lable at ScienceDirect

Environmental Pollution 241 (2018) 701e709

Contents lists avai

Environmental Pollution

journal homepage: www.elsevier .com/locate/envpol

Interactions of polymeric drug carriers with DDT reduce theircombined cytotoxicity*

Xuejiao Zhang a, 1, Lei Lei a, e, 1, Haiyan Zhang b, Siyu Zhang a, d, Weiwei Xing c, Jin Wang c,Haibo Li b, d, Qing Zhao a, *, Baoshan Xing d

a Key Laboratory of Pollution Ecology and Environmental Engineering, Institute of Applied Ecology, Chinese Academy of Sciences, Shenyang, Chinab School of Resources & Civil Engineering, Northeastern University, Shenyang, 110004, Chinac Liaoning Beifang Environmental Technology Co., LTD., Shenyang, 110161, Chinad Stockbridge School of Agriculture, University of Massachusetts, Amherst, MA, 01003, USAe University of Chinese Academy of Sciences, Chinese Academy of Sciences, Beijing, 100049, China

a r t i c l e i n f o

Article history:Received 4 April 2018Received in revised form30 May 2018Accepted 4 June 2018Available online 7 June 2018

Keywords:Polymeric nanomicellesOrganic pollutantsPartitioning behaviorCombined cytotoxicityBiointerfacial behavior

* This paper has been recommended for acceptanc* Corresponding author.

E-mail address: zhaoqing@iae.ac.cn (Q. Zhao).1 These authors contributed equally to the manusc

https://doi.org/10.1016/j.envpol.2018.06.0100269-7491/© 2018 Elsevier Ltd. All rights reserved.

a b s t r a c t

Attention has been paid to the environmental distribution and fate of nanomedicines. However, theireffects on the toxicity of environmental pollutants are lack of knowledge. In this study, the negativelycharged poly (ethylene glycol)-b-poly (L-lactide-co-glycolide) (mPEG-PLA) and positively chargedpolyethyleneimine-palmitate (PEI-PA) nanomicelles were synthesized and served as model drug carriersto study the interaction and combined toxicity with dichlorodiphenyltrichloroethane (DDT). DDT exertedlimited effect on the biointerfacial behavior of mPEG-PLA nanomicelles, whereas it significantly miti-gated the attachment of PEI-PA nanomicelles on the model cell membrane as monitored by quartz crystalmicrobalance with dissipation (QCM-D). The cytotoxicity of DDT towards NIH 3T3 cells was greatlydecreased by either co-treatment or pre-treatment with the nanomicelles according to the results ofreal-time cell analysis (RTCA). The cell viability of NIH 3T3 exposed to DDT was increased up to 90% bythe co-treatment with mPEG-PLA nanomicelles. Three possible reasons were proposed: (1) decreasedamount of free DDT in the cell culture medium due to the partitioning of DDT into nanomicelles; (2)mitigated cellular uptake of nanomicelle-DDT complexes due to the complex agglomeration or elec-trostatic repulsion between complexes and cell membrane; (3) detoxification effect in the lysosome uponendocytosis of nanomicelle-DDT complexes.

© 2018 Elsevier Ltd. All rights reserved.

1. Introduction

With the rapid development of nanotechnology, the demandand production of manufactured nanoparticles (MNPs) are under-going unprecedented growth, consequently resulting in growingdischarges into the environment (Sun et al., 2016; Sun et al., 2014;Yin et al., 2017). Thus, significant attention has been paid to theenvironmental fate and ecological impact of MNPs, such asfullerene, metal oxide, carbon nanotubes, and graphene, etc(Colvin, 2003; Mueller and Nowack, 2008; Petersen et al., 2011;Tsang et al., 2017). Nowadays, the application of MNPs is extended

e by Charles Wong.

ript.

to nanotherapeutics and at least 250 nanomedicines have beenapproved or under approval for clinical applications (Etheridgeet al., 2013). The global market of nanomedicines is estimated tobe USD 178e528 billion in 2019 (Evers, 2012). Baun and Hansenfirst claimed the necessity to explore the environmental exposureand risk of nanomedicines in 2008 (Baun and Hansen, 2008).Manufacturing processes and disposal of containers are regularpathways that lead to the discharges of nanomedicines into theenvironment. In addition, some of the administered nanomedicinescan be excreted from the patients via renal or hepatic clearance,producing partial-, fully-, and non-metabolized nanomedicines(Baun and Hansen, 2008; Yegin et al., 2017). Afterwards, thereleased nanomedicines can reach the sewer system in hospitals,pharmaceutical factories, and residential areas, and eventually gothrough thewastewater treatment plants (WWTPs) (Sanchez et al.,2011). However, almost one-third of the world's population are notprovided with sewage systems (Mahapatra et al., 2013) and the

X. Zhang et al. / Environmental Pollution 241 (2018) 701e709702

traditional WWTPs cannot completely eliminate nanomaterials(Chen et al., 2011; Limbach et al., 2008). As a result, nanomedicinescan enter into rivers, lakes, and oceans after being discharged withthe effluents from WWTPs. Therefore, attention should be paid tothe environmental fate and risks of nanomedicines (Chen et al.,2015; Yegin et al., 2017; Zhang and Akbulut, 2011).

