Pseudomonas aeruginosa inhibits in-vitro Candida biofilm development.

Abstract

Elucidation of the communal behavior of microbes in mixed species biofilms may have a major impact on understanding infectious diseases and for the therapeutics. Although, the structure and the properties of monospecies biofilms and their role in disease have been extensively studied during the last decade, the interactions within mixed biofilms consisting of bacteria and fungi such as Candida spp. have not been illustrated in depth. Hence, the aim of this study was to evaluate the interspecies interactions of Pseudomonas aeruginosa and six different species of Candida comprising C. albicans, C. glabrata, C. krusei, C. tropicalis, C. parapsilosis, and C. dubliniensis in dual species biofilm development.
A significant reduction in colony forming units (CFU) of C. parapsilosis (90 min), C. albicans and C. tropicalis (90 min, 24 h and 48 h), C. dubliniensis and C. glabrata, (24 h and 48 h) was noted when co-cultured with P. aeruginosa in comparison to their monospecies counterparts (P < 0.05). A simultaneous significant reduction in P. aeruginosa numbers grown with C. albicans (90 min and 48 h), C. krusei (90 min, 24 h and 48 h),C. glabrata, (24 h and 48 h), and an elevation of P. aeruginosa numbers co-cultured with C. tropicalis (48 h) was noted (P < 0.05). When data from all Candida spp. and P. aeruginosa were pooled, highly significant mutual inhibition of biofilm formation was noted (Candida P < 0.001, P. aeruginosa P < 0.01). Scanning Electron Microscopy (SEM) and Confocal Laser Scanning Microscopy (CLSM) analyses confirmed scanty architecture in dual species biofilm in spite of dense colonization in monospecies counterparts.
P. aeruginosa and Candida in a dual species environment mutually suppress biofilm development, both quantitatively and qualitatively. These findings provide a foundation to clarify the molecular basis of bacterial-fungal interactions, and to understand the pathobiology of mixed bacterial-fungal infections.

Full text

Page 1

Bandara et al. BMC Microbiology 2010, 10:125
http://www.biomedcentral.com/1471-2180/10/125
Open Access
RESEARCH ARTICLE
Pseudomonas aeruginosa inhibits in-vitro Candida
biofilm development
BioMed Central
© 2010 Bandara et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons
Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
Research article
HMHN Bandara1, JYY Yau1, RM Watt1, LJ Jin2 and LP Samaranayake*1
Abstract
Background: Elucidation of the communal behavior of microbes in mixed species biofilms may have a major impact
on understanding infectious diseases and for the therapeutics. Although, the structure and the properties of
monospecies biofilms and their role in disease have been extensively studied during the last decade, the interactions
within mixed biofilms consisting of bacteria and fungi such as Candida spp. have not been illustrated in depth. Hence,
the aim of this study was to evaluate the interspecies interactions of Pseudomonas aeruginosa and six different species
of Candida comprising C. albicans, C. glabrata, C. krusei, C. tropicalis, C. parapsilosis, and C. dubliniensis in dual species
