FEMS Microbiology Reviews 25 (2001) 513^529
www.fems-microbiology.org
Survival of enteric bacteria in seawater
Yael Rozen, Shimshon Belkin *
Environmental Sciences, Fredy and Nadine Herrmann Graduate School of Applied Science, Hebrew University of Jerusalem, Jerusalem 91904, Israel
Accepted 19 June 2001
First published online 16 July 2001
Abstract
Enteric bacteria exposed to the marine environment simultaneously encounter a variety of abiotic and biotic challenges. Among the
former, light appears to be critical in affecting seawater survival; previous growth history plays a major part in preadaptation of the cells,
and stationary phase cells are generally more resistant than exponentially growing ones. Predation, mostly by protozoa, is probably the
most significant biotic factor. Using Escherichia coli as a model, a surprisingly small number of genes was found that, when mutated,
significantly affect seawater sensitivity of this bacterium. Most prominent among those is rpoS, which was also dominant among genes
induced upon transfer to seawater. ß 2001 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All
rights reserved.
Keywords : Enteric bacterium; Seawater survival; Stress response; Viable but nonculturable; Wastewater; Escherichia coli
Contents
1.
2.
3.
4.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
513
513
513
516
516
517
518
518
518
520
520
521
521
521
522
522
522
523
524
524
Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
525
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
525
5.
6.
7.
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Terms and abbreviations . . . . . . . . . . . . . . . . . . . . . . .
Viability and colony formation . . . . . . . . . . . . . . . . . .
Abiotic factors a¡ecting survival . . . . . . . . . . . . . . . . .
4.1. Light . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.2. Salinity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.3. pH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.4. Nutrient deprivation . . . . . . . . . . . . . . . . . . . . . . .
4.5. Temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.6. Hydrostatic pressure . . . . . . . . . . . . . . . . . . . . . . .
4.7. Sedimentation . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4.8. Previous growth history and stress cross-protection
Biotic factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.1. Grazing and competition . . . . . . . . . . . . . . . . . . . .
5.2. Bacteriophages . . . . . . . . . . . . . . . . . . . . . . . . . . .
5.3. Antibiotics and toxins . . . . . . . . . . . . . . . . . . . . . .
Molecular mechanisms involved in seawater survival . .
6.1. E¡ect of speci¢c mutations . . . . . . . . . . . . . . . . . .
6.2. Genes induced upon seawater exposure . . . . . . . . .
Concluding remarks . . . . . . . . . . . . . . . . . . . . . . . . . .
* Corresponding author. Tel. : +972 (2) 658 4192; Fax: +972 (2) 658 5559.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
E-mail address: [email protected] (S. Belkin).
0168-6445 / 01 / $20.00 ß 2001 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved.
PII: S 0 1 6 8 - 6 4 4 5 ( 0 1 ) 0 0 0 6 5 - 1
FEMSRE 725 5-12-01
514
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
1. Introduction
3. Viability and colony formation
The factors controlling the survival of enteric bacteria in
the marine environment have intrigued scientists for decades. Driven by obvious public health concerns as well as
by broader attempts to understand bacterial responses to
environmental stress, numerous studies have explored the
fate of Escherichia coli and other enteric bacteria following
their exposure to seawater.
Many of these e¡orts were motivated by the need to
properly evaluate the risk posed by such microorganisms
when released into the sea, either to the health of bathers
in recreational waters or to the safety of ¢sheries or marine agriculture. Consequently, colony formation-based
coliform die-o¡ rates were often the main parameter
used to characterize the bacterial responses, under a variety of biotic and abiotic test conditions. Only in recent
years is there an increasing number of reports attempting
to assay additional viability criteria, as well as to preliminarily explore the molecular mechanisms involved and
their genetic regulation.
Over the years there have been very few publications
summarizing available information in this ¢eld [1,2]; the
most recent of these dates to the mid-80s and focuses
mainly on the di¡erent viability states of enteric bacteria.
The aim of the present review is therefore to partially
remedy this situation by providing an updated compilation
of the relevant scienti¢c literature. For this purpose, following a discussion of the intricacies of colony formation
and other modes of viability testing upon seawater exposure, we shall review the main biotic and abiotic factors
reported to a¡ect the sensitivity and survival of enteric
bacteria in the sea; we shall then proceed to summarize
the relatively scant available information concerning the
molecular control of these e¡ects, and point at future directions which will need to be undertaken.
As pointed out above, in many of the investigations into
the survival of E. coli and other enteric bacteria in the sea,
colony formation ability was used as the main or only
viability parameter. Roszak and Colwell [3], reviewing survival strategies of microorganisms, and Grimes et al. [2]
highlighted the limitations of this approach. The `viable
but nonculturable' (VBNC) state concept was introduced
to describe cells that remain metabolically active but are
unable to divide in or on nutritional media that normally
support their growth. This state was demonstrated for
many enteric bacterial species during seawater incubation
in the dark [4^10] and in the light [11,12]. Gauthier [13], in
his review of the environmental parameters associated
with the VBNC state, discusses the e¡ects of temperature,
radiation, osmolarity and organic matter concentration as
possible triggers for the VBNC state. He points out that,
as in other cases, the multi-stress situation in seawater,
along with the cells' previous history, plays an important
role in their possible entry into VBNC status.
Since marine bacterial pollution is routinely monitored
by culturing methods, the ¢ndings that enteric VBNC bacteria in seawater can be resuscitated [14,15] or retain their
pathogenicity [6] are signi¢cant. Other studies, however,
demonstrated loss of pathogenicity concomitantly with
culturability [8] or failed to resuscitate VBNC cells of several species using various methods [16,17]. The latter reports, according to their authors, cast a doubt on the
validity of the entire VBNC concept.
It is thus clear that interpretation of seawater viability
studies depends, to a large extent, upon the diagnostic
methodologies adopted. Furthermore, the results are also
bound to be in£uenced by the speci¢c conditions during
and even before the actual experiments, including the bacterial strains used, their growth history, and the media
they were allowed to grow in and/or form colonies on.
Other complications may arise from di¡erent parameters
monitored or units of measurements used. For these reasons, comparing results of di¡erent experiments may be
problematic. Nevertheless, while summarizing available information, we shall also try to point at emerging general
trends and patterns.
It is broadly accepted that the ¢rst ability enteric bacterial cells lose in seawater is the capacity to form colonies
on a solid medium. Later [7,9,18^21], or in some cases in
parallel [4,5,9,16,17], the viability represented as direct viable count (DVC) is lost. Total counts seem to remain almost constant during all reported experiments [4,5,7,9,16^
19,21].
