monitoring analysis purposes. Regarding the mobile phase composed of aqueous triethylamine the best resolution values (up to Rs = 2.40) were achieved when using 0.45% triethylamine with acetic acid (pH adjusted to 4.0) with ethanol 20/80 (v/v) at 1.0 mL min‐1. However, despite the
9
good resolution obtained, the same limitations previously identified were observed relating to the high retention and low detector signal. Regarding ristocetin A‐bonded CSP, reversed elution mode provided a resolution value of Rs = 1.2 in less than 10 minutes at a pH value of 3.6 adjusted with acetic acid with a mobile phase of 0.45% triethylamine in aqueous solution and ethanol as the organic solvent (80/20, v/v), at a flow rate of 0.4 mL min‐1. Although lower pH would improve the resolution, lower pH values were not attempted due to restriction of pH range for these CSPs. The acidity of the mobile phase improved the ionic binding interactions between the ofloxacin enantiomers and the functional groups of the ristocetin A‐bonded CSP. Oven temperature was adjusted in order to combine better chromatographic performance and lower retention time. Fixing all the other parameters, different column oven temperatures were attempted and optimal results were achieved at 32 °C. Further optimization was also attempted in order to achieve suitable conditions for LC‐MS analysis. Ammonium formate in water (concentration 20 mM) pH 4.25 / ethanol (80/20, v/v) as mobile phase, and column oven temperature set to 40 °C with a flow rate of 0.2 mL min ‐1 presented enantioselectivity and resolution for LC‐MS analysis (α = 1.44 and Rs = 1.06). The mobile phase adjustments were crucial to achieve enantioselectivity, resolution and suitable conditions for further LC‐MS analysis in order to confirm the identity of the transformation product in the biodegradation study. Table 1 summarizes the optimized chromatographic conditions achieved for each of the CSPs tested regarding the enantioseparation of ofloxacin at 1 mg L‐1. Results show the influence of the mobile phase compositions onto the CSPs evaluated. Good resolutions were achieved under reversed elution mode using ristocetin A and teicoplanin‐ bonded CSPs. The presence of carboxylic acids and amines in the structure of these chiral selectors, the anionic and cationic binding interfaces, and its functionalities imply strong interactions when dealing with ionizable molecules and the zwitterionic features of ofloxacin. Along with polar organic and polar ionic modes, reversed elution mode is amongst the most effective and successful conditions when employing macrocyclic antibiotic‐based CSPs [25]. Despite the high resolution achieved on the teicoplanin‐bonded CSP, the high retention and the low signal response of the analytes turned this CSP unsuitable for monitoring the biodegradation assays. Therefore, and assuming a compromise between time and chromatographic efficiency, the ristocetin A‐bonded CSP has proven to be the most appropriate CSP to proceed with validation and further monitoring of the biodegradation studies. 10
3.2. Method validation Selectivity of the method was verified by chromatographic comparison between ofloxacin standards and the spiked blank matrix, activated sludge in mineral salts medium. Although the matrix content included several ionizable salts present in the mineral salts medium, these salts did not interfere with the chromatographic performance of the ristocetin A‐bonded CSP (Table 3). Linearity was assessed for ofloxacin enantiomers ranging from their quantification limits to 350 µg L‐1, and analyzed by linear regression using the peak areas and the theoretical concentrations. Concentration range, calibration equation, and correlation level (r2 > 0.99) obtained for each enantiomer (Table 3) presented values in accordance with the requirements specified by the international guidelines followed [21]. Accuracy, intra‐ and inter‐day precisions, and recovery were assessed with the chromatographic results obtained with three quality control standard solutions (as described in section 2.4.) with enantiomeric concentrations of 15, 175, and 300 µg L‐1. Accuracy percentages were between 96.47 and 99.56% for (R)‐ofloxacin and 95.70 and 99.85% for (S)‐ofloxacin (Table 3). Repeatability (intra‐ day precision) and intermediate precision (inter‐day precision) were expressed as RSD percentages and were lower than 3.69 and 4.56%, respectively (Table 3). Precision values obtained agree with those demanded by the international guidelines (RSD values under 20%) [21]. Recovery assessment was used to estimate the potential losses of the analytes by sorption to biomass, which might affect their quantification in the biodegradation assays employed (process detailed in section 2.4.). Recovery percentages varied between 93.17 and 98.92%. Detection and quantification limits achieved were 2.5 and 5.0 µg L‐1 for each enantiomer, respectively. These values (Table 3) proved to be suitable for monitoring of the target enantiomers during the biodegradation progression. The method presented is considered a green approach for the quantification of ofloxacin enantiomers, without any organic solvent being used in sample preparation and with ethanol as organic solvent for chromatographic resolution. 3.3. Biodegradation assays Biodegradation of ofloxacin and levofloxacin was followed by quantification of the target analytes with the validated chiral method. Ofloxacin and levofloxacin were added individually at 500 and 250 µg L‐1, respectively, to the mineral salts medium inoculated with the activated sludge. The ability of the activated sludge inoculum to degrade the individual compounds, in the absence and 11
in the presence of acetate as an additional carbon source, was evaluated under light and dark conditions. The biodegradation patterns monitored for 46 days are shown in Fig. 2a‐d. Biodegradation occurred faster in the first 3 days, corresponding to almost 50% of the total degradation reached at the end of the assays. After that, degradation occurred slowly until the end of the experiments. In the first 3 days the activated sludge inoculum was able to consume almost twice as much of ofloxacin and levofloxacin in the presence of acetate than in its absence. Degradation achieved under light and dark conditions was similar. The degree of degradation after 46 days was always greater in the presence of acetate under light and dark conditions, and this applied to both enantiomers, either when present in a racemic mixture or singly supplied. Regarding supplementation with ofloxacin, the (S)‐enantiomer was degraded to a greater extent than the (R)‐enantiomer both under all tested conditions. Concerning supplementation with levofloxacin, biodegradation of the (S)‐enantiomer was greater in the presence of acetate. This distinction was more pronounced in the assays exposed to light. Without any acetate addition, levofloxacin was less consumed by the activated sludge inoculum when compared to the (S)‐ enantiomer consumption observed in the assays with ofloxacin at the end of 46 days. The presence of acetate increased the degradation extent of the single enantiomer, either with or without light exposure, when compared to the (S)‐enantiomer consumption in the racemic mixture. Chromatographic patterns obtained from samples supplemented with levofloxacin showed the formation of a compound with a retention time corresponding to that of the (R)‐ enantiomer of ofloxacin, suggesting a possible enantiomerization phenomenon mediated through biodegradation of the (S)‐enantiomer by activated sludge. Fig. 3 shows the overall degradation of the target enantiomers at an initial concentration of 250 µg L‐1, at the end of 46 days. Higher degradation rates were achieved in the assays with levofloxacin and acetate addition, where the single enantiomer was consumed up to 69 and 72%, under light and dark conditions, respectively. More extensive degradation rates were always verified in the settings with the additional carbon source. In the experiments performed with ofloxacin, the (S)‐enantiomer degradation was higher than the (R)‐enantiomer under light (58 and 43%) and dark conditions (58 and 45%) without acetate and also under light (67 and 61%) and dark conditions (72 and 67%) in the presence of acetate. The lowest degradation extent was observed for the (R)‐enantiomer in the assay with ofloxacin without acetate addition and exposed to the light.
