1 23

Ecotoxicology ISSN 0963-9292Volume 21Number 8 Ecotoxicology (2012) 21:2205-2213DOI 10.1007/s10646-012-0975-3

Surface-functionalization effects on uptakeof fluorescent polystyrene nanoparticles bymodel biofilms

Brian A. Nevius, Yung Pin Chen, JohnL. Ferry & Alan W. Decho

1 23

Your article is protected by copyright and

all rights are held exclusively by Springer

Science+Business Media, LLC. This e-offprint

is for personal use only and shall not be self-

archived in electronic repositories. If you

wish to self-archive your work, please use the

accepted author’s version for posting to your

own website or your institution’s repository.

You may further deposit the accepted author’s

version on a funder’s repository at a funder’s

request, provided it is not made publicly

available until 12 months after publication.

Surface-functionalization effects on uptake of fluorescentpolystyrene nanoparticles by model biofilms

Brian A. Nevius • Yung Pin Chen • John L. Ferry •

Alan W. Decho

Accepted: 4 July 2012 / Published online: 18 July 2012

� Springer Science+Business Media, LLC 2012

Abstract A study was conducted to investigate the role

of nanoparticle (NP) surface functionalization/charge on

their uptake by biofilms. Biofilms, bacterial colonies

attached to surfaces via extracellular polymers, are effec-

tive at removing suspended nanomaterials from the aque-

ous phase. However, the mechanisms regulating particle

uptake are unknown. Here, it was shown that the mecha-

nism was strongly dependent on the nanoparticle surface

ionization, and not the core composition of the NP. Uptake

experiments were conducted using laboratory-cultured

biofilms. The biofilms were incubated in the presence of

fluorescent polystyrene NPs with either negatively-charged

surfaces (i.e. functionalized with sulfated (SO4--NP) or

carboxylated (COO--NP) groups) or positively-charged

surfaces (functionalized with primary amines, Amine-P).

Particles with negatively-charged sulfated surfaces associ-

ated most strongly to biofilms across all experimental

conditions. Associations of positively-charged amine

particles with biofilms were greatest at high ionic condi-

tions resembling those of seawater, but were sensitive to

changes in ionic strength. Sorption of COO--NPs was

lowest, relative to other particle types, and was not sensi-

tive to ionic strength. The results of this study support an

emerging precedent that biofilms may be an effective

player in the binding and sequestration of nanoparticles in

aqueous systems.

Keywords Functionalized-NPs � Bacteria � Biofilms �Sorption � EPS

Introduction

Nanomaterials enter the environment through a manifold of

anthropogenic and natural processes (Farre et al. 2009).

Their effects on human and environmental health are lar-

gely unknown, in part because their attenuation mecha-

nisms and pathways are poorly mapped. In the aquatic

compartment nanoparticle (NP) attenuation is recognized

to be a function of several different parameters, including

ionic strength, dissolved salts and organic matter and the

availability of biologically derived structures that can act as

nanomaterial sinks. Bacteria are an essential component of

natural and many engineered systems, and often the first

organisms to interact with introduced compounds. In the

environment, most bacteria occur as biofilms (Fig. 1)

where cells are attached and surrounded by a secreted

matrix of extracellular polymeric substances (EPS) (Mayer

et al. 1999; Flemming and Wingender 2010). These com-

munities can be strongly cohesive and have been shown to

be an effective matrix for the trapping, sorption and

binding of metal ions, organic molecules and even viruses

Electronic supplementary material The online version of thisarticle (doi:10.1007/s10646-012-0975-3) contains supplementarymaterial, which is available to authorized users.

B. A. Nevius � Y. P. Chen � A. W. Decho (&)

Department of Environmental Health Sciences, Arnold School

of Public Health, University of South Carolina, Columbia,

SC 29208, USA

e-mail: awdecho@mailbox.sc.edu

J. L. Ferry

Department of Chemistry and Biochemistry, College of Arts

and Sciences, University of South Carolina, Columbia,

SC 29208, USA

J. L. Ferry � A. W. Decho

Nanocenter at the University of South Carolina,

Columbia, SC 29208, USA

123

Ecotoxicology (2012) 21:2205–2213

DOI 10.1007/s10646-012-0975-3

Author's personal copy

(Mayer et al. 1999; Flemming and Wingender 2010;

Beveridge 1989).

At present little is known concerning interactions of

bacterial biofilms with NPs (Neal 2008). Recent studies

have shown significant accumulations of NPs occurring in

the biofilms of riverine- (Battin et al. 2009) and marine-

(Ferry et al. 2009) mesocosms. Recently, gold (Au) NPs

were added to a marine/estuarine mesocosm ecosystem

containing different biological compartments including

sediments, Spartina, bivalves, shrimp, free-floating plank-

ton and multispecies microbial biofilms. Introduced NPs

rapidly partitioned from the aqueous (introduction) phase

to other phases in the system, with the microbial biofilms

the most significant sink after two weeks of equilibration.

In that study, biofilms accounted for more than 60 % of the

total mass of added NPs. Similar accumulations were found

using riverine mesocosms with titanium dioxide (TiO2)

NPs (Battin et al. 2009). These initial studies suggest an

important role of biofilms for influencing the overall

environmental partitioning of NPs within natural systems.