Polymeric nanomicelles have been recognized as one of themost commonly used drug carriers for nanomedicines because oftheir superior properties, including solubilizing hydrophobic mol-ecules, high payload capacity, increased bioavailability, excellentstability in the aqueous solution, and prolonged blood circulation(Cheng et al., 2012; Peer et al., 2007; Petros and DeSimone, 2010).However, nanomicelles have received inadequate concernregarding their environmental risks since they are normally highlybiocompatible and nontoxic. However, once they enter into thenatural environment, nanomicelles will inevitably encounter theco-existing pollutants and interact with them (Qu et al., 2013).Given the excellent ability of nanomicelles to solubilize hydro-phobic drug molecules, it is reasonable to expect that they willinteract also with hydrophobic pollutants. According to previousstudies, the interactions of pollutants with MNPs could changetheir bioavailability and toxicity (Bruton et al., 2015; Caballero-Díazet al., 2014; Cui et al., 2016; Deng et al., 2017; Zheng et al., 2012). Forinstance, synergistic toxicity of nano-TiO2 with DDT or bisphenol Awas induced owing to the efficient interactions (Shi et al., 2010;Zheng et al., 2012), in which nano-TiO2 served as carriers toconcentrate the organic pollutants and to increase their bioavail-ability via “Trojan horse effect” (Deng et al., 2017). The combinationof non-toxic concentration of Ni2þ and single walled carbonnanotubes (SWCNTs) exhibited elevated toxicity towards macro-phages, which was attributed to the inhibition of exocytosis-dominated P2X7 receptor by Ni2þ, leading to the higher accumu-lation of SWCNTs (Cui et al., 2016). The complexation of nano-particles and pollutant molecules could also alleviate theirindividual toxicity by decreasing the bioavailability (Glomstadet al., 2016; Li et al., 2015). Therefore, the effects of interactions oforganic pollutants with nanomicelles on their combined bioavail-ability and toxicity are worthwhile to understand.

Dichlorodiphenyltrichloroethane (DDT) was widely used as anorganochlorine pesticide for several decades all over the world.Considering its adverse impacts on the ecological environment andhuman health, it has been forbidden since 1980s (Smith et al.,2012). However, due to its long-term persistence, lipophilicity,and low degradability, DDT is still existing in the environment inmany areas nowadays (Van Dyk et al., 2010). For example, it wasdetected in the surface soil from Xianfeng County in China, with amean level of 40 ng/g and a maximum concentration of 430 ng/g(Ma et al., 2016). Thus, DDT was chosen as a model hydrophobicpollutant in this study. To facilitate study of the fate and toxicityexperiments, the concentrations used for the laboratory experi-ments were greater than typical environmentally relevant levels.

Herein, we studied the interaction of DDT with polymericnanomicelles and the effect of interaction on their biointerfacialbehavior and combined cytotoxicity. Two model drug carriers,negatively charged methoxy poly (ethylene glycol)-poly (lactide)(mPEG-PLA) and positively charged polyethyleneimine-palmitate(PEI-PA) nanomicelles were fabricated and applied to interactwith DDT. Supported lipid bilayers (SLBs) composed of zwitterionic1,2-dioleoyl-sn-glyero-3-phosphocholine (DOPC) were used asmodel cell membranes to investigate the biointerfacial behavior ofDDT, polymeric nanomicelles, and their complexes by quartz crystalmicrobalance with dissipation (QCM-D). The combined cytotoxicitytowards NIH 3T3 (mouse fibroblast cells) was determined by alabel-free real-time cell analysis (RTCA) technique.

2. Materials and methods

2.1. Materials

DDT was purchased from Beijing Bailingwei Chemical Technol-ogy Co. Ltd. mPEG5000-PLA10000 was obtained from ShanghaiYare Biotechnology Co. Ltd. N-succinimidyl palmitate was boughtfrom Saen chemistry Technology Co. Ltd. N,N-dimethylformamide(DMF) was purchased from Beijing Hanlongda DevelopmentTechnology Co. Ltd. Polyethyleneimine (PEI, branched,Mw¼ 1800 g/mol) was purchased from Beijing Yinuokai Technol-ogy Co. Ltd. 1,2-Dilauroyl-sn-glycero-3-phosphocholine (DLPC) wasobtained from Shanghai Aladdin biochemical Technology Co. Ltd.N-(2-hydroxyethyl)-piperazine-N0-(2-ethanesulfonic acid) (HEPES)was purchased from Shanghai Yuexv Technology Co. Ltd. All theother reagents were of chromatographic grade unless otherwisestated. For cell culture experiments, NIH 3T3 cells were cultured inDulbecco's modified Eagle's Medium: nutrient mixture F-12(DMEM/F12) supplemented with 10% fetal bovine serum (FBS) and100 U/mL penicillin/streptomycin under standard culturing con-ditions (humidified atmosphere of 5% CO2 at 37 �C).

2.2. Synthesis of polyethyleneimine-palmitate (PEI-PA)

15.9mg PEI was dissolved in 4.8mL DMF in a 50mL flask. N-succinimidyl palmitate (15mg/mL in DMF) was added dropwise tothe flask and the reaction mixture was stirred overnight. Then thecrude product was transferred to a dialysis tube (molecular weightcut-off 2000 g/mol) and extensively dialyzed against deionizedwater, followed by lyophilization to obtain pure PEI-PA. Thestructures of mPEG-PLA and PEI-PA were characterized by 1H NMR(400MHz). 1H NMR (mPEG-PLA, CDCl3, 25 �C): d¼ 1.5 (CH3-CH-),3.51e3.63 (-CH2-O-), 5.02e5.14 (-CO-CH-O-) (Fig. 1a); 1H NMR (PEI-PA, CD3OD, 25 �C): d¼ 0.9 (CH3-CH2-), 1.2e1.3 (CH3-CH2-), 2.3e2.4(-NH-CO-CH2-), 2.4e3.2 (-NH-CH2- or -N-CH2-) (Fig. 1b).

2.3. Preparation and characterization of polymeric nanomicelles

Both polymeric nanomicelles were prepared by nano-precipitation method (Zhang et al., 2013). In brief, 2mL PEI-PA(2mg/mL) acetone solution was added dropwise to 10mL Milli-Qwater under vigorous stirring. After 24 h, acetone was evaporatedand the suspension was filtered through a cellulose membrane(pore size¼ 0.22 mm) to obtain the PEI-PA nanomicelles, whichwere stored at 4 �C. The mPEG-PLA nanomicelles were preparedfollowing the similar procedure except that water was used as thesolvent and acetone was the precipitant. The size distribution andzeta potential of the polymeric nanomicelles were measured bydynamic light scattering (DLS, Zetasizer Nano ZS90, Malvern). Themorphology of polymeric nanomicelles was visualized by trans-mission electron microscope (TEM, JEOL JEM 2100F, Japan).