12
In the non‐inoculated control assays, only small extent of degradation of ofloxacin enantiomers was observed (under 18%) after 46 days (data not shown), suggesting that the perceived degradation extent in the biotic conditions was mediated by microbial activity. Although several published studies have reported the biodegradation of ofloxacin by isolated bacteria or by microbial consortia [8, 9], namely Labrys portucalensis F11 with degradation extents up to 91% after 28 days when using 3.5 µM of ofloxacin (ca. 1.2 mg L‐1) [9], no enantioselective approach has been presented before. In this study the activated sludge inoculum was able to partially degrade ofloxacin and levofloxacin, presenting an enantioselective pattern in the degradation of the racemate and enantiomerization of the single enantiomer. A different study on biodegradation of chiral pharmaceuticals using higher concentrations (1 mg L‐1) and an activated sludge inoculum reported both enantioselective and non‐enantioselective degradations for metoprolol, and atenolol and fluoxetine, respectively [14]. The degradation rates were enhanced by the supplementation with acetate as observed elsewhere, and it should be pointed out that the concentrations used in this study were lower than others reported [14, 26]. When dealing with low concentrations of an organic compound the addition of an easily degradable carbon source may reduce the threshold level needed to activate degradation enzymes with advantage for the degrading inoculum [27‐29]. Moreover, mechanisms of cometabolism induced by the presence of a growth substrate other than the target pollutant have been used as common approaches in the study of biotransformation of recalcitrant contaminants [30]. 3.4. Liquid‐chromatography mass spectrometry analysis Selected samples with observable formation of transformation products in the LC‐FD analysis (Fig. 4) were analyzed by LC‐MS with LTQ Orbitrap with exact mass. The chromatographic analysis were performed with the optimal conditions for LC‐MS, as described in section 2.3. Exact mass profiles of samples corresponding to the four different biodegradation assays, considering solely the supplementation with levofloxacin, with and without addition of acetate, and with and without exposure to light, were assessed. In all the samples corresponding to day 31 the presence of the R‐ enantiomer of ofloxacin was confirmed. The R‐enantiomer was not present in the samples at day 0, as expected, since in these assays supplementation was performed only with the S‐enantiomer. Fig. 5 (a‐b) shows the LTQ Orbitrap MS spectra on (S)‐ofloxacin and (R)‐ofloxacin of cultures (exposed to light and with acetate addition) supplied with (S)‐ofloxacin at days 0 (top) and at day 13
31 (bottom). The presence of (S)‐ofloxacin is confirmed in the samples from both days (Fig. 5 a, top and bottom MS spectra) whereas the (R)‐ofloxacin presence is only detectable at day 31 (Fig. 5 b, bottom MS spectrum). The identification of each enantiomer was possible due to the exact mass spectrometry instrument used (m < 5 ppm). Accordingly, it was confirmed that during the degradation process a partial conversion of (S)‐ofloxacin to (R)‐ofloxacin was observed. These results highlight the importance of studying and managing the enantiomeric fractions of chiral contaminants in the environment and the need to control this type of phenomenon in biological transformations. To the best of our knowledge, this work reports the enantiomerization of levofloxacin for the first time. These outcomes endorse enantioselectivity of biodegradation and emphasize the importance of chirality in environmental risk assessment of pharmaceuticals. Additionally, the LC method developed can be easily adapted for other applications beyond environmental analysis such as biological fluids or food analysis. 4. Conclusions An integrated LC approach was used to follow the biodegradation and to confirm the transformation pattern observed in assays supplemented with ofloxacin and levofloxacin. The optimized and validated LC‐FD enantioselective method used to follow the biodegradation of ofloxacin and levofloxacin by an activated sludge inoculum proved to be suitable for this purpose in its linearity range with several advantages: low mobile phase consumption; adequate sensitivity and selectivity demonstrated by the fluorescence detector; and no need to perform time consuming and expensive pre‐concentration processes. The biodegradation of ofloxacin as a racemate followed an enantioselective pattern, with the (S)‐enantiomer being slightly more degraded. Biodegradation of levofloxacin exhibited approximately the same degradation extent observed in the racemic mixture. Enantiomerization was observed in the biodegradation assays supplemented with the single (S)‐enantiomer with formation of the (R) enantiomer. Enantioselectivity and enantiomerization during the biodegradation were confirmed by EMS with LTQ Orbitrap XL.
14
Acknowledgements Authors wish to thank Fundação para a Ciência e Tecnologia – FCT for financial support under the project Fluoropharma PTDC/EBB‐EBI/111699/2009, PhD grant attributed to Alexandra S. Maia SFRH/BD/86939/2012, QREN‐POPH, European Social Fund, MCTES, PEst (FCOMP‐01‐0124‐FEDER‐ 022718; PEst‐OE/EQB/LA0016/2011, PEst‐OE/SAU/UI4040/2014. This research was partially supported by the Strategic Funding UID/Multi/04423/2013 through national funds provided by FCT – Foundation for Science and Technology and European Regional Development Fund (ERDF), in the framework of the programme PT2020and by CESPU (PHARMADRUGS‐CESPU‐2014). The authors thank Parada WWTP for supplying the activated sludge, Virgínia Gonçalves for her collaboration, and also CEMUP – Materials Centre of the University of Porto and the Laboratory for Structural Elucidation for the mass spectrometry analysis.
15
References [1] A.L. Demain, S. Sanchez, Microbial drug discovery: 80 years of progress, The Journal of antibiotics, 62 (2009) 5‐16. [2] M.I. Andersson, A.P. MacGowan, Development of the quinolones, J. Antimicrob. Chemother., 51 (2003) 1‐11. [3] K.P. Fu, S.C. Lafredo, B. Foleno, D.M. Isaacson, J.F. Barrett, A.J. Tobia, M.E. Rosenthale, In vitro and in vivo antibacterial activities of levofloxacin (l‐ofloxacin), an optically active ofloxacin, Antimicrob. Agents Chemother., 36 (1992) 860‐866. [4] Y. Fu, X. Duan, X. Chen, H. Zhang, J. Zhang, W. Li, Chiral discrimination of ofloxacin enantiomers using DNA double helix regulated by metal ions, Chirality, 26 (2014) 249‐254. [5] S.L. Kang, M.J. Rybak, B.J. McGrath, G.W. Kaatz, S.M. Seo, Pharmacodynamics of levofloxacin, ofloxacin, and ciprofloxacin, alone and in combination with rifampin, against methicillin‐ susceptible and ‐resistant Staphylococcus aureus in an in vitro infection model, Antimicrob. Agents Chemother., 38 (1994) 2702‐2709. [6] P. Sukul, M. Spiteller, Fluoroquinolone antibiotics in the environment., Rev. Environ. Contam. Toxicol., 191 (2007) 131‐162. [7] X. Van Doorslaer, J. Dewulf, H. Van Langenhove, K. Demeestere, Fluoroquinolone antibiotics: an emerging class of environmental micropollutants, Sci. Total Environ., 500‐501 (2014) 250‐269. [8] A.S. Maia, A.R. Ribeiro, C.L. Amorim, J.C. Barreiro, Q.B. Cass, P.M.L. Castro, M.E. Tiritan, Degradation of fluoroquinolone antibiotics and identification of metabolites/transformation products by liquid chromatography‐tandem mass spectrometry, J. Chromatogr. A, 1333 (2014) 87‐ 98. [9] C.L. Amorim, I.S. Moreira, A.S. Maia, M.E. Tiritan, P.M.L. Castro, Biodegradation of ofloxacin, norfloxacin and ciprofloxacin as single and mixed substrates by Labrys portucalensis F11, Appl. Microbiol. Biotechnol., 98 (2013) 3181‐3190. [10] C.L. Amorim, A.S. Maia, R.B.R. Mesquita, A.O.S.S. Rangel, M.C.M. van Loosdrecht, M.E. Tiritan, P.M.L. Castro, Performance of aerobic granular sludge in a sequencing batch bioreactor exposed to ofloxacin, norfloxacin and ciprofloxacin, Water Res., 50 (2014) 101‐113. [11] A.R. Ribeiro, P.M.L. Castro, M.E. Tiritan, Chiral pharmaceuticals in the environment, Environ. Chem. Lett., 10 (2012) 239‐253. [12] A.R. Ribeiro, A.S. Maia, Q.B. Cass, M.E. Tiritan, Enantioseparation of chiral pharmaceuticals in biomedical and environmental analyses by liquid chromatography: An overview, J. Chromatogr. B, 968 (2014) 8‐21. [13] A.R. Ribeiro, C.M. Afonso, P.M.L. Castro, M.E. Tiritan, Enantioselective biodegradation of pharmaceuticals, alprenolol and propranolol, by an activated sludge inoculum, Ecotoxicol. Environ. Saf., 87 (2013) 108‐114. [14] A.R. Ribeiro, C.M. Afonso, P.M.L. Castro, M.E. Tiritan, Enantioselective HPLC analysis and biodegradation of atenolol, metoprolol and fluoxetine, Environ. Chem. Lett., 11 (2013) 83‐90. [15] T.J. Ward, K.D. Ward, Chiral separations: a review of current topics and trends, Anal. Chem., 84 (2012) 626‐635. [16] M. Tian, H.S. Row, K.H. Row, Chiral separation of ofloxacin enantiomers by ligand exchange chromatography, Monatshefte für Chemie ‐ Chemical Monthly, 141 (2010) 285‐290. [17] W. Bi, M. Tian, K.H. Row, Chiral separation and determination of ofloxacin enantiomers by ionic liquid‐assisted ligand‐exchange chromatography, Analyst, 136 (2011) 379‐387.
