APMIS 115: 891–9, 2007 C 2007 The Authors

Printed in Denmark . All rights reserved

Journal Compilation C 2007 APMIS
ISSN 0903-4641

Quantification of biofilm in microtiter plates:

overview of testing conditions and practical recommendations
for assessment of biofilm production by staphylococci
SRDJAN STEPANOVIĆ,1 DRAGANA VUKOVIĆ,1 VERONIKA HOLA,2
GIOVANNI DI BONAVENTURA,3 SLOBODANKA DJUKIĆ,1 IVANA ĆIRKOVIĆ1 and
FILIP RUZICKA2
1
Department of Bacteriology, Institute of Microbiology and Immunology, School of Medicine, Belgrade,
Serbia, 2Institute for Microbiology, Faculty of Medicine, Masaryk University, Brno, Czech Republic and
3
Laboratory of Clinical Microbiology, Aging Research Center Ce.S.I., ‘‘G. D’Annunzio’’
University of Chieti-Pescara, Chieti, Italy

Stepanović S, Vuković D, Hola V, Di Bonaventura G, Djukić S, Ćirković I, Ruzicka F. Quantification

of biofilm in microtiter plates: overview of testing conditions and practical recommendations for
assessment of biofilm production by staphylococci. APMIS 2007;115:891–9.
The details of all steps involved in the quantification of biofilm formation in microtiter plates are
described. The presented protocol incorporates information on assessment of biofilm production by
staphylococci, gained both by direct experience as well as by analysis of methods for assaying biofilm
production. The obtained results should simplify quantification of biofilm formation in microtiter
plates, and make it more reliable and comparable among different laboratories.
Key words: Biofilm; microtiter plate; Staphylococcus; quantification.
Srdjan Stepanović, Institute of Microbiology and Immunology, School of Medicine, Dr Subotića 1,
11000 Belgrade, Serbia. e-mail stepan/afrodita.rcub.bg.ac.yu

The study of microbial biofilms has received sig- (3–5), but no standardized protocol for assess-
nificant attention over the past decades. Biofilm ment of biofilm formation by different bacterial
is defined as an assemblage of microbial cells that species has been established so far. Quantifi-
is associated with a surface and enclosed in cation of biofilms started with a method based
an extracellular matrix principally of polysac- on the cultivation of biofilm on the wall of a
charide material (1). Biofilm-associated organ- test tube and subsequent detection by stain for
isms are fundamentally different from popula- biofilm recognition (6). Later, the wells of
tions of suspended cells. Bacteria in biofilm microtiter plate were used as culture vessel, and
display elevated resistance to antibiotics, disin- the results were measured spectrophotometr-
fectants, as well as to host immune system clear- ically (7). Currently a number of different
ance (1, 2). The importance of biofilm is well rec- methods are used, such as tube test, microtiter
ognized in medical, environmental, and indus- plate test, radiolabeling, microscopy, Congo red
trial contexts. agar plate test, etc. (3–5). Nevertheless, the
A number of methods have been developed microtiter plate method remains among the
for cultivation and quantification of biofilms most frequently used assays for investigation of
biofilm, and a number of modifications have
Received 8 November 2006. been developed for the in vitro cultivation and
Accepted 13 February 2007. quantification of bacterial biofilms.
891
 STEPANOVIĆ et al.

