combined fluorescent antibody assay and viability staining for the assessment of the physiological...

Combined fluorescent antibody assay and viability stainingfor the assessment of the physiological statesof Escherichia coli in seawaters

G. Caruso, M. Mancuso and E. CrisafiCNR Istituto per l’Ambiente Marino Costiero – Section of Messina, Messina, Italy

2003/0013: received 8 January 2003, revised 12 February 2003 and accepted 11 March 2003

ABSTRACT

G. CARUSO, M. MANCUSO AND E. CRISAFI . 2003.

Aims: A comparison of methods that combine the use of immune sera with specific fluorescent probes for testing

viability at single cell level was performed in order to estimate different living attributes of Escherichia coli in natural

seawater samples.

Methods and Results: Cell culturability was assayed by plate method, respiratory activity and membrane integrity

were determined by an indirect fluorescent antibody assay, combined with 5-cyano-2, 3 ditolyl tetrazolium chloride

and propidium iodide, respectively. Results showed the coexistence of different physiological states within the E. coli

population, of which a large fraction (46%) of cells was actively respiring.

Conclusions: The methodological approach used offer interesting perspectives in water pollution monitoring,

particularly when the differentiation between dead and living E. coli cells is required for a more precise assessment

of the bacteriological quality of seawaters.

Significance and Impact of the Study: The study suggests the importance of knowledge of the viability status of

faecal bacteria in aquatic environments as a fundamental issue for the preservation of public health; the availability

of rapid analytical procedures for this purpose may find significant applications in the evaluation of the sanitary risk

consequent to water use.

Keywords: epifluorescence microscopy, Escherichia coli, fluorescent probes, monitoring, seawater.

INTRODUCTION

In water quality assessment, the detection of the micro-

organism Escherichia coli, usually provides evidence of faecal

contamination and consequently indicates the potential

presence of pathogenic microbes in the aquatic environment.

A consistent number of investigations have focused on the

existence of different metabolic levels and vital states for this

microorganism in experimental conditions (Porter et al.

1995; Pyle et al. 1995; Muela et al. 1999; Ericsson et al.2000; Petit et al. 2000); in natural environments, however,

the physiological state of this bacterium has scarcely been

characterized and the relative proportion in situ of viable

and/or active, inactive and dead cells is not yet well known.

Conversely, knowledge of the viability properties of the cells

is of great significance in order to ascertain the real health

risk for humans deriving from the use of polluted seawaters

(Lopez-Amoros et al. 1995; Pommepuy et al. 1996). Con-

ventional culture methods lack the sensitivity to detect

environmental pathogens as they fail to evaluate injured or

stressed cells that are unable to reproduce on growth media.

Also the direct microscopic methods for the detection and

enumeration of E. coli by immunofluorescence, recently

used in seawater analysis as an alternative to standard plate

counts (Caruso et al. 2000, 2002), allow to determine the

total abundance of this bacterium only, regardless of its

physiological state. Direct counts do not take into accountCorrespondence to: Dr Gabriella Caruso, CNR Istituto per l’Ambiente marino

costiero – Section of Messina, Messina, Italy1 (e-mail: [email protected]).

ª 2003 The Society for Applied Microbiology

Journal of Applied Microbiology 2003, 95, 225–233

whether enumerated cells are able to metabolize, grow,

respire and divide (Roszack and Colwell 1987).

Many studies have been addressed to the survival of

allochthnous bacteria in marine environments as well as to

the abiotic (solar radiation, nutrient availability, temperature

and osmotic stress) and biotic (predation) factors that affect

it (see Barcina et al. 1997; Rozen and Belkin 2001 for a

review). It has been shown that after release into natural

aquatic environments through wastewater discharge, they

may enter a ‘viable but non-culturable’ (VBNC, according to

Roszack and Colwell 1987) state; this condition has been

proposed as a survival strategy adopted by bacteria in

response to environmental stresses (Hood and Ness 1982;

Xu et al. 1982; Oliver 1993). In this state, cells loose

culturability becoming unable to form colonies on conven-

tional growth media, while they maintain active metabolism

as well as cellular integrity and also potential pathogenicity

(Barer et al. 1993; Pommepuy et al. 1996; Kell et al. 1998).

The non-recovery of indicator bacteria or pathogens on

culture media, i.e. in the case of false negative results, cannot

prove itself the good quality of seawaters, as VBNC forms

are a potential reservoir for pathogenic bacteria, that may

cause outbreaks of infectious diseases when environmental

conditions become favourable. Therefore it is of fundamen-

tal importance from a sanitary point of view to assess the

actual viability of faecal bacteria in aquatic environments,

i.e. whether enteropathogenic E. coli cells are really injured

or if they still have a pathogenic potential (Pommepuy et al.

1996; Petit et al. 2000). This concept is a critical issue of

primary concern in assessing seawater quality and needs

further investigation.

Recent advances in the fluorescent dye technology have

resulted in the availability of a number of new fluorochromes

to be used as probes for the assessment of a variety of cell

attributes and physiological functions; these dyes may be

applied in bacterial ecology studies to monitor cell viability

and to obtain estimates of relative activity within the

microbial community (McFeters et al. 1995; Porter et al.

1996).

