Targeting microbial biofilms using Ficin, a nonspecific plant protease

Abstract

Biofilms, the communities of surface-attached bacteria embedded into extracellular matrix, are ubiquitous microbial consortia securing the effective resistance of constituent cells to environmental impacts and host immune responses. Biofilm-embedded bacteria are generally inaccessible for antimicrobials, therefore the disruption of biofilm matrix is the potent approach to eradicate microbial biofilms. We demonstrate here the destruction of Staphylococcus aureus and Staphylococcus epidermidis biofilms with Ficin, a nonspecific plant protease. The biofilm thickness decreased twofold after 24 hours treatment with Ficin at 10 μg/ml and six-fold at 1000 μg/ml concentration. We confirmed the successful destruction of biofilm structures and the significant decrease of non-specific bacterial adhesion to the surfaces after Ficin treatment using confocal laser scanning and atomic force microscopy. Importantly, Ficin treatment enhanced the effects of antibiotics on biofilms-embedded cells via disruption of biofilm matrices. Pre-treatment with Ficin (1000 μg/ml) considerably reduced the concentrations of ciprofloxacin and bezalkonium chloride required to suppress the viable Staphylococci by 3 orders of magnitude. We also demonstrated that Ficin is not cytotoxic towards human breast adenocarcinoma cells (MCF7) and dog adipose derived stem cells. Overall, Ficin is a potent tool for staphylococcal biofilm treatment and fabrication of novel antimicrobial therapeutics for medical and veterinary applications.

Full text

1
Scientific RepoRts | 7:46068 | DOI: 10.1038/srep46068
www.nature.com/scientificreports
Targeting microbial biolms using
Ficin, a nonspecic plant protease
Diana R. Baidamshina1,*, Elena Y. Trizna1,*, Marina G. Holyavka2, Mikhail I. Bogachev3,
Valeriy G. Artyukhov2, Farida S. Akhatova1, Elvira V. Rozhina1, Rawil F. Fakhrullin1 &
Airat R. Kayumov1
Biolms, the communities of surface-attached bacteria embedded into extracellular matrix, are
ubiquitous microbial consortia securing the eective resistance of constituent cells to environmental
impacts and host immune responses. Biolm-embedded bacteria are generally inaccessible for
antimicrobials, therefore the disruption of biolm matrix is the potent approach to eradicate microbial
biolms. We demonstrate here the destruction of Staphylococcus aureus and Staphylococcus epidermidis
biolms with Ficin, a nonspecic plant protease. The biolm thickness decreased two-fold after 24 hours
treatment with Ficin at 10 μg/ml and six-fold at 1000 μg/ml concentration. We conrmed the successful
destruction of biolm structures and the signicant decrease of non-specic bacterial adhesion to the
surfaces after Ficin treatment using confocal laser scanning and atomic force microscopy. Importantly,
Ficin treatment enhanced the eects of antibiotics on biolms-embedded cells via disruption of
biolm matrices. Pre-treatment with Ficin (1000 μg/ml) considerably reduced the concentrations of
ciprooxacin and bezalkonium chloride required to suppress the viable Staphylococci by 3 orders of
magnitude. We also demonstrated that Ficin is not cytotoxic towards human breast adenocarcinoma
cells (MCF7) and dog adipose derived stem cells. Overall, Ficin is a potent tool for staphylococcal biolm
treatment and fabrication of novel antimicrobial therapeutics for medical and veterinary applications.
Biolms are formed by the surface-attached bacterial cells arranged into complex communal tertiary structures
and embedded into an extracellular matrix1,2. e bulk of the matrix is formed by extracellular polymeric sub-
stances (EPS) that typically constitute up to 95% of the biolm and consist of biopolymers (i.e polysaccharides,
proteins, lipids and nucleic acids) produced and secreted by the constituent bacteria. e matrix supports the
three-dimensional structure of the biolm and protects the cells from various environmental impacts.
Bacterial cells in biolms are extremely resistant to medicinal treatment and immune system attacks, that
leads to chronic reinfections1,3,4. Many opportunistic bacteria (i.e. Staphylococcus, Micrococcus, Klebsiella,
Pseudomonas, etc.) form biolms on chronic and acute dermal wounds impeding their healing, causing reinfec-
tion and sepsis1,3,4. Accordingly, the colonization with S. epidermidis and/or S. aureus is a common cause of intra-
and extravascular catheter-associated infection, implants, wound surfaces and mucous membranes5. As a result,
bacterial biolms appear a signicant clinical challenge leading to increased patient morbidity and mortality
from infectious diseases6,7. erefore, the prevention of biolm formation and disruption of already established
biolms is crucially important for clinical treatment of infectious diseases8–10.
Destroying the biolm matrix backbone, for example via enzymatic lysis, is an advantageous approach for
biolms eradication6. Numerous bacterial enzymes, such as glycosidases, proteases, and DNases degrade various
components of biolms stimulating cells detachment and increasing cellular susceptibility to antimicrobials11.
In particular, the glycoside hydrolase dispersin B produced by Aggregatibacter actinomycetemcomitans has been
shown to sensitize S. epidermidis biolm-embedded cells to antimicrobials action12,13. Dispersin B injection in
combination with triclosan reduced the catheter colonization density by S. aureus in rabbits in vivo14. Another
glycoside hydrolase, alginate lyase, successfully enhanced the activity of aminoglycosides against P. aeruginosa
biolms both in vitro15,16 and in vivo17. DNase (NucB) from Bacillus licheniformis induced rapid dispersal of
biolm formed by B. subtilis, E. coli and M. luteus18. Recombinant human DNase I (rhDNase) has been shown to
disperse preformed S. aureus biolms and increase the susceptibility of S. aureus biolm cells to antiseptics6. In
1Kazan Federal University, Institute of Fundamental Medicine and Biology, Kazan, Republic of Tatarstan, Russian
Federation. 2Voronezh State University, Medicine and Biology Faculty, Voronezh, Russian Federation. 3St Petersburg
Electrotechnical University, Biomedical Engineering Research Centre, St. Petersburg, Russian Federation. *These
authors contributed equally to this work. Correspondence and requests for materials should be addressed to A.R.K.
(email: kairatr@yandex.ru)
Received: 26 September 2016
Accepted: 08 March 2017
Published: 07 April 2017
OPEN