16
[18] H. Yan, F. Qiao, Rapid screening of ofloxacin enantiomers in human urine by molecularly imprinted solid‐phase extraction coupled with ligand exchange chromatography, J. Liq. Chromatogr. Rel. Technol., 37 (2014) 1237‐1248. [19] J. Grellet, B. Ba, M.C. Saux, High‐performance liquid chromatographic separation of fluoroquinolone enantiomers: a review, J. Biochem. Bioph. Methods, 54 (2002) 221‐233. [20] K. Valliappan, M. Sai Sandeep, Multiple Response Optimization of a HPLC Method for the Determination of Enantiomeric Purity of S‐Ofloxacin, Chromatographia, 77 (2014) 1203‐1211. [21] ICH, Q2B ‐ Validation of Analytical Procedures: Methodology, International Conference on Harmonization Expert Working Group, 1996. [22] D.W. Armstrong, Y. Tang, S. Chen, Y. Zhou, C. Bagwill, J.R. Chen, Macrocyclic antibiotics as a new class of chiral selectors for liquid chromatography, Anal. Chem., 66 (1994) 1473‐1484. [23] A.R. Ribeiro, A.S. Maia, I.S. Moreira, C.M. Afonso, P.M.L. Castro, M.E. Tiritan, Enantioselective quantification of fluoxetine and norfluoxetine by HPLC in wastewater effluents, Chemosphere, 95 (2014) 589‐596. [24] A.R. Ribeiro, L.H. Santos, A.S. Maia, C. Delerue‐Matos, P.M. Castro, M.E. Tiritan, Enantiomeric fraction evaluation of pharmaceuticals in environmental matrices by liquid chromatography‐ tandem mass spectrometry, J. Chromatogr. A, 1363 (2014) 226‐235. [25] A. Berthod, Chiral recognition mechanisms with macrocyclic glycopeptide selectors, Chirality, 21 (2009) 167‐175. [26] M.F. Carvalho, A.S. Maia, M.E. Tiritan, P.M.L. Castro, Bacterial degradation of moxifloxacin in the presence of acetate as a bulk substrate, J. Environ. Manage., 168 (2016) 219‐228. [27] T. Egli, How to live at very low substrate concentration, Water Res., 44 (2010) 4826‐4837. [28] K. Kovar, V. Chaloupka, E. Th, A Threshold Substrate Concentration Is Required to Initiate the Degradation of 3‐Phenylpropionic Acid in Escherichia coli, Acta Biotechnol., 22 (2002) 285‐298. [29] N.A. Zhou, A.C. Lutovsky, G.L. Andaker, H.L. Gough, J.F. Ferguson, Cultivation and characterization of bacterial isolates capable of degrading pharmaceutical and personal care products for improved removal in activated sludge wastewater treatment, Biodegradation, 24 (2013) 813‐827. [30] L. Delgadillo‐Mirquez, L. Lardon, J.‐P. Steyer, D. Patureau, A new dynamic model for bioavailability and cometabolism of micropollutants during anaerobic digestion, Water Res., 45 (2011) 4511‐4521.
17
Figure captions Fig. 1 – Chemical structures of ofloxacin enantiomers. Fig. 2 – Biodegradation patterns during 46 days for: a) racemic ofloxacin; b) racemic ofloxacin with acetate addition; c) (S)‐ofloxacin; and d) (S)‐ofloxacin with acetate addition, under light and dark conditions, based on LC‐FD quantification. Fig. 3 – Overall biodegradation percentages after 46 days of ofloxacin enantiomers performed by the activated sludge inoculum with and without acetate addition and under light and dark conditions, based on LC‐FD quantification. Fig. 4 – LC‐FD chromatograms of (S)‐ofloxacin biodegradation samples at days 0 and 31. Conditions: mobile phase 0.45% TEA (pH 3.6 CH3COOH)/ ethanol (80/20, v/v); flow rate 0.4 mL min‐1; column oven temperature 32 °C; λ exc/em = 290/460 nm. Fig. 5 – LTQ Orbitrap MS spectra of ofloxacin enantiomers on biodegradation samples of (S)‐ ofloxacin (with acetate addition and exposed to light) at days 0 (top) and 31 (bottom) for: a) (S)‐ ofloxacin (retention time: 17 min); and b) (R)‐ofloxacin (retention time: 13 min).`
18
Fig 1
Fig 2
19
b Fig 3
Fig 4
20
Fig 5a
21
Fig 5b
22
Table captions Table 1 – Ofloxacin [1 mg L‐1] enantioseparation results achieved in optimization tests with five different CSPs. Chiral stationary phase
Chromatographic parameters
elution mobile phase composition mode
flow rate (mL min‐1)
column oven temp. (oC)
k1
(S,S)‐Whelk‐ O1
NP
n‐hexane/ethanol/CH3COOH: 25/75/0.01
1.0
n.a.
8.25
1.15 1.24
V
POM
methanol: 100
1.0
n.a.
58.53 1.00 0.00
TAG
PIM
methanol/ethanol/CH3COOH/TEA: 30/70/0.1/0.1
0.8
40
20.30 1.16 0.36
T
RP
0.45% TEA pH 4 CH3COOH/ethanol: 20/80
1.0
38
15.19 1.21 2.40
R
RP
0.45% TEA pH 3.6 CH3COOH/ethanol: 80/20
0.4
32
4.30
column
Pirkle
Instrument parameters
Rs
Chirobiotic
1.46 1.15
NP: normal‐phase; PIM: polar ionic mode; POM: polar organic mode; RP: reversed‐phase; TEA: triethylamine; n.a.:not applicable
Table 2 – Chromatographic results obtained for ofloxacin enantioseparation on Chirobiotic CSPs under reversed‐phase elution conditions. Chiral stationary phase Chirobiotic column
Instrument parameters
eluents (aqueous/organic)
TAG T
Chromatographic parameters
20 mM CH3COONH4/ ethanol
R
mobile flow rate phase (v/v) (mL min‐ 1 )
column oven temp. (oC)
k1
50a/50
0.60
40
19.93 1.19
0.78
50a/50
1.00
40
9.63
1.23
2.04
50 /50
1.00
45
8.80
1.20
1.58
50b/50
0.20
n.a.
8.16
1.55
0.95
50/50
0.2
40
2.98
1.36
0.89
60/40
0.3
40
1.62
1.46
0.92
80/20
0.2
40
3.28
1.44
1.06
85c/15
0.65
32
38.52 0.00
0.00
a
Rs
R TAG
20 mM HCOONH4 pH 4.3/ ethanol
0.45% TEA (CH3COOH)/
23
ethanol
T
0.60
35
5.21
1.33
0.57
60a/40
1.00
38
7.50
1.27
1.59
a
1.00
38
5.98
1.27
1.47
a
55 /45
50 /50
1.00
38
11.94 1.23
1.20
40a/60
1.00
25
7.66
1.33
1.27
a
1.00
38
5.53
1.29
1.61
a
40 /60
20 /80
1.00
38
15.19 1.21
2.40
R
88c/12
32
3.85
1.34
0.70
0.60
35
1.54
1.49
0.98
a
0.60
32
1.46
1.58
1.08
85 /15
0.70
a
86 /14
a
50c/50
a
85 /15
0.65
32
2.03
1.51
1.06
80d/20
0.40
33
1.41
1.61
1.12
80d/20
0.50
33
3.53
1.45
1.09
0.40
32
4.30
1.46
1.15
d
80 /20
pH 4; bpH 6; cpH 3.5; dpH 3.6; TEA: triethylamine; n.a.:not applicable
24
Table 3 – Ofloxacin chemical characteristics, separation performance obtained with ristocetin A‐bonded CSP under the validated conditions, and validation results. Enantiom er
Chemical Chromatograph properties ic parameters (average values)
(S)‐ ofloxacin
Limits
Validation parameters
correlati on level (r2)
detectio quantificati n limit on limit (µg L‐1) (µg L‐1)
accurac y (%)
Rs
rang e (µg L‐1)
1.3 7
‐‐‐
‐‐‐
5.00 y = ‐ 1788307 350 3x ‐ 14844
0.9986
2.5
5.0
96.47 93.17 1.88 99.56 98.38 2.68
2.06 4.04
2.3
1. 6
1.1 9
5.00 y = ‐ 1804215 350 9x ‐ 21760
0.9985
2.5
5.0
95.70 95.01 1.91 99.85 98.92 3.69
2.11 4.56
C18H20FN3 O4 / 361.38
Linearity
structure k' / molecular formula / molecular weight (g mol‐1)
(R)‐ ofloxacin
Method validation
linear regressio n
recover y (%)
intra‐ day precisio n (% RSD)
inter‐ day precisio n (% RSD)
S1570-0232(16)30413-5 http://dx.doi.org/doi:10.1016/j.jchromb.2016.06.026 CHROMB 20112
To appear in:
Journal of Chromatography B
Received date: Revised date: Accepted date:
9-4-2016 12-6-2016 15-6-2016
Please cite this article as: Alexandra S.Maia, Paula M.L.Castro, Maria Elizabeth Tiritan, Integrated liquid chromatography method in enantioselective studies: biodegradation of ofloxacin by an activated sludge consortium, Journal of Chromatography B http://dx.doi.org/10.1016/j.jchromb.2016.06.026 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Integrated liquid chromatography method in enantioselective studies: biodegradation of ofloxacin by an activated sludge consortium
Alexandra S. Maia1,2, Paula M. L. Castro2 and Maria Elizabeth Tiritan1,3,4 1
CESPU, Instituto de Investigação e Formação Avançada em Ciências e Tecnologias da Saúde, Rua Central de
Gandra, 1317, 4585‐116 Gandra PRD, Portugal 2
Universidade Católica Portuguesa, CBQF ‐ Centro de Biotecnologia e Química Fina – Laboratório Associado,
Escola Superior de Biotecnologia, Rua Arquiteto Lobão Vital, Apartado 2511, 4202‐401 Porto, Portugal 3
Laboratório de Química Orgânica e Farmacêutica, Departamento de Ciências Químicas, Faculdade de
Farmácia, Universidade do Porto, Rua de Jorge Viterbo Ferreira, 228, 4050‐313 Porto, Portugal. 4
Centro Interdisciplinar de Investigação Marinha e Ambiental (CIIMAR/CIMAR), Universidade do Porto, Rua
dos Bragas 289, 4050‐123 Porto, Portugal. Corresponding author: Maria Elizabeth Tiritan () CESPU, Instituto de Investigação e Formação Avançada em Ciências e Tecnologias da Saúde, Rua Central de Gandra, 1317, 4585‐116 Gandra PRD, Portugal E‐mail: [email protected]; [email protected] Telephone: +351 224 157 204 Fax: +351 224 157 100
Highlights
An integrated liquid chromatography method approach allowed the evaluation of the enantioselective biodegradation of ofloxacin and levofloxacin by an activated sludge consortium Enantiomerization was observed in the biodegradation of the isolated (S)‐enantiomer with formation of the (R) enantiomer. Enantioselectivity and enantiomerization during the biodegradation were confirmed by exact mass spectrometry with LTQ Orbitrap XL.
