Microbes and Infection, 2, 2000, 1523−1535

© 2000 Éditions scientifiques et médicales Elsevier SAS. All rights reserved

S1286457900013071/REV

Review

Use of fluorescent probes to assess

physiological functions of bacteria
at single-cell level
Fabien Joux, Philippe Lebaron*
Observatoire océanologique, université Pierre et Marie Curie, UMR-7621 CNRS, Institut national des sciences de l’Univers,
BP 44, F-66651 Banyuls-sur-Mer cedex, France

ABSTRACT – A wide diversity of fluorescent probes is currently available to assess the

physiological state of microorganisms. The recent development of techniques such as solid-phase
cytometry, the increasing sensitivity of fluorescence tools and multiparametric approaches
combining taxonomic and physiological probes have improved the effectiveness of direct methods in
environmental and industrial microbiology. © 2000 Éditions scientifiques et médicales Elsevier
SAS
bacteria / fluorescent probes / physiology / cytometry

1. Introduction fluorescent dye technology offering probes for a variety of

One of the most basic questions that microbiologists
address is whether a bacterial cell is alive or dead. The
answer to this question is far from simple and the question
remains unanswered after 20 years of intense research and
permanent controversy [1]. Life is generally characterised
by: (i) the presence of structure; (ii) changeable genetic
information; (iii) metabolism or functional activity, and (iv)
the ability to reproduce and grow [2]. However, living
bacteria are generally only characterised by their ability to
reproduce, probably because the capacity of a cell to
multiply as determined by culture has long been the single
method available to microbiologists for bacterial viability
assessment. Therefore, culturability and viability are often
considered synonymous terms.
During the last two decades, an increasing need for
alternative techniques has been stimulated by: (i) the need
for real-time monitoring methods for viability assessment
[3]; (ii) the existence of injured and ‘viable but non-
culturable’ (VNC) cells sometimes encountered when cells
are stressed (e.g. heat, oxidative, osmotic stress, or starva-
tion) [1, 4], and (iii) the discovery of an important diversity
of species for which suitable in vitro culture conditions
have not yet been defined and thus, which cannot be
recognised as ‘alive or dead’ by the culturable/non-
culturable dichotomy [5]. The development of these alter-
native methods was also concomitant with advances in
* Correspondence and reprints.
E-mail address: lebaron@obs-banyuls.fr (Philippe Lebaron).
cellular functions. The VNC state was initially suggested
by Roszak et al. [4] for readily culturable bacteria which
under particular circumstances become non-culturable
but retain ‘viability’. Nevertheless, the viability of VNC
cells was never defined by culture but only by the detec-
tion of some forms of activity in cells, which are not
directly related to the capacity of these cells to reproduce.
For these reasons, the term ‘active but non-culturable’
(ANC) might be more precise than viable but non-
culturable [1].
Fluorescence-based methods have remained very use-
ful for a wide diversity of applications ranging from indus-
trial to environmental microbiology. These tools are used
for viability/activity assessment in food, pharmaceutical
and cosmetic industries, and in the natural environment,
including fresh and marine waters. The increased use of
fluorescent probes is also due to improvements in the
quantitative and qualitative sensitivity of instruments. The
aim of this review is to describe the different strategies
used for the detection of cell fluorescence, the different
cellular target sites and fluorescent probes which are
actually used in activity assays, and the main applications
of these probes.
2. Analytical instruments
Instrumentation strategy is of great importance, since
all instruments have different ranges of application,
depending on their quantitative and qualitative sensitivi-

Microbes and Infection 1523

2000, 1523-1535
Review Joux and Lebaron

Table I. Characteristics of the different fluorescent dyes exposed in the text.

Characteristic Absorption (nm) Emission (nm) Molecular weight
Dehydrogenase activity (hydrolysis product)
CTC (CTC formazan, CTF) 450 580–660 332
Esterase activity (hydrolysis product)
FDA (fluorescein) 473 514 416
CFDA (carboxyfluoroscein) 492 517 460
CFDA-AM (carboxyfluoroscein) 492 517 532
BCECF-AM (BCECF) 482 520 615
Calcein-AM (calcein) 494 517 995
ChemChrome (*) 488 520 *
Membrane potential
Rh123 507 529 381
DiOC6(3) 484 501 573
DiBAC4(3) 493 516 517
Oxonol VI 599 634 316
Probe efflux
Ethidium bromide 518 605 394
Membrane integrity
SYTO-9 (membrane permeant stain) * (blue) * (green) *
SYTO-13 (membrane permeant stain) 488 509 ∼ 400
Propidium iodide 535 617 668
Sytox Green 502 523 600
PO-PRO-3 539 567 605
CSE * (blue) * (orange) *
* Unspecified by the manufacturer. Source of information: Chemunex (Ivry-sur-Seine, Paris, France), Molecular Probes (Eugene, Oreg., USA), Polysciences
Europe (Germany).

ties. The quantitative limits, accuracy, speed of analysis,
type of samples which can be analysed and light sources
of these instruments are important selection criteria. All
fluorochromes are characterised by their absorption and
emission spectra (table I). The characteristics of selected
dyes must be compatible with those of the instrumentation
used for fluorescence detection.
Epifluorescence microscopy (EFM) of fluorochrome-
stained cells has become the standard method by which
bacteria are enumerated. Most microscopes are equipped
with a mercury or a mercury–xenon lamp. The xenon
lamp provides a continuous spectrum, whereas the mer-
cury lamp emits spectral lines (365, 405, 435, 546 and
578 nm) with a weak background continuum at other
wavelengths. Xenon lamps should be used in preference
to mercury lamps only for dyes whose absorption spectra
does not match any of the strong mercury lines. The main
limitation in the use of microscopy is that it is time con-
suming and subjective [6]. Moreover, the detection limit of
microscopy is low and it becomes very tedious to detect
less than 103 targeted cells per cm2 of filter – this limit
corresponding to the detection of one cell per microscopic
field. Thus, bacteria often have to be much more concen-
trated for microscopic examination than for plate counting
after filtration. This can be quite problematic for samples
containing non-cellular particles and when targeted cells
are diluted within non-targeted cells. Therefore, micro-
scopic examination does not facilitate the simple detec-
tion of rare events.
Flow cytometry (FCM) has became more popular and
offers the ability to perform analysis on individual cells at
numbers which begin to be more representative of nature
1524
[20]. FCM allows the rapid and simultaneous measure-
ment of various cell parameters with a precision of a few
per cent. Analyses are commonly performed at a flow rate
between 10 and 100 µL per min and count rates ranging
from 100 to 1 000 cells/s. Most flow cytometers are
equipped to sort cells with specific characteristics for
further analyses. However, FCM is not adapted to the
detection of rare events since it is difficult to detect less
than 100 bacteria per mL, which is still 10–100-fold better
than current microscopic procedures. For instance, at a
concentration of 100 bacteria per mL and at a mean flow
rate of 50 µL/min, 20 min are needed to detect 100 events.
Furthermore, when targeted cells represent less than one
cell in one thousand or one cell in one million of non-
targeted cells, flow cytometric analysis may range from
difficult to impossible [7].
Laser scanning cytometer (LSC) offers the advantage of
rapid detection of bacteria on filters. LSC is a microscope-
based cytofluorometer which possesses the attributes of
both flow and image cytometry [8]. A 10-µm laser spot
allows the measurement of emitted fluorescence in addi-
tion to forward scattered light every 5 or 10 µm over
sequential fields. The rapid scanning speed (0.3 m/s) and
the low-level resolution enable faster processing of the
data. However, the field-by-field scanning is an intrinsic
limitation, and the 10-µm step does not always allow
resolution of individual cells. This is probably why this
instrument has received little application in bacteriology.
Recently, a solid-phase cytometer (SPC) equivalent to a
rapid laser scanning device (ChemScan; Chemunex, Ivry
sur Seine, France) has been developed [9, 10]. SPC differs
from LSC in that the former is not a microscope-based
Microbes and Infection
2000, 1523-1535
Fluorescent physiological probes for bacteria
instrument. Cells are concentrated on a membrane filter as
for microscopy, and sample illumination and data collec-
tion are achieved without the use of an objective. This
allows high scanning speeds (1 m/s) and scanning of the
whole membrane (25-mm diameter) in one scan pattern
with a fully overlapping laser beam. The scanning is
performed in under 3 min and any event detected in the
scanning procedure can be positioned for observation in
the visual field of a conventional epifluorescence micro-
scope in which the stage is driven by the instrument’s
computer [11]. This procedure is of great interest since it
allows the possibility of an immediate visual control and
confirmation of the result. The system is also able to
differentiate between labelled microorganisms and autof-
luorescent particles present in the sample based on the
optical and electronic characteristics of the generated
signals. SPC is increasingly used in routine water analyses
for the detection of rare events (Escherichia coli,
Cryptosporidium spp., Giardia spp.), in some industrial
applications for real-time control of pharmaceutical-grade
water and in different histological and cytological controls
[11]. This technique allows the detection of a single cell on
a 25-mm diameter membrane after filtration.
