In Vitro Bactericidal and Bacteriolytic Activity of

Ceragenin CSA-13 against Planktonic Cultures and

Biofilms of Streptococcus pneumoniae and Other
Pathogenic Streptococci
Miriam Moscoso1,3., Marı́a Esteban-Torres1,3.¤, Margarita Menéndez2,3, Ernesto Garcı́a1,3*
1 Centro de Investigaciones Biológicas, CSIC, Madrid, Spain, 2 Departamento de Quı́mica-Fı́sica Biológica, Instituto Quı́mica-Fı́sica Rocasolano, CSIC, Madrid, Spain, 3 CIBER
de Enfermedades Respiratorias (CIBERES), Mallorca, Illes Balears, Spain

Abstract
Ceragenin CSA-13, a cationic steroid, is here reported to show a concentration-dependent bactericidal/bacteriolytic activity
against pathogenic streptococci, including multidrug-resistant Streptococcus pneumoniae. The autolysis promoted by CSA-
13 in pneumococcal cultures appears to be due to the triggering of the major S. pneumoniae autolysin LytA, an N-
acetylmuramoyl-L-alanine amidase. CSA-13 also disintegrated pneumococcal biofilms in a very efficient manner, although at
concentrations slightly higher than those required for bactericidal activity on planktonic bacteria. CSA-13 has little hemolytic
activity which should allow testing its antibacterial efficacy in animal models.

Citation: Moscoso M, Esteban-Torres M, Menéndez M, Garcı́a E (2014) In Vitro Bactericidal and Bacteriolytic Activity of Ceragenin CSA-13 against Planktonic
Cultures and Biofilms of Streptococcus pneumoniae and Other Pathogenic Streptococci. PLoS ONE 9(7): e101037. doi:10.1371/journal.pone.0101037
Editor: Herminia de Lencastre, Rockefeller University, United States of America
Received February 21, 2014; Accepted June 2, 2014; Published July 9, 2014
Copyright: ß 2014 Moscoso et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by Dirección General de Investigación Cientı́fica y Técnica (grant numbers SAF2009-10824, BFU2009-10052, BFU2012-36825,
and SAF2012-39444-C02-01) and Centro de Investigación Biomédica en Red de Enfermedades Respiratorias (CIBERES), which is an initiative of the ISCIII. The
funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* Email: [email protected]
. These authors contributed equally to this work.
¤ Current address: Laboratorio de Biotecnologı́a Bacteriana, Instituto de Ciencia y Tecnologı́a de Alimentos y Nutrición, ICTAN-CSIC, Madrid, Spain

Introduction lifetimes of antimicrobial peptides. Under physiological conditions,

ceragenins are polycationic; they therefore associate strongly with
Despite advances in medicine, infectious diseases remain a anionic bacterial cell surfaces, displacing patches of the membrane
major cause of death and inflict social and economic upheaval on [6]. This causes a loss of membrane integrity along with changes in
millions of people worlwide. Respiratory infections alone are its permeability, eventually leading to the death of affected cells.
responsible for 4 million deaths every year, with Streptococcus Ceragenins show strong bactericidal activity against Gram-
pneumoniae the predominant causal agent [1]. Indeed, it is the negative and Gram-positive bacteria (mainly enterococci and
leading cause of non-invasive infections such as bacterial staphylococci), and there is little likelihood that natural resistance
pneumonia, sinusitis and otitis media (all biofilm-related infections can be generated against them. It has also been reported that
[2]), and is commonly the culprit in life-threatening invasive ceragenins potentiate the effect of certain antibiotics on planktonic
conditions such as bacteremia/sepsis and meningitis. In develop- cells, as well as on mature biofilms produced by a number of
ing countries, pneumococcal septicemia is a major cause of infant Pseudomonas aeruginosa strains [7].
mortality, leading to the deaths of more than 1.2 million infants The present work examines the antimicrobial activity of the
every year [3]. HIV infection, sickle-cell anemia, and a variety of most potent ceragenin, CSA-13 (Fig. 1), against planktonic
chronic organ failure conditions, increase the risk of serious cultures and biofilms of S. pneumoniae and other pathogenic
pneumococcal disease. streptococci.
The alarming, global increase in multidrug resistance, partic-
ularly in pneumococcal strains resistant to b-lactams, macrolides, Materials and Methods
tetracyclines and sulfonamides, has renewed interest in the
development of novel drugs and strategies for controlling the Ethics statement
spread of antibiotic resistant bacteria [4]. Ceragenins, also known The study was submitted to the Consejo Superior de
as CSAs (for Cationic Steroid Antibiotics), are based on a cholic Investigaciones Cientı́ficas (CSIC) Ethics Committee which
acid scaffolding, and mimic antimicrobial peptides [5]. Ceragen- declared it to be carried out with the appropriate ethical
ins, however, are easier to synthesize than the latter and are much safeguards.
more stable. Further, since they are not peptide-based, they are
free from breakdown by proteases, a phenomenon that limits the

PLOS ONE | www.plosone.org 1 July 2014 | Volume 9 | Issue 7 | e101037

 Anti-Streptococcal Activity of Ceragenin CSA-13

Figure 1. Structure of the ceragenin CSA-13.

