Journal of Biomaterials and Nanobiotechnology, 2012, 3, 508-518
http://dx.doi.org/10.4236/jbnb.2012.324052 Published Online October 2012 (http://www.SciRP.org/journal/jbnb)
Spions Increase Biofilm Formation by Pseudomonas
aeruginosa
Carl Haney, John J. Rowe, Jayne B. Robinson
Department of Biology, University of Dayton, Dayton, USA.
Received July 18th, 2012; revised August 25th, 2012; accepted September 14th, 2012
ABSTRACT
Limited research has suggested iron oxide nanoparticles (FeNP) have an inhibitory effect against several different gen-
era of bacteria: Staphylococcus, Bacillus and Pseudomonas spp. In this study we looked at the effect of three different
sets of Fe3O4 nanoparticles (FeNPs) on the development of Pseudomonas aeruginosa PAO1 biofilms. Two of the tested
NPs were SPIONs (Superparamagnetic Iron Oxide Nanoparticles). Exposure of cells to the SPIONs at concentrations
up to 200 µg/ml resulted in an increase in biofilm biomass by 16 h under static conditions and a corresponding increase
in cell density in the bulk liquid. In contrast, these biofilms had decreased levels of extracellular DNA (eDNA). Fe(II)
levels in the supernatants of biofilms formed in the presence of FeNPs exceeded 100 µM compared with 20 µM in con-
trol media without cells. Spent cell supernatants had little effect on Fe(II) levels. Cells also had an effect on the aggre-
gation behavior of these nanoparticles. SPIONs incubated with cells exhibited a decrease in the number and size of
FeNP aggregates visible using light microscopy. SPIONs resuspended in fresh media or spent culture supernatants
formed large aggregates visible in the light microscope upon exposure to a supermagnet; and could be pelleted mag-
netically in microtitre plate wells. In contrast, SPION FeNPs incubated with cells were unaffected by exposure to the
supermagnet and could not be pelleted. The results of this study indicate a need to reconsider the effects of FeNPs on
bacterial growth and biofilm formation and the effect the bacterial cells may have on the use and recovery of SPIONs.
Keywords: Spion; Iron Oxide Nanoparticles; Biofilms; Pseudomonas Aeruginosa; Magnetism; Nanotechnology
1. Introduction
Pseudomonas aeruginosa is a ubiquitous gram-negative,
rod shaped bacterium best known as an opportunistic
pathogen. According to the CDC, P. aeruginosa accounts
for 10% of all hospital acquired infections and a fatality
rate of nearly 50% in patients with cancer, cystic fibrosis,
and burns. Survival and virulence of this bacterium is
attributable to the biofilms it produces [1]. Biofilms are
sessile colonies which form on both biotic and abiotic
surfaces. These biofilms can withstand phagocytic envi-
ronments and exhibit 100 - 1000 fold greater resistance
to antibiotics [2]. Biofilms develop and are held together
by secreted macromolecules called the extra-polymeric
substances (EPS) which form the biofilm matrix. This
EPS consists of extracellular DNA (eDNA), polysaccha-
rides, and proteins.
Iron is a key nutrient required by most living organ-
isms. Due to the insolubility of the ferric (Fe(III)) ion and
because the soluble ferrous (Fe(II)) ion is readily oxi-
dized at neutral pH, this abundant element is a limited
resource for which organisms must compete [3]. Simi-
larly, iron’s efficiency as an electron donor and acceptor
makes it susceptible to generating toxic oxygen radicals
[4].
Iron is a powerful regulator of gene expression in P.
aeruginosa. Iron regulation is mediated by both the Fer-
ric uptake regulator (Fur) repressor protein as well as the
density-dependent cell-to-cell signaling quorum sensing
pathways [5,6]. Musk et al. [7] demonstrated that iron
concentrations between 1 - 100 µM are required for P.
aeruginosa to form and develop robust biofilms. Outside
this range, P. aeruginosa can only exist in the planktonic
environment. Under low iron conditions twitching motil-
ity increases thereby limiting the ability of the bacteria to
transition to a sessile subpopulation and therefore the
development of a mature bacterial mushroom shaped
colony [8-10]. Yang et al. [11] showed a decrease in the
levels of eDNA in the EPS with increasing iron concen-
tration. Berlutti et al. [12] found iron limitation of 1 µM
induced twitching motility while concentrations between
10 and 100 µM stimulated cell aggregation and biofilm
development. Additionally, sufficient levels of iron also
play a role in the development of the bacterial mushroom
caps indicative of mature biofilms [13]. Further, both

Spions Increase Biofilm Formation by Pseudomonas aeruginosa 509
ferric and ferrous ions have been shown to be potent
cationic cross-linkers and can increase the viscosity of
the biofilm matrix [14].