Staphylococci are the most common causa- tain a mixture of phenotypes that differ in their abil-
tive agents of medical device-associated infec- ity to form biofilm (see below for phase variations).
tions. The crucial step in the pathogenesis of Inoculum
these infections is the formation of stable bi- Prior to inoculation, tested strains may be culti-
ofilm on the surface of the implanted biomed- vated either in broth or on solid medium. Incubation
ical device (8). The ability to produce an adher- of staphylococci in liquid medium is more frequently
ent multilayered biofilm on implanted bio- used. The strain is transferred from the stock culture
materials is considered an important virulence onto appropriate non-inhibitory medium (i.e. TSA or
blood agar) and incubated overnight aerobically
factor of staphylococci (8). Biofilm production 35 æC–37 æC. After verifying purity of the strain, three
by staphylococci has been extensively investi- to four well-isolated identical colonies are suspended
gated, particularly in Staphylococcus epiderm- in 5 ml of TSB and incubated without shaking for 18
idis and Staphylococcus aureus. h∫30 min (3). After incubation, the stationary-phase
The purpose of this paper is to present in a culture is vortexed and thereafter diluted 1:2 to
step-by-step manner all phases in the quantifi- 1:1000 (3, 7, 12–25), but most frequently 1:100, in
medium for bioflim cultivation (TSB supplemented
cation of biofilm formation by staphylococci in
with glucose, see below). The diluted bacteria are vor-
a microtiter plate, and to present the most im- texed and then inoculated into a microtiter plate (200
portant variations of the method. mL per well).
Another approach is to transfer the strain from the
stock culture, prior to assaying for biofilm produc-
tion, onto TSA, Mueller Hinton agar, or blood agar
TESTING CONDITIONS and to incubate overnight to 24 h aerobically at
35 æC–37 æC. After verifying purity of the strain, a few
Storage of bacterial strains for biofilm testing colonies with identical morphology are suspended in
Preservation is important in the maintenance of sterile distilled water or physiological saline. The tur-
bacteria for research applications. Freeze drying (i.e. bidity of the bacterial suspension is adjusted to
lyophilization) (9) and, more frequently, cryopreserv- match turbidity comparable to that of the 0.5
ation are used for archiving of bacteria isolates (10). McFarland standard (∂108 CFU/ml). It is preferable
For cryopreservation, staphylococci are inoculated to perform this step by using a photometric device.
onto appropriate non-inhibitory medium (e.g. tryptic The obtained suspension should be vortexed for at
soy agar (TSA), blood agar) and incubated overnight least 1 min.
aerobically 35 æC–37 æC. After verifying the purity of Although both the above-mentioned protocols are
the strains, one colony is emulsified in appropriate acceptable, it is significant that staphylococci grown
broth (e.g. tryptic soy broth (TSB), brain heart in- on solid medium differ in expression of cell-associ-
fusion broth (BHI)). Following overnight incubation ated molecules compared to staphylococci grown in
at 35 æC–37 æC, the broth is vortexed and poured into liquid medium (26, 27). Probably all staphylococci
a suitable tube, and previously sterilized glycerol is possess capsule or slime layers that are regarded as
aseptically added to reach a final concentration of soft polyelectrolyte layers surrounding the bacteria
10–15%. Alternatively, a few colonies from the over- (27). The presence of such a layer decreases the en-
night incubated agar medium are emulsified in appro- ergy barrier due to electrostatic repulsion in the inter-
priate broth supplemented with 10–15% glycerol to action of the organisms with negatively charged sub-
obtain a suspension greater than 2 on the McFarland strata and, thus, plays an important role in their ad-
scale, and after labeling, immediately frozen (11). hesion (27). The cell surface softness of S. epidermidis
Bacteria can be stored at ª20 æC for 1 to 3 years, at strains grown on agar medium increased sometimes
ª70 æC for 1 to 10 years, while freezing in liquid ni- by a factor of two, compared to the cell surface soft-
trogen preserves bacteria for up to 30 years (10). ness for the strains grown in liquid medium (27).
Storage at ª20 æC should be used as a temporary Therefore, preparation of the inoculum from broth
measure only. After storage, a rapid thawing of cul- or agar growing staphylococci may directly influence
tures and a quick transfer of bacteria to an appropri- the biofilm produced by these bacteria, since ad-
ate growth medium is recommended (10). hesion is the first step in biofilm formation. It is well
Experimental infections using stored bacteria have known that infecting bacteria are often surface-as-
shown that virulence properties are usually main- sociated, and their cell surface can therefore be ex-
tained after lyophilization or freezing at ª70 æC (10). pected to be more similar to that of bacteria grown
However, after storage and thawing it was shown that on solid medium than to that found in organisms
some fastidious bacteria may pose problems in terms grown in liquid media (28). Hence, the inoculum pre-
of viability and stability of their antigenic, molecular pared from bacteria grown on agar probably more
and biochemical properties (10). After retrieval from closely resembles the in vivo situation, compared to
storage, the pure culture of staphylococci may con- the inoculum prepared from bacteria grown in broth.