Accordingly, understanding the factors that influence NP

uptake by these communities is a fundamental step in their

broader environmental fate.

In this paper, it is reported for the first time that NP

surface functional groups, and not core composition,

strongly determine sorption to biofilms. Controlled biofilm

sorption experiments conducted over 36 h time-frames

showed that SO4-- and COO--NP (20 nm), and positively-

charged amine-particles (200 nm) exhibited different

accumulation profiles over a range of ionic conditions.

Materials and methods

Biofilm culture systems

The marine euryhaline bacterium Alteromonas macleodii

(Genbank Accession # 0111138) was used to produce bio-

films. Preliminary experiments showed that this bacterium has

rapid growth and produces repeatable biofilm formation over

a salinity range of 2–50 g l-1. Biofilms were grown on glass

slides within flow-through CDC Biofilm Reactors (CBR)

(BioSurface Technol., Bozeman, MT, USA). CBRs are che-

mostat reactors that allow a gradual but constant exchange of

nutrients in order to facilitate growth of attached biofilm

communities (Donlan et al. 2004). All biofilms were cultured

in 10 % Difco WL Nutrient Broth (Sigma Chem. Co., St.

Louis, MO, USA) supplemented with artificial seawater

(ASW) consisting of the following major ion concentrations

(30 g/l, or 513.4 mM total per liter), expressed as mM per

liter: Cl-, 468 mM; Na?, 401 mM; Mg 2?, 45.68 mM;

SO42?, 24.08 mM; Ca2?, 8.914 mM; K?, 8.545 mM; Br-,

0.711 mM; B(OH)3, 0.394 mM; Sr2? 0.078 mM;

Fl 0.583 mM; HCO3-, 2.01 mM. This was called the ‘High-

Salinity ASW trt’ (hereafter), and was used approximate

conditions representative of the outer estuary. A ‘Low-

Salinity ASW’ trt (hereafter) represented a 910 dilution

(de-ionized H2O) of high-salinity water. Biofilms were grown

(25 �C) with constant stirring (100 rpm; exchange-rate of

0.1 ml/min) on glass microscope slides vertically-oriented on

the inner periphery of the CBR vessels. Under these condi-

tions, biofilms took approximately 6–7 days to allow for

adequate biofilm growth (i.e. to form firm, coatings on glass

slides that were visible to the eye). The biofilms were further

characterized using confocal scanning laser microscopy

(CSLM), fourier-transform infrared spectroscopy (FT-IR),

protein and carbohydrate analyses (see below).

Nanoparticles

Nanoparticles consisted of fluorescent spheres having a con-

stant polystyrene core composition, but varying in the types

of surface functional groups (i.e. carboxyl, sulfate or amine

groups) (Table 1): (a) Carboxylate-nanoparticles (COO--

NPs): Surfaces of COO--NPs (20 ± 4 nm dia.; excit./emiss.,

575/620 nm) are relatively hydrophilic, with each nanoparticle

containing a relatively high density (ca. 3,290 per nanoparticle)

of (pendent) carboxylic acids on their surface. (b) Sulfated-

Fig. 1 Design of sorption experiments. For each salinity condition,

three replicate biofilm treatments (Trt), and Alginate- and blank slide-

controls were used. During a batch sorption incubation, each of the

500 ml vessels (11 cm 9 8 cm) contained 400 ml of water and

contained duplicate, standard glass microscope slides (2.5 9 7.5 cm).

A third ‘no-slide’ control was used to determine vessel sorption. A

fourth biofilm Trt vessel contained a slide that was used for later

confocal imaging. Nine replicate water samples were collected at

each 3 h interval to measure nanoparticle fluorescence. Prior to

sorption experiments, biofilms were grown on clean standard glass

slides for 7 days in 1 l CDC biofilm reactors (CBRs) vessels, which

contained approximately 400 ml of medium with constant exchange

at a flow-rate of 0.1 ml per min., and stirring at 100 rpm

2206 B. A. Nevius et al.

123

Author's personal copy

nanoparticles (SO4--NPs): SO4

--nanoparticles (i.e. 20 ± 4

nm dia., 485/528 nm; ca. 2,540 pendant sites per nanoparticle)

are relatively hydrophobic enabling passive sorption of almost

any protein (pKa \ 2). (c) Amine-particles (Amine-Ps): amine-

modified particles are functionalized with amine groups (i.e.

200 ± 11 nm dia.; 580/605 nm; ca. 429,000 pendant sites per

nanoparticle) to provide hydrophilic particles with positively-

charged amine groups (FluoSpheres�, Invitrogen Inc.). Zeta

potentials (mV) of NPs (and washed A. macloedii bacterial

cells) were measured using a ZetaPALS Zeta-Potential Ana-

lyzer (Brookhaven Instr., NY, USA). These nanoparticles have

been used in many previous studies for their efficient fluores-

cence yield, purity, and uniformity of size (Bloem et al. 1995;

Kolodny et al. 2001; Lai et al. 2007).

Experimental approach

Once the biofilms were developed, they were removed

from the CBRs, gently rinsed, to remove excess growth

medium, in water having the same ionic conditions as those

used for growth (but lacking growth medium), then

immediately transferred into previously-prepared batch

sorption vessels (see below) containing nanoparticles, and

water having similar ionic conditions (high- or low-salinity)

as those used for biofilm growth, but with no added growth

medium. An additional treatment, using 513.2 mM NaCl,

termed ‘High-Salinity NaCl’, was used to examine the

effects of univalent cations present in seawater.

