Environ Sci Pollut Res DOI 10.1007/s11356-013-2436-9

RESEARCH ARTICLE

Quantifying interactions between propranolol and dissolved organic matter (DOM) from different sources using fluorescence spectroscopy Na Peng & Kaifeng Wang & Guoguang Liu & Fuhua Li & Kun Yao & Wenying Lv

Received: 25 June 2013 / Accepted: 6 December 2013 # Springer-Verlag Berlin Heidelberg 2014

Abstract Beta blockers are widely used pharmaceuticals that have been detected in the environment. Interactions between beta blockers and dissolved organic matter (DOM) may mutually alter their environmental behaviors. To assess this potential, propranolol (PRO) was used as a model beta blocker to quantify the complexation with DOM from different sources using the fluorescence quenching titration method. The sources of studied DOM samples were identified by excitation–emission matrix spectroscopy (EEMs) combined with fluorescence regional integration analysis. The results show that PRO intrinsic fluorescence was statically quenched by DOM addition. The resulting binding constants (log K oc) ranged from 3.90 to 5.20, with the surface-water-filtered DOM samples claiming the lower log K oc and HA having the highest log K oc. Log K oc is negatively correlated with the fluorescence index, biological index, and the percent fluorescence response (P i,n) of protein-like region (P I,n) and the P i,n of microbial byproduct-like region (P II,n) of DOM EEMs, while it is correlated positively with humification index and the P i,n of UVC humic-like region (P III,n). These results indicate that DOM samples from allochthonous materials rich in aromatic and humic-like components would strongly bind PRO in aquatic systems, and autochthonous DOM containing high protein-like components would bind PRO more weakly. Keywords Propranolol . Dissolved organic matter . Binding constant . Fluorescence quenching EEMs . FRI Responsible editor: Céline Guéguen N. Peng (*) : G. Liu (*) : F. Li : K. Yao : W. Lv Faculty of Environmental Science and Technology, Guangdong University of Technology, Guangzhou 510006, China e-mail: [email protected] e-mail: [email protected] N. Peng : K. Wang Faculty of Chemistry and Environment, Jiaying University, Meizhou 514015, China

Introduction Since the discovery of propranolol (PRO) in the 1960s, betaadrenergic receptor antagonist drugs (so-called beta blockers) have become important therapies for the treatment of angina, glaucoma, heart failure, high blood pressure, and other related conditions (Black and Stephenson 1962; Toda 2003). It is not surprising that several of these compounds can be detected in the environment as a consequence of current beta blocker use by patients in many countries. As a subclass of pharmaceuticals and personal care products (PPCPs), beta blockers have been detected at levels of up to 2 μg/L in sewage treatment plant effluents (Huggett et al. 2002) and 240 ng/L in rivers (Zuccato et al. 2005). Beta blockers have been reported to cause harmful effects on aquatic organisms. They have been shown to decrease the fecundity of Oryzias latipes (Japanese medaka) after PRO exposure at concentrations as low as 0.5 μg/L (Huguet et al. 2009). Cardiovascular dysfunction is one possible consequence of exposing fish to beta blockers, even at concentrations as low as nanograms per liter, leading to impaired fitness (e.g., reduced growth and fecundity) (Owen et al. 2007). Therefore, it is necessary to investigate the fates and behaviors of beta blockers in aquatic systems. Dissolved organic matter (DOM) is ubiquitous and compositionally diverse in natural inland waters with different spatial and temporal distributions resulting from multiple sources and sinks (Brown et al. 2004; Stedmon et al. 2007). It is composed of a complex mixture of substances, including nonhumic solutes such as amino acids, hydrocarbons, carbohydrates, fats, waxes, resins, low molecular weight acids, and aquatic humic substances. DOM composition is determined by the relative contribution and turnover from its precursor materials (Hiriart-Baer et al. 2008; Stedmon et al. 2007), which include both terrestrial (allochthonous) and microbial (autochthonous) organic matter. When compared to autochthonous DOM, allochthonous DOM tends to have a higher