892
 QUANTIFICATION OF BIOFILM IN MICROTITER PLATES

Irrespective of the method used for preparation of When preparing the inocula, it is also important
the inoculum, the inoculum size should be carefully to be aware that phase variations involving biofilm
determined. It was shown that the size of the inocu- formation are seen in staphylococci (3, 19, 36–
lum considerably influences the amount of biofilm 39). This means that upon subcultivation of single
produced, i.e. biofilm density increases with increas- colony of a biofilm-forming strain, biofilm-negative
ing initial inoculum (3, 29). Thus, standardized in- variants may occur (19, 38). These variations were
oculum should be used. However, in most studies noted after serial passages or after retrieval from
dealing with biofilm quantification, the exact size of storage at ª70 æC (3, 38). Presence of phase vari-
the inoculum has not been determined. We consider ations in staphylococci is easily achieved by observ-
that an initial inoculum of ∂108 CFU/ml is most ing colonial morphology on Congo red agar (3, 19,
suitable. The same inoculum was adopted as stan- 37–39). If more than one colonial morphotype is
dard in the well-known reference procedure for anti- present, the predominant morphotype should be se-
microbial susceptibility testing; procedures for ob- lected for further testing (3). Working cultures may
taining such inoculum are well described in (11). be stored on appropriate medium (i.e. TSA or blood
Therefore, if liquid medium is used for preparation agar) at 4 æC, but due to the phase variations, the
of the inocula, before 1:100 dilution in a fresh broth number of subcultures made before testing must be
medium, the turbidity of the bacterial suspension kept to an absolute minimum (3).
should be adjusted to obtain turbidity optically com-
parable to that of the 0.5 McFarland standard (∂108
CFU/ml). Subsequent 1:100 dilution of this suspen- Medium for biofilm cultivation
sion will result in the final testing inoculum of ∂106 Although a clearcut recommendation for single
CFU. medium appropriate for testing biofilm formation
Another point of interest in preparing inocula is to by different staphylococcal species is difficult, our
avoid inoculation of preexisting cell clusters (3), since choice––based upon direct experience and the overall
they may lead to false-positive results. Therefore, pre- data––would be TSB (which commonly contains
pared cell suspensions must be vortexed. Alterna- 0.25% glucose) supplemented with an additional 1%
tively, preexisting cell clusters in the testing suspen- glucose.
sions may be broken up with a syringe fitted with a Composition of the medium is probably the most
23-gauge needle and brief vortexing (3). Still, some important factor influencing the ability of bacteria to
investigators consider that vortexing may not be suf- produce biofilm under in vitro conditions (3, 31, 32,
ficient to break up all preexisting cell clusters if the 40–43). This was investigated in a number of studies
suspension is made from bacteria grown in standard and the results showed that BHI is sometimes better
TSB medium or bacteria grown on solid medium (30, than TSB (25, 32, 41–43), although it has also been
31). Biofilm formation by staphylococci is generally shown that some strains of a given staphylococcal
enhanced in the presence of glucose and incubation species produce greater biofilm quantities in TSB,
of bacteria in a broth which provides optimal con- while others do so in BHI (32). Nevertheless, TSB is
ditions for biofilm formation, i.e. TSB (standard TSB the medium most frequently used to grow staphylo-
medium commonly contains 0.25% glucose), may coccal biofilms (3, 7, 12–17, 19, 22, 31, 34, 35, 37, 40,
prejudice the test due to possible presence of pre- 44).
existing cell clusters. Accordingly, the inoculated bac- Supplementation of the medium with glucose in-
teria should optimally be precultured in a broth creases the ability of staphylococci to form biofilm
which does not support biofilm formation by a given (7, 16, 25, 30, 32–34). Therefore, liquid medium (TSB
bacterial species. TSB lacking glucose is a broth or BHI) supplemented with an additional 0.25% to
which generally does not support biofilm formation 4% of glucose was frequently used (3, 17, 18, 25, 32,
by staphylococci (16, 32, 33). Thus, it was recom- 42, 45–47). Subinhibitory antibiotic concentrations,
mended that staphylococci should be incubated in ethanol and NaCl also stimulate the production of
TSB medium without glucose or other sugars (30, biofilm and, accordingly, were used as supplements to
31), and diluted in TSB with glucose (see below). basic medium (20, 43, 46, 48–50). In addition, some
However, standard TSB which contains glucose was investigators showed that TSB without additional
used in the great majority of previous studies and, glucose may not be useful for the detection of biofilm
moreover, availability of TSB without glucose may formation by S. aureus (16, 32). To identify biofilm-
pose a problem for some laboratories. It should be positive S. aureus isolates, use of at least two media,
also noted that some staphylococcal strains produce TSBπ1% glucose and BHIπ2% glucoseπ2% sucrose,
biofilm even in TSB without glucose (7, 34, 35), and was recommended (32).
especially some strains had greater adherence in TSB Furthermore, the significant variation in quantities
without glucose than in TSB with glucose (7). Al- of biofilm produced was detected using the TSB me-
though we are aware of the limitations of vortexing, dium manufactured by different suppliers (31, 40,
we propose this as an acceptable solution for elimin- 51), as well as using different lots of the same medium
ation of preexisting cell clusters. (31, 51).