To prepare for each batch-sorption experiment, NPs

were sonicated for approximately 3 min prior to addition to

sorption vessels. Sorption vessels consisted of 500 ml glass

beakers containing 400 ml of water with one of three dif-

ferent ionic conditions (Table 2). Then, an approximately

equal concentration of NPs (ca. 5.72 ± 0.38 9 1016;

mean ± SD) was added to each of seven treatment or

control vessels (Fig. 1), and allowed to stir (in absence of

biofilms) for 24 h prior to the start of each experiment.

To commence a sorption experiment, biofilm-coated

slides (and biofilm-free control slides) were suspended

vertically within the vessels containing one of three dif-

ferent types of NP suspensions, and incubated under con-

stant stirring (100 rpms). A sorption equilibrium time-

frame of 36 h, determined from preliminary experiments,

was used for nanoparticle sampling during sorption

experiments. To measure NP sorption versus time, nine

replicate aliquots (0.2 ml) of vessel water were collected

approx. 2 cm below the water surface at time zero (i.e. start

of experiment), and over a series of time points throughout

the 36 h sorption experiment. Samples from each time

point were measured for NP fluorescence (FLx800

Microplate Reader, BioTek Inst.). Means (±SD; n = 9)

were determined for each 3-h interval and correlated with

NP concentrations. The precision of replicate (n = 9)

fluorescence measurements (\0.5 %) indicated that a rel-

atively homogenous suspension of NPs occurred in solu-

tions throughout experiments. Although NP fluorescence

was homogeneous among replicate water samples within

each time point throughout experiments, the degree to

which NPs were individually-dispersed versus aggregated

(within suspensions) could not be determined. The relative

sorption to biofilms at each time point was calculated from

the relative loss of NPs from the aqueous phase correcting

for controls (e.g. vessel sorption). NP measurements were

based on fluorescence emissions for each type of NP, with

their conversion to concentration determined using stan-

dard curves (Supplemental Figure 1). Three replicate

treatment vessels were used for each time-course mea-

surement; a fourth was used for observations of biofilms

using confocal scanning laser microscopy (CSLM, see

below). Replicate high-NaCl treatments were used to test

Table 1 Properties of surface-functionalized nanoparticles

Nanoparticle

type

Product, cat.

no., lot no.aDiameter

(nm)aZeta

potential (mV)

Mean surface

charge (meq/g)

Charged

groups/

particlea

Fluorescence

absorption

(nm)a

Fluorescence

emission

(nm)a

Amine Fluospheres, F-8763; 29965 W 190 ± 11 21.89 ± 0.61 0.1 429,000 575 ± 15 620 ± 40

COO--NPs Fluospheres, F-8782; 538961 24 ± 3 -40.24 ± 1.74 0.7 3,290 575 ± 15 620 ± 40

SO4--NPs Fluospheres, F-8845; 454238 22 ± 3 -36.01 ± 0.59 0.7 2,540 485 ± 20 528 ± 20

A. macleodii cells GenBank no. # 0111138 – -42.10 ± 0.46 – – – –

Nanoparticles used in sorption experiments were surface-functionalized with carboxyl (COO--), sulfated (SO4--) or amine-functional groups

(Fluospheres, Invitrogen Inc.). Zeta potentials of nanoparticles and A. macleodii bacterial cells, measured in deionised water, were expressed in

millivolts (mV). Values represent mean ± SD; n = 3. Mean surface charge is expressed as milliequivalents of exchangeable lights per g

particles. Charged groups per particle represent the total(s) on the particle surfacea Information from manufacturer

Surface-functionalization effects on uptake of fluorescent 2207

123

Author's personal copy

effects of univalent ions on NP binding at high ionic

conditions. Two different types of controls were used: (1)

slide-only controls, consisting of blank slides (free of

biofilms) and (2) no-slide controls, using only vessels, were

used to differentiate non-biofilm sorption during experi-

ments. A total of 63 separate sorption experiments

were conducted with constant pH (8.0 ± 1) maintained

throughout.

Sample measurements

At the conclusion of 36 h incubations, stirring was ceased

and slides were gently removed from each treatment vessel

and NP sorption to biofilms was quantified. All material

(cells, NPs, EPS) attached to each slide was removed by

carefully scraping slides with a sterile, flat razor. Slides

were washed with 0.5 ml sterile deionized-H2O to remove

any additional material. Samples were measured for fluo-

rescence, and converted to (ca.) NP concentration as pre-

viously described (Supplemental Figure 1). Then, cells and

EPS were placed in microcentrifuge tubes, and separated

by repeated (39) centrifugation (10,0009g; 10 min)

(Decho 1990). Supernatants were used for EPS extraction.

Bulk concentration factors (Cf) of nanoparticles on biofilms

were determined under the pH and ionic conditions for

each experimental treatment. Percent recovery of NPs was

calculated as the portion of total-added NPs recovered from

the aqueous phase ? biofilm ? vessel-sorbed (Table 2).