2.4. Interaction of DDT with polymeric nanomicelles

0.2mg PEI-PA or mPEG-PLA nanomicelles were added intodifferent concentrations of DDT aqueous solution (2e140 mg/mL,containing 0.1% methanol) in a 300mL glass bottle. The glass bot-tles were sealed with Teflon-lined screw caps and continuouslyshaken at 150 rpm under 25 �C for 5 days. After centrifugation for30min to precipitate the nanomicelles, 40mL of supernatant wasremoved and extracted with n-hexane to obtain DDT. The con-centration of extracted DDT from solution without nanomicelleswas used as control. All the experiments were performed in trip-licate and the results represented the mean value.

The concentration of DDT in the extract was determined by gas

Fig. 1. 1H NMR spectra of mPEG-PLA in CDCl3 (a) and PEI-PA in CD3OD (b); Size distributions of mPEG-PLA (c) and PEI-PA (e) nanomicelles in Milli-Q water by DLS measurements;TEM images of mPEG-PLA (d) and PEI-PA (f) nanomicelles.

X. Zhang et al. / Environmental Pollution 241 (2018) 701e709 703

chromatograph (GC, Agilent 6890N) equipped with a63Ni electroncapture detector (Liu et al., 2014; Shi et al., 2010). Separation wasconducted on a HP-5 quartz capillary column (30m� 0. 32mm� 0.25 mm). High purity nitrogen was used as the carrier gas at a linearvelocity of 1mL/min. The injected volume was 1 mL and the tem-perature of injector and detector was 250 and 300 �C, respectively.The column temperature was initially 100 �C, increased at a rate of15 �C/min to 180 �C and held for 0.5min, followed by rising at aspeed of 20 �C/min to 250 �C and holding for 1min, and finally,raised at a rate of 5 �C/min to 295 �C and held for 3min. Eachsample was analyzed in triplicate. The calibration curve of DDT wascalculated to be linear (R2¼ 0.999) in the concentration range from0.1 to 800 ng/mL.

2.5. Preparation of DLPC vesicles

The DLPC vesicles were prepared according to the previouslyreported technique (Cho et al., 2010; Liu and Chen, 2015; Liu andChen, 2016; Zhang and Yang, 2011). Briefly, 0.2mL of 25 g/L DLPCsolution in chloroform was dried in an Erlenmeyer flask with agentle nitrogen stream to form a DLPC thin film on the bottom ofthe flask, followed by desiccation under vacuum for at least 4 h toremove the residual chloroform. The DLPC film was hydrated by

adding 5mL HEPES buffer (10mM HEPES, 150mM NaCl, pH 7.4)and the suspension was magnetically stirred for 30min. Then thesuspensionwas extruded using amini-extruder (Avanti Polar LipidsInc.) through a polycarbonate membrane (50 nm, Whatman) backand forth for at least 15 times to produce unilamellar vesicles withthe average hydrodynamic diameter of 70.03± 0.49 nm(PDI¼ 0.091). The stock suspension was sealed in glass vials withnitrogen and stored at 4 �C, which was used within 4 days afterpreparation.

2.6. Effect of partitioning on biointerfacial behavior of DDT andnanomicelles

The interactions of nanomicelles, DDT, and their complexes withthe model cell membrane were monitored by QCM-D (E1, Q-Sense,Biolin Scientific, Sweden). A 5MHz AT-cut quartz crystal sensorwith a silica-coated surface (QSX 303, Q-Sense) was mounted ineach module (Yi and Chen, 2013). The measurements were per-formed at several harmonics (z¼ 3, 5, 7, 9, 11, and 13) and all thedata presented were recorded at the third harmonic (Frost et al.,2016). Before the experiments, SiO2-coated sensors were treatedin UV-ozone for 30min, then rinsed with deionized water, anddried in air. All measurements were conducted under constant flow

X. Zhang et al. / Environmental Pollution 241 (2018) 701e709704

rate (100 mL/min) at 37 �C. A SLB was generated on a SiO2-coatedcrystal sensor according to the published method (Keller andKasemo, 1998). First, the sensor was rinsed with HEPES buffer un-til a stable baseline (frequency signal drifts of less than 0.2 Hzwithin 10min) was obtained. Then 5mL of 0.05 g/L DLPC vesiclesuspension was introduced into the flow module and a DLPC SLBwas formed due to the deposition and self-assembly of vesicles.After rinsing the sensor with HEPES buffer to achieve a stablebaseline, the nanomicelles (100 mg/mL), DDT (1 mg/mL), or theircomplex mixtures (the weight ratio of nanomicelles to DDT was100: 1) prepared in the same buffer solutionwere introduced to thechamber. At the end of the experiments, the chamber was rinsedwith HEPES buffer for 10min to remove the un-adsorbed com-pound. All the complex mixtures were prepared in accordance tothe method described in Section 2.4.

2.7. Cytotoxicity

The effect of interaction with nanomicelles on DDT's toxicitytowards mammalian cell line, NIH 3T3, was evaluated by RTCA. Theelectrode in E-plate 16 (ACEA biosciences, USA) can detect thechanges of impedance during cell attachment, which was reflectedby the cell index (CI). NIH 3T3 cells were seeded into the E-plate16 at a density of 5000 cells/well and incubated for 24 h. After-wards, the culture medium was replaced with fresh medium con-taining polymeric nanomicelles, DDT, or their complex mixtures atpredetermined concentrations (polymeric nanomicelles: 5 mg/mL;DDT: 0.4, 4, 10, 20, 30, and 40 mg/mL). The complex mixtures wereprepared following the procedure described in Section 2.4. Cellswithout any treatment were used as control. For statistical analysis,the CI values were normalized at the point of treatment andexpressed as NCI. The percentage of response represented the NCIratio of the testing group to the control group (Zhang et al., 2017).