doi:10.1371/journal.pone.0101037.g001

Bacterial strains and growth conditions characteristics and their ability to penetrate viable bacterial cells.
The streptococcal strains used in this study are listed in Table 1. SYTO 9 stains all cells green, while propidium iodide penetrates
These were routinely grown in C medium [8] supplemented (or those with a damaged cell membrane, staining them red.
not) with 0.08% yeast extract (C+Y medium), at 37uC without
shaking, or on reconstituted tryptose blood agar base plates (Difco Biofilm formation assay
Laboratories) supplemented with 5% defibrinated sheep blood. The conditions adopted for biofilm formation by pneumococcal
Escherichia coli BL21(DE3) (pMMN1) [9] was grown in Luria- cells and other streptococci in 96-well polystyrene microtiter plates
Bertani (LB) medium with ampicillin (100 mg/ml) [10] at 37uC (Costar 3595, Corning Inc.) were those previously described [14].
with aeration, as a source of the major pneumococcal autolysin Cells in late exponential growth were diluted in fresh C medium
LytA, an N-acetylmuramoyl-L-alanine amidase (NAM-amidase; and about 4.56106 CFU were dispensed into each well. After 6 h
EC 3.5.1.28) [11]. of incubation at 34uC, the biofilm formed was stained with 0.2%
crystal violet for 15 min and rinsed to remove non-adherent
Transformation procedures bacteria. After solubilizing the biofilm in 95% ethanol, the A595
S. pneumoniae R6 was transformed with chromosomal DNA from was determined using an Anthos 2020 microplate absorbance
the pneumococcal R924 strain (lytA::aphIII) [12] by treating reader (Anthos Labtec Instruments).
precompetent cells with 100 ng/ml of synthetic competence- For observation of the biofilms by confocal laser scanning
stimulating peptide 1 (CSP1), as previously described [13]. microscopy (CLSM), bacteria (about 4.56104 CFU) were grown
Transformants were selected by plating in CAT agar supplement- on glass-bottom dishes (WillCo-dish, WillCo Wells B.V.) for 12 h
ed with 3% defibrinated sheep blood, followed by challenge with a at 34uC and then stained with the LIVE/DEAD BacLight
10 ml overlay containing kanamycin (250 mg/ml) [14]. bacterial viability kit. The biofilms were observed at 6100
magnification using a Leica TCS-SP2-AOBS-UV CLSM
Antibacterial susceptibility assays equipped with an argon ion laser. The excitation/emission
The antibacterial activity of CSA-13, kindly provided by P.B. maxima were around 480/500 and 490/653 nm for SYTO 9
Savage (Department of Chemistry and Biochemistry, Brigham and PI respectively. Images were analyzed using LCS software
Young University, Provo, Utah, USA), was determined by (Leica). Projections were obtained in the x–y (scans at 0.5 mm
standard macrodilution methods according to the Clinical and intervals) and x–z planes (scans at 3 mm intervals).
Laboratory Standards Institute guidelines using an inoculum of
<46105 CFU/ml [15]. The MIC was defined as the lowest Expression and purification of NAM-amidase LytA
concentration of CSA-13 that prevented visible growth after 24 h Escherichia coli BL21(DE3) (pMMN1) was incubated in LB
of culture in Mueller-Hinton broth supplemented with 2.5% lysed medium containing ampicillin (100 mg/ml) and 0.4 mM isopro-
horse blood (Oxoid), or C medium, at 37uC. pyl-b-D-thiogalactopyranoside for the overproduction of LytA,
following the previously described protocol [9]. After bacterial
Time-kill experiments disruption in a French pressure cell press, the insoluble fraction
Unless otherwise stated, exponentially growing cultures of the was separated by ultracentrifugation (100,0006g, 1 h, 4uC), and
streptococcal strains were incubated in C+Y medium to an the supernatant loaded into a DEAE-cellulose column to purify
absorbance at 550 nm (A550) of about 0.2 corresponding to the LytA protein in a single step [16]. The purified enzyme was
approximately 1–26108 CFU/ml, depending on the streptococcal dialyzed against 20 mM sodium phosphate buffer, pH 6.9, and the
strain. CSA-13 was then added at different concentrations to protein concentration determined spectrophotometrically.
aliquots of the culture, and incubation continued without shaking
at 37uC. Growth and lysis were monitored via A550 values. The Enzymatic assay of LytA activity using radioactively
viability of a bacterial population was assessed by colony counting labeled cell walls
on blood agar plates and/or visualized by fluorescence microsco- Pneumococcal cell walls were radioactively labeled with
py, using the LIVE/DEAD BacLight bacterial viability kit [methyl-3H] choline as previously described [17]. Cell wall
(Invitrogen-Molecular Probes) to stain cells over a 15 min period degradation assays were performed according to standard
in the dark. Bacteria were observed at 640 and 6100 procedures [18], measuring the amount of radioactivity released
magnifications using a Leica DM4000B fluorescence microscope into the supernatant. One unit (U) of enzymatic activity was
equipped with L5 (bandpass 480/40) and N2.1 (bandpass 515– defined as the amount of enzyme needed to release 1 mg (about
560) filter sets for SYTO 9 and propidium iodide (PI). Both stains 700 net cpm) of labeled cell wall material in 10 min at 37uC.
are nucleic acid-binding agents but they differ in their spectral

PLOS ONE | www.plosone.org 2 July 2014 | Volume 9 | Issue 7 | e101037

 Anti-Streptococcal Activity of Ceragenin CSA-13

Table 1. Streptococcal strains used in the present work, and associated CSA-13 MIC values.

Relevant characteristics
Bacterial species/strainsa (MIC; mg/ml)b CSA-13 susceptibility Reference/Origind
c
(MIC; mg/ml)

Mitis group
Streptococcus pneumoniae
R6 Unencapsulated laboratory strain, lytA+ 4 (1) [57]
P103 Unencapsulated laboratory strain, 4 (1) This study
lytA::aphIII
Spain23F-1 Serotype 23F, lytA+ 4 (1) [58]
(PEN, 1–2; TET, 8; CHL, .8)
8249 Serotype 19A, lytA+ 4 (1) [59]
(PEN, 6; IPM, 0.35; VAN, 0.4)
S3 Serotype 23F, DlytA 8 [60]
(PEN, 8; CTX, .16; VAN, 0.5)
SPC2162 Serotype 19A, lytA+ 8 [61]
(PEN, 16)
SPC2552 Serotype 23F, lytA+ 4 [61]
(PEN, 32)
ST942 Non-typeable, lytA+ 8 [24,62]
Streptococcus sp. strain lytA+ (1) [37]
782/96 (PEN, 0.25; TET, 128; ERY, .128)
Streptococcus sp. strain lytA+ (1) [37]
11923/96 (PEN, 8; TET, 0.5; ERY, 4)
Streptococcus gordoniiT Type strain (2) CECT
Streptococcus mitis NCTC 12161T Type strain (2) NCTC
Streptococcus oralis NCTC Type strain (1) NCTC
11427T
Streptococcus oralisT Type strain, lytA+ (1) [27]
(pLSE5)
Streptococcus pseudopneumoniae CCUG Type strain, lytA+ (1) [63], CCUG
49455T
Streptococcus sanguinis Type strain (2) CECT
CECT 480T
Mutans group
Streptococcus mutans Type strain (1) CECT
CECT 479T
Pyogenic group
Streptococcus agalactiae Type strain (1) CECT
CECT 183T
Streptococcus pyogenes Type strain (1) CECT
CECT 985T

aT
, type strain.
b
CHL, chloramphenicol; CTX, cefotaxime; ERY, erythromycin; IPM, imipenem; PEN, penicillin; TET, tetracycline; VAN, vancomycin.
c
MICs in C medium are indicated in parentheses.
d
CCUG, Culture Collection, University of Göteborg; CECT, Colección Española de Cultivos Tipo; NCTC, National Collection of Type Cultures.
doi:10.1371/journal.pone.0101037.t001