www.nature.com/scientificreports/
2
Scientific RepoRts | 7:46068 | DOI: 10.1038/srep46068
addition, two glycoside hydrolases from Pseudomonas aeruginosa eciently destroyed the Pseudomonas biolm
backbone19.
Proteases are believed to be one of the most eective enzymes in biolm eradication via hydrolysis of both
matrix proteins and adhesins (proteins providing cells attachment onto solid surfaces and other bacteria)20,21 as
well as by the cleavage of signaling peptides of intercellular communication of gram-positive bacteria22. Recently,
several groups reported the ecacy of proteases as wound healing agents simultaneously exhibiting anti-biolm
properties, such as degradation of the biolm matrix structural components and destruction of its backbone23–26.
e serine protease Esp from S. epidermidis has been demonstrated to inhibit the biolm formation by S. aureus
and to eradicate the already preformed biolms10. Similar eects have been shown for the elastase LasB from
P. aeruginosa and proteinase K10. Finally, the metalloprotease serratopeptidase (SPEP) produced by Serratia marc-
escens is widely used as an anti-inammatory agent, successfully inhibiting biolm formation and enhancing the
ecacy of ooxacin against biolms of both P. aeruginosa and S. epidermidis27. Two other enzymes, glycosidase
pectinase and protease subtilisin A have been shown to suppress the biolm formation of Escherichia coli and
enhance the cell sensitivity to ampicillin24. Chymotrypsin derived from maggot excretions/secretions was shown
to disrupt a protein component of staphylococcal biolms28. e treatment of Listeria monocytogenes with sub-
lethal concentrations of serratiopeptidase from Serratia marcescens reduced their ability to form biolms and to
invade host cells29.
In this paper we show that Ficin (EC 3.4.22.3), a nonspecic sulydryl protease isolated from the latex of the
Ficus tree, disrupts the staphylococcal biolm backbone, thus signicantly increasing the eciency of conven-
tional antibiotics.
Results and Discussion
Staphylococcal biolms disruption by Ficin. Over decades, a number of proteolytic enzymes have been
adopted in clinical practice as wound healing agents destroying the cell debris and necrotic tissues. Recently,
several proteases were reported to exhibit anti-biolm properties and to increase the susceptibility of biolm-em-
bedded bacterial cells to antibiotics23–26. We investigated whether Ficin is able to disrupt bacterial biolms formed
by S. aureus and S. epidermidis, the bacteria colonizing wounds and thus retarding wound healing30. To do so,
the bacteria were grown in BM broth earlier developed31,32 for 72 h on 24-well TC-treated plates that provided
a representative and repeatable formation of the rigid biolm strongly attached to the surfaces, in contrast to
Müller-Hinton broth, Tr ypticase soy broth or LB-medium (Fig.1). Next the plates were washed twice by fresh BM
followed by incubation during 24 h in the fresh BM broth in the presence of Ficin at concentrations of 10, 100 and
1000 μ g/ml, since the recommended concentrations of proteolytic enzymes used for wounds healing (like Trypsin
and Chymotrypsin) are 1–2 mg/ml33,34. en, the culture liquid was discarded and the residual biolms were
quantied by crystal violet staining. e control wells were subjected to all procedures described above except the
enzyme addition aer the wash and medium replacement. Wells were stained with crystal violet and their absorb-
ance was taken as 100%. Our data indicate that Ficin eectively destroyed the established 3-days old biolms
formed by both S.aureus and S.epidermidis which can be typically observed on wounds35 and cause nosocomial
infections (Fig.2). Even at 10 μ g/ml of Ficin only ca. 55–65% of the initial biolm mass remained as conrmed
by crystal violet staining, and biolms were almost completely eliminated at higher Ficin concentration (1000 μ
g/ml) (OD570 < 0.1). Remarkably, the other proteolytic enzymes such as trypsin or papain could decrease the
staphylococcal biolm for 20–30% only at 100 μ g/ml and on 50–60% at 1000 μ g/ml36 conrming higher eciency
of Ficin for the treatment of staphylococcal biolms.
To verify the stability of Ficin in the culture liquid, the proteolytic activity was measured using azocaseine as
substrate37 in wells aer the enzyme addition. During the rst 4 hours more that 90% of the initial activity was
detectable in the liquid (Fig.S1), and approximately half of activity remained in the cultures aer 24 hours incu-
bation suggesting high stability of the enzyme.
The biolm structure after Ficin treatment. To test the hydrolysis of the protein components of the
biolm matrix by Ficin, the preformed 3 day-old biolms were treated with enzyme in the presence of Congo red,
a specic dye staining the amyloid proteins (Fig.3). e control wells incubated with Congo red in the absence of
Figure 1. e biolm formation by S. aureus and S. epidermidis cultivated in Basal medium (BM), Luria-
Bertani broth (LB), Müller-Hinton broth (MH), or Trypticase soy broth (TSB) on 35-mm polystyrol
adhesive plates. 72 hours-old biolms were stained by crystal violet.