Abstract Ofloxacin is a chiral fluoroquinolone commercialized as racemate and as its enantiomerically pure form levofloxacin. This work presents an integrated liquid chromatography (LC) method with fluorescence detection (FD) and exact mass spectrometry (EMS) developed to assess the enantiomeric biodegradation of ofloxacin and levofloxacin in laboratory‐scale microcosms. The optimized enantioseparation conditions were achieved using a macrocyclic antibiotic ristocetin A‐ bonded CSP (150 x 2.1 mm i.d.; particle size 5 µm) under reversed‐phase elution mode. The method was validated using a mineral salts medium as matrix and presented selectivity and linearity over a concentration range from 5 µg L‐1 (quantification limit) to 350 µg L‐1 for each enantiomer. The method was successfully applied to evaluate biodegradation of ofloxacin enantiomers at 250 µg L‐1 by an activated sludge inoculum. Ofloxacin (racemic mixture) and (S)‐ enantiomer (levofloxacin) were degraded up to 58 and 52%, respectively. An additional degradable carbon source, acetate, enhanced biodegradation up to 23%. (S)‐enantiomer presented the highest extent of degradation (66.8%) when ofloxacin was supplied along with acetate. Results indicated slightly higher biodegradation extents for the (S)‐enantiomer when supplementation was done with ofloxacin. Degradation occurred faster in the first 3 days and proceeded slowly until the end of the assays. The chromatographic results from LC‐FD suggested the formation of the (R)‐enantiomer during levofloxacin biodegradation which was confirmed by LC‐MS with a LTQ Orbitrap XL. Keywords Ofloxacin; Levofloxacin; Macrocyclic antibiotic‐based stationary phases; Enantioselectivity; Biodegradation; Activated sludge
2
1. Introduction Quinolone antibiotics, discovered in the 1960s when nalidixic acid was developed, have been gaining great pharmaceutical interest [1, 2]. Between 1985 and 1990 ofloxacin (Fig. 1), which belongs to the second generation of fluoroquinolones, became commercially available in some European countries. Ofloxacin is nowadays largely used in human antimicrobial therapies and in veterinary medicine, the latter with both therapeutic and prophylactic value. Ofloxacin is a racemic mixture and the (S)‐enantiomer corresponds to levofloxacin. The enantiomerically pure form is up to fourfold more active against numerous bacterial strains than the racemic mixture and with up to 128 times more antibacterial activity when compared to the (R)‐enantiomer [3, 4]. Levofloxacin prevailed in time‐kill studies against methicillin‐susceptible Staphylococcus aureus, when compared to ofloxacin and ciprofloxacin [5]. Because of their fluorinated nature, fluoroquinolones exhibit higher recalcitrance and persistence in the environment [6] and have been labeled as emerging environmental micropollutants [7]. These characteristics make them potential targets for environmental risk assessment studies regarding occurrence of pharmaceutical compounds. In recent years, the biodegradability of fluoroquinolones in different conditions has been assessed in several works [8‐10]. Although it is known that biotransformation processes can be enantioselective [11, 12] only a few classes of pharmaceuticals have been included in enantioselective environmental studies [12‐14]. Monitoring the enantiomeric fraction (EF) of chiral pollutants by liquid chromatography in complex natural matrices is a powerful tool to study the enantioselective behavior of such molecules in the environment [11]. However, the difficulty to establish suitable conditions for liquid chromatography with mass spectrometry (LC‐MS) analysis is a drawback in enantioselective studies. The classic approach to select the right chiral stationary phase (CSPs) is based on trial and error studies and usually starts with CSPs with broad applicability such as polysaccharide and macrocyclic antibiotic‐based CSPs [15]. In recent years works regarding environmental analysis of chiral pharmaceuticals report the use of protein‐based, polysaccharide‐based and macrocyclic antibiotic‐based CSPs [12] and other derivatization methods [16‐18]. Methods for enantioseparation of chiral fluoroquinolones using other CSPs than crown ethers‐based and protein‐based are still scarce [19, 20] and none are compatible with LC‐MS. This work describes the development and optimization of a LC method for the resolution of ofloxacin enantiomers. The optimized analytical method was validated in accordance to 3
international criteria and applied to the quantification of the EF of ofloxacin and its (S)‐enantiomer levofloxacin during biodegradation assays by an activated sludge inoculum from a municipal wastewater treatment plant. LC‐MS with a LTQ Orbitrap XL exact mass spectrometer confirmed the identity of the enantiomers. To the best of our knowledge this is the first report on the enantioseparation of ofloxacin by Chirobiotic based CSPs and validation of the enantioselective method to follow the biodegradation of its enantiomers. 2. Material and methods 2.1. Chemicals and materials Standards of the fluoroquinolones ofloxacin and levofloxacin were purchased from Sigma‐Aldrich. The standards presented a purity degree > 98%. Chromatographic gradient grade solvents ethanol, methanol, isopropanol, and acetonitrile, were obtained from Fisher Scientific UK (Leicestershire, UK); hexane was obtained from Merck (Darmstadt, Germany). Triethylamine and diethylamine, both with ≥ 99% purity, were acquired from Sigma‐Aldrich (St. Louis, USA). Acetic acid, formic acid, and trifluoroacetic acid were purchased from VWR International (Fontenay‐sous‐Bois, France), Merck (Darmstadt, Germany) and Acros Organics (New Jersey, USA), respectively. Ammonium acetate and ammonium formate were both obtained from Sigma‐Aldrich (St. Louis, USA). Ultrapure water was supplied by a Milli‐Q water system. All chromatographic solvents were filtered prior to use with 0.45 µm glass microfiber filters from Whatman™. Ofloxacin 1000 mg L‐1 and levofloxacin ((S)‐ofloxacin) 500 mg L‐1 stock solutions were prepared by dissolving the standards in a mixture of water:acetic acid 10% (50/50, v/v). These solutions were stored in amber bottles at ‐20 °C. Working standard solutions were obtained by dilution of stock solutions in ultrapure water or in ethanol (depending on the elution mode in use) to 1 mg L‐1 and prepared weekly. 2.2. Instrumentation Chiral analysis was accomplished in a Shimadzu UFLC Prominence equipment, using two pumps LC‐20AD, an autosampler SIL‐20AC, a column oven CTO‐20AC, a degasser DGU‐20A5, a fluorescence detector RF‐10AXL, and a system controller CBM‐20A. The detection wavelengths were set at 290 and 460 nm for excitation and emission, respectively. Data acquisition was performed using LC Solution, Version 1.24 SP1 from Shimadzu. CSPs included: (S,S)‐Whelk‐O1 and L‐Phenylglycine (250 x 4.6 mm i.d., particle size 5 µm, pore size 100 Å) columns, both from Regis 4
Technologies, Inc. (Morton Grove, IL, USA); and macrocyclic antibiotic‐based CSPs. The latter were used to perform a sequential optimization of ofloxacin enantiomers separation, namely Chirobiotic V (vancomycin‐bonded CSP), Chirobiotic T (teicoplanin‐bonded CSP), Chirobiotic TAG (teicoplanin aglycone‐bonded CSP), and Chirobiotic R (ristocetin A‐bonded CSP). Vancomycin and teicoplanin CSPs with 150 x 4.6 mm i.d. and particle size 5 µm, and teicoplanin aglycone and ristocetin A CSPs with 150 x 2.1 mm i.d. and particle size 5 µm. All the macrocyclic antibiotic‐based CSPs were provided from SUPELCO analytical (Sigma‐Aldrich, Steinheim, Germany). An HPLC Accela with Accela PDA detector, Accela Autosampler and Accela 600 Pump (Thermo Fischer Scientific, Bremen, Germany) coupled to a LTQ Orbitrap XL mass spectrometer (Thermo Fischer Scientific, Bremen, Germany) was used to analyze biodegradation samples. Data acquisition software was LTQ Tune Plus 2.5.5 and Xcalibur 2.1.0. MS detection settings were as follows: capillary voltage of the electrospray ionization (ESI) 3100 V; capillary temperature 275 °C, sheath gas and auxiliary gas flow rates (nitrogen) set to 40 and 10, respectively (arbitrary unit as provided by the software seetings), capillary voltage 36 V, and tube lens voltage 110 V. The identification was performed in a positive ionization full scan mode ranging from 100 to 1000 m/z. 2.3. Optimization of chromatographic conditions Optimization experiments were performed in isocratic mode. Flow rate variations were between 0.2 and 1.0 mL min‐1. Sample injection volume was 10 µL. Mobile phase compositions were: hexane and an additional modifier, ethanol, or isopropanol in normal elution mode approach; and methanol, ethanol, isopropanol, and acetonitrile solo or in combinations, with different proportions of triethylamine and acetic acid (0.01 to 0.1%) in polar and polar ionic elution mode, respectively. More extensive experiments were done in reversed‐phase elution mode, combining different aqueous buffers, namely ammonium acetate and ammonium formate, concentrations ranging from 10 to 50 mM, and diethylamine and triethylamine solutions, 0.1 to 0.45% (v/v), at different pH values adjusted with acetic, formic or trifluoroacetic acids and organic solvents (methanol, ethanol and acetonitrile). The optimized conditions were attained with ristocetin A‐ bonded CSP under reversed‐phase mode with isocratic elution and a flow rate of 0.4 mL min‐1, using a mobile phase of 0.45% triethylamine aqueous solution pH 3.6 adjusted with acetic acid, and ethanol as the organic solvent (80/20, v/v). The column oven temperature was set to 32 °C. The elution order was recognized according to levofloxacin ((S)‐ofloxacin) retention time based on the injection of the enantiomer standard. The optimal chromatographic conditions for LC‐MS 5