The effect of nanomicelles on the cytotoxicity of DDT wasfurther investigated by sequential treatment with nanomicellesand DDT. Cells were seeded at a density of 5000 cells/well andincubated for 24 h. Then the nontoxic concentration of nano-micelles (5 mg/mL) was introduced. After 6.5 h, the culture mediumcontaining nanomicelles was removed and the cells were rinsedwith PBS for three times to get rid of the nanomicelles attached onthe cell surface. Then the cells were supplemented with fresh cul-ture medium containing various concentrations of DDT (10, 20, 30,and 40 mg/mL) and further incubated for 24 h. The cell viability wasmonitored by RTCA.

3. Results and discussion

3.1. Fabrication and characterization of mPEG-PLA and PEI-PAnanomicelles

Both mPEG-PLA and PEI-PA nanomicelles were constructed bythe self-assembly of amphiphilic polymers in the process ofnanoprecipitation. By adjusting the initial polymer concentrationand volume ratio of water to organic solvent, monodispersedspherical mPEG-PLA and PEI-PA nanomicelles were obtained(Fig. 1d, f). They exhibited similar hydrodynamic diameters of87.2± 5.3 and 85.6± 1.8 nm (Fig. 1c, e) and inverse zeta potentialsof �11.7± 1.4 and 49.9± 1.4mV, respectively.

3.2. Interaction of DDT with mPEG-PLA and PEI-PA nanomicelles

The experiments were performed using a batch equilibrationtechnique at 25 �C. Linear partitioning isotherms were obtained forboth mPEG-PLA and PEI-PA nanomicelles (Fig. 2a), suggestingpartitioning rather than physical adsorption was the main process

for the interaction of DDT with nanomicelles (CHIOU et al., 1979).Previous studies also confirmed a partitioning mechanism of hy-drophobic drug loading into mPEG-PLA nanomicelles (Huh et al.,2005). The partitioning affinity of DDT to mPEG-PLA nanomicelleswas much higher than that to PEI-PA nanomicelles, which might beoriginated from the different micelle structures (Fig. 2b). Theamphiphilic mPEG-PLA copolymer formed a uniform core-shellnanomicelle that contained a hydrophilic PEG shell surrounding asingle hydrophobic PLA core. In contrast, the self-assembly ofbranched PEI-PA could generate multiple hydrophobic domains inthe nanomicelle structure. It is reasonable that the partitioning ofDDT to the homogeneous hydrophobic core of mPEG-PLA nano-micelles was much easier than that to the multiple hydrophobicdomains of PEI-PA nanomicelles. Therefore, the partitioning of DDTin mPEG-PLA nanomicelles was more preferable compared to thatin PEI-PA nanomicelles.

3.3. Effects of partitioning on the biointerfacial behavior ofnanomicelles and DDT

Biological membranes are important barriers and interfacesbetween many living organisms and pollutants. In order to un-derstand the effect of partitioning on the bioavailability of nano-micelles and DDT, the affinity of nanomicelles, DDT, and theircomplexes to the cell membrane was evaluated by QCM-D. SLBscomposed of DOPC were used as model cell membranes. HEPESbuffer was introduced into the QCM-D chamber until stable fre-quency and dissipation baselines were obtained (Fig. 3a). WhenDOPC vesicles in HEPES buffer were allowed to flow through thechamber, decreased Df and increased DD were initially establisheddue to the deposition of vesicles on the sensor. Then Df waspartially recovered within 2min, which was ascribed to the releaseof encapsulated solution from the ruptured vesicles, resulting in theself-assembly of SLBs on the sensor surface. The overall Df and DDinduced by the formation of SLB in this study were �25± 0.50 Hzand 0.28� 10�6 ± 0.1� 10�6, which was in accordance with theprevious report (Yi and Chen, 2013).

After the SLBs were formed on the surface of SiO2-coated sensor,DDT solution (1.0 mg/mL in HEPES) was subsequently introduced(Fig. 3b). The equilibrium was not reached during the entireexperimental period (more than 10 h) with a temporary maximumfrequency decrease of 159 Hz. The massive DDT deposition on themodel cell membrane was probably due to the diffusion of DDTmolecules into the hydrophobic inner layer of SLBs. The frequencyand dissipation were almost unchanged upon rinsing with HEPES,demonstrating a strong and irreversible partitioning of DDT in thecell membrane.

When the suspension of mPEG-PLA nanomicelles was intro-duced, Df and DDwere not observed (Fig. 3c), illustrating there wasno specific interaction between the negatively charged mPEG-PLAnanomicelles and the lipid bilayers. PEG was considered as astealth corona for nanoparticles to avoid nonspecific protein in-teractions and cellular uptake by the mononuclear phagocyte sys-tem and to prolong the blood circulations (Fontana et al., 2001; Grefet al., 2000; Sch€ottler et al., 2016). Moreover, PEGylated nano-particles showed the lowest uptake into the skin cells compared tothose with other surface coatings (Verma and Stellacci, 2010).Therefore, the PEG shell could protect mPEG-PLA nanomicellesfrom adhering on the SLBs.

The complex mixture of mPEG-PLA nanomicelles and DDT(mPEG-PLA þ DDT) induced a frequency decrease of 4.40 Hz in thefirst 4 min after flowing through the chamber, whereas the fre-quency subsequently rose almost back to the level before intro-ducing the complex mixture (Fig. 3d and Fig. S1). The negativecharges of mPEG-PLA nanomicelles were partially shielded by the

Fig. 2. (a) Partitioning isotherms of DDT in mPEG-PLA (,) and PEI-PA (△); (b) Schematic partitioning mechanisms of DDT in two nanomicelles.