Confocal scanning laser microscopy (CSLM) is a com-
puterised microscope that couples a laser light source to a
light microscope. This technique allows for the generation
of three-dimensional digital images of microorganisms.
This technique combined with the use of fluorescent
probes has proven useful for the analyses of the three-
dimensional distribution of total and active cells in solid
structures (e.g., biofilms, aggregates, endosymbionts,
rhizophere and soil) [12]. However, CSLM is neither
adapted to routine applications for cell counting nor to the
detection of rare events.
The argon ion laser is the most widely used light source
for FCM, LSC, SPC and CSLM. Argon ion lasers provide
emission lines at several wavelengths ranging from 351.1
to 514.5 nm: the most widely used is the single line at
488 nm. In the case of instruments equipped with a single
laser source, the excitation wavelength is fixed and the
strategy for staining is limited to the range of probes and
stains excitable at this wavelength. The use of probes with
contrasting wavelengths is usually required for multipa-
rameter measurement (e.g., combination of a nucleic acid
dye for the quantification of total bacteria with physiologi-
cal and taxonomic probes) (see section 6). In this instance,
contrasting wavelength means a combination of excita-
tion and emission wavelengths, which allows discrimina-
tion of each probe in the presence of the others. Double
staining procedures with a single laser excitation source
(often 488 nm) are limited since both dyes may have a
common excitation wavelength and different emission
wavelengths with a minimal overlap.
The choice of an instrument is primarily driven by its
technical characteristics but also by its cost. EFM is now
considered traditional equipment for fluorescence appli-
cations. FCM is more frequently used not only for research
but also in many industrial applications. CSLM is expen-
sive and not adapted to routine bacterial analyses. Other
instruments are still under validation for different applica-
tions. The quantitative sensitivity of the instrument and the
Microbes and Infection
2000, 1523-1535
Review
filterability of the products to be analysed will also influ-
ence the choice of the instrument. The lack of consistency
sometimes reported between microscopic and flow cytom-
etry counts of total bacteria in microcosm studies and
mainly the higher variations of FCM counts ([13], unpub-
lished data from our group) may be due to day-to-day
fluctuations in flow cytometry operations, which require
strong quality controls. When controls are made, EFM,
FCM and SPC provide accurate counts of bacterial cells
([9] Lebaron, unpublished data).
3. Definition of terms
Defining cell death and cell viability is philosophically
and experimentally difficult. A viable cell should be con-
sidered a cell capable of fission to produce similarly viable
progeny under realisable culture conditions [1]. More
generally, the term ‘viable’ should be restricted to circum-
stances where evidence that a cell is able to divide is a
priori or a posteriori provided. From an operational point
of view, viability is demonstrated by culture, and cell
proliferation can be detected at both macroscopic (mac-
rocolonies or turbidity) or microscopic (microcolonies)
scale. The main advantage of direct methods based on
fluorescent probes is the lack of incubation. Thus, there is
clear evidence that methods based on the use of physi-
ological probes cannot directly address cell division but
only the presence of some active functions or the integrity
of cell structures. Therefore, cells in which any kind of
metabolic activity can be detected should be called active
cells. When metabolic activity is not detectable but integ-
rity of essential envelope structures is preserved, cells are
called inactive cells. These cells may be active but not
enough to be detected as active cells or may be dying cells
with intact membranes. Lastly, cells with damaged mem-
branes are considered dead cells, whereas those with
intact membranes are considered intact cells. Dead cells
can be detected until cell lysis and are called ‘ghost’ cells
when cellular envelopes are maintained, whereas the
nucleoid is lost [14]. Most of these cellular categories were
already described by Neve-von Caron et al. [15].
The question of whether the active but non-culturable
(ANC) cells are able or not to return to a growing state
when their environment becomes favourable is at the
centre of the current debate on the ANC state (for a
complete review, see [1]). The demonstration of the revers-
ibility of this non-growing state is based on resuscitation
studies but most of these are open to criticism. In most
studies, it is generally uncertain whether the ANC cells are
capable of resuscitation or whether the increase in cell
numbers reported in a large number of publications on this
issue is merely the result of growth of a few viable cells [1].
This is due to the existence of an important heterogeneity
of physiological states within a population and to our
inability to provide the proof that a cell in a given cellular
state as determined by any given method is able to grow
and to divide.
1525
Review
Figure 1. Different cellular target sites for physiological and taxonomic fluorescent dyes.
4. Physiological target sites
This section presents the different probes used to assess
different physiological functions and cellular structures.
The interpretation of these dyes in terms of cellular activity
or viability is discussed in more details in sections 5 and 6.
Figure 1 summarises the different physiological target sites
of these probes.
4.1. Membrane potential
The electrochemical potential occurring through the
plasma membrane of metabolising bacteria is generated
by respiration or by ATP hydrolysis. It results from the
selective permeability of biological membranes to a vari-
ety of cations and anions leading to a difference of electric
potential across the membrane. Inside, the cell is nega-
tively charged compared with outside the cell, and mem-
brane potential (MP) plays a central role in different cell-
life processes (ATP synthesis, active transport, mobility,
regulation of intracellular pH, etc.). Voltage-sensitive dyes
have been developed to measure MP in bacteria. Depend-
ing on the charge of the dye, they are accumulated in
polarised (cationic dyes) or depolarised (anionic dyes)
cells. Reliability of staining is confirmed by observing if
dye uptake is sensitive to uncouplers (e.g., carbonyl cya-
nide m-chlorophenyl hydrazone; CCCP) or ionophores
(e.g., nigericin, valinomycin). In appropriate conditions,
the amount of dye taken up can be directly related to the
level of energy metabolism in the cell.
1526
Joux and Lebaron
Rhodamine 123 (Rh123) is a lipophilic, cationic dye
commonly used to detect MP [16]. However, staining with
Rh123 often requires a pretreatment step of the cells,
generally performed by adding EDTA, to permeabilise the
outer membrane of Gram-negative bacteria [17]. When
antibiotic or disinfectant treatments are studied, the per-
meabilisation step can introduce some bias by enhancing
the toxic effects of these compounds (i.e. they penetrate
more easily inside the cells). Furthermore, the pretreat-
ment conditions may vary depending on the environment
(e.g., salinity) and between species. This is why MP dyes
have received little applications at the community level.
An additional problem is that Rh123 staining requires
several cell-washing steps, which are time consuming and
may result in cell losses. The cationic carbocyanine dyes
(e.g., 3,3'-dihexyloxacarbocyanine; DiOC6(3)) have also
been used to estimate MP of bacteria. Advantages of these
dyes are the absence of permeabilization and washing
steps. However, nonspecific binding of carbocyanine dyes
to hydrophobic regions of the cell and quenching of the
fluorescence of intracellular dye have been reported [17].
Thus, careful calibration of the staining procedure is
required to avoid false positive signals.
MP can also be determined by the anionic lipophilic
oxonols. Accumulation inside bacterial cells is favoured
by a reduction in the magnitude of the MP, allowing dye
molecules to concentrate within the cell, and bind to
lipid-rich components. Bis-(1,3-dibutylbarbituric acid) tri-
methine oxonol [DiBAC4(3)] (BOX) has been reported to
Microbes and Infection
2000, 1523-1535
Fluorescent physiological probes for bacteria
be useful to detect depolarised cells of numerous Gram-
positive and Gram-negative bacterial species. In some
protocols, oxonol dyes combine a pretreatment with EDTA
or EGTA 1–5 mM to increase membrane permeability [15,
18]. Without permeabilisation, uptake of oxonols would
be more related to membrane integrity rather than mem-
brane depolarisation [15]. Staining only dead cells with
oxonols may be difficult when the treatments studied lead
to the disruption of the cells (e.g., surfactants, antibiotics).
In these cases, it is important to determine total cell counts
and then to deduce the remaining fraction of living cells.
For instance, Comas and Vives-Rego [19] suggested the
simultaneous detection of depolarised and total cell popu-
lation by combining BOX with the nucleic acid dye SYTO-
17.