analysis with LTQ Orbitrap consisted in ammonium formate in water (concentration 20 mM) pH 4.25 / ethanol (80/20, v/v) and elution was performed in isocratic mode with a flow rate of 0.2 mL min ‐1. The injection volume was 10 µL and the column oven temperature was set to 40 °C. 2.4. Method validation The optimized method conditions were subsequently validated according to the International Conference on Harmonization guidelines [21] and in accordance to our previously reports concerning quantification in biodegradation [8, 13, 14]. Selectivity, linearity and range of application, accuracy, intra‐ and inter‐day precision, recovery, and detection and quantification limits were evaluated. Selectivity was evaluated by comparison of chromatograms obtained by the injection of ofloxacin and levofloxacin standard solutions prepared in ultrapure water with those prepared in mineral salts medium (further details in section 2.5.) in the presence of the activated sludge inoculum. Linearity was assessed considering seven ofloxacin standard concentrations, each prepared in triplicate by dilution with mineral salts medium, ranging from 10 to 700 µg L−1. Calibration curves were calculated by linear regression, establishing the correspondence between peaks areas and the known prepared concentration. Intra‐ and inter‐day precisions, expressed as relative standard deviation percentages (%RSD), were estimated using three quality control standard solutions prepared at three different concentrations within the application range, all in triplicate. Accuracy was calculated as the agreement percentage between method quantitative results of the three quality control standards and the real amount of compound added. Blank samples composed by mineral salts medium with activated sludge were spiked with ofloxacin at the three quality control concentrations and centrifuged at 14,000 rpm for 10 min, after two hours of shaking at 110 rpm. The supernatant aliquots collected were used to calculate the recovery percentage of those fortified samples when compared with the same ofloxacin concentrations prepared in ultrapure water. Detection and quantification limits were calculated by the signal/noise ratio. The ratios were determined by comparing the measurable signals from samples with known low concentrations of the enantiomers with those of blank samples and establishing the minimum concentration at which each enantiomer could be detected and quantified. Signal‐ to‐noise ratios of 3:1 and 10:1 were considered for estimating the detection and quantification limits of each enantiomer, respectively [21]. 2.5. Biodegradation assays 6
The activated sludge inoculum used in the biodegradation assays was collected from the secondary treatment aerated tanks of a municipal wastewater treatment plant (Parada, Maia, Portugal), and preserved in amber glass flasks at 4 °C until usage. The inoculum was washed with mineral salts medium for three cycles prior to its use. The mineral salts medium used followed the composition per liter: Na2HPO42H2O, 2.67 g; KH2PO4, 1.40 g; MgSO47H2O, 0.20 g; (NH4)2SO4, 0.5 g and 10 mL of a trace elements solution with the following composition per liter: NaOH, 2.0 g; Na2EDTA22H2O, 12.0 g; FeSO47H2O 2.0 g; CaCl2, 1.0 g; Na2SO4, 10.0 g; ZnSO4, 0.4 g; MnSO44H2O, 0.4 g; CuSO45H2O, 0.1 g; Na2MoO42H2O, 0.1 g; H2SO4 98%, 0.5 mL. The biodegradation assays were performed in batch mode using 100 mL flasks containing 25 mL of mineral salts medium inoculated with the activated sludge inoculum, previously washed, in order to obtain an optical density of ca. 0.3 at λ = 600 nm according to work published elsewhere [13]. Ofloxacin and levofloxacin standard solutions were added to the flasks to obtain initial concentrations of 500 µg L‐1 and 250 µg L‐1 of the racemic mixture and the (S)‐enantiomer, respectively. Biodegradation was assessed under light and dark conditions, and with and without the presence of acetate as an extra carbon source at an initial concentration of 200 mg L‐1 (as sodium acetate). All experiments were prepared in duplicate, using glass flasks with a volume capacity of fourfold the medium volume used to assure the aeration of the cultures, and conducted aerobically at 25 °C with constant shaking (110 rpm). Control assays without inoculation, under light and dark conditions, were also included. The assays were monitored for 46 days, and the enantiomers were quantified using the validated LC‐FD method by injecting 10 µL aliquots of the supernatant obtained after centrifuging 1 mL samples at 14,000 rpm for 10 min. 3. Results and discussion 3.1. Optimization of ofloxacin enantioseparation Several CSPs were selected and assessed to achieve enantioseparation of ofloxacin. Evaluation started with two Pirkle‐type columns, namely L‐Phenylglycine and (S,S)‐Whelk‐O1. Each CSP was evaluated under normal, polar organic, polar ionic and reversed elution modes, using isocratic conditions. L‐Phenylglycine was not able to resolve ofloxacin enantiomers in many trials and was therefore abandoned. (S,S)‐Whelk‐O1 CSP demonstrated enantioselectivity and resolution power under normal elution mode. Enantioselectivity and resolution were = 1.15 and Rs = 1.24, respectively, (Table 1) at mobile phase composed by hexane/ethanol/CH3COOH (25/75/0.01, v/v). However, even with an augmented polarity of this mobile phase (hexane:ethanol below 50:50, v/v 7
) the retention time was too high and Rs was lower than 1.2. To bypass this situation addition of triethylamine (0.01%, v/v) to the mobile phase was evaluated. Although this modification led to a decrease in the retention time, it showed no advantage in the enantioresolution and additionally it contributed to a decline of the detector signal strength. (S,S)‐Whelk‐O1 was also evaluated under polar ionic mode with different organic eluents, namely methanol, ethanol, isopropanol, and acetonitrile with triethylamine and acetic acid as additives (0.01%, v/v). However, the additives did not improve the chromatographic parameters. Assessment of reversed elution mode using the same organic solvents referred above, and water with triethylamine and acetic acid (pH 3 and 5) also did not result in suitable resolutions. Regarding the versatility of Chirobiotic™ columns [22] and the robustness demonstrated before with Chirobiotic V in monitoring biodegradation [13, 14] and wastewater effluents [23, 24], vancomycin, teicoplanin aglycone, teicoplanin, and ristocetin A‐bonded CSPs were exhaustively evaluated in order to achieve suitable enantioseparation of ofloxacin. Taking into account the good performance verified in previous works [13, 14, 23], the trials started with the vancomycin‐ bonded CSP in polar ionic mode of elution with a mobile phase composed of ethanol/methanol/triethylamine (50/50/0.075, v/v/v) at pH 6.7 adjusted with acetic acid, according to that proposed by Ribeiro et al. [14] for the enantioseparation of three basic pharmaceuticals. Subsequently, several pH conditions were assessed within this mobile phase, modifying the acetic acid content. Chromatographic results were unsatisfactory with total absence of resolution for the ofloxacin enantiomers under these conditions (data not shown). Reversed elution mode was also considered, and a mobile phase of ammonium acetate/methanol mixture, with different aqueous buffer concentrations and pH values, was used as the starting point. Methanol and ethanol proportions between 20 and 90% were evaluated. Reversed elution mode also did not show acceptable results (data not shown). High retention was observed, with some chromatographic runs lasting up to 90 minutes. Considering teicoplanin aglycone, teicoplanin, and ristocetin A‐bonded CSPs, the first attempt was carried out under polar ionic mode to evaluate the influence of different organic solvents and the basic and acid modifiers. Mobile phases constituted by ethanol or methanol with triethylamine/acetic acid (0.1/0.1, v/v) solo or in combination of different volume ratios (between 25:75 to 80:20 of ethanol:methanol) were evaluated under isocratic conditions with flow rates ranging from 0.8 to 1.0 mL min‐1.Teicoplanin and ristocetin A CSPs presented high retentions in all the conditions tested. Teicoplanin aglycone‐bonded CSP was able to partially resolve ofloxacin 8