Fig. 3. Frequency (blue) and dissipation (red) shifts during the formation of DOPC SLBs (a) and the deposition of DDT (b), mPEG-PLA nanomicelles (c), mPEG-PLA þ DDT (d), PEI-PAnanomicelles (e), and PEI-PA þ DDT (f) on the model cell membrane monitored by QCM-D. (For interpretation of the references to colour in this figure legend, the reader is referredto the Web version of this article.)

X. Zhang et al. / Environmental Pollution 241 (2018) 701e709 705

X. Zhang et al. / Environmental Pollution 241 (2018) 701e709706

DDT partitioning with the zeta potential increased from �11.7to �5.51mV, which allowed slight interaction between the com-plexes and SLBs due to their alleviated electrostatic repulsion.However, the interaction was too weak to maintain the complexeson SLBs, so that the deposition was reversible over continuousrinse. Therefore, the partitioning interaction reduced DDT'sbioavailability, but did not significantly affect the bioavailability ofmPEG-PLA nanomicelles.

In the case of introducing the positively charged PEI-PA nano-micelles, significant mass variations were detected on the SLBcovered QCM-D sensor with Df¼�54.4 Hz and DD¼ 9.46� 10�6

(Fig. 3e), which was not affected by rinsing with HEPES buffer,depicting an irreversible deposition of PEI-PA nanomicelles on theSLBs due to electrostatic interactions. It has been reported thatcationic PEI could cause significant membrane damage ofmammalian cells, which was, on the one hand, due to the elec-trostatic interactions, on the other hand, resulted from the in-teractions between PEI and membrane proteins and phospholipids(Fischer et al., 2003; Kunath et al., 2003). Since the zeta potential ofPEI-PA nanomicelles was decreased from 49.9 to 26.0 mV bycomplexing DDT, the interaction between the complex mixture ofPEI-PA nanomicelles and DDT (PEI-PA þ DDT) and the SLBs wasmitigated as indicated by the less frequency shift of PEI-PA þ DDT(Df¼�30.9 Hz) (Fig. 3f) compared to that of free DDT(Df¼�159 Hz) and PEI-PA nanomicelles (�54.4 Hz) (Fig. 3b,e).Therefore, the partitioning of DDT in the PEI-PA nanomicelles wascapable of alleviating their individual interactions with the cellmembrane, which would then decrease their bioavailability.

3.4. Cytotoxicity of nanomicelles, DDT, and their complexes

Co-existence of nanoparticles and environmental pollutants ispossible to impose unexpected toxic effects on human health. In

Fig. 4. The proliferation profiles of NIH 3T3 cells incubated with mPEG-PLA þ DDT (a) andDDT, and their complex mixtures; (d) Cell viability after 24 h treatment with PEI-PA nanomnanomicelles was 5 mg/mL and DDT's concentration was from 0.4 to 40 mg/mL in both com

the previous study, nontoxic concentration of nano-TiO2 increasedthe cytotoxicity of trace amount of DDT or bisphenol A due to theimproved cellular uptake (Shi et al., 2010; Zheng et al., 2012). Otherreports claimed that nano-TiO2 could reduce the internalizationand cytotoxicity of benzo [a]pyrene to lung carcinoma cells (A549)by competitively binding with the aromatic hydrocarbon receptors(Dhasmana et al., 2014). Therefore, it is essential to assess thecombined cytotoxicity of DDTand nanomicelles upon simultaneousor sequential exposure tomammalian cells. RTCAwas performed toevaluate the cytotoxicity of DDT, nanomicelles, and their complexesto NIH 3T3 cells. DDT induced a dose-dependent inhibition of thecell proliferationwhen the concentrationwas increased from 0.4 to40 mg/mL reflected by the decreased NCI (Fig. S2a). The inhibition ofcell proliferation occurred at a dose of 10 mg/mL or above (Fig. S2b).Then we screened the cytotoxicity of nanomicelles at the concen-trations of 5, 40, and 100 mg/mL via RTCA (Figs. S2c and d) and chose5 mg/mL, which was nontoxic to NIH 3T3 cells for both nano-micelles, to conduct the following toxicity evaluations.

When the cells were exposed to different concentrations ofmPEG-PLA þ DDT and PEI-PA þ DDT complex mixtures, the cellviability (NCItreated group/NCIcontrol) was comparable to control(NCIcontrol¼ 100%) after 24 h (Fig. 4), indicating there was nocytotoxicity induced by mPEG-PLA þ DDT and PEI-PA þ DDT. Ac-cording to the partitioning equation, the amount of free DDT in thecomplex mixtures was calculated to be 0.003 and 5.17 mg/mL formPEG-PLA þ DDT and PEI-PA þ DDT, respectively (Table S1), whichwere both lower than the toxic concentration of DDT (above 10 mg/mL). The hydrodynamic diameter of mPEG-PLA and PEI-PA nano-micelles was increased from 87.2± 5.3 and 85.6± 1.8 to 252± 35.9and 164± 6.4 nm, respectively. The reason was attributed to thatsome DDT molecules partitioned on the nanomicelles’ surface,which lowered the surface charge, enhanced the surface hydro-phobicity, and therefore, promoted the micelle agglomeration. It

PEI-PA þ DDT (c); (b) Cell viability after 24 h treatment with mPEG-PLA nanomicelles,icelles, DDT, and their complex mixtures. The concentration of mPEG-PLA and PEI-PAplex mixtures.

Fig. 5. The proliferation profiles of NIH 3T3 cells first exposed to 5 mg/mL of mPEG-PLA nanomicelles (a) and PEI-PA nanomicelles (b) for 6.5 h and then incubated in DDT-containingculture medium (10, 20, 30, and 40 mg/mL) for another 24 h; Cell viability after sequential exposure to mPEG-PLA nanomicelles (b) or PEI-PA nanomicelles (d) and various con-centrations of DDT.