893
 STEPANOVIĆ et al.

Microtiter plate (B) The inoculated plate should be covered with a

The plates for biofilm cultivation should be sterile lid and incubated aerobically for 24 h∫30 min at
flat-bottomed 96-well polystyrene tissue culture- 35 æC–37 æC under static conditions.
treated microtiter plates with a lid. Some investigators incubate microtiter plates over-
The process of biofilm formation begins with the night (15, 17–19, 49), or for 18 h (3, 7, 16, 23, 24,
initial attachment of bacteria to a surface, which is 52), some for 20–24 h (3, 30, 31), while some authors
affected by a number of factors, including the charac- use prolonged incubation of 48 h (3, 53, 55). The dur-
teristics of the surface to which the bacteria adhere. ation of incubation is important since the density of
There are a number of different microtiter-plate types biofilm is dependent on incubation time (3, 25, 47).
designed for different purposes. Microtiter plate pre- We recommend the incubation for 24 h which was
pared for tissue culture implies that its surface was used in the great majority of the biofilm studies (3,
specially treated to allow excellent cell attachment 13, 14, 20–22, 25, 35, 43, 44, 50). Besides the incuba-
and proliferation. Some strains produce biofilm only tion time, other factors such as atmospheric con-
on tissue culture plates, and many strains show weak- ditions, temperature, dynamic/static conditions, etc.
er biofilm positivity on non-tissue culture microtiter also influence the biofilm production by tested bac-
plates (7, 43). In addition, disruption and falling off teria (3, 44, 47, 56–60).
in pieces of the biofilm layer during the washing pro-
cedure is lower in tissue culture-treated microtiter Washing
plates. After incubation, the contents of the wells are de-
The bottom of the microtiter plate may be U- canted into a discard container. Each well is washed
shaped, V-shaped, or flat. Although on rare occasions three times with 300 mL of sterile phosphate-buffered
U-bottomed microtiter plates were used for biofilm saline (PBS; pH 7.2) using an appropriate micropip-
quantification (17, 22, 52), flat-bottomed polystyrene ette. PBS should previously be warmed at room tem-
microtiter plates were most frequently used (7, 18, 25, perature. Following every washing step, the wells are
34, 35, 49, 53). emptied by flicking the plates. Prior to fixation of the
It is important to be aware that tissue culture- biofilm, the plates should be drained in an inverted
treated plates supplied by different manufacturers position.
may not provide identical conditions for biofilm culti- Washing of biofilm is another step of the utmost
vation and quantification (54), although the differ- importance since it is supposed to remove all non-
ence was not noted by all authors (7). Nevertheless, adherent cells while simultaneously providing preser-
in order to keep assay conditions constant, it is sug- vation of the biofilm integrity. There are two critical
gested that plates supplied by the same manufacturer points to be addressed within the recommended
be used. washing protocol: first, number of washings and, sec-
ond, the technique used for washing.
Cultivation of biofilm It is clear that insufficient washing may lead to
(A1) If inoculum is prepared from bacteria grown false-positive results, while, on the other hand, ex-
in broth, then the culture is diluted 1:100 in TSB cessive washing may lead to false-negative results.
supplemented with 1% glucose, and 200 mL is poured Different protocols comprise two (3, 13, 17, 18, 20,
into the well. The negative control wells contain broth 53), three (3, 16, 19, 21, 22, 35, 43, 44, 49, 50), or
only: 200 mL of TSB supplemented with 1% glucose four washing steps (3, 7, 14, 15, 24, 25, 30, 31). We
per well. believe that the two-step washing protocol is not fully
(A2) If inoculum is prepared from bacteria culti- effective while both three- and four-step protocols are
vated on agar plates, then the wells of the microtiter acceptable. The three-step washing protocol is ap-
plate are filled with 180 mL of TSB supplemented plied in laboratories participating in this survey.
with 1% glucose. Thereafter, a 20 mL quantity of pre- As far as the washing technique is concerned, there
viously prepared bacterial suspensions is added to is an obvious variety of methodologies applied. Some
each well. The negative control wells contain broth investigators use a washing roboter, which is able to
only: 200 mL of TSB supplemented with 1% glucose wash very gently, but may lead to false-positive re-
per well. sults. Other authors propose washing of biofilm by
To reduce the possibility of contamination, this immersion of plates into large containers of water or
step should be performed in a biological safety cabi- PBS, since they consider that use of pipettes as well
net. Since phenotypic expression of biofilm forma- as mechanical plate washers may lead to disruption
tion is highly susceptible to various in vitro con- of the biofilm (3). We found that careful pipetting
ditions, in order to minimize errors and provide re- does not compromise integrity of the biofilm and rec-
liable analysis of the data obtained it is essential to ommend washing using micropipettes and emptying
perform testing of each strain at least in triplicate by flicking as a simple and effective method. How-
(three wells per strain). In addition, each test should ever, turning the microtiter plate upside down will
be carried out three times. Six wells should be used not empty the wells because of capillary and inter-
for the negative control. Up to 30 strains may be molecular forces. It is necessary to splash the content
tested per one 96-well microtiter plate. out of the microtiter plate––though this may result in

894
 QUANTIFICATION OF BIOFILM IN MICROTITER PLATES

the formation of aerosol and possible contamination acid mucopolysaccharides, can be used (7). Other
of the environment, and should therefore be carried dyes, such as 0.1%–1% safranin (3, 13, 17, 21, 23, 53)
out very carefully. or trypan blue (14), were also used for the staining.
Irrespective of the technique used, biofilm integrity However, crystal violet was the most frequently used
should be carefully monitored during washing, and stain (3, 7, 12, 15, 16, 18–20, 22, 24, 25, 30, 31, 33,
all wells with clearly disrupted biofilm layer should 35, 37, 43–45, 49, 50). Although crystal violet stains
be excluded from further calculations. only staphylococcal cells and not the slimy material,
It is advisable to measure the growth of tested its use is acceptable because the former washing steps
strains in the microtiter plate prior to washing. The wash off all non-adherent cells, so only the resting
strain may fail to grow or the growth may be reduced adherent cells will be stained.
due to a variety of possible factors and/or technical In some experiments the biofilm was stained for
errors. As a consequence, no biofilm at all or smaller only 30 s (usually when safranin was used as the dye)
quantities of biofilm will be produced. There are sev- (13, 23), in others for 1 min (3), 3 min (21), 5 min
eral ways to measure bacterial growth, but the easiest (31, 35), 10 min (37), or 15 min (22). We consider
way is to measure the turbidity in wells using a that staining for 15 min is most appropriate, since
microtiter-plate reader (21, 61). It is important to the dye has to penetrate through the biofilm.
measure turbidity without previous shaking of the (B) After the microplate is air dried at room tem-
microtiter plate to ensure that the integrity of the bi- perature, the dye bound to the cells should be resol-
ofilm is not disturbed. ubilized, i.e. eluted from attached cells with 150 mL
of 95% ethanol per well. Ethanol should be gently
Fixation added and thereafter the microtiter plate covered
After washing, the remaining attached bacteria with the lid (to minimize evaporation) should be left
should be heat-fixed by exposing them to hot air at at room temperature for at least 30 min without
60 æC for 60 min. shaking.
It was shown that Bouin’s reagent is the most re- The addition of ethanol enables indirect measure-
liable biofilm fixative, followed by air drying (1 h at ment of bacteria attached both to the bottom and
60 æC) (12). However, air drying appears to be the walls of the wells. Biofilms are as diverse as the micro-
method of choice, due to safety concerns regarding organisms which produce them. Floating biofilms, or
Bouin’s reagent, which contains explosive chemicals pellicles, that form at the liquid-air interface of
(12). As an alternative to air drying, fixation by meth- standing cultures represent one type of biofilm (62).
anol can also be used since it has proved to be almost Nevertheless, this deposit is not considered to be in-
as effective as Bouin’s reagent (12). Following fixation dicative of biofilm production by staphylococci (7).
with 150 mL of methanol for 20 min, the microtiter Although the final volume of liquid per well is 200
plates should be emptied by simple flicking, and left mL, in order to exclude the deposit at the liquid-air
to air dry overnight in an inverted position at room interface, only 150 mL of crystal violet and 150 mL of
temperature. ethanol are used for staining and resolubilization of
the dye, respectively. For the same reason, ethanol
should be gently added to the wells, and shaking of
Staining the microtiter plate to speed up the process of resolu-
Two methods were used for staining and sub- bilization is prohibited. Alternatively, 33% glacial
sequent quantification of biofilm. In the first method, acetic acid (35) or methanol can also achieve resolu-
originally described by Christensen et al. (7), only bilization of the dye.
the biofilm formed on the bottom of the well was
measured. The modified Christensen’s method in- Measurement of results
cludes resolubilization of the dye and measures the The optical density (OD) of each well stained with
biofilm formed both on the bottom and walls of the crystal violet is measured at 570 nm using a microtit-
well. Results obtained by these two methods may dif- er-plate reader.
fer significantly (35). For the reasons described below, The problem of common OD readers is that they
we recommend the second approach. measure the OD only at one point in the middle of
(A) The adherent biofilm layer formed in each the well. Thus, if the thickness of the biofilm at that
microtiter-plate well is stained with 150 mL of crystal point significantly differs from the rest of the well,
violet used for Gram staining (2% Hucker crystal vi- the measurement will not be accurate. However, ho-
olet) for 15 min at room temperature. After staining, mogeneous resolubilization of the dye bound to the
the stain should be aspirated with a pipette and ex- bacterial cells in the biofilm layer achieved by the rec-
cess stain should be rinsed off by placing the microtit- ommended protocol enables indirect but precise
er plate under running tap water. Washing is con- measurement of the biofilm production.
tinued until the washings are free of the stain. In some experiments, the product of decolorization
Crystal violet stains bacterial cells but not the was transferred to a new microtiter plate for OD
slimy material (7). To demonstrate the presence of measurement (20). Although this costly and time-
slimy material, Alcian blue, which selectively stains consuming procedure may be beneficial, we consider