Iron oxide nanoparticles (NPs) are commercially im-
portant types of these particles [15]. In recent years, there
has been an increase in interest and research directed to-
wards superparamagnetic iron oxide nanoparticles (SPI-
ONS). SPIONS have shown promise in areas such as
cellular therapy and tissue repair. Additionally, recent
research with Fe3O4 SPIONs has shown them to be toxic
to several different important pathogenic bacteria: Sta-
phylococcus epidermidis, Klebsiella pneumonia, and Strep-
tococcus pyogenes and P. aeruginosa [16,17]. These stu-
dies focused on planktonic bacteria but not on the bio-
films which they produce. In this study, the effects of
SPIONs on 16 h static P. aeruginosa biofilms are de-
scribed and evaluated.
2. Materials and Methods
2.1. Bacterial Strain, Media and Culture
Conditions
All experiments were performed with the wild-type strain
of P. aeruginosa PAO1. Cells were grown in Minimal
Salts Medium (MSM) as described by Moulton and
Montie [18]: 7 g of K2HPO4, 3 g of KH2PO4, 1 g of
(NH4)2SO4, 0.05 g of MgSO4*7H2O and 2.5 mg of
FeCl3*6H2O per liter of Milli-Q purified water. The me-
dium was supplemented with glucose as the sole carbon
and energy source at 0.4% (w/v) which will be referred
hereafter as MSG. For iron deplete media, FeCl3 was not
added and the medium was supplemented with 250 µM
2,2 dipyridyl, an iron chelator.
2.2. Iron Nanoparticle Preparation
Three different FeNPs were tested in this study. They
will be referred to as: Brown, US Research, or Novacentrix.
The Brown Nanoparticles were a generous gift from
Brown University (T. J. Webster). They were synthe-
sized by the Center for Biomedical Engineering at Brown
University using a co-precipitate method as described by
University of California-Davis [19]. These nanoparticles
were delivered in a 70% ethanol stock solution of ap-
proximately 8 mg/ml. This stock solution was diluted
into fresh 70% ethanol to a new working concentration
stock of 2 mg/ml which was verified via dry weight after
desiccation. To prepare NPs for addition to cells aliquots,
the stock solution was centrifuged in an Eppendorf Cen-
trifuge 5415D at 15,000 g for 20 min to obtain a pellet.
The supernatant was removed and replaced with the
same volume of sterile Milli-Q water followed by 20 min
of sonication using a Branson 1510R-MT (70 W, 42 kHz)
at room temperature to disperse the FeNPs followed by
brief vortexing to ensure a homogenous solution.
The US Research nanoparticles were purchased as a
dry powder from US Research Nanomaterial Inc. (Hous-
ton, TX). A stock solution was made by suspending the
powder in Milli-Q water at a concentration of 2 mg/ml.
Prior to adding the nanoparticles to cultures, the stock
solution was vigorously sonicated and vortexed as de-
scribed above.
The Novacentrix nanoparticle were obtained from
Wright-Patterson Air Force Base as a 1mg/ml stock solu-
tion in water. Wright-Patterson AFB purchased the
nanoparticles from NovaCentrix (Austin, TX). Prior to
use, the nanoparticles were sonicated and vortexed as
described above for the other nanoparticles.
2.3. Nanoparticle Characterization
In order to help identify the basis for the differences in
toxicity among the NPs we used transmission electron
microscopy (TEM) and dynamic light scattering (DLS)
to obtain the mean size, size distribution, and zeta poten-
tials. The Novacentrix FeNPs were the largest with a
mean of 34.7 nm while the nanoparticles from Brown
University and US Research Nanomaterial Inc. were
similar at 8.79 nm and 11.18 nm respectively (Figure 1).
Additionally, all three sets of nanoparticles were gener-
ally spherical in shape and had the tendency to aggregate.
The nanoparticles from US Research Nanomaterial ex-
hibiting the greatest aggregation as shown by the large
size distribution and zeta potential centered around zero
(Figure 1).
An analysis obtained from Novacentrix revealed that
these FeNPs contained heavy metals such as silver, cop-
per, chromium, manganese and nickel at levels that
would be toxic to bacteria (Table 1). This is not surpris-
ing given the method of manufacture of the Novacentrix
NPs simply involved the grinding together of steel bars.
In contrast, the FeNPs from Brown University and US
Research Nanomaterial Inc. were synthesized using pure
reagents. The presence of heavy metals in the Novacen-
trix NPs likely accounts for their observed toxicity and
thus these FeNPs were excluded from subsequent ex-
periments.
2.4. Biofilm Formation and Assay
For this study we modified the biofilm assay protocol
described by O’Toole and Kolter [20] as follows. Cells
were grown in MSG medium overnight at 37˚C with
shaking. The overnight culture was diluted in fresh MSG
medium to an OD590 nm of 0.15 (4 × 108 cells per milliliter).
This diluted culture was aliquoted into 15ml conical tubes
to which FeNPs or iron salts (FeCl3*6H2O or
FeSO4*7H2O) were added to the stated concentrations.
Controls did not receive nanoparticles. Cell suspensions

Spions Increase Biofilm Formation by Pseudomonas aeruginosa
510

Spions Increase Biofilm Formation by Pseudomonas aeruginosa 511
Table 1. Contaminants present in Novacentrix nanoparticles. Table provided from Novacentrix showing the contaminants
present in their nanoparticles. Values provided are the concentration in weight percentage with an uncertainty of ±50%.