X. Zhang et al. / Environmental Pollution 241 (2018) 701e709 707

has been reported that cells preferred to internalize smallernanoparticles with PEG shell (Hu et al., 2007). Thus, agglomerationof mPEG-PLA nanomicelles induced by DDT partitioning mightmitigate their cellular uptake. Likewise, the decreased surfacecharge of PEI-PA nanomicelles due to DDT partitioning wouldsuppress their affinity to the cell membrane, which was discussedin detail in Section 3.3. Thus, the partitioning of DDT in bothnanomicelles could decrease its available concentration in theculture medium and its bioavailability, which subsequently low-ered the cytotoxicity. Furthermore, given endocytosis was a generalpathway for the cell internalization of nanomicelles, thenanomicelle-DDT complexes would be located in the lysosome,where macromolecules, carbohydrates, dead organelles, and otherunwanted materials could be digested due to the presence ofdozens of hydrolytic enzymes (de Duve, 2005). In contrast, freeDDT, as a small molecule, could diffuse directly into the cells(Hjelmborg et al., 2008). Accordingly, the partitioning of DDT in thenanomicelles could alter its cellular uptake process, and therebydecrease its cytotoxicity.

We also evaluated the effect of nanomicelles on DDT's cyto-toxicity by sequentially treating the cells with nanomicelles andDDT (Fig. 5). The viability of DDT treated cells was obviouslyenhanced upon pre-treatment with mPEG-PLA or PEI-PA nano-micelles. Similar observation was reported in a previous study,which demonstrated that the toxicity of polychlorinated biphenyls(PCBs) was dramatically attenuated by pre-treatment with gra-phene oxide (Liu et al., 2016). The underlying mechanism wasneeded to further explore in the future.

4. Conclusion

The interaction of DDT with both negatively charged mPEG-PLAand positively charged PEI-PA nanomicelles was a partitioning-induced process instead of physical adsorption that was suitablefor most nanomaterials. The DDT partitioning could induceagglomeration, shield the surface charge, and enhance the surfacehydrophobicity of both nanomicelles. The partitioning of DDTexerted negligible influence on the biointerfacial behavior ofmPEG-PLA nanomicelles, but significantly alleviated the interactionbetween PEI-PA nanomicelles and the model cell membrane. Thecytotoxicity of nanomicelle-DDT complexes was much lower thanthat of free DDT. There are three aspects contributing to the low-ered toxicity: (1) The available concentration of free DDT in theculture medium after partitioning was much lower than the toxicDDT concentration; (2) The bioavailability of DDT was decreaseddue to partitioning in the polymeric nanomicelle; (3) The endocy-tosis of nanomicelle-DDTcomplexes facilitated them locating in thelysosome, where the cells were protected from outside substances’attack by enzymatic digestion. Also, the tolerant capacity of NIH3T3 cells towards DDT was enhanced by the pre-treatment withnanomicelles, but the underlying mechanism was not clear andneeded to explore in the future. The results in this study will poseimportant implications to the interaction of polymeric drug carrierswith hydrophobic organic pollutants and their combined toxicity inthe aquatic environment.

X. Zhang et al. / Environmental Pollution 241 (2018) 701e709708

Acknowledgement

This work was supported by the National Natural ScienceFoundation of China (No. 41603120, 41703107, and 21671115), theNational Key Research and Development Program of China (No.2016YFD0800300), Youth Innovation Promotion Association CASawarded to S.-Y.Z. (2017e2020), Key Laboratory of Industrial Ecol-ogy and Environmental Engineering (MOE), and Hundreds TalentsProgram of Chinese Academy of Sciences awarded to X.-J.Z.(2015e2020) and Q.Z. (2014e2019).

Appendix A. Supplementary data

Supplementary data related to this article can be found athttps://doi.org/10.1016/j.envpol.2018.06.010.

References

Baun, A., Hansen, S.F., 2008. Environmental challenges for nanomedicine. Nano-medicine 3, 605e608.

Bruton, T.A., Pycke, B.F.G., Halden, R.U., 2015. Effect of nanoscale zero-valent irontreatment on biological reductive dechlorination: a review of current under-standing and Research needs. Crit. Rev. Environ. Sci. Technol. 45, 1148e1175.

Caballero-Díaz, E., Guzm�an-Ruiz, R., Malag�on, M.M., Simonet, B.M., Valc�arcel, M.,2014. Effects of the interaction of single-walled carbon nanotubes with 4-nonylphenol on their in vitro toxicity. J. Hazard Mater. 275, 107e115.

Chen, G., Liu, X., Su, C., 2011. Transport and retention of TiO2 rutile nanoparticles insaturated porous media under low-ionic-strength conditions: measurementsand mechanisms. Langmuir 27, 5393e5402.

Chen, I.C., Zhang, M., Teipel, B., de Araujo, I.S., Yegin, Y., Akbulut, M., 2015. Transportof polymeric nanoparticulate drug delivery systems in the proximity of silicaand sand. Environ. Sci. Technol. 49, 3575e3583.

Cheng, Z., Al Zaki, A., Hui, J.Z., Muzykantov, V.R., Tsourkas, A., 2012. Multifunctionalnanoparticles: cost versus benefit of adding targeting and imaging capabilities.Science 338, 903e910.

CHIOU, C.T., PETERS, L.J., FREED, V.H., 1979. A physical concept of soil-water equi-libria for nonionic organic compounds. Science 206, 831e832.

Cho, N.-J., Frank, C.W., Kasemo, B., Hook, F., 2010. Quartz crystal microbalance withdissipation monitoring of supported lipid bilayers on various substrates. Nat.Protoc. 5, 1096e1106.

Colvin, V.L., 2003. The potential environmental impact of engineered nano-materials. Nat. Biotechnol. 21, 1166.