Hemolytic activity assays 8006g for 5 min; the supernatant was then removed and the
Defibrinated sheep blood (from Biomedics or Oxoid) or A550 determined. The percentage of hemolytic activity of the
human blood from a volunteer (EG) was used in hemolytic drug at different concentrations was estimated as (A2A0/Amax2
activity assays. The blood (5 ml) was centrifuged and the cells A0)6100, where A0 is the absorbance associated with the
washed thoroughly with phosphate-buffered saline (PBS). Cells background hemolysis occurring during incubation with PBS,
were collected by centrifugation at 5,000 rpm for 10 min in a and Amax the absorbance at 100% hemolysis after incubation in
Sorvall SS-34 rotor, resuspended in 5 ml of PBS, and stored at distilled water (for sheep blood) or 0.01% Triton X-100 (for
4uC. The number of red blood cells (RBC) was adjusted to human blood) [19].
ensure total lysis with distilled water. Then, 0.6 ml of RBC
suspension in PBS was mixed with 0.6 ml of CSA-13 solution Circular dichroism spectra
using final drug concentrations proportional to the MIC. The Circular dichroism (CD) spectra were recorded at 20uC
mixtures were incubated at 37uC for 1 h and centrifuged at using a J-810 spectropolarimeter (Jasco Corporation) equipped

PLOS ONE | www.plosone.org 3 July 2014 | Volume 9 | Issue 7 | e101037

 Anti-Streptococcal Activity of Ceragenin CSA-13

Figure 2. Effect of CSA-13 on streptococcal species. Exponentially growing cultures were incubated in C+Y to an A550 of about 0.2. CSA-13 was
then added to aliquots of the cultures and incubation continued without shaking at 37uC. (a) S. pneumoniae R6. (b) S. pneumoniae Spain23F-1. (c)
Streptococcus sp. strain 11923/96. (d) S. pseudopneumoniae. (e) S. gordonii. (f) S. mitis. (g) S. sanguinis. (h) S. mutans. (i) S. agalactiae. (j) S. pyogenes.
Solid circles represent untreated, control cultures. CSA-13 was added (open symbols) at the concentrations: triangles, 1 mg/ml; circles, 10 mg/ml;
squares, 25 mg/ml. (k) Lytic effect of CSA-13 (10 mg/ml; open symbols) added to a culture of the pneumococcal strain R6 at the times indicated by the
arrows. (l) Bactericidal effect of CSA-13 on different streptococcal species (initial A550 corresponding to 1–26108 CFU/ml). Grey, dashed and blackened
bars correspond, respectively, to bacterial survival after 0, 2 h, and 6 h incubation in the presence (+) or absence (–) of the ceragenin (10 mg/ml).
Standard error bars are shown.
doi:10.1371/journal.pone.0101037.g002

with a Peltier holder. Far UV spectra were recorded in Results

0.1 cm-pathlength cells at a protein concentration of 0.16 mg/
ml [20]. The observed ellipticities were converted to mean Determination of CSA-13 MICs for the streptococcal
residue ellipticities [h] assuming a mean molecular mass per strains
residue of 114.8 Da after subtraction of the buffer or the ligand Planktonic cultures of eight pneumococcal strains (including
spectra. the R6 unencapsulated laboratory strain and multidrug-resistant
clinical isolates) and of several primary or opportunistic
Statistical analyses pathogens (Table 1), were tested for their sensitivity to CSA-13.
The two-tailed Student t test, performed using GraphPad InStat MICs were determined in Mueller-Hinton broth supplemented
version 3.0 software (GraphPad Software, San Diego, CA), was with 2.5% lysed horse blood, and in C medium in the absence of
used to determine the differences between means. Data are such blood since some authors have reported that certain
representative of results obtained from at least three independent antimicrobial agents, such as miltefosine, bind to serum
experiments. components, reducing the effective concentration [21,22]. CSA-
13 showed potent antimicrobial activity against all the strepto-
coccal strains tested. MICs of 1–2 mg/ml were recorded in the C

PLOS ONE | www.plosone.org 4 July 2014 | Volume 9 | Issue 7 | e101037

 Anti-Streptococcal Activity of Ceragenin CSA-13

Figure 3. Fluorescence microscopy images of S. pneumoniae R6 cells either untreated (panels a–c) or treated with CSA-13 (100 mg/
ml; 3 h) (panels d–f) and stained with the BacLight bacterial viability kit. Live and non-viable bacteria fluoresce green (b, e) and red (c, f)
respectively. Panels a and d are phase contrast micrographs. The arrows indicate cells treated with CSA-13 and lacking any cytoplasmic content. Panel
d (enlarged inset) shows four apparently empty cocci. Some cellular material appears to be undergoing release from two bacteria (arrowheads).
doi:10.1371/journal.pone.0101037.g003

medium tests, and of up to 8 mg/ml in the lysed blood-containing The difference between the MIC and the concentration of
medium (Table 1). ceragenin with in vitro activity can be explained in that the efficacy
of the antibiotic depends inversely on the initial density of
Bactericidal and lytic effects of CSA-13 on streptococcal microorganisms [23]. The multidrug-resistant Spain23F-1 strain
cultures also underwent lysis in the presence of CSA-13, even at
Time-kill experiments were performed with CSA-13 and concentration of 2.56MIC (lower than that causing lysis in the
various streptococcal species. CSA-13 triggered rapid pneumo- R6 strain). Interestingly, S. pseudopneumoniae and Streptococcus sp.
coccal lysis at concentrations of $5 mg/ml when added during the 11923/96, which synthesize a partly active LytA-like autolysin
early exponential phase of growth (Figs. 2a and b). Interestingly, a [24], also lysed rapidly in response to CSA-13 (Figs. 2c and d).
lytic effect was recorded in S. pneumoniae R6 when the drug was Other streptococci that underwent lysis after exposure to CSA-13
added during the mid-exponential and stationary phases (Fig. 2k). (10 mg/ml) were S. gordonii, S. mutans and S. agalactiae. Moreover,