biofilm development.
Results: A significant reduction in colony forming units (CFU) of C. parapsilosis (90 min), C. albicans and C. tropicalis (90
min, 24 h and 48 h), C. dubliniensis and C. glabrata, (24 h and 48 h) was noted when co-cultured with P. aeruginosa in
comparison to their monospecies counterparts (P < 0.05). A simultaneous significant reduction in P. aeruginosa
numbers grown with C. albicans (90 min and 48 h), C. krusei (90 min, 24 h and 48 h),C. glabrata, (24 h and 48 h), and an
elevation of P. aeruginosa numbers co-cultured with C. tropicalis (48 h) was noted (P < 0.05). When data from all Candida
spp. and P. aeruginosa were pooled, highly significant mutual inhibition of biofilm formation was noted (Candida P <
0.001, P. aeruginosa P < 0.01). Scanning Electron Microscopy (SEM) and Confocal Laser Scanning Microscopy (CLSM)
analyses confirmed scanty architecture in dual species biofilm in spite of dense colonization in monospecies
counterparts.
Conclusions: P. aeruginosa and Candida in a dual species environment mutually suppress biofilm development, both
quantitatively and qualitatively. These findings provide a foundation to clarify the molecular basis of bacterial-fungal
interactions, and to understand the pathobiology of mixed bacterial-fungal infections.
Background
In most natural environments, microorganisms exist pre-
dominantly as biofilms rather than as free floating plank-
tonic cells [1]. A biofilm can be defined as a complex
functional community of one or more species of
microbes encased in extra cellular polymeric substances
and, attached to one another or to a solid surface [2]. Bio-
films can be composed of a single microbial species or
more commonly, mixed species such as bacteria and
fungi [3,4]. Perhaps the most studied example of the bio-
film in humans is the dental plaque[5]. Microorganisms
in the biofilm characteristically display a phenotype that
is markedly different from that of their free floating coun-
terparts [1]. For instance, they are resistant to antimicro-
bial agents in comparison to planktonic cells [6-8]. As
more than 65% of biofilms with human microbial infec-
tions are caused by biofilms [5], there is an urgent need to
understand biofilm behaviour.
The genus Candida comprises more than 150 patho-
genic and nonpathogenic yeast species. Among these, C.
albicans, C. tropicalis, C. parapsilosis, C. krusei, C. kefyr,
C. glabrata and C. guillermondii are recognized as medi-
cally important pathogens [9]. C. albicans is the most
prevalent yeast isolated from humans (47-75%) followed
by C. tropicalis (7%), C. glabrata (7%), C. krusei (5%), C.
parapsilosis (< 5%) and C. guillermondii (< 5%) [9]. Com-
mon Candidal habitats of humans include the gut, skin
and mucosal surfaces, while one half of the human popu-
lation carry Candida in their oral cavities[10].
* Correspondence: lakshman@hkucc.hku.hk
1 Faculty of Dentistry, The University of Hong Kong, Oral Biosciences, 5/F, Prince
Phillip Dental Hospital, 34, Hospital road, Sai Ying Pun, Hong Kong
Full list of author information is available at the end of the article