Ag 0.075 Al <0.001 As 0.006 Au <0.001B 0.001Ba <0.001Be <0.001 Bi <0.001
C N.D. Ca 0.006 Cd <0.001 Ce <0.001Co <0.001Cr 0.022Cs N.D. Cu 0.240
Dy <0.001 Er <0.001 Eu <0.001 Fe MajorGa <0.001 Gd <0.001 Ge <0.001 Hf <0.001
Hg <0.001 Ho <0.001 In <0.001 Ir <0.001K <0.001La <0.001Li <0.001 Lu <0.001
Mg <0.001 Mn 0.44 Mo 0.007 Na 0.001 Nb <0.001 Nd <0.001Ni 0.047 Os <0.001
P 0.011 Pb <0.001 Pd <0.001 Pr <0.001Pt <0.001Rb N.D.Re <0.001 Rh <0.001
Ru <0.001 S N.D. Sb <0.001 Sc <0.001 Se <0.001Si 0.10 Sm <0.001 Sn 0.006
Sr <0.001 Ta <0.001 Tb <0.001 Te <0.001 Th <0.001Ti <0.001Tl <0.001 Tm <0.001
U <0.001 V <0.001 W <0.001 Y <0.001Yb <0.001Zn 0.007Zr <0.001
were aliquoted at 100 µl per well into Falcon BD PVC
96 well microtiter plate, covered with parafilm, and in-
cubated statically in a 37˚C incubator for 16 h.
Following the 16 h incubation, 50 µl of 0.15% (w/v)
crystal violet solution was added to each well and plates
were incubated at room temperature for 30 min. After
incubation the microtiter plates were rinsed five times in
Milli-Q water and allowed to dry completely overnight.
To each dried well, 200 µl of 97% ethanol was added and
incubated at room temperature for 10 min to re-solubilize
the crystal violet stained biomass. The solubilized biomass
(150 µl) was transferred to sterile NUNC 96 well round
bottom plates and absorbance readings were taken using
a Wallac Victor2 1420 Multilabel Counter at 590 nm.
2.5. Measurement of Extracellular DNA in
Biofilms
Biofilms were grown as described above. Growth me-
dium from the 16 h biofilms was removed and replaced
with 125 µl of 1X TE Buffer. The biofilm was resus-
pended in the TE buffer and cells removed by centrifuga-
tion in a 1.5 ml microcentrifuge tube at 10,000 g for two
min. 100 µl of the supernatant was transferred to a black
96 well microtiter plate. A PicoGreen solution was pre-
pared as per the manufacturer’s (Molecular Probes Inc.,
Eugene, OR) instructions and 20 µl added to each well.
Samples were incubated for 3 min in the dark at room
temperature and fluorescence measured in a Wallac Vic-
tor2 1420 Multilabel Counter using 485/535 nm excita-
tion and emission wavelengths.
2.6. Quantification of Fe(II) Iron Levels in
Biofilm Supernatants
Synthesis of ferrous tri-dipyridyl chloride {Fe(C10H8N2)Cl2}
standard. A solution consisting of 2.7 g of ferrous sulfate
heptahydrate solubilized in distilled water was poured in
a second solution of 42 mM of 2,2 dipyridyl dissolved in
distilled water. This combined solution was then added to
200 ml of distilled water with hexafluorophosphate pre-
sent in excess. The resulting precipitate was filtered out
using a glass fritted filter and allowed to air dry over-
night. After 24 h the dried precipitate was dissolved in
150 ml of acetone saturated with tetraethyl ammonium
chloride and stirred. The precipitated ferrous tridipyridyl
chloride {Fe(C10H8N2)Cl2} was collected by filtration
using a fritted glass filter and allowed to air dry. Once
completely dry, the ferrous tri-dipyridyl chloride was
scraped into a plastic vial and stored at room tempera-
ture.
A standard curve (Figure 2) was prepared by dissolv-
ing 0.5 mg of the ferrous tridipyridyl chloride into 100ml
of MSG to make an 84.1 mM solution. Serial dilutions
were used to create solutions of the following concentra-
tions: 67.3, 50.4, 33.6, 16.8 and 1.7 µM. 100 µl aliquots
of each dilution was added to a 96 well microtiter plate,
incubated at room temperature for 30 min and absorb-
ance readings taken at 520 nm.