Figure 4. Growth and lysis curves and survival of S. oralis (a, b) and S. oralis (pLSE5) expressing LytA (c, d) treated with CSA-13. Cells
were grown in C+Y at 37uC to an A550 of about 0.2. CSA-13 was then added to an aliquot of the culture and incubation continued without shaking at
37uC. Solid circles represent untreated control cultures. CSA-13 was added (open symbols) at 2.5 mg/ml (triangles) or 10 mg/ml (circles). For panels b
and d, grey, dashed and blackened bars correspond, respectively, to bacteria survival after 0, 2 h, and 6 h incubation in the presence (+) or absence (–)
of the ceragenin (10 mg/ml). Standard error bars are shown.
doi:10.1371/journal.pone.0101037.g004

PLOS ONE | www.plosone.org 5 July 2014 | Volume 9 | Issue 7 | e101037

 Anti-Streptococcal Activity of Ceragenin CSA-13

the growth of S. mutans and S. agalactiae was slowed at 1 mg/ml of

CSA-13, and the cultures lysed with concentrations ranging
between 2.5 and 10 mg/ml. S. mitis, S. sanguinis and S. pyogenes
stopped growing when treated with CSA-13, but did not lyse
(Figs. 2f, g and j). Interestingly, even in the absence of lysis, CSA-
13 showed notable bactericidal activity, causing a fall in bacterial
survival of up to 4–5-log units after 6 h incubation with 10 mg/ml
(Fig. 2l).
It was noted that bacterial growth stopped but lysis did not
occur when pneumococcal cultures were treated with concentra-
tions of CSA-13 $25 mg/ml. Nevertheless, micrographs of strains
treated with 100 mg/ml CSA-13 showed only non-viable bacteria
represented by cocci that apparently lacked intracellular content
(Fig. 3d–f).

CSA-13 triggers LytA activity

Bile (or sodium deoxycholate) solubility is one of the classic tests
for distinguishing S. pneumoniae from all other a-hemolytic
streptococci [25]. NAM-amidase LytA is responsible for the
solubilization of pneumococci by deoxycholate [26]. To determine
whether NAM-amidase LytA is involved in the lytic effect of CSA-
13, cultures of S. oralis (pLSE5) expressing the pneumococcal LytA
autolysin [27], and S. oralis, which lacks the lytA gene, were
incubated with CSA-13 in the early exponential phase of growth.
S. oralis (pLSE5) underwent lysis after exposure to $2.5 mg/ml, a
response that contrasted to that shown for S. oralis (Figs. 4a and c).
Moreover, when treated with 10 mg/ml CSA-13 over a 6 h period,
S. oralis (pLSE5) showed a viable count reduction of 3-logs
compared to a 2-log reduction for S. oralis (which lacks lytA)
(Figs. 4b and d). These results suggest that LytA may be
responsible for as much as 1-log unit of the cell killing observed.
The lytA mutant strain P103, which underwent no lysis in the
stationary phase of growth either in the absence or presence of the
CSA-13, did undergo lysis after the addition of purified LytA
(10 mg/ml) and then CSA-13 (5 mg/ml) (Fig. 5a) –i.e., it was
‘‘cured’’ [28] to the wild-type phenotype. This provides further
evidence that the lysis events observed were LytA-mediated. In
addition, a rapid loss of viability was observed when the P103
strain was incubated with CSA-13 (a ca. 4-log reduction in viable
cells after 3 h of treatment), even in the absence of any detectable
autolysis (Fig. 5a).
The inhibitory effect of CSA-13 on the lysis of pneumococci
when at concentrations $25 mg/ml may be due to a direct (or
indirect) inhibitory effect of the drug on LytA activity. In vitro
experiments showed that ceragenin caused a dose-dependent
inhibition of NAM-amidase LytA activity (Fig. 5b). A reduction in
LytA activity of 85% was observed at 25 mg/ml CSA-13. CD
spectroscopy showed that a clear structural alteration of LytA takes
place upon incubation with 50 mg/ml CSA-13 (Fig. 5c). In the far-
Figure 5. Triggering of the LytA autolysin by CSA-13. (a) An UV region of the LytA spectrum, two negative bands were seen at
exponentially growing of S. pneumoniae strain P103 (lytA::aphIII) was 210 and 230 nm, a known spectral characteristic [29]. The low
incubated in C+Y medium with (open symbols) or without (solid absolute value of the ellipticity at 210 nm, and the negative band
symbols) pure LytA enzyme (10 mg/ml) for 30 min at 37uC. The culture at 230 nm, reflect a high contribution of aromatic amino acid side
was then diluted to an A550 of 0.2 and divided into two portions. CSA-13
was added at 5 mg/ml (diamonds) to one while the other was left chains. Moreover, choline addition to LytA (Fig. 5c) produces a
untreated (circles). Incubation was continued at 37uC. Survival of the positive band at 224 nm, a red shift in the small negative band at
culture treated only with CSA-13 was determined by plating at different 230 nm, and a slight increase in the negative ellipticity at 210 nm
incubation times (dotted line). (b) Effect of CSA-13 on the activity of cell [29]. Interestingly, the changes in the CD spectrum of LytA
wall hydrolase LytA using radioactively labeled pneumococcal cell walls incubated with CSA-13 were partially reverted by choline
as substrate. Data represent the percentage activity of LytA in the
absence of CSA-13 and are the means of three independent chloride, indicating that the ceragenin mainly affects the aromatic
experiments. (c) CD spectra of LytA in the far-UV region in the absence acid-rich, C-terminal domain of the NAM-amidase responsible for
and presence of CSA-13 (50 mg/ml) and/or 48 mM choline chloride cell wall binding [11]. These results taken together are in
(cho). accordance with the inhibition of lysis in pneumococcal cultures
doi:10.1371/journal.pone.0101037.g005 treated with high concentrations of CSA-13 (Fig. 2).

