Multispecies biofilm in an artificial wound bed-A novel model for in vitro assessment of solid antimicrobial dressings

Full text

Multispecies biofilm in an artificial wound bed—A novel model for in vitro
assessment of solid antimicrobial dressings
J. Kucera
a,d,1
, M. Sojka
b,c,
⁎
,1
,V.Pavlik
b,e,1
, K. Szuszkiewicz
a
,V.Velebny
a,b
,P.Klein
a,1
a
Wound Healing Research Group, ContiproPharma, Dolní Dobrouč, Czech Republic
b
Cell Physiology Research Group, ContiproBiotech, Dolní Dobrouč, Czech Republic
c
Institute of Microbiology, Faculty of Medicine, Slovak Medical University, Bratislava, Slovakia
d
Department of Histology and Embryology, Faculty of Medicine in Hradec Kralove, Charles University in Prague, Czech Republic
e
Department of Dermatology, Third Faculty of Medicine, Charles University in Prague, Prague, Czech Republic
abstractarticle info
Article history:
Received 17 February 2014
Received in revised form 4 May 2014
Accepted 5 May 2014
Available online 28 May 2014
Keywords:
Artificial wound bed
Multi-species wound biofilm model
Anti-biofilm substances
In vitro testing
Wound infections represent a major problem, particularly in patients with chronic wounds. Bacteria in the
wound exist mainly in the form of biofilms and are thus resistant to mostantibiotics and antimicrobials. A simple
and cost-effective in vitro model of chronic wound biofilms applied for testing treatments and solid devices, es-
peciallywound dressings, is presented in thiswork. The method is based on thewell-established Lubbock chronic
wound biofilm transferred onto an artificial agar wound bed. The biofilm formed by four bacterial species
(Staphylococcus aureus,Enterococcus faecalis,Bacillus subtilis and Pseudomonas aeruginosa) was stable for up to
48 h post-transplant. The applicability of the model was evaluated by testing two common iodine wound treat-
ments. These observations indicate that this method enables assessing the effects of treatments on established
resilient wound biofilms and is clinically highly relevant.
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
Microbial biofilms are structured communities of bacterial cells
enclosed in a self-produced polymeric matrix and adherent to an inert
or living surface (Costerton et al., 1999). Sessile and planktonic microbi-
al cells are phenotypically and physiologically different (Donlan and
Costerton, 2002). Bacteria forming biofilms are highly resistant to
many traditional therapies. Bacteria in biofilms can adapt to a sessile
state by down-regulating cellular activity and encapsulating in a mas-
sive structure of extracellular polysaccharides (Brady et al., 2008;
Sutherland, 2001). There is a growing recognition that biofilms are
one of the principal causes of wound chronicity (Wolcott et al., 2010).
Over 90% of chronic wounds contain bacteria and fungi from the skin,
oral mucosa, enteric tract or the environment. Together these bacteria
form a multispecies biofilm construct (Attinger and Wolcott, 2012;
Price et al., 2009). Novel treatments for wound biofilms have been re-
cently developed, potentially saving many lives by preventing systemic
infections (Wolcott et al., 2010).
In order to develop antimicrobial therapies and test treatments,
it is essential to have appropriate microbiological models. Most meth-
odologies used to study antimicrobials and test medical devices use
planktonic microbial cultures (Costerton et al., 1999). Several wound
biofilm models were described previously to study different aspects of
wound biofilms. These models use multiple species and aim to mimic
the polymicrobial nature of wound biofilms (reviewed by Coenye and
Nelis, 2010). Werthén et al. (2010) developed a model of wound biofilm
without a solid surface and grown in the presence of a simulated body
fluid composed of peptone and foetal calf serum. One of the more so-
phisticated biofilm models is based on tissue-engineered skin (Charles
et al., 2009).
The first chronic wound biofilm model was developed by Sun et al.
(2008) at the Medical Biofilm Research Institute in Lubbock, Texas,
and was named the “Lubbock chronic wound biofilm (LCWB) model”.
This model was shown to be a realistic in vitro multispecies biofilm
which grows and matures rapidly, is cost effective and easy to set up.
Only liquid or semi-solid substances with putative inhibitory effects
on biofilm formation were tested on the LCWB model (Dowd et al.,
2009). This model was also modified for the high throughput testing
of anti-biofilm properties of different woundcare products on staphylo-
coccal biofilms (Brackman et al., 2013). Furthermore, the LCWB model
was successfully transplanted into murine skin wounds to induce for-
mation of wound biofilm (Dalton et al., 2011).