Environ Sci Pollut Res

aromaticity (from lightning incorporation) and lower nitrogen and sulfur contents (Brown et al. 2004; Miller and Chin 2005). Allochthonous DOM has a higher molecular weight distribution than autochthonous DOM (Chin et al. 1994). DOM plays a key role in aquatic systems because its strong interactions with inorganic and organic contaminants may influence their solubility, mobility, bioavailability, and toxicity. Interactions between DOM and hydrophobic organic contaminants have been studied vigorously (Laor and Rebhun 2002; Lee and Kuo 1999; Lu et al. 2012; Pan et al. 2007a, 2008), and interaction studies between DOM and PPCPs, which began later, have also provided some interesting results. Hernandez-Ruiz et al. have reported that the presence of DOM at environmentally relevant concentrations can give rise to PPCP interactions that could potentially influence their environmental transport (Hernandez-Ruiz et al. 2012). Results by Bai et al. (2008) showed that the hydrophobic adsorption/ partitioning may be the major force for carbamazepine binding to humic substances. Carmosini and Lee (2009) reported that the predominant ciprofloxacin sorption in reference humic materials was done by a cation exchange mechanism, with sorption to bio-waste DOM contributing to additional weaker mechanisms. Therefore, it seems that the interactions between PPCPs and DOM may be influenced by PPCPs characteristics such as polarity and DOM composition simultaneously. The objective of the present study was to investigate the interactions between PRO, a representative beta blocker, and DOM from different sources using the fluorescence quenching method. The DOM composition was evaluated by excitation–emission matrix spectroscopy (EEMs) combined with regional integration (FRI) analysis. The interaction mechanisms between PRO and DOM were also discussed.

Materials and methods Reagents and materials Propranolol hydrochloride (CAS = 318-98-9, 99.0 %) was purchased from Sigma (USA) and used as received. Fulvic acid (FA) with a purity of 90 % was obtained from the Aladdin Corporation (Shanghai, China). Chemically pure humic acid (HA) was purchased from Kelong Corporation (Chengdu, China). Suwannee River fulvic acid (SRFA) and Pony Lake fulvic acid (PLFA) were purchased in freeze-dried form from the International Humic Substances Society (IHSS, St. Paul, MN, USA) and used as received. Stock solutions of the above chemicals (with the exception of HA) were prepared from solids in ultra-pure water and stored at 4 °C in the dark before use. The HA was dissolved in an alkaline solution. Sodium chloride, sodium nitrate, sodium nitrite, sodium hydroxide,

and nitric acid of analytical reagent grade were purchased from Fuchen Reagent Corporation (Tianjin, China). Surface-water-filtered DOM and water-extracted sediment DOM preparation Four surface-water-filtered DOM solutions were prepared from two rivers and two lakes in Meizhou City, and were designated as DOMR1, DOMR2, DOML1, and DOML2, respectively. Each water sample was filtered with a 0.45-μm Millipore syringe filter to remove undissolved particulate materials and then stored at 4 °C in the dark. The waterextracted sediment DOM was named DOMS1 and extracted from the sediment of river 1. DOMS1 was isolated by shaking the sediment with water (10 mL = 1 g) for 2 h and centrifuged at 10,000×g. The supernatant was filtered with a 0.45-μm Millipore syringe filter and stored at 4 °C in the dark. These five DOM samples combined with commercial FA and HA, SRFA, and PLFA solutions were used as studied DOM samples from different sources. The DOC concentrations of DOM were determined by high-temperature combustion in a TOC Analyzer (Shimadzu, TOC-V). Fluorescence quenching Interactions between PRO and DOM samples were studied by using the fluorescence quenching method. A 1-mL PRO stock solution of 100 mg/L was added to a series of 10-mL volumetric flasks. A certain volume of the DOM stock solution was then added and diluted to 10 mL of pH 7.0 phosphate buffer solution. The mixture was hand-shaken for 5 min before the fluorescence intensity value was recorded because equilibrium was reached within a short time from the result of a preliminary experiment. Measurements were made with a fluorescence spectrophotometer (Perkin Elmer, LS55) at λ Ex =290 and λ Em =360 nm. The slit was 5 nm and scan speed was 1,000 nm/min. The voltage was set at 650 V. For the temperature-dependent experiment, 10.0 mg/L PRO was quenched by FA at 20 °C, 35 °C, and 50 °C. For the pH-dependent experiment, 10.0 mg/L PRO was quenched by FA/HA at pHs 5.0, 6.0, 7.0, 8.0, and 9.0. For the ionicstrength-dependent experiment, 10.0 mg/L PRO was quenched by FA/HA at NaCl concentrations of 0.001 M, 0.01 M, and 0.1 M. DOM fluorescence spectra analysis DOM fluorescence measurements were made by LS55 fluorescence spectrophotometer. The slit was set at 5 nm for both excitation and emission and the voltage was set at 700 V. A series of emission spectra were collected over a range of excitation wavelengths to provide a complete representation of the sample fluorescence in the form of an excitation–

Environ Sci Pollut Res

emission matrix (EEM), in which the fluorescence intensity was presented as a function of the excitation wavelength on one axis and the emission wavelength was on the other. A wavelength step size of 10 nm was used for the EEM spectra collection. The excitation wavelength ranged from 200 to 490 nm, and the emission wavelength ranged from 300 to 550 nm. All blank spectra were subtracted, and the fluorescence intensity was expressed in arbitrary units. Each experiment was carried out in triplicate, and the results were reported as an average. Considering the inner filter effect, all fluorescence intensity data for DOM samples were corrected by the method reported by the following equation (Gauthier et al. 1986; Pan et al. 2007b): F cor ¼ f cor *F obs f cor ¼