EPS were extracted in cold (-20 �C) ethanol (EtOH;

70 % final conc.) from centrifuged supernatants (Decho

1990). EPS was dried, weighed and later analyzed. Fourier-

transform infrared spectroscopy (FT-IR) was used to

determine the presence of specific functional groups in EPS

(approx 1 mg) using a Nexus 670 spectrometer (Thermo-

Nicolet Inc.) equipped with attenuated total reflectance

(ATR) and multibounce germanium crystal. Absorbance

data were collected between 680 and 4,000 cm-1 (resolu-

tion 8 cm-1, 64 scans co-added and averaged). Confocal

scanning laser microscopy (CSLM) imaging (Decho et al.

2003) was conducted using an SP5 CSLM System (Leica

Microsystems) equipped with four excitation lasers (405,

488, 594, 633 nm). Cells were stained with SYTO 45

(Invitrogen), and EPS with fluorescein isothiocyanate

(FITC)- or Alexa 594-concanavalin A lectin (Bockelmann

et al. 2002).

Statistical analyses

Statistical analyses were performed on semi-log trans-

formed NP fluorescence data using least square linear

regression. P values less than 0.05 were considered sta-

tistically significant. All statistical analyses were

Table 2 Sorption nanoparticles (NPs) to bacterial biofilms incubated under different ionic conditions

NP surface-functional

group

Ionic conc. Slope y-intercept R2 K (#nanos/cm2) Cfa % Recovery t� loss

of NP (h)b

SO4--NPs L-ASW 0.0070 19.193 0.939*** 5.4 9 1017 2.071 9 106 92 ± 15.3 [36

H-ASW 0.0175 16.037 0.979*** 7.8 9 1017 1.735 9 106 91 ± 15.7 12.1

H-NaCl 0.0140 19.158 0.780*** 1.6 9 1017 3.353 9 105 71 ± 10.7 5.3

COO--NPs L-ASW 0.0011 17.974 0.562* 1.2 9 1016 1.192 9 105 67 ± 1.9 [36

H-ASW 0.0017 18.007 0.920*** 1.1 9 1016 1.018 9 105 63 ± 3.0 [36

H-NaCl 0.0018 18.083 0.770*** 3.7 9 1015 nd 53 ± 1.2 [36

Amine-P L-ASW 0.0036 16.476 0.617** 1.6 9 1015 7.826 9 105 92 ± 20.9 [36

H-ASW 0.0080 16.499 0.870*** 1.0 9 1015 6.212 9 105 89 ± 2.6 27.2

H-NaCl 0.0012 16.454 0.963*** 1.5 9 1014 1.698 9 104 58 ± 2.7 [36

Results showing the sorptive accumulation of surface-functionalized NPs to biofilms of the bacterium A. macloedii, incubated under different

ionic conditions. Values for slope, y-intercept, and coefficient of determination (R2) are derived least squares linear regression using semi-

log-transformed data. Significance by regression: * P \ 0.05; ** P \ 0.01; *** P \ 0.001. Concentration factor (Cf) values represent ratios of

concentrations (lg/g) of bound NPs in biofilms over concentrations of unbound NPs present in aqueous phase at experiment conclusion. Ionic

concentrations are expressed as: H-ASW is High-Salinity ASW (513.4 mM per liter); L-ASW is Low-Salinity ASW (51.34 mM/l); H-NaCl is

High Salinity NaCl (513.4 mM). High- and Low-Salinity artificial seawater salts (ASW) were used for High- and Low-Salinity ASW treatments

and included mono- and divalent-cations. A high-salinity sodium chloride treatment (H-NaCl) was used to include only monovalent ions Na?

and Cl-. K is the binding constant, and is based on biofilm area

ND not determineda The concentration factor (Cf) applies to a steady state, and was the ratio of bound/unbound particles at the conclusion of the experiment.

%Recovery represents a mass balance, expressed as a percentage, of the total nanoparticles recovered from all fractions compared to the added

nanoparticlesb The t� represents the half-life (h) for loss of nanoparticles from overlying water during sorption experiments

2208 B. A. Nevius et al.

123

Author's personal copy

performed using statistical analysis systems (SAS) soft-

ware (SAS Inst., Cary, NC, USA).

Results

The results of sorption incubations showed that for the three

types of NPs, over 80 % of the total NP sorption occurred

within 24 h (Fig. 1). The mean percent recoveries of SO4--,

COO-- and amines-NPs after 36 h incubations were

85 ± 13.9, 61 ± 2.03, and 80 ± 8.73 % (mean ± SD;

n = 3), respectively (Table 2). Coefficients of determination

(R2) indicated high precision among values within a given

experimental time point (Table 2). Controls indicated that less

than 3 % of total added NPs were sorbed to glass vessels

(mean ± SD; 104 ± 7.8 9 1015 NPs). No significant dif-

ferences were found for nanoparticles sorbed to vessels

(P = 0.799) among the different treatments and salinities

used.

SO4--NPs showed highest net accumulations at high

ionic concentrations, irrespective of whether mono- (High-

Salinity NaCl) or divalent- (High-Salinity ASW) ions were

present (Table 2). Accumulations of COO--NPs were less

than 15 % across all salinity conditions, while amine-

particles showed greater than 50 % binding in the presence

of divalent ions (i.e. High-Salinity ASW) and less than

10 % binding in the presence of only monovalent ions

present (i.e. High-Salinity NaCl) (Fig. 1).