Long-term experiments (7^300 days) in which various
viability parameters were measured revealed additional
changes in the state of the cell with incubation time in
seawater. Garcia-Lara et al. [18] reported on E. coli 536,
previously grown at 30³C in a minimal medium (M9G),
which was transferred to sterile seawater (20³C) for
2. Terms and abbreviations
CFU, colony forming unit, also referred to as `culturable cells', viable counts; CTC, 5-cyano-2,3-ditolyltetrazolium chloride, a dye for direct (microscopic) enumeration
of respiring cells ; DVC, direct viable count, a direct (microscopic) enumeration of substrate-responsive cells ; TC,
total count, all microscopically countable cells, direct
count of cells retaining cellular integrity usually stained
by nucleic acid stains such as DAPI (4P,6-diamidino-2phenylindole) or acridine orange; VBNC, viable but nonculturable, cells that retain some form of viability, but can
no longer divide in or on media that usually support their
growth.
FEMSRE 725 5-12-01
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
30 days. Most of the cells (a 4 log reduction) lost their
colony forming ability, but maintained constant total
counts (TC). Cellular DNA concentrations increased signi¢cantly during the ¢rst 5 days, and [3 H]thymidine incorporation rates kept increasing for 13 days, following which
both parameters were maintained at a stable level. Electron transfer activity and cell-attached exoproteolytic activity remained constant during the entire experiment,
while cell-free exoproteolytic activity was drastically decreased [18].
Another experiment [5,9] measured multicellular parameters in Salmonella typhimurium cells during incubation in
sterile arti¢cial seawater (20³C) in the dark for 19 days.
Colony forming ability, enumerated as colony forming
units (CFU), was lost ¢rst and decreased by 6 logs within
the course of the experiment. Cells showing metabolic activity in the presence of nutrients (DVC) decreased in concentration to an undetectable level (more than a 4 log
reduction) in 4 days. Real respiration (with no nutrient
addition) and potential respiration (after nutrient addition), both assayed by 5-cyano-2,3-ditolyltetrazolium chloride (CTC) reduction, decreased by 3 logs within 10 and
13 days, respectively. The other three parameters measured ^ cellular integrity (TC), genomic integrity (cells
the DNA content of which was equal or superior to one
equivalent genome) and membrane permeability ^ remained constant during the entire experiment [5,9].
Somewhat di¡erent results were obtained in a long-term
experiment (294 days) in which the decay of ¢ve stationary
phase enteric bacterial species (E. coli, Klebsiella pneumoniae, Enterococcus faecalis, Enterobacter aerogenes, and
Salmonella choleraesuis) in arti¢cial seawater microcosms
(20³C) was monitored using di¡erent methods. Plate
counts (CFU) and DVC declined gradually and in parallel,
while TC remained constant during the entire experiment,
indicating cell membrane integrity. Respiratory activity
was a¡ected to various extents in the di¡erent species
[17].
In similar experiments conducted by Smith et al. [21] at
the polar environment of McMurdo station in Antarctica,
the survival of four stationary phase enteric bacteria (E.
coli, S. typhimurium, Yersinia enterocolitica and E. faecalis) in di¡usion chambers (31.8³C) was measured by several methods (TC, DVC, CTC, CFU) for 54^56 days. All
strains retained stable TC, but colony formation ability
(CFU), CTC reduction and DVC were decreased; the decay was gradual for the ¢rst three strains and rapid for E.
faecalis (CTC reduction capability loss within 10 days).
The e¡ect of sterilized seawater (22³C) on exponentially
growing S. typhimurium was monitored during 32 days
[20]. In that period, less than a 10-fold decrease was observed in all cell count parameters studied ^ TC, DVC and
CFU. Cellular carbohydrates, lipids, proteins and DNA
gradually decreased to a level of 5^25% at day 32; RNA
content did not decrease signi¢cantly [20]. Morphological
changes were in the following order: cell size reduction,
515
cell wall damage, a wrinkling of the outer membrane, degenerative cellular forms, loss of outer membrane, and
compression of nuclear region into the center of the cell.
Hydrophobicity and adherence to eukaryotic cells (Hep2)
decreased with seawater incubation time while susceptibility to phagocytosis (peritoneal macrophages) increased.
All of the above correlated with the virulence of S. typhimurium as tested on mice: mortality of infected animals
(as recorded at day 7) decreased from 80% at the ¢rst day
to 30% by the 16th day [20].
As can be expected, the ability to form colonies on solid
media is largely dependent on the medium used. Morinigo
et al. [22] de¢ned sublethal cellular injury as the loss of
ability to form colonies on selective media, while retaining
it on less selective ones. This phenomenon was observed in
E. coli, Streptococcus faecalis and Salmonella spp., all of
which formed fewer colonies on selective media (Endo
agar, KF agar, and xylose-lysine-deoxycholate agar, respectively) than on trypticase soy agar [22]. In a di¡erent
study [23], the survival of a marine E. coli isolate in a
dialysis bag submerged in seawater was examined. While
CFUs enumerated on seawater-based nutrient agar remained constant for 6 days, on standard Levine EMB
agar the counts declined to zero by the ¢fth day. The
same protective e¡ect of the recovery medium salinity
was observed by Gauthier et al. [24], who found that the
highest recovery rate of E. coli K12 incubated in seawater
was on a nonselective medium (nutrient agar) supplemented with sodium chloride (15 g l31 ).
Assuming that enzymatic activities in VBNC cells may
still be detectable, disparities may be expected between
colony-based counts and fecal coliform enumeration assays based on enzymatic activities. While in studies monitoring sewage pollution in marine waters the results were
well correlated [25,26], when enzyme activity of pure cultures was compared to their colony forming ability such
discrepancies were indeed recorded [19,27,28]. Two studies
described L-D-galactosidase activity in seawater even when
no culturable cells could be detected [27,28]. In one of
these an environmental E. coli isolate was tested for culturability, viability and L-D-galactosidase activity for 85
days in seawater. Culturability decreased to zero after 60
days while viability (2-(b-iodophenyl)-3-(b-nitrophenyl)-5phenyltetrazolium (INT) chloride reduction) decreased by
1^2 logs during the same period. L-D-Galactosidase activity, after an initial increase, remained constant for the
duration of the experiment, even when the number of culturable cells reached zero [27]. L-D-Galactosidase activity
as an indication of viability was also measured by Pommepuy et al. [28] using E. coli H10407. Once again, a
decrease in culturable cells was observed while enzymatic
activity remained constant. When the experiment was repeated under visible light exposure, L-D-galactosidase activity was also a¡ected, although it was still detectable
when no colony forming cells remained.