In the framework of the multidisciplinary project Cluster-

10 SAM funded by Italian Ministry for University and

Scientific Research, specific studies have been addressed to

improve and update the methods currently available for the

detection of faecal pollution in aquatic environments. The

main objective of our research has been the application of

new microscopic procedures for the quantitative determin-

ation of the living and dead fraction of E. coli in seawaters.

Two different dual staining protocols have been compared,

that combine the indirect fluorescent antibody (FA) assay

previously used (Caruso et al. 2000, 2002) with the fluor-

escent dye propidium iodide (PI), a membrane-compro-

mised cell indicator, or FA assay with a metabolic activity

marker, 5-cyano-2, 3-ditolyl-tetrazolium chloride (CTC). PI

is defined as an exclusion stain, as it is excluded by living

cells, having intact cell membranes, because of its high

molecular size, whereas only cells with damaged or

compromised membranes allow this stain to be internalized

and bind to the nucleic acids. Therefore it is usually used as

a probe for the physiological assessment of the loss in

membrane integrity, and in turn, as an indicator of the loss

in cell viability. The CTC is a tetrazolium redox dye

currently in use as an indicator of respiratory activity

because it is converted by the enzymes of the electron

transport system to the red fluorescent formazan, which

accumulates in granules within active respiring cells (CTC+

cells) (Rodriguez et al. 1992; del Giorgio and Scarborough

1995; Pyle et al. 1999).

In this paper we report the detailed procedures of these

rapid methods and a comparison of data obtained in natural

marine samples assayed for E. coli viability, in order to

discriminate between different physiological states that may

occur during the life cycle of bacteria. Some considerations

in the light of the recent controversy concerning the real

existence of VBNC forms are reported.

MATERIALS AND METHODS

Collection and treatment of samples

Seawater samples were collected using sterile Niskin bottles

from some coastal sites of the Messina Straits with various

degrees of faecal contamination. For each station, a volume

of 2 l was drawn, stored at +5�C, and analysed within 2 h of

sampling. Water subsamples (100 ml or less) were treated in

different ways according to the methods of determination

reported below.

Plate counts. Viable counts were determined by mem-

brane filtration on m-faecal coliforms (m-FC)2 agar (Difco)

plates, according to APHA (American Public Health

Association 1992)3 .

Double PI–FA staining procedure. For the estimation of

the non-viable fraction, the protocol of FA labelling of cells

(Caruso et al. 2000) was modified with simultaneous PI

staining. Briefly, an aliquot of sample (100 ml) was filtered

on a 0Æ22 lm pore size black polycarbonate membrane

(Nuclepore; Whatman Inc. Nucleopore, Newton, MA,

USA4 ) and the filter stained in the dark with PI (Sigma, St

Louis, MO, USA; final concentration: 0Æ01 lg ml)1). After

treatment with hydrolysed gelatine (pH 7Æ2), the filter was

first labelled with Murex E. coli agglutinating sera specific

for enteropathogenic serotypes (pools 2 + 3 + 4, 1 : 40

dilution, 30 min at room temperature) and then with goat

anti rabbit IgG fluorescein isothiocyanate (FITC)-conju-

gated (Sigma) (1 : 80 dilution, 30 min at room temperature).

226 G. CARUSO ET AL.

ª 2003 The Society for Applied Microbiology, Journal of Applied Microbiology, 95, 225–233

A duplicate sample was fixed in formaldehyde (3Æ7%) and

used as a negative control, according to Rodriguez et al.

(1992). Prior to their use in microscopic assays, the specificity

for E. coli of the reagents was confirmed using a collection of

homologous and heterologous bacteria (Caruso et al. 2002).

Observation was performed under a Zeiss Axioplan 2

epifluorescence microscope (Carl Zeiss Vision GmbH,

Munchen, Germany)5 coupled with an image analysis

system, equipped with a 100 W mercury lamp and filter

sets specific for fluorescein (blue light, BP 450–490, FT 510

and LP 520) and Rhodamine (green light, BP 510–560, FT

580 and LP 590), respectively. When excited under green

light, dead (PI-positive) E. coli cells were viewed as red

fluorescing cells, because of PI (emission peak: 617–

623 nm); by switching to the blue light filter, they were

easily distinguished within the total (living + non-living) E.

coli population, green fluorescing because of FITC (emission

peak: 520 nm).

Double CTC–FA staining procedure. The fraction of

viable active cells, showing respiratory (electron transport)

activity, was determined by simultaneous CTC and FA

staining. Labelling with CTC (Polysciences, Warrington,

PA, USA) was performed according to our reference

procedures (Rodriguez et al. 1992; del Giorgio and Scar-

borough 1995), with some modifications. Briefly, 100 ml of

sample was incubated with 5 mmol l)1 CTC (final concen-

tration) for 90 min (instead of 4 h as reported in literature)

in the dark at room temperature and then filtered through a

0Æ22-lm polycarbonate black membrane (Nuclepore). Filters

were labelled and observed according to the steps described

above for PI–FA procedure.

Total FA labelled cells were quantified under blue light

(450–490 nm) for FITC excitation; the viable and actively

respiring cells showed bright red fluorescence because of

CTC-formazan crystals (Pyle et al. 1995) and were scored as

CTC-positive (CTC+) cells by observation under a green

combination filter block, such as the Rhodamine filter set

(Pyle et al. 1995; del Giorgio et al. 1997).