To our knowledge, few models were described for testing anti-
biofilm activity of wound dressings and other solid materials. Lipp
et al. (2010) used a drip-flow reactor model with monospecies biofilms
only. Hammond et al. (2011) developed a burn wound “biofilm”model
that comprised burn wound bacterial isolates grown on cellulose discs
Journal of Microbiological Methods 103 (2014) 18–24
⁎Corresponding au thor at: Institute of Microbiology, Facu lty of Medicine, Slov ak
Medical University, Limbova 12, 833 03Bratislava, Slovakia. Tel.: +421 2 59370736.
E-mail address: martin.sojka@szu.sk (M. Sojka).
1
These authors contributed equally to this work.
http://dx.doi.org/10.1016/j.mimet.2014.05.008
0167-7012/© 2014 Elsevier B.V. All rights reserved.
Contents lists available at ScienceDirect
Journal of Microbiological Methods
journal homepage: www.elsevier.com/locate/jmicmeth

and placed on agar plates. Different antibiotic ointments soaked in
gauze were applied on the discs.
Recognizing the need for adequate in vitro biofilm models for evalu-
ating solid anti-microbial wound dressings, we employed the well-
established superior multispecies LCWB model. We transferred the
pre-cultured biofilm onto an artificial wound bed and verified the appli-
cability of this model for the testing of wound dressings. Here we de-
scribe the evaluation of this biofilm.
2. Materials and methods
2.1. Bacteria
Staphylococcus aureus,Enterococcus faecalis,Bacillus subtilis and
Pseudomonas aeruginosa, originally isolated from patients with chronic
infected wounds hospitalized at the University Hospital, Hradec Kralove
(Czech Republic), were used in this study. Cryopreserved bacterial
strains were grown for 24 h at 37 °C on Columbia agar plates supple-
mented with sheep blood (Oxoid, Germany). Sodium chloride peptone
broth (buffered peptone bouillon, BPB; Merck, Germany) was generally
used for dilutions and measuring optical density of cultures.
2.2. Modified Lubbock chronic wound biofilm (LCWB) model
We used the previously described LCWB model (Sun et al., 2008)
with some modifications and amendments to pre-form the matured
biofilms for treatment: briefly, 6 ml of liquid biofilm formation medium
containing Bolton broth base (Sigma, Germany), 1% gelatine, 50% por-
cine plasma and 5% freeze-thawed porcine erythrocytes was dispensed
into sterile 1.6 × 10 cm polystyrene tubes (Gama, Czech Republic). To
control for possible variability in biofilm formation caused by the differ-
ent batches of blood in the culture medium, four biological replicates
were prepared in duplicate, each with blood from a different pig.Optical
density-normalized cultures of four bacterial species were mixed to-
gether and 10 μlof10
6
CFU/ml culture were inoculated into the tubes
by ejecting the pipette tips along with the mixed bacterial suspen-
sion. The inoculated tubes were incubated at 37 °C in an orbital shaker
(1.5 ×g) for up to 48 h. The biofilms were harvested at selected time
intervals (12, 24, 36 and 48 h post-inoculation (p.i.)). Biofilms harvested
48 h post-inoculation were used to model chronic would biofilms and
treated.
2.3. Treatment of established biofilms
Petri dishes with a two-layer nutrient medium composed of Bolton
broth supplemented with 1% (w/v) gelatin and 1.2% (w/v)agar
(Sigma, Germany) were prepared as follows: a 2-mm thin layer of nutri-
ent medium was poured into Petri dishes. One sterile 20 × 8 mm PTFE-
coated magnetic stirring bar was put onto the agar in each dish after so-
lidification of the nutrient medium. A second 2-mm layer of nutrient
medium was subsequently added. After the medium had completely
congealed, the stirring bars were carefully and aseptically removed
from the agar creating oval-shaped artificial wound beds. Pre-formed
48 hour-old biofilm was removed from the tube, washed with BPB
and the pipette tip extracted from the biofilm using sterile forceps and
a scalpel. The biofilm was placed into the “wound bed”in the nutrient
medium and covered with a piece (2 × 5 cm) of 100% cotton 8-ply
gauze sponge (Batist, Czech Republic), soaked with a test substance
(Fig. 1). Biofilm cultures were incubated at 37 °C for 24 and 48 h respec-
tively. After treatment, the biofilms were harvested from the artificial
wound bed using sterile forceps and a Lang eye spoon, homogenized,
and the bacteria were enumerated.
In our study aimed at model optimization and characterization,
two commonly used antimicrobial wound treatments were applied to
the biofilm model: polyvinypyrrolidone–iodine complex (0.2 mg of
iodine/cm
2
)–2 ml of 10% Alfadin (Bioveta, Czech Republic) per gauze
and cadexomer-iodine complex (0.2 mg of iodine/cm
2
)–2 ml of 11%
Iodosorb gel (Smith and Nephew, USA), or 2 ml of concentrated Iodosorb
gel per gauze (1.8 mg of iodine/cm
2
), respectively. Gauze pieces soaked
with 2 ml of BPB were used as controls.