Cui, X., Wan, B., Guo, L.-H., Yang, Y., Ren, X., 2016. Insight into the mechanisms ofcombined toxicity of single-walled carbon nanotubes and nickel ions in mac-rophages: role of P2X7 receptor. Environ. Sci. Technol. 50, 12473e12483.

de Duve, C., 2005. The lysosome turns fifty. Nat. Cell Biol. 7, 847.Deng, R., Lin, D., Zhu, L., Majumdar, S., White, J.C., Gardea-Torresdey, J.L., et al., 2017.

Nanoparticle interactions with co-existing contaminants: joint toxicity, bio-accumulation and risk. Nanotoxicology 11, 591e612.

Dhasmana, A., Sajid Jamal, Q.M., Mir, S.S., Bhatt, M.L.B., Rahman, Q., Gupta, R., et al.,2014. Titanium dioxide nanoparticles as guardian against environmentalcarcinogen benzo[alpha]pyrene. PLoS One 9, e107068.

Etheridge, M.L., Campbell, S.A., Erdman, A.G., Haynes, C.L., Wolf, S.M.,McCullough, J., 2013. The big picture on nanomedicine: the state of investiga-tional and approved nanomedicine products. Nanomed. Nanotechnol. Biol.Med. 9, 1e14.

Evers, P., 2012. Global Markets for Stem Cells. BCC Research.Fischer, D., Li, Y., Ahlemeyer, B., Krieglstein, J., Kissel, T., 2003. In vitro cytotoxicity

testing of polycations: influence of polymer structure on cell viability and he-molysis. Biomaterials 24, 1121e1131.

Fontana, G., Licciardi, M., Mansueto, S., Schillaci, D., Giammona, G., 2001. Amoxi-cillin-loaded polyethylcyanoacrylate nanoparticles: influence of PEG coating onthe particle size, drug release rate and phagocytic uptake. Biomaterials 22,2857e2865.

Frost, R., Svedhem, S., Langhammer, C., Kasemo, B., 2016. Graphene Oxide and LipidMembranes: Size-dependent interactions. Langmuir 32, 2708e2717.

Glomstad, B., Altin, D., Sørensen, L., Liu, J., Jenssen, B.M., Booth, A.M., 2016. Carbonnanotube properties influence adsorption of phenanthrene and subsequentbioavailability and toxicity to pseudokirchneriella subcapitata. Environ. Sci.Technol. 50, 2660e2668.

Gref, R., Lück, M., Quellec, P., Marchand, M., Dellacherie, E., Harnisch, S., et al., 2000.‘Stealth’ corona-core nanoparticles surface modified by polyethylene glycol(PEG): influences of the corona (PEG chain length and surface density) and ofthe core composition on phagocytic uptake and plasma protein adsorption.Colloids Surf. B Biointerfaces 18, 301e313.

Hjelmborg, P.S., Andreassen, T.K., Bonefeld-Jørgensen, E.C., 2008. Cellular uptake oflipoproteins and persistent organic compoundsdan update and new data.Environ. Res. 108, 192e198.

Hu, Y., Xie, J., Tong, Y.W., Wang, C.-H., 2007. Effect of PEG conformation and particle

size on the cellular uptake efficiency of nanoparticles with the HepG2 cells.J. Contr. Release 118, 7e17.

Huh, K.M., Lee, S.C., Cho, Y.W., Lee, J., Jeong, J.H., Park, K., 2005. Hydrotropic polymermicelle system for delivery of paclitaxel. J. Contr. Release 101, 59e68.

Keller, C.A., Kasemo, B., 1998. Surface specific kinetics of lipid vesicle adsorptionmeasured with a quartz crystal microbalance. Biophys. J. 75, 1397e1402.

Kunath, K., von Harpe, A., Fischer, D., Petersen, H., Bickel, U., Voigt, K., et al., 2003.Low-molecular-weight polyethylenimine as a non-viral vector for DNA de-livery: comparison of physicochemical properties, transfection efficiency andin vivo distribution with high-molecular-weight polyethylenimine. J. Contr.Release 89, 113e125.

Li, L., Fern�andez-Cruz, M.L., Connolly, M., Conde, E., Fern�andez, M., Schuster, M.,et al., 2015. The potentiation effect makes the difference: non-toxic concen-trations of ZnO nanoparticles enhance Cu nanoparticle toxicity in vitro. Sci.Total Environ. 505, 253e260.

Limbach, L.K., Bereiter, R., Müller, E., Krebs, R., G€alli, R., Stark, W.J., 2008. Removal ofoxide nanoparticles in a model wastewater treatment plant: influence ofagglomeration and surfactants on clearing efficiency. Environ. Sci. Technol. 42,5828e5833.

Liu, F., Tian, H., He, J., 2014. Adsorptive performance and catalytic activity ofsuperparamagnetic Fe3O4@nSiO2@mSiO2 coreeshell microspheres towardsDDT. J. Colloid Interface Sci. 419, 68e72.

Liu, X., Chen, K.L., 2015. Interactions of graphene oxide with model cell membranes:probing nanoparticle attachment and lipid bilayer disruption. Langmuir 31,12076e12086.

Liu, X., Chen, K.L., 2016. Aggregation and interactions of chemical mechanical pla-narization nanoparticles with model biological membranes: role of phosphateadsorption. Environ. Sci. J. Integr. Environ. Res. Nano 3, 146e156.

Liu, Y., Wang, X., Wang, J., Nie, Y., Du, H., Dai, H., et al., 2016. Graphene oxide at-tenuates the cytotoxicity and mutagenicity of PCB 52 via activation of genuineautophagy. Environ. Sci. Technol. 50, 3154e3164.

Ma, J., Pan, L.-B., Yang, X.-Y., Liu, X.-L., Tao, S.-Y., Zhao, L., et al., 2016. DDT, DDD, andDDE in soil of Xiangfen County, China: residues, sources, spatial distribution,and health risks. Chemosphere 163, 578e583.