The highest sorptive concentration to biofilms was

observed with negatively-charged SO4--NPs. Mean bind-

ing constants (K), indicated that SO4--NPs showed the

highest concentration by biofilms under all salinity condi-

tions. Highest K values occurred at relatively high ionic

conditions (e.g. High-Salinity ASW and NaCl), followed

by Low-Salinity ASW (Table 2). Concentration factors

(Cf), a measure of the relative partitioning of NPs between

biofilms and the bulk medium, were 2.1 9 106 and 1.7 9

106, for SO4--NPs at Low-Salinity and High-Salinity ASW

biofilms, respectively.

Sorptive concentration of COO--NPs to biofilms under

all ionic conditions was less than 15 %), and Cf’s of

COO--NPs to biofilms were an order of magnitude lower

(e.g. 1.2 9 105 at Low-Salinity ASW) than SO4--NPs

(Table 2). Regression analysis results showed a significant

(P \ 0.05) R2 and slope, indicating a positive net

accumulation of COO--NPs to biofilms over the 36 h time-

period. Positively-charged amine-particles exhibited high-

est sorption to biofilms at high ASW concentration.

Approximately 30 % of the recovered particles were found

to be associated with biofilms. Mean Cf’s for amine-

particle sorption to biofilms were calculated between

7.8 9 105 and 6.2 9 105, occurring at Low- and High-

Salinity ASW, respectively.

Fourier-transform infrared (FT-IR) spectroscopic anal-

yses of EPS extracted from A. macleodii biofilms (Fig. 2)

showed a range of peaks corresponding to a C=O stretch

(1,419 cm-1), amides (1,621 cm-1), and amines (2,360 cm-1).

Imaging using confocal scanning laser microscopy

(CSLM) indicated fluorescence signatures corresponding to

SO4-- and COO--NPs.

Discussion

Results showed that for the three types of NPs, a relatively

large portion ([80 %) of sorption occurred over 24 h

(Fig. 1) with lesser net accumulation thereafter. Mean

percent recoveries of SO4--, COO-- and amines-NPs after

36 h were 85 ± 13.9, 61 ± 2.03, and 80 ± 8.73 %

(mean ± SD; n = 3) respectively (Table 2). Coefficients

of determination (R2), calculated from analyses of log-

transformed data using least squares linear regression

analyses (Table 2), showed that replicate sorption values

collected within a given experimental time point had high

precision (i.e. low variability among replicates). Controls

indicated that minimal (i.e. typically \3 % of total added)

sorption to glass (vessels) occurred (mean ± SD;

104 ± 7.8 9 1015 NPs). No significant differences were

found for nanoparticles that were sorbed to vessels

(P = 0.799) among the different treatments and salinities

used.

Salinity conditions affected sorption differently depend-

ing on the surface functionalization of the NP. SO4--NPs

exhibited greatest net accumulations at high ionic con-

centrations irrespective of whether mono- (High-Salinity

NaCl) or divalent- (High-Salinity ASW) ions were present

(Table 2). Accumulations of COO--NPs remained rela-

tively minimal (15 %) over all salinity conditions, while

amine-particles showed greatest binding (50 %) in the

presence of divalent ions (i.e. High-Salinity ASW) but

minimal binding (\ 10 %) with only monovalent ions

present (i.e. High-Salinity NaCl). Thus, the three ionic

conditions resulted in very different sorption patterns for

NPs, with the latter depending on their surface function-

alization (Fig. 1).

Largest sorptive concentration to biofilms occurred with

negatively-charged S-NPs. Examinations of mean binding

constants (K), {i.e. a conditional distribution coefficient

since not constant) which represented numbers of bound

NPs per cm2 for a given set of experiment conditions,

showed that S-NPs were concentrated most efficiently by

biofilms under all salinity conditions. Biofilms typically

exhibit much heterogeneity in density and thickness over

small (e.g. um) spatial scales (Flemming and Wingender

2010). Therefore, it was necessary to calculate K based on

biofilm area. Highest K values occurred at relatively high

Surface-functionalization effects on uptake of fluorescent 2209

123

Author's personal copy

ionic conditions (e.g. High-Salinity ASW and NaCl), fol-

lowed by Low-Salinity ASW (Table 2). However, irre-

spective of ionic conditions, SO4--NPs were more

concentrated onto biofilms by an order of magnitude, when

compared to either COO-- or amine-functionalized NPs

under similar conditions. Concentration factors (Cf) were

calculated to determine the relative partitioning of NPs

between biofilms and the bulk medium. The Cf represents

the ratio of the concentration of bound NPs in biofilm

phase (lg/g) over the concentration in aqueous phase

(lg/g) at the termination of experiments. Cf‘s for SO4--

NPs at Low-Salinity and High-Salinity ASW biofilms were

2.1 9 106 and 1.7 9 106, respectively.