2.4. Biofilm processing
Harvested biofilms were washed in BPB as follows; excess medium
was removed with sterile cotton and biofilms were weighed. Subse-
quently, the biofilms were homogenized using a rotor-stator laboratory
homogenizer (UltraTurrax, IKA, Germany). The biofilm homogenates
were divided in three equal portions and used for quantification of
biofilm bacteria and RNA isolation. Homogenates for molecular assays
were resuspended in RNAlater (Life Technologies, USA), incubated
overnight at 4 °C, pelleted and stored at −80 °C.
2.5. Quantitative cultures of biofilm bacteria
Homogenized biofilms were initially diluted 1:10 in BPB and vigor-
ously vortexed for 2–3 min. The suspended cells were then diluted 10-
fold, and 10 μl aliquots of each dilution and undiluted homogenate
Fig. 1. Treatment of pre-formed biofilms transferred to artificial wound bed. Schematic
drawing displays the cross-section of the Petri dish with a two-layer nutrient medium
and centrally cultured biofilm covered with test antimicrobial agent/wound dressing.
Table 1
qPCR primers specifications.
Primer pair Target Sequences 5′–3′Final concentration Reference
16S 16S rDNA/rRNA TCCTACGGGAGGCAGCAGT
GGACTACCAGGGTATCTAATCCTGTT
100 nM Nadkarni et al., 2002
SA S.aureus
nuc
GCGATTGATGGTGATACGGTT
AGCCAAGCCTTGACGAACTAAAGC
300 nM Hein et al., 2001
EF E.faecalis
16S rDNA/rRNA
CCCTTATTGTTAGTTGCCATCATT
ACTCGTTGTACTTCCCATTGT
500 nM Rinttilä et al., 2004
PA P.aeruginosa
16S rDNA/rRNA
CAAAACTACTGAGCTAGAGTACG
TAAGATCTCAAGGATCCCAACGGCT
600 nM Matsuda et al., 2007
BS B.subtilis
16S rDNA/rRNA
CCCTTATTGTTAGTTGCCATCATT
GGACTACCAGGGTATCTAATCCTGTT
100 nM This study
Nadkarni et al., 2002
icaA S.aureus
adhesin
TGAACCGCTTGCCATGTG
CACGCGTTGCTTCCAAAGA
200 nM Rode et al., 2007
ebrA E.faecalis
GntR family protein
TCGTCGTCATGGCAAAGGAA
AGCAATCCGCAACCGACTTA
500 nM This study
19J. Kucera et al. / Journal of Microbiological Methods 103 (2014) 18–24

spotted onto agar plates. To differentiate between the four bacterial
species used in this model selective media (Columbia CAP agar
(Oxoid, UK) for S. aureus,E. faecalis and B. subtilis; ENDO agar (Oxoid,
UK) for P. aeruginosa, kanamycin aesculin azide agar (Oxoid, UK) for
E. faecalis) and non-selective Columbia agar with sheep blood were
used. Selective properties of used media together with the growth
characteristics on blood supplemented media wereused to differentiate
between bacterial species. After incubation at 37 °C for 16 h, the CFU
counts of each particular species were determined. The viable bacterial
counts were expressed as CFU per mg of original biofilm.
2.6. Bacteria quantification and gene expression assays using qPCR
Propidiummonoazide (PMA) binds to extracellular DNA and, when
exposed to light, creates crosslinks (Nocker et al., 2009). DNA modified
by PMA is a suboptimal qPCR template. This property enabled us to
quantify total DNA in biofilmsand DNA in intact cells which is protected
from PMA. Prior to genomic DNA isolation, each of the homogenized
biofilm samples was split into two aliquots. To one of the aliquots
PMA (Biotium, USA) was added to a final concentration of 40 μM. The
samples were then incubated for 10 min in the dark at room tempera-
ture with occasional shaking. Both the PMA-treated and untreated
samples were placed on ice on a shaker and exposed to a 1000 W halo-
gen lamp for 5 min. The samples were centrifuged (8000 ×g,10 min),
the bacterial pellet was resuspended in lysis buffer (20 mg/ml lyso-
zyme, 20 mM Tris–HCl, pH 8.0; 2 mM EDTA; 1.2% Triton-X 100, Sigma,
Germany) and genomic DNA was isolated with DNeasy Blood and
Tissue Mini Kit in a QIAcube isolator (Qiagen, UK). An initial lysis step
of 60 min was used and the second elution volume was 150 μl. After
DNA isolation, a third manual elution with 50 μl of AE buffer was per-
formed and the eluates were pooled. DNA concentration was deter-
mined spectrophotometrically.