Page 2

Bandara et al. BMC Microbiology 2010, 10:125
http://www.biomedcentral.com/1471-2180/10/125
Page 2 of 9
Pseudomonas aeruginosa is an aerobic Gram-negative
bacterium that causes community acquired infections,
such as ulcerative keratitis, otitis externa, skin and soft
tissue infections and, nosocomial infections including
pneumonias, urinary tract infections, infections in surgi-
cal sites and burns [24,25]. Indeed, out of all nosocomial
infections in different ethnic communities, 11-13.8% is
found to be caused by P. aeruginosa [11-13]. United States
Cystic Fibrosis Foundation Patients Registry (2004), has
stated that 57.3% of all reported respiratory cultures con-
tained P. aeruginosa indicating its important role in caus-
ing chronic and recurrent infections in cystic fibrotic
patients [14]. Lee et al [15] have demonstrated that P.
aeruginosa is the most commonly identified cause of sep-
ticemia in primary immunodeficiency and some 20% of
bacteriaemia in acute leukemic patients [16,17]. Inci-
dence of P. aeruginosa bacteriaemias in HIV affected
patients is approximately 10 times higher than that of the
normal population [18].
Pathogenic interactions between C. albicans and P.
aeruginosa have recently been demonstrated by a number
of groups [19,20]. The antifungal behaviour of P. aerugi-
nosa against Candida spp. was first reported in early
nineties by Kerr et al [20]. Subsequently others have
shown that P. aeruginosa kills C. albicans by forming a
dense film on fungal filaments, though, it neither binds
nor kills the yeast-form of C. albicans [19]. Thein et al
[21] have also reported that P. aeruginosa ATCC 27853 at
a concentration gradient elicited a significant inhibition
of Candida albicans biofilms.
Although, the structure and the properties of monospe-
cies biofilms and their role in disease have been exten-
sively studied during the last decade [22,23], the
interactions within mixed biofilms consisting of bacteria
and fungi including Candida spp. have not been studied
in depth. Furthermore, the majority of the previous stud-
ies on interactions between Candida and bacteria in
mixed biofilms have focused on C. albicans and there are
only a few studies on non-albicans Candida spp. biofilms
in a mixed species environment. Hence, the aims of this
study were to evaluate the interactions of a reference lab-
oratory strain of P. aeruginosa and six different Candida
species, C. albicans, C. glabrata, C. tropicalis, C. parapsi-
losis, C. dubliniensis, and C. krusei in a dual species bio-
films environment over a period of 2 days by both
quantitative assays (Colony Forming Unit assay - CFU)
and, qualitative evaluations using Scanning Electron
Microscopy (SEM) and Confocal Laser Scanning micros-
copy (CLSM).
Results
Candida and P. aeruginosa dual species biofilm growth
After 90 min of biofilm development with P. aeruginosa, a
significant, 57-88%, reduction in Candida counts was
noted for C. albicans (57%, P = 0.005),C. dubliniensis
(69%, P < 0.001),C. tropicalis (18%, P = 0.010) and C.
parapsilosis (74%, P = 0.030) while P. aeruginosa did not
impart such an effect on C. glabrata and C. krusei com-
pared with the controls (Table 1). Conversely, after 90
min, a significant reduction in CFU of P. aeruginosa was
observed in the presence of C. albicans (81%, P = 0.002)
C. krusei (62%, P = 0.002) but not with the other four
Candida species (Table 1).
However, after prolonged incubation for 24 hours, a
significant, 58-91% reduction in the counts of C. albicans
(67%, P < 0.001), C. tropicalis (88%, P < 0.001) C. dublin-
iensis (91%, P < 0.001) and C. glabrata (58%, P= 0.024)
was noted in dual species biofilms with P. aeruginosa
(Table 1) although C. krusei and C. parapsilosis counts
were unaffected in comparison to the monospecies con-
trols. On the other hand, mean CFU of P. aeruginosa
decreased significantly in the presence of C. krusei (41%,
P = 0.022), C. dubliniensis (48%, P = 0.003) and C.
glabrata (83%, P < 0.001) after 24 h, while the other three
Candida species had no significant effect on P. aerugi-
nosa numbers at this time point (Table 1).
Most remarkable results were observed on further
incubation for 48 hours, C. albicans (99%, P < 0.001), C.
tropicalis (100%, P < 0.001) and C. glabrata (100%, P <
0.001) growth was almost totally suppressed in dual spe-
cies biofilms with P. aeruginosa while the remaining Can-
dida species were unaffected (Table 1). Simultaneously
the mean CFU of P. aeruginosa decreased in co cultures
of C. albicans (32%, P = 0.009) C. krusei (48%, P = 0.010),
and C. glabrata (78%, P < 0.001). Conversely, P. aerugi-
nosa counts significantly increased in the presence of C.
tropicalis (72%, P = 0.002). Such an effect was not seen
after 48 h with the two remaining Candida species,C.
dubliniensis and C. parapsilosis (Table 1).
Despite these variable results, at different time inter-
vals, when data from all Candida spp. were pooled and
analyzed, a highly significant inhibition of Candida bio-
film formation by P. aeruginosa (P < 0.001) and a simulta-
neous significant inhibition of P. aeruginosa biofilm
development by Candida at all three time intervals (P <
0.01) was noted.
Confocal laser scanning microscopy
CLSM with Live and Dead stain confirmed, in general,
that Candida spp. and P. aeruginosa have mutually sup-
pressive effects on each other at every stage of biofilm
formation, in comparison to their monospecies counter-
parts. CLSM showed a reduction in both Candida and P.
aeruginosa cells that were adherent after 90 min, con-
firming the data from CFU assay. Few dead C. albicans
cells were also visible (Figure 1A, B and 1C).
In 24 h-dual species biofilms, mutual suppression of C.
dubliniensis and P. aeruginosa was clearly seen, confirm-