Sorptive concentration of COO--NPs to biofilms under

all ionic conditions was minimal (\ 15 %), and Cf‘s were

an order of magnitude lower (e.g. 1.2 9 105 at Low-

Salinity) than SO4--NPs (Table 2). Results of regression

analyses indicated significant (P \ 0.05) R2 and slopes

indicating a net accumulation of COO--NPs to biofilms

over the 36 h time-period. However, accumulations did not

vary as a function of salinity conditions. Positively-charged

amine-particles exhibited highest sorption to biofilms at

Low-Salinity ASW

Time (h)

0 6 12 18 24 30 36

Aq

ueo

us-

Ph

ase

Nan

op

arti

cles

/ N

ano

par

ticl

es a

t T 0

0.0

0.2

0.4

0.6

0.8

1.0

SO4-

COO-

Amine

High-Salinity ASW

Time (h)

0.0

0.2

0.4

0.6

0.8

1.0

SO4-

COO-

Amine

High-Salinity NaCl

Time (h)

0.0

0.2

0.4

0.6

0.8

1.0

SO4-

COO-

Amine

Aq

ueo

us-

Ph

ase

Nan

op

arti

cles

/ N

ano

par

ticl

es a

t T 0

Aq

ueo

us-

Ph

ase

Nan

op

arti

cles

/ N

ano

par

ticl

es a

t T 0

0 6 12 18 24 30 36

0 6 12 18 24 30 36

Fig. 2 Sorption of NPs with

different surface functional

groups to biofilms. Graphs show

results of 36 h time-course

experiments examining sorption

of NPs having different surface

functional groups to biofilms

under different salinity

conditions: Low-salinity

artificial seawater (ASW);

High-salinity ASW; High-

salinity NaCl. NPs had a

polystyrene core and were

surface-functionalized with

either: SO4--; carboxylic acid-;

or amine-functional groups.

Sorption of NPs to biofilms

could not be measured directly

until the conclusion of a given

time course experiment.

Therefore, each line represents

the decrease of NPs versus time,

measured from aqueous phase.

Relative decrease was used to

infer (i.e. by difference) biofilm

sorption, correcting for controls.

Values represent measurements

of nanoparticle disappearance

from the medium. Each time

point value represents the

mean ± SD bars (n = 9).

Y-axes represent NP

concentrations normalized

against the time zero (T0)

concentration. All experiments

were conducted at a constant pH

(8.0 ± 1)

2210 B. A. Nevius et al.

123

Author's personal copy

high ASW concentration. Approximately 30 % of the

recovered particles were found to be associated with bio-

films. Mean Cf‘s for amine-particle sorption to biofilms

ranged between 7.8 9 105 and 6.2 9 105, occurring at

Low- and High-Salinity ASW, respectively. The relatively

high Cf estimates observed for positively-charged particles

in experiments (621 9 104) were within range of those

determined for positively-charged gold nanorods

(Cf = 1.53 9 104) bound to natural marine biofilms (Ferry

et al. 2009). In that study, marine mesocosms contained

seawater having a range of divalent cations, and a wide

range of other natural environmental components including

sediments, seagrass, bivalves, shrimp and plankton. Mass

balance analyses showed that biofilms alone accounted for

approximately 60 % of added NPs; with Cf’s considered to

be conservative estimates (Ferry et al. 2009).

NPs and the EPS matrix

The extracellular polymeric secretions (EPS) of biofilms

(Decho 1990; Flemming and Wingender 2010) represent a

primary site for the binding of charged ions. Binding of

cations and anions to EPS occurs largely through func-

tional groups located on EPS molecules (Decho 1990).

Analyses of EPS extracted from A. macleodii biofilms by

fourier-transform infrared (FT-IR) spectroscopy (Fig. 2)

indicated a range of functional groups that can form

complexes with NPs via cation- or anion-bridging. These

include carboxyls as evidenced by a C=O stretch

(1,419 cm-1), amides (1,621 cm-1), and amines (2,360 cm-1),

and are in agreement with the literature data (Braissant

et al. 2007). Carboxylic acids, for example, impart EPS

molecules with a net negative-charge at neutral pH and

facilitate binding of metal cations (Ferris et al. 1989; Jang

et al. 1990; Foster et al. 2000; Spagnoli et al. 2005).

Hydrophobic or electrostatic forces can also contribute to

binding between the charged NPs and the biofilm matrix.

The polysaccharides (present in natural EPS), specifically

their -CH groups could exhibit hydrophobic interactions in

the complexation of SO4--NPs. Binding may also occur

through transient bidentate bridging involving divalent

cations (e.g. Ca2? and Mg2?) present in ASW, and occurs

between carboxyl groups on uronic acids and NP functional

groups. However, sorption of SO4--NPs occurred in the

presence of High-Salinity NaCl, therefore, not all binding

could be explained by divalent cations. This suggests that

at least some of the binding under the experimental con-

ditions was simply related to an ionic strength effect. It was

not known, however, which functionalities on EPS directly

contributed to binding with NP surface groups. The FT-IR

study was designed to analyze which functional groups

present on EPS might contribute to biofilm binding,

therefore, analyses were not conducted to determine how

nanoparticle exposure affected EPS functional groups.

Also, our study could not differentiate between NPs that

were sorbed to the biofilm surface versus those which had

migrated into the biofilm (Fig. 3).

Currently, several unanswered questions must be

addressed before the mechanisms of NP binding can be

fully understood. It was not known if charged NPs sorb to

the surface of biofilms or, once bound, can migrate into the

biofilm matrix. It was postulated that brownian motion, in

concert with the transient nature of cation bridges, can

allow NPs to migrate deeper into biofilms over time.