Applications of MP dyes to natural samples are scarce
and inconclusive [20], probably because of both the effects
of the permeabilisation step, which vary depending on
species and the interaction between charged dyes and
chemical compounds present in the sample.
4.2. Enzyme activity
4.2.1. Dehydrogenase activity
Cell-specific assays to detect the respiratory activity of
bacteria have been developed based on the use of different
tetrazolium salts. Tetrazolium dyes are reduced from a
colourless complex to a brightly coloured, intracellular,
formazan precipitate by components of the electron trans-
port system and/or a variety of dehydrogenase enzymes
present in active bacterial cells. Since electron transport is
directly related to cellular energy metabolism in respiring
cells, the ability of cells to reduce tetrazolium compounds
can be considered an indicator of bacterial activity. A
variant approach is the use of the redox dye 5-cyano-2,3-
ditolyl tetrazolium chloride (CTC). CTC is reduced by
bacteria to a water-insoluble, red fluorescent formazan
product. It allows the quantification of the metabolic
activity of bacteria under both aerobic and different anaero-
bic conditions [21, 22]. However, different factors may
affect the CTC reduction during incubation [23–25].
Although CTC counts are commonly determined by epif-
luorescence microscopy, flow cytometry can be used when
possible, to overcome the rapid fading of the fluorescence
signal [26], yielding equal or higher counts [25, 27]. In
contrast to other probes, samples can be fixed after stain-
ing and stored at –20 °C until analysis.
The addition of nutrients during incubation with CTC is
sometimes used to improve the fluorescent signal and/or
the number of respiring cells [28]. In this case, controls are
needed to ensure that growth does not occur during incu-
bation or that growth can be controlled by means of
antibiotics (e.g., cephalexin) without interfering with meta-
bolic activity [18, 29]. Total cell counts can be determined
by counterstaining the cells with a fluorescent nucleic
acid stain (commonly DAPI).
CTC is commonly used in microbial ecology, for both
aquatic and terrestrial systems. Applications include drink-
ing water [30], biofilms [31], lake and seawater [32],
sediments [33, 34], and subsurface soils [33]. However,
the universality of CTC to detect respiring cells in natural
samples remains controversial. Sherr et al. [32] found that
Microbes and Infection
2000, 1523-1535
Review
only a single strain did not reduce CTC over a large
number of marine bacterial strains examined (n = 27). A
similar trend was reported from the correlation found
between total and CTC counts in deep sediment samples,
which suggests that CTC also stains anaerobic bacteria
[34]. In contrast, the lack of universality of CTC was
pointed out by Yamaguchi and Nasu [35] for environmen-
tal applications. Toxic effects of CTC on bacterial metabo-
lism have been reported [36], and this toxicity may explain
why CTC counts are sometimes lower than active counts
determined by other metabolic assays (e.g., microautora-
diographic counts, or esterase activity) [36, 37].
4.2.2. Esterase activity
Detection of esterase activity is measured using lipo-
philic, uncharged and non-fluorescent fluorogenic sub-
strates. Once within active cells, the substrate is cleaved
by non-specific esterases releasing a polar fluorescent
product (fluorescein or fluoroscein derivatives) retained
inside cells having an intact membrane. Contrary to
eukaryotic cells, Gram-negative bacteria are generally
impermeable to the lipophilic fluorogenic probes, and a
permeabilisation step of the outer membrane is required.
Esterases are present in all living organisms, and these
enzymes can be used to provide information on the meta-
bolic state of bacterial cells. Although enzyme synthesis
requires energy, the enzyme–substrate reaction does not,
and this assay alone should be considered energy inde-
pendent. However, dead or dying cells with damaged
membranes rapidly leak the dye, even if they retain some
residual esterase activity. Consequently, fluorogenic sub-
strates for esterases often serve as activity and cell integrity
probes that measure both enzymatic activity, which is
required to activate their fluorescence, and cell-membrane
integrity, which is required for intracellular retention of
their fluorescent products.
Fluorescein diacetate (FDA) is known to give weak
fluorescence signals, since fluorescein is poorly retained
inside the cells [17]. In contrast, hydrophobic FDA deriva-
tives are cleaved into hydrophilic products that are retained
more efficiently inside cells with an intact membrane.
Among these, acetoxymethyl ester (calcein-AM) was
shown ineffective to label different species, with the excep-
tion of Staphylococcus aureus [38, 39]. Comparison made
by Jepras et al. [40] of different fluorogenic esters shows
that carboxyfluorescein diacetate (CFDA) is superior to
FDA (fluorophore retention problems) and carboxyfluo-
rescein diacetate acetoxy methyl ester (CFDA-AM) (solu-
bility problems). The ability of a range of fluorogenic esters
to stain several bacterial species was investigated by Dia-
per and Edwards [38]. They found that ChemChrome B
(Chemunex) (a commercial preparation of unknown for-
mulation) is superior to FDA, CFDA, and 2',7'-bis-
(2-carboxyethyl)-5-(and-6)-carboxyfluorescein acetoxy
methyl ester (BCECF-AM), as it stains the widest diversity
of Gram-negative and Gram-positive species.
The main limitations encountered with the fluorogenic
esterases are related to a poor dye uptake, an active dye
extrusion (see section 4.3) [41, 42], and problems of low
labelling efficiency for some bacterial species in pure
cultures such as Pseudomonas spp. [41, 43]. Most of these
1527
Review
limitations were recently overcome by the ChemChrome
V6 staining kit (Chemunex) which combines a counter-
stain to reduce non-specific fluorescence and a specific
buffer to reduce dye extrusion [44].
Porter et al. [20] found that BCECF-AM and Calcein-AM
were inappropriate dyes to enumerate viable bacteria in
freshwater environments. Inversely, these authors observed
a clear signal of fluorescence for active cells stained with
CFDA and ChemChrome B, and the latter was found more
efficient with river water samples. Universality and high
quality of staining was reported for CFDA with both pol-
luted and non-polluted river waters [35]. Reynolds and
Fricker [43] used ChemChrome B to determine the frac-
tion of living bacteria in a large number of potable waters
and they found that this fluorogenic ester was always
superior to plate counts, and often significantly higher
than CTC counts. The application of ChemChrome V6 to
different water samples was recently reported by Catala et
al. [44].
4.3. Pump activity
Several observations have confirmed that different dyes
can be loaded into bacteria and are subsequently actively
removed by energised cells. This is the case for Rh123
[45], BCECF (from BCECF-AM) [41], carboxyfluoroscein
(from CFDA) [42], and ethidium bromide [15]. Active
extrusion of these dyes can lead to biased results. To limit
pump interference, Nebe-von Caron et al. [15] suggested
the addition of sodium azide in staining solutions.
When active dye extrusion is not inhibited, probe efflux
can be used as an additional measure of cell activity [15,
42]. Actively pumping cells are determined using suitable
growth conditions and are sometimes detected using
ethidium bromide at the single-cell level [15]. However,
pump activity assays have only been applied to a few
species in culture and are not universal enough to be
applied to environmental samples.
4.4. Membrane integrity
The loss of membrane integrity represents significant
damage for cells due to multiple functions linked to the
plasma membrane (permeability barrier, transport, respi-
ratory activity, etc.). The maintenance of membrane integ-
rity is commonly measured in eukaryotic cells as an indi-
cator of cell damage or cell death [46]. Assessment of
membrane integrity of bacteria is complicated due to their
complex membrane structure. The envelope of Gram-
negative bacteria consists of three interacting layers: the
outer membrane, the rigid petidoglycan layer and the
inner membrane (plasma membrane). For Gram-positive
cells the outer membrane is absent. Additionally, some
Gram-negative and Gram-positive bacteria have a highly
hydrated polysaccharide layer outside the cell, called the
capsule. Membrane integrity analysis is based on the
capacity of the cells to exclude fluorescent dye com-
pounds, which when used at low concentrations do not
normally cross intact membranes. Most of the membrane
integrity assays use nucleic acid stains, due to the high
concentrations of nucleic acids within the cells and the
large fluorescence enhancement exhibited by nucleic acid
stains upon binding, leading to a clear separation between
1528
Joux and Lebaron
intact and dead cells. Loss of membrane integrity as mea-
sured by uptake of membrane-impermeant dyes is gener-
ally considered irreversible. However restoration of the
membrane integrity has already been reported by Votya-
kova et al. [47] (revealed by the use of the membrane-
impermeant fluorescent DNA stain PO-PRO-3) during
resuscitation of starved Micrococcus luteus.