FA staining procedure. The number of total E. coli cells

was determined on a 100-ml subsample fixed in 2%

formaldehyde (final concentration) and treated according

to the indirect FA labelling procedure described in detail by

Caruso et al. (2000).

For all the microscopic procedures, a minimum of 20 fields

was counted at random, and cell abundance was expressed as

the mean value of cells counted per 100 ml of sample.

Statistical analysis

Prior to statistical analysis, data were log-transformed to

stabilize the variance and attain normality. Differences

between mean counts were tested for statistical significance

using analysis of variance (ANOVAANOVA); the Pearson correlation

coefficient was calculated to determine whether values

obtained with different methods were statistically correlated.

RESULTS

Preliminary assays with different concentrations of fluoro-

chrome and incubation times were first necessary to

establish the optimal staining conditions. Various amounts

of PI (5, 1, 0Æ5, 0Æ1, 0Æ05 and 0Æ01 lg ml)1, final concen-

trations) were tested in order to find which was appropriate

as a working concentration (data not reported). Incubation

for 5 min with 1 ml of 0Æ01 lg ml)1 PI solution (final

concentration) was effective to stain dead E. coli cells.

Controls performed with fixed cells showed that PI pene-

trated within non-viable cells. The concentration used was

found to be appropriate to avoid high background fluores-

cence resulting from excessive amounts of fluorescent dye,

observed during previous trials in which final concentrations

ranging from 5 to 0Æ1 lg ml)1 were applied. Also a step of

treatment with hydrolysed gelatine was included to reduce

the non-specific fluorescence. The incubation time of 5 min

was chosen as appropriate for optimal staining.

With respect to CTC, as no modifications were performed

to the staining protocol suggested by del Giorgio and

Scarborough (1995) in terms of the amount of reagent,

5 mmol l)1 final concentration of CTC was used in the

assay. The shortening of the incubation time to 2 h, instead

of 4 h reported in the literature, and the elimination of the

filtration step with NaCl, gave results comparable with those

achieved by using the traditional protocol and avoided

excessive amounts of debris accumulating on the filter.

When viewed under epifluorescence microscope, no intra-

cellular formazan crystals were observed in samples fixed

with formaldehyde before CTC incubation, used as negative

controls.

After CTC–FA and PI–FA labelling, E. coli cells were

clearly detected through microscopic observation, as shown

in Figs 1 and 2.

Bacterial concentrations, with mean values and standard

deviation, found in the coastal samples analysed with the

different methods are reported in Table 1. The relative

percentage of the values obtained by each method with

respect to the total FA count was also calculated (Fig. 3).

Total FA values varied between 6Æ00 · 102 and 2Æ54 ·105 cells (100 ml))1, while the concentration of viable and

culturable cells, obtained by m-FC counts, was from one

to two orders of magnitude lower, ranging from 7Æ84 · 100

to 1Æ13 · 104 CFU (100 ml))1 and accounted for 8%

(mean value) of the total E. coli abundance only. This

quantitative difference suggested the persistence in natural

samples of a large percentage of cells in a viable state, but

ESCHERICHIA COLI VIABIL ITY ASSAYS 227


not detectable by culture method. The fraction of viable and

metabolically active E. coli cells, identified by CTC–

FA staining, was comprised between 4Æ80 · 101 and

1Æ50 · 105 cells (100 ml))1; it represented 46% (mean

value) of total bacterial density as determined by FITC

labelling. Dead cells, detected by PI–FA labelling, ranged

from 6Æ00 · 101 to 5Æ70 · 103 cells (100 ml))1 and were

equivalent to 22% (mean value) of the total bacterial density.

Their mean values were in the same order of magnitude of

those of culturable cells.

The comparison between the numbers of viable E. coli

cells determined by CTC–FA staining and those obtained

by m-FC culture method showed that active cell numbers

were, on average, one order of magnitude higher than plate

counts; CTC+ cells included, in fact, a fraction of in situ still

active (i.e. with detectable respiration) but non-culturable

cells (ABNC, a new definition that has been proposed in

alternative to VBNC, see Barer et al. 1993; Lebaron et al.

1999), being damaged in reproductive ability or unable to

metabolize nutrients. The statistical analysis by ANOVAANOVA test

(F ¼ 2Æ69, n ¼ 23) and Pearson correlation coefficient

(r ¼ 0Æ84, P < 0Æ01, n ¼ 23) suggested, however, that

m-FC and CTC+ cell counts were significantly correlated

and that both methods detected the same viable fraction of

the E. coli population.

The incidence of viable active cells was highly variable

(Fig. 3) and, in one case, it reached a peak value of 87% of

the total count. The high percentage of actively respiring

cells was probably related to the large nutrient inputs and to

the availability of organic matter for bacterial metabolism; in

Fig. 2 Natural sample stained with PI–FA procedure. Epifluorescence

microphotographs obtained by using a Zeiss Axioplan 2 microscope

equipped with filters 09 for FITC and 14 for Rhodamine and a 100X

oil immersion Neofluar objective. Photographs were digitally processed

by Zeiss Axiovision image system and Zeiss Axioviewer. (a) Cells

labelled with Escherichia coli agglutinating sera and fluorescein

isothiocyanate (FITC)-conjugated goat anti rabbit IgG appear as

fluorescing rods; (b) same field of (a), with dead cells stained by

propidium iodide. Bar, 4 lm

Fig. 1 Natural sample stained with CTC–FA procedure.

