M. Shigetaa
G. Tanakaa
H. Komatsuzawab
M.Sugaib
H. Suginakab
T. Usuia
Permeation of Antimicrobial
Agents through Pseudomonas
aeruginosa Biofilms: A Simple
Method
a
Department of Urology,
Hiroshima University School of
Medicine and
b
Department of Microbiology,
Hiroshima University School of
Dentistry, Hiroshima, Japan
Abstract
In this study, we evaluated the permeation of piperacillin
(PIPC), imipenem (IPM), amikacin (AKM), gentamicin
(GM), ofloxacin (OFLX), levofloxacin (LVFX), ciprofloxacin
(CPFX) and sparfloxacin (SPFX) through Pseudomonas aeruginosa biofilm with a simple new method. Bacteria used were
a leucine-requiring mucoid mutant. Bacteria were grown on
the membrane of a cell culture insert in chemically defined
medium and incubated at 37 °C for 5 days. At days 0, 1, 3 and
5, the penetration rates through the biofilms were measured.
PIPC and IPM demonstrated relatively high permeation both
with penetration rates at day 5 of 50%, whereas AMK and
GM, which are aminoglycosides, showed low permeation both
with penetration rates after day 1 of less than 25%. Among the
4 fluoroquinolones, LVFX and SPFX demonstrated excellent
permeation with penetration rates that reached 100% from
day 0 to 5, while OFLX and CPFX showed almost the same
permeation as IPM. This method of measuring penetration
rates of antimicrobial agents through biofilm is very simple
and useful for the evaluation of antibiotics against biofilmforming bacteria.
Introduction
Bacteria form biofilms by adhering to surfaces using exopolysaccharides termed glycocalyx [1]. Biofilm can cause chronic infection
such as urinary tract infection in patients with
an urinary catheter [2], Antimicrobial agents
are relatively ineffective against biofilmforming bacteria when compared with planktonic bacteria [3], making it difficult to eradi-
cate the pathogens in chronic urinary tract
infection with catheter. It is suggested that
this resistance is due to a barrier effect of glycocalyx, and/or to a lower growth rate caused
by nutrient deprivation [4] and/or to β-lactamase production [5], We recently reported
that the growth rate is an important factor in
the bactericidal activity of β-lactams against
biofilm-forming bacteria, but not for that of
fluoroquinolones, using a leucine-requiring
Pseudomonas aeruginosa mutant (HU1), the
growth rate of which could be controlled by
varying the leucine concentration in the medium [6]. In this study, we used leucinerequiring P. aeruginosa HU1 grown at a low
concentration of leucine as a model of slowgrowing biofilm-forming bacteria and evaluated the permeation of antimicrobial agents
through HU1 biofilms with a simple new
method.
Materials and Methods
Bacterial Strain
A leucine-requiring mucoid mutant, HU1, was
used. This strain is a derivative of the mucoid clinical
strain 4568 isolated from the urine sample of a patient
with urinary tract infection at the Hiroshima University Hospital by chemical mutagenesis, as described previously [6]. Escherichia coli NIHJ and Staphylococcus
aureus FDA 209P were used as control strains for evaluation of antibiotic activities.
Antimicrobial Agents
The antibiotics used were piperacillin (PIPC; Toyama Chemical, Tokyo, Japan), imipenem (IPM; Banyu
Seiyaku, Osaka, Japan), amikacin (AMK; Meiji Seika,
Tokyo, Japan), gentamicin (GM; Schering Plau, Osaka, Japan), ofloxacin (OFLX; Daiichi Seiyaku, Osaka,
Japan), levofloxacin (LVFX; Daiichi Seiyaku), ciprofloxacin (CPFX; Bayer, Leverkusen, Germany) and
sparfloxacin (SPFX; Dainippon Pharmacy, Osaka, Japan). These drugs were provided by the respective
pharmaceutical companies.
Susceptibility Testing
The minimum inhibitory concentrations (MICs) of
PIPC, IPM, AMK, GM, LVFX, CPFX and SPFX
were determined by a microbroth dilution method.
The medium used was trypticase-soy broth (TSB; Becton-Dickinson Microbiology System, Franklin Lakes,
N.J., USA), and the initial suspensions, prepared by
diluting broth cultures of HU1 overnight, contained
106 CFU/ml. The MIC was defined as the lowest concentration which prevented visible growth after incubation without shaking for 24 h at 37 °C.