www.nature.com/scientificreports/
3
Scientific RepoRts | 7:46068 | DOI: 10.1038/srep46068
Ficin were red-stained. In the presence of Ficin a signicant decrease of the staining intensity could be observed
for both S. aureus and S. epidermidis plates, indicating the degradation of the protein backbone of the biolm.
en, to investigate how Ficin aects the biolm structure, the biolms of S. aureus and S.epidermidis treated
with Ficin were analyzed by confocal laser scanning microscopy (CLSM). For imaging, both S. aureus and S.
epidermidis were grown for 48 h in 500 μ l of BM broth in cell imaging coverglass slides (Eppendorf) to form the
biolm. en 250 μ l of broth was replaced with the fresh aliquote containing Ficin reaching the nal concentra-
tion of 1000 μ g/ml. Aer 24 h incubation the cells were stained with DioC6 and propidium iodide as described in
Materials and Methods and analyzed using CLSM (Fig.4A).
In control wells the biolms of both strains reached 20–22 μ m and formed pronounced mushroom-shaped
structures with cell agglomerates (Fig.4A, control lane). In the presence of Ficin a signicant suppression of the
S. aureus biolm was observed, while less pronounced eect was detected for S. epidermidis. Also the structure
of the biolm has been changed. Unlike in the control sample, in the Ficin-treated samples a mushroom-like
structure of the staphylococcal biolm disappeared, whereas the uniform layer of the cells could be observed
suggesting the destruction of the protein backbone of the matrix. In contrast to the control, this layer could be
easily removed by pipetting suggesting its low adherence to the surface. Notably, the fractions of dead cells were
rather comparable in wells with or without protease, demonstrating no expressed antimicrobial activity of Ficin
and suggesting the absence of a direct evolutionary pressure on the bacterial resistance development.
To verify that the observed changes in the biolm structure are caused by enzymatic action of Ficin, the estab-
lished biolms were also treated with enzyme in the presence of protease inhibitors mix. As shown on Fig.4B, nei-
ther inhibitor alone nor inhibited Ficin caused changes in the biolm structure and cell viability of Staphylococci,
conrming that Ficin destroyed the biolm by hydrolyzing proteins of its matrix.
For a deeper investigation of the staphylococcal biolm structural changes aer treatment with Ficin, both
treated and untreated biolms were imaged using atomic force microscopy (Fig.5). AFM data conrms that
Ficin treatment leads to ecient eradication of the biolms. While the overall morphology of the isolated cells
Figure 2. e biolm disruption by Ficin. S. aureus (A) and S. epidermidis (B) were grown in BM broth for
72 h to form a rigid biolm, the mature biolms were gently washed by BM and a fresh BM broth was loaded.
Ficin was added until nal concentrations of 10, 100 or 1000 μ g/ml and incubation was followed for 24 h. e
residual biolms were quantied by crystal-violet staining.
Figure 3. Evaluation of matrix proteins hydrolysis with Ficin. Bacteria were grown on BM medium for 72 h
to form a biolm, then a medium was replaced by the fresh one containing Ficin (1000 μ g/ml) and Congo red
and incubation was continued for the next 24 h.

www.nature.com/scientificreports/
4
Scientific RepoRts | 7:46068 | DOI: 10.1038/srep46068
in Ficin-treated samples remained unaected, the cell density was severely reduced. In control samples, the cells
formed a typical conuent multilayer biolm, as shown in AFM images (Fig.5). Noteworthy, the non-specic tip
adhesion mapping showed that the force is almost identical in the cells located in either lower or upper visible bio-
lm levels. In the case of Ficin-treated biolms, the AFM imaging revealed the island-like cell clusters on the plate
surface. e morphology of Ficin-treated cells was apparently unaected compared to non-treated cells, while the
density of the cell layers was considerably lower. e Peak Force Tapping atomic force microscopy (AFM) allowed
to obtain high-resolution images of the quality matching that of the contact mode AFM without damaging the
cells. Moreover, unlike in tapping mode AFM, the topography data of microbial cells could be obtained with no
typical edge defects due to tip parachuting. Consequently, our AFM topography images of the biolms represent
the precise nanoscale reconstruction of the actual structure of biolms grown on polymer surfaces conrming
the biolm removal aer Ficin treatment.
Further, in S. aureus biolms treated with Ficin the non-specic adhesion of non-functionalized AFM probe
tip was somewhat reduced, unlike in intact biolms, indicating that the specic adhesion of the cells to substrates
might be also reduced. On the other side, only 2–4 fold decrease of S. epidermidis and S. aureus biolms layer aer
Ficin treatment was observed in CLSM microphotographs, while both crystal violet and Congo red quantica-
tion showed 5–7-fold reduction of the biolm (Figs2 and 3). Since the AFM images demonstrated a monolayer
of residual cell clusters on the surface, we hypothesized that the 5–8 μ m layer observed with CLSM (Fig.4) might
represent the sedimented cells which are not well-adherent to the surface anymore. is hypothesis is partially
conrmed by the non-specic tip adhesion AFM data for Ficin-treated samples, which appeared to be some-
what lower when compared to control samples, indicating that the specic adhesion of the cells to the substrates
might also be reduced. Non-specic adhesion forces of cell surfaces to silicon nitride AFM tips by no means can
be directly extrapolated onto the adhesive properties of bacteria to real substrates. However, this may serve as
a good indicator of certain physiological eects occurring in Ficin-treated bacteria forming biolms. Together
with Congo red staining data (Fig.2) and cell density observations (Fig.3) from the Peak Force Tapping AFM
nanomechanical data (Fig.5) this suggests that Ficin apparently hydrolyses both the biolm matrix and proteins
participating in the adhesion of microbial cells, thus signicantly reducing their ability to form biolms as shown
for other proteases.
Ficin treatment enhances the ecacy of antimicrobials against biolm-embedded Staphylococci.
Aer being embedded into the matrix of the biolm, bacteria become almost inaccessible for biocides and
Figure 4. Confocal laser scanning microscopy. S. aureus and S. epidermidis 48 h-old biolms were established
in cell imaging cover slips (Eppendorf) and treated with Ficin in absence (A) or presence (B) of protease
inhibitors. Aer 24 h incubation cells were stained with DioC6 and propidium iodide to evaluate the cell
viability. e scale bars indicate 5 µm.