A wide diversity of impermeant nucleic acid stains can
be used; Molecular Probes (Eugene, Oreg., USA) offers a
large variety of dyes with molecular weights ranging from
402–1 355, among which propidium iodide (PI) is the
most commonly used. In order to simultaneously detect
dead and intact cells, Molecular Probes has developed the
Live/Dead BacLight kit containing two nucleic acid stains
(SYTO-9 and PI) which differ in their spectral characteris-
tics and their ability to penetrate intact bacterial mem-
branes. SYTO-9 penetrates inside cells with both intact or
damaged membranes, staining the cells green, whereas PI
only penetrates cells with damaged membranes, staining
the cells red. When the dyes are used in combination, cells
with intact membrane show a green fluorescence while
cells with damaged membranes show a red fluorescence
(SYTO-9 emission contributes to the excitation of PI by
energy transfer). According to the manufacturer, SYTO-9
should penetrate intact membranes of a large number of
Gram-negative and Gram-positive bacteria. However,
Langsrud and Sundheim [48] found that 30% of Pseudomo-
nas aeruginosa strains tested (n = 18) did not accumulate
SYTO-9. To overcome this problem, other membrane-
permeable stains have been tested in combination with PI
(e.g., SYTO-13 [19]).
The Live/Dead BacLight staining kit was recently used
by different authors to estimate the number of live cells in
marine planktonic environments using epifluorescence
microscopy [49, 50]. The highly variable percentage of
viable cells reported from these studies may result from
cells exhibiting a continuum of in-between colours. In
order to obtain a better discrimination between green and
red fluorescent cells, Gasol et al. [50] recommended the
filter should be rinsed with isopropanol after staining.
Recently, Boulos et al. [30] reported the application of the
Live/Dead BacLight kit to drinking water and similar trends
were found between Live/Dead BacLight and CTC in the
absence of stress. In contrast, stress results in a more rapid
decrease of viable counts as determined by CTC com-
pared to Live/Dead BacLight kit. Defives et al. [51] used
the Live/Dead BacLight kit to determine intact and dead
cell concentrations in natural mineral water during storage
and mechanical bottling. Results suggest that the Live/
Dead BacLight kit is a reliable stain for both autochtho-
nous and allochthonous bacterial strains introduced by
the bottling system.
Recently, Molecular Probes has introduced a new
impermeant nucleic acid dye (Sytox Green). According to
the manufacturer, Sytox Green can be used to detect both
Gram-negative and Gram-positive bacteria with damaged
membranes. However, Lebaron et al. [52] reported limita-
tions in the use of Sytox Green alone for death assessment
in the case of treatment leading to DNA degradation (e.g.,
long-term starvation). In this case, cells with damaged
membranes and damaged DNA display a low fluores-
Microbes and Infection
2000, 1523-1535
Fluorescent physiological probes for bacteria
cence signal due to the degradation of target sites and may
be considered cells having intact membrane and intact
DNA since these cells are, as living cells, not stained with
the dye. This limitation may exist for all membrane-
impermeant DNA stains.
4.5. Nucleic acids
The detection of damaged DNA, such as breaks in the
DNA strands, is often used to characterise apoptosis in
eukaryotic cells [46]. Application of this method to bacte-
ria may be biased by the maintenance of intact DNA for
extended periods after cell death has occurred [28, 53].
For instance, changes in the topology of the DNA mol-
ecule such as those occurring in starved cells may be
misinterpreted as DNA damage [28, 53]. Changes in the
fluorescence of DAPI-stained bacteria have been used to
estimate the efficiency of drinking water disinfection by
sodium hypochlorite [54]. However, death was not a
determining factor in the loss of DAPI fluorescence for all
disinfection treatments, because bacteria killed at high
doses of monochloramine were still fluorescent after DAPI
staining [54]. Recently, evidence has been presented that
bacterial abundance in seawater enumerated by conven-
tional DAPI staining and EFM include cells that do not
contain a genome (‘ghost’ cells) due to unspecific binding
of DAPI [14]. Using a destaining procedure with isopro-
panol to eliminate non-specific binding of DAPI, Zweifel
and Hasgtröm [14] found that a large fraction (68–92%) of
total counts can be scored as ghost cells. However, doubts
persist about the significance of ghost cells [55].
The cellular rRNA content of bacterial cells can be
quantified by fluorescence in situ rRNA hybridization
(FISH) of oligonucleotides carrying a fluorochrome.
Because rRNA content in many bacterial species varies
depending on their growth rate and decrease rapidly in
inactive cells, FISH has been proposed to estimate the
physiological state of cells [56]. Moreover, in the case of
complex communities, this assay could be developed to
detect activity of specific bacteria using appropriate oligo-
nucleotide probes [12]. However, in dynamic environ-
ments and when cells are submitted to stress treatment
(e.g., cold stress, acetic acid, or ethanol), the rRNA content
is a poor indicator of activity due to the high stability of
rRNA [57]. The recent development of FISH techniques
using mRNA or pre-rRNAs (precursors in rRNA synthesis)
as target molecules and those which determine the expres-
sion of specific functional genes may provide more reli-
able methods to assess the activity of individual cells
within complex bacterial communities [58, 59].
5. Heterogeneity of physiological states
at population and community levels
Most investigations at the population level were per-
formed in microcosm, to further understand the cellular
and molecular responses of targeted species to different
stress conditions. An important part of these studies aimed
to identify a definite live versus dead state. Studies at the
community level were performed to characterise the physi-
ological structure of natural communities with the aim of
Microbes and Infection
2000, 1523-1535
Review
discriminating live and dead cells and more recently, to
investigate if at least one assay could be used to detect and
to enumerate growing cells responsible for the production
of bacterial biomass. Both types of studies have given rise
to important semantic confusion around the definition of
activity and viability as already discussed in section 3.
5.1. Heterogeneity at the population level
Probes targeting different cellular functions and the
technique of flow cytometry have been used to analyse the
physiological heterogeneity of bacterial cells within tar-
geted populations in different stress conditions. Many of
these dyes share common excitation/emission properties,
prohibiting simultaneous labelling of samples and there-
fore the collection of multiple parameters on a given cell.
However, the speed of flow cytometry allows the staining
of identical sub-samples by several dyes and then the
comparison of results. Figure 2 illustrates the succession of
cell changes that occurred at different rates in a Salmo-
nella typhimurium population during starvation in artifi-
cial seawater. The proportion of culturable cells decreased
rapidly (state 1) followed by a decrease in the proportion
of respiring cells without or with nutrient addition (states 2
and 3, respectively). After the loss of these functions, cell
integrity is maintained for a few days (state 4) and, after the
loss of membrane integrity, cells with intact DNA (state 5)
are rapidly subjected to an apparent DNA degradation
(state 6) following lysis or formation of non-nucleoid-
containing bacteria (state 7). According to the definition of
physiological states given in section 3, cells in states 2 and
3 may be considered ANC, those in state 4 intact cells, in
states 5 and 6 dead cells and in state 7 ghost cells.
These results and those from many other studies suggest
that there is a strong heterogeneity of physiological states
within bacterial populations subjected to environmental
stress, e.g., [28, 29, 47, 53, 60, 61]. Because each cellular
function is lost at a different rate, the activity of non-
culturable cells is represented by a decreasing number of
active physiological functions during the time of stress and
the fraction of active cells generally decreases up to the
loss of membrane integrity. Today, the significance of each
cellular state in terms of viability remains unclear, mainly
because of our inability to analyse the recovery of specific
cells under favourable growth conditions. In a recent study
combining flow cytometry and cell sorting, it was shown
that S. typhimurium cells with depolarised membranes
could be recovered after cell sorting [15]. It suggests that
depolarisation is a sensitive measure of cell damage but a
poor indicator of cell death. The loss of membrane integ-
rity as determined by the penetration of impermeant
nucleic acid dyes is more often used as an indicator of
death, but inversely, the integrity of the membrane cannot
be considered alone as an indicator of viability but is
commonly used as an indicator of cell integrity [62, 63].
Interpretation of results may also vary depending on the
conditions affecting the physiological state of the cells. For
instance, cell death occurs rapidly after exposure to UV or
gamma radiation due to DNA damage, but the radiation
does not affect dehydrogenase activity, membrane poten-
tial, membrane integrity and β-D-galactosidase activity
[64]. Therefore, this implies a delay following death before
1529
Review
Figure 2. Evolution of the different cell states in a S. typhimurium
population during starvation survival in artificial seawater. Each
cell state (or lysed/non-nucleoid-containing cells) is expressed as a
percentage of the original total cell count. The methods used are
developed in the text. Results published in Joux et al. [28].