Epifluorescence microphotographs obtained by using a Zeiss

Axioplan 2 microscope equipped with filters 09 for FITC and 14

for Rhodamine and a 100X oil immersion Neofluar objective.

Photographs were digitally processed by Zeiss Axiovision image

system and Zeiss Axioviewer. (a) Cells labelled with Escherichia coli

agglutinating sera and fluorescein isothiocyanate (FITC)-conjugated

goat anti rabbit IgG appear as fluorescing rods; (b) same field of (a),

showing fluorescence in correspondence of actively respiring cells

because of the intracellular accumulation of fluorescent CTC-formazan

granules. Bar, 4 lm



contrast, the lowest number (4Æ80 · 101 cells 100 ml)1,

equal to 8% of the total count) of viable active E. coli,

observed in one sample collected in June 2002, could be

caused by a rapid cell decline (‘die-off’) of allochthonous

bacteria under adverse environmental conditions.

The number of living cells, with integer membranes, as

estimated by comparing total and dead (PI+) cell counts,

was frequently higher than the number of living active cells

determined by CTC staining. This discrepancy may be

explained by the presence of a percentage of cells that

possess living attributes other than the respiratory activity

and therefore undetected by CTC assay.

A fraction of cells that showed no apparent reduction of

CTC, CTC), numerically characterized by the difference

between the total FA and viable active (respiring) cells, was

identified (Table 1, Fig. 3) and corresponded to dormant or

moribund cells, i.e. cells that had low activity or were not

active. They accounted, on average, for 23% of the total

population, predominating in the samples collected in

October 2002.

In the transition from the viable to the non-viable state,

the last step, represented by cells irreversibly compromised

in their membrane integrity (dead or injured cells), was

distinguished by PI staining. The number of PI positive cells

was inversely correlated with plate and CTC counts,

particularly during the first sampling (r ¼ )0Æ98 and

)0Æ92, P < 0Æ01), confirming that cells labelled by PI

represented the non-viable fraction within the population

and, therefore, that PI labelling was a good indicator of this

physiological state.

Table 1 Bacterial counts, with geometric mean and geometric S.D.S.D., obtained with different methods in coastal samples collected from Messina

Straits. FA, fluorescent antibody stained cells; m-FC, colony-forming units (CFU) on m-FC agar plates; CTC+, active respiring cells, and CTC),

inactive non-respiring cells as determined by CTC and fluorescent antibody (CTC–FA) staining; PI, cells that have lost membrane integrity, stained

by propidium iodide and fluorescent antibody (PI–FA) staining. CTC was used at a 5 mmol l)1 final concentration, PI at a 0Æ01 lg ml)1 final

concentration

Total FA (viable + non-viable) m-FC (culturable) CTC+ active (respiring)*8 CTC) inactive� PI+ dead�

Sample Date Cells (100 ml))1 CFU (100 ml))1 Cells (100 ml))1 Cells (100 ml))1 Cells (100 ml))1