www.nature.com/scientificreports/
5
Scientific RepoRts | 7:46068 | DOI: 10.1038/srep46068
antibiotics. We tested whether Ficin would increase the eciency of antibiotics against surface-adherent bacteria
due to the biolm damage. Both S. aureus and S. epidermidis strains were sensitive to ciprooxacin according to
EUCAST rules (http://mic.eucast.org/), therefore this antibiotic was chosen as a model antimicrobial drug. e
MIC values of ciprooxacin established by the broth microdilution method were 2 μg/ml for S.aureus and 1 μg/ml
for S. epidermidis. e MBCs were 8 μg/ml and 4 μg/ml, respectively.
To test the eect of ciprooxacin on S. aureus and S. epidermidis biolm-embedded cells in the presence of
Ficin, 48-h biolms were prepared on 96-well TC-treated plates. e established biolms were washed twice by
fresh BM broth to remove non-adherent cells, and incubated for the next 24 h in the fresh BM broth in the pres-
ence of Ficin and ciprooxacin as indicated (Fig.6, Fig.S2). Ciprooxacin was added to the nal concentrations
of 1× , 2× , 4× and 8× MBCs, the nal concentration of Ficin was xed at 1000 μ g/ml. Aer 24 h incubation, the
culture liquids with planktonic and detached cells were saved, the biolms were washed twice by sterile 0.9%
NaCl. en the viability of both detached and biolm-embedded cells was analyzed by drop plate assay. e
experiments were carried out in biological triplicates with three independently treated samples in each one, the
latter being averaged in each biological replicate, the dierences between groups were analyzed by using Pearson’s
Chi-squared test and were considered signicant at p < 0.05.
e viability of both S. aureus and S. epidermidis cells in either biolm or culture liquid was insignicantly
aected by the enzyme (Fig.6, Fig.S2). When ciprooxacin was added into the broth, its 8 × MBC reduced the
amount of S. aureus and S. epidermidis detached cells by nearly 2 and 3 orders of magnitude, respectively (Fig.S2).
Figure 5. Atomic force microscopy (Peak Force Tapping mode) of intact and Ficin-treated S. aureus
and S. epidermidis biolms. Bacteria were grown in BM broth for 72 h to form a rigid biolm, the
mature biolms were gently washed by BM and a fresh BM broth was loaded. Ficin was added until nal
concentrations of 1000 μ g/ml and incubation was followed for 24 h. e residual biolms were washed, xed
with glutardialdehyde and analyzed with AFM. (A) – height (topography); (B) – peak force error image; (C) –
adhesion force image.

www.nature.com/scientificreports/
6
Scientific RepoRts | 7:46068 | DOI: 10.1038/srep46068
In the presence of Ficin half of the initial antibiotic concentration was required to achieve the same eect, proba-
bly, due to the possible disintegration of detached bacterial clumps by the enzyme. Signicant dierences between
ciprooxacin-treated cells in presence or absence of Ficin were observed at 8 × MBC of antibiotic. e CFUs
of the biolm-embedded cells of both strains decreased only 10-fold in the presence of ciprooxacin even at
8 × MBC, while in the presence of Ficin the decrease up to 3 orders of magnitude could be observed (Fig.6) with
signicance of 0.05 at 8 × MBC for S. aureus and 4–8 × MBC for S. epidermidis. At lower Ficin concentration
(100 μ g/ml) the increase of ciprooxacin ecacy was also observed although less pronounced (not shown). e
increase of ciprooxacin ecacy against biolm-embedded Staphylococci was also veried using the confocal
laser scanning microscopy. For that, the cells were grown for 48 h in 500 μ l of BM broth in cell imaging cover-
glass slides (Eppendorf) to prepare the biolm. en 250 μ l of broth was replaced with the fresh one containing
Figure 6. e Ficin treatment increases the ecacy of ciprooxacin against biolm-embedded
Staphylococci. Ficin (1000 μg/ml) and ciprooxacin (1–8 × MBC) were added to 48 hours-old biolms of
S. aureus and S. epidermidis. Aer 24 h incubation, the biolms were washed twice with sterile 0.9% NaCl.
e adherent cells were scratched, resuspended and their viability was analyzed by using drop plate assay
(A,B). Alternatively, 48 hours-old biolms of S. aureus and S. epidermidis were incubated 24 h in presence of
Ficin (1000 μ g/ml) and ciprooxacin (8 × MBC) in cell imaging coverglass slides and analyzed with confocal
scanning microscopy (C–J). Signicant dierences between 10 log10 of the viable cell counts aer treatment
with ciprooxacin in either absence of presence of Ficin according to Pearson’s Chi-squared homogeneity test
(p < 0.05) are indicated in the gure. e scale bars indicate 5 µm.

www.nature.com/scientificreports/
7
Scientific RepoRts | 7:46068 | DOI: 10.1038/srep46068
Ficin (1000 μ g/ml) and ciprooxacin (8 × MBC). Aer 24 h cultivation the cells were stained with DioC6 and
propidium iodide, as described in Materials and Methods, and analyzed with CLSM (Fig.6). In wells containing
only ciprooxacin most cells were stained in green suggesting their viability (Fig.6D,H), with only a few dead
cells being detectable. In contrast, in the wells with both antibiotic and protease nearly no viable cells could be
detected, and considerably reduced quantity of red-stained cells could be observed.
e eciency of other antimicrobials regularly used for outer treatment of wounds also increased in presence
of Ficin. In particular, Ficin treatment led to the twofold decrease of the ecient concentration of Benzalkonium
chloride, the biocide belonging to quaternary ammonium salts (Fig.7, Fig.S3). Here, the signicant dierences
between Ficin treated and untreated cells were observed at low concentrations of antimicrobial (1–2 × MBC) for
Figure 7. e Ficin treatment increases the ecacy of benzalkonium chloride against biolm-embedded
Staphylococci. Ficin (1000 μ g/ml) and benzalkonium chloride (1–8 × MBC) were added to 48 hours-old
biolms of S. aureus and S. epidermidis. Aer 24 h incubation, the biolms were washed twice with sterile 0.9%
NaCl. e adherent cells were scratched, resuspended and their viability was analyzed by using drop plate assay
(A,B). Alternatively, 48 hours-old biolms of S. aureus and S. epidermidis were incubated 24 h in presence of
Ficin (1000 μ g/ml) and benzalkonium chloride (8 × MBC) in cell imaging coverglass slides and analyzed with
confocal scanning microscopy (C–J). Signicant dierences between 10 log10 of the viable cell counts aer
treatment with benzalkonium chloride in either absence of presence of Ficin according to Pearson’s Chi-squared
homogeneity test (p < 0.05) are indicated in the gure. e scale bars indicate 5 µm.

