w a t e r r e s e a r c h 6 8 ( 2 0 1 5 ) 4 0 4 e4 1 3

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/watres

Interactions between protein-like and humic-like components in dissolved organic matter revealed by fluorescence quenching Zhigang Wang a,b, Jing Cao a,b, Fangang Meng a,b,* a

School of Environmental Science and Engineering, Sun Yat-sen University, Guangzhou 510275, China Guangdong Provincial Key Laboratory of Environmental Pollution Control and Remediation Technology (Sun Yat-sen University), Guangzhou 510275, China

b

article info

abstract

Article history:

Numerous reports have documented the interactions of fluorescent dissolved organic

Received 7 June 2014

matter (FDOM) with other compounds such as metals and trace contaminants by charac-

Received in revised form

terizing the fluorescence quenching of the FDOM components. As FDOM is composed of

8 October 2014

numerous components, inter-component interactions can potentially take place. This

Accepted 9 October 2014

study investigated the interactions between protein-like and humic-like components in

Available online 18 October 2014

FDOM using titration experiments and end-member mixing tests. We found that the cooccurrence of protein-like and humic-like components in FDOM samples resulted in an

Keywords:

overlap behavior between their fluorescence peaks related to inter-component in-

Fluorescent dissolved organic

teractions. Our results suggest that the fluorescence of the protein-like components could

matter (FDOM)

be greatly quenched by the humic-like components in the FDOM samples, e.g., the humic-

Fluorescence quenching

like components from Suwannee River and Nordic Reservoir FDOM yielded significant

Natural organic matter (NOM)

quenching effect for tyrosine (52% and 46%, respectively) and tryptophan (35% and 36%,

Proteins

respectively) in the titration experiments. The fluorescence of the humic-like components,

Humic substances

however, was not impacted by the protein-like components. With the help of complexation modeling, we found that the binding capability between protein-like and humic-like components was dependent on their sources. This study could enhance our current knowledge on the role of FDOM in water and it is also important to the monitoring of FDOM by fluorescence spectroscopy. © 2014 Elsevier Ltd. All rights reserved.

1.

Introduction

Dissolved organic matter (DOM) plays an important role in both natural waters and engineered systems (e.g., water treatment facilities) (Coble, 2007; Henderson et al., 2009; Ishii

and Boyer, 2012). Because of the complexity of DOM (Leenheer and Croue, 2003), accurate and informative monitoring is a challenge when attempting to increase the understanding of DOM characteristics (Henderson et al., 2009; Ishii and Boyer, 2012). In recent years, advanced analytical techniques, such as Fourier-transform ion cyclotron mass

* Corresponding author. Guangdong Provincial Key Laboratory of Environmental Pollution Control and Remediation Technology, Sun Yat-sen University, Guangzhou 510275, China. Tel.: þ86 20 39335060; fax: þ86 20 84110267. E-mail address: [email protected] (F. Meng). http://dx.doi.org/10.1016/j.watres.2014.10.024 0043-1354/© 2014 Elsevier Ltd. All rights reserved.

w a t e r r e s e a r c h 6 8 ( 2 0 1 5 ) 4 0 4 e4 1 3

spectrometry (FT-ICR MS) (Tfaily et al., 2013; Tremblay et al., 2007; Zhang et al., 2014), multidimensional nuclear magnetic resonance (NMR) spectroscopy (Pautler et al., 2012; Tfaily et al., 2013) and excitation-emission matrix (EEM) fluorescence spectroscopy (Carstea et al., 2014; Pautler et al., 2012; Tfaily et al., 2013; Tremblay et al., 2007), have been actively used for the characterization of DOM. Monitoring of EEM fluorescence has been cited as a powerful technique for the characterization of FDOM (Nebbioso and Piccolo, 2013). Modeling of parallel factor (PARAFAC) analysis, which can decompose the EEM spectrum into independent fluorophore groups (Stedmon and Bro, 2008), has substantially improved the capacity of EEM in characterizing FDOM (Nebbioso and Piccolo, 2013; Stedmon and Bro, 2008). As such, EEM measurements and PARAFAC analysis have been increasingly used to explore organic matter in water (e.g., freshwaters  et al., 2014; Mladenov et al., 2011; Murphy et al., 2008) (Jaffe and human-impacted waters (Baker, 2001; Goldman et al., 2012; Meng et al., 2013; Yang et al., 2014a; Zhang et al., 2011)). However, the EEM measurements have not fully considered the potential impacts of the interactions between fluorophore groups on the measured data. Aquatic DOM contains numerous chemical molecules with a high content of oxygenated reactive functional groups, such as carboxylic, phenolic and alcoholic groups (Plaza et al., 2006). The presence of these groups enables DOM to have high complexation capacities (Plaza et al., 2006; Tipping, 2002), such as complexation with metals (Chappaz and Curtis, 2013; Riedel et al., 2012, 2013; Yamashita and Jaffe, 2008; Yan and Korshin, 2014) and pharmaceuticals (Hernandez-Ruiz et al., 2012). The binding propensity of FDOM or DOM with other compounds is usually characterized by changes in optical properties (Chappaz and Curtis, 2013; Plaza et al., 2006; Yamashita and Jaffe, 2008; Yan et al., 2013). It has been demonstrated that the combined use of EEM measurements and PARAFAC analysis was sensitive enough to determine the binding capacities of FDOM, which were often revealed by fluorescence quenching (Wu et al., 2011; Yamashita and Jaffe, 2008). Previous studies were mostly focused on how and to what extent other compounds (e.g., heavy metals and pharmaceuticals) interacted with FDOM. However, we should note that FDOM itself is composed of numerous molecular assemblies (Peuravuori and Pihlaja, 2004; Piccolo, 2001; Romera-Castillo et al., 2014) that can potentially take part in inter-molecule or inter-component interactions. Further investigation of FDOM inter-component interactions is of high interest for the understanding of FDOM transport in water ecosystem and the monitoring of fluorophores by fluorescence spectroscopy. Human activities have seriously impacted the concentrations, compositions and characteristics of DOM in urbanized rivers, lakes and coastal oceans (Baker, 2001; Goldman et al., 2012; Meng et al., 2013; Mladenov et al., 2011; Mostofa et al., 2013; Murphy et al., 2008; Zhang et al., 2011). The discharge of treated wastewater is one of the most important anthropogenic inputs to DOM in water (Baker, 2001; Goldman et al., 2012). Note that effluent organic matter (EfOM) has different compositions and characteristics from naturally occurring organic matter (NOM) (Goldman et al., 2012; Neale et al., 2012; Yang et al., 2014b), e.g., EfOM is composed of soluble microbial products and non-biodegraded wastes (Meng et al., 2009) and