enantiomers within ca. 25 minutes, when ethanol/methanol (30/70, v/v) with triethylamine/acetic acid (0.1/0.1, v/v) was used as mobile phase, but low resolution (Rs < 0.5) was observed (Table 1). Teicoplanin aglycone, teicoplanin and ristocetin A‐bonded CSPs were also subjected to extensive experiments under reversed elution mode. The influence of the organic modifier was studied for ethanol and methanol with several concentrations of aqueous buffer and pH values, the most important aspects assessed in the optimization. Triethylamine, diethylamine, ammonium acetate and ammonium formate were tried in aqueous solvents, using different concentrations. Triethylamine and ammonium acetate showed the most promising results (Table 2). Different pH values were assessed, ranging from 3.5 to 6.0, respecting the proposals reported in the Chirobiotic™ columns guidelines. As in polar ionic mode, the acid and basic modifiers play an important role in reversed elution mode adjusting ionization states and thus interfering in the resolution. Also in the reversed elution mode, the hydrophobic, charge‐charge, and steric interactions are the prevailing mechanisms [25]. Acidification of the aqueous buffers with acetic, formic, and trifluoroacetic acids was evaluated. Acetic acid revealed to be the most advantageous concerning the compromise between the binomial resolution/selectivity and maintenance of the column integrity, avoiding the employment of halogenated acids. Generally, better results were achieved when the aqueous buffer was prepared at lower pH values (Table 2). The influence of triethylamine concentration in the enantioresolution was also exhaustively evaluated. Triethylamine content in the aqueous solvent varied between 0.1 and 0.45% (v/v) and chromatographic results showed increased resolution, better peak shapes and lower tailing effect as the proportion of triethylamine was augmented. Table 2 shows the chromatographic results obtained in reversed elution mode and the parallelisms observed between the results achieved with teicoplanin aglycone, teicoplanin, and ristocetin A‐bonded CSPs. The mobile phase composed of ammonium acetate 20 mM buffer with acetic acid (pH adjusted to 4.0) in combination with ethanol (50/50, v/v), using a 1.0 mL min‐1 flow rate, resulted in the best achieved resolution value (Rs = 2.04) for the teicoplanin CSP. However, despite the baseline resolution accomplished, the high retention (k1 = 9.63 for the first enantiomer) resulted in a long chromatographic run (over 77 minutes) that along with a low detector response was incompatible for monitoring analysis purposes. Regarding the mobile phase composed of aqueous triethylamine the best resolution values (up to Rs = 2.40) were achieved when using 0.45% triethylamine with acetic acid (pH adjusted to 4.0) with ethanol 20/80 (v/v) at 1.0 mL min‐1. However, despite the
9
good resolution obtained, the same limitations previously identified were observed relating to the high retention and low detector signal. Regarding ristocetin A‐bonded CSP, reversed elution mode provided a resolution value of Rs = 1.2 in less than 10 minutes at a pH value of 3.6 adjusted with acetic acid with a mobile phase of 0.45% triethylamine in aqueous solution and ethanol as the organic solvent (80/20, v/v), at a flow rate of 0.4 mL min‐1. Although lower pH would improve the resolution, lower pH values were not attempted due to restriction of pH range for these CSPs. The acidity of the mobile phase improved the ionic binding interactions between the ofloxacin enantiomers and the functional groups of the ristocetin A‐bonded CSP. Oven temperature was adjusted in order to combine better chromatographic performance and lower retention time. Fixing all the other parameters, different column oven temperatures were attempted and optimal results were achieved at 32 °C. Further optimization was also attempted in order to achieve suitable conditions for LC‐MS analysis. Ammonium formate in water (concentration 20 mM) pH 4.25 / ethanol (80/20, v/v) as mobile phase, and column oven temperature set to 40 °C with a flow rate of 0.2 mL min ‐1 presented enantioselectivity and resolution for LC‐MS analysis (α = 1.44 and Rs = 1.06). The mobile phase adjustments were crucial to achieve enantioselectivity, resolution and suitable conditions for further LC‐MS analysis in order to confirm the identity of the transformation product in the biodegradation study. Table 1 summarizes the optimized chromatographic conditions achieved for each of the CSPs tested regarding the enantioseparation of ofloxacin at 1 mg L‐1. Results show the influence of the mobile phase compositions onto the CSPs evaluated. Good resolutions were achieved under reversed elution mode using ristocetin A and teicoplanin‐ bonded CSPs. The presence of carboxylic acids and amines in the structure of these chiral selectors, the anionic and cationic binding interfaces, and its functionalities imply strong interactions when dealing with ionizable molecules and the zwitterionic features of ofloxacin. Along with polar organic and polar ionic modes, reversed elution mode is amongst the most effective and successful conditions when employing macrocyclic antibiotic‐based CSPs [25]. Despite the high resolution achieved on the teicoplanin‐bonded CSP, the high retention and the low signal response of the analytes turned this CSP unsuitable for monitoring the biodegradation assays. Therefore, and assuming a compromise between time and chromatographic efficiency, the ristocetin A‐bonded CSP has proven to be the most appropriate CSP to proceed with validation and further monitoring of the biodegradation studies. 10
3.2. Method validation Selectivity of the method was verified by chromatographic comparison between ofloxacin standards and the spiked blank matrix, activated sludge in mineral salts medium. Although the matrix content included several ionizable salts present in the mineral salts medium, these salts did not interfere with the chromatographic performance of the ristocetin A‐bonded CSP (Table 3). Linearity was assessed for ofloxacin enantiomers ranging from their quantification limits to 350 µg L‐1, and analyzed by linear regression using the peak areas and the theoretical concentrations. Concentration range, calibration equation, and correlation level (r2 > 0.99) obtained for each enantiomer (Table 3) presented values in accordance with the requirements specified by the international guidelines followed [21]. Accuracy, intra‐ and inter‐day precisions, and recovery were assessed with the chromatographic results obtained with three quality control standard solutions (as described in section 2.4.) with enantiomeric concentrations of 15, 175, and 300 µg L‐1. Accuracy percentages were between 96.47 and 99.56% for (R)‐ofloxacin and 95.70 and 99.85% for (S)‐ofloxacin (Table 3). Repeatability (intra‐ day precision) and intermediate precision (inter‐day precision) were expressed as RSD percentages and were lower than 3.69 and 4.56%, respectively (Table 3). Precision values obtained agree with those demanded by the international guidelines (RSD values under 20%) [21]. Recovery assessment was used to estimate the potential losses of the analytes by sorption to biomass, which might affect their quantification in the biodegradation assays employed (process detailed in section 2.4.). Recovery percentages varied between 93.17 and 98.92%. Detection and quantification limits achieved were 2.5 and 5.0 µg L‐1 for each enantiomer, respectively. These values (Table 3) proved to be suitable for monitoring of the target enantiomers during the biodegradation progression. The method presented is considered a green approach for the quantification of ofloxacin enantiomers, without any organic solvent being used in sample preparation and with ethanol as organic solvent for chromatographic resolution. 3.3. Biodegradation assays Biodegradation of ofloxacin and levofloxacin was followed by quantification of the target analytes with the validated chiral method. Ofloxacin and levofloxacin were added individually at 500 and 250 µg L‐1, respectively, to the mineral salts medium inoculated with the activated sludge. The ability of the activated sludge inoculum to degrade the individual compounds, in the absence and 11
in the presence of acetate as an additional carbon source, was evaluated under light and dark conditions. The biodegradation patterns monitored for 46 days are shown in Fig. 2a‐d. Biodegradation occurred faster in the first 3 days, corresponding to almost 50% of the total degradation reached at the end of the assays. After that, degradation occurred slowly until the end of the experiments. In the first 3 days the activated sludge inoculum was able to consume almost twice as much of ofloxacin and levofloxacin in the presence of acetate than in its absence. Degradation achieved under light and dark conditions was similar. The degree of degradation after 46 days was always greater in the presence of acetate under light and dark conditions, and this applied to both enantiomers, either when present in a racemic mixture or singly supplied. Regarding supplementation with ofloxacin, the (S)‐enantiomer was degraded to a greater extent than the (R)‐enantiomer both under all tested conditions. Concerning supplementation with levofloxacin, biodegradation of the (S)‐enantiomer was greater in the presence of acetate. This distinction was more pronounced in the assays exposed to light. Without any acetate addition, levofloxacin was less consumed by the activated sludge inoculum when compared to the (S)‐ enantiomer consumption observed in the assays with ofloxacin at the end of 46 days. The presence of acetate increased the degradation extent of the single enantiomer, either with or without light exposure, when compared to the (S)‐enantiomer consumption in the racemic mixture. Chromatographic patterns obtained from samples supplemented with levofloxacin showed the formation of a compound with a retention time corresponding to that of the (R)‐ enantiomer of ofloxacin, suggesting a possible enantiomerization phenomenon mediated through biodegradation of the (S)‐enantiomer by activated sludge. Fig. 3 shows the overall degradation of the target enantiomers at an initial concentration of 250 µg L‐1, at the end of 46 days. Higher degradation rates were achieved in the assays with levofloxacin and acetate addition, where the single enantiomer was consumed up to 69 and 72%, under light and dark conditions, respectively. More extensive degradation rates were always verified in the settings with the additional carbon source. In the experiments performed with ofloxacin, the (S)‐enantiomer degradation was higher than the (R)‐enantiomer under light (58 and 43%) and dark conditions (58 and 45%) without acetate and also under light (67 and 61%) and dark conditions (72 and 67%) in the presence of acetate. The lowest degradation extent was observed for the (R)‐enantiomer in the assay with ofloxacin without acetate addition and exposed to the light.