895
 STEPANOVIĆ et al.

that it is not of crucial importance for the accuracy is a strong biofilm producer. There are other strong
of the results. biofilm-producing strains, such as S. epidermidis
CCM 7221, deposited in the Czech Collection of
Interpretation of the results Microorganisms, which also may be used as the con-
The interpretation of obtained results requires trol strain.
definition of the cut-off value that separates biofilm-
producing from non-biofilm-producing strains. Some
investigators calculate this on the basis of three stan- CONCLUSIONS
dard deviations above that of uninoculated medium
(negative control) (7), or above that of biofilm-negative The presented protocol incorporated accumu-
control strain (14, 52). Other investigators use prede- lated knowledge on assessment of biofilm pro-
fined values for the cut-off point: 0.1 (22, 30, 31, 33), duction by staphylococci gained both by direct
0.12 (3, 19, 24, 45, 49, 50), or 0.2 (21). In some experi- experience in the field of biofilm research in lab-
ments, the result of the positive control strain was used
as the starting point for calculation (15), while in an- oratories participating in this survey as well as
other a strain was considered positive if its OD was by comprehensive analysis of methods for as-
twice that of the negative control strain (18). saying biofilm production by staphylococci
Our recommendation is that results obtained available in the literature. The study provides
should be averaged and expressed as numbers. The details of all the steps involved in quantification
average OD values should be calculated for all tested of biofilm formation by staphylococci, which
strains and negative controls, since all tests are per-
are quite frequently lacking in the literature,
formed in triplicate and repeated three times. Second,
the cut-off value (ODc) should be established. It is and identifies the most important problems met
defined as three standard deviations (SD) above the when performing the assay. This should make
mean OD of the negative control: ODcΩaverage OD the procedure more easily applicable, which is
of negative controlπ(3¿SD of negative control). Fi- of particular importance for those newly start-
nal OD value of a tested strain is expressed as average ing in this field of research. The recognition of
OD value of the strain reduced by ODc value (ODΩ a single medium suitable for quantification of
average OD of a strain ªODc). ODc value is calcu-
lated for each microtiter plate separately. If a negative
biofilm production by all staphylococci and
value is obtained, it should be presented as zero, clear-cut recommendations for washing pro-
while any positive value indicates biofilm production. cedure are the two most critical points. It is
For easier interpretation of the results, strains may quite clear that complete standardization of the
be divided into the following categories (35): no bi- procedure for quantification of biofilm produc-
ofilm producer (0), weak biofilm producer (π or 1), tion by staphylococci, in particular the two
moderate biofilm producer (ππ or 2) and strong bi- above-mentioned points, will require further
ofilm producer (πππ or 3), based upon the previously
calculated OD values (for this type of calculation aver- scientific effort. All variations in testing con-
age OD value of the strain should not be reduced by ditions primarily influence biofilm formation by
ODc value): OD ⱕODcΩno biofilm producer; ODc weak––and some moderate––biofilm producers.
⬍OD ⱕ2¿ODcΩweak biofilm producer; 2¿ODc An important limitation of this method, as well
⬍OD ⱕ4¿ODcΩmoderate biofilm producer; 4¿ODc as all other in vitro methods, for biofilm meas-
⬍ODΩstrong biofilm producer. urement is that it does not exactly reflect in vivo
We consider that the important advantage of the
situations. Guidelines for extrapolation of the
described guidelines for interpretation of the results
is the use of the negative control as the starting point data obtained under ideally optimized in vitro
for all calculations. Application of positive or nega- testing conditions, which provide maximal bi-
tive control strains, as the reference for calculations, ofilm formation, to an actual in vivo situation
may lead to certain inconsistencies due to the numer- remain to be established.
ous above-mentioned factors influencing the biofilm
production. The work of SS, DV, SD, and IC was supported by
the Ministry of Science, Republic of Serbia (project
Quality control grant no. 145069B).
Four strains deposited by Christensen et al. (7) in
the ATCC collection (ATCC 35981, ATCC 35982,
ATCC 35983 and ATCC 35984) may be used as con- REFERENCES
trol strains for quality assurance of the test. The most
frequently used strain is S. epidermidis ATCC 35984 1. Donlan RM. Biofilms: microbial life on surfaces.
(designated by Christensen et al. as RP62A), which Emerg Infect Dis 2002;8:881–90.