405

has a higher protein abundance than NOM (NOM is often dominated by humic-like substances). Thus, wastewaterimpacted FDOM in urbanized rivers could appear as a significant Peak T (tryptophan-like fluorophores) and/or a Peak B (tyrosine-like fluorophores) at lower Ex/Em wavelengths in the EEM spectra (Baker, 2001; Henderson et al., 2009; Meng et al., 2013). Generally, the mixing behavior of FDOM with different sources could give rise to interactions between their end-members pools (Yang and Hur, 2014), leading to changes in their optical properties (Hur et al., 2006; Myat et al., 2013). We expect that different fluorophore groups in wastewaterimpacted FDOM, such as protein-like and humic-like substances, will likely interact with each other (Myat et al., 2014); however; few studies have thus far been performed to assess the potential for fluorescence quenching within FDOM components. The aim of this study is to reveal the fluorescence quenching behavior within FDOM components and to reveal to what extent they interact with each other. Interactions between protein-like components (tyrosine and tryptophan) and humic-like components (two NOM samples purchased from the International Humic Substances Society (IHSS)) were ascertained using titration experiments. This study is novel in revealing the naturally occurring fluorescence quenching within the FDOM components.

2.

Materials & methods

2.1.

DOM source materials

Suwannee River NOM (SRNOM, 2R101N) and Nordic Reservoir NOM (NRNOM, 1R108N) were obtained from the IHSS. LTryptophan (99.0e101.0%), L-Tyrosine (>99%) and fatty-acidfree bovine serum Albumin (BSA) were obtained from Sigma. Solutions of the two NOM samples (50 mg/L), tryptophan (5 mg/L), tyrosine (5 mg/L) and BSA (50 mg/L) were prepared using Milli-Q water. Large particles as a result of incomplete solubilization in the NOM solutions were removed by filtering with pre-rinsed membranes (0.22 mm, PVDF, Millipore, USA). EfOM samples were collected from the effluent (after the final disinfection using the chlorination method) of a local wastewater treatment facility, which employed anaerobic-anoxicoxic (A2/O) process treating domestic wastewater. The samples were stored in acid-cleaned plastic containers, and transport to our laboratory within 1 h. Then, they were stored at 4  C until needed. The sampling campaign was conducted on morning (7:30), noon (11:30) and evening (21:00) of three different days. The EfOM sample used in this study was the mixture of these samplings. The dissolved organic carbon (DOC), total nitrogen (TN), and total phosphorous (TP) of the samples were determined to be 12.6 mg-C/L, 1.5 mg-N/L, and 0.24 mg-P/L, respectively, on average.

2.2.

Fluorescence titration

Concentrations of amino acids in natural waters are often quit low, e.g., concentrations of tryptophan and tyrosine in a bay were determined to be about 0.8 mg/L and 3.1 mg/L, respectively (Yamashita and Tanoue, 2003b). In comparison, the

406

w a t e r r e s e a r c h 6 8 ( 2 0 1 5 ) 4 0 4 e4 1 3

tryptophan-like and tyrosine-like components in wastewaterimpacted FDOM samples can yield comparable or higher fluorescence intensity with humic-like components in the samples (Meng et al., 2013). In the fluorescence titration experiments, thus, the concentrations of tryptophan and tyrosine solutions were finally fixed at 250 mg/L, which resulted in comparable fluorescence intensity with that of humic-like in wastewater-impacted FDOM samples. Before the titration, the vials were cleaned thoroughly with acid and Milli-Q water and then dried thoroughly at 105  C. The fluorescence titration of tryptophan or tyrosine with NOM was as following. Using an automatic syringe, 5 mL aliquots of a diluted solution of tryptophan or tyrosine were titrated into 10-mL brown sealed vials containing the SRNOM or NRNOM solutions. The concentrations of NRNOM and SRNOM in the final solutions ranged from 0 to 10 mg/L, and the tryptophan or tyrosine concentration in the final solutions was fixed at 250 mg/L. The pH of all the titrated solutions was adjusted to approximately 7.0 using 0.1 M NaOH or HCl. To avoid the influence of background ionic strength on the fluorescence measurements, the ionic strength of all solutions was adjusted to 0.01 M using NaCl (Hur et al., 2006; Lu and Jaffe, 2001). To ensure full complexation, shaking of the titrated samples did for 2 h at room temperature in the dark. The titration was conducted in duplicate to obtain reliable data. We noted that the preparation and shake of the samples did not lead to alterations or loss of fluorescence intensity of either NOM or amino acids. In a similar way, the fluorescence titration of NOM (5 mgDOC/L) with tryptophan or tyrosine (0e200 mg/L) was conducted. The only difference was that the tryptophan or tyrosine was titrated into 10-mL brown sealed vials containing the SRNOM or NRNOM solutions with a fixed concentration of 5 mg-DOC/L in the final solutions. The tryptophan or tyrosine concentration in the final solutions was in the range of 0e200 mg/L.

2.3.

End-member mixing experiments

Differing from tryptophan or tyrosine, the EfOM and large-size proteins (BSA) are of low fluorophore abundance. So, the interactions between NOM and EfOM/BSA were investigated using laboratory mixing experiments with two end-members (i.e., NOM and EfOM, NOM and BSA). The BSA has been widely used as a proxy to model the transport and behavior of largesized proteins in water or wastewater. More importantly, the BSA is of high purity which could warrant that there is no interference of other chemicals on the interactions between protein-like and humic-like components. The mixtures were prepared at eleven DOC percentage ratios (i.e., 100:0, 90:10, 80:20, 70:30, 60:40, 50:50, 40:60, 30:70, 20:80, 10:90 and 0:100). The total DOC of all the mixtures was kept at 10 mg/L. Prior to EEM measurements, the mixtures were shaken for 2 h at room temperature in dark. Similar to the tritration experiments, the pH and background ionic strength of all mixtures were adjusted.