State 1: culturable cells as determined on nutrient agar
(bioMérieux, France) plates.
State 2: non-culturable cells showing real respiration by the CTC
(Polysciences Europe, Germany) method.
State 3: non-culturable cells showing no real respiration but
potential respiration as determined by the CTC method after
nutrient (1/10 R2A broth) supply.
State 4: non-culturable cells showing no respiration but mem-
brane integrity maintenance as determined by the Live/Dead
BacLight kit (Molecular Probes).
State 5: non-culturable cells showing loss of membrane integrity
with no DNA change as determined by Hoechst 33342 (Sigma)
staining.
State 6: non-culturable cells showing cellular integrity with
DNA change determined by Hoechst staining.
State 7: lysed or non-nucleoid-containing cells.
enzymes are degraded and membranes are damaged. In
this case, there is no protocol available to rapidly assess
cell viability and the use of culture should be also consid-
ered with caution, since DNA repair mechanisms may
restore viability [65]. Some recent and promising investi-
gations have been made to improve direct methods by
combining traditional probes targeting enzyme activities
or membrane integrity with a probe targeting DNA dam-
age such as fragmentation [66].
The heterogeneity of cellular physiological states
reported in most microcosm-stressed populations is always
observed by using a large inoculum of cells (> 105 cells per
mL) because of the low quantitative sensitivity of both flow
cytometry and microscopy. As a consequence, high cell
densities may introduce a bias in the temporal evolution of
the physiological states of individual cells due to the lysis
of some cells and the use of released organic compounds
by the most metabolically active cells. Consequently, some
physiological states may exist at smaller time scales than
those found in most laboratory studies. The recent devel-
opment of more sensitive instruments such as solid-phase
cytometry may help to provide a more accurate descrip-
tion of the temporal variations in physiological states, and
therefore to measure the persistence time of active but
non-culturable cells within stressed populations. This infor-
mation remains of great importance in the case of patho-
1530
Joux and Lebaron
gens and should be determined for each species and for a
wide variety of stress conditions. If ANC forms are present
and are able to persist in the natural environment, then it
will be important to further evaluate the public health
significance of this state.
5.2. Heterogeneity at the community level
In natural waters, culture methods give microorganism
counts that represent less than 1% of the total bacterial
cells and overlook an unknown but probably large propor-
tion of living cells (that do not grow on the conventional
substrates). On the other hand, total counts determined by
nucleic acid dyes (such as DAPI or acridine orange) include
dead or inactive cells. The detection and enumeration of
active cells (living bacteria with any detectable metabolic
activity), growing cells (bacteria which participate in the
production of biomass) and dead cells (representing
organic particles) in natural bacterial assemblages would
be of great help and most useful in aquatic microbial
ecology. It would help to better understand the mecha-
nisms behind the regulation of bacterial activity. Different
probes and methods have been used to estimate the pro-
portion of both active and dead cells and only a few recent
investigations have tried to detect and to enumerate grow-
ing cells. However, consensus on what each of the probes
tested in these studies measures has not yet been reached.
The detection of active cells within complex commu-
nities requires the use of universal probes which target the
widest diversity of species. Therefore, only a few probes
(formazan salt reduction, esterase activity, and rRNA
probes) have been applied to natural communities to
describe the physiological heterogeneity of bacterial cells.
Despite the different drawbacks of CTC reported in section
4.2.1, studies suggest that CTC provides an ecologically
meaningful measure of bacterial activity [26, 67]. Sherr et
al. [32] suggested that CTC-positive cells represent those
bacteria characterised by a high level of metabolic activ-
ity, and that cells which show no apparent reduction of
CTC have either low or no metabolic activity. This assump-
tion is far from clear, and the interpretation of CTC counts
remains controversial. The recent and increasing use of
CTC counts in microbial ecology and at the same time the
well-known limitations of the method illustrate the impor-
tant need, for microbial ecologists, to identify which cells
are responsible for bacterial community activity. As an
alternative to CTC, growing cells can also be determined
by flow cytometry as the fraction of cells with a high DNA
content [68]. The fraction of cells with a high DNA content
is easily discriminated by flow cytometric analysis of SYBR
Green-stained cells, and in a recent study, Gasol et al. [50]
indicated that high-DNA bacteria are the dynamic mem-
bers of the bacterial assemblage. General acceptance of
this approach to determine the dynamic members of natu-
ral communities will require further investigation to under-
stand how it works and what its results mean.
Studies which perform a comparison of methods are
scarce and more intercalibration between methods is cur-
rently needed. Karner and Fuhrman [37] have applied
different methods to assess bacterial activity in seven
different marine water samples and they found an average
proportion of 56% of cells containing detectable rRNA,
Microbes and Infection
2000, 1523-1535
Fluorescent physiological probes for bacteria
49% of cells able to take up added radiolabelled com-
pounds, 29% of nucleoid-containing cells and only 0.7%
of respiring cells. Parthuisot et al. [69] found that active
bacteria determined by esterase activity measurements in
different water samples are higher than those determined
by CTC and CFU and that CTC counts are sometimes
lower than CFU. From these results and other studies [55,
70], there is clear evidence that there is an important
heterogeneity of physiological states within natural com-
munities. However, each method yields different estimates
of the proportion of bacterial cells that are active. The
physiological and ecological significance of each cell
state remains unclear and the answer, if it exists, to this
question will require further investigations to relate a given
cellular physiological state to additional information such
as its growth potential, enzyme activity or taxonomic
identity.
6. Combination of fluorescent probes
The combination of fluorescent probes in single assays
provides confident tools and some of these are actually
under validation for industrial applications such as water
quality assessment in the pharmaceutical industry [43].
However, although physiological probes are very useful
for different industrial and environmental applications,
they cannot address the detection of specific active cells
such as pathogenic microorganisms. Most of these appli-
cations are based on the combination of physiological and
taxonomic probes and require instruments which can
detect rare events.
6.1. Combination of different
physiological fluorescent probes
A more accurate evaluation of cell activity should be
obtained by combining physiological probes targeting
different cellular functions. Assays in which both
membrane-based and metabolism-based probes are used
simultaneously provide information on whether the dye
assays accurately reflect cell activity. Probes should be
selected with excitation and emission wavelengths which
allow discrimination of each probe in the presence of the
other(s). Therefore, dyes with a narrow emission spectrum
may be better suited to multicolour applications. The
selection will also be dependent on the instrumentation
used for fluorescence analysis (see section 2). One should
also keep in mind that the selection of fluorophores in
multiple-staining procedures may take into account
molecular interactions which can result in quenching, a
decrease in the fluorescence signal. This decrease is due to
changes in the excited states of the fluorophores, resulting
in energy dissipation by non-radiative transitions. These
physico-chemical changes may be caused by the interac-
tion between fluorophores.
In multiple-staining assays, PI (dead cells, red fluores-
cence) is often combined with CFDA (cells with esterase
activity, green fluorescence) [35, 42], Rh123 (polarised
cells, green fluorescence) [18, 71] or BOX (depolarised
cells, green fluorescence) [18, 71]. The CTC dye can also
be combined with the detection of cell depolarisation
Microbes and Infection
2000, 1523-1535
Review
(BOX) or cell polarisation (Rh123) [18]. The CV6/CSE
labelling kit from Chemunex offers the possibility of simul-
taneously detecting cells with an esterase activity (CV6)
and membrane integrity (CSE). Application of this staining
procedure to S. typhimurium cells starved in artificial sea-
water allowed a more accurate assessment of active cell
counts by counterstaining cells with permeabilised mem-
branes [44]. A nice illustration of multiple-staining assays
was provided by Nebe-von Caron et al. [15] using a
combination of three fluorescent dyes to measure mem-
brane potential (BOX), pump activity (ethidium bromide)
and membrane integrity (PI).
Most of these studies remained methodological studies
and suffer from a lack of validation. Therefore, the signifi-
cance of these methods and their possible use for viability
assessment is unknown. ChemChrome V6 from Chemu-
nex is one of the most commonly used methods which
combines a fluorogenic ester for esterase activity assess-
ment with a counterstain which penetrates inside cells
with permeabilised membranes [44, 69]. This method is
widespread in different industrial applications (pharma-
ceutical, food and cosmetic industries) and when coupled
with flow cytometry provide real-time estimation of active
cell counts. Although active counts are sometimes signifi-
cantly higher than viable counts determined by CFU, they
are considered very useful to detect production problems
and perhaps yield a more accurate estimation of viable
counts than do culture methods.