The same E. coli strain was used by Fiksdal et al. [19],
FEMSRE 725 5-12-01
516
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
who measured its 4-methylumbelliferyl heptanoate hydrolase activity in seawater. During the ¢rst 16 days CFUs
decreased by 2^3 logs while enzyme activity appeared to
increase.
Another possible indication of viability was provided
for E. coli strains JM83 and JM101 that were shown to
maintain plasmids (pBR322 and pUC8, respectively) at a
high copy number for at least 21 days in arti¢cial seawater
microcosms, even though both strains lost their ability to
form colonies on a solid medium [29].
Resuscitation of VBNC cells was demonstrated using
several methods. The restoration of colony forming ability
by betaine was achieved in osmotically stressed E. coli
CA8000 [14]. The cells were exposed to 0.8 M NaCl in a
minimal medium, reducing CFU counts by 75% within 2 h.
The addition of betaine (2 mM) restored colony forming
ability in about 80% of the cells within a similar time
period. Colony forming ability could still be restored after
24 h in NaCl, even in the presence of the protein synthesis
inhibitor chloramphenicol. The latter observation indicates that the betaine-driven increase in CFU was not
due to cryptic growth of culturable cells [14]. A di¡erent
resuscitation strategy was employed by Roszak et al. [15]:
Salmonella enteritidis C1 cells were revived 4 days after the
nonculturable state was achieved, by a 25 h incubation in
full strength veal infusion broth followed by colony formation on veal infusion agar.
The question whether E. coli can actually enter a viable
but nonculturable state was critically examined by Bogosian et al. [16] using stationary phase strain W3110 cells in
water and soil. In sterile arti¢cial seawater, total counts
were constant during the 56 days of the experiment ; at the
same time, viability decreased by 3 or 5 logs (at 20³C and
37³C, respectively) regardless of the parameter monitored
(CFU, DVC or growth in liquid medium). Attempts to
resuscitate these VBNC cells by various means including
incubation in di¡erent media, a shift to a low temperature,
and an osmotic adaptation, did not succeed [16].
Another study attempted to test the hypothesis that the
presence of culturable cells may be required for the recovery of the nonculturable ones. Cells of di¡erent strains (E.
coli, K. pneumoniae, E. faecalis, E. aerogenes, and S. choleraesuis) were incubated in arti¢cial seawater for nearly
300 days. Resuscitation was attempted by nutrient addition or a temperature shift (20³C to 4³C or to 37³C) both
in the presence and in the absence of culturable cells. At
all time points tested, only cells of the culturable strains
could be recovered [17].
In their study of the potential pathogenicity of stationary phase S. typhimurium C52 VBNC (driven by sterile
arti¢cial seawater and UV-C stress) cells, Caro et al. [8]
claimed the concomitant loss of culturability and pathogenicity (tested in a mouse model), independent of the
level of viability. Cell viability as determined by respiratory activity (CTC reduction), cytoplasmic membrane
(£uorescent staining) and genomic integrity (TC) remained
unchanged, but metabolic activity (DVC) declined together with culturability and could not be distinguished from
the latter. In contrast, Pommepuy et al. [6] demonstrated
pathogenic e¡ects of E. coli H10407 cells induced to the
VBNC state by a combined seawater and light exposure.
Enteropathogenicity was demonstrated by a rabbit intestinal loop assay that was con¢rmed by GM1-ELIZA, a
sensitive assay for enterotoxin production. As DVC counts
were almost constant during the experiments, and no culturable cells could be detected, it was interpreted that the
entire population produced the enterotoxin rather than a
few active cells [6].
A general conclusion that may be drawn from many of
the studies described above is that the hostile situation
encountered by enteric bacteria in seawater promotes the
loss of colony forming ability while maintaining di¡erent
aspects of viability. As already pointed out in the past,
their fate can thus be considered a special case of the
VBNC phenomenon [13]. While it is di¤cult to point
out which, if any, of the several stress factors in this environment may be more important, it is tempting to ¢nd
similarities between this situation and the responses to
starvation, often a major player in the entry into the
VBNC condition [13].
4. Abiotic factors a¡ecting survival
4.1. Light
Survival of enteric bacteria in the sea is greatly a¡ected
both by UV and visible light. In fact, light is considered to
be the single most important contributor to bacterial dieo¡ in the sea [30^33], although its e¡ects are restricted to
shallow depths [33]. A depth-dependent e¡ect was also
implicated in the importance of the radiation wavelength
[34,35].
The toxic e¡ect of light on E. coli in seawater was demonstrated to cause a rapid decrease of colony forming
ability [9,12,32,36]. The UV-B (280^320 nm) portion of
the solar spectrum is the most bactericidal, causing direct
photobiological DNA damage [37]. At higher wavelengths,
photochemical mechanisms become more important, usually acting through photosensitizers and tending to be
more injurious in the presence of oxygen [33].
Fecal microorganisms di¡ered in their sensitivity to light
in seawater [33^35], and greater sunlight exposure was
required to inactivate enterococci compared to fecal coliforms ; the latter group was also found to be more sensitive to light inactivation than fecal bacteriophages (somatic coliphages, F3 DNA phages, F3 RNA phages
and Bacillus fragilis phages) [35].
Light was also demonstrated to exert a signi¢cant negative e¡ect on culturability in an in situ experiment, where
natural sewage populations of E. coli and enterococci
where placed 1^1.5 m deep in seawater for 7 days. The
FEMSRE 725 5-12-01
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
e¡ect was pronounced when batch culture cells were used,
but not in di¡usion chambers [38].
In a di¡erent study, E. coli and S. typhimurium were
exposed to light (approx. 1 MJ m32 h31 ) in seawater for
4 and 10 h, respectively. Total counts were not a¡ected,
while DVC and CFU in both strains decreased by 4^5 logs
[12]. Pommepuy et al. [6] demonstrated that following a 26
h exposure to natural sunlight in seawater, E. coli cells
showed stable DVC and TC counts, but CFUs decreased
from 2U106 ml31 to an undetectable level. Somewhat different e¡ects were obtained under arti¢cial visible light (40
klux) ; E. coli H10407 concentrations were una¡ected
when measured as TC, while DVC and CFU exhibited
approx. 1.5 and 4 log reduction, respectively, within a 40
h experiment [39]. Similar results were presented [9] for E.
coli exposed to visible light (125 W m32 ) for the same
duration: no decrease in TC, approx. a 1 log decrease in
DVC and a 6 log decrease in CFU. In non-illuminated
bacteria, only a 1.5 log decrease in the CFU was measured, with no e¡ect on either TC or DVC [9].