2.4. Fluorescence measurements and PARAFAC modeling Measurements of three-dimensional EEM spectra were performed at room temperature (~25  C) using a fluorescence

spectrophotometer with a Xenon lamp light source (F-4500, Hitachi, Japan). The EEM spectra of samples and Milli-Q water (blank EEMs) were scanned over an excitation range of 240e400 nm (increments of 5 nm) and an emission range of 250e550 nm (increments of 2 nm) with a slit size of 5 nm. The scan rate for all samples was set to 1200 nm/min. Rayleigh light scattering was eliminated using a 290 nm emission cutoff filter. Prior to measuring, highly concentrated samples were diluted with Milli-Q water to achieve absorbance values of less than 0.1 cm1 at 254 nm, which meets the requirement for eliminating inner-filter effects during fluorescence measurements (Miller et al., 2010). According to the manufacturer's suggested protocols, the spectrophotometer was corrected for excitation and emission. The sample EEMs were obtained by subtracting the water blank EEM and then correcting the dilution. The relative standard deviation of fluorescence intensity was less than 5% for three replicated measurements of the samples. Variations in the resultant fluorescence intensity from lamp intensity fluctuations were calibrated using both a quinine sulfate (QS) solution and Milli-Q water after every five samples were measured. Fluorescence intensities were standardized by quinine sulfate units (QSU) (Yamashita and Tanoue, 2003a). The correction and calculation of the EEMs were conducted using the software Matlab7.0 (Math Works Inc., USA). We constructed a PARAFAC model for the NOM mixtures (NOM-tryptophan, NOM-tyrosine and NOM-BSA) and their end-members. A total of 174 EEMs were modeled with PARAFAC. The modeling process was conducted using the DOMFluor Toolbox, as suggested by Stedmon and Bro (Andersen and Bro, 2003). Validity of the PARAFAC model and the appropriate number of components were assessed and determined primarily based on two conditions: (1) EEM residuals should be dominated by instrument noise rather than visible fluorophore-like peaks (Stedmon et al., 2003; Ziegelgruber et al., 2013) and (2) a perfect model should have a low sum of squared error and be able to pass a core consistency test (CC > 80%) (Andersen and Bro, 2003). The appropriate component number was determined using split-half analysis and random initialization (Andersen and Bro, 2003). Because the PARAFAC model of the DOM from various sources was validated (CC > 90%), we presented the fluorophore groups based on PARAFAC modeling rather than direct “peak-picking” on the EEM spectra.

2.5.

Complexation modeling

Inter-component actions were described using the complexation model reported elsewhere (Plaza et al., 2006; Ryan and Weber, 1982). A prerequisite for this model is the assumption that inter-component bindings occur at independent and identical sites with a 1:1 stoichiometry. Thus, for the application of this model, the independent fluorescence components generated by PARAFAC modeling are more appropriate than direct “peak-picking” of intensity (Yamashita and Jaffe, 2008). The binding properties between the fluorescence components can be quantified using the following nonlinear equation:

w a t e r r e s e a r c h 6 8 ( 2 0 1 5 ) 4 0 4 e4 1 3

407

Fig. 1 e EEM spectra of the four endemembers and two mixtures. a, d, EEM spectra of tryptophan (250 mg/L; a) and tyrosine (250 mg/L; d); b, e, EEM spectra of SRNOM (9.36 mg/L; b) and NRNOM (9.36 mg/L; e); c, f, EEM spectra of tryptophan-SRNOM (250 mg/L for tryptophan and 9.36 mg/L for SRNOM, c); and tyrosine-NRNOM (250 mg/L for tyrosine and 9.36 mg/L for NRNOM, f).



 1 I ¼I0 þ IML þ I0 2KM CL
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi  1 þ KM CL þ KM CM  ð1 þ KM CL þ KM CM Þ2  4K2M CL CM where I and I0 are the fluorescence intensity (FImax) at the quencher concentration of CM and at the beginning of the titration (without any addition of quencher), respectively; IML is the limiting value below which the FI does not change as the quencher dosage is increased; and CL and KM are the total ligand concentration and the conditional stability constant, respectively. The data were fitted using the above model equation with Statistical Package for the Social Science (SPSS) software. The proportion of the initial FI corresponding to its binding of fluorophores (f) was determined by the following equation: f¼

3.

ðI0  IML Þ  100 I0

Results

3.1. Fluorescence characteristics of the end-members and their mixtures EEM spectra of the four end-members and two of their mixtures (tryptophan-SRNOM and tyrosine-NRNOM) are shown in Fig. 1. For brevity, additional EEM spectra are detailed in the Supplementary Materials file (see Fig. S1). It can be seen that the EEM spectra of tryptophan and tyrosine are characterized

by a peak at Ex/Em ¼ 275/350 nm (Peak T) and Ex/Em ¼ 275/ 300 nm (Peak B), respectively. The two NOM samples had similar shapes and locations of peak regions in the EEM spectra, with two significant peaks at Ex/Em ¼ 245/450 nm and Ex/Em ¼ 330/450 nm, referred to as Peak A and Peak C, respectively (Henderson et al., 2009). The occurrence of Peak A and Peak C is indicative of the presence of terrestrial humiclike substances. Fluorescence peaks were not found in the low Ex/Em region of the EEM spectra of the NOM samples, indicating that the two NOM samples contained negligible protein-like components. In addition, Peaks A and C in the two NOM samples overlapped with each other. It is interesting to note that Peaks T and B elongated significantly into Peak A in the EEM spectra of the tryptophan-SRNOM and tyrosineNRNOM mixtures, leading to changes in the shape of Peak A for the mixtures. Moreover, it can be seen that the Ex/Em of Peak A in the mixture (Fig. 1c and f), compared with that in the NOM samples (Fig. 1b and e) occurred blue shift by ca. 10 nm and 20 nm, respectively. The overlap was usually attributable to the inter-molecular energy transfer, or the presence of multiple fluorophores, or both (Boehme and Coble, 2000). We can rule out the possibility of the presence of multiple fluorophores because the two types of end-members generate fluorescence peaks at two significantly different regions (low Ex/Em for tryptophan and tyrosine and high Em for SRNOM and NRNOM). Thus, the significant overlap between Peak B or T and Peak A was due to the energy transfer between proteinlike and humic-like components (Wang and Zhang, 2014) as the DOM is usually present in water in the form of molecular assemblies held together by weak dispersive forces (Peuravuori and Pihlaja, 2004; Piccolo, 2001; Romera-Castillo

408

w a t e r r e s e a r c h 6 8 ( 2 0 1 5 ) 4 0 4 e4 1 3

et al., 2014). Thus, the overlap between Peak B/T and Peak A was attributable to interactions of components rather than independent fluorescence joint. Nevertheless, there was no significant overlap between Peak C and Peak B/T in the EEM spectra of the mixture (Fig. 1c and f). A possible explanation for this could be due to the fact that the Peak A-related compounds are much richer in the NOM than the Peak C-related ones (the fluorescence intensity ratio of Peak A to Peak C were determined to be 2.7 and 2.3 for SRNOM and NRNOM, respectively), thus having more opportunity to interact with the dosed amino acids. To make clear this phenomenon, the interactions between protein-like fluorophores (i.e., Peak B/T) and the Peak A-only and Peak C-only fluorophores should be investigated.