PLOS ONE | www.plosone.org 6 July 2014 | Volume 9 | Issue 7 | e101037

 Anti-Streptococcal Activity of Ceragenin CSA-13

Figure 6. Disaggregation of biofilms in the presence of CSA-13. S. pneumoniae R6 (a) and P103 (lytA::aphIII) (b) were inoculated into 200 ml of
C medium in the wells of a microtiter plate (4.56106 CFU/ml). After biofilm development (6 h at 34uC), CSA-13 was added at different concentrations,
and incubation allowed to proceed for 1 h at 34uC before staining with crystal violet to quantify biofilm formation (open bars). Percentage viability
(solid lines) after treatment with CSA-13 was determined by plating on blood agar plates. Standard error bars are shown. Asterisks indicate a P value,
0.05.
doi:10.1371/journal.pone.0101037.g006

Effect of CSA-13 on biofilm formation by S. pneumoniae Hemolytic activity of CSA-13

To test the capacity of CSA-13 to destroy pneumococcal CSA-13 showed relatively little hemolytic activity at concentra-
biofilms, the percentage of biofilm remaining after treatment with tions equivalent to the MIC in most of the streptococci examined
different concentrations of CSA-13 above the MIC for 1 h at 34uC (Table 2). Indeed, less than 10% hemolysis was recorded in human
was determined. At concentrations of $10 mg/ml, CSA-13 blood at CSA-13 concentrations of #10 mg/ml. Total permeabi-
effectively disintegrated the biofilms produced by S. pneumoniae lization of RBC was only achieved with $50 mg/ml CSA-13 (not
R6 (Fig. 6a). However, they had no apparent effect on the biofilm shown). All concentrations of CSA-13 below 50 mg/ml showed less
produced by the lytA mutant P103, as determined by crystal violet lytic activity against human RBC than against sheep RBC
staining, although it killed more than 95% of the pneumococci (Table 2).
present, as estimated by enumeration of viable counts (Fig. 6b).
The viability of pneumococcal biofilms in the presence of CSA-13 Discussion
was also examined by CLSM (Fig. 7). Biofilms of S. pneumoniae R6
grown on glass plates for 12 h at 34uC were treated with CSA-13 S. pneumoniae is a major human pathogen and a leading cause of
(5 or 25 mg/ml) for 90 min and the cells stained using the bacterial pneumonia, bacteremia and meningitis in adults, and of otitis
viability BacLight kit. A noticeable decrease in biofilm thickness media in children. Given the rapid dissemination of antibiotic
was observed after treatment with 5 mg/ml of CSA-13 (Fig. 7b), resistance in S. pneumoniae, along with other Gram-positive
and almost all biofilm cells were killed (i.e., they showed red pathogens quoted as ‘‘superbugs’’ [30], new therapeutic strategies
fluorescence) when a dose of 25 mg/ml was used (Fig. 7c). are needed to fight the increasing prevalence of multidrug-resistant

Figure 7. Confocal laser scanning microscopy image of the viability of biofilm-grown S. pneumoniae R6 in the presence of CSA-13.
Pneumococcal strains (4.56104 CFU) were grown on glass-bottom dishes (WillCo-dish) in 2 ml of C medium for 12 h at 34uC. The pneumococcal
biofilms were rinsed with C medium to remove non-adherent bacteria, and then incubated with CSA-13 at 5 mg/ml (b) or 25 mg/ml (c) for 90 min at
34uC. Panel a shows untreated control biofilm. Cells in the biofilms were stained with the BacLight bacterial viability kit to reveal living (green) and
dead (red) bacteria.
doi:10.1371/journal.pone.0101037.g007

PLOS ONE | www.plosone.org 7 July 2014 | Volume 9 | Issue 7 | e101037

 Anti-Streptococcal Activity of Ceragenin CSA-13

Table 2. Hemolytic activity of CSA-13 on sheep and human blood.

CSA-13
(mg/ml) % Hemolytic activitya

Sheep blood Human blood

0.05 7.4 0.0

0.1 8.6 0.0
0.5 13.4 0.1
1 21.1 0.3
2.5 32.9 1.4
5 41.7 3.1
10 52.7 9.0
25 58.7 20.1

a
Percentage hemolytic activity estimated as (A2A0/Amax2A0)6100, where A0 represents background hemolysis occurring during incubation with PBS, and Amax the
absorbance at 100% hemolysis after incubation in distilled water (for sheep blood) or 0.01% Triton X-100 (for human blood).
doi:10.1371/journal.pone.0101037.t002