2.7. Quantification of Fe(II) in Biofilm Bulk
Liquid
Biofilms were generated as described above. Following
the 16 h incubation, 5 µl of 35 mM stock of 2,2 dipyridyl
prepared in Milli-Q water was added to each well. The
bulk liquid of two wells (200 µL) was transferred to 1.5
ml microcentrifuge tubes and centrifuged at 10,000 g for
10 min. Following centrifugation 100 µl of the super-
natant was added to a NUNC 96 well round bottom
plates. After 30 min had elapsed from the time the 2,2
dipyridyl was added, the absorbance at 520 nm was

Spions Increase Biofilm Formation by Pseudomonas aeruginosa
512
Figure 2. Fe(II) colorimetric standard curve. Ferrous tridipyridyl chloride was used as the standard. A standard curve was
prepared by dissolving 0.5 mg of the ferrous tridipyridyl chloride into 100 ml of MSG to make an 84.1 mM solution. Serial
dilutions were used to create solutions of the following concentrations: 67.3, 50.4, 33.6, 16.8 and 1.7 µM. 100 µl aliquots of
each dilution was added to a 96 well microtiter plate, incubated at room temperature for 30 min and absorbance readings
taken at 520 nm.
measured on a BioTek Synergy 4 Multi-detection Mi-
croplate Reader and the results recorded.
To measure the effect of FeNPs on Fe(II) levels in
spent supernatant, biofilms were cultured as described
above. Following the 16 h incubation period, bulk liquid
was removed, centrifuged for 10 min at 10,000 g and
filtered through a 0.2 µm filter to remove all cells. Either
no NPs or 200 µg/ml were added to the filtered spent
supernatants and then 100 µL aliquots transferred to a
Falcon BD PVC 96 well microtiter plate. The plate was
then covered in parafilm and incubated at 37˚C statically
for 24 h after which absorbance at 520 nm was measured
as described above.
2.8. Time-Lapse Microscopy
Biofilms were cultured as described above for biofilm
assays. Wet mounts were prepared at 0, 2, 4, 6, 8, and 24
h and viewed under an Olympus BXC51 System Micro-
scope at 1000× magnification.
2.9. Viable Cell Counts
Biofilms were cultured as described above for biofilm
assays. After 16 h, the biofilm bulk liquid from three
replicate samples was diluted in MSG and plated in trip-
licate on the LB agar. Colony forming units were
counted after incubation at 37˚C for 24 h.
3. Results
3.1. Effect of Fe3O4 Nanoparticles on Biofilm
Biomass
The effect of Fe3O4 nanoparticles (FeNPs) on biofilm
formation by P. aeruginosa PAO1 was measured using
the crystal violet biofilm assay modified from that origi-
nally described by O’Toole and Kolter [20]. As seen in
Figure 3(a), addition of Novacentrix nanoparticles at
50µg/ml to planktonic cells resulted in a substantial re-
duction in the amount of biofilm biomass by 16 h, and
standard plate counts revealed there to be no viable bac-
teria present in the bulk liquid. In the presence of the iron
chelator 2,2 dipyridyl, added to simulate iron depleted
growth conditions, the toxicity of the nanoparticles was
ameliorated. Interestingly, wells containing cells grown
in media supplemented with 2,2 dipyridyl and exposed to
FeNPs at concentrations of 25 µg/ml or higher turned a
deep rose color. This color was not observed in wells
without cells. 2,2 dipyridyl is often used for colorimetric
assays to determine the Fe(II) concentrations. The pres-
ence of the rose coloring indicated elevated levels of
Fe(II) when NPs were incubated with cells.
In contrast, nanoparticles from US Research and
Brown (Figures 3(b) and (c)) did not show any toxicity
even at concentrations of 200 µg/ml in standard plate
count assays. In fact there was a significant increase in
biofilm biomass both with or without 2,2 dipyridyl com-
pared to biofilms grown without nanoparticles present.
There was also a 2.5 fold increase in the cell density of
populations exposed to US Research nanoparticles com-
pared with the control without FeNPs. and 2,2 dipyridyl
at 25 µg/ml. As before, the rose color was not observed
in wells containing NPs only.
3.2. Levels of Extracellular DNA and Fe(II) in
Biofilms Exposed to FeNPs
As shown in Figure 4(a), with increasing levels of

Spions Increase Biofilm Formation by Pseudomonas aeruginosa 513
(a)
(b)
(c)
Figure 3. Biofilm biomass comparison using different nanoparticles. Fe3O4 nanoparticles from (a) Novacentrix, (b) US Re-
search, and (c) Brown. Overnight cultures, grown in MSG, or MSG supplemented with 250 µM 2,2 dipyridyl to simulate iron
depleted conditions, were diluted in fresh MSG media to an absorbance of 0.15 at 590 nm. Nanoparticles were added at the
specified concentrations. Cells, or cells + nanoparticles, were transferred to 96-well PVC plates in 100 µl aliquots, covered in
parafilm, and incubated statically for 16 hours at 37˚C. The biofilms were then stained with 0.25% (w/v) crystal violet for 30
min at room temperature, rinsed five times in Milli-Q water and allowed to dry overnight. The next day, the crystal violet
was resolubilized in ethanol, transferred to a 96 well round bottom microtiter plate and absorbance taken at 590 nm. “*”
denotes p < 0.05 compared with PAO1 values without nanoparticles present. Additionally, viable plate counts from Novacen-
trix and US Research has been included.