3.2. Interactions between protein-like and humic-like components PARAFAC modeling can successfully decompose the EEM spectra of the end-members and their mixtures into four independent components (see Fig. 2), with a validation of 90% (>the requirement of 80% (Andersen and Bro, 2003)). PARAFAC modeling yielded two humic-like components (C1 and C3) and two protein-like components (C2 and C4). C1 and C3 were attributable to the presence of terrestrial humic substances, which only appear in the NOM-containing samples. In contrast, C2 and C4 were mainly present in the

tryptophan- and tyrosine-containing samples, respectively. Thus, the fluorescence components belonging to the four end-members (SRNOM, NRNOM, tryptophan and tyrosine) were successfully extracted from the EEM spectra of their mixtures. Changes in the maximum fluorescence intensity (FImax) of protein-like end-members (C2 and C4) with the addition of humic-like end-members are presented in Fig. 3a. The duplicate titration experiments had good reproducibility of data curves, as indicated by the nominal deviations. Fig. 3a shows that the FImax levels of tryptophan (C2) and tyrosine (C4) decreased markedly with the addition of SRNOM or NRNOM, implying that the fluorescence intensity of proteinlike components from tryptophan or tyrosine were substantially quenched by the humic-like components from SRNOM or NRNOM. SRNOM and NRNOM produced a larger quenching effect for tyrosine (52% and 46%) than tryptophan (35% and 36%). In addition, as shown in Table 1 the fitted stability constants (KM) of SRNOM with the two protein end-members (KM ¼ 0.48 and 0.68 L/mg for tyrosine and tryptophan, respectively) were also much higher than those of NRNOM (KM ¼ 0.27 and 0.15 L/mg for tyrosine and tryptophan, respectively), which was likely caused by the lower total ligand concentration of SRNOM than NRNOM, as indicated by CL. Similarly, the f values of SRNOM (70.1% and 44.7% for tyrosine and tryptophan, respectively, with a fixed concentration of 250 mg/L) were also lower than those of NRNOM (80.7% and 74.8% for tyrosine and tryptophan, respectively).

Fig. 2 e EEM contours of the four components obtained by the DOM FluorePARAFAC model. a, Ex/Em ¼ 250e275(280e400)/ 370e500 nm (Components 1, C1); b, Ex/Em ¼ 250e300/320e410 nm (Components 2, C2); c, Ex/Em ¼ 250e270(290e350)/ 310e450 nm (Components 3, C3); d, Ex/Em ¼ 260e290/290e340 nm (Components 4, C4); C1 and C3 were only present in NOM-containing samples, whereas C2 and C4 were primarily present in the tryptophan- and tyrosine-containing samples. The contribution of NOM to the C2 and C4 in the mixtures of NOM-tryptophan or NOM-tyrosine was corrected, although their abundances are at very low levels.

409

w a t e r r e s e a r c h 6 8 ( 2 0 1 5 ) 4 0 4 e4 1 3

Fig. 3 e Changes in fluorescence intensity during the fluorescence titration. a, Changes in fluorescence intensity of the tryptophan (C2) and tyrosine (C4) solutions at a concentration of 250 mg/L when dosed with SRNOM or NRNOM (0-10 mg-DOC/L); b, Changes in fluorescence intensity of the two humic-like components belonging to SRNOM at a concentration of 5 mg-DOC/L when dosed with tryptophan or tyrosine (0e200 mg/L).

The differences imply that the interaction behavior depends on the origins of the end-members. Despite the different quenching effects, these results suggest that the presence of humic-like components in FDOM could lead to a significant

underestimation of protein-like components during fluorescence spectroscopy measurement. The interactions of NOMeNOM, tryptophanetryptophan and tyrosineetyrosine could be ruled out because the FImax of each end-member generally increased in a linearly concentration-dependent manner as the concentration varied in a certain range (0e10 mg/L for the two NOM and 0e250 mg/L for the two amino acids) (see Fig. S3). However, the fluorescence titration of humic-like end-members (at a fixed final DOC concentration of 5 mg/L for SRNOM or NRNOM) with the addition of protein-like end-members (with concentrations increasing from 0 to 200 mg/L for tryptophan or tyrosine) showed that the dosage of protein-like components did not cause a fluorescence quenching effect in the humic-like components (see Fig. 3b). In addition, the UVevis spectra of SRNOM samples did not change significantly with dosage of tyrosine or tryptophan (0e250 mg/L) (see Fig. S4). All these phenomena suggest that the humic-like and protein-like components took place unidirectional interactions rather than bidirectional interactions. A potential reason for the unidirectional interactions could be due to the fact that the NOM concentrations (1e10 mg/L) were much higher than those of amino acids (0e200 mg/L for tryptophan or tyrosine). End-member mixing experiments for BSA and NOM revealed similar evidence that the fluorescence of proteinlike components with much larger sizes (66 KDa) could be substantially quenched by the humic-like components (see Fig. 4), whereas the humic-like components (C1 and C3) in the NOM end-member were not impacted by the dosage of BSA. It appears that BSA and NOM also occurred unidirectional interactions despite of their equivalent DOC range to NOM (0e10 mg/L for both NOM and BSA). Thus, the humic-like components in the FDOM samples could be properly quantified by fluorescence spectroscopy irrespective of the presence of protein-like components. Experimental evidence from fluorescence titration and end-member mixing implies that the presence of protein-like compounds, regardless of their size and concentrations, did not result in changes in the fluorescence intensity of humic-like components in NOM. Nevertheless, the end-member mixing experiments for EfOM and NOM in our current study revealed that there was a conservative mixing between EfOM and NOM; that is, the measured FImax of all the protein-like or humic-like components were close to the theoretical values (see Fig. S5). This is in good agreement with a recent study reported by Goldman et al. (Goldman et al., 2012). The conservative mixing behavior of the protein-like components in the EfOM-NOM mixing experiments was due to the fact that the fluorescence of protein-like components in the EfOM samples has been already fully quenched by the humic-like components

Table 1 e Fitting parameters of the complexation model on the data obtained in the titration experiments, i.e., the conditional stability constant (KM), total ligand concentration (CL), the limiting fluorescence intensity value below which the FI does not change (IML), the fraction of the fluorescence intensity that corresponding to binding fluorophores (f). KM (L/mg) NRNOM þ tyrosine NRNOM þ tryptophan SRNOM þ tyrosine SRNOM þ tryptophan