www.nature.com/scientificreports/
8
Scientific RepoRts | 7:46068 | DOI: 10.1038/srep46068
both detached and biolm-embedded cells. Similar eect could be observed for gentamycin (Fig.S4), although
less pronounced, probably due to the low sensitivity of strains used to this antimicrobial.
To analyze the antimicrobial enhancement eciency in the presence of Ficin in more details, respective
dose-response curves were plotted providing residual CFUs function as a function of antibiotic concentrations
(see Fig.8, upper panel for S. aureus, lower panel for S. epidermidis). Rough estimates of the dose-response
curves were obtained by linear regression applied in logarithmic scale. Figure8 shows that the addition of Ficin
signicantly increased the sensitivity of both S. aureus and S. epidermidis cells to Ciprooxacin leading to 10-fold
discrepancy for 8 × MBC. In contrast, for Benzalkonium chloride and Gentamicin treatment of both S. aureus
and S. epidermidis cells, the eect of Ficin was clearly observed already at 1 × MBC, likely indicating higher
susceptibility of the biolm-embedded cells to respective antibiotics. To achieve comparable eect without Ficin
treatment, the antibiotic concentrations had to be increased 4- to 16-fold. Accordingly, our results indicate that
treatment with Ficin reduces the required antimicrobial dose likely due to the increased susceptibility of the
biolm-embedded cells.
For detached cells (see Fig.S5 in the SupplementaryInformation available), the above eects were less pro-
nounced, and the discrepancy between cells treated with either antibiotics and cin or with antibiotics alone was
less signicant, while still some limited enhancement of treatment ecacy could be observed at large antibiotic
concentrations, probably due to the destruction of detached cell clumps by Ficin.
Altogether, these observations suggest that Ficin destroys the biolm backbone making the cell accessible
for antimicrobials. Similar eect has been observed previously for subtilisin A and some 2(5 H)-furanone deriv-
atives38,39, suggesting that disruption of biolms could be one of the factors of how proteases speed the wound
healing. Furthermore, a signicant decrease in the bacterial biolm thickness was observed, this way conrming
that the biolm was nearly completely eradicated and suggesting the combination of the Ficin with antibiotics as
a promising approach for the development of wounds treatment therapeutics.
Cytotoxicity evaluation. To investigate the cytotoxicity of Ficin, the metabolic MTS-assay was performed
employing MCF7 cells, human skin broblasts and dog adipose derived stem cells (ADSC) (see Table1). No
suppression of the dehydrogenase activity by the enzyme was detected within the concentrations tested aer the
cells were treated by the enzyme for 24 h. Additionally, to test the inuence of long-term Ficin treatment, the car-
cinoma and stem cells were grown in the presence of Ficin samples over 3 days. Aer every 24 h the culture liquid
was removed from part of the wells and cells were live/dead stained and analyzed with dierential uorescence
microscopy using Carl Zeiss Observer 2.0 microscope. No signicant increase in the fraction of necrotic MCF7
Figure 8. Dose-response curves for biolm-embedded Staphylococci treated with antimicrobials in either
presence (green) or absence (blue) of Ficin (1000 μg/ml). Full lines denote regression lines, while dashed lines
denote corresponding 95% condence intervals.

www.nature.com/scientificreports/
9
Scientific RepoRts | 7:46068 | DOI: 10.1038/srep46068
or stem cells (see FigsS6andS7) was detected in either control wells or wells with Ficin at concentrations of
10–1000 μ g/ml, indicating Ficin safety for potential biomedical applications at least under conditions been tested.
Conclusion
Our results conrm that Ficin, a nonspecic sulydryl protease from Ficus tree, eectively disrupts the biolm
matrix backbone of S. aureus and S. epidermidis, which colonize skin, catheters and cause nosocomial infections.
e eciency of biolm disruption activity has been also conrmed using atomic force and uorescence micros-
copy of treated and non-treated biolms. As a result, the presence of protease led to at least twofold decrease
of antimicrobials (ciprooxacin and benzalkonium chloride) concentrations required to reduce the number of
viable biolm-embedded cells. Ficin is not cytotoxic, as we have conrmed using viability assays with adipose
derived stem cells and MCF7 carcinoma cells. Importantly, Ficin did not aect the growth rate and morphology of
either cell lines. We believe that Ficin appears a safe and eective agent for external wound treatment to suppress
the biolm formation and reduce the reinfection risk. Although the detailed investigation of the practical aspects
of wound healing with Ficin requires further thorough investigations, our current results indicate that Ficin is an
advantageous tool for therapeutic antibiolm treatment.
Materials and Methods
A commercially available Ficin obtained from MP Biomedicals, USA (0.2 U/mg) was used in this study.
Bacterial strains and growth conditions. Staphylococcus aureus subsp. aureus (ATCC® 29213™ ) and
Staphylococcus epidermidis (clinical isolate, obtained from the Kazan Institute of Epidemiology and Microbiology,
Kazan, Russia) were used for the biofilm assays. Bacterial strains were cultivated using LB medium. The
Müller-Hinton broth (Fluka) or Trypticase soy broth (Sigma) did not provide stable biolm formation by both
Staphylococcus aureus and Staphylococcus epidermidis, as is has been determined in preliminary studies (Fig.1),
thus the modied Basal medium (BM) (glucose 5 g, peptone 7 g, MgSO4 × 7H2O 2.0 g and CaCl2 × 2H2O 0.05 g in
1.0 liter tap water) was used for the biolm formation assays31,39. Bacteria were grown for 48–72 hours as indicated
under static conditions at 37 °C to obtain rigid biolm structures.
Biofilm staining. To investigate the effect of Ficin on bacterial biofilms, a bacterial suspension
(2–9 × 106 CFU ml−1) was inoculated in BM broth and grown in 96-well plates (200 μ l per well) or 34-mm plates
(2 ml per plate). All plates (polystyrol) were TC-treated and obtained from Eppendorf. Aer 72 h of growth the
biolm was formed, the old medium was exchanged by the new one, the Ficin was added and the incubation was
continued for the next 24 h. To analyze the hydrolysis of the protein backbone of the biolm matrix by Ficin, a
Congo Red solution40 (nal concentration 50 μg/ml) was added to the preformed biolm together with Ficin.
For crystal violet staining, the culture supernatant was discarded, and the wells were washed several times
with phosphate-buered saline (PBS) to remove non-adherent cells. e samples obtained were then stained with
crystal violet as described previously41. Briey, the plates were air dried for 20 min, and the surface-attached cells
were stained with 200 μ l of 1% crystal violet solution for 20 min. Subsequently, the crystal violet was removed and
the plates were washed 3 times with tap water. Aer 30 min of air drying, 200 μ l or 2 ml of 96% ethanol was added
to dissolve the cell-bound crystal violet, and the absorbance was measured at 570 nm using the microplate reader
Tecan Innite 200 Pro. e wells incubated with the cell-free medium were also stained and used as a reference.
Evaluation of antibacterial activity. e minimum inhibitory concentration (MIC) of antimicrobials
was determined by the broth microdilution method in BM broth in 96-well non-treated cell culture plates
(Eppendorf) in three independent repeats. e concentrations of antibiotic aer a series of two-fold dilutions
were in the range of 0.5–512 μ g/ml. Wells were seeded with 200 μ l of the bacterial culture (3 × 107 CFU/ml) and
incubated at 37 °C. e MIC was determined as the lowest concentration of compound for which no visible bac-
terial growth could be observed aer 24 h of incubation. To determine the minimum bactericidal concentration
(MBC), 5 μ l of culture liquid from the wells with no visible growth were inoculated into 5 ml of LB broth and cul-
tivated for 24 h. e MBC was determined as the lowest concentration of compound for which no visible bacterial
growth could be observed.
To evaluate the antibacterial activity against biolm-embedded cells, rigid biolms were preformed by 48 h
growth in BM broth as indicated, the plates were washed twice with sterile broth, followed by the exchange of the
old medium by the new one. Ficin and antibiotics were added as indicated and the incubation was continued for
the next 24 h. e viability of both biolm-embedded and biolm-detached cells in culture liquid was investigated
by both drop plate approach and CLSM.
Drop plate assay. To evaluate the viability of detached and planktonic cells with drop plate assay, a series
of 10-fold dilutions of liquid culture from each well were prepared in 3 technical repeats and 50 μ l of suspension
was dropped by 10 μ l-drops onto LB plates42. CFUs were counted from the two last drops containing countable
amount of colonies and averaged. To evaluate the viability of biolm-embedded cells, the wells were washed twice
Final concentration
of Ficin, μg/ml
MCF7 cells Dog adipose derived stem cells
10 100 1000 10 100 1000
Residual activity of
dehydrogenase, % 122 ± 12.3 83 ± 12.5 105 ± 7.9 98 ± 0.21 90 ± 0.21 83 ± 0.21
Table 1. Cytotoxicity of Ficin in metabolic MTS test (residual activity, percentage of the solvent control).