1 20/6/2002 1Æ21E + 03 9Æ02E + 01 8Æ00E + 02 3Æ95E + 01 2Æ80E + 02

2 20/6/2002 1Æ52E + 03 1Æ96E + 02 8Æ00E + 02 3Æ70E + 02 1Æ50E + 02

3 20/6/2002 8Æ92E + 02 4Æ51E + 01 4Æ50E + 02 9Æ65E + 01 3Æ00E + 02

4 20/6/2002 6Æ00E + 02 7Æ84E + 00 4Æ80E + 01 1Æ14E + 02 4Æ30E + 02

5 20/6/2002 7Æ58E + 02 1Æ37E + 01 1Æ80E + 02 1Æ64E + 02 4Æ00E + 02

6 20/6/2002 6Æ80E + 02 1Æ37E + 01 2Æ30E + 02 3Æ63E + 01 4Æ00E + 02

7 20/6/2002 1Æ42E + 03 1Æ65E + 02 7Æ80E + 02 2Æ78E + 02 2Æ00E + 02

8 20/6/2002 1Æ71E + 03 2Æ96E + 02 8Æ90E + 02 4Æ59E + 02 6Æ00E + 01

1A 10/1/2002 4Æ27E + 03 4Æ57E + 02 1Æ80E + 03 5Æ13E + 02 1Æ50E + 03

2A 10/1/2002 4Æ80E + 03 5Æ24E + 02 1Æ90E + 03 2Æ76E + 02 2Æ10E + 03

3A 10/1/2002 8Æ00E + 03 1Æ70E + 03 4Æ20E + 03 6Æ00E + 02 1Æ50E + 03

4A 10/1/2002 3Æ24E + 04 1Æ04E + 02 2Æ28E + 04 3Æ80E + 03 5Æ70E + 03

5A 10/1/2002 3Æ05E + 04 8Æ80E + 01 8Æ00E + 03 1Æ81E + 04 4Æ30E + 03

6A 10/1/2002 2Æ54E + 05 1Æ13E + 04 1Æ50E + 05 9Æ25E + 04 2Æ00E + 02

7A 10/1/2002 1Æ50E + 05 1Æ01E + 04 4Æ47E + 04 9Æ48E + 04 4Æ00E + 02

8A 10/1/2002 1Æ58E + 03 2Æ20E + 02 7Æ40E + 02 5Æ31E + 02 8Æ90E + 01

1B 10/10/2002 2Æ56E + 03 2Æ81E + 02 1Æ26E + 03 2Æ39E + 02 7Æ80E + 02

2B 10/10/2002 2Æ98E + 03 3Æ40E + 02 1Æ29E + 03 1Æ85E + 02 1Æ17E + 03

3B 10/10/2002 4Æ47E + 03 8Æ12E + 02 2Æ35E + 03 3Æ68E + 02 9Æ40E + 02

4B 10/10/2002 1Æ08E + 04 7Æ50E + 01 9Æ40E + 03 1Æ09E + 03 2Æ40E + 02

5B 10/10/2002 1Æ42E + 04 6Æ50E + 01 3Æ70E + 03 7Æ84E + 03 2Æ60E + 03

6B 10/10/2002 1Æ18E + 04 4Æ80E + 02 7Æ50E + 03 3Æ46E + 03 3Æ60E + 02

7B 10/10/2002 5Æ60E + 04 6Æ50E + 03 2Æ10E + 04 2Æ81E + 04 4Æ00E + 02

Mean 5Æ28E + 03 2Æ41E + 02 2Æ23E + 03 8Æ74E + 02 5Æ39E + 02

S.D.S.D. 5Æ61E + 00 7Æ02E + 00 6Æ56E + 00 9Æ81E + 00 3Æ28E + 00

Min 6Æ00E + 02 7Æ84E + 00 4Æ80E + 01 3Æ63E + 01 6Æ00E + 01

Max 2Æ54E + 05 1Æ13E + 04 1Æ50E + 05 9Æ48E + 04 5Æ70E + 03

*CTC-positive, FA-stained cells.

�CTC-negative, FA-stained cells.

�PI-positive, FA-stained cells.



With respect to the significance of E. coli viability counts

performed in a temporal series, the ANOVAANOVA test showed that

significant differences were recorded among the mean

CTC+ and PI+ values obtained during the first and the

third samplings (F ¼ 6Æ05 and 5Æ03, P < 0Æ05, respectively).

DISCUSSION

Improvement of currently available techniques constitutes

the challenge for future developments in the assessment of

microbiological water quality. Standard culture methods

need consistent reconsideration and the proposal of updated

analytical approaches is welcome (Sartory and Watkins

1999). The combined viability and FA staining protocols

developed and applied in our study are one example of

direct microscopic techniques specifically designed for the

simultaneous enumeration and physiological probing of a

selected bacterial species. Through the combination of the

procedures proposed, different progressive physiological

states within the bacterial community were discriminated:

cells that had potential for growth (viable and culturable)

were determined by plate counts, cells that were metabolically

active by respiratory activity (CTC+ cells), inactive cells

(CTC) cells) that included ‘dormant’ or ‘moribund’ cells

(Lebaron et al. 1999), while ‘dead’ or ‘injured’ cells were

evidenced by labelling with PI. Dormant cells, identified

as non-culturable and metabolically inactive cells with

membrane integrity, are recognized as a transient step

between viable and membrane-compromised states (Gregori

et al. 2001). Dead cells represent the last step in the

succession of cellular states that occur during E. coli

starvation, including the loss of culturability, metabolic

activity and DNA content (Muela et al. 1999).

The high percentages of active (respiring) cells found in our

study showed that, at the concentrations used (5 mmol l)1),

CTC did not have toxic effects on bacterial metabolism that

have been found (Ullrich et al. 1996) for high concentrations

(near 10 mmol l)1). The low numbers of CTC+ cells

obtained in other studies carried out in oligotrophic environ-

ments had been explained by possible toxicity phenomena, or

by the inability of CTC to detect the presence of cells with low

metabolic activity, that are scored as negative to the assay

(Gasol and del Giorgio 2000), or by the fact that CTC

measures only respiratory activity and not all the variety of

metabolic activities, with consequent underestimation of the

real fraction of metabolically active cells.

The dual CTC–FA and PI–FA staining procedures

represent an improvement in FA technique, previously

applied for the detection and enumeration of E. coli in

seawater (Caruso et al. 2000, 2002), which expand its

potential application in aquatic microbial ecology. The

assessment of the physiological heterogeneity of E. colipopulation in aquatic ecosystems has a great significance for

its surveillance in the field. The retention in the VBNC state

of some signs of active metabolism and virulence factors

indicates the potential of recoverability and pathogenesis in

appropriate hosts (Smith et al. 1994; Pommepuy et al.

1996); therefore, methods able to discriminate in situ

whether microorganisms are alive or dead hold a key

importance with respect to sanitary monitoring of indicator

bacteria or pathogens (Pyle et al. 1995, 1999; McFeters et al.