Spions Increase Biofilm Formation by Pseudomonas aeruginosa
514
FeNPs, there was a decrease in eDNA in the supernatants
of the biofilms. Media supplemented with 250 µM 2,2
dipyridyl restored eDNA levels to those measured for
cells grown without FeNPs (Figure 4(b)). Since 2,2
dipyridyl chelates Fe(II), our results suggest that the ob-
served decrease in eDNA may be related to Fe(II) re-
leased from the FeNPs. In this scenario, increased iron
concentrations would negatively affect Pseudomonas
Quorum Sensing (PQS) and the corresponding eDNA
release from cell lysis. We measured Fe(II) concentra-
tions in biofilm supernatants formed in the presence of
FeNPs. Using ferrous tridipyridyl chloride as the stan-
dard in MSG medium we generated a standard curve
(Figure 2). After biofilms were formed as previously
described, 2,2 dipyridyl was added before cells were re-
moved from the bulk liquid (Note: Adding 2,2 dipyridyl
after the cells were removed caused lower levels of Fe(II)
to be observed). Levels of Fe(II) were significantly
higher in samples with cells and FeNPs at 8, 12, and 16 h
(Figure 5) and reached levels of approximately 150 µM
by 16 h compared with 19 µM in cultures without FeNPs
and less than 4 µM in media with nanoparticles but no
cells present. These results show that cells or cell prod-
ucts are required for the release of the ferrous ion from
the nanoparticles. Since similar results were observed in
both the Brown and US Research Nanomaterials, we
opted to focus our attention strictly on the US Research
nanoparticles.
3.3. Effect of Spent Supernatants on Fe2+ Levels
Generated by FeNPs
Since the presence of cells was required for the release of
the Fe(II) from the nanoparticles, we examined whether
it was cell metabolism, extracellular molecules, or both
responsible for this increase in Fe(II). To answer this
question, we compared Fe(II) levels in spent media from
cultures grown with and without the presence of cells.
Spent supernatants of 16 h biofilms were collected by
centrifugation followed by filtration through a 0.2 µm
filter to remove all cells. To the spent supernatant of each
subset (those grown with and without cells), 200 µg/ml
of FeNPs were added, parallel controls receiving no NPs.
These samples were placed in a 96 well microtiter plate
and allowed to incubate statically for an additional 24 h
at 37˚C prior to measuring the amount of Fe(II) ion pre-
sent. Figure 6 follows a culture of cells grown with or
without the nanoparticles after 16 h and of the super-
natant for an additional 24 h with or without additional
FeNPs added. Fe(II) ion measurements taken immedi-
ately after the cells were removed demonstrated high
levels of Fe(II) present in the supernatant of cells grown
in the presence of nanoparticles. After an additional 24 h,
Fe(II) ion concentration in the spent supernatant de-
(a)
(b)
Figure 4. eDNA in biofilms. Overnight cultures grown in
MSG were diluted in fresh MSG to an OD590 nm of 0.15.
FeNPs were added at the specified concentrations to the
diluted cells and 100 µl volumes aliquoted into wells of a
PVC 96-well microtiter plate. Plates were incubated stati-
cally for 16 h at 37˚C. After 16 h, the bulk liquid was re-
moved and replaced with 125 µl of 1X TE Buffer. The
biofilm was resuspended in the TE buffer and centrifuged
at 10,000 xg for two minutes to pellet bacterial cells. 100 µl
of the supernatant was transferred to a black 96 well mi-
crotiter plate. PicoGreen solution, prepared as per the
manufacturer’s protocol, was added to each well in 20 µl
aliquotes. The samples were incubated for 3 min in the dark
at room temperature and fluorescence measured using
485/535 nm excitation and emission wavelengths. “*” de-
notes p < 0.05 as compared to PAO1 without nanoparticles
added. (a) Cultures grown in MSG, (b) Cultures grown in
MSG supplemented with 250 µM 2,2 dipyridyl.

Spions Increase Biofilm Formation by Pseudomonas aeruginosa 515
creased to similar levels as was seen with PAO1 cells
without the addition of nanoparticles.
3.4. Effect of Non-NP Iron on Biofilm Biomass
In order to determine whether it was the increase in Fe(II)
generated from FeNPs by the cells that was causing the
increase in biofilm biomass, we supplemented the growth
media with Fe(II) or Fe(III) not of FeNP origin. Ferric
chloride or ferrous sulfate was added at 0, 10, 100, or
1000 µM to cells in MSG at T0 during the biofilm assays.
As seen in Figure 7, addition of iron as FeCl3 or FeSO4
had no statistically significant effect on biofilm biomass
at any of the concentrations tested.