bacterial infections. The bacterial membrane remains an interest- Undisrupted cocci apparently lacking intracellular content were
ing target since it represents a well-conserved structural element detected after treatment of pneumococcal cells with 100 mg/ml
across Gram-positive and indeed Gram-negative bacteria, and CSA-13 (Fig. 3d). Similar changes have previously been observed,
because resistance to drugs that attack the membrane would using transmission electron microscopy, in E. coli cells treated with
require major –and likely unviable– changes in membrane CSAs [41]. Changes indicating wall damage have also been
structure and composition [31]. In fact, CSA-13 resistance in reported in P. aeruginosa after treatment with CSA-13, as
Gram-negative bacteria only correlates with membrane modifica- determined by atomic force microscopy [42]. In agreement with
tions that are unstable in the absence of the drug, and in Gram- the latter report, CSA-13 was here confirmed to show relatively
positive Staphylococcus aureus no CSA-13 resistance has ever been little hemolytic activity against RBC at concentrations near the
seen, at least under in vitro conditions [32]. MIC for the present strains (Table 2). Cytotoxicity experiments
In this study, we have analyzed the susceptibility of S. pneumoniae have shown CSA-13 to be mildly toxic to fibroblasts at 100 mg/ml,
and other pathogenic streptococci to CSA-13, the most potent but not at 75 mg/ml [43]. Interestingly, the cytolytic activity of
member of the ceragenin class. For all the streptococci tested, the CSA-13 is greatly reduced in the presence of pluronic acid F-127;
MIC values were moderately higher in Mueller-Hinton broth consequently, it might be useful for combating streptococcal
supplemented with 2.5% horse blood than in C medium (Table 1). infections in this combination [34,42,44].
This is probably due to interaction with serum components The autolysis caused in pneumococcal cultures by CSA-13 (and
reducing the effective concentration of the drug in the former also in S. oralis cultures harboring the plasmid pLSE5 encoding
medium. Interestingly, CSA-13 showed comparable activity LytA) was due to the triggering of the LytA NAM-amidase, as
against the laboratory pneumococcal strain R6, the multi-resistant demonstrated by phenotypic ‘‘curing’’ experiments (Fig. 5a). The
clinical isolates as strains Spain23F-1 and 8249, the vancomycin- rapid lysis of pneumococcal strains (and other streptococcal
tolerant strain S3, and the highly b-lactam-resistant strains cultures producing a LytA-like autolysin) is very reminiscent of
SPC2162 and SPC2552 (Table 1). The bactericidal activity of the lytic response of S. pneumoniae to the detergent action of sodium
CSA-13 against clinical isolates of resistant S. aureus and P. deoxycholate (the bile solubility test) or miltefosine [22,25,26].
aeruginosa strains was reported in earlier studies [31,33,34]. The Since the mid 1970s it has been known that mild detergents, such
present results show the susceptibility of S. mutans and S. pyogenes to as sodium deoxycholate, or cell wall inhibitors such a b-lactams,
CSA-13 to be similar to that of S. mutans Ingbritt and other clinical fosfomycin and D-cycloserine, cause the release of lipoteichoic
isolates of these species [35,36]. acid from the bacterial cell membrane in a number of streptococci
CSA-13 showed a concentration-dependent activity against all [45,46]. Pneumococcal lipoteichoic acid is a specific inhibitor of
the pathogenic streptococci tested. In time-kill analyses, the the LytA NAM-amidase [47], and destabilization of the
pneumococcal strains, S. pseudopneumoniae and Streptococcus sp. lipoteichoic acid–LytA complex triggers bacterial autolysis [48].
11923/96 underwent autolysis in the presence of 2.5–20 mg/ml It should be noted that ceragenins and sodium deoxycholate are
CSA-13 (Figs. 2c, d). This is similar to that observed for cholic acid-derivatives, and a similar response to both agents might
miltefosine, an alkyllysophospholipid derivative used for the oral be anticipated [5]. Importantly, our results showed that CSA-13
treatment of visceral leishmaniasis [22]. These strains synthesize possesses bactericidal activity even when no bacteriolysis occurs.
an autolysin closely related to the pneumococcal autolytic NAM- This is not completely unexpected since previous work [49] has
amidase LytA but in contrast with the behavior of the ‘‘typical’’ S. shown that, although an enzymatically active LytA autolysin is
pneumoniae autolysin, these enzymes are inhibited by 1% sodium required for b-lactams-promoted lysis (see above), pneumococci
deoxycholate [22,37]. This lytic effect of CSA-13 was also are actually killed even though lysis fails to occur.
observed in S. gordonii, S. mutans and S. agalactiae, presumably Bacterial biofilms are implicated in chronic and persistent
caused by the triggering of bacterial and/or prophage peptido- infections [50] due to the protection they offer against antibiotics
glycan hydrolases [38–40]. In contrast, other streptococcal species and host immune defenses [51–53]. Biofilm formation by S.
(i.e., S. mitis, S. sanguinis, and S. pyogenes) did not lyse with CSA-13. pneumoniae on abiotic and biotic substrates (such as adenoid and
However, even when no bacterial lysis was observed, CSA-13 mucosal epithelia) has been documented [54]. Our results reveal
efficiently killed the streptococcal cells (Figs. 2, 4, and 5a). noticeable bactericidal activity by CSA-13 against pneumococcal

PLOS ONE | www.plosone.org 8 July 2014 | Volume 9 | Issue 7 | e101037

 Anti-Streptococcal Activity of Ceragenin CSA-13

cells growing in biofilms, a phenomenon associated with the Acknowledgments

disintegration of biofilm structure (Figs. 6a and 7). However,
concentrations slightly higher than those required for bactericidal We are indebted to P. B. Savage for providing ceragenin CSA-13. We
thank A. Burton for revising the English version. We also thank E. Cano
activity against planktonic bacteria were required to remove
for skillful technical assistance and M. Seisdedos for her help with CLSM
pneumococcal biofilms. Recent reports have shown that CSA-13 is imaging.
effective against biofilms formed by P. aeruginosa, Moraxella
catarrhalis, Helicobacter pylori, S. aureus and Enterococcus faecalis
[34,55,56].
Author Contributions
In summary, CSA-13 is an efficient antimicrobial agent against Conceived and designed the experiments: M. Moscoso ME-T M.
S. pneumoniae and other human streptococcal pathogens in vitro; it Menéndez EG. Performed the experiments: M. Moscoso ME-T M.
also destroys pneumococcal biofilms in a very efficient manner. Menéndez EG. Analyzed the data: M. Moscoso ME-T M. Menéndez EG.
Contributed reagents/materials/analysis tools: M. Moscoso ME-T M.
This molecule warrants further investigation as a therapeutic
Menéndez. Wrote the paper: M. Moscoso ME-T M. Menéndez EG.
weapon for use against streptococci, particularly multidrug-
resistant pneumococci and strains causing biofilm-related infec-
tions.