1999). A recent research (Bogosian et al. 1998) has, however,

excluded the existence of enteric bacteria in the VBNC state,

concluding that the decline in E. coli cell numbers in water

was as a result of cell death; criticism to this conclusion has

been forwarded by Ohtomo and Saito (2001), who have

proved the existence of E. coli cells in the VBNC state

showing the recovery of ‘injured’ cells from saline-stressed

cultures.

6/20/2002

10/01/2002

10/10/2002

10080

Per

cent

age

of th

e to

tal F

A

6040200

100806040200

100

80

60

40

20

0

1 2 3 4 5 6 7 8

1A 2A 3A 4A 5A 6A 7A 8A

Stations1B 2B 3B 4B

m-FC CTC+ CTC– PI

5B 6B 7B

Fig. 3 Percentage contribution to the total Escherichia coli population

of different bacterial fractions as indicative of various physiological

states: culturable, active respiring, inactive non-respiring, and dead

cells, in the three samplings performed in Messina Straits [total cell

counts were obtained by fluorescent antibody (FA) assay; culturable

cells were estimated on m-FC agar plates; active respiring cells and

inactive non-respiring cells were determined by CTC–FA staining;

dead cells were stained by propidium iodide]



The combined immunofluorescence–viability approach

may reflect the true viable count more accurately than

immunofluorescence alone. In fact, whether CFU counts

result in biased estimations (generally an underestimation),

as they do not consider a large proportion of bacteria that do

not grow on conventional substrates, microscopic counts

certainly overestimate the real bacterial concentration,

including living, both active and/or inactive, dormant and

dead bacteria. Approaches for assessing bacterial viability are

based on the demonstration of culturability, metabolic

activity or maintenance of cellular structures such as the

cytoplasmatic membrane (Lloyd and Hayes 1995; McDoug-

ald et al. 1998), but it is known that the assay of one

property of living cells does not constitute sufficient proof of

viability (Weichart 2000) and none of the methods for the

measurement of cellular activity may be considered adequate

predictors of culturability or viability (Kell et al. 1998; Barer

et al. 1999; Liu 2000). On the other hand, the failure of

culture media to recover bacteria may be partially explained

by the negative impact that selective agents (such as lauryl

sulphate, tergitol 7 and rosolic acid), usually included in the

media conventionally used for E. coli enumeration, may have

on cell viability. The actual quantification of viable enteric

bacteria may be significantly compromised by these com-

pounds, to which microbes are normally resistant, while they

become more sensitive when cells are stressed (Smith et al.

1994); in this case, they may undergo a few divisions but in

insufficient numbers to produce bacterial colonies (Joux and

Lebaron 1997).

Our analytical protocols are similar to those currently in

use in flow cytometry (FCM), where a wide range of

fluorescent probes specifically target various aspects of

bacterial single-cell activity (Gasol and del Giorgio 2000)

and allow the differentiation of cells according to their

physiological state (Lopez-Amoros et al. 1995; Porter et al.1996; Joux et al. 1997). FA staining protocols combined

with PI or CTC as fluorogenic markers for cell viability have

been reported for the detection of Salmonella typhimurium

(Clarke and Pinder 1998) or E. coli (Yamaguchi et al. 2001)

in food samples by FCM. Rapid assessment of E. coli

viability in sterile lake water was performed by Porter et al.

(1995) through antibody labelling, fluorescent dyes and

FCM.

The methodological approach here proposed is charac-

terized by speed and simplicity of execution, specificity,

sensitivity and reduced costs. Simultaneous labelling of cells

with antibodies and viability markers does not interfere with

the specificity of the antigen–antibody binding, as FITC-

labelled antibodies specifically recognize the antigenic

determinants on the cell surface, while PI or CTC bind to

the nucleic or cytoplasmatic portion, respectively. The

viability substrates used did not overlap their emission

spectra with the emission spectra of the antibody labels.

The dual labelling protocol also has a potential application

in ecological studies, where it may be used for the detection

of the viability of specific bacterial species of ecological

interest, i.e. involved in biogeochemical cycles, provided that

immune sera are available and specific for target bacteria. A

combination of the direct FA technique with the tetrazolium

salt INT (2-p-iodophenyl-3-p-nitrophenyl-5-phenyl tetra-

zolium chloride) was previously applied to natural water

samples for the enumeration of actively respiring cells of

Thiobacillus ferrooxidans (Baker and Mills 1982). Fliermans

and Schmidt (1975) combined microautoradiography and

FA staining for the direct microscopic enumeration of viable

Nitrobacter cells in soils. In studies concerning the function

of microbial assemblages in natural environments, assess-

ment of bacterial viability in natural samples may give

additional information on the fraction of active cells that

effectively play a role in the microbial processes involved in

the ecosystem functioning (del Giorgio and Scarborough

1995; Sherr et al. 1999; Gregori et al. 2001).

ACKNOWLEDGEMENTS

This work has been financially supported by the Italian

MIUR (Ministry for University and Scientific Research)

within the frame of the Cluster-10 SAM Project (2000–

2003).

REFERENCES

APHA (American Public Health Association) (1992) Standard Methods

for the Examination of Water and Waste Water, 18th edn. Washing-

ton, DC: American Public Health Association.