Kapuscinski and Mitchell [40] showed sublethal injuries
to an E. coli isolate incubated in autoclaved ¢ltered seawater exposed to sunlight. The injuries were apparent during the recovery stage on complete (lactose nutrient agar)
or minimal (minimal glucose agar) media, and were not
observed in the dark. Addition of either pyruvate or catalase to the minimal medium restored colony forming ability to the levels of the complete medium, implying at least
a partial involvement of peroxide damage.
The involvement of direct photooxidative processes in
the toxicity of visible light to E. coli was pointed out by
Gourmelon et al. [11]. The e¡ect was signi¢cantly lower
under anaerobic conditions, and was also alleviated in the
presence of a hydroxyl radical scavenger (thiourea), an
iron chelator (desferrioxamine B), or catalase.
The role of reactive oxygen species in the toxic e¡ect of
visible light on E. coli in seawater was also studied by the
use of mutants, defective in speci¢c genes involved in antioxidant defense systems. Exposure to arti¢cial light (0.9
mmol photons m32 s31 ) for 50^100 h highlighted the importance of a functional rpoS for cell survival as determined by colony counts [41]. RpoS is a c factor involved
in various environmental stresses and its presence was
shown to have a protective e¡ect on stationary phase E.
coli in seawater [42,43]. The RpoS protective e¡ect against
light was evident only in stationary phase cells, and a
mutation in the rpoS gene caused a CFU decrease of
about 5 logs compared to the wild-type. In the stationary
phase, mutants defective in genes involved in resistance to
hydrogen peroxide (katE katG (catalases), dps (DNA binding protein) and katE katG dps) were all more sensitive to
light (1^2 logs) than their corresponding wild-types [41]. In
exponentially growing cells, a mutation in oxyR, the regulatory gene of the adaptive response to H2 O2 , led to an
increased sensitivity to light (approx. 1 log) [41], further
suggesting that the deleterious e¡ects may be associated
517
with H2 O2 production. However, the increased sensitivity
of the rpoS mutant suggested that rpoS-dependent protection against deleterious e¡ects may be independent of
H2 O2 removal [41]. Possible involvement of superoxide
radicals was indicated by an enhanced sensitivity of the
double superoxide dismutase mutation (sodA sodB); individually, each of these mutations had no e¡ect. Other
mutants the seawater survival of which was not a¡ected
in the light were the double mutants otsA otsB (trehalose
synthesis), fpg uvrA, and recA (DNA repair), xthA (exonuclease III), and katE. Interestingly, a mutation in hemA
(glutamyl-tRNA reductase), involved in N-aminolevulinic
acid formation from glutamate, had a protective e¡ect
[39].
A certain mitigation of the hazardous e¡ect of solar
radiation may be provided by dissolved organic material
[44], chlorophyll, and particulate matter [45,46]. Such an
e¡ect is more pronounced in relatively eutrophic coastal
and estuarine areas, and is particularly relevant for UV-B
radiation (280^320 nm) owing to selective absorption of
shorter wavelengths. Consequently, bacteria in such areas
are primarily exposed to visible light (400^775 nm) and, to
a lesser extent, to UV-A (320^400 nm) radiation.
An environmental condition that was found to enhance
the bactericidal e¡ects of visible light was salinity, suggesting a synergistic e¡ect of the combined stress factors. It
was proposed [39,41] that the salinity-driven sensitization
is mediated by endogenous components, such as protoporphyrin or ubiquinone biosynthesis intermediates.
4.2. Salinity
When released into the sea, enteric bacteria are subjected to an immediate osmotic upshock, and their ability
to overcome this by means of several osmoregulatory systems could largely in£uence their subsequent survival in
the marine environment [24,47].
Survival of E. coli in seawater/distilled water mixtures at
di¡erent ratios (0, 25, 50, 75 and 100% seawater) for 48 h
showed an optimal survival (74%) at 25% seawater. Survival in the end members ^ distilled water and seawater ^
was 60% and 8%, respectively [48]. A generally similar
e¡ect was reported by Anderson et al. [49], who monitored
the survival of an E. coli isolate for 8 days in seawater at
selected salinities (1, 1.5, 2.5, and 3%): decreasing salinity
was accompanied by increasing survival. Similar results
were obtained when NaCl solutions were used instead of
seawater [48].
The interactive e¡ects of salinity (sea salts, 3.7%), light
(sunlight, approx. 300 W m32 ) and the presence of organic
matter (100 mg l31 glucose) were studied by Troussellier et
al. [9]. Salinity was found to increase E. coli sensitivity
(CFU) only in the presence of light, regardless of the presence of organic matter. In the same study it was also
shown that an osmotic upshock combined with nutrient
deprivation inhibited several transmembrane transport
FEMSRE 725 5-12-01
518
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
systems within 60 min in di¡erent E. coli strains [9]. During a much longer incubation (30 days), Flatau et al. [50]
found di¡erent accumulation patterns for histidine (constant accumulation) and maltose (increased during the ¢rst
2 weeks and than slowly decreased).
Upon an osmotic upshift, bacterial cells accumulate or
synthesize speci¢c osmoprotectant molecules, in order to
equalize osmotic pressure and avoid drastic loss of water
from the cytoplasm [51]. The accumulation or synthesis of
such molecules (trehalose, glycine betaine, glutamic acid)
in Salmonella manhattan and S. typhimurium in estuarine
waters (3.5% salinity, 20³C) was detected by H-NMR. S.
manhattan accumulated trehalose, and an unidenti¢ed substance, whereas S. typhimurium accumulated mostly glycine betaine [9]. Trehalose synthesis was also observed for
wastewater-borne S. manhattan in oligotrophic seawater
[52].
Glutamate added to ¢lter-sterilized seawater was shown
to enhance E. coli MC4100 culturability. The e¡ect appeared to be logarithmically correlated to glutamate concentration ; glycine betaine uptake and its protective e¡ect
were both enhanced in the presence of glutamate [53].
The essential role of osmoregulatory mechanisms in enteric bacterial survival in seawater was demonstrated in
several studies. Munro et al. [47,54] and Gauthier et al.
[24] found that cells preadapted to high osmolarity are
highly resistant to seawater. This, however, depended
upon the preadaptation media. Among the induced systems that were shown to be of value in conferring a protective e¡ect for growing E. coli MC4100 were potassium
transport, glycine betaine synthesis or transport, trehalose
synthesis [47] and glutamate accumulation [55]. The protective advantage of glycine betaine in nutrient-free seawater seemed to vary according to di¡erent reports [47,56].
The resistance of E. coli grown in saline media (M9, 0.5
M NaCl) to seawater was suppressed by an osmotic downshock that resulted in a loss of several osmolytes previously accumulated [55]. An osmotic downshock (distilled
water or wastewater) increased loss of culturability
[55,57].