However, brownian motion for an entity as large as a

nanoparticle (e.g. 20 nm dia.) could be quite slow. This

would depend, in part, on the relative spacing between

adjacent polymer molecules. In fact, the larger the particle

the more likely that random motion will average to zero

0.000

0.020

0.040

0.060

0.080

0.100

500 1000 1500 2000 2500 3000 3500 4000

Ab

sorb

ance

Wavenumber (cm )-1

EPS

Amine

(a)

(b)

100 nm

-COO

1091.4

-amide

-amine -OH -COOH

1621.3

2360.2 3379.0 1419.4

Fig. 3 Biofilm EPS. a Confocal scanning laser micrograph (CSLM)

of A. macleodii biofilm showing amine-particles bound to EPS

(bacteria not visible). EPS are stained (green) with concanavalin-A

lectin. Individual amine-NPs (red fluorescence; 200 nm dia.) in the

biofilm are barely visible (in image); however groups of particles

brightly fluoresce. b Infrared spectra of EPS extracted from A.macloedii bacterial biofilms (used in experiments) showing several

absorption peaks representing functional groups on EPS that can

potentially bind surface-functionalized NPs. Attenuated total reflec-

tance—fourier-transform infrared (ATR-FTIR) spectroscopy was used

to determine functional groups. Assignment of wavenumber peaks:

1,091 cm-1 = carbohydrate (C–O) stretching vibrations; 1,419 cm-1 =

carboxylic (C=O) stretch; 1,621 cm-1 = amides (can also be attrib-

uted to more C=O bonds typical of carboxyl groups; and 2,360 cm-1

and 3,379 cm-1 represent amines and O–H, respectively. Wavenum-

ber is equal to 1/wavelength. Spectrum values represent means of 64

co-added scans (Color figure online)

Surface-functionalization effects on uptake of fluorescent 2211

123

Author's personal copy

and the particle will not move. Virus particles (40–100 nm),

however, which are larger than the SO4-- and COO--NPs

(20 nm dia.) used in the study, have been shown to effi-

ciently penetrate biofilms (Briandet et al. 2008). Also,

small particles are known to enter and move through water

channels within the EPS matrix, which have been observed

in some biofilms (Stoodley et al. 1994). Finally, there is a

growing realization that functionalized NPs can become

coated with sorbed organic molecules forming a ‘protein

corona’, which coats the NP (Lundqvist et al. 2008). The

corona surrounding the NP then interacts with surfaces or

matrices during binding. In the present study, efforts were

made to minimize organic molecules in the aqueous phase

during sorption experiments. However, there remained the

possibility that some NP binding may have resulted from

molecules sorbed to the NPs. These are important consid-

erations for understanding binding and subsequent pene-

trations of biofilms by nanoparticles.

Biofilms were imaged using confocal scanning laser

microscopy (CSLM), and fluorescence signatures of NPs

were concurrently detected. Individual 20 nm NPs are

outside the resolving power of CSLM (i.e. light

microscopy). However, fluorescence peaks indicative of

SO4-- and COO--NPs were detectable and showed accu-

mulations over 2-dimensional areas of biofilms (data not

shown). Higher magnification imaging was used to illustrate

the binding of much larger 200 nm dia. amine-particles on

EPS (Fig. 2), although these particles are larger than the

typical size range (1–100 nm) of nanoparticles. It is not

known if the increased size of the amine- particles influ-

enced their binding when compared with the 20 nm

nanoparticles.

The results of this study show that biofilms were

effective at removing suspended nanomaterials from the

aqueous phase. While the mechanisms regulating particle

uptake are not yet known, it was shown that the mecha-

nisms are strongly dependent on the surface nanoparticle

ionization. In both natural- and engineered-systems, NP

surface properties will contribute to their dispersion,

sorptive accumulation, and even bioaccumulation in

organisms. While previous studies of NPs and microbial

systems have focused on core composition (e.g. Ag, TiO2)

(Fabrega et al. 2009; Hetrick et al. 2009; Lellouche et al.

2009; Tong et al. 2010), the present work is the first report

that surface charges on NPs, and not their core properties,

strongly affect sorption and bioconcentration by biofilms.

Biofilms are a rapidly-emerging area of importance in

understanding the bioconcentration of NPs, but one that is

only marginally understood. Biofilms and their sorbed

components, such as NPs, represent a fundamental step for

entry into food webs and humans.

Acknowledgments The authors thank Dr. Cathy Murphy (Univer-

sity of Illinois at Urbana-Champaign) for assistance in zeta-potential

measurements. This work was supported by the University of South

Carolina NanoCenter, and the National Science Foundation (BME-

1032579). The authors declare that they have no conflict of interest.

References

Battin TJ, Kammer FVD, Weilhartner A, Ottofuelling S, Hofmann T

(2009) Nanostructured TiO2: transport behavior and effects on

aquatic microbial communities under environmental conditions.

Environ Sci Technol 43:8098–8104

Beveridge TJ (1989) Role of cellular design in bacterial metal

accumulation and mineralization. Annu Rev Microbiol 43:

147–171

Bloem J, Veninga M, Shepherd J (1995) Fully automatic determina-

tion of soil bacterial numbers, cell volumes, and frequency of

dividing cells by confocal laser scanning microscopy and image

analysis. Appl Environ Microbiol 61:926–936

Bockelmann U, Manz W, Neu TR, Szewzyk U (2002) Investigation

of lotic microbial aggregates by a combined technique of

fluorescent in situ hybridization and lectin-binding-analysis.