qPCR was employed for quantification of bacteria and bacterial gene
expression. The primers are shown in Table 1. Power SYBR Green
Master Mix 2X (Life Technologies, USA) was used for quantification of
bacterialDNA with primer pairs 16S, SA, EF, PA and BS usingthe follow-
ing cycling conditions: 95 °C, 10 min; 40 cycles of 95 °C, 15 s; 62 °C,
1 min. Three OD-adjusted samples containing the four bacterial species
in exponential growth phase were simultaneously aliquoted for DNA
isolation and plated for estimating CFU/ml counts. A calibration curve
was set up with the DNA isolated from calibrated bacterial suspensions.
Quantification cycle values and the corresponding CFU/ml values were
subsequently transformed to CFU/mg of biofilm.
For studying the expression of biofilm markers, RNA was isolated
with RNAzol (MRC, USA) according to the user manual. Reverse tran-
scription of 0.5 μg of total RNA was accomplished with High Capacity
RNA to cDNA Master Mix (Life Technologies, USA). The reactions were
set up with Ssofast Evagreen Supermix (Bio-Rad, USA) and primer
pairs SA, EF, icaA and ebrA (Generi Biotech, Czech Republic; see
Table 1). The cycling conditions were 95 °C, 5 min; 40 cycles of 95 °C,
5 s 60 °C, 1 min. The resulting quantification cycle valueswere analysed
with 2
−ΔΔCt
method(Livak and Schmittgen, 2001). As a reference genes
were chosen staphylococcal nuclease A (SA primers) and enterococcal
16S rRNA (EF primers). The gene expression was expressed as a relative
change in gene expression of icaA or ebrA between planktonic and bio-
film bacteria.
2.7. Histological examination of biofilms
Biofilms cultured in the test-tubes for 12, 24, 36 and 48 h were
fixed with 4% paraformaldehyde-PBS at room temperature for at least
24 h. Longitudinal paraffin sections of 3–5μm were stained with
haematoxylin-eosin for visualizing biofilm morphology, with Gram
staining for differentiation between gram-positive and gram-negative
bacterial species and with PAS-AB (Periodic Acid-Schiff Alcian Blue) to
visualize extracellular polymeric substances according to standard
protocols. Furthermore, specific immunohistochemical (IHC) detec-
tion was performed with anti-S. aureus (ab37644, Abcam), anti-
P. aeruginosa (ab68538, Abcam) and anti-E. faecalis (ab19980, Abcam)
antibodies and anti-mouse or anti-rabbit EnVision + HRP.DAB (Dako)
detection kit, respectively. Biofilm morphology was observed and docu-
mented under 20× to 1000× magnification using an Eclipse 50i (Nikon,
Japan) microscope with attached DS-Fi1 (Nikon, Japan) camera. NIS-
Elements AR 3.2 (Laboratory Imaging, Czech Republic) imaging soft-
ware was used. Evaluation and description of morphology were
recorded and transcribed into text with NovaVoice software (NovaSoft,
Czech Republic).
2.8. Scanning electron microscopy (SEM)
Scanning electron microscopy was performed with dewaxed vacu-
um desiccated histological sections of biofilms, sputter coated with
gold in a SC7620 instrument (Quorum Technologies Ltd., UK). Samples
0
50
100
150
200
250
300
350
400
450
12 24 36 48
weight [mg]
hrs p.i.
Fig. 2. Biofilm wet weights at selected time intervals post-inoculation (p.i.). Bars show
mean ± SD; n= 8 (four biological replicates, each in duplicate).
0
1
2
3
4
5
6
7
0 12243648
hrs p.i.
N (log10/mg BF)
A
0
1
2
3
4
5
6
7
8
9
12 24 36 48
hrs p.i.
SA
EF
PA
BS
N (log10/mgBF)
B
Fig. 3. Bacterialcountsinbiofilms after 12, 24, 36 and 48 h post-inoculation (p.i.)
determined in culture (A) or quantitative PCR (B). Y-axis displays log
10
of bacterial num-
bers per mg of biofilm mass. Mean ± SD, n= 8 for cultivation methods, n=4forqPCR.
(SA—S.aureus,EF—E.faecalis,PA—P.aeruginosa,BS—Bacillus subtilis.)
20 J. Kucera et al. / Journal of Microbiological Methods 103 (2014) 18–24

were analysed with a VEGA II LSU scanning electron microscope
(Tescan, Czech Republic). Samples were scanned with a secondary
electron detector at 10 kV beam voltage in high vacuum at various
magnifications.