0.275 ± 0.149 ± 0.479 ± 0.684 ±

0.0637 0.00780 0.108 0.153

CL (mg/L) 8.00 10.98 5.65 4.11

± 1.16 ± 0.345 ± 0.781 ± 0.741

R2

IML

IML/I0

0.9986 0.9962 0.9943 0.9756

1.21 5.76 1.780 13.71

0.193 0.252 0.299 0.553

f ¼ (1  IML/I0)  100 (%) 80.70 ± 74.78 ± 70.09 ± 44.67 ±

7.55 3.74 3.54 1.33

410

w a t e r r e s e a r c h 6 8 ( 2 0 1 5 ) 4 0 4 e4 1 3

Fig. 4 e Relation of measured and theoretical fluorescence intensity with the BSA proportion in the BSAeSRNOM mixture. The two endemembers for the mixing experiments were prepared with a DOC of 10 mg/L a, Relation of the protein-like components (C2 and C4) with the BSA percent in the BSA-SRNOM mixture; b, Relation of the humic-like components (C1 and C3) with the BSA percent in the BSA-SRNOM mixture. The theoretical data were calculated using mass balance and the measured fluorescence intensity of pure BSA and SRNOM.

and some other compounds such as heavy metals in the EfOM itself. As such, the protein-like components in the EfOM could not be changed further when being dosed with more humic-like substances.

4.

Discussion

4.1. Increasing understanding of the fluorescence quenching of DOM As aforementioned, the interactions of DOM with other compounds have often been studied by characterizing the fluorescence quenching of FDOM components (Chappaz and Curtis, 2013; Hernandez-Ruiz et al., 2012; Riedel et al., 2012; Yamashita and Jaffe, 2008). However, we should note that these investigations mainly focused on how the dosed compounds impact the fluorescence feature of FDOM, without sufficient consideration of the inherent interactions among

the FDOM components. Our current results demonstrate the FDOM can take place inter-component interactions. This result can be supported by previous findings that the FDOM is present in the form of supramolecular assembly structure, which is held together by weak dispersive forces (e.g., p-p interactions and van der Waals forces), in water (Peuravuori and Pihlaja, 2004; Piccolo, 2001; Romera-Castillo et al., 2014). As such we can expect that proteins or amino acids and humic substances can potentially form supramolecular assemblies. According to the significant overlap between Peak B or T and Peak A (see Fig. 1), intermolecular energy transfer is an important mechanism responsible for the fluorescence quenching of the protein-like components by the humic-like components. It also suggests that the molecular assemblies of the two components interact with each other tightly, as efficient intermolecular energy transfer requires small donor˚ ~ 100 A ˚ ) (Clegg, 1995). According to to-acceptor distance (10 A the fact that the emission intensity and the fluorescence lifetime of DOM decreased with increasing molecular size, previous efforts have pointed out the presence of intermolecular energy transfer between chromophores of DOM such as individual chemical molecules (Boyle et al., 2009; Richard et al., 2011). In addition, Wang and Zhang recently reported that presence of energy transfer from polycyclic aromatic hydrocarbons to FDOM occurs because of the significant overlap between the emission spectra of polycyclic aromatic hydrocarbons and absorbance spectra of FDOM (Wang and Zhang, 2014). In fact, the humic substances themselves can also present intermolecular energy transfer between their electron donor groups (e.g., indoles, polyhydroxylated aromatics) and acceptor groups (e.g., oxidized aromatics and quinones) (Boyle et al., 2009; Del Vecchio and Blough, 2004); and the extent of the intermolecular energy transfer seemed to increase with increasing molecular complexity of the humic substances (Mignone et al., 2012). Previously, it has been reported that the fluorescence quenching curve of the humic-like components showed good relationship with the dosed heavy metals; whereas, that of the protein-like components fluctuated (Wu et al., 2011; Yamashita and Jaffe, 2008). Especially, the fluorescence enhancement of protein-like components after significant Cu2þ complexation of FDOM increased sharply, which is clearly related to a reduction on protein-like to humic-like interaction due to metal-induced steric hindrance (Yamashita and Jaffe, 2008). Our current results, plus previous efforts on intermolecular energy transfer and molecular assemblies, suggest that the fluctuation could be in part due to the competition of humic-like and protein-like components when binding with heavy metals. Further investigation on the competition of humic-like and protein-like components during their interactions with heavy metals could aid in revealing such an assumption.

4.2. Significance for understanding the transport and fate of DOM in waters Because of its strong binding/complexation capacity, DOM could affect the mobility and transport of other compounds such as trace contaminants (Williams et al., 2000), heavy metals (Linnik, 2003; Lu and Jaffe, 2001) and nano materials

w a t e r r e s e a r c h 6 8 ( 2 0 1 5 ) 4 0 4 e4 1 3

(Pallem et al., 2009) in waters or soils. The DOM-amended transport of these compounds is usually due to the formation of molecular assemblies by weak dispersive forces (Peuravuori and Pihlaja, 2004; Piccolo, 2001; Romera-Castillo et al., 2014) and the changes in the surface properties because of hydrophobic interactions through partitioning (Mei et al., 2009). Based on the current study, we can suggest that protein-like components could be trapped by humic-like components; and then they form molecular assemblies, which ultimately lead to fluorescence quenching of the protein-like components. Thus, the transport of protein-like components could be amended because of the strong binding role of humic-like components. In addition, another important consequence of interactions between protein-like and humic-like components could be the changes in photodegradation and biodegradation of the FDOM. It can be expected that the degradation of one given DOM molecule should strongly depend on the physical form of the molecular assemblies, e.g., the surface sorbed DOM molecules have more opportunities to adsorb sunlight or to contact with microbes. A recent study conducted by Carlos et al. has suggested that the hydrophobic compounds located inside the aggregates of humic substances could be protected against photosensitized degradation by singlet oxygen species, because of the preferential consumption of the singlet oxygen species by the humic substances themselves (Carlos et al., 2011). Thus, because of the formation of molecular assemblies, the photodegradation and biodegradation of the mixture of protein-like and humic-like components are expected to be quite different from their end-members.