12
In the non‐inoculated control assays, only small extent of degradation of ofloxacin enantiomers was observed (under 18%) after 46 days (data not shown), suggesting that the perceived degradation extent in the biotic conditions was mediated by microbial activity. Although several published studies have reported the biodegradation of ofloxacin by isolated bacteria or by microbial consortia [8, 9], namely Labrys portucalensis F11 with degradation extents up to 91% after 28 days when using 3.5 µM of ofloxacin (ca. 1.2 mg L‐1) [9], no enantioselective approach has been presented before. In this study the activated sludge inoculum was able to partially degrade ofloxacin and levofloxacin, presenting an enantioselective pattern in the degradation of the racemate and enantiomerization of the single enantiomer. A different study on biodegradation of chiral pharmaceuticals using higher concentrations (1 mg L‐1) and an activated sludge inoculum reported both enantioselective and non‐enantioselective degradations for metoprolol, and atenolol and fluoxetine, respectively [14]. The degradation rates were enhanced by the supplementation with acetate as observed elsewhere, and it should be pointed out that the concentrations used in this study were lower than others reported [14, 26]. When dealing with low concentrations of an organic compound the addition of an easily degradable carbon source may reduce the threshold level needed to activate degradation enzymes with advantage for the degrading inoculum [27‐29]. Moreover, mechanisms of cometabolism induced by the presence of a growth substrate other than the target pollutant have been used as common approaches in the study of biotransformation of recalcitrant contaminants [30]. 3.4. Liquid‐chromatography mass spectrometry analysis Selected samples with observable formation of transformation products in the LC‐FD analysis (Fig. 4) were analyzed by LC‐MS with LTQ Orbitrap with exact mass. The chromatographic analysis were performed with the optimal conditions for LC‐MS, as described in section 2.3. Exact mass profiles of samples corresponding to the four different biodegradation assays, considering solely the supplementation with levofloxacin, with and without addition of acetate, and with and without exposure to light, were assessed. In all the samples corresponding to day 31 the presence of the R‐ enantiomer of ofloxacin was confirmed. The R‐enantiomer was not present in the samples at day 0, as expected, since in these assays supplementation was performed only with the S‐enantiomer. Fig. 5 (a‐b) shows the LTQ Orbitrap MS spectra on (S)‐ofloxacin and (R)‐ofloxacin of cultures (exposed to light and with acetate addition) supplied with (S)‐ofloxacin at days 0 (top) and at day 13
31 (bottom). The presence of (S)‐ofloxacin is confirmed in the samples from both days (Fig. 5 a, top and bottom MS spectra) whereas the (R)‐ofloxacin presence is only detectable at day 31 (Fig. 5 b, bottom MS spectrum). The identification of each enantiomer was possible due to the exact mass spectrometry instrument used (m < 5 ppm). Accordingly, it was confirmed that during the degradation process a partial conversion of (S)‐ofloxacin to (R)‐ofloxacin was observed. These results highlight the importance of studying and managing the enantiomeric fractions of chiral contaminants in the environment and the need to control this type of phenomenon in biological transformations. To the best of our knowledge, this work reports the enantiomerization of levofloxacin for the first time. These outcomes endorse enantioselectivity of biodegradation and emphasize the importance of chirality in environmental risk assessment of pharmaceuticals. Additionally, the LC method developed can be easily adapted for other applications beyond environmental analysis such as biological fluids or food analysis. 4. Conclusions An integrated LC approach was used to follow the biodegradation and to confirm the transformation pattern observed in assays supplemented with ofloxacin and levofloxacin. The optimized and validated LC‐FD enantioselective method used to follow the biodegradation of ofloxacin and levofloxacin by an activated sludge inoculum proved to be suitable for this purpose in its linearity range with several advantages: low mobile phase consumption; adequate sensitivity and selectivity demonstrated by the fluorescence detector; and no need to perform time consuming and expensive pre‐concentration processes. The biodegradation of ofloxacin as a racemate followed an enantioselective pattern, with the (S)‐enantiomer being slightly more degraded. Biodegradation of levofloxacin exhibited approximately the same degradation extent observed in the racemic mixture. Enantiomerization was observed in the biodegradation assays supplemented with the single (S)‐enantiomer with formation of the (R) enantiomer. Enantioselectivity and enantiomerization during the biodegradation were confirmed by EMS with LTQ Orbitrap XL.
14
Acknowledgements Authors wish to thank Fundação para a Ciência e Tecnologia – FCT for financial support under the project Fluoropharma PTDC/EBB‐EBI/111699/2009, PhD grant attributed to Alexandra S. Maia SFRH/BD/86939/2012, QREN‐POPH, European Social Fund, MCTES, PEst (FCOMP‐01‐0124‐FEDER‐ 022718; PEst‐OE/EQB/LA0016/2011, PEst‐OE/SAU/UI4040/2014. This research was partially supported by the Strategic Funding UID/Multi/04423/2013 through national funds provided by FCT – Foundation for Science and Technology and European Regional Development Fund (ERDF), in the framework of the programme PT2020and by CESPU (PHARMADRUGS‐CESPU‐2014). The authors thank Parada WWTP for supplying the activated sludge, Virgínia Gonçalves for her collaboration, and also CEMUP – Materials Centre of the University of Porto and the Laboratory for Structural Elucidation for the mass spectrometry analysis.
15
References [1] A.L. Demain, S. Sanchez, Microbial drug discovery: 80 years of progress, The Journal of antibiotics, 62 (2009) 5‐16. [2] M.I. Andersson, A.P. MacGowan, Development of the quinolones, J. Antimicrob. Chemother., 51 (2003) 1‐11. [3] K.P. Fu, S.C. Lafredo, B. Foleno, D.M. Isaacson, J.F. Barrett, A.J. Tobia, M.E. Rosenthale, In vitro and in vivo antibacterial activities of levofloxacin (l‐ofloxacin), an optically active ofloxacin, Antimicrob. Agents Chemother., 36 (1992) 860‐866. [4] Y. Fu, X. Duan, X. Chen, H. Zhang, J. Zhang, W. Li, Chiral discrimination of ofloxacin enantiomers using DNA double helix regulated by metal ions, Chirality, 26 (2014) 249‐254. [5] S.L. Kang, M.J. Rybak, B.J. McGrath, G.W. Kaatz, S.M. Seo, Pharmacodynamics of levofloxacin, ofloxacin, and ciprofloxacin, alone and in combination with rifampin, against methicillin‐ susceptible and ‐resistant Staphylococcus aureus in an in vitro infection model, Antimicrob. Agents Chemother., 38 (1994) 2702‐2709. [6] P. Sukul, M. Spiteller, Fluoroquinolone antibiotics in the environment., Rev. Environ. Contam. Toxicol., 191 (2007) 131‐162. [7] X. Van Doorslaer, J. Dewulf, H. Van Langenhove, K. Demeestere, Fluoroquinolone antibiotics: an emerging class of environmental micropollutants, Sci. Total Environ., 500‐501 (2014) 250‐269. [8] A.S. Maia, A.R. Ribeiro, C.L. Amorim, J.C. Barreiro, Q.B. Cass, P.M.L. Castro, M.E. Tiritan, Degradation of fluoroquinolone antibiotics and identification of metabolites/transformation products by liquid chromatography‐tandem mass spectrometry, J. Chromatogr. A, 1333 (2014) 87‐ 98. [9] C.L. Amorim, I.S. Moreira, A.S. Maia, M.E. Tiritan, P.M.L. Castro, Biodegradation of ofloxacin, norfloxacin and ciprofloxacin as single and mixed substrates by Labrys portucalensis F11, Appl. Microbiol. Biotechnol., 98 (2013) 3181‐3190. [10] C.L. Amorim, A.S. Maia, R.B.R. Mesquita, A.O.S.S. Rangel, M.C.M. van Loosdrecht, M.E. Tiritan, P.M.L. Castro, Performance of aerobic granular sludge in a sequencing batch bioreactor exposed to ofloxacin, norfloxacin and ciprofloxacin, Water Res., 50 (2014) 101‐113. [11] A.R. Ribeiro, P.M.L. Castro, M.E. Tiritan, Chiral pharmaceuticals in the environment, Environ. Chem. Lett., 10 (2012) 239‐253. [12] A.R. Ribeiro, A.S. Maia, Q.B. Cass, M.E. Tiritan, Enantioseparation of chiral pharmaceuticals in biomedical and environmental analyses by liquid chromatography: An overview, J. Chromatogr. B, 968 (2014) 8‐21. [13] A.R. Ribeiro, C.M. Afonso, P.M.L. Castro, M.E. Tiritan, Enantioselective biodegradation of pharmaceuticals, alprenolol and propranolol, by an activated sludge inoculum, Ecotoxicol. Environ. Saf., 87 (2013) 108‐114. [14] A.R. Ribeiro, C.M. Afonso, P.M.L. Castro, M.E. Tiritan, Enantioselective HPLC analysis and biodegradation of atenolol, metoprolol and fluoxetine, Environ. Chem. Lett., 11 (2013) 83‐90. [15] T.J. Ward, K.D. Ward, Chiral separations: a review of current topics and trends, Anal. Chem., 84 (2012) 626‐635. [16] M. Tian, H.S. Row, K.H. Row, Chiral separation of ofloxacin enantiomers by ligand exchange chromatography, Monatshefte für Chemie ‐ Chemical Monthly, 141 (2010) 285‐290. [17] W. Bi, M. Tian, K.H. Row, Chiral separation and determination of ofloxacin enantiomers by ionic liquid‐assisted ligand‐exchange chromatography, Analyst, 136 (2011) 379‐387.