2.9. Statistical analysis
Biofilm weights were log-transformed and analysed with ANOVA.
Post hoc comparison of the data was done with Tukey multiple com-
parisons of means (95% family-wise confidence level). The Wilcoxon
Signed-Rank Test for one sample was computed to compare percent
of live bacteria to the theoretical 100%. ANOVA was performed to test
the effect of various anti-biofilm agents at 24 and 48 h after biofilm
transfer, followed by Tukey multiple comparisons of means (95%
family-wise confidence level, Tukey HSD). The gene expression levels
of planktonic and biofilm bacteria were compared with Student's
t-tests. The above computations were done in the R software environ-
ment (R Core Team, 2013).
3. Results and discussion
3.1. Dynamics of bacterial populations during biofilm formation
The wet weights of the LCWB determined after homogenization
are depicted in Fig. 2. Biofilm wet weight did not significantly change
during the course of the experiment. Three of the four biological repli-
cate groupsmade with blood from different pigs showed similarweight.
Biofilms cultured in medium supplemented with plasma from one par-
ticular animal tended to weigh more than the others (pb0.01). How-
ever, bacterial counts per mg of the biofilm did not differ significantly
among biological replicates.
Selective culture and qPCR were used for quantifying bacteria during
biofilm formation (Fig. 3). Population growth patterns observed in cul-
ture and by qPCR correlated well. A marked increase in viable bacterial
counts was recorded at 24h p.i. as compared with 12 h p.i., especially in
case of S.aureus and P.aeruginosa.Biofilms cultured for 24–48 h
contained 10
5
–10
6
CFU/mg (10
6
–10
7
genomes/mg) of three surviving
bacterial species. E. faecalis was the predominant species, followed
*
0
20
40
60
80
100
120
140
12 24 36 48
hrs p.i.
Enterococcus faecalis
**
0
20
40
60
80
100
120
140
160
12 24 36 48
hrs p.i.
Pseudomonas aeruginosa
*
*
**
0
20
40
60
80
100
120
12 24 36 48
% alive
hrs p.i.
Staphylococcus aureus
Fig. 4. Viability of biofilm bacteria determined by qPCR following exposure to PMA. Graphs sho w for each species the mean percent PMA protected bacterial genomes
within uncompromised cell walls (live), normalized against total bacterial genome concentration set to 100%, at four time points post-inoculation. Error bars show standard deviation
(*pb0.05 for one sample t-test, n=4).
Fig. 5. Micromorphologyof model biofilms culturedin the test-tubes for 48 h. Relevant serial sections are presented, showing S.aureus (a, b, c), P.aeruginosa (d, e, f) and E.faecalis(g, h, i).
Gram stain(a, d, g), IHC localization of particular bacterial species(b, e, h) and PAS-AB staining of extracellular polysaccharide substances producedby bacteria (c, f, i) are shown. PAS-AB
staining shows production of acidic polysascharides by S.aureus and neutral polysaccharides by P.aeruginosa. No polysaccharide substances were produced by E.faecalis. Arrows show
particular bacterial populations in biofilms. Bars represent 100 μm.
21J. Kucera et al. / Journal of Microbiological Methods 103 (2014) 18–24

by S.aureus and P.aeruginosa. The ubiquitous and generally non-
pathogenic bacterium B.subtilis fell below detection level in the course
of the experiment. A few living B.subtilis bacteria were detected until
36 h p.i. by means of bacterial culture. This species was found only occa-
sionally by qPCR. B. subtilis was detectable neither by cultivation tech-
niques nor by qPCR after 48 h p.i.
Total genomic DNA (gDNA) and gDNA from live cells was quantified
by means of differential nucleic acid amplification after exposure to
PMA (Fig. 4). Genomic DNA from E.faecalis and P.aeruginosa was
contained mostly in intact cells. There was a minor increase of extracel-
lular DNA, presumably from dead E.faecalis and P.aeruginosa cells after
36 h (21% and 15%, respectively). In contrast, approximately 40% of
the S.aureus gDNA was extracellular, i.e., likely to originate from dead
cells. The data did not show a clear association between biofilm incuba-
tion time and fraction of dead S.aureus cells. Staphylococcal autolysis
probably occurred even in the earliest phase of LCWB formation. Extra-
cellular DNA rise in biofilm mainly by cellular lysis and can also be ex-
creted under specific conditions (Barnes et al., 2012). Also, the lysis
may be result of cell death either due to stress or is a part of organized
build-up of extracellular DNA mass. One should bear in mind that differ-
ential amplification of genomic DNA is not capable of distinguishing the
cause of cell death.