Baker, K.H. and Mills, A.L. (1982) Determination of the number of

respiring Thiobacillus ferrooxidans cells in water samples by using

combined fluorescent antibody-2-(p-iodophenyl)-3-(p-nitrophenyl)-

5-phenyltetrazolium chloride staining. Applied and Environmental

Microbiology 43, 338–344.

Barcina, I., Lebaron, P. and Vives-Rego, J. (1997) Survival of

allochthnous bacteria in aquatic systems: a biological approach.

FEMS Microbiology Ecology 23, 1–9.

Barer, M.R., Gribbon, L.T., Harwood, C.R. and Nwoguh, C.E. (1993)

The viable but non-culturable hypothesis and medical bacteriology.

Reviews in Medical Microbiology 4, 183–191.

Barer, M.R., Gribbon, L.T., Whiteley, A.S., Smith, R., Newton, A.

and O’Donnell, A.G. (1999) Relationships between cellular activity

and culturability. In Microbial Biosystems: New Frontiers ed. Bell,

C.R., Brylinsky, M. and Johnson-Green, P. Halifax, Canada:

Atlantic Canada Society for Microbial Ecology http://plato.

arcadiau.ca/isme/Symposium23/barer.pdf.6

Bogosian, G., Morris, J.L. and O’Neil, J.P. (1998) A mixed culture

recovery method indicates that enteric bacteria do not enter the

viable but nonculturable state. Applied and Environmental Microbio-

logy 64, 1736–1742.

Caruso, G., Zaccone, R. and Crisafi, E. (2000) Use of the indirect

immunofluorescence method for detection and enumeration of



Escherichia coli in seawater samples. Letters in Applied Microbiology

31, 274–278.

Caruso, G., Crisafi, E. and Mancuso, M. (2002) Immunofluorescence

detection of Escherichia coli in seawater: a comparison of various

commercial antisera. Journal of Immunoassay and Immunochemistry

23, 479–496.

Clarke, R.G. and Pinder, A.C. (1998) Improved detection of bacteria

by flow cytometry using a combination of antibody and viability

markers. Journal of Applied Microbiology 84, 577–584.

Ericsson, M., Hanstorp, D., Hagberg, P., Enger, J. and Nystrom, T.

(2000) Sorting out bacterial viability with optical tweezers. Journal of

Bacteriology 182, 5551–5555.

Fliermans, C.B. and Schmidt, E.L. (1975) Autoradiography and

immunofluorescence combined for autoecological study of single cell

activity with Nitrobacter as a model system. Applied Microbiology 30,

676–684.

Gasol, J.M. and del Giorgio, P.A. (2000) Using flow cytometry for

counting natural planktonic bacteria and understanding the structure

of planktonic bacterial communities. Scientia Marina 64, 197–224.

del Giorgio, P.A. and Scarborough, G. (1995) Increase in the

proportion of metabolically active bacteria along gradients of

enrichment in freshwaters and marine plankton: implications for

estimates of bacterial growth and production rates. Journal of

Plankton Research 17, 1905–1924.

del Giorgio, P.A., Prairie, Y.T. and Bird, D.F. (1997) Coupling

between rates of bacterial production and the abundance of

metabolically active bacteria in lakes, enumerated using CTC

reduction and flow cytometry. Microbial Ecology 34, 144–154.

Gregori, G., Citterio, S., Ghiani, A., Labra, M., Sgorbati, S., Brown, S.

and Denis, M. (2001) Resolution of viable and membrane-compro-

mised bacteria in freshwater and marine waters based on analytical

flow cytometry and nucleic acid double staining. Applied and

Environmental Microbiology 67, 4662–4670.

Hood, M.A. and Ness, G.E. (1982) Survival of Vibrio cholerae and

Escherichia coli in estuarine waters and sediments. Applied and


Joux, F. and Lebaron, P. (1997) Ecological implication of an improved

direct viable count method for aquatic bacteria. Applied and


Joux, F., Lebaron, P. and Troussellier, M. (1997) Succession of cellular

states in a Salmonella typhimurium population during starvation

in artificial seawater microcosms. FEMS Microbiology Ecology 22,

65–76.

Kell, D.B., Kaprelyants, A.S., Weichart, D.H., Harwood, C.R. and

Barer, M.R. (1998) Viability and activity in readily culturable

bacteria: a review and discussion of the practical issues. Antonie Van

Leeuwenhoek 73, 169–187.

Lebaron, P., Bernard, L., Baudart, J. and Courties, C. (1999) The

ecological role of VBNC cells in the marine environment. In

Microbial Biosystems: New Frontiers ed. Bell, C.R., Brylinsky, M. and

Johnson-Green, P. Halifax, Canada: Atlantic Canada Society for

Microbial Ecology http://plato.arcadiau.ca/isme/Symposium23/

lebaron.pdf.7

Liu, S.V. (2000) Viable but non-culturable (VBNC) microorganisms: a

misnomer or a whistle-blower? Logical Biology 1, 17–20.

Lloyd, D. and Hayes, A.J. (1995) Vigor, vitality and viability of

microorganisms. FEMS Microbiology Letters 133, 1–7.