Page 3

Bandara et al. BMC Microbiology 2010, 10:125
http://www.biomedcentral.com/1471-2180/10/125
Page 3 of 9
Table 1: The mean CFU counts (± SD) of Candida spp. and P. aeruginosa from both monospecies and dual species biofilms
at 90 min, 24 h and 48 h.
Time
interval
Candida CFU (106) ± SD P value
P. aeruginosa CFU (106) ± SDP value
Control (MSB)Test (DSB)Control (MSB) Test (DSB)
Candida
albicans
90 min
12.60 ± 2.19 5.29 ± 1.52 0.005 3.44 ± 2.200.66 ± 0.690.002
24 h
15.22 ± 3.315.00 ± 2.60 < 0.001
876.89 ± 206.39719.56 ± 266.530.200
48 h
31.89 ± 6.600.22 ± 0.44 < 0.0011358.89 ± 323.59922.22 ± 186.600.009
Candida krusei
90 min 2.43 ± 1.462.71 ± 0.66 0.352
7.32 ± 3.82 2.78 ± 1.29 0.003
24 h3.39 ± 2.00 2.49 ± 0.730.301
987.78 ± 341.79583.33 ± 218.920.022
48 h0.09 ± 0.14 0.22 ± 0.44 0.867
140.00 ± 48.73 73.33 ± 35.71 0.010
Candida
tropicalis
90 min
9.81 ± 3.05 3.87 ± 2.290.004
1.42 ± 1.252.26 ± 0.710.070
24 h
27.67 ± 5.92 3.44 ± 1.59< 0.001
431.11 ± 66.23 471.11 ± 162.900.534
48 h
4.22 ± 2.050.00 ± 0.00 < 0.00198.89 ± 75.74351.11 ± 162.510.002
Candida
parapsilosis
90 min
10.60 ± 6.711.26 ± 1.34< 0.001
4.87 ± 1.663.83 ± 2.31 0.228
24 h 2.11 ± 2.32 0.78 ± 0.440.364412.22 ± 208.55 277.78 ± 162.69 0.121
48 h0.89 ± 0.60 0.44 ± 0.73 0.120 183.33 ± 69.64179.56 ± 50.020.859
Candida
glabrata
90 min10.81 ± 2.9010.12 ± 3.970.659 9.91 ± 9.018.17 ± 5.030.691
24 h
35.78 ± 21.7215.00 ± 21.080.024 328.89 ± 88.9456.67 ± 15.81 < 0.001
48 h
28.22 ± 17.140.11 ± 0.33 < 0.001 128.89 ± 69.54 28.89 ± 17.64< 0.001
Candida
dubliniensis
90 min
9.34 ± 3.212.94 ± 1.50 < 0.001
9.83 ± 2.336.51 ± 4.350.070
24 h
5.81 ± 2.46 0.54 ± 0.88< 0.001 878.89 ± 286.07461.11 ± 142.780.003
48 h0.00 ± 0.000.00 ± 0.001.00097.78 ± 48.1652.22 ± 50.940.056
P < 0.05 was considered statistically significant. Significant differences are shown in bold text.
ing CFU data. Thus, sparsely developed C. dubliniensis
biofilm was seen with few dead cells in contrast to its
dense monospecies biofilm, while P. aeruginosa numbers
were greatly reduced compared to its monospecies coun-
terpart (Figure 1D, E and 1F).
Similarly, after 48 h, sparsely distributed C. tropicalis
blastospores were noted in dual species biofilms with few,
scattered P. aeruginosa cells and a scant biofilm once
again confirming the aforementioned quantitative CFU
findings. Some dead cells and cellular debris were also
observed compared to dense monospecies biofilm
growth of C. tropicalis control (figure 1G, H and 1I).
Scanning Electron Microscopy
Although species specific growth variations could be
noted, in general, single species biofilms of all Candida
species demonstrated profuse growth and dense coloni-
zation of the substrate on SEM observation (Figure 2).
After 90 min, i.e. adhesion phase, the control monospe-
cies Candida and P. aeruginosa cells were seen well-
adherent and uniformly distributed on the polystyrene
surface. Yeast blastospores were seen aggregated either in
pairs or clumps with some budding yeasts. During 24 h of
initial colonization phase, monospecies biofilms of both
Candida and P. aeruginosa showed increased numbers of

Page 4

Bandara et al. BMC Microbiology 2010, 10:125
http://www.biomedcentral.com/1471-2180/10/125
Page 4 of 9
Figure 1 CLSM images of monospecies (Candida spp. or P. aeruginosa) and dual species (Candida spp. and P. aeruginosa) biofilms. (A). Ad-
hesion of C. albicans for 90 min, (B). Adhesion of C. albicans and P. aeruginosa for 90 min, (C). Adhesion of P. aeruginosa for 90 min. Note the mutual
inhibition of adhesion of both pathogens in dual species environment. (D) Initial colonization of C. dubliniensis for 24 h (E). Initial colonization of C.
dubliniensis and P. aeruginosa for 24 h, (F). Initial colonization of P. aeruginosa for 24 h. Note the impaired biofilm formation after 24 h in the dual species
biofilm due to mutual inhibition of these organisms. (G) Maturation of C. tropicalis for 48 h, (H). Maturation of C. tropicalis and P. aeruginosa for 48 h, (I).
maturation of P. aeruginosa for 48 h. Note the altered and scant biofilm maturation in dual species biofilm as a result of mutual inhibition of C. tropicalis
and P. aeruginosa.