Fig. 6. Scanning electron micrographsof the 48 hour-old biofilm model formed in test-tubes.Complex extracellular polymeric substances embedding the cells ofE.faecalis (a, b) S.aureus
(c, d) and P.aeruginosa (e, f)are shown. Bars represent20 μminfigures a, c, e and 2 μminfiguresb, d, f; squares in figuresa, c, e indicate areasof which the detailsare presented in figures
b, d, f.
22 J. Kucera et al. / Journal of Microbiological Methods 103 (2014) 18–24

3.2. Morphological characteristics and spatial distributionof bacteriain the
biofilm
The harvested pre-formed biofilm had a solid–slimy consistency
with macroscopic appearance characterized by the“central hollow”
formed by the pipette tip, the “superficial biofilm area”at the medium/
air interface and the “bottom area”at the pipette tip nozzle. In general,
no major changes in microscopic morphology were observed in the
course of biofilm formation before the transfer to artificial wound bed.
Extracellular fibrous network formed the scaffold for biofilm-
forming bacteria grown in colonies, as was previously shown by Sun
et al. (2008) and Dalton et al. (2011).E. faecalis was found in discrete
colonies of various densities of bacterial cells, 10–20 μm in diameter,
which tended to grow to more than 100 μm over the time. PAS-AB
staining showed no specific polysaccharide production by E.faecalis
(Fig. 5 a, b, c). S. aureus formed highly compact colonies that produced
acidic polysaccharides (light blue colour in PAS-AB), and were found
mainly at the top of the central hollow (Fig. 5 d, e, f). P. aeruginosa colo-
nies were of different size and morphology (more diffuse than those of
other species) and were found predominantly in the biofilm cavities
and disruptions. P. aeruginosa was present mainly on the superficial
area of the biofilms at all observed time intervals, except that at 12 h
p.i. PAS-AB staining showed large amounts of neutral polysaccharides
(purple to rose colour) surrounding the P.aeruginosa colonies (Fig. 5
g, h, i). The ultrastructure of the biofilm was also evaluated by SEM. A
complex, interconnected fibrous network of extracellular polymeric
substances embedding the microcolonies of E.faecalis (Fig. 6a, b)
S.aureus (Fig. 6c,d) and P.aerugino sa (Fig. 6e, f) was visible. The findings
are similar to those reported by Sun et al. (2008).
3.3. Expression of biofilm RNA markers
To assess whether the bacteria in pre-cultured LCWB werein biofilm
phenotype, transcription of biofilm-related genes was evaluated.
Expression of two candidate genes –icaA and ebrA –was previously
shown to increase in S.aureus or E.faecalis, respectively, during biofilm
formation on polystyrene tips or cellulose coupons (Rode et al., 2007;
Ballering et al., 2009). We compared gene expression of planktonic
bacteriagrownfor16or24hwithbiofilm bacteria (Fig. 7). The expres-
sion of ebrA was significantly increased in the biofilm and did not
change over the time. Interestingly, the expression of ebrA was lower
in planktonic cells after 16 h than after 24 h. However, although icaA
was expressed in S.aureus, there was no difference between planktonic
and biofilm bacteria. Differences in icaA transcription among various
S.aureus strains were found previously (Rode et al., 2007), which may
support icaA-independent mechanism of biofilm formation. The results
suggest a shift in gene expression towards the biofilm phenotype at
least for E.faecalis.
3.4. Effect of iodine treatment on established biofilms
Based on the results described above, we decided to use 48 hour-old
pre-formed biofilms for transfer to the artificial wound bed and treat-
ment. The stability and reproducibility of untreated model biofilms
*
***
****
0
1
2
3
4
5
12 24 36 48
Relative expression (log2)
hrs p.i.
A
-3
-2
-1
0
1
2
3
4
5
12 24 36 48
Relative expression (log2)
hrs p.i.
Biofilm vs.
planctonic 16 hrs
Biofilm vs.
planctonic 24 hrs
B
Fig. 7. Expression of biofilmmRNA markers. A—ebrA expression in E.faecalis.B—icaA expression in S.aureus. Bars represent mean expression of ebrA or icaA relative to planktonic bacteria
(±SD) cultivated16 h (dark shade) or 24 h (light shade). Expression is shown on a log
2
scale. n=8 for biofilm,n= 5 for planktonic bacteria. (*pb0.05,Student's t-testcomparing biofilm
and planktonic bacteriafor the given time point.)