896
 QUANTIFICATION OF BIOFILM IN MICROTITER PLATES

2. Donlan RM, Costerton JW. Biofilms: survival tors in Staphylococcus caprae strains responsible
mechanisms of clinically relevant microorgan- for human bone infections following implantation
isms. Clin Microbiol Rev 2002;15:167–93. of orthopaedic material. Microbiology 1999;145:
3. Deighton MA, Capstick J, Domalewski E, van 2033–42.
Nguyen T. Methods for studying biofilms pro- 16. Baldassarri L, Bertuccini L, Ammendolia MG,
duced by Staphylococcus epidermidis. Methods Arciola CR, Montanaro L. Effect of iron limi-
Enzymol 2001;336:177–95. tation on slime production by Staphylococcus au-
4. Arciola CR, Campoccia D, Gamberini S, Cervel- reus. Eur J Clin Microbiol Infect Dis 2001;20:343–
lati M, Donati E, Montanaro L. Detection of 5.
slime production by means of an optimised Congo 17. Cramton SE, Gerke C, Gotz F. In vitro methods to
red agar plate test based on a colourimetric scale study staphylococcal biofilm formation. Methods
in Staphylococcus epidermidis clinical isolates Enzymol 2001;336:239–55.
genotyped for ica locus. Biomaterials 2002;23: 18. Fowler VG Jr, Fey PD, Reller LB, Chamis AL, Co-
4233–9. rey GR, Rupp ME. The intercellular adhesin locus
5. Harraghy N, Seiler S, Jacobs K, Hannig M, Meng- ica is present in clinical isolates of Staphylococcus
er MD, Herrmann M. Advances in in vitro and in aureus from bacteremic patients with infected and
vivo models for studying the staphylococcal fac- uninfected prosthetic joints. Med Microbiol Im-
tors involved in implant infections. Int J Artif Or- munol 2001;189:127–31.
gans 2006;29:368–78. 19. Ziebuhr W, Loessner I, Krimmer V, Hacker J.
6. Christensen GD, Simpson WA, Bisno AL, Beach- Methods to detect and analyze phenotypic vari-
ey EH. Adherence of slime-producing strains of ation in biofilm-forming staphylococci. Methods
Staphylococcus epidermidis to smooth surfaces. Enzymol 2001;336:195–205.
Infect Immun 1982;37:318–26. 20. Beenken KE, Blevins JS, Smeltzer MS. Mutation
7. Christensen GD, Simpson WA, Younger JJ, Bad- of sarA in Staphylococcus aureus limits biofilm
dour LM, Barrett FF, Melton DM, et al. Adher- formation. Infect Immun 2003;71:4206–11.
ence of coagulase-negative staphylococci to plas- 21. Moretro T, Hermansen L, Holck AL, Sidhu MS,
tic tissue culture plates: a quantitative model for Rudi K, Langsrud S. Biofilm formation and the
the adherence of staphylococci to medical devices. presence of the intercellular adhesion locus ica
J Clin Microbiol 1985;22:996–1006. among staphylococci from food and food pro-
8. Mack D, Becker P, Chatterjee I, Dobinsky S, Kno- cessing environments. Appl Environ Microbiol
bloch JK, Peters G, et al. Mechanisms of biofilm 2003;69:5648–55.
formation in Staphylococcus epidermidis and 22. Vasudevan P, Nair MK, Annamalai T, Venkitana-
Staphylococcus aureus: functional molecules, re- rayanan KS. Phenotypic and genotypic character-
gulatory circuits, and adaptive responses. Int J ization of bovine mastitis isolates of Staphylococ-
Med Microbiol 2004;294:203–12. cus aureus for biofilm formation. Vet Microbiol
9. Morgan CA, Herman N, White PA, Vesey G. Pres- 2003;92:179–85.
ervation of micro-organisms by drying; a review. J 23. Tormo MA, Knecht E, Gotz F, Lasa I, Penades
Microbiol Methods 2006;66:183–93. JR. Bap-dependent biofilm formation by patho-
10. De Paoli P. Biobanking in microbiology: from genic species of Staphylococcus: evidence of hori-
sample collection to epidemiology, diagnosis and zontal gene transfer? Microbiology 2005;151:
research. FEMS Microbiol Rev 2005;29:897–910. 2465–75.
11. Courvalin P, Soussy CJ, editors. 1996 Report of 24. Arciola CR, Campoccia D, Baldassarri L, Donati
the Comité de l’Antibiogramme de la Société ME, Pirini V, Gamberini S, et al. Detection of bi-
Française de Microbiologie. Clin Microbiol Infect ofilm formation in Staphylococcus epidermidis
1996; 2:(Supplement 1). from implant infections. Comparison of a PCR-
12. Baldassarri L, Simpson WA, Donelli G, Chris- method that recognizes the presence of ica genes
tensen GD. Variable fixation of staphylococcal with two classic phenotypic methods. J Biomed
slime by different histochemical fixatives. Eur J Mater Res A 2006;76:425–30.
Clin Microbiol Infect Dis 1993;12:866–8. 25. Mathur T, Singhal S, Khan S, Upadhyay DJ, Fat-
13. Heilmann C, Gerke C, Perdreau-Remington F, ma T, Rattan A. Detection of biofilm formation
Gotz F. Characterization of Tn917 insertion mu- among the clinical isolates of staphylococci: an
tants of Staphylococcus epidermidis affected in bi- evaluation of three different screening methods.
ofilm formation. Infect Immun 1996;64:277–82. Indian J Med Microbiol 2006;24:25–9.
14. Mulder JG, Degener JE. Slime-producing prop- 26. Cheung AL, Fischetti VA. Variation in the expres-
erties of coagulase-negative staphylococci isolated sion of cell wall proteins of Staphylococcus aureus
from blood cultures. Clin Microbiol Infect 1998; grown on solid and liquid media. Infect Immun
4:689–94. 1988;56:1061–5.
15. Allignet J, Galdbart JO, Morvan A, Dyke KG, 27. Kiers PJ, Bos R, van der Mei HC, Busscher HJ.
Vaudaux P, Aubert S, et al. Tracking adhesion fac- The electrophoretic softness of the surface of