Lopez-Amoros, R., Comas, J. and Vives-Rego, J. (1995) Flow

cytometric assessment of Escherichia coli and Salmonella typhimurium

starvation-survival in seawater using rhodamine 123, propidium

iodide, and oxonol. Applied and Environmental Microbiology 61,

2521–2526.

McDougald, D., Rice, S.A., Weichart, D. and Kjelleberg, S. (1998)

Nonculturability: adaptation or debilitation? FEMS Microbiology

Ecology 25, 1–9.

McFeters, G.A., Yu, F.P., Pyle, B.H. and Stewart, P.S. (1995)

Physiological assessment of bacteria using fluorochromes. Journal of

Microbiological Methods 21, 1–13.

McFeters, G.A., Pyle, B.H., Lisle, J.T. and Broadaway, S.C. (1999)

Rapid direct methods for enumeration of specific, active bacteria in

water and biofilms. Journal of Applied Microbiology Symposium

Supplement 85, 193S–200S.

Muela, A., Arana, I., Justo, J.I., Seco, C. and Barcina, I. (1999)

Changes in DNA content and cellular death during a starvation-

survival process of Escherichia coli in river water. Microbial Ecology

37, 62–69.

Ohtomo, R. and Saito, M. (2001) Increase in the culturable cell number

of Escherichia coli during recovery from saline stress: possible

implication for resuscitation from the VBNC state. Microbial Ecology

42, 208–214.

Oliver, J.D. (1993) Formation of viable but non culturable cells. In

Starvation in Bacteria ed. Kjelleberg, S. pp. 239–272. New York:

Plenum Press.

Petit, M., George, I. and Servais, P. (2000) Survival of Escherichia coli

in freshwater: b-D-glucuronidase activity measurements and char-

acterization of cellular states. Canadian Journal of Microbiology 46,

679–684.

Pommepuy, M., Butin, M., Derrien, A., Gourmelon, M., Colwell,

R.R. and Cormier, M. (1996) Retention of enteropathogenicity by

viable but nonculturable Escherichia coli exposed to seawater and

sunlight. Applied and Environmental Microbiology 62, 4621–4626.

Porter, J., Edwards, C. and Pickup, R.W. (1995) Rapid assessment of

physiological status in Escherichia coli using fluorescent probes.

Journal of Applied Bacteriology 79, 399–408.

Porter, J., Deere, J., Pickup, R. and Edwards, C. (1996) Fluorescent

probes and flow cytometry: new insights into environmental

bacteriology. Cytometry 23, 91–96.

Pyle, B.H., Broadaway, S.C. and McFeters, G.A. (1995) A rapid, direct

method for enumerating respiring enterohemorrhagic Escherichia coli

O157:H7 in water. Applied and Environmental Microbiology 61, 2614–

2619.

Pyle, B.H., Broadaway, S.C. and McFeters, G.A. (1999) Sensitive

detection of Escherichia coli O157:H7 in food and water by immuno-

magnetic separation and solid-phase laser cytometry. Applied and


Rodriguez, G.G., Phipps, D., Ishiguro, K. and Ridhway, H.F. (1992)

Use of a fluorescent redox probe for direct visualization of actively

respiring bacteria. Applied and Environmental Microbiology 58, 1801–

1808.

Roszack, D.B. and Colwell, R.R. (1987) Metabolic activity of bacterial

cells enumerated by direct viable count. Applied and Environmental


Rozen, Y. and Belkin, S. (2001) Survival of enteric bacteria in seawater.

FEMS Microbiology Reviews 725, 1–17.



Sartory, D.P. and Watkins, J. (1999) Conventional culture for water

quality assessment: is there a future? Journal of Applied Microbiology

Symposium Supplement 85, 225S–233S.

Sherr, B.F., del Giorgio, P. and Sherr, E.B. (1999) Estimating

abundance and single-cell characteristics of respiring bacteria via the

redox dye CTC. Aquatic Microbial Ecology 18, 117–131.

Smith, J.J., Howington, J.P. and McFeters, G.A. (1994) Survival,

physiological response, and recovery of enteric bacteria exposed to a

polar marine environment. Applied and Environmental Microbiology

60, 2977–2984.

Ullrich, S., Karrasch, B., Hoppe, H.-G., Jeskulke, K. and Mehrens, M.

(1996) Toxic effects on bacterial metabolism of the redox dye

5-cyano-2,3-ditolyl tetrazolium chloride. Applied and Environmental


Weichart, D. (2000) How viable is the VBNC hypothesis? Logical

Biology 1, 21–24.

Xu, H.S., Roberts, N., Singleton, F.L., Attwell, R.W., Grimes, D.J.

and Colwell, R.R. (1982) Survival and viability of nonculturable

Escherichia coli and Vibrio cholerae in the estuarine and marine

environment. Microbial Ecology 8, 313–323.

Yamaguchi, N.Y., Sasada, M. and Nasu, M. (2001) Rapid detection of

respiring Escherichia coli O157:H7 in food by flow cytometry. In 9th

International Symposium on Microbial Ecology (ISME-9), Amster-

dam (The Netherlands), August 26–31, 2001, Abstracts, p. 39.



combined fluorescent antibody assay and viability staining for the assessment of the physiological...

Documents