The survival of non-adapted E. coli cells was not affected by impaired K‡ transport or glycine betaine uptake,
while trehalose synthesis ability was apparently important
[47].
4.3. pH
Seawater pH normally ranges between 7.5 and 8.5, and
is in£uenced by temperature, pressure, and the photosynthetic and respiratory activities of microorganisms [58]. An
acidic pH (5) was found to be most favorable for E. coli
survival (in the 5.0^9.0 range) in both seawater and NaCl
solutions, and sensitivity increased with the increase in pH
[48]. The authors concluded that seawater pH, normally
around 8, contributes to the deleterious e¡ects on E. coli
survival.
4.4. Nutrient deprivation
It is a little recognized fact that in the presence of su¤cient nutrient levels, E. coli can grow in seawater almost as
well as it does in rich laboratory media. In fact, it has been
shown that in nutrient-enriched seawater E. coli competed
successfully with ¢ve marine bacterial isolates [59]. LopezTorres et al. [60], in their study of K. pneumoniae and E.
coli in membrane di¡usion chambers at di¡erent coastal
areas in Puerto Rico, showed increased survival of E. coli
and to a lesser extent of K. pneumoniae (respiring cells) in
a site receiving rum distillery e¥uents. The density of E.
coli declined immediately after the e¥uent discharge
stopped, suggesting that the organic load improved their
survival [60].
Regularly, however, the amounts of inorganic and organic nutrients in seawater are dramatically lower than
those of laboratory media or wastewaters, and bacteria
released into the marine environment have to contend
with starvation conditions. This was already shown by
Carlucci and Pramer [48], who demonstrated that E. coli
survival improved with increasing nutrient concentrations,
both organic and inorganic. The e¡ect of organics was
pronounced when peptone and sewage volatile solids
were added, but not glucose. The e¡ect of sewage volatile
solids was more pronounced in ¢lter-sterilized seawater,
indicating a possible competition for available nutrients
with indigenous marine micro£ora.
Similar results were obtained by Troussellier et al. [9],
who studied the combined e¡ects of salinity, light and
nutrient availability. When only nutrient deprivation (glucose) was imposed, a small energy charge decrease was
observed but CFU and transport abilities were maintained. Combined with the other stress factors, nutrient
deprivation led to the inactivation of membrane transport
and to a pronounced reduction in energy charge.
Our own results (Rozen and Belkin, unpublished) indicate that while the presence of nutrients actually allows E.
coli to grow in seawater, their absence does not necessarily
a¡ect the survival (colony formation) of non-growing cells.
In fact, cells which were induced to grow in nutrient-enriched seawater were more susceptible to various mutations (kdpABC trkA, kdpABC trkE (potassium transport),
nhaA (Na‡ /H‡ antiporter), hns (histone-like protein), sodA
sodB and rpoE (cE )) that had no e¡ect in unamended
seawater (see Table 1).
4.5. Temperature
Seawater temperature is another obvious shock confronted by microorganisms the optimal growth of which
occurs around 37³C [61]. However, the optimal temperature for survival is not necessarily the same as the one
for growth. Indeed, most reports indicate enhanced stability of E. coli at lower temperatures. This was shown by
Carlucci and Pramer [62] in a 48-h experiment in natural
FEMSRE 725 5-12-01
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
seawater (5^40³C), and by Vasoncelos and Swartz [63] in a
6-day experiment in di¡usion chambers exposed to temperatures ranging from 8.9³C to 14.5³C.
A similar tendency was reported by Lessard and Sieburth [38], who measured the viability (CFU) of natural
sewage populations of E. coli and enterococci in di¡usion
chambers from February to August, with experimental
temperatures ranging from 0³C to 20³C. Though other
environmental changes may have contributed to bacterial
survival, it was signi¢cantly higher at the lower temperatures.
Short-term experiments (6 h) measuring survival (TC,
DVC, CFU) of Salmonella montevideo in ¢ltered seawater
exposed to visible light (1.03^1.12 MJ m32 h31 ) at three
di¡erent temperatures (5³C, 15³C, 25³C), showed no
change in any of the parameters tested, and no di¡erences
among the three temperatures possibly due to the short
time span of the experiment [12].
4.6. Hydrostatic pressure
The release of wastewaters into deep waters was suggested as a solution for immediate public health hazards
caused by near o¡shore disposal. While the die-o¡ rate of
fecal microorganisms in o¡shore conditions was emphasized in many studies, their resistance and fate under deepsea conditions were studied to a much lesser extent.
The response of E. coli to elevated hydrostatic pressure
was described at both the whole cell physiology and proteins levels. A pressure upshift to 546 atm dramatically
inhibited E. coli W3110 growth measured as optical density, CFU or TC (epi£uorescence direct counts) [64]. A
linear decrease in the rate of protein synthesis was observed with increasing pressure up to 1092 atm, where it
essentially ceased. Increased pressure also caused the modulation of synthesis of speci¢c proteins. Two-dimensional
gels revealed the induction of 55 proteins; 11 of them were
shown to be heat-shock proteins and four were cold-shock
proteins [64].
Under simulated deep-sea conditions (seawater at low
temperature (4³C) and high hydrostatic pressure (up to
1000 atm) [65]) the survival of pure cultures of four human
enteric bacteria as well as a sewage bacterial population
was found to be species-dependent. Following a 12-day
incubation at di¡erent hydrostatic pressures, the survival
in some cases (Clostridium perfringens, Vibrio parahaemolyticus and an aerobic sewage bacterial population) was
negatively a¡ected by the pressure, while E. coli and S.
faecalis exhibited higher survival rates at 250 and 500 atm
than at 1 or 1000 atm [65].
It should be noted that the e¡ects reported for hydrostatic pressure (even at 1000 atm) were variable and apparently species-speci¢c. The dumping of raw or partially
treated sewage into deep waters, therefore, does not guarantee a reduction in the health risks inherent in wastewater-borne microorganisms.
519
4.7. Sedimentation
Several studies, focusing on the occurrence of coliforms,
fecal coliforms or enterococci in estuarine water and coastal sediments, revealed that greater numbers (10- or a 100fold higher) of these organisms are found in the sediments
than in the water above them [66,67]. This phenomenon
had been the impetus for detailed studies determining that
these organisms indeed survive better in sediments than in
the water [66,68], and for studies attempting to reveal the
factors involved in the protective e¡ect of the sediment
[66,67,69,70].