*
*
*
*
*
*
*
*
*
*
*
0
1
2
3
4
5
6
7
8
9
10
BPB
control
BPB
control
PVP-I
(0,2)
C-I (0,2) C-I (1,8) BPB
control
PVP-I
(0,2)
C-I (0,2) C-I (1,8)
0 hrs p.t. 24 hrs p.t. 48 hrs p.t.
log10 CFU/mg
S. aureus
E. faecalis
B. subtilis
P. aeruginosa
Fig. 8. Countsof viable bacteria in untreated and treatedbiofilms 24 and 48 h post-transfer and post-treatment (p.t.).The Y-axis displaysthe log
10
of bacterialcounts per mg biofilm mass.
Shown are means ± SD (*, pb0.05, Tukey HSD comparing bacterial counts for untreated and treated biofilm at a given time point, n = 4; BPB—buffered peptone bouillon, PCP-I—poly-
vinylpyrrolidone-iodine, C-I—cadexomer-iodine.)
23J. Kucera et al. / Journal of Microbiological Methods 103 (2014) 18–24

were tested together with possible effects of different forms of iodine—
an antimicrobial agent commonly used in wound care. Selective cul-
tures of untreated and treated biofilmsweremadetoevaluatethe
response of bacterial populations to transfer to artificial wound bed
and to treatment. The data show relatively stable bacterial counts in un-
treated biofilms for up to 48 h (Fig. 8).
The solid–slimy macroscopic appearance of the biofilms did not sig-
nificantly change during the course of the experiment. The comparison
of viable bacterial counts at selected time points in treated biofilms with
those in untreated biofilms enabled an assessment of the anti-biofilm
properties of the tested substances. Only stabilized iodine in the form
of cadexomer-iodine at an initial concentration of 1.8 mg I
2
/cm
2
signif-
icantly reduced the number of biofilm bacteria for up to 48 h, whereas
polyvinypyrrolidone-iodine and cadexomer-iodine at a concentration
of 0.2 mg I
2
/cm
2
appeared ineffective against established polymicrobial
biofilms.
3.5. Conclusions
In this study, the benefits of using the presented chronic wound
biofilm model were clearly demonstrated. Multispecies biofilms were
stable for at least 48 h after transfer to the artificial wound bed and
thus allow for long term testing of anti-biofilm treatments on matured
biofilms. Use of the artificial wound bed in the model enables to
mimic the situation in chronic infected wound, where biofilm is only
in partial contact with wound dressing. As illustrated for two iodine
treatments, the model described here could facilitate the cost-effective
and simple in vitro evaluation for anti-biofilm activity of virtually any
topically applied substance or wound dressing.
Acknowledgement
We would like to express our gratitude to Veronika Hekrlova, Lenka
Nespechalova B.Sc., Pavla Hlavackova, Darina Majercikova and Lucie
Simunkova B.Sc. for their extensive technical assistance. This research
was conducted under financial support provided by Technology Agency
of the Czech Republic (project TA03011029 - New wound dressings
with programmed release of active substances for biofilm inhibition)
and by the Charles University in Prague, Faculty of Medicine in Hradec
Kralove (grant SVV-2014-260058).
References
Attinger, C., Wo lcott, R., 2012. Clinically addressing biofilm in chronic wounds. Adv.
Wound Care 1, 127–132.
Ballering, K.S., Kristich, C.J., Grindle, S.M., Oromendia, A., Beattie, D.T., Dunny, G.M., 2009.
Functional genomics of Enterococcus faecalis: multiple novel genetic determinants for
biofilm formation in the core genome. J. Bacteriol. 191, 2806–2814.
Barnes, A.M.T., Ballering, K.S., Leibman, R.S., Wells, C.L., Dunny, G.M., 2012. Enterococcus
faecalis produces abundant extracellular structures containing DNA in the absence
of cell lysis during early biofilm formation. mBio 3 (4), e00193-12. http://dx.doi.
org/10.1128/mBio.00193-12.
Brackman, G., De Meyer, L., Nelis, H.J., Coenye, T.,2013. Biofilm inhibitoryand eradicating
activity of wound care products against Staphylococcus aureus and Staphylococcus
epidermidis in an in vitro chronic wound model.
Brady, R.A., Leid, J.G., Calhoun, J.H., Costerton, J.W., Shirtliff, M.E., 2008. Osteomyelitis and
the role of biofilms in chronic infection. Immunol. Med. Microbiol. 52, 13–22.
Charles, C.A., Ricotti, C.A., Davis, S.C., Mertz, P.M., Kirsner, R.S., 2009 . Use of tissue-
engineered skin to studyin vitro biofilm development.Dermatol. Surg.35, 1334–1341.
Coenye, T., Nelis, H.J., 2010. In vitro and in vivo model systems to study microbial biofilm
formation. J. Microbiol. Methods 83, 89–105.
R. Core Team, 2013. R: A language and environment for statistical computing. R Founda-
tion for Statistical Computing, Vienna, Austria.