4.3. Significance for FDOM measurements by fluorescence spectroscopy Proteins are ubiquitous in freshwater and particularly in wastewater-impacted waters, e.g., the FImax of proteins accounts for a profound portion in the human-impacted rivers or in wastewater effluents, e.g., up to 50% of the total fluorescence in the urbanized Zhujiang river was contributed by proteins without consideration of fluorescence quenching (Meng et al., 2013). Operationally, the presence of protein-like fluorescent components in the FDOM is indicative of either microbial activity (freshwater) (Stedmon and Markager, 2005; Stedmon et al., 2007) or anthropogenic inputs (wastewater-impacted water) (Goldman et al., 2012). The FImax of protein-like components in freshwater, except for algal laden waters, is often much lower than that in wastewater-impacted water, which could be due to the low quantity of proteins in waters, according to the good linear relationship between protein concentrations and FImax levels (Yamashita and Tanoue, 2003a). Moreover, our results imply that fluorescence quenching of proteins by the humic-like components could be another potential reason for the low FImax of the protein-like components, which could result in the underestimation of proteins by fluorescence spectroscopy. We found that the measured FImax of protein-like components (250 mg/L of tryptophan or tyrosine) decreased by approximately 35%e52% when a mixture's NOM content was increased to approximately 10 mg/L. As such, the actual abundance of protein-like fluorescent components in FDOM should be higher than what is measured by fluorescence

411

spectroscopy. The degree of naturally occurring fluorescence quenching in FDOM samples, in particular for the wastewaterimpacted FDOM, should be considered when calculating actual fluorescence abundances of each component. A possible alternative solution is to separate protein-like components from humic-like components in their molecular assemblies based on their different properties such as molecular size distribution. For instance, the protein-like components are enriched in the lowest and highest molecular size fractions obtained by size exclusion chromatography (Romera-Castillo et al., 2014). In addition, we should note that the fluorescence quenching of other potential protein-like compounds, such as polyphenols produced by plant leaves (e.g., red mangrove) (Cuss and Gueguen, 2012; Maie et al., 2007), by humic-like compounds should be further revealed in future study. Additionally, it is of some interest to assess or distinguish the contributions of static and dynamic quenching using additional methods such as laser flash photolysis (Cottrell et al., 2013).

5.

Conclusions

DOM plays important roles in both geochemistry and environmental science. In this work, we explored the intercomponent interactions of protein-like and humic-like components and revealed the naturally occurring fluorescence quenching within the FDOM components. The main conclusions can be drawn as follows: (1) The presence of protein-like and humic-like components in a given FDOM resulted in an overlap behavior between peaks A and C. Moreover, the presence of humic-like components led to a blue shift of Ex and Em of Peak A by ca. 10 nm and 20 nm, respectively. (2) The measured FImax of protein-like components (250 mg/ L of tryptophan or tyrosine) decreased by approximately 35%e52% when a mixture's NOM content was increased to approximately 10 mg/L. The fluorescence of the humic-like components, however, was not impacted by the protein-like components (tryptophan or tyrosine). End-member mixing experiments for BSA and NOM revealed similar evidence. (3) The fitted stability constants (KM) of SRNOM with the two protein end-members (KM ¼ 0.48 and 0.68 L/mg for tyrosine and tryptophan, respectively) were much higher than those of NRNOM (KM ¼ 0.27 and 0.15 L/mg for tyrosine and tryptophan, respectively), which was likely caused by the lower total ligand concentration of SRNOM than NRNOM. Similarly, the f values of SRNOM (70.1% and 44.7% for tyrosine and tryptophan, respectively) were also lower than those of NRNOM (80.7% and 74.8% for tyrosine and tryptophan, respectively). The binding capability between protein-like and humic-like components was dependent on their sources. (4) The findings of this work have two important implications: (1) the inter-component interactions of FDOM have a substantial potential to change the dynamics and fate of DOM in water ecosystems; and (2) the currently used fluorescence measurements cannot reveal the actual abundance of protein-like components

412

w a t e r r e s e a r c h 6 8 ( 2 0 1 5 ) 4 0 4 e4 1 3

in FDOM, which should be higher than what is measured by fluorescence spectroscopy.

Acknowledgments The project was supported by the New Century Excellent Talents in University from the Ministry of Education of China (NCET-11-0537) and the Fundamental Research Funds for the Central Universities (No. 12lgpy52). The authors would like to thank Dr. Y. Yamashita and Dr. K.M.G. Mostofa for their constructive comments and suggestions on this manuscript.

Appendix A. Supplementary data Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.watres.2014.10.024.

references

Andersen, C.M., Bro, R., 2003. Practical aspects of PARAFAC modeling of fluorescence excitation-emission data. J. Chemom. 17 (4), 200e215. Baker, A., 2001. Fluorescence excitation-emission matrix characterization of some sewage-impacted rivers. Environ. Sci. Technol. 35 (5), 948e953. Boehme, J.R., Coble, P.G., 2000. Characterization of colored dissolved organic matter using high-energy laser fragmentation. Environ. Sci. Technol. 34 (15), 3283e3290. Boyle, E.S., Guerriero, N., Thiallet, A., Del Vecchio, R., Blough, N.V., 2009. Optical properties of humic substances and CDOM: relation to structure. Environ. Sci. Technol. 43 (7), 2262e2268. Carlos, L., Pedersen, B.W., Ogilby, P.R., Martire, D.O., 2011. The role of humic acid aggregation on the kinetics of photosensitized singlet oxygen production and decay. Photochem. Photobiol. Sci. 10 (6), 1080e1086. Carstea, E.M., Baker, A., Bieroza, M., Reynolds, D.M., Bridgeman, J., 2014. Characterisation of dissolved organic matter fluorescence properties by PARAFAC analysis and thermal quenching. Water Res. 61, 152e161. Chappaz, A., Curtis, P.J., 2013. Integrating empirically dissolved organic matter quality for WHAM VI using the DOM optical properties: a case study of Cu-Al-DOM interactions. Environ. Sci. Technol. 47 (4), 2001e2007. Clegg, R.M., 1995. Fluorescence resonance energy transfer. Curr. Opin. Biotechnol. 6, 103e110. Coble, P.G., 2007. Marine optical biogeochemistry: the chemistry of ocean color. Chem. Rev. 107 (2), 402e418. Cottrell, B.A., Timko, S.A., Devera, L., Robinson, A.K., Gonsior, M., Vizenor, A.E., Simpson, A.J., Cooper, W.J., 2013. Photochemistry of excited-state species in natural waters: a role for particulate organic matter. Water Res. 47 (14), 5189e5199. Cuss, C.W., Gueguen, C., 2012. Determination of relative molecular weights of fluorescent components in dissolved organic matter using asymmetrical flow field-flow fractionation and parallel factor analysis. Anal. Chim. Acta 733, 98e102. Del Vecchio, R., Blough, N.V., 2004. On the origin of the optical properties of humic substances. Environ. Sci. Technol. 38 (14), 3885e3891.