A longer survival time of E. coli in sediments than in
seawater was demonstrated in several cases, and was attributed to the higher organic matter content of the sediment [66,68^70]. One of the compounds shown to be
present was glycine betaine [71]; indeed, marine E. coli
isolates were shown to accumulate glycine betaine from
autoclaved estuarine sediments when mixed with a minimal medium, a phenomenon that was not observed without the sediment [70]. Based on these results, a study trying to assess the protective e¡ect of glycine betaine on E.
coli survival in three media (seawater, low- and high-organic carbon sediments) was conducted [56]. In all three
media, the addition of glycine betaine had a protective
e¡ect on the survival of E. coli MC4100 (CFU), and an
uptake of radioactively labeled glycine betaine could be
detected. The e¡ect of glycine betaine on E. coli survival
was clearly related to nutrient availability.
4.8. Previous growth history and stress cross-protection
Prior to their arrival into the marine environment,
wastewater bacteria may be exposed to a variety of environmental conditions. In some cases they are discharged
directly from boats or bathers, in others they remain for
di¡erent periods in wastewater reservoirs, and/or are carried out to the sea through natural rivers or arti¢cial conduits. In several studies it was found that the survival
ability of enteric bacteria depends on their history preceding the seawater shock [9,18,24,42,47,54,55,72^76].
Garcia-Lara et al. [18] have shown that preadaptation
by growth in an arti¢cial seawater medium supplemented
with complex organics (tryptone, peptone and yeast extract) endowed the cells with a much higher viability in
seawater. Munro et al. [54] demonstrated the same phenomenon using four enteric bacteria (Shigella dysenteriae,
E. coli, K. pneumoniae and S. typhimurium). For four
other strains, typically found in urban wastewater (Staphylococcus aureus, S. faecalis, Pseudomonas aeruginosa and
Aeromonas hydrophila), such pretreatment had no e¡ect
[54].
One of the dominant factors in£uencing seawater survival is the growth phase during which the transfer to
seawater occurred [9,76]. Aerobically grown E. coli
MC4100 cells exhibited high sensitivity to seawater (deter-
FEMSRE 725 5-12-01
520
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
mined after 2 and 6 days of incubation) when exposed
during the lag phase, lower sensitivity in the exponential
phase and a minimal one during the stationary phase [76].
In the light, the di¡erence between exponential and stationary phase cells was much less pronounced [9]. For
anaerobically grown cells, a di¡erent pattern was observed
after 1 or 2 days of exposure, but not after 6 days [76].
In general, cells grown under anaerobic conditions were
found to be much more sensitive than aerobically grown
cells [73,76], but this e¡ect was not observed when TC was
the parameter measured [73]. This correlates well with the
possible involvement of reactive oxygen species in the
stress imposed by seawater, and hence with the signi¢cance of the adaptation to oxidative conditions.
In order to determine the nature of the protection acquired in the preadaptation process, E. coli MC4100 cells
were exposed to ¢ve di¡erent types of stress for up to 4 h:
thermal (48³C), oxidative (H2 O2 , 1 mM), acid (pH 5.1),
osmotic (0.5 M NaCl) and starvation (a nutrient-de¢cient
bu¡er). The cells were then incubated in seawater (22^
25³C) for 6 days. All pretreatments led to a better resistance to seawater compared to non-treated cells [42]. In a
di¡erent report, a 44³C pretreatment was found to decrease such resistance (CFU), as did growth at pH 6 or
8 compared to pH 7 [73].
Munro et al. [42] concluded that the relative adaptive
importance of the various environmental conditions may
be ranked as follows : osmotic s starvation s oxidative
s acidic s thermal. Apart from osmotic stress, all others
induced higher resistance to seawater in an rpoS‡ strain
than in the rpoS mutant, suggesting that the cross-protection endowed by oxidative, acidic, thermal and nutritional
stresses was rpoS-dependent [42].
There was an increased resistance to seawater of cells
preadapted to high osmolarity, by 0.5 M NaCl [42,47,73]
or by 0.8 M saccharose [47]. The preacquired capabilities
for K‡ transport, glycine betaine synthesis or transport,
and trehalose synthesis, helped to increase the ability of E.
coli to survive in seawater [47]. A negative e¡ect on resistance was reported to be exerted by a hypoosmotic shock
prior to seawater exposure [55,57].
Another pretreatment variable tested was medium composition. Gauthier et al. [73] reported signi¢cant di¡erences between media only after more than 7 days, but with
no correlation to the richness or complexity of the medium. An additional adaptive parameter tested by Gauthier et al. [75] was phosphate concentration : low phosphate
grown cells, in which alkaline phosphatase synthesis was
induced, were shown to remain culturable after longer
periods in seawater.
5. Biotic factors
Although mostly very poor in nutrient concentrations,
the marine environment is nevertheless inhabited by rela-
tively diverse biological populations that may a¡ect the
fate of enteric bacteria entering the system. Indeed, the
observations that the numbers of enteric bacteria decreased in natural seawater more than in sterile seawater
[24,77] certainly suggest the involvement of biological processes. Support for this hypothesis was provided by Le
Guyader et al. [78], who monitored the viability (CFU)
of an E. coli culture for 13 days in seawater and sediment.
In the presence of seawater indigenous £ora, E. coli concentrations decreased earlier and at a faster rate compared
to sterile conditions.
Among the major biological mechanisms that have been
implicated in a¡ecting enteric bacterial concentrations in
seawater are predation [7,77,79^85], competition
[59,78,79,86] and bacteriophages [82,87].
5.1. Grazing and competition
Various studies indicate that the main predators of bacteria in the marine environment are protozoa [7,77,84,85].
Using ¢lters with di¡erent pore sizes and antibiotics to
suppress indigenous bacterial activity, it was shown that
bacterial competition, antagonism, and even bacterial predation were relatively unimportant in coliform removal. It
was also shown that the reduction in E. coli populations
paralleled an increase in the number of protozoa [84]. In
the same vein, the survival of fecal coliforms was signi¢cantly improved in the presence of cyclohexamide, a compound expected to inhibit eukaryotic organisms. Interestingly, Clostridium perfringens and fecal streptococci were
not a¡ected by the addition of this inhibitor, which could
be explained either by di¡erential resistance of protozoan
species to cyclohexamide or by selective predation on the
di¡erent bacterial species [7].
Another demonstration of the role protozoan predation
plays in enteric bacterial disappearance from marine
waters was provided by Gonzalez et al. [77], who enumerated E. coli and E. faecalis for 5 days in natural and
¢ltered (0.2 Wm) seawater. Both CFU and direct counts
decreased signi¢cantly (3^5 logs) in the presence of natural
microbiota, an e¡ect that was attributed to protist activity.