Mahapatra, I., Clark, J., Dobson, P.J., Owen, R., Lead, J.R., 2013. Potential environ-mental implications of nano-enabled medical applications: critical review.Environ. Sci. J. Integr. Environ. Res. Process. Impacts 15, 123e144.

Mueller, N.C., Nowack, B., 2008. Exposure modeling of engineered nanoparticles inthe environment. Environ. Sci. Technol. 42, 4447e4453.

Peer, D., Karp, J.M., Hong, S., Farokhzad, O.C., Margalit, R., Langer, R., 2007. Nano-carriers as an emerging platform for cancer therapy. Nat. Nanotechnol. 2,751e760.

Petersen, E.J., Zhang, L., Mattison, N.T., O'Carroll, D.M., Whelton, A.J., Uddin, N., et al.,2011. Potential release pathways, environmental fate, and ecological risks ofcarbon nanotubes. Environ. Sci. Technol. 45, 9837e9856.

Petros, R.A., DeSimone, J.M., 2010. Strategies in the design of nanoparticles fortherapeutic applications. Nat. Rev. Drug Discov. 9, 615e627.

Qu, X., Alvarez, P.J.J., Li, Q., 2013. Applications of nanotechnology in water andwastewater treatment. Water Res. 47, 3931e3946.

Sanchez, W., Sremski, W., Piccini, B., Palluel, O., Maillot-Mar�echal, E., Betoulle, S.,et al., 2011. Adverse effects in wild fish living downstream from pharmaceuticalmanufacture discharges. Environ. Int. 37, 1342e1348.

Sch€ottler, S., Becker, G., Winzen, S., Steinbach, T., Mohr, K., Landfester, K., et al., 2016.Protein adsorption is required for stealth effect of poly(ethylene glycol)- andpoly(phosphoester)-coated nanocarriers. Nat. Nanotechnol. 11, 372e377.

Shi, Y., Zhang, J.-H., Jiang, M., Zhu, L.-H., Tan, H.-Q., Lu, B., 2010. Synergistic geno-toxicity caused by low concentration of titanium dioxide nanoparticles andp,p0-DDT in human hepatocytes. Environ. Mol. Mutagen. 51, 192e204.

Smith, E., Weber, J., Rofe, A., Gancarz, D., Naidu, R., Juhasz, A.L., 2012. Assessment ofDDT relative bioavailability and bioaccessibility in historically contaminatedsoils using an in vivo mouse model and fed and unfed batch in vitro assays.Environ. Sci. Technol. 46, 2928e2934.

Sun, T.Y., Bornh€oft, N.A., Hungerbühler, K., Nowack, B., 2016. Dynamic probabilisticmodeling of environmental emissions of engineered nanomaterials. Environ.Sci. Technol. 50, 4701e4711.

Sun, T.Y., Gottschalk, F., Hungerbühler, K., Nowack, B., 2014. Comprehensive prob-abilistic modelling of environmental emissions of engineered nanomaterials.Environ. Pollut. 185, 69e76.

Tsang, M.P., Kikuchi-Uehara, E., Sonnemann, G.W., Aymonier, C., Hirao, M., 2017.Evaluating nanotechnology opportunities and risks through integration of life-cycle and risk assessment. Nat. Nanotechnol. 12, 734.

Van Dyk, J.C., Bouwman, H., Barnhoorn, I.E.J., Bornman, M.S., 2010. DDT contami-nation from indoor residual spraying for malaria control. Sci. Total Environ. 408,2745e2752.

Verma, A., Stellacci, F., 2010. Effect of surface properties on nanoparticleecell in-teractions. Small 6, 12e21.

Yegin, Y., Yegin, C., Oh, J.K., Orr, A., Zhang, M., Nagabandi, N., et al., 2017. Ecotoxiceffects of paclitaxel-loaded nanotherapeutics on freshwater algae, Raphidocelissubcapitata and Chlamydomonas reinhardtii. Environ. Sci. J. Integr. Environ. Res.Nano 4, 1077e1085.

Yi, P., Chen, K.L., 2013. Interaction of multiwalled carbon nanotubes with supportedlipid bilayers and vesicles as model biological membranes. Environ. Sci. Technol.47, 5711e5719.

Yin, Y., Tan, Z., Hu, L., Yu, S., Liu, J., Jiang, G., 2017. Isotope tracers to study theenvironmental fate and bioaccumulation of metal-containing engineered

X. Zhang et al. / Environmental Pollution 241 (2018) 701e709 709

nanoparticles: techniques and applications. Chem. Rev. 117, 4462e4487.Zhang, M., Akbulut, M., 2011. Adsorption, desorption, and removal of polymeric

nanomedicine on and from cellulose surfaces: effect of size. Langmuir 27,12550e12559.

Zhang, M., Soto-Rodriguez, J., Chen, I.C., Akbulut, M., 2013. Adsorption and removaldynamics of polymeric micellar nanocarriers loaded with a therapeutic agenton silica surfaces. Soft Matter 9, 10155e10164.

Zhang, X., Yang, S., 2011. Nonspecific adsorption of charged quantum dots on

supported zwitterionic lipid bilayers: real-time monitoring by quartz crystalmicrobalance with dissipation. Langmuir 27, 2528e2535.

Zhang, X., Zhang, Z., Zhang, S., Li, D., Ma, W., Ma, C., et al., 2017. Size effect on thecytotoxicity of layered black phosphorus and underlying mechanisms. Small 13,1701210.

Zheng, D., Wang, N., Wang, X., Tang, Y., Zhu, L., Huang, Z., et al., 2012. Effects of theinteraction of TiO2 nanoparticles with bisphenol A on their physicochemicalproperties and in vitro toxicity. J. Hazard Mater. 199, 426e432.