6.2. Combination of physiological
and taxonomic fluorescent probes
In the case of complex communities, physiological
fluorescent probes provide useful information on the per-
centage of active, inactive or dead cells of the overall
community. However, detection of the activity of specific
bacteria is of primary importance in different microbio-
logical areas. For instance, it could be used to detect active
pathogens in food and waters, or to detect active specific
populations implicated in different transformations (biore-
actors, environment). Fluorescent antibodies and fluores-
cent oligonucleotide rRNA are taxonomic probes that can
be used at the single-cell level to detect bacteria of interest
including pathogens. One problem encountered with
hybridisation of fluorescent oligonucleotide rRNA con-
cerns the permeabilisation of cells and amplification of
fluorescence signals. The combination of taxonomic and
physiological fluorescent probes is a key challenge for
microbiologists since there is an increasing need to pro-
vide numbers of specific cells, depending on their physi-
ological state (at least viable or active state).
Whiteley et al. [72] described the combination of fluo-
rescent in situ hybridisation (rRNA probes) with the cyto-
chrome oxidase assay. Similar combinations with a tetra-
zolium salt necessitate a fastidious sequential procedure,
inconceivable in routine, in which tetrazolium reduction
is evaluated prior to probe hybridisation [72]. Different
authors have demonstrated the feasibility of combining
immunostaining with CTC [73], Rh123 [74] and Chem-
Chrome staining [75]. The fluorochrome conjugated to the
monoclonal antibody must be carefully chosen in order to
avoid fluorescence overlap with the physiological dye.
1531
Review
Recently, Pyle et al. [76] combined an immunomagnetic
separation (IMS) with the analysis of respiratory activity
(CTC staining) to assess the number of active E. coli
O157:H7 present in meat and water. In this procedure,
E. coli O157:H7 cells are isolated from the sample by
using a magnetic particle concentrator after addition of
immunomagnetic beads coated with anti-O157 antibody.
Then, cells are incubated with CTC (3 h), stained with
FITC-labelled antibody (20 min) and filtered for micro-
scopic enumeration or solid-phase laser cytometry. This
procedure results in high rates of recovery but cell injury is
observed [76]. However, the combination of taxonomic
probes with the CTC dye has little interest since CTC
counts are well known to underestimate the number of
viable targeted cells (see sections 4.2.1 and 5.2). Further-
more, most of these assays were performed by using an
artificial mixture of strains or artificial contamination.
Such a combination should be treated cautiously for envi-
ronmental monitoring, due to additional manipulative
steps that could reduce cell recovery.
One additional limitation to the actual use of these
methods is that in most cases, cells of interest are highly
diluted within autochthonous populations, and field appli-
cations have long been limited because instruments such
as microscopy and flow cytometry do not allow the detec-
tion of rare events [77]. Detection of bacterial cells at a
frequency of 10–5 to 10–7 is a challenge in various fields of
microbiology (industry, medicine, environment, etc.). The
recent development of solid-phase cytometry has covered
this gap and this technology offers promising perspectives.
The ChemScan technology is now increasingly used in the
industry for the rapid and accurate detection of E. coli,
Cryptosporidium spp. and Giardia spp. in waters with a
high sensitivity of detection as well as for the quantifica-
tion of active bacteria in process waters (pharmaceutical
industries). For instance, it is possible to detect a single cell
of E. coli or any other species for which specific antibodies
are available in 1 L of tap water and in the presence of
106–108 non-targeted cells (Lebaron, unpublished data).
This detection limit is similar to that of culture techniques
which combine a filtration step. Furthermore, SPC offers
the advantage that it does not require the use of selective
growth conditions and thus, allows the detection of injured
cells which are viable cells not detected on selective
media. However, although the instrument can detect three
fluorescences, applications are still limited because it is
equipped with a single source of excitation. Therefore,
today the use of at least two probes for physiological
assessment and an additional probe (taxonomic) for the
detection of specific cells remains difficult if not impos-
sible. Many applications will require the use of a second
source of light.
7. Conclusions and future prospects
There is clear evidence that the application of fluores-
cent dyes will provide microbiologists with a set of valu-
able tools to complement existing molecular and conven-
tional methods by providing information concerning
specific physiological activities at the level of the indi-
1532
Joux and Lebaron
vidual bacterial cell. However, despite the continuous
development of new fluorescent dyes, no universal stain
or staining procedure that is suitable for all applications
has been found, and will probably never be found. The
efficiency of dyes and protocols needs to be evaluated for
each application and studies at the population level remain
essential to understand how different species react in a
given assay and how the response varies depending on the
stress conditions applied to these species.
The development of multiparameter analysis which
combines at least two physiological probes is very prom-
ising but requires further investigations and validation
work to ensure that these methods are reproducible and
robust enough for routine applications. Validation should
be made in regard to specific applications since no method,
as is the case for culturability, can be robust enough to be
universal and suitable for all applications. For instance,
activity assessment after UV or gamma radiation requires
methods which are not based on membrane integrity or
enzyme activities. Inversely, methods based on a combi-
nation of probes which target membrane permeability and
enzyme activity can be very useful and are already rou-
tinely used to quantify active cells in water, food products,
drugs and cosmetics, to control the efficiency of freeze-
drying procedures in bacterial collections and probably
other applications. Differences between active and cultur-
able cell counts do not constitute a limitation to the use of
physiological probes if active cell counts are obtained
rapidly and allow the detection of any interpretable
changes in terms of quality control. In these cases, limits
that are based on the standard cultivation techniques,
when they exist, should be reconsidered in favour of
alternative methods.
The traditional debate and conflict around viability and
activity as assessed by physiological probes is somehow
paradoxical. On the one hand, traditional culture methods
are based on the dividing capacity of a cell which requires
a few hours and sometimes a few days to be detected and
on the other hand, physiological probes which have been
developed for real-time monitoring of cell viability do not
allow the control of cell division. Therefore, viability
assessment cannot be achieved directly by physiological
probes. When assays based on the use of physiological
probe(s) are available, they should be evaluated and com-
pared to standard methods. However, these probes have
never been applied to the detection of rare events such as
pathogens because the instrumentation used for fluores-
cence analysis was until recently inappropriate for the
detection of rare events. This last point as well as the price
of these instruments probably explains why alternative
methods have received little field application and valida-
tion and why it is difficult if not impossible today to
provide an accurate evaluation of the potential use of
these probes as alternative methods to the standard culti-
vation techniques.
As stated above, the adoption of these new tools will
also require new standards in bacterial quality assessment
since, in most applications, counts are generally higher
than CFU counts which were used to define the actual
standards. The development of new fluorescent dyes more
closely related to cellular energy metabolism and assays
Microbes and Infection
2000, 1523-1535
Fluorescent physiological probes for bacteria
combining different probes remain important to improve
and/or to further develop the use of alternative methods in
a wide variety of bacterial applications. Furthermore, the
development of new assays, which combine both physi-
ological and taxonomic probes, should be encouraged to
provide new tools for the detection of active specific cells.
For these combined staining procedures, development of
low-cost instruments equipped with at least two excitation
sources is essential.
References
[1] Barer M.R., Harwood C.R., Bacterial viability and cultur-
ability, Adv. Microb. Physiol. 41 (1999) 93–137.
[2] Nebe-von Caron G., Badley R.A., Viability assessment of
bacteria in mixed populations using flow cytometry,
J. Microscopy 179 (1995) 55–66.
[3] Sidorowicz S.V., Withmore T.N., Prospects for new tech-
niques for rapid bacteriological monitoring of drinking
water, J. IWEN 9 (1995) 92–97.
[4] Roszak D.B., Grimes D.J., Colwell R.R., Viable but non-
recoverable stage of Salmonella enteritidis in aquatic system,
Can. J. Microbiol. 30 (1984) 334–338.
[5] Ward D., Weller R., Bateson M.M., 16S rRNA sequences
reveal numerous uncultured microorganisms in a natural
community, Nature 345 (1990) 63–65.
[6] Porter J., Deere D., Pickup R., Edwards C., Fluorescent
probes and flow cytometry: new insights into environmen-
tal bacteriology, Cytometry 23 (1996) 91–96.
[7] Collier J.L., Campbell L., Flow cytometry in molecular
aquatic ecology, Hydrobiologia 401 (1999) 33–53.
[8] Darzynkiewicz Z., Bedner E., Li X., Gorczyca W.,
Melamed M.R., Laser-scanning cytometry: a new instru-
mentation with many applications, Exp. Cell Res. 249
(1999) 1–12.
[9] Reynolds D.T., Slade R.B., Sykes N.J., Jonas A.,
Fricker C.R., Detection of Crysporidium oocysts in water:
techniques for generating precise recovery data, J. Appl.