Medium
The media used were TSB and minimum medium
(MM) (7 g of K2HP04, 3 g of KH2PO4,0.5 g of sodium
citrate, 0.1 g of MgSO4-7 H2O, 1 g of (NH4)2SO4 and 3
g of glucose, dissolved in 1,000 ml of water), containing
100 mg/1 of leucine (L100).
Biofilm Formation on the Membrane of the Cell
Culture Insert
Cells were grown overnight in 10ml of TSB at
37 °C and washed with phospate-buffered saline
(PBS). Bacteria were resuspended in 4 ml of MM containing LI00 to 107 CFU/ml and then placed into cell
culture inserts (0.4 μm, 23.4 mm in diameter, BectonDickinson Labware, Franklin Lakes, NJ. USA)
(fig. 1). Cell culture inserts were placed into 6-well cell
culture plates (Becton-Dickinson Labware; one cell
culture insert/well) and incubated for 2 h at 37° C to
allow the bacteria to adhere to the membrane at the
bottom of each insert. After several washings with PBS
to remove nonadherent cells, the bacteria on the membrane of the cell culture insert were resuspended in
4ml of MM containing LI00 at 37 °C (day 0). The
medium was changed every 8 h following several washings with PBS.
Measurement of Permeation of Antimicrobial
Agents through HU1 Biofilms
A cell culture insert containing biofilm incubated
in MM containing L100 was removed at days 0, 1, 3
and 5. After several washings with PBS to remove nonadeherent cells, the bacteria on the membrane of the
cell culture insert were incubated in 4 ml of MM containing LI00 together with respective antimicrobial
agents (100μg/ml). The cell culture inserts were suspended in 6-well cell culture plates containing 4 ml of
fresh TSB. At 3 h, 200 μl of TSB from these cell culture
plates were taken and passed through a filter (sterile
Acrodisk, 0.2 μm, 13mm in diameter, HT TufTryn
membrane; Gelman Science, Ann Arbor, Mich, USA)
to measure the concentration of antimicrobial agent
penetrated through the HU1 biofilm by a microdilution method. The concentrations of penetrated antimicrobial agent at day 0 were defined as 100% for each
Fig. 1. Schema of cell culture insert and cell culture
plate.
Fig. 2. Growth curve of biofilm-forming mucoid
P.aeruginosa HU1 (leucine-requiring mutant) on a
cell desk during incubation in MM in the presence of
LI 00.
agent. E. coli NIHJ was used as the indicator bacterium for the assay of PIPC, IPM, OFLX, LVFX, CPFX
and SPFX. S. aureus FDA 209P was used for the assay
of AMK and GM. All experiments were performed in
triplicate, and the results were expressed as the median
percentage.
Results
Minimum Inhibitory Concentrations The
MICs of PIPC, IPM, AMK, GM,
OFLX, LVFX, CPFX and SPFX for the
planktonic HU1 were 8, 4, 2, 1, 0.5, 0.1, 0.1
and 0.4 μg/ml, respectively.
Growth Curve of Biofilm-FormingHUl
Growth curves of biofilm-forming HU1 on
cell desks (round type, 13.5 mm in diameter;
Sumitomo, Tokyo, Japan) suspended in MM
containing LI00 are shown in figure 2. This
method has been used in a previous study [6].
The growth rate of biofilm-forming HU1 was
very slow. The growth curve of those cultured
in MM containing LI00 showed biphasic
growth: the cells grew exponentially until day
3 and remained stationary afterwards. The
approximate doubling time of biofilm-forming HU1 in the exponential growth phase was
7.6 h.
Permeation of Antimicrobial Agents
through HUlBiofilm
The penetration rate of antimicrobial
agents through HU1 biofilm is shown in figure 3a, b. PIPC and IPM, which are β-lactams, demonstrated relatively high permeation both with penetration rates at day 5 of
50%, while AMK and GM, which are aminoglycosides, showed low permeation, both with
penetration rates after day 1 of less than 25%.
Among the 4 fluoroquinolones, LVFX and
SPFX demonstrated excellent permeation
antibiotic diffusion [7]. Baltimore et al. [8]
have reported that alginate, a major component of glycocalyx produced by mucoid-type
P. aeruginosa, is an important factor in the
resistance to aminoglycosides, but not to βlactams [8], This can be explained by differDiscussion
ences in the charges of antibiotics; aminoglyIt has been suggested that the barrier effect cosides are positively charged, while β-lacof glycocalyx is an important factor in the tams are uncharged. The former bind to the
resistance of biofilm to antimicrobial agents electronegative alginate, while the latter do
[4] and that glycocalyx may directly protect not. Therefore, aminoglycoside activity is dithe bacteria against antibiotics by retarding minished in the presence of alginate [8]. On
both with penetration rates of 100% from day
0 to 5, while OFLX and CPFX showed almost
the same permeation as IPM.