www.nature.com/scientificreports/
10
Scientific RepoRts | 7:46068 | DOI: 10.1038/srep46068
with 0.9% NaCl to remove the non-adherent cells, and the biolms were suspended in 0.9% NaCl by scratching
the well bottoms with subsequent treatment in an ultrasonic bath for 2 min to facilitate the disintegration of bac-
terial clumps. Viable cells were counted by the drop plate method as described above.
Biolm assay with CLSM. To evaluate the viability of biolm-embedded cells, bacterial suspension was
inoculated in BM broth and grown on cell imaging cover slips (Eppendorf) under static conditions. Aer 48 h
of growth, half of the medium was exchanged by the fresh medium. Next Ficin and antimicrobials were added
as described previously and further incubated for 24 h. e samples were then stained for 5 min with the 3,3′
-Dihexyloxacarbocyanine iodide (Sigma) at nal concentration of 0.02 μ g/ml (green uorescence) and propidium
iodide (Sigma) at nal concentration of 3 μ g/ml (red uorescence) to dierentiate between bacteria with intact
and damaged cell membranes (live and dead cells). Confocal laser scanning microscopy images (CLSM) were
obtained with a Carl Zeiss LSM 780 confocal microscope with Ζ -series images taken in 1-μ m slices.
Atomic force microscopy (AFM). Atomic force microscopy images of the air-dried microbial bio-
lms were collected using Dimension Icon scanning probe microscope (Bruker, USA) operating in PeakForce
Tap p i n g ™ mode. For AFM imaging in air the biolms were grown in BM-broth on 34-mm plates (TC-treated,
Eppendorf, 2 ml per plate) and treated with Ficin as described above. en the treated biolms were washed with
water and xed with glutaraldehyde (0.1% aqueous solution) for 4 hours. Aer subsequent washing with water
the plates were dried in air and imaged at ambient conditions. Scan Asyst-Air probes (Bruker) having nominal
length 115 μ m, tip radius 2 nm, spring constant 0.4 N\m were used throughout. e images were obtained at 512
lines\scan at 0.8–0.9 Hz scan rate. e images were acquired in height (topography), peak force error and adhe-
sion channels. e raw AFM imaging data obtained were processed and analysed using Nanoscope Analysis v.1.7
soware (Bruker).
Cytotoxicity assay. e MCF7 cells and dog adipose derived stem cells36 were cultured in DMEM sup-
plemented with 10% FBS, 2 mM L-glutamine, 100 μg/ml penicillin and 100 μg/ml streptomycin. e cells were
seeded in 96-well plates at the density of 3000 cells per well and allowed to attach overnight. e cells were cul-
tured at 37 °C and 5% CO2 in the presence of Ficin. Aer 48 h of cultivation the cells were subjected to MTS-assay
based on the cellular reduction of MTS (3-(4,5-dimethyl-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl
)-2H-tetrazolium) by the mitochondrial dehydroxygenase using phenazine methosulfate (PMS) as the electron
coupling reagent (Promega Cell Proliferation Assay kit). e MTS tetrazolium compound was bioreduced by the
viable cells into a colored formazan product which was measured using Tecan Innite 200Pro at 550 nm.
Statistical analysis. Experiments were carried out in biological triplicates (i.e. newly prepared cultures and
medium) with 3 independent repeats in each one. e fraction of not viable cells was estimated as the relative
fraction of the red cells among all cells in the combined images obtained by overlaying of the green and the red
uorescence microphotographs (10 images per each sample). e statistical signicance of the biolm destruc-
tion in the series of Ficin-treated samples was assessed using the Mann-Whitney U-test for independent samples
separately for each of three tested enzyme concentrations. Since the drop plate assay results were assessed from
10-fold dilutions, where typically only in the two latter dilutions the number of colonies was countable, to assess
the statistical signicance, we compared 10 log10(c), where c is the obtained cell number, using the Pearson’s
chi-squared homogeneity test. For both tests signicant dierences were reported at p < 0.05. Dose-response
curves were estimated by linear regression in the logarithmic scale for both × MBC and relative CFU counts, with
95% condence intervals for the regression coecients.
References
1. Yang, L. et al. Combating biolms. Fems Immunology and Medical Microbiology 65, 146–157, doi: 10.1111/j.1574-695X.2011.00858.x
(2012).
2. ichards, J. J. & Melander, C. Controlling Bacterial Biolms. Chembiochem 10, 2287–2294, doi: 10.1002/cbic.200900317 (2009).
3. Donlan, . M. Biolms: Microbial life on surfaces. Emerging Infectious Diseases 8, 881–890 (2002).
4. Worthington, . J., Blacledge, M. S. & Melander, C. Small-molecule inhibition of bacterial two-component systems to combat
antibiotic resistance and virulence. Future Medicinal Chemistry 5, 1265–1284, doi: 10.4155/fmc.13.58 (2013).
5. Barth, E., Myrvi, Q. M., Wagner, W. & Gristina, A. G. In vitro and In vivo comparative colonization of staphylococcus-aureus and
staphylococcus-epidermidis on orthopedic implant materials. Biomaterials 10, 325–328, doi: 10.1016/0142-9612(89)90073-2
(1989).
6. Blacledge, M. S., Worthington, . J. & Melander, C. Biologically inspired strategies for combating bacterial biolms. Current
Opinion in Pharmacology 13, 699–706, doi: 10.1016/j.coph.2013.07.004 (2013).
7. Costerton, J. W., Stewart, P. S. & Greenberg, E. P. Bacterial biofilms: A common cause of persistent infections. Science 284,
1318–1322, doi: 10.1126/science.284.5418.1318 (1999).
8. Gaddy, J. A. & Actis, L. A. egulation of Acinetobacter baumannii biolm formation. Future Microbiology 4, 273–278, doi: 10.2217/
fmb.09.5 (2009).
9. ogers, S. A., Huigens, . W., Cavanagh, J. & Melander, C. Synergistic Effects between Conventional Antibiotics and
2-Aminoimidazole-Derived Antibiolm Agents. Antimicrobial Agents and Chemotherapy 54, 2112–2118, doi: 10.1128/aac.01418-09
(2010).
10. Par, J. H., Lee, J. H., Cho, M. H., Herzberg, M. & Lee, J. Acceleration of protease eect on Staphylococcus aureus biolm dispersal.
Fems Microbiology Letters 335, 31–38, doi: 10.1111/j.1574-6968.2012.02635.x (2012).
11. aplan, J. B. Biolm Dispersal: Mechanisms, Clinical Implications, and Potential erapeutic Uses. Journal of Dental esearch 89,
205–218, doi: 10.1177/0022034509359403 (2010).
12. Izano, E. A., Amarante, M. A., her, W. B. & aplan, J. B. Dierential roles of poly-N-acetylglucosamine surface polysaccharide and
extracellular DNA in Staphylococcus aureus and Staphylococcus epidermidis biolms. Applied and Environmental Microbiology 74,
470–476, doi: 10.1128/aem.02073-07 (2008).
13. Donelli, G. et al. Synergistic activity of dispersin B and cefamandole nafate in inhibition of staphylococcal biolm growth on
polyurethanes. Antimicrobial Agents and Chemotherapy 51, 2733–2740, doi: 10.1128/aac.01249-06 (2007).