Microbiol. 87 (1999) 804–813.
[10] Mignon-Godefroy K., Guillet J.C., Butor C., Solid-phase
cytometry for detection of rare events, Cytometry 27 (1997)
336–344.
[11] Butor C., Duquenne O., Mignon-Godefroy K., Mougin C.,
Guillet J.G., Solid-phase cytometry allows rapid in situ
quantification of human papilloma virus infection in biopsy
material, Cytometry 29 (1997) 292–297.
[12] Aßmus B., Schloter M., Kirchhof G., Hutzler P., Hart-
mann A., Improved in situ tracking of rhizosphere bacteria
using dual staining with fluorescence-labeled antibodies
and rRNA-targeted oligonucleotides, Microb. Ecol. 33
(1997) 32–40.
[13] Porter J., Deere D., Hardman M., Edwards C., Pickup R.,
Go with the flow –use of flow cytometry in environmental
microbiology, FEMS Microbiol. Ecol. 24 (1997) 93–101.
[14] Zweifel U.L., Hagström Å., Total counts of marine bacteria
include a large fraction of non-nucleoid-containing bacteria
(ghosts), Appl. Environ. Microbiol. 61 (1995) 2180–2185.
Microbes and Infection
2000, 1523-1535
Review
[15] Nebe-von Caron G., Stephens P., Badley R.A., Assessment
of bacterial viability status by flow cytometry and single
cell sorting, J. Appl. Microbiol. 84 (1998) 988–998.
[16] Kaprelyants A.S., Kell D.B., Dormancy in stationary-phase
cultures of Micrococcus luteus: flow cytometric analysis of
starvation and resuscitation, Appl. Environ. Microbiol. 59
(1993) 3187–3196.
[17] Diaper J.P., Tither K., Edwards C., Rapid assessment of
bacterial viability by flow cytometry, Appl. Microbiol.
Biotechnol. 38 (1992) 268–272.
[18] López-Amorós R., Castel S., Comas-Riu J., Vives-Rego J.,
Assessment of E. coli and Salmonella viability and starvation
by confocal laser microscopy and flow cytometry using
rhodamine 123, DiBAC4(3), propidium iodide, and CTC,
Cytometry 29 (1997) 298–305.
[19] Comas J., Vives-Rego J., Assessment of the effects of grami-
cidin, formaldehyde, and surfactants on Escherichia coli by
flow cytometry using nucleic acid and membrane potential
dyes, Cytometry 29 (1997) 58–64.
[20] Porter J., Diaper J., Edwards C., Pickup R., Direct mea-
surements of natural planktonic bacterial community
viability by flow cytometry, Appl. Environ. Microbiol. 61
(1995) 2783–2786.
[21] Smith J.J., McFeters G.A., Mechanisms of INT (2-(4-
iodophenyl)-3-(4-nitrophenyl)-5-phenyltetrazolium chlo-
ride), and CTC (5-cyano-2,3-ditolyl tetrazolium chloride)
reduction in Escherichia coli K-12, J. Microbiol. Methods 29
(1997) 161–175.
[22] Bhupathiraju V.K., Hernandez M., Landfear D., Alvarez-
Cohen L., Application of a tetrazolium dye as an indicator
of viability in anaerobic bacteria, J. Microbiol. Methods 37
(1999) 231–243.
[23] Pyle B.H., Broadaway S.C., McFeters G.A., Factors affect-
ing the determination of respiratory activity on the basis of
cyanoditolyl tetrazolium chloride reduction with mem-
brane filtration, Appl. Environ. Microbiol. 61 (1995)
4304–4309.
[24] Smith J.J., McFeters G.A., Effects of substrates and phos-
phate on INT (2-(4-iodophenyl)-3-(4-nitrophenyl)-5-
phenyl tetrazolium chloride) and CTC (5-cyano-2,3-ditolyl
tetrazolium chloride) reduction in Escherichia coli, J. Appl.
Bacteriol. 80 (1996) 209–215.
[25] Sieracki M.E., Cucci T.L., Nicinski J., Flow cytometric
analysis of 5-cyano-2,3-ditolyl tetrazolium chloride activ-
ity of marine bacterioplankton in dilution cultures, Appl.
Environ. Microbiol. 65 (1999) 2409–2417.
[26] Lovejoy C., Legendre L., Klein B., Tremblay J.É.,
Ingram R.G., Therriault J.C., Bacterial activity during
early winter mixing (Gulf of St. Lawrence, Canada), Aquat.
Microb. Ecol. 10 (1996) 1–13.
[27] del Giorgio P.A., Prairie Y.T., Bird D.F., Coupling between
rates of bacterial production and the abundance of meta-
bolically active bacteria in lakes, enumerated using CTC
reduction and flow cytometry, Microb. Ecol. 34 (1997)
144–154.
[28] Joux F., Lebaron P., Troussellier M., Succession of cellular
states in a Salmonella typhimurium population during starva-
tion in artificial seawater microcosms, FEMS Microbiol.
Ecol. 22 (1997) 65–76.
1533
Review
[29] Caro A., Got P., Baleux B., Physiological changes of Salmo-
nella typhimurium cells under osmotic and starvation condi-
tions by image analysis, FEMS Microbiol. Lett. 179 (1999)
265–273.
[30] Boulos L., Prévost M., Barbeau B., Coallier J., Desjar-
dins R., Live/Dead Baclight: application of a new rapid
staining method for direct enumeration of viable and total
bacteria in drinking water, J. Microbiol. Methods 37 (1999)
77–86.
[31] Huang C.T., Yu F.P., McFeters G.A., Stewart P.S., Nonuni-
form spatial patterns of respiratory activity within biofilms
during disinfection, Appl. Environ. Microbiol. 61 (1995)
2252–2256.
[32] Sherr B.F., del Giorgio P.A., Sherr E.B., Estimating abun-
dance and single-cell characteristics of respiring bacteria
via the redox dye CTC, Aquat. Microb. Ecol. 18 (1999)
117–131.
[33] Yu W., Doods W.K., Banks M.K., Skalsky J., Strauss E.A.,
Optimal staining and sample storage time for direct micro-
scopic enumeration of total and active bacteria in soil with
two fluorescent dyes, Appl. Environ. Microbiol. 61 (1995)
3367–3372.
[34] Miskin I., Rhodes G., Lawlor K., Saunders J.R.,
Pickup R.W., Bacteria in post-glacial freshwater sediments,
Microbiology 144 (1998) 2427–2439.
[35] Yamaguchi N., Nasu M., Flow cytometric analysis of bac-
terial respiratory and enzymatic activity in the natural
aquatic environment, J. Appl. Microbiol. 83 (1997) 43–52.
[36] Ullrich S., Karrasch B., Hoppe H.G., Jeskulke K., Meh-
rens M., Toxic effects on bacterial metabolism of the redox
dye 5-cyano-2,3-ditolyl tetrazolium chloride, Appl. Envi-
ron. Microbiol. 62 (1996) 4587–4593.
[37] Karner M., Furhman J.A., Determination of active marine
bacterioplankton: a comparison of universal 16S rRNA
probes, autoradiography, and nucleic staining, Appl. Envi-
ron. Microbiol. 63 (1997) 1208–1213.
[38] Diaper J.P., Edwards C., The use of fluorogenic esters to
detect viable bacteria by flow cytometry, J. Appl. Bacteriol.
77 (1994) 221–228.
[39] Comas J., Vives-Rego J., Enumeration, viability and het-
erogeneity in Staphylococcus aureus cultures by flow cytom-
etry, J. Microbiol. Methods 32 (1998) 45–53.
[40] Jepras R.I., Carter J., Pearson S.C., Paul F.E., Wilkin-
son M.J., Development of a robust flow cytometric assay for
determining numbers of viable bacteria, Appl. Environ.
Microbiol. 61 (1995) 2696–2701.
[41] Molenaar D., Bolhuis H., Abee T., Poolman B., Kon-
ings W.N., The efflux of a fluorescent probe is catalyzed by
an ATP-driven extrusion system in Lactococcus lactis, J. Bac-
teriol. 174 (1992) 3118–3124.
[42] Bunthof C., VanDen Braak S., Breeuwer P., Rom-
bouts F.M., Abee T., Rapid fluorescence assessment of the
viability of stressed Lactococcus lactis, Appl. Environ. Micro-
biol. 65 (1999) 3681–3689.
[43] Reynolds D.T., Fricker C.R., Application of laser scanning
for the rapid and automated detection of bacteria in waters
samples, J. Appl. Microbiol. 86 (1999) 785–795.