3.5. Effect of P. aeruginosa Cells and Spent
Culture Supernatants on FeNP Aggregation
and Loss of Magnetism
Microscopy was used to identify if cells or spent culture
supernatants affected the aggregation of FeNPs. FeNPs
were added to diluted overnight cultures as well as the
spent supernatant of the culture and the mixtures were
aliquoted into 96 well plates. At T = 0, 2, 4, 8 and 24 h
after the addition of the FeNPs wet mounts of the sam-
ples were prepared and observed microscopically. As
shown in Figure 8, the prevalence of observable FeNP
aggregates was lower in samples containing cells. Fol-
lowing observation the samples were exposed to a su-
permagnet by slowing drawing the magnet under the
sample on the microscope slide. Prior to applying a
magnet to the samples, all samples had similar irregular
shaped aggregates. After applying the magnet; however,
iron aggregates in the samples without cells presented an
elongated icicle shaped topology whereas the samples
with cells did not appear to change. Figure 8 shows ob-
servations at 8 h but similar results were seen at all
other time points including at 24. Similarly, at 8 h a mag-
net was placed under the 96 well plate. The nanoparticles
in fresh and spent media formed a pellet on the bottom
(Figures 9(a) and (c)) whereas samples with cells and
FeNP’s did not form a pellet in response to the magnet
and stayed in solution as seen in row E of Figure 9.
Similar results were also obtained at 24 h (data not
shown).
4. Discussion
Previous research has shown FeNPs inhibit bacterial
growth and thus may also inhibit the development of
biofilms [16,17]. In this study we report an increase in
biofilm biomass when P. aeruginosa PAO1 cells were
exposed to SPIONs from US Research and Brown Uni-
versity. In contrast, nanoparticles from Novacentrix were
inhibitory; however, this toxicity can be attributed to the
presence of high levels of heavy metals in these NPs
(Table 1) which are known to be toxic to bacteria. This
finding reinforces the need to ensure a thorough charac-
terization of nanoparticles to avoid misinterpretation of
their effects.
The FeNPs obtained from US Research and Brown
University are unlikely to contain the impurities present
in the Novacentrix NPs as they were synthesized using
well characterized and pure reagents. With both of these
FeNPs a statistically significant increase in biofilm bio-
mass was observed at a concentration of 200 µg/ml. In
addition to the increase in biomass, a corresponding in-
crease in cell density was also seen. Our results therefore
Figure 5. Fe(II) ion concentrations in biofilm bulk liquid. Biofilms were generated as previously described. Following the 16 h
incubation, 5 µl of 35 mM stock of 2,2 dipyridyl prepared in Milli-Q water was added to each well. The bulk liquid of two wells
was transferred 1.5 ml microcentrifuge tubes, centrifuged at 10,000 g for 10 min. Following centrifugation 100 µl of the super-
natant was added to a NUNC 96 well round bottom plates. After 30 min had elapsed from the time the 2,2 dipyridyl was added,
the absorbance at 520 nm was taken. “*” denotes p < 0.05 as compared to PAO1 without nanoparticles added at each timepoint.

Spions Increase Biofilm Formation by Pseudomonas aeruginosa
516
Figure 6. Effect of spent supernatant on Fe(II) ion release. Concentrations of Fe(II) ions of a diluted overnight culture was
measured following 16 hours with (A) or without (B) nanoparticles present. Cells were removed via centrifuging 10,000 g for
10 minutes followed by filtering through a 0.2 µm filter. To half of each supernatant (A and B), an additional 200 µg/ml of
iron NPs were added (D and F) and no additional nanoparticles were added to the other half (C and E). Aliquots of 100 µl of
each were added to a 96 well plate and allowed to incubate statically for an additional 24 h at 37˚C before 2,2 dipyridyl was
added and Fe(II) measurements taken.
Figure 7. Effect of non-NP iron on biofilm biomass. Overnight cultures were diluted in fresh MSG media supplemented with
either ferric chloride or ferrous sulfate at the indicated concentrations. Biofilm biomass was measured following a static 16 h
incubation at 37˚C.
do not support the use of FeNPs for the inhibition of
biofilms as previously suggested [16,17].
Our study is the first to, our knowledge, that shows
bacterial cells interact with FeNPs to release Fe(II). In
the presence of cells Fe(II) levels exceeded 100 µM by 8
h compared with less than 20 µM in the absence of cells.
Cell-free spent supernatants did not have this effect. The
increased levels of biofilm biomass observed for cells
grown in the presence of FeNPs may be due to Fe(II)
supplied by FeNPs; however, similar levels of Fe(II)
supplied by the addition of ferrous sulfate did not result
in a corresponding increase in biofilm biomass. This
might be explained if the slow release of Fe(II) from the
FeNPs advantages the cells differently than the addition
of similar levels of FeSO4 at the outset, or low levels
early followed by higher levels later on enhance biofilm
formation. Alternately, the FeNPs effect on growth is not
simply due to increases in Fe(II). Increases in Fe(II) lev-
els were not observed until 4 h after the addition of
FeNPs. Even in the presence of 2,2 dipyryidyl, which
binds in a 3:1 stoichometry with Fe(II), by eight hours all
of the 2,2 dipyridyl would be saturated and the amount of
Fe(II) in the culture would begin to increase which would
lend credence that it is the timing of the release of Fe(II)

Spions Increase Biofilm Formation by Pseudomonas aeruginosa 517
Before Magnet After Magnet
Fresh Media
+ FeNPs
Spent Media
+ FeNPs
Cells
+ FeNPs
Figure 8. Effect of cells or spent medium on FeNP aggre-
gates after magnet exposure. Suspensions were sampled at
T = 0, 2, 4, 8 and 24 h (T = 8 shown). Wet mounts were
prepared and viewed under an Olympus BXC51 System
Microscope at 1000× magnification. Suspensions were ex-
posed to the magnet by passing the magnet under the mi-
croscope slide in a slow continuous manner in a single di-
rection.