www.nature.com/scientificreports/
11
Scientific RepoRts | 7:46068 | DOI: 10.1038/srep46068
14. Darouiche, . O., Mansouri, M. D., Gawande, P. V. & Madhyastha, S. Antimicrobial and antibiolm ecacy of triclosan and
DispersinB () combination. Journal of Antimicrobial Chemotherapy 64, 88–93, doi: 10.1093/jac/dp158 (2009).
15. Alawash, M. A., Soothill, J. S. & Schiller, N. L. Alginate lyase enhances antibiotic illing of mucoid Pseudomonas aeruginosa in
biolms. Apmis 114, 131–138, doi: 10.1111/j.1600-0463.2006.apm_356.x (2006).
16. Alipour, M., Suntres, Z. E. & Omri, A. Importance of DNase and alginate lyase for enhancing free and liposome encapsulated
aminoglycoside activity against Pseudomonas aeruginosa. Journal of Antimicrobial Chemotherapy 64, 317–325, doi: 10.1093/jac/
dp165 (2009).
17. Bayer, A. S. et al. Eects of alginase on the natural-history and antibiotic-therapy of experimental endocarditis caused by mucoid
pseudomonas-aeruginosa. Infection and Immunity 60, 3979–3985 (1992).
18. Nijland, ., Hall, M. J. & Burgess, J. G. Dispersal of Biolms by Secreted, Matrix Degrading, Bacterial DNase. Plos One 5, doi:
10.1371/journal.pone.0015668 (2010).
19. Baer, P. et al. Exopolysaccharide biosynthetic glycoside hydrolases can be utilized to disrupt and prevent Pseudomonas aeruginosa
biolms. Science Advances 2, doi: 10.1126/sciadv.1501632 (2016).
20. Leroy, C., Delbarre-Ladrat, C., Ghillebaert, F., Compere, C. & Combes, D. Eects of commercial enzymes on the adhesion of a
marine biolm-forming bacterium. Biofouling 24, 11–22, doi: 10.1080/08927010701784912 (2008).
21. Sharafutdinov, I. et al. HtrA Protease from Bacillus subtilis Suppresses the Bacterial Fouling of the at Sin Injuries. Bionanoscience
6, 564–567, doi: 10.1007/s12668-016-0281-2 (2016).
22. Schallenberger, M. A., Niessen, S., Shao, C. X., Fowler, B. J. & omesberg, F. E. Type I Signal Peptidase and Protein Secretion in
Staphylococcus aureus. Journal of Bacteriology 194, 2677–2686, doi: 10.1128/jb.00064-12 (2012).
23. ristensen, J. B. et al. Antifouling enzymes and the biochemistry of marine settlement. Biotechnology Advances 26, 471–481, doi:
10.1016/j.biotechadv.2008.05.005 (2008).
24. Lequette, Y., Boels, G., Clarisse, M. & Faille, C. Using enzymes to remove biolms of bacterial isolates sampled in the food-industry.
Biofouling 26, 421–431, doi: 10.1080/08927011003699535 (2010).
25. Pavluhina, S. V. et al. Noneluting Enzymatic Antibiolm Coatings. Acs Applied Materials & Interfaces 4, 4708–4716, doi: 10.1021/
am3010847 (2012).
26. allinger, B., Prasetyo, E. N., Nyanhongo, G. S. & Guebitz, G. M. Antimicrobial enzymes: An emerging strategy to ght microbes
and microbial biolms. Biotechnology Journal 8, 97-+ , doi: 10.1002/biot.201200313 (2013).
27. Selan, L., Berlutti, F., Passariello, C., Comodiballanti, M. . & aller, M. C. Proteolytic-enzymes - a new treatment strategy for
prosthetic infections. Antimicrobial Agents and Chemotherapy 37, 2618–2621 (1993).
28. Harris, L. G., Nigam, Y., Sawyer, J., Mac, D. & Pritchard, D. I. Lucilia sericata Chymotrypsin Disrupts Protein Adhesin-Mediated
Staphylococcal Biolm Formation. Applied and Environmental Microbiology 79, 1393–1395, doi: 10.1128/aem.03689-12 (2013).
29. Longhi, C. et al. Protease treatment aects both invasion ability and biolm formation in Listeria monocytogenes. Microbial
Pathogenesis 45, 45–52, doi: 10.1016/j.micpath.2008.01.007 (2008).
30. urnheer, T. & Belibasais, G. N. Incorporation of staphylococci into titanium-grown biolms: an invitro “submucosal” biolm
model for peri-implantitis. Clinical Oral Implants esearch 27, 890–895, doi: 10.1111/clr.12715 (2016).
31. ayumov, A. . et al. Inhibition of biolm formation in Bacillus subtilis by new halogenated furanones. Journal of Antibiotics 68,
297–301, doi: 10.1038/ja.2014.143 (2015).
32. ayumov, A. . et al. New Derivatives of Pyridoxine Exhibit High Antibacterial Activity against Biolm-Embedded Staphylococcus