897
 STEPANOVIĆ et al.

Staphylococcus epidermidis cells grown in a liquid Fitzpatrick F, Humphreys H, et al. Genetic and
medium and on a solid agar. Microbiology 2001; phenotypic analysis of biofilm phenotypic vari-
147:757–62. ation in multiple Staphylococcus epidermidis iso-
28. Schleifer KH, Kroppenstedt RM. Chemical and lates. J Med Microbiol 2004;53:367–74.
molecular classification of staphylococci. Soc 40. Rohde H, Knobloch JK, Horstkotte MA, Mack
Appl Bacteriol Symp Ser 1990;19:9S–24S. D. Correlation of biofilm expression types of
29. Stepanovic S, Djukic N, Opavski N, Djukic S. Sig- Staphylococcus epidermidis with polysaccharide
nificance of inoculum size in biofilm formation by intercellular adhesin synthesis: evidence for in-
staphylococci. New Microbiol 2003;26:129–32. volvement of icaADBC genotype-independent
30. Mack D, Siemssen N, Laufs R. Parallel induction factors. Med Microbiol Immunol 2001;190:105–
by glucose of adherence and a polysaccharide 12.
antigen specific for plastic-adherent Staphylococ- 41. Stepanovic S, Dakic I, Opavski N, Jezek P, Ranin
cus epidermidis: evidence for functional relation to L. Influence of the growth medium composition
intercellular adhesion. Infect Immun 1992;60: on biofilm formation by Staphylococcus sciuri.
2048–57. Ann Microbiol 2003;53:63–74.
31. Mack D, Bartscht K, Fischer C, Rohde H, de 42. Hola V, Ruzicka F, Votava M. Impact of surface
Grahl C, Dobinsky S, et al. Genetic and biochem- coating on the adherence of slime producing and
ical analysis of Staphylococcus epidermidis biofilm nonproducing Staphylococcus epidermidis. New
accumulation. Methods Enzymol 2001;336:215– Microbiol 2004;27:305–8.
39. 43. Kennedy CA, O’Gara JP. Contribution of culture
32. Knobloch JK, Horstkotte MA, Rohde H, Mack media and chemical properties of polystyrene
D. Evaluation of different detection methods of tissue culture plates to biofilm development by
biofilm formation in Staphylococcus aureus. Med Staphylococcus aureus. J Med Microbiol 2004;53:
Microbiol Immunol 2002;191:101–6. 1171–3.
33. Dobinsky S, Kiel K, Rohde H, Bartscht K, Knob- 44. Perez-Giraldo C, Rodriguez-Benito A, Moran FJ,
loch JK, Horstkotte MA, et al. Glucose-related Hurtado C, Blanco MT, Gomez-Garcia AC.
dissociation between icaADBC transcription and Influence of the incubation atmosphere on the
biofilm expression by Staphylococcus epidermidis: production of slime by Staphylococcus epiderm-
evidence for an additional factor required for idis. Eur J Clin Microbiol Infect Dis 1995;14:359–
polysaccharide intercellular adhesin synthesis. J 62.
Bacteriol 2003;185:2879–86. 45. Gelosia A, Baldassarri L, Deighton M, van Nguy-
34. Deighton MA, Balkau B. Adherence measured by en T. Phenotypic and genotypic markers of
microtiter assay as a virulence marker for Staphy- Staphylococcus epidermidis virulence. Clin Micro-
lococcus epidermidis infections. J Clin Microbiol biol Infect 2001;7:193–9.
1990;28:2442–7. 46. Fitzpatrick F, Humphreys H, O’Gara JP. Evidence
35. Stepanovic S, Vukovic D, Dakic I, Savic B, Svabic- for icaADBC-independent biofilm development
Vlahovic M. A modified microtiter-plate test for mechanism in methicillin-resistant Staphylococ-
quantification of staphylococcal biofilm forma- cus aureus clinical isolates. J Clin Microbiol 2005;
tion. J Microbiol Methods 2000;40:175–9. 43:1973–6.
36. Christensen GD, Baddour LM, Madison BM, Pa- 47. Hola V, Ruzicka F, Votava M. Dynamic of biofilm
risi JT, Abraham SN, Hasty DL, et al. Colonial formation in relation to nutrition, temperature
morphology of staphylococci on Memphis agar: and time. Scripta Medica – in press.
phase variation of slime production, resistance to 48. Knobloch JK, Bartscht K, Sabottke A, Rohde H,
beta-lactam antibiotics, and virulence. J Infect Dis Feucht HH, Mack D. Biofilm formation by
1990;161:1153–69. Staphylococcus epidermidis depends on functional
37. Ziebuhr W, Heilmann C, Gotz F, Meyer P, Wilms RsbU, an activator of the sigB operon: differential
K, Straube E, et al. Detection of the intercellular activation mechanisms due to ethanol and salt
adhesion gene cluster (ica) and phase variation in stress. J Bacteriol 2001;183:2624–33.
Staphylococcus epidermidis blood culture strains 49. Cho SH, Naber K, Hacker J, Ziebuhr W. Detec-
and mucosal isolates. Infect Immun 1997;65:890– tion of the icaADBC gene cluster and biofilm for-
6. mation in Staphylococcus epidermidis isolates
38. Ziebuhr W, Krimmer V, Rachid S, Lossner I, Gotz from catheter-related urinary tract infections. Int
F, Hacker J. A novel mechanism of phase variation J Antimicrob Agents 2002;19:570–5.
of virulence in Staphylococcus epidermidis: evi- 50. Fitzpatrick F, Humphreys H, Smyth E, Kennedy
dence for control of the polysaccharide intercellu- CA, O’Gara JP. Environmental regulation of bi-
lar adhesin synthesis by alternating insertion and ofilm formation in intensive care unit isolates of
excision of the insertion sequence element IS256. Staphylococcus epidermidis. J Hosp Infect 2002;
Mol Microbiol 1999;32:345–56. 52:212–8.
39. Handke LD, Conlon KM, Slater SR, Elbaruni S, 51. Mack D, Rohde H, Dobinsky S, Riedewald J, Ne-