1534
Joux and Lebaron
[44] Catala P., Parthuisot N., Bernard L., Baudart J., Lemarch-
and K., Lebaron P., Effectiveness of CSE to counterstain
particles and dead bacterial cells with permeabilised mem-
branes: application to viability assessment in waters, FEMS
Microbiol. Lett. 178 (1999) 219–226.
[45] Ueckert J., Breeuwer P., Abee T., Stephens P., Nebe-von
Caron G., TerSteeg P.F., Flow cytometry applications in
physiological study and detection of foodborne microor-
ganisms, Int. J. Food Microbiol. 28 (1995) 317–326.
[46] Shapiro H.M., Pratical flow cytometry, 2nd edition, Alan
R. Liss, New York, 1988.
[47] Votyakova V., Kaprelyants A.S., Kell D.B., Influence of
viable cells on the resuscitation of dormant cells in Micro-
coccus luteus cultures held in an extended stationary phase:
the population effect, Appl. Environ. Microbiol. 60 (1994)
3284–3291.
[48] Langsrud S., Sundheim G., Flow cytometry for rapid assess-
ment of viability after exposure to a quaternary ammonium
compound, J. Appl. Bacteriol. 81 (1996) 411–418.
[49] Decamp O., Rajendran N., Assessment of bacterioplankton
viability by membrane integrity, Mar. Poll. Bull. 36 (1998)
739–741.
[50] Gasol J.M., Zweifel U.L., Peters F., Furhman J.A., Hag-
ström Å., Significance of size and nucleic acid content
heterogeneity as measured by flow cytometry in natural
planktonic bacteria, Appl. Environ. Microbiol. 65 (1999)
4475–4483.
[51] Defives C., Guyard S., Oularé M.M., Mary P., Hornez J.P.,
Total counts, culturable and viable, and non-culturable
microflora of a French mineral water: a case study, J. Appl.
Microbiol. 86 (1999) 1033–1038.
[52] Lebaron P., Catala P., Parthuisot N., Effectiveness of SYTOX
Green stain for bacterial viability assessment, Appl. Envi-
ron. Microbiol. 64 (1998) 2697–2700.
[53] Joux F., Lebaron P., Troussellier M., Changes in cellular
states of the marine bacterium Deleya aquamarina under
starvation conditions, Appl. Environ. Microbiol. 63 (1997)
2686–2694.
[54] Saby S., Sibille I., Mathieu L., Paquin J.L., Block J.C.,
Influence of water chlorination on the counting of bacteria
with DAPI (4',6-diamidino-2-phenylindole), Appl. Envi-
ron. Microbiol. 63 (1997) 1564–1569.
[55] Choi J.W., Sherr E.B., Sherr B.F., Relation between
presence-absence of a visible nucleoid and metabolic activ-
ity in bacterioplankton cells, Limnol. Oceanogr. 41 (1996)
1161–1168.
[56] Ruimy R., Breittmayer V., Boivin V., Christen R., Assess-
ment of the state of activity of individual bacterial cells by
hybridization with a ribosomal RNA targeted fluorescently
labelled oligonucleotidic probe, FEMS Microbiol. Ecol. 15
(1994) 207–214.
[57] Tolker-Nielsen T., Halberg Larsen M., Kyed H., Molin S.,
Effect of stress treatments on the detection of Salmonella by
in situ hybridization, Int. J. Food Microbiol. 35 (1997)
251–258.
[58] Chen F., Gonzáles J.M., Dustman W.A., Moran M.A.,
Hodson R.E., In situ reverse transcription, an approach to
characterize genetic diversity and activities of prokaryotes,
Appl. Environ. Microbiol. 63 (1997) 4907–4913.
Microbes and Infection
2000, 1523-1535
Fluorescent physiological probes for bacteria
[59] Licht T.R., Tolker-Nielsen T., Holmstrøm K., Krog-
felt K.A., Molin S., Inhibition of Escherichia coli precursor-
16S rRNA processing by mouse intestinal contents, Envi-
ron. Microbiol. 1 (1999) 23–32.
[60] Caro A., Got P., Lesne J., Binard S., Baleux B., Viability and
virulence of experimentally stressed nonculturable Salmo-
nella typhimurium, Appl. Environ. Microbiol. 65 (1999)
3229–3232.
[61] Lisle J.T., Pyle B.H., McFeters G.A., The use of multiple
indices of physiological activity to access viability in chlo-
rine disinfected Escherichia coli O157:H7, Lett. Appl. Micro-
biol. 29 (1999) 42–47.
[62] Novo D.J., Perlmutter N.G., Hunt R.H., Shapiro H.M.,
Multiparameter flow cytometric analysis of antibiotic effects
on membrane potential, membrane permeability, and bac-
terial counts of Staphylococcus aureus and Micrococcus luteus,
Antimicrob. Agents Chemother. 44 (2000) 827–834.
[63] Weichart D., McDougald D., Jacobs D., Kjelleberg S., In
situ analysis of nucleic acids in cold-induced nonculturable
Vibrio vulnificus, Appl. Environ. Microbiol. 63 (1997)
2754–2758.
[64] Fiksdal L., Tryland I., Effect of UV light irradiation, star-
vation and heat on Escherichia coli β-D-galactosidase activ-
ity and other potential viability parameters, J. Appl. Micro-
biol. 87 (1999) 62–71.
[65] Joux F., Jeffrey W.H., Lebaron P., Mitchell D.L., Marine
bacteria isolates display diverse responses to ultraviolet-B
radiation, Appl. Environ. Microbiol. 65 (1999) 3820–3827.
[66] Rohwer F., Azam F., Detection of DNA damage in prokary-
otes by terminal deoxyribonucleotide transferase-mediated
dUTP nick end labeling, Appl. Environ. Microbiol. 66
(2000) 1001–1006.
[67] Smith E.M., Coherence of microbial respiration rate and
cell-specific bacterial activity in a coastal planktonic com-
munity, Aquat. Microb. Ecol. 16 (1998) 27–35.
[68] Jellet J.F., Li W.K.W., Dickie P.M., Boraie A., Kepkay P.E.,
Metabolic activity of bacterioplankton communities
assessed by flow cytometry and single carbon substrate
utilization, Mar. Ecol. Prog. Ser. 136 (1996) 213–225.
Microbes and Infection
2000, 1523-1535
Review
[69] Parthuisot N., Catala P., Lemarchand K., Baudart J., Leb-
aron P., Evaluation of ChemChrome V6 for bacterial viabil-
ity assessment in waters, J. Appl. Microbiol. (2000)
370–380.
[70] Joux F., Lebaron P., Ecological implication of an improved
direct viable count method for aquatic bacteria, Appl.
Environ. Microbiol. 63 (1997) 3643–3647.
[71] López-Amorós R., Comas J., Vives-Rego J., Flow cytomet-
ric assessment of Escherichia coli and Salmonella typhimurium
starvation–survival in seawater using rhodamine 123, pro-
pidium iodide, and oxonol, Appl. Environ. Microbiol. 61
(1995) 2521–2526.
[72] Whiteley A.S., O’Donnell A.G., Macnaughton S.J.,
Barer M.B., Cytochemical colocalization and quantitation
of phenotypic and genotypic characteristics in individual
bacterial cells, Appl. Environ. Microbiol. 62 (1996)
1873–1879.
[73] Pyle B.H., Broadaway S.C., McFeters G.A., A rapid, direct
method for enumerating respiring enterohemorragic
Escherichia coli O157:H7 in water, Appl. Environ. Micro-
biol. 61 (1995) 2614–2619.
[74] Clarke R.G., Pinder A.C., Improved detection of bacteria
by flow cytometry using a combination of antibody and
viability markers, J. Appl. Microbiol. 84 (1998) 577–584.
[75] Hennington E.W., Krocova Z., Sandström G., Forsman M.,
Flow cytometric assessment of the survival ratio of Fran-
cisella tularensis in areobiological samples, FEMS Microbiol.
Ecol. 25 (1998) 241–249.
[76] Pyle B.H., Broadaway S.C., McFeters G.A., Sensitive detec-
tion of Escherichia coli O157:H7 in food and water by
immunomagnetic separation and solid-phase laser cytom-
etry, Appl. Environ. Microbiol. 65 (1999) 1966–1972.
[77] Pinder A.C., Purdy P.W., Poulter S.A.G., Clark D.C.,
Validation of flow cytometry for rapid enumeration of
bacterial concentrations in pure cultures, J. Appl. Bacte-
riol. 69 (1990) 92–100.
1535