Mitchell et al. [81] demonstrated that seawater samples
plated on E. coli lawns on solid agar resulted in bacterial
colonies surrounded by clearing zones; these were attributed to bacteria utilizing the cell wall of E. coli as a carbon
source. Also observed were plaques, possibly caused by
obligate parasites such as Bdellovibrio.
Mitchell and Nevo [80] isolated from seawater a Pseudomonas sp. capable of utilizing Flavobacterium capsular
polysaccharide as its sole carbon source. They also isolated bacteria capable of growth on capsules of Azotobacter, Rhizobium, Arthrobacter, and the cell wall of E.
coli. Activity of the isolated Pseudomonas against living
cells of Arthrobacter and E. coli was tested by joint growth
in arti¢cial seawater with 0.1% peptone. The authors concluded that growth of both `terrestrial' bacteria was mark-
FEMSRE 725 5-12-01
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
edly suppressed by the marine bacterium, though this was
based on optical density rather than on a more accurate
enumerating method.
The role of competition with marine bacteria has been
investigated in several studies [59,81]. Experiments measuring E. coli survival in autoclaved seawater with di¡erent sizes of marine microbial inocula showed a clear e¡ect
on E. coli CFU counts. Within 7 days, the presence of 107
marine bacteria ml31 decreased the E. coli population by
about 4 logs, while in autoclaved seawater almost no decrease was detected [81]. Furthermore, when a mixed marine bacteria/E. coli culture was reinoculated with E. coli
after 7 days, its elimination from the medium proceeded
rapidly with no lag phase. The enhanced removal rates
suggested to the authors the existence of micro£ora parasitic or lytic to E. coli [81].
Another approach that was taken in order to assess the
role of competitive elimination of Enterobacteriaceae from
seawater was based on selecting and isolating organisms
de¢ned as successful competitors under various dilution
rates or di¡erent limiting nutrient concentrations in chemostat experiments [59]. E. coli was found to be a successful competitor in rich media but a very poor one under the
low nutrient concentrations characterizing natural seawater [59].
5.2. Bacteriophages
Many studies detected coliphages in marine waters subject to sewage contamination [35,82,87^90]. It was also
shown that phages detected in seawater were active against
E. coli, Aerobacter aerogenes, and Serratia marinorubra
[87]. The presence of coliphages was investigated in relation to their usage as indicators for fecal contamination
[90^92], and in many cases a positive correlation between
fecal phages, enteric viruses, and other pathogens has been
recorded [88,91^94]. Nevertheless, the presence of enteric
bacterial infectious phages does not necessarily indicate
their actual activity in removing coliforms from marine
water.
Carlucci and Pramer [87] have shown that bacteriophages were e¡ective in reducing E. coli population sizes
only under nutrient-rich conditions, suggesting a very minor role for bacteriophages, if any, under natural conditions. This was corroborated by Penon et al. [95], who
found no signi¢cant deleterious e¡ect of the seawater fraction passing 0.2-Wm ¢lters (containing bacteriophages but
not bacteria or protozoa) on the survival of fecal bacteria
in seawater. The 0.2^2-Wm fraction, in contrast, had a
signi¢cant e¡ect.
5.3. Antibiotics and toxins
Several studies from the 40s through the 60s suggested
that the deleterious e¡ects of seawater on enteric bacteria
that could not be explained by other factors known at the
521
time, were caused by toxic substances believed to be antibiotics produced by microorganisms [79,96^98]. Similarly,
there was also a possible indication of the negative e¡ects
of an algal toxin [87].
All reported attempts to ¢nd an antibiotic e¡ect of seawater on enteric bacteria yielded no evidence that such
compounds are produced under natural conditions and
contribute to the elimination of their targets from seawater.
6. Molecular mechanisms involved in seawater survival
Many of the studies quoted to this point share a similar
characteristic : colony formation, and to a lesser extent
other viability criteria, serve as a practically sole parameter in determining the factors controlling the sensitivity or
survival of E. coli and other enteric bacteria in seawater. It
appears as if a substantial gap exists between these reports
and the enormous wealth of information presently available on bacterial physiology, biochemistry and molecular
biology. This gap is especially pronounced in the latter
category ; many of the molecular responses of E. coli
and Salmonella to the environmental parameters mentioned above (salinity, starvation, pH etc.) were studied
in great detail, but very few attempts were reported to
make use of these data in respect to bacterial survival in
seawater. These reports, most of them from the last decade, will be summarized below, along with a few recent
¢ndings from our own laboratory.
6.1. E¡ect of speci¢c mutations
Past attention to the molecular mechanisms determining
survival of enteric bacteria such as E. coli in arti¢cial seawater or sterile seawater in the dark revealed very few
genes crucial to seawater survival. Only six mutations
were shown to have a signi¢cant e¡ect. The most dominant among them was in rpoS [9,39,41^43]. The other
genes in which mutations were reported to be signi¢cant
were otsA [47], relA, spoT (but only on top of a relA
deletion [43]) and one or both membrane porins ompC
and ompF [99].
As already noted, the most dominant gene for seawater
survival of the above list was rpoS [9,39,41^43]. An rpoS
mutation was shown to decrease seawater survival (CFU)
of stationary phase E. coli cells by approx. 3 logs over
8 days [43]. The RpoS (cs ) transcription factor controls
the expression of a large number of genes involved in
cellular responses to a diverse number of stresses, including starvation, osmotic stress, acid shock, cold shock, heat
shock, oxidative damage, and transition to stationary
phase. A list of over 50 genes under the control of rpoS
has been compiled [101]. It normally acts as a positive
regulator, but some genes are under its negative control.