Figure 9. Effect of supermagnet exposure on FeNP aggre-
gates in wells. Fresh media (row A), spent media (row C),
and cells (row E) in which 200 µg/ml of FeNPs were added,
incubated statically for 8 hours, and then exposed to a su-
permagnet placed beneath the wells.
that is important to biofilm development. Our findings
indicate that the FeNPs used in this study behave, or at
least can be manipulated to behave, as a source of ele-
mental iron for use by the cell; the temporal effect of
Fe(II) availability should be looked at more closely.
Iron has been shown to have an effect on eDNA levels
in biofilms via the PQS quorum sensing system [11].
Increased iron ion concentrations leads to decreases in
PQS production which in turn results in decreased cell
lysis and corresponding eDNA release. Our studies show
an increase in Fe(II) levels in media when FeNPs are
incubated with P. aeruginosa, and a corresponding de-
crease in the amount of eDNA. These results support
previous experiments on the effect of iron on eDNA re-
lease and supports our findings that P. aeruginosa can
use FeNPs as an iron source.
Our results also indicate that P. aeruginosa cells affect
both the aggregation and response to magnetic force of
the FeNPs. At 8 h, nanoparticle aggregates were not as
prevalent in samples with cells. Additionally, exposure of
FeNPs in the presence of cells to a supermagnet did not
affect the shape of FeNP aggregates nor could the aggre-
gates be pelleted in microtitre plates. These observations
support our data showing that cells are interacting with
the nanoparticles in some way to alter their properties.
In this study, we have shown the use of Fe3O4 na-
noparticles increase biofilm biomass with a correspond-
ing increase in cell density and a decrease in eDNA in
the biofilm matrix. Additionally, we demonstrate that the
cells interact with nanoparticles to: 1) Release Fe(II); 2)
decrease the prevalence of FeNP aggregates over time,
and 3) alter the response of the particles to magnetic
forces. We hypothesize this may be due to some type of
reducing agent such as leukopyacyanin which may cause
the release of Fe(II) ions from the nanoparticles [21,22]
thereby altering the surface of the nanoparticles and al-
tering their physical properties. Additionally, bacterial
surfactants may play a role in keeping the iron from ag-
gregating thereby ensuring a larger surface area from
which the reducing agent can interact and would thus
also limit our ability to observe the aggregates through a
light microscope. LDS characterization of FeNPs in the
presence of natural surfactant should provide some in-
sight into this hypothesis.
There is an increased focus on the use of superpara-
magnetic iron oxide nanoparticles (SPIONS) in biomedi-
cal applications. Our findings have implications in the
clinical and ecological sense, two areas where FeNPs are
being looked at to inhibit microbial growth. With the
booming nanotechnology industry and the use of FeNPs
in many household items, such effects must be considered.
5. Acknowledgements
We wish to acknowledge financial support by USAF to
Carl Haney and Jayne Robinson.
REFERENCES
[1] K. Todar, “Opportunistic Infections Caused by Pseudo-

Spions Increase Biofilm Formation by Pseudomonas aeruginosa
518
monas aeruginosa,” Todar’s Online Textbook of Bacteri-
ology, 2008. http://www.textbookofbacteriology.net
[2] N. Hoiby, T. Bjarnsholt, M. Givskov, S. Molin and O. Cio-
fu, “Antibiotic Resistance of Bacterial Biofilms,” Inter-
national Journal of Antimicrobial Agents, Vol. 35, No. 4,
2010, pp. 322-332.
doi:10.1016/j.ijantimicag.2009.12.011
[3] M. L. Vasil and U. A. Ochsner, “The Response of Pseu-
domonas aeruginosa to Iron: Genetics, Biochemistry and
Virulence,” Molecular Microbiology, Vol. 34, No. 3,
1999, pp. 399-413.
doi:10.1046/j.1365-2958.1999.01586.x
[4] G. Papanikolaou and K. Pantopoulos, “Iron Metabolism
and Toxicity,” Toxicology and Applied Pharmacology,
Vol. 202, No. 2, 2005, pp. 199-211.
[5] U. A. Ochsner and J. Reiser, “Autoinducer-Mediated Re-
gulation of Rhamnolipid Biosurfactant Synthesis in Pseu-
domonas aeruginosa,” Proceedings of the National Aca-
demy of Sciences of the United States of America, Vol. 92,
No. 14, 1995, pp. 6424-6428.
[6] T. R. de Kievit, “Quorum Sensing in Pseudomonas aeru-
ginosa Biofilms,” Environmental Microbiology, Vol. 11,
No. 2, 2009, pp. 279-288.
doi:10.1111/j.1462-2920.2008.01792.x
[7] D. J. Musk, D. A. Banko and P. J. Hergenrother, “Iron
Salts Perturb Biofilm Formation and Disrupt Existing
Biofilms of Pseudomonas aeruginosa,” Chemistry & Bi-
ology, Vol. 12, No. 7, 2005, pp. 789-796.
doi:10.1016/j.chembiol.2005.05.007
[8] P. K. Singh, M. R. Parsek, E. P. Greenberg and M. J.