References
1. UNICEF WHO (2006) Pneumonia: the forgotten killer of children. Available: 22. Llull D, Rivas L, Garcı́a E (2007) In vitro bactericidal activity of the
http://www.unicef.org/spanish/publications/files/Pneumonia_The_ antiprotozoal drug miltefosine against Streptococcus pneumoniae and other
Forgotten_Killer_of_Children.pdf. Accessed 25 April 2014. pathogenic streptococci. Antimicrob Agents Chemother 51: 1844–1848.
2. Coates H, Thornton R, Langlands J, Filion P, Keil AD, et al. (2008) The role of 23. Udekwu KI, Parrish N, Ankomah P, Baquero F, Levin BR (2009) Functional
chronic infection in children with otitis media with effusion: evidence for relationship between bacterial cell density and the efficacy of antibiotics.
intracellular persistence of bacteria. Otolaryngol Head Neck Surg 138: 778–781. J Antimicrob Chemother 63: 745–757.
3. Kadioglu A, Weiser JN, Paton JC, Andrew PW (2008) The role of Streptococcus 24. Llull D, López R, Garcı́a E (2006) Characteristic signatures of the lytA gene
pneumoniae virulence factors in host respiratory colonization and disease. Nat Rev provide a rapid and reliable diagnosis of Streptococcus pneumoniae infections. J Clin
Microbiol 6: 288–301. Microbiol 44: 1250–1256.
4. Maestro B, Sanz JM (2007) Novel approaches to fight Streptococcus pneumoniae. 25. Lund E, Henrichsen J (1978) Laboratory diagnosis, serology and epidemiology
Recent Pat Antiinfect Drug Discov 2: 188–196. of Streptococcus pneumoniae. Methods Microbiol 12: 241–262.
5. Lai XZ, Feng Y, Pollard J, Chin JN, Rybak MJ, et al. (2008) Ceragenins: cholic 26. Mosser JL, Tomasz A (1970) Choline-containing teichoic acid as a structural
acid-based mimics of antimicrobial peptides. Acc Chem Res 41: 1233–1240. component of pneumococcal cell wall and its role in sensitivity to lysis by an
6. Epand RM, Epand RF, Savage PB (2008) Ceragenins (cationic steroid autolytic enzyme. J Biol Chem 245: 287–298.
compounds), a novel class of antimicrobial agents. Drug News Perspect 21: 27. Ronda C, Garcı́a JL, López R (1988) Characterization of genetic transformation
307–311. in Streptococcus oralis NCTC 11427: expression of the pneumococcal amidase in S.
7. Bozkurt-Guzel C, Savage PB, Gerceker AA (2011) In vitro activities of the novel oralis using a new shuttle vector. Mol Gen Genet 215: 53–57.
ceragenin CSA-13, alone or in combination with colistin, tobramycin, and 28. Tomasz A, Waks S (1975) Enzyme replacement in a bacterium: phenotypic
ciprofloxacin, against Pseudomonas aeruginosa strains isolated from cystic fibrosis correction by the experimental introduction of the wild type enzyme into a live
patients. Chemotherapy 57: 505–510. enzyme defective mutant pneumococcus. Biochem Biophys Res Commun 65:
8. Lacks S, Hotchkiss RD (1960) A study of the genetic material determining an 1311–1319.
enzyme activity in Pneumococcus. Biochim Biophys Acta 39: 508–518. 29. Medrano FJ, Gasset M, López-Zúmel C, Usobiaga P, Garcı́a JL, et al. (1996)
9. Moscoso M, Domenech M, Garcı́a E (2010) Vancomycin tolerance in clinical Structural characterization of the unligated and choline-bound forms of the
and laboratory Streptococcus pneumoniae isolates depends on reduced enzyme major pneumococcal autolysin LytA amidase. Conformational transitions
activity of the major LytA autolysin or cooperation between CiaH histidine induced by temperature. J Biol Chem 271: 29152–29161.
kinase and capsular polysaccharide. Mol Microbiol 77: 1052–1064. 30. Hancock REW (2005) Mechanisms of action of newer antibiotics for Gram-
10. Sambrook J, Russell DW (2001) Molecular Cloning. A Laboratory Manual. positive pathogens. Lancet Infect Dis 5: 209–218.
Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press. 31. Chin JN, Rybak MJ, Cheung CM, Savage PB (2007) Antimicrobial activities of
11. López R, Garcı́a E (2004) Recent trends on the molecular biology of ceragenins against clinical isolates of resistant Staphylococcus aureus. Antimicrob
pneumococcal capsules, lytic enzymes, and bacteriophage. FEMS Microbiol Agents Chemother 51: 1268–1273.
Rev 28: 553–580. 32. Pollard JE, Snarr J, Chaudhary V, Jennings JD, Shaw H, et al. (2012) In vitro
12. Moscoso M, Claverys JP (2004) Release of DNA into the medium by competent evaluation of the potential for resistance development to ceragenin CSA-13.
Streptococcus pneumoniae: kinetics, mechanism and stability of the liberated DNA. J Antimicrob Chemother 67: 2665–2672.
Mol Microbiol 54: 783–794. 33. Chin JN, Jones RN, Sader HS, Savage PB, Rybak MJ (2008) Potential synergy
13. Martin B, Prudhomme M, Alloing G, Granadel C, Claverys JP (2000) Cross- activity of the novel ceragenin, CSA-13, against clinical isolates of Pseudomonas
regulation of competence pheromone production and export in the early control aeruginosa, including multidrug-resistant P. aeruginosa. J Antimicrob Chemother
of transformation in Streptococcus pneumoniae. Mol Microbiol 38: 867–878. 61: 365–370.
14. Moscoso M, Garcı́a E, López R (2006) Biofilm formation by Streptococcus 34. Leszczyńska K, Namiot A, Cruz K, Byfield FJ, Won E, et al. (2011) Potential of
pneumoniae: role of choline, extracellular DNA, and capsular polysaccharide in ceragenin CSA-13 and its mixture with pluronic F-127 as treatment of topical
microbial accretion. J Bacteriol 188: 7785–7795. bacterial infections. J Appl Microbiol 110: 229–238.
15. Clinical Laboratory Standards Institute (2006) Methods for dilution antimicro- 35. Isogai E, Isogai H, Takahashi K, Okumura K, Savage PB (2009) Ceragenin
bial susceptibility tests for bacteria that grow aerobically; approved standard, 7th CSA-13 exhibits antimicrobial activity against cariogenic and periodontopathic
ed. Document M7-A7. Wayne, Pa: Clinical Laboratory Standards Institute. bacteria. Oral Microbiol Immunol 24: 170–172.
16. Sánchez-Puelles JM, Sanz JM, Garcı́a JL, Garcı́a E (1992) Immobilization and 36. Epand RF, Savage PB, Epand RM (2007) Bacterial lipid composition and the
single-step purification of fusion proteins using DEAE-cellulose. Eur J Biochem
antimicrobial efficacy of cationic steroid compounds (Ceragenins). Biochim
203: 153–159.
Biophys Acta 1768: 2500–2509.
17. Garcia-Bustos JF, Chait BT, Tomasz A (1987) Structure of the peptide network
37. Obregón V, Garcı́a P, Garcı́a E, Fenoll A, López R, et al. (2002) Molecular
of pneumococcal peptidoglycan. J Biol Chem 262: 15400–15405.
peculiarities of the lytA gene isolated from clinical pneumococcal strains that are
18. Höltje JV, Tomasz A (1976) Purification of the pneumococcal N-acetylmuramyl-
bile insoluble. J Clin Microbiol 40: 2545–2554.
L-alanine amidase to biochemical homogeneity. J Biol Chem 251: 4199–4207.
38. Liu Y, Burne RA (2011) The major autolysin of Streptococcus gordonii is subject to
19. Papazafiri P, Avlonitis N, Angelou P, Calogeropoulou T, Koufaki M, et al.
complex regulation and modulates stress tolerance, biofilm formation, and
(2005) Structure-activity relationships of antineoplastic ring-substituted ether
extracellular-DNA release. J Bacteriol 193: 2826–2837.
phospholipid derivatives. Cancer Chemother Pharmacol 56: 261–270.
39. Dufour D, Lévesque CM (2013) Cell death of Streptococcus mutans induced by a
20. Dı́ez-Martı́nez R, de Paz H, Bustamante N, Garcı́a E, Menéndez M, et al.
quorum-sensing peptide occurs via a conserved streptococcal autolysin.
(2013) Improving the lethal effect of Cpl-7, a pneumococcal phage lysozyme
J Bacteriol 195: 105–114.
with broad bactericidal activity, by inverting the net charge of its cell wall-
40. Burnside K, Lembo A, Harrell MI, Gurney M, Xue L, et al. (2011) Serine/
binding module. Antimicrob Agents Chemother 57: 5355–5365.
threonine phosphatase Stp1 mediates post-transcriptional regulation of hemo-
21. Goppelt-Struebe M, Winter I (1995) Effects of hexadecylphosphocholine on fatty
lysin, autolysis, and virulence of group B Streptococcus. J Biol Chem 286: 44197–
acid metabolism: relation to cytotoxicity. Cancer Chemother Pharmacol 35:
44210.
519–526.