The synthesis and accumulation of cs were shown to be
FEMSRE 725 5-12-01
522
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
Table 1
E¡ect of speci¢c mutations on E. coli seawater survival and on growth in nutrient-supplemented seawater
Strain
Relevant mutation
Survival
in the darka
FRAG-5 [107]
TK110 [107]
TK121 [107]
TK133 [107]
TK142 [107]
TK2205 [47]
TK401 [107]
TK2240 [47]
PLB3260 [108]
PLB3261 [108]
SG477 [109]
SBM26 [99]
SG480 [109]
UE14 [110]
RK33Z [111]
TA15D2 [112]
JJ2 [113]
FF4169 [104]
RO22 [105]
EF047 [114]
RH90 [115]
QC2410 [116]
UM122 [117]
CA8306 [118]
CA8445 [119]
CA8445-1 [119]
CF1652 [120]
CF1693 [120]
QC2408 [116]
QC2445 [121]
UM120 [117]
QC2476 [116]
QC2463 [116]
QC2473 [122]
QC2474 [122]
QC772 [123]
QC773 [123]
QC774 [123]
QC2472 [116]
QC2589 [122]
QC2413 [116]
SASX41B [39]
HB101 [39]
DM803 [124]
QC227 [39]
BW295 [39]
CAG9333 [125]
AB1899 [126]
CAG22188 [127]
TN3151 [128]
CLG302 [75]
W2961 [129]
K10 [130]
C101a [130]
C75a [131]
C75b [132]
C112a [132]
kdpABC
kdpABC, kefB (trkB)
kdpABC, kefC (trkC)
kdpABC, trkA
kdpABC, trkE
kdpABC, trkA, trkD
kdpABC, trkA, trkD
trkA, trkD
envZ (ompF)
envZ (ompC)
envZ (ompC)
envZ (ompF, ompC)
envZ (ompF, ompC)
treA
nhaA
hns
osmB
otsA
otsAB
proP, proU
rpoS
rpoS
rpoS
cyaA
cyaA, crp, rpsL
crp, rpsL
relA
relA, spoT
dps
fur
katE
katE3 , katG
katE, katG, dps
soxR
soxS
sodA
sodB
sodA, sodB
sodA, sodB
sodA, sodB, zwf
oxyR
hemA
recA
lexA (SOS repressed)
fpg, uvrA
xthA
rpoH (groE constitutive)
lon
rpoE
uspA
phoA
phoE
pit
pit, pstA
pit, phoR
pit, phoS
pit, phoT
3 [47], [Rozen]
3
3
3
3
3 [47]
3 [Rozen]
3 [47]
+ [99]
+ [99]
3 [Rozen]
+ [99]
3 [Rozen]
3
3
3
Protective e¡ect [47]
+ [47]
Survival
in the lightb
Growthc
3d [47]
3
3
3
+
+
+d [47]
+d [47]
3
3
+
3
+
+
+d [47]
3 [41]
3 [47]
+ [42,43]
+ [Rozen]
Survival in the dark
after preadaptation
+ [9,41]
+e [47]
+f [42]
+
+ [9,41]
3
3
3
+ [43]
+ [43]
3
3
3
3
3
3
3
3
3
+ [9,41]
3
3
3 [39]
+ [41]
+ [41]
3
3
3
3
3 [9,39]
3 [9,39]
+ [9,39]
3
3
+
+
+ [41]
Protective e¡ect [39]
3 [9,39]
3
3
3 [9,39]
3 [9,39]
3
3
3
3
3 [75]
+, signi¢cant e¡ect of the mutation; 3, no e¡ect ; no sign, mutation not tested.
FEMSRE 725 5-12-01
+g [75]
+g [100]
3g [100]
+g [100]
+g [100]
+g [100]
+g [100]
3
3
+
3
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
523
a
Survival experiments referred to as [Rozen] or with no reference noted, are reported here for the ¢rst time : stationary phase cells (24 h growth in a
MOPS^glucose medium [133], 200 rpm, 37³C) were washed three times and resuspended in ¢lter-sterilized (0.22 Wm) seawater for 7 days, 200 rpm,
26³C. Experiments were repeated at least three times, and the mutation was noted as + when the wild-type strain survived signi¢cantly better than the
mutated strain. In this study, `signi¢cantly better survival' was de¢ned as at least a 10-fold di¡erence in viable counts (CFU) over a 7-day period. Munro et al. [43]: stationary phase cells were used. In the other experiments [47,75,99] exponential phase cells were used.
b
Experiments were in ¢lter-sterilized arti¢cial seawater (Instant Ocean), 20³C, 180 rpm, arti¢cial visible light mainly s 400 nm, (0.9 mmol photons m32
s31 ) [9,39,41].
c
All growth experiments were done in our laboratory; overnight cultures were diluted 100-fold in ¢ltered (0.22 Wm) LB [134] and seawater-based LB.
Growth (37³C, 200 rpm) was followed by culture turbidity for 9 h. Experiments were repeated at least three times, and the mutation was noted as +
when the wild-type strain grew signi¢cantly better (relative growth yield at least 2-fold higher) than the mutated strain. Relative growth yield was de¢ned as the ratio of the yield in seawater-based LB to that in LB.
d
Previously grown in minimal medium supplemented with NaCl [47].
e
Previously grown in minimal medium supplemented with glycine betaine [47].
f
Previously exposed for 1 h at 37³C to oxidative (H2 O2 , 1 mM), acid (pH 5.1), osmotic (0.5 M NaCl), starvation (a nutrient-de¢cient bu¡er) or thermal
stress (48³C) [42].
g
Previously grown in a phosphate-limited minimal medium (0.1 mM K2 HPO4 [75] or 0.1 mM Na2 HPO4 [100]).
6
controlled at the transcription, translation, holoenzyme
complex formation and proteolysis levels. Transcriptional
control of rpoS involves guanosine 3P,5P-bispyrophosphate
(ppGpp) and polyphosphate as positive regulators, and
the cAMP receptor protein^cAMP complex (CRP^
cAMP) as a negative one [101].
Munro et al. [43] have demonstrated that in addition to
rpoS, a relA deletion and the double relA spoT mutation
also a¡ected seawater survival of E. coli. The absence of
these two genes from stationary phase E. coli was shown
to cause an approx. 4 log decrease in E. coli colony formation ability over 8 days. RelA and SpoT are enzymes,
ppGpp synthetase I and II respectively, that are responsible for the synthesis of ppGpp [102]. Among the multiple
e¡ects shown to be caused by the increase in intracellular
ppGpp concentration [102] is an enhancement of rpoS expression [103]. The hypothesis that the intracellular level
of ppGpp could a¡ect seawater survival due to lower rpoS
expression was suggested [43], but was not experimentally
tested.
An additional gene shown to be important in determining E. coli's sensitivity to seawater was otsA [47]; within
an 8-day experiment, colony formation ability of the mutant (E. coli FF4169) decreased 10^100-fold compared to
the wild-type. The otsA gene codes for trehalose-6-phosphate synthase, an enzyme that catalyzes the synthesis of
the osmoprotectant trehalose; it is transcriptionally activated both by osmotic stress [104] and by growth into the
stationary phase [105]. Not surprisingly, otsA is RpoScontrolled [105,106], and it is not unlikely that at least
part of the deleterious e¡ect of the rpoS mutation on E.
coli seawater survival is due to the absence of trehalose.
Mutants lacking OmpC, OmpF or both (E. coli strain
PLB3261, PLB3260 and SBM26, respectively) also seemed
to be impaired in their seawater survival. The relative signi¢cance of the two porins depended upon pregrowth conditions: ompF was more dominant following exponential
growth in low osmolarity, while ompC appeared to be
more important for survival following exponential growth
in high osmolarity [99]. It should be noted that