Welsh, “A Component of Innate Immunity Prevents Bac-
terial Biofilm Development,” Nature, Vol. 417, No. 6888,
2002, pp. 552-555.
[9] G. M. Patriquin, E. Banin, C. Gilmour, R. Tuchman, E. P.
Greenberg and K. Poole, “Influence of Quorum Sensing
and Iron on Twitching Motility and Biofilm Formation in
Pseudomonas aeruginosa,” Journal of Bacteriology, Vol.
190, No. 2, 2008, pp. 662-671. doi:10.1128/JB.01473-07
[10] C. Y. O’May, K. Sanderson, L. F. Roddam, S. M. Kirov
and D. W. Reid, “Iron-Binding Compounds Impair Pseu-
domonas aeruginosa Biofilm Formation, Especially un-
der Anaerobic Conditions,” Journal of Medical Microbi-
ology, Vol. 58, No. 6, 2009, pp. 765-773.
doi:10.1099/jmm.0.004416-0
[11] L. Yang, K. B. Barken, M. E. Skindersoe, A. B. Chris-
tensen, M. Givskov and T. Tolker-Nielsen, “Effects of
Iron on DNA Release and Biofilm Development by
Pseudomonas aeruginosa,” Microbiology (Reading, Eng-
land), Vol. 153, No. 5, 2007, pp. 1318-1328.
[12] F. Berlutti, C. Morea, A. Battistoni, S. Sarli, P. Cipriani,
F. Superti, M. G. Ammendolia and P. Valenti, “Iron
Availability Influences Aggregation, Biofilm, Adhesion
and Invasion of Pseudomonas aeruginosa and Burk-
holderia cenocepacia,” International Journal of Immuno-
pathology and Pharmacology, Vol. 18, No. 4, 2005, pp.
661-670.
[13] E. Banin, M. L. Vasil and E. P. Greenberg, “Iron and Pseu-
domonas aeruginosa Biofilm Formation,” Proceedings of
the National Academy of Sciences of the United States of
America, Vol. 102, No. 31, 2005, pp. 11076-11081.
[14] X. Chen and P. S. Stewart, “Role of Electrostatic Interac-
tions in Cohesion of Bacterial Biofilms,” Applied Micro-
biology and Biotechnology, Vol. 59, No. 6, 2002, pp.
718-720. doi:10.1007/s00253-002-1044-2
[15] Y. Ju-Nam and J. R. Lead, “Manufactured Nanoparticles:
An Overview of Their Chemistry, Interactions and Poten-
tial Environmental Implications,” The Science of the Total
Environment, Vol. 400, No. 1-3, 2008, pp. 396-414.
doi:10.1016/j.scitotenv.2008.06.042
[16] E. N. Taylor and T. J. Webster, “The Use of Superpara-
magnetic Nanoparticles for Prosthetic Biofilm Preven-
tion,” International Journal of Nanomedicine, Vol. 4,
2009, pp. 145-152.
[17] S. Ravikumar, R. Gokulakrishnan, K. Selvanathan and S.
Samayanan, “Antibacterial Activity of Metal Oxide Na-
noparticles against Ophthalmic Pathogens,” International
Journal of Pharmaceutical Research and Development,
Vol. 3, No. 5, 2011, pp. 122-127.
[18] R. C. Moulton and T. C. Montie, “Chemotaxis by Pseu-
domonas aeruginosa,” Journal of Bacteriology, Vol. 137,
No. 1, 1979, pp. 274-280.
[19] Y. S. Kang, S. Risbud, J. F. Rabolt and P. Stroeve, “Syn-
thesis and Characterization of Nanometer-Size Fe3O4 and
Γ-Fe2O3 Particles,” Chemistry of Materials, Vol. 8, No. 9,
1996, pp. 2209-2211.
[20] G. A. O’Toole and R. Kolter, “Initiation of Biofilm For-
mation in Pseudomonas fluorescens WCS365 Proceeds
via Multiple, Convergent Signalling Pathways: A Genetic
Analysis,” Molecular Microbiology, Vol. 28, No. 3, 1998,
pp. 449-461. doi:10.1046/j.1365-2958.1998.00797.x
[21] W. M. Huston, A. J. Potter, M. P. Jennings, J. Rello, A. R.
Hauser and A. G. McEwan, “Survey of Ferroxidase Ex-
pression and Siderophore Production in Clinical Isolates
of Pseudomonas aeruginosa,” Journal of Clinical Micro-
biology, Vol. 42, No. 6, 2004, pp. 2806-2809.
doi:10.1128/JCM.42.6.2806-2809.2004
[22] C. D. Cox, “Role of Pyocyanin in the Acquisition of Iron
from Transferrin,” Infection and Immunity, Vol. 52, No. 1,
1986, pp. 263-270.