Table 1. pK and charge of antimicrobial agents at pH 6.6
ND = Not determined; Ν = not charged; Ρ = positively charged.
the other hand, Hodges and Gordon [9] have
reported the possibility of a non-charge-related
binding of antimicrobial agents to alginate
because ciproxan and β-lactams were inhibited by alginate at a neutral pH. The pK of each
antimicrobial agent used in this study is
shown in table 1. At pH 6.6, which is the pH
of MM containing LI00, AMK is positively
charged, and the other antibiotics are uncharged. Our results supported Hodges' [9]
theory and suggested that permeation of antimicrobial agents through biofilm varied depending on the type of antibiotic and the age
of the biofilm. With the method of Yasuda et
al. [10], we confirmed that the glycocalyx in
HU1 biofilm incubated in MM with L100
contained alginate (data not shown). In spite
of the presence of alginate, LVFX and SPFX
demonstrated excellent permeation from day
1 to 5, and their permeation was not affected
at all. PIPC, IPM, OFLX and CPFX showed
relatively high penetration rates. However, on
day 5, the penetration of these antibiotics was
inhibited to 50%. Lastly, aminoglycosides
were the most strongly inhibited of all agents
tested, with penetration rates after day 1 of
only 25%.
The results of this and a previous study [6]
suggest that a barrier effect of glycocalyx and a
lower growth rate are important factors in the
resistance of biofilm-forming bacteria to an-
timicrobial agents. For the treatment of biofilm-forming bacteria, antibiotics should be
selected based on good permeation through
glycocalyx and good bactericidal activity
against slow-growing bacteria. Our new method may prove useful for evaluating the clinical
potential of antimicrobial agents against biofilm-forming bacteria.
Acknowledgements
This study was partially supported by an educational grant from the Tsuchiya foundation, Hiroshima,
Japan.
References
1 Costerton JW, Irvin RT, Cheng KJ:
The bacterial glycocalyx in nature
and disease. Annu Rev Microbioi
1981;35:299-324.
2 Hoyle BD, Alcantara J, Costerton
JW: Pseudomonas aeruginosa biofilm as a diffusion barrier to piperacillin. Antimicrob Agents Chemother!992;36:2054-2056.
3 Prosser BLT, Taylor D, Dix BA,
Cleeland R: Method of evaluating
effect of antibiotics on bacterial biofilm. Antimicrob Agents Chemother
1987;31:1502-1506.
4 Gilbert P, Collier PJ, Brown MRW:
Influence of growth rate on suscepti
bility to antimicrobial agents. Biofilm, cell cycle, dormancy, and strin
gent response. Antimicrob Agents
Chemother 1990;34:1865-1868.
5 Giwercman B, Jensen ET, Hoiby N,
Kharazmi A, Costerton JW: Intro
duction of β-lactamase production
in Pseudomonas aeruginosa biofilm.
Antimicrob Agents Chemother
199l;35:1008-1010.
6 Shigeta M, Komatsuzawa H, Sugai
M, Suginaka H, Usui T: The effect
of growht rate of Pseudomonas aeru
ginosa biofilms on the susceptibility
to antimicrobial agents. Chemother
apy, in press.
7 Gordon CA, Hodges NA, Marriott
C: Antibiotic interaction and diffu
sion through alginate and exopolysaccharide of cystic fibrosis-derived
Pseudomonas aeruginosa. Antimi
crob Agents Chemother 1988;22:
667-674.
8 Baltimore RS, Cross AS, Dobek AS:
The inhibitory effect of sodium algi
nate on antibiotic activity against
mucoid and non-mucoid of P. aeruginosa. J Antimicrob Chemother
1987;20:815-823.
9 Hodges NA, Gordon CA: Protection
of Pseudomonas aeruginosa against
ciprofloxacin and β-lactams by ho
mologous alginate. Antimicrob Che
mother 1991;35:2450-2452.
10 Yasuda H, Ajiki Y, Kawada H, Yokota T: Interaction between biofilms
formed by Pseudomonas aeruginosa
and clarithromycin. Antimicrob
Agents Chemother 1993;37:17491755.