J Microbiol Meth 49:75–87

Braissant O, Decho AW, Dupraz C, Glunk C, Przekop KM, Visscher

PT (2007) Exopolymeric substances of sulfate-reducing bacteria:

interactions with calcium at alkaline pH and implication for

formation of carbonate minerals. Geobiology 5:401–411

Briandet R, Lacroix-Gueu P, Renault M, Lecart S, Meylheuc T,

Bidnenko E, Steenkeste K, Bellon-Fontaine MN, Fontaine-

Aupart MP (2008) Fluorescence correlation spectroscopy to

study diffusion and reaction of bacteriophages inside biofilms.

Appl Environ Microb 74:2135–2143

Decho AW (1990) Microbial exopolymer secretions in ocean

environments—their role(s) in food webs and marine processes.

Oceanogr Mar Biol Ann Rev 28:73–153

Decho AW, Kawaguchi T, Allison MA, Louchard EM, Reid RP,

Stephens FC, Voss KJ, Wheatcroft RA, Taylor BB (2003) Sediment

properties influencing upwelling spectral reflectance signatures: the

‘‘biofilm gel effect’’. Limnol Oceanogr 48:431–443

Donlan RM, Piede JA, Heyes CD, Sanii L, Murga R, Edmonds P, El-

Sayed I, El-Sayed MA (2004) Model system for growing and

quantifying Streptococcus pneumoniae biofilms in situ and in

real time. Appl Environ Microb 70:4980–4988

Fabrega J, Renshaw JC, Lead JR (2009) Interactions of silver

nanoparticles with Pseudomonas putida biofilms. Environ Sci

Technol 43:9004–9009

Farre M, Gajda-Schrantz K, Kantiani L, Barcelo D (2009) Ecotoxicity

and analysis of nanomaterials in the aquatic environment. Anal

Bioanal Chem 398:81–95

Ferris FG, Schultze S, Witten TC, Fyfe WS, Beveridge TJ (1989)

Metal interactions with microbial biofilms in acidic and neutral

pH environments. Appl Environ Microb 55:1249–1257

Ferry JL, Craig P, Hexel C, Sisco P, Frey R, Pennington PL, Fulton

MH, Scott IG, Decho AW, Kashiwada S, Murphy CJ, Shaw TJ

(2009) Transfer of gold nanoparticles from the water column to

the estuarine food web. Nat Nanotechnol 4:441–444

Flemming HC, Wingender J (2010) The biofilm matrix. Nat Rev

Microbiol 8:623–633

Foster LJR, Moy YP, Rogers PL (2000) Metal binding capabilities of

Rhizobium etli and its extracellular polymeric substances.

Biotechnol Lett 22:1757–1760

2212 B. A. Nevius et al.

123

Author's personal copy

Hetrick EM, Shin JH, Paul HS, Schoenfisch MH (2009) Anti-biofilm

efficacy of nitric oxide-releasing silica nanoparticles. Biomate-

rials 30:2782–2789

Jang LK, Brand W, Resong M, Mainieri W, Geesey GG (1990)

Feasibility of using alginate to absorb dissolved copper from

aqueous-media. Environ Prog 9:269–274

Kolodny LA, Willard DM, Carillo LL, Nelson MW, Van Orden A

(2001) Spatially correlated fluorescence/AFM of individual

nanosized particles and biomolecules. Anal Chem 73:1959–1966

Lai SK, O’Hanlon DE, Harold S, Man ST, Wang YY, Cone R, Hanes

J (2007) Rapid transport of large polymeric nanoparticles in

fresh undiluted human mucus. Proc Natl Acad Sci USA

104:1482–1487

Lellouche J, Kahana E, Elias S, Gedanken A, Banin E (2009)

Antibiofilm activity of nanosized magnesium fluoride. Bioma-

terials 30:5969–5978

Lundqvist M, Stigler J, Elia G, Lynch I, Cedervall T, Dawson KA

(2008) Nanoparticle size and surface properties determine the

protein corona with possible implications for biological impacts.

Proc Natl Acad Sci USA 105:14265–14270

Mayer C, Moritz R, Kirschner C, Borchard W, Maibaum R,

Wingender J, Flemming HC (1999) The role of intermolecular

interactions: studies on model systems for bacterial biofilms. Int

J Biol Macromol 26:3–16

Neal AL (2008) What can be inferred from bacterium-nanoparticle

interactions about the potential consequences of environmental

exposure to NPs? Ecotoxicology 17:362–371

Spagnoli C, Korniakov A, Ulman A, Balazs EA, Lyubchenko YL,

Cowman MK (2005) Hyaluronan conformations on surfaces: effect

of surface charge and hydrophobicity. Carbohydr Res 340:929–941

Stoodley P, Debeer D, Lewandowski Z (1994) Liquid flow in biofilm

systems. Appl Environ Microb 60:2711–2716

Tong MP, Ding JL, Shen Y, Zhu PT (2010) Influence of biofilm on

the transport of fullerene (C-60) NPs in porous media. Water Res

44:1094–1103

Surface-functionalization effects on uptake of fluorescent 2213

123

Author's personal copy