Cells. Biomed esearch International, doi: 10.1155/2015/890968 (2015).
33. Braun, M., Zaremba, W. & ietzmann, M. Enhancement of wound healing by an ointment containing proteolytic enzymes and
vitamins. Pratische Tierarzt 88, 110–114 (2007).
34. Chambers, L. et al. Degradation of extracellular matrix components by dened proteinases from the greenbottle larva Lucilia
sericata used for the clinical debridement of non-healing wounds. British Journal of Dermatology 148, 14–23, doi:
10.1046/j.1365-2133.2003.04935.x (2003).
35. Hogan, S., Stevens, N. T., Humphreys, H., O’Gara, J. P. & O’Neill, E. Current and Future Approaches to the Prevention and Treatment
of Staphylococcal Medical Device-elated Infections. Current Pharmaceutical Design 21, 100–113 (2015).
36. Trizna, E. et al. Soluble and immobilized papain and trypsin as destroyers of bacterial biolms. Genes and Cells 10, 106–112 (2016).
37. Sabirova, A. . et al. A novel secreted metzincin metalloproteinase from Bacillus intermedius. Febs Letters 584, 4419–4425, doi:
10.1016/j.febslet.2010.09.049 (2010).
38. Trizna, E. Y. et al. io Derivatives of 2(5H)-Furanone As Inhibitors against Bacillus subtilis Biolms. Acta Naturae 7, 102–107
(2015).
39. Trizna, E., Latypova, L., urbangalieva, A., Bogachev, M. I. & ayumov, A. 2(5H)-Furanone Derivatives as Inhibitors of
Staphylococcal Biolms. Bionanoscience 6, 423–426, doi: 10.1007/s12668-016-0258-1 (2016).
40. Baugh, S., Phillips, C. ., Eanayaa, A. S., Piddoc, L. J. V. & Webber, M. A. Inhibition of multidrug eux as a strategy to prevent
biolm formation. Journal of Antimicrobial Chemotherapy 69, 673–681, doi: 10.1093/jac/dt420 (2014).
41. O’To ole, G. A., & olter, . Initiation of biolm formation in Pseudomonas uorescens WCS365 proceeds via multiple, convergent
signalling pathways: a genetic analysis. Molecular Microbiology. 28, 449–461, doi: 10.1046/j.1365-2958.1998.00797.x (1998).
42. Herigstad, B., Hamilton, M. & Heersin, J. How to optimize the drop plate method for enumerating bacteria. Journal of
Microbiological Methods. 44, 121–129, doi: 10.1016/s0167-7012(00)00241-4 (2001).
Acknowledgements
Prof. Dr. Albert Rizvanov is gratefully acknowledged for providing the eukaryotic cell lines. is work has
been supported by the Russian Science Foundation (project No. 15-14-00046), performed in the framework
of the Program of competitive development of Kazan Federal University and using the equipment of the
Interdisciplinary shared use center at Kazan Federal University. e uorescent images were analyzed using in-
house soware developed at Saint-Petersburg Electrotechnical University as a part of the basic state assignment
by the Ministry of Education and Science of the Russian Federation.
Author Contributions
D.B., E.T., M.H., F.A. and E.R. perfomed the experiments, A.K. and R.F. conducted the experiment(s), M.B., A.K.
and V.A. analyzed the results. All authors reviewed the manuscript.
Additional Information
Supplementary information accompanies this paper at http://www.nature.com/srep
Competing Interests: e authors declare no competing nancial interests.

www.nature.com/scientificreports/
12
Scientific RepoRts | 7:46068 | DOI: 10.1038/srep46068
How to cite this article: Baidamshina, D. R. et al. Targeting microbial biolms using cin, a nonspecic plant
protease. Sci. Rep. 7, 46068; doi: 10.1038/srep46068 (2017).
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional aliations.
is work is licensed under a Creative Commons Attribution 4.0 International License. e images
or other third party material in this article are included in the article’s Creative Commons license,
unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license,
users will need to obtain permission from the license holder to reproduce the material. To view a copy of this
license, visit http://creativecommons.org/licenses/by/4.0/
© e Author(s) 2017