16
[18] H. Yan, F. Qiao, Rapid screening of ofloxacin enantiomers in human urine by molecularly imprinted solid‐phase extraction coupled with ligand exchange chromatography, J. Liq. Chromatogr. Rel. Technol., 37 (2014) 1237‐1248. [19] J. Grellet, B. Ba, M.C. Saux, High‐performance liquid chromatographic separation of fluoroquinolone enantiomers: a review, J. Biochem. Bioph. Methods, 54 (2002) 221‐233. [20] K. Valliappan, M. Sai Sandeep, Multiple Response Optimization of a HPLC Method for the Determination of Enantiomeric Purity of S‐Ofloxacin, Chromatographia, 77 (2014) 1203‐1211. [21] ICH, Q2B ‐ Validation of Analytical Procedures: Methodology, International Conference on Harmonization Expert Working Group, 1996. [22] D.W. Armstrong, Y. Tang, S. Chen, Y. Zhou, C. Bagwill, J.R. Chen, Macrocyclic antibiotics as a new class of chiral selectors for liquid chromatography, Anal. Chem., 66 (1994) 1473‐1484. [23] A.R. Ribeiro, A.S. Maia, I.S. Moreira, C.M. Afonso, P.M.L. Castro, M.E. Tiritan, Enantioselective quantification of fluoxetine and norfluoxetine by HPLC in wastewater effluents, Chemosphere, 95 (2014) 589‐596. [24] A.R. Ribeiro, L.H. Santos, A.S. Maia, C. Delerue‐Matos, P.M. Castro, M.E. Tiritan, Enantiomeric fraction evaluation of pharmaceuticals in environmental matrices by liquid chromatography‐ tandem mass spectrometry, J. Chromatogr. A, 1363 (2014) 226‐235. [25] A. Berthod, Chiral recognition mechanisms with macrocyclic glycopeptide selectors, Chirality, 21 (2009) 167‐175. [26] M.F. Carvalho, A.S. Maia, M.E. Tiritan, P.M.L. Castro, Bacterial degradation of moxifloxacin in the presence of acetate as a bulk substrate, J. Environ. Manage., 168 (2016) 219‐228. [27] T. Egli, How to live at very low substrate concentration, Water Res., 44 (2010) 4826‐4837. [28] K. Kovar, V. Chaloupka, E. Th, A Threshold Substrate Concentration Is Required to Initiate the Degradation of 3‐Phenylpropionic Acid in Escherichia coli, Acta Biotechnol., 22 (2002) 285‐298. [29] N.A. Zhou, A.C. Lutovsky, G.L. Andaker, H.L. Gough, J.F. Ferguson, Cultivation and characterization of bacterial isolates capable of degrading pharmaceutical and personal care products for improved removal in activated sludge wastewater treatment, Biodegradation, 24 (2013) 813‐827. [30] L. Delgadillo‐Mirquez, L. Lardon, J.‐P. Steyer, D. Patureau, A new dynamic model for bioavailability and cometabolism of micropollutants during anaerobic digestion, Water Res., 45 (2011) 4511‐4521.
17
Figure captions Fig. 1 – Chemical structures of ofloxacin enantiomers. Fig. 2 – Biodegradation patterns during 46 days for: a) racemic ofloxacin; b) racemic ofloxacin with acetate addition; c) (S)‐ofloxacin; and d) (S)‐ofloxacin with acetate addition, under light and dark conditions, based on LC‐FD quantification. Fig. 3 – Overall biodegradation percentages after 46 days of ofloxacin enantiomers performed by the activated sludge inoculum with and without acetate addition and under light and dark conditions, based on LC‐FD quantification. Fig. 4 – LC‐FD chromatograms of (S)‐ofloxacin biodegradation samples at days 0 and 31. Conditions: mobile phase 0.45% TEA (pH 3.6 CH3COOH)/ ethanol (80/20, v/v); flow rate 0.4 mL min‐1; column oven temperature 32 °C; λ exc/em = 290/460 nm. Fig. 5 – LTQ Orbitrap MS spectra of ofloxacin enantiomers on biodegradation samples of (S)‐ ofloxacin (with acetate addition and exposed to light) at days 0 (top) and 31 (bottom) for: a) (S)‐ ofloxacin (retention time: 17 min); and b) (R)‐ofloxacin (retention time: 13 min).`
18
Fig 1
Fig 2
19
b Fig 3
Fig 4
20
Fig 5a
21
Fig 5b
22
Table captions Table 1 – Ofloxacin [1 mg L‐1] enantioseparation results achieved in optimization tests with five different CSPs. Chiral stationary phase
Chromatographic parameters
elution mobile phase composition mode
flow rate (mL min‐1)
column oven temp. (oC)
k1
(S,S)‐Whelk‐ O1
NP
n‐hexane/ethanol/CH3COOH: 25/75/0.01
1.0
n.a.
8.25
1.15 1.24
V
POM
methanol: 100
1.0
n.a.
58.53 1.00 0.00
TAG
PIM
methanol/ethanol/CH3COOH/TEA: 30/70/0.1/0.1
0.8
40
20.30 1.16 0.36
T
RP
0.45% TEA pH 4 CH3COOH/ethanol: 20/80
1.0
38
15.19 1.21 2.40
R
RP
0.45% TEA pH 3.6 CH3COOH/ethanol: 80/20
0.4
32
4.30
column
Pirkle
Instrument parameters
Rs
Chirobiotic
1.46 1.15
NP: normal‐phase; PIM: polar ionic mode; POM: polar organic mode; RP: reversed‐phase; TEA: triethylamine; n.a.:not applicable
Table 2 – Chromatographic results obtained for ofloxacin enantioseparation on Chirobiotic CSPs under reversed‐phase elution conditions. Chiral stationary phase Chirobiotic column
Instrument parameters
eluents (aqueous/organic)
TAG T
Chromatographic parameters
20 mM CH3COONH4/ ethanol
R
mobile flow rate phase (v/v) (mL min‐ 1 )
column oven temp. (oC)
k1
50a/50
0.60
40
19.93 1.19
0.78
50a/50
1.00
40
9.63
1.23
2.04
50 /50
1.00
45
8.80
1.20
1.58
50b/50
0.20
n.a.
8.16
1.55
0.95
50/50
0.2
40
2.98
1.36
0.89
60/40
0.3
40
1.62
1.46
0.92
80/20
0.2
40
3.28
1.44
1.06
85c/15
0.65
32
38.52 0.00
0.00
a
Rs
R TAG
20 mM HCOONH4 pH 4.3/ ethanol
0.45% TEA (CH3COOH)/
23
ethanol
T
0.60
35
5.21
1.33
0.57
60a/40
1.00
38
7.50
1.27
1.59
a
1.00
38
5.98
1.27
1.47
a
55 /45
50 /50
1.00
38
11.94 1.23
1.20
40a/60
1.00
25
7.66
1.33
1.27
a
1.00
38
5.53
1.29
1.61
a
40 /60
20 /80
1.00
38
15.19 1.21
2.40
R
88c/12
32
3.85
1.34
0.70
0.60
35
1.54
1.49
0.98
a
0.60
32
1.46
1.58
1.08
85 /15
0.70
a
86 /14
a
50c/50
a
85 /15
0.65
32
2.03
1.51
1.06
80d/20
0.40
33
1.41
1.61
1.12
80d/20
0.50
33
3.53
1.45
1.09
0.40
32
4.30
1.46
1.15
d
80 /20
pH 4; bpH 6; cpH 3.5; dpH 3.6; TEA: triethylamine; n.a.:not applicable
24
Table 3 – Ofloxacin chemical characteristics, separation performance obtained with ristocetin A‐bonded CSP under the validated conditions, and validation results. Enantiom er
Chemical Chromatograph properties ic parameters (average values)
(S)‐ ofloxacin
Limits
Validation parameters
correlati on level (r2)
detectio quantificati n limit on limit (µg L‐1) (µg L‐1)
accurac y (%)
Rs
rang e (µg L‐1)
1.3 7
‐‐‐
‐‐‐
5.00 y = ‐ 1788307 350 3x ‐ 14844
0.9986
2.5
5.0
96.47 93.17 1.88 99.56 98.38 2.68
2.06 4.04
2.3
1. 6
1.1 9
5.00 y = ‐ 1804215 350 9x ‐ 21760
0.9985
2.5
5.0
95.70 95.01 1.91 99.85 98.92 3.69
2.11 4.56
C18H20FN3 O4 / 361.38
Linearity
structure k' / molecular formula / molecular weight (g mol‐1)
(R)‐ ofloxacin
Method validation
linear regressio n
recover y (%)
intra‐ day precisio n (% RSD)
inter‐ day precisio n (% RSD)