Goldman, J.H., Rounds, S.A., Needoba, J.A., 2012. Applications of fluorescence spectroscopy for predicting percent wastewater in an urban stream. Environ. Sci. Technol. 46, 4374e4381. Henderson, R.K., Baker, A., Murphy, K.R., Hamblya, A., Stuetz, R.M., Khan, S.J., 2009. Fluorescence as a potential monitoring tool for recycled water systems: a review. Water Res. 43 (4), 863e881. Hernandez-Ruiz, S., Abrell, L., Wickramasekara, S., Chefetz, B., Chorover, J., 2012. Quantifying PPCP interaction with dissolved organic matter in aqueous solution: combined use of fluorescence quenching and tandem mass spectrometry. Water Res. 46 (4), 943e954. Hur, J., Williams, M.A., Schlautman, M.A., 2006. Evaluating spectroscopic and chromatographic techniques to resolve dissolved organic matter via end member mixing analysis. Chemosphere 63 (3), 387e402. Ishii, S.K.L., Boyer, T.H., 2012. Behavior of reoccurring PARAFAC components in fluorescent dissolved organic matter in natural and engineered systems: a critical review. Environ. Sci. Technol. 46 (4), 2006e2017. , R., Cawley, K., Yamashita, Y., 2014. Advances in the Jaffe Physicochemical Characterization of Dissolved Organic Matter: Impact on Natural and Engineered Systems. ACS Publisher, pp. 27e73. Leenheer, J.A., Croue, J.P., 2003. Characterizing aquatic dissolved organic matter. Environ. Sci. Technol. 37 (1), 18ae26a. Linnik, P.N., 2003. Complexation as the most important factor in the fate and transport of heavy metals in the Dnieper water bodies. Anal. Bioanal. Chem. 376 (3), 405e412. Lu, X.Q., Jaffe, R., 2001. Interaction between Hg(II) and natural dissolved organic matter: a fluorescence spectroscopy based study. Water Res. 35 (7), 1793e1803. Maie, N., Scully, N.M., Pisani, O., Jaffe, R., 2007. Composition of a protein-like fluorophore of dissolved organic matter in coastal wetland and estuarine ecosystems. Water Res. 41 (3), 563e570. Mei, Y., Wu, F.C., Wang, L.Y., Bai, Y.C., Li, W., Liao, H.Q., 2009. Binding characteristics of perylene, phenanthrene and anthracene to different DOM fractions from lake water. J. Environ. Sci. China 21 (4), 414e423. Meng, F., Drews, A., Mehrez, R., Iversen, V., Ernst, M., Yang, F., Jekel, M., Kraume, M., 2009. Occurrence, source, and fate of dissolved organic matter (DOM) in a pilot-scale membrane bioreactor. Environ. Sci. Technol. 43 (23), 8821e8826. Meng, F., Huang, G., Yang, X., Li, Z., Li, J., Cao, J., Wang, Z., Sun, L., 2013. Identifying the sources and fate of anthropogenically impacted dissolved organic matter (DOM) in urbanized rivers. Water Res. 47 (14), 5027e5039. Mignone, R.A., Martin, M.V., Vieyra, F.E.M., Palazzi, V.I., de Mishima, B.L., Martire, D.O., Borsarelli, C.D., 2012. Modulation of optical properties of dissolved humic substances by their molecular complexity. Photochem. Photobiol. 88 (4), 792e800. Miller, M.P., Simone, B.E., McKnight, D.M., Cory, R.M., Williams, M.W., Boyer, E.W., 2010. New light on a dark subject: comment. Aquat. Sci. 72 (3), 269e275. Mladenov, N., Sommaruga, R., Morales-Baquero, R., Laurion, I., Camarero, L., Dieguez, M.C., Camacho, A., Delgado, A., Torres, O., Chen, Z., Felip, M., Reche, I., 2011. Dust inputs and bacteria influence dissolved organic matter in clear alpine lakes. Nat. Commun. 2 (1). Mostofa, K.M.G., Yoshioka, T., Mottaleb, A., Vione, D., 2013. Photobiogeochemistry of Organic Matter: Principles and Practices in Water Environments. Springer, New York. Murphy, K.R., Stedmon, C.A., Waite, T.D., Ruiz, G.M., 2008. Distinguishing between terrestrial and autochthonous organic matter sources in marine environments using fluorescence spectroscopy. Mar. Chem. 108 (1e2), 40e58. Myat, D.T., Mergen, M., Zhao, O., Stewart, M.B., Orbell, J.D., Merle, T., Croue, J.P., Gray, S., 2013. Effect of IX dosing on