PLOS ONE | www.plosone.org 9 July 2014 | Volume 9 | Issue 7 | e101037

 Anti-Streptococcal Activity of Ceragenin CSA-13

41. Ding B, Guan Q, Walsh JP, Boswell JS, Winter TW, et al. (2002) Correlation of 53. Domenech M, Ramos-Sevillano E, Garcı́a E, Moscoso M, Yuste J (2013) Biofilm
the antibacterial activities of cationic peptide antibiotics and cationic steroid formation avoids complement immunity and phagocytosis of Streptococcus
antibiotics. J Med Chem 45: 663–669. pneumoniae. Infect Immun 81: 2606–2615.
42. Bucki R, Sostarecz AG, Byfield FJ, Savage PB, Janmey PA (2007) Resistance of 54. Domenech M, Garcı́a E, Moscoso M (2012) Biofilm formation in Streptococcus
the antibacterial agent ceragenin CSA-13 to inactivation by DNA or F-actin and pneumoniae. Microb Biotechnol 5: 455–465.
its activity in cystic fibrosis sputum. J Antimicrob Chemother 60: 535–545. 55. Nagant C, Tré-Hardy M, El-Ouaaliti M, Savage P, Devleeschouwer M, et al.
43. Polat ZA, Savage PB, Genberg C (2011) In vitro amoebicidal activity of a (2010) Interaction between tobramycin and CSA-13 on clinical isolates of
ceragenin, cationic steroid antibiotic-13, against Acanthamoeba castellanii and its Pseudomonas aeruginosa in a model of young and mature biofilms. Appl Microbiol
cytotoxic potential. J Ocul Pharmacol Ther 27: 1–5. Biotechnol 88: 251–263.
44. Nagant C, Savage PB, Dehaye JP (2012) Effect of pluronic acid F-127 on the 56. Nagant C, Pitts B, Stewart PS, Feng Y, Savage PB, et al. (2013) Study of the
toxicity towards eukaryotic cells of CSA-13, a cationic steroid analogue of effect of antimicrobial peptide mimic, CSA-13, on an established biofilm formed
antimicrobial peptides. J Appl Microbiol 112: 1173–1183. by Pseudomonas aeruginosa. Microbiologyopen 2: 318–325.
45. Horne D, Hakenbeck R, Tomasz A (1977) Secretion of lipids induced by 57. Ottolenghi E, Hotchkiss RD (1962) Release of genetic transforming agent from
inhibition of peptidoglycan synthesis in streptococci. J Bacteriol 132: 704–717. pneumococcal cultures during growth and disintegration. J Exp Med 116: 491–
519.
46. Horne D, Tomasz A (1977) Tolerant response of Streptococcus sanguis to beta-
58. Muñoz R, Coffey TJ, Daniels M, Dowson CG, Laible G, et al. (1991)
lactams and other cell wall inhibitors. Antimicrob Agents Chemother 11: 888–
Intercontinental spread of a multiresistant clone of serotype 23F Streptococcus
896.
pneumoniae. J Infect Dis 164: 302–306.
47. Höltje J-V, Tomasz A (1975) Lipoteichoic acid: a specific inhibitor of autolysin 59. Liu HH, Tomasz A (1985) Penicillin tolerance in multiply drug-resistant natural
activity in pneumococcus. Proc Natl Acad Sci USA 72: 1690–1694. isolates of Streptococcus pneumoniae. J Infect Dis 152: 365–372.
48. López R, Garcı́a E, Garcı́a P, Garcı́a JL (2004) Cell wall hydrolases. In: 60. Sung H, Shin HB, Kim M-N, Lee K, Kim E-C, et al. (2006) Vancomycin-
Tuomanen EI, Mitchell TJ, Morrison DA, Spratt BG, editors. The tolerant Streptococcus pneumoniae in Korea. J Clin Microbiol 44: 3524–3528.
Pneumococcus. Washington, D.C.: ASM Press. 75–88. 61. Soriano F, Cafini F, Aguilar L, Tarragó D, Alou L, et al. (2008) Breakthrough in
49. López R, Ronda C, Garcı́a E (1990) Autolysins are direct involved in the penicillin resistance? Streptococcus pneumoniae isolates with penicillin/cefotaxime
bactericidal effect caused by penicillin in wild type and in tolerant pneumococci. MICs of 16 mg/L and their genotypic and geographical relatedness.
FEMS Microbiol Lett 66: 317–322. J Antimicrob Chemother 62: 1234–1240.
50. Wolcott RD, Ehrlich GD (2008) Biofilms and chronic infections. JAMA 299: 62. Berrón S, Fenoll A, Ortega M, Arellano N, Casal J (2005) Analysis of the genetic
2682–2684. structure of nontypeable pneumococcal strains isolated from conjunctiva. J Clin
51. Lewis K (2008) Multidrug tolerance of biofilms and persister cells. Curr Top Microbiol 43: 1694–1698.
Microbiol Immunol 322: 107–131. 63. Arbique JC, Poyart C, Trieu-Cuot P, Quesne G, Carvalho MdGS, et al. (2004)
52. Jensen PØ, Givskov M, Bjarnsholt T, Moser C (2010) The immune system vs. Accuracy of phenotypic and genotypic testing for identification of Streptococcus
Pseudomonas aeruginosa biofilms. FEMS Immunol Med Microbiol 59: 292–305. pneumoniae and description of Streptococcus pseudopneumoniae sp. nov. J Clin
Microbiol 42: 4686–4696.

PLOS ONE | www.plosone.org 10 July 2014 | Volume 9 | Issue 7 | e101037