Page 5

Bandara et al. BMC Microbiology 2010, 10:125
http://www.biomedcentral.com/1471-2180/10/125
Page 5 of 9
cellular layers with recognizable extracellular matrix.
After 48 h, the single species biofilms of both pathogens
were relatively thick and multilayered, although the extra-
cellular matrix was scarcely visible.
However, on visual examination by SEM, dual species
biofilms demonstrated reduction of yeast blastospores at
each stage of biofilm formation compared to their mono-
species counterparts. Specially in the maturation stage at
48 h, this reduction was marked and recognizable. The
former biofilms were also less dense than the monospe-
cies controls, and demonstrated few layers of cells, pro-
fuse cellular debris, together with degrading and
morphologically altered yeast cells. Interestingly, most of
the bacteria were seen attached to the blastospores (fig-
ure 2E and 2H). Bacterial density varied in the presence
of different Candida species at different time intervals. In
general, P. aeruginosa distribution was scanty and nonde-
script in the dual species environment (Figure 2B, E and
2H).
Quantitatively, smaller numbers of clumped C. albi-
cans, together with some degrading blastospores, were
observed with P. aeruginosa at the end of the adhesion
phase, and the latter was also lesser in number compared
to the monospecies variant (Figure 2A, B and 2C). A thin,
scant biofilm, formed by a lesser numbers of morphologi-
cally altered C. glabrata was noted after initial coloniza-
tion (Figure 2C, D and 2E). Furthermore, a few,
morphologically altered blastospores of C. tropicalis were
visible in mature dual species biofilm with P. aeruginosa
at 48 h. In contrast, P. aeruginosa demonstrated thicker
biofilms in the presence of C. tropicalis, compared to its
mature monospecies variant (Figure 2G, H and 2I).
Discussion
Candida and P. aeruginosa are major pathogens of
device-associated nosocomial infections for virtually all
types of indwelling devices [24]. It has also been stated
that, the coexistence of Pseudomonas spp. and C. albicans
in elderly is a potential indicator of high risk for pneumo-
nia [25]. Recent experimental studies have identified sim-
ilarities in environmental factors such as its physical and
chemical nature where P. aeruginosa and C. albicans
coexist [26]. As a result, these two microorganisms have
become obvious candidates and models for the study of
biofilm infections in order to develop potential methods
for the control of device-associated nosocomial infec-
tions[24].
The principle aim of this study was to evaluate the qual-
itative and quantitative effects of P. aeruginosa on various
stages of in-vitro biofilm formation of six different Can-
dida species. Our results indicate that both Candida and
P. aeruginosa mutually inhibit biofilm development to
varying degrees at different stages of biofilm formation.
However, the most important conclusion of our study is
the ability of P. aeruginosa to almost totally inhibit C.
albicans, C. glabrata and C. tropicalis in 48 h biofilms.
Using a CFU assay, we report here for the first time, the
quantitative effect of P. aeruginosa on biofilm formation
of six different Candida species in a time dependant
manner. Our results indicate that P. aeruginosa had sig-
nificant inhibitory effects on several Candida spp. such
as, C. albicans, C. dubliniensis, C. tropicalis, and C.
parapsilosis. In contrast, El-Azizi [27] found that
Pseudomonas had no significant effect on C. albicans
adhesion and biofilm growth, regardless of adding pre-
formed Pseudomonas biofilms to C. albicans or vice
versa. As there appeared to be differences in the mode of
attachment of P. aeruginosa to yeast form of C. albicans
or its filamentous form [28], mixed biofilm development
between these two organisms could be a function of these
characteristics.
Thein et al [21] from our group reported that, on pro-
long incubation for 2 days, P. aeruginosa ATCC 27853 at a
concentration gradient, elicited a significant inhibition of
C. albicans biofilm with a mean reduction in the number
of viable Candidal cells ranging from 38% to 81%. Our
results extend their work further and indicate that P.
aeruginosa suppresses several other Candida species on
incubation for upto two days, for instance, C. dubliniensis
at 24 h and,C. albicans, C. glabrata and C. tropicalis both
at 24 h and 48 h. In this context, Kaleli et al [29] investi-
gated the anticandidial activity of 44 strains of P. aerugi-
nosa, isolated from a number of specimens of intensive
care patients, against four Candida species (C. albicans,
C. tropicalis, C. parapsilosis and C. krusei) by a cross
streak assay and subcutaneous injections of both bacte-
rial and fungal suspensions into mice. They found that all
Pseudomonas strains tested inhibited all four Candida
species to varying degrees. C. albicans and C. krusei were
the most inhibited while C. tropicalis were the least [29].
In contrast, our data show that the most significant inhi-
bition elicited by P. aeruginosa was C. albicans and C.
tropicalis while, the least was C. krusei. Grillot et al [30]
observed complete or partial inhibition of C. albicans, C.
tropicalis, C. parapsilosis and C. glabrata by P. aeruginosa
in pure and mixed blood cultures using in-vitro yeast
inhibition assays and suggested that preclusion of yeast
recovery from blood cultures in mixed infections, such as
polymicrobial septicemia, may be due to suppression of
yeast by P. aeruginosa. In another study Kerr [20] demon-
strated that nine Candida species, out of eleven tested,
including C. krusei, C. kefyr, C. guillermondii, C. tropi-
calis, C. lusitaniae, C. parapsilosis, C. pseudotropicalis, C.
albicans and Torulopsis glabrata were suppressed by P.
aeruginosa. This in-vitro susceptibility test was per-
formed with ten different strains of P. aeruginosa
obtained from the sputum of three patients. Moreover, C.
albicans was the most susceptible to growth inhibition