w a t e r r e s e a r c h 6 8 ( 2 0 1 5 ) 4 0 4 e4 1 3

polypropylene and PVDF membrane fouling control. Water Res. 47 (11), 3827e3834. Myat, D.T., Stewart, M.B., Mergen, M., Zhao, O., Orbell, J.D., Gray, S., 2014. Experimental and computational investigations of the interactions between model organic compounds and subsequent membrane fouling. Water Res. 48, 108e118. Neale, P.A., Antony, A., Gernjak, W., Leslie, G., Escher, B.I., 2012. Natural versus wastewater derived dissolved organic carbon: implications for the environmental fate of organic micropollutants. Water Res. 45 (14), 4227e4237. Nebbioso, A., Piccolo, A., 2013. Molecular characterization of dissolved organic matter (DOM): a critical review. Anal. Bioanal. Chem. 405 (1), 109e124. Pallem, V.L., Stretz, H.A., Wells, M.J.M., 2009. Evaluating aggregation of gold nanoparticles and humic substances using fluorescence spectroscopy. Environ. Sci. Technol. 43 (19), 7531e7535. Pautler, B.G., Woods, G.C., Dubnick, A., Simpson, A.J., Sharp, M.J., Fitzsimons, S.J., Simpson, M.J., 2012. Molecular characterization of dissolved organic matter in glacial ice: coupling natural abundance H-1 NMR and fluorescence spectroscopy. Environ. Sci. Technol. 46 (7), 3753e3761. Peuravuori, J., Pihlaja, K., 2004. Preliminary study of lake dissolved organic matter in light of nanoscale supramolecular assembly. Environ. Sci. Technol. 38 (22), 5958e5967. Piccolo, A., 2001. The supramolecular structure of humic substances. Soil Sci. 166 (11), 810e832. Plaza, C., Brunetti, G., Senesi, N., Polo, A., 2006. Molecular and quantitative analysis of metal ion binding to humic acids from sewage sludge and sludge-amended soils by fluorescence spectroscopy. Environ. Sci. Technol. 40 (3), 917e923. Richard, C., Coelho, C., Guyot, G., Shaloiko, L., Trubetskoj, O., Trubetskaya, O., 2011. Fluorescence properties of the < 5 kDa molecular size fractions of a soil humic acid. Geoderma 163 (1e2), 24e29. Riedel, T., Biester, H., Dittmar, T., 2012. Molecular fractionation of dissolved organic matter with metal salts. Environ. Sci. Technol. 46 (8), 4419e4426. Riedel, T., Zak, D., Biester, H., Dittmar, T., 2013. Iron traps terrestrially derived dissolved organic matter at redox interfaces. Proc. Natl. Acad. Sci. U S A 110 (25), 10101e10105. Romera-Castillo, C., Chen, M.L., Yamashita, Y., Jaffe, R., 2014. Fluorescence characteristics of size-fractionated dissolved organic matter: implications for a molecular assembly based structure? Water Res. 55, 40e51. Ryan, D.K., Weber, J.H., 1982. Fluorescence quenching titration for determination of complexing capacities and stability constants of fulvic acid. Anal. Chem. 54, 986e990. Stedmon, C.A., Bro, R., 2008. Characterizing dissolved organic matter fluorescence with parallel factor analysis: a tutorial. Limnol. Oceanogr. 6, 572e579. Stedmon, C.A., Markager, S., 2005. Resolving the variability in dissolved organic matter fluorescence in a temperate estuary and its catchment using PARAFAC analysis. Limnol. Oceanogr. 50 (2), 686e697. Stedmon, C.A., Markager, S., Bro, R., 2003. Tracing dissolved organic matter in aquatic environments using a new approach to fluorescence spectroscopy. Mar. Chem. 82 (3e4), 239e254. Stedmon, C.A., Thomas, D.N., Granskog, M., Kaartokallio, H., Papadimitriou, S., Kuosa, H., 2007. Characteristics of dissolved organic matter in Baltic coastal sea ice: allochthonous or autochthonous origins? Environ. Sci. Technol. 41 (21), 7273e7279. Tfaily, M.M., Hamdan, R., Corbett, J.E., Chanton, J.P., Glaser, P.H., Cooper, W.T., 2013. Investigating dissolved organic matter decomposition in northern peatlands using complimentary

413

analytical techniques. Geochim. Cosmochim. Acta 112, 116e129. Tipping, E., 2002. Cation Binding by Humic Substances. Cambridge University Press, New York. Tremblay, L.B., Dittmar, T., Marshall, A.G., Cooper, W.J., Cooper, W.T., 2007. Molecular characterization of dissolved organic matter in a North Brazilian mangrove porewater and mangrove-fringed estuaries by ultrahigh resolution Fourier Transform-Ion Cyclotron Resonance mass spectrometry and excitation/emission spectroscopy. Mar. Chem. 105 (1e2), 15e29. Wang, H.B., Zhang, Y.J., 2014. Mechanisms of interaction between polycyclic aromatic hydrocarbons and dissolved organic matters. J. Environ. Sci. Health Part A Tox. Hazard. Subst. Environ. Eng. 49 (1), 78e84. Williams, C.F., Agassi, M., Letey, J., Farmer, W.J., Nelson, S.D., Ben-Hur, M., 2000. Facilitated transport of napropamide by dissolved organic matter through soil columns. Soil Sci. Soc. Am. J. 64 (2), 590e594. Wu, J., Zhang, H., He, P.J., Shao, L.M., 2011. Insight into the heavy metal binding potential of dissolved organic matter in MSW leachate using EEM quenching combined with PARAFAC analysis. Water Res. 45 (4), 1711e1719. Yamashita, Y., Jaffe, R., 2008. Characterizing the interactions between trace metals and dissolved organic matter using excitation-emission matrix and parallel factor analysis. Environ. Sci. Technol. 42 (19), 7374e7379. Yamashita, Y., Tanoue, E., 2003a. Chemical characterization of protein-like fluorophores in DOM in relation to aromatic amino acids. Mar. Chem. 82 (3e4), 255e271. Yamashita, Y., Tanoue, E., 2003b. Distribution and alteration of amino acids in bulk DOM along a transect from bay to oceanic waters. Mar. Chem. 82 (3e4), 145e160. Yan, M., Wang, D., Korshin, G.V., Benedetti, M.F., 2013. Quantifying metal ions binding onto dissolved organic matter using log-transformed absorbance spectra. Water Res. 47 (7), 2603e2611. Yan, M.Q., Korshin, G.V., 2014. Comparative examination of effects of binding of different metals on chromophores of dissolved organic matter. Environ. Sci. Technol. 48, 3177e3185. Yang, L., Hur, J., 2014. Critical evaluation of spectroscopic indices for organic matter source tracing via end member mixing analysis based on two contrasting sources. Water Res. 59, 80e89. Yang, X., Li, Z., Meng, F., Wang, Z., Sun, L., 2014a. Photochemical alteration of biogenic particles in wastewater effluents. Chin. Sci. Bull. http://dx.doi.org/10.1007/s11434-014-0519-8. Yang, X., Meng, F., Huang, G., Sun, L., Lin, Z., 2014b. Sunlightinduced changes in chromophores and fluorophores of wastewater-derived organic matter in receiving waters e the role of salinity. Water Res. 62, 281e292. Zhang, F., Harir, M., Moritz, F., Zhang, J., Witting, M., Wu, Y., Schmitt-Kopplin, P., Fekete, A., Gaspar, A., Hertkorn, N., 2014. Molecular and structural characterization of dissolved organic matter during and post cyanobacterial bloom in Taihu by combination of NMR spectroscopy and FTICR mass spectrometry. Water Res. 57, 280e294. Zhang, Y.L., Yin, Y., Feng, L.Q., Zhu, G.W., Shi, Z.Q., Liu, X.H., Zhang, Y.Z., 2011. Characterizing chromophoric dissolved organic matter in Lake Tianmuhu and its catchment basin using excitation-emission matrix fluorescence and parallel factor analysis. Water Res. 45 (16), 5110e5122. Ziegelgruber, K.L., Zeng, T., Arnold, W.A., Chin, Y.P., 2013. Sources and composition of sediment pore-water dissolved organic matter in prairie pothole lakes. Limnol. Oceanogr. 58 (3), 1136e1146.