2:3dAex 2:3sAem 10gAem 1−10−dAex 1−10−sAem

ð1Þ ð2Þ

where F obs and F cor are the fluorescence intensity before and after correction, respectively. A ex and A em represent for the absorbance at excitation and emission wavelengths, respectively. The parameters of d , g , and s are the geometric dimensions of the fluorescence cuvette, which are 1, 0.25, and 0.1 cm, respectively (Pan et al. 2007b). Fluorescence quantification A quantitative method named FRI was proposed to analyze EEM plots (Chen et al. 2003b). In this method, fluorescence intensity was integrated beneath each of four EEM regions previously characterized as follows (Birdwell and Engel 2010; Chen et al. 2003b): (I) “protein-like”, (II) “microbial byproduct-like”, (III) “UVC humic-like”, and (IV) “UVA humic-like”. The percent of the fluorescence response of each region was calculated (P i,n =Φ i,n/Φ T,n × 100 %), where Φ i,n is the normalized EEM area volumes of region i, and Φ T,n is the total normalized EEM fluorescence deriving from all regions i. Φ i,n and Φ T,n can be calculated as the method proposed by Chen et al. (2003b).

Results and discussion PRO fluorescence quenching by DOM The PRO fluorescence emission spectrum ranging from 300 to 500 nm with the addition of FA is shown as Fig. 1. The spectrum shows that the PRO fluorescence intensity gradually decreases with increasing FA concentration and there is no change in the spectral shape and no shift at the maxima. The results indicate that the PRO fluorophore could be quenched by FA and that neither PRO conformation nor

Fig. 1 Fluorescence emission spectra of PRO (10 mg/L) with and without FA at pH 7.0 and 25 °C after inner filter correction. The DOC concentrations of FA were 0, 3.125, 6.25, 9.375, 12.5, and 18.75 mg/L

rigidity changed significantly during its interaction with FA (Wu et al. 2013). There are two main types of quenching mechanisms in solution, namely dynamic quenching and static quenching (Pan et al. 2012; Thipperudrappa and Hanagodimath 2013). Dynamic quenching occurs in response to a collision between the excited fluorophore and a quencher. Static quenching requires the formation of a non-fluorescent ground-state complex between the fluorophore and quencher. In order to distinguish the quenching mechanisms, the fluorescence quenching data are analyzed with the Stern– Volmer equation as follows (Lakowicz 2006; Valeur 2002): F o = F ¼ 1 þ K sv ½QŠ

ð3Þ

where F o and F are the fluorescence intensity in the absence and presence of a quencher, K SV is Stern–Volmer constant, and [Q] is the DOC concentration of DOM (mg/L). Figure 2 displays the Stern–Volmer plots for PRO quenching by DOM samples. The Stern–Volmer plots are linear, which generally indicates that one type of quenching mechanism is predominant. Static and dynamic quenching can usually be differentiated by calculating the bimolecular quenching constant (Kq), which reflects the quenching or the accessibility efficiency of the fluorophores to the quencher. Diffusion-controlled quenching typically results in Kq values near 1 × 1010 L/(mol s) (Lakowicz 2006). K sv ¼ K q τ 0 ½QŠ

ð4Þ

where τ 0 is the fluorescence lifetime of the fluorophore in the absence of a quencher, and K q is the bimolecular quenching rate constant.

Environ Sci Pollut Res

Fig. 2 Stern–Volmer plots for different DOM samples (a DOMR1, DOMR2, DOML1, DOML2, DOMS1; b FA, HA, SRFA, PLFA)

DOM in aquatic systems have been reported at sizes up to 20–2,000 kDa (Gjessing 1970; Rashid and King 1969). Fulvic acids, which make up the acid-soluble fraction of DOM and comprise the majority of aquatic DOM, are considered smaller than humic acids and have molecular weight distributions of 200–2,000 Da (Leenheer et al. 2001; Remucal et al. 2012). In this study, we use 200 Da–2,000 kDa as the molecular weight of DOM to calculate the K q with τ 0 of PRO at 10 ns (Bisby et al. 2012). The calculated K q range from 1.60 × 10 11 L/(mol · s) to 6.95 × 10 17 L/(mol · s), which is far greater than the K q of diffusion-controlled quenching, indicating that PRO quenching by different DOM samples is static. The quenching mechanism can be further differentiated by its temperature dependence. The quenching rate constant increases for dynamic Fig. 3 Stern–Volmer plot at different temperatures

quenching with increasing temperature, whereas the increased temperature is likely to yield lower values for the static quenching constant (Pan et al. 2012). As shown in Fig. 3, the Stern–Volmer slope decrease with temperature, thereby indicating static quenching (i.e., complexation) occurs. For static quenching, the dependence of fluorescence intensity on quencher concentration is easily derived by considering the binding constant for complex formation. This constant could be calculated by using the following function: F o = F ¼ 1 þ K oc ½QŠ

ð5Þ

where K oc is the binding constant between PRO and DOM, [Q ] is the DOC concentration of DOM and is expressed as milligrams per liter. K oc is identical to K sv

Environ Sci Pollut Res

Fig. 4 EEM plots of DOM from different sources (a DOMR1, b DOMR2, c DOML1, d DOML2, e DOMS1, f FA, g HA, h SRFA, i PLFA)

since the two constants are calculated by using the Stern–Volmer equation. The calculated log K oc values range from 3.90 to 5.20, which agrees well with the sorption coefficients of some selected endocrine disruptors by DOM (Yamamoto et al. 2003). The log K oc for surface-water-filtered DOM samples are lower than that for FA and HA. The log K oc for HA is almost as 2.4 times as that of FA, which is consistent with the results that the K oc values for HA were at 1.5–2 orders of magnitude higher than those for FA for the same PAH (Danielsen et al. 1995) and the same endocrine disruptor (Yamamoto et al. 2003). The log K oc of SRFA is 4.70 and almost four orders of magnitude higher than that of PLFA. The results show that the K oc values vary considerably considering the DOM source, indicating K oc may highly depend on DOM physical–chemical properties. Therefore, to understand the underlying relationships between DOM from different sources and PRO, the composition and precursor source have also been analyzed and identified.

DOM EEM spectra and fluorescence indices The EEM fluorescence spectroscopy is often used to provide information on the prevalence and structural composition of DOM since fluorescence spectra vary by excitation wavelength and DOM composition (Bakera et al. 2003; Chen et al. 2003a; Hiriart-Baer et al. 2008; Stedmon et al. 2003). The EEM plots for nine DOM types are shown in Fig. 4. DOMR1 and DOMR2 have four same peaks (Birdwell and Engel 2010; Hendersona et al. 2009; Hernandez-Ruiz et al. 2012), with a high intensity peak I at Ex/Em 220(290)/300– 360 nm ascribed to protein-like molecules, a secondary peak at Ex/Em 290–310/370–410 (peak II) which has been attributed to microbial byproduct or anthropogenic input to natural waters (Stedmon and Markager 2005) and has been proposed as a precursor for humic (Burdige et al. 2004). The two DOM samples also have a peak at Ex/Em 240–260/400–460 nm (peak III) attributed to UVC humic-like molecules, in addition to a minor peak at Ex/Em 320–360/420–460 nm (peak IV) correlated to UVA humic-like molecules. The existence of

Environ Sci Pollut Res

peak I and peak II in the EEM of DOMR1 and DOMR2 indicate rivers 1 and 2 have been polluted by protein-like substances that may come from sewage inputs (Bakera et al. 2003; Hendersona et al. 2009). DOML1 and DOM L2 all have peaks II, III, and IV. Compared to DOM L1, the Em of peak III for the DOM L2 exhibits a blue shift which indicates the DOM from river 2 may be less humified than the DOM from river 1. DOMS1 was extracted from river 1 sediment, but it does not include peak I, which indicates the sediment has not been polluted by sewage or protein-like substances yet or has been humified and used by microorganisms. Only the peak IV from FA solution indicates the purity of this sample is high. HA has peaks III, IV, and V (Ex/Em 420/520 nm) which correlated to soil HA (Chen et al. 2003a). SRFA and PLFA all have peaks I, II, III, and IV, but the intensity of peak I in PLFA is much stronger than in SRFA which indicates high N-containing moieties in PLFA (Brown et al. 2004). The peak III of SRFA is red-shifted to that of PLFA, which is in accordance with that SRFA has more aromatic moieties and more humified (Guerard et al. 2009). The ratio of the emission intensity at λ Em 450 nm to that at λ Em 500 nm following excitation at λ Ex 370 nm is referred to as the fluorescence index (FI), which provides a metric for distinguishing DOM derived from terrestrial or microbial sources. FI values of 1.4 or less indicate DOM of terrestrial origin and values of 1.9 or higher correspond to microbederived material (Mcknight et al. 2001). As shown in Table 1, DOMR1 and PLFA have FI values >1.9 which indicate these two DOM samples derived from autochthonous OM (Birdwell and Engel 2010). FA, HA, and SRFA having FI values

Quantifying interactions between propranolol and dissolved organic matter (DOM) from different sources using fluorescence spectroscopy.

Beta blockers are widely used pharmaceuticals that have been detected in the environment. Interactions between beta blockers and dissolved organic mat...
507KB Sizes 0 Downloads 0 Views