898
 QUANTIFICATION OF BIOFILM IN MICROTITER PLATES

delmann M, Knobloch JK, et al. Identification of aerobic conditions. J Clin Microbiol 1990;28:
three essential regulatory gene loci governing ex- 2578–9.
pression of Staphylococcus epidermidis polysac- 57. Stepanovic S, Vukovic D, Jezek P, Pavlovic M, Sva-
charide intercellular adhesin and biofilm forma- bic-Vlahovic M. Influence of dynamic conditions
tion. Infect Immun 2000;68:3799–807. on biofilm formation by staphylococci. Eur J Clin
52. Fox LK, Zadoks RN, Gaskins CT. Biofilm pro- Microbiol Infect Dis 2001;20:502–4.
duction by Staphylococcus aureus associated with 58. Gotz F. Staphylococcus and biofilms. Mol Micro-
intramammary infection. Vet Microbiol 2005;107: biol 2002;43:1367–78.
295–9. 59. Stepanovic S, Djukic N, Djordjevic V, Djukic S.
53. Ozerdem Akpolat N, Elci S, Atmaca S, Akbayin Influence of the incubation atmosphere on the
H, Gul K. The effects of magnesium, calcium and production of biofilm by staphylococci. Clin
EDTA on slime production by Staphylococcus epi- Microbiol Infect 2003;9:955–8.
dermidis strains. Folia Microbiol (Praha) 2003;48: 60. Svabic-Vlahovic M, Dakic I, Vukovic D, Ste-
649–53. panovic S. Biofilm production by Staphylococcus
54. Stepanovic S, Cirkovic I, Mijac V, Svabic-Vlahovic epidermidis at 37 æC and 32 æC. 13th European
M. Influence of the incubation temperature, at- Congress of Clinical Microbiology and Infectious
mosphere and dynamic conditions on biofilm for- Diseases, Glasgow, United Kingdom, May 10–13,
mation by Salmonella spp. Food Microbiol 2003. Clin Microbiol Infect 2003;9(Suppl 1-CD),
2003;20:339–43. P926.
55. Petrelli D, Zampaloni C, D’Ercole S, Prenna M, 61. Lindqvist R. Estimation of Staphylococcus aureus
Ballarini P, Ripa S, et al. Analysis of different gen- growth parameters from turbidity data: character-
etic traits and their association with biofilm for- ization of strain variation and comparison of
mation in Staphylococcus epidermidis isolates methods. Appl Environ Microbiol 2006;72:4862–
from central venous catheter infections. Eur J Clin 70.
Microbiol Infect Dis 2006;25:773–81. 62. Branda SS, Vik S, Friedman L, Kolter R. Bio-
56. Barker LP, Simpson WA, Christensen GD. Differ- films: the matrix revisited. Trends Microbiol 2005;
ential production of slime under aerobic and an- 13:20–6.

899