Costerton, J.W., Stewart, P.S., Greenberg, E.P., 1999. Bacterial biofilms: a commoncause of
persistent infections. Science 284, 1318–1322.
Dalton, T., Dowd, S.E., Wolcott, R.D., Sun, Y., Watters, C., Griswold, J.A., Rumbaugh, K.P.,
2011. An in vivo polymicr obial biofilm woundinfection model to study interspeciesin-
teractions. PLoS One 6 (11), e27317. http://dx.doi.org/10.1371/journal.pone.0027317.
Donlan, R.M., Costerton, J.W., 2002. Biofilms: survival mechanisms of clinically relevant
microorganisms. Clin. Microbiol. Rev. 15, 167–193.
Dowd, S.E., Sun, Y., Smith, E., Kennedy, J.P., Jones, C.E., Wolcott, R., 2009. Effects of biofilm
treatments on the multi-species Lubbock chronic wound biofilm model. J. Wound
Care 18, 508–512.
Hammond, A.A., Miller, K.G., Kruczek, C.J., Dertien, J., Colmer-Hamood, J.A., Griswold, J.A.,
Horswill, A.R., Hamood,A.M., 2011. An in vitro biofilm model to examine the effect of
antibiotic ointments on biofilms produced by burn wound bacterial isolates. Burns
37, 312–321.
Hein, I., Lehner, A., Rieck, P., Klein, K., Brandl, E., Wagner, M., 2001. Comparison of differ-
ent approaches to quantify Staphylococcus aureus cells by real-time quantitative PCR
and application of this technique forexamination of cheese. Appl. Environ. Microbiol.
67, 3122–3126.
Lipp, C., Kirker, K., Agostinho, A., James, G., Stewart, P., 2010. Testing wound dressings
using an in vitro wound model. J. Wound Care 19, 220–226.
Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-
time quantitative PCR and the 2
−ΔΔCT
method. Methods 25, 402–408.
Matsuda, K., Tsuji, H., Asahara, T., Kado, Y., Nomoto,K., 2007. Sensitive quantitative detec-
tion of commensal bacteria by rRNA-targeted reverse transcription-PCR. Appl. Envi-
ron. Microbiol. 73, 32–39.
Nadkarni, M.A., Martin, F.E., Jacques, N.A., Hunter, N., 2002. Determination of bacterial
load by real-time PCR using a broad-range (universal) probe and primers set. Micro-
biology 76, 2 57–266.
Nocker, A., Mazza, A., Masson, L., Camper, A.K., Brousseau, R., 2009. Selective detection of
live bacteria combining propidiummonoazide sample treatment with microarray
technology. J. Microbiol. Methods 76, 253–261.
Price, L.B., Liu, C.M., Melendez, J.H., Frankel, Y.M., Engelthaler, T., Aziz, M., Bowers, J.,
Rattray, R., Ravel, J., Kingsley, C., Keim, P.S., Lazarus, G.S., Zenilman, J.M., 2009. Com-
munity analysis of chronic wound bacteria using 16S rRNA gene-based pyrosequenc-
ing: impact of diabetes and antibiotics on chronic wound microbiota. PLoS One 4,
e6462.
Rinttilä,T., Kassinen, A., Malinen, E., Krogius, L., Palva, A., 2004.Development of an exten-
sive set of 16S rDNA-targeted primers for quantification of pathogenic and indige-
nous bacteria in faecal samples by real-time PCR. J. Appl. Microbiol. 97, 1166–1177.
Rode, T.M., Solveig, L., Askild, H., Møretrø, T., 2007. Different patterns of biofilm formation
in Staphylococcus aureus under food-related stress conditions. Int. J. Food Microbiol.
11, 372–383.
Sun, Y., Dowd, S.E., Smith, E., Rhoads, D.D ., Wolcott, R.D., 2008. In vitro multispecies
Lubbock chronic wound biofilm model. Wound Repair Regen. 16, 805–813.
Sutherland, I.W., 2001. Biofilm exopolysaccharides: a strong and sticky framework. Mi-
crobiology 147, 3–9.
Werthén, M., Henriksson, L., Jensen, P.Ø., Sternberg, C., Givskov, M., Bjarnsholt, T., 2010.
An in vitro model of bacterial infections in wounds and other soft tissues. APMIS
118, 156–164.
Wolcott, R.D., Rhoads, D.D., Bennett, M.E., Wolcott, B.M., Gogokhia, L., Costerton, J.W.,
Dowd, S.E., 2010. Chronic wounds and the medical biofilm paradigm. J. Wound Care
19, 45–53.
24 J. Kucera et al. / Journal of Microbiological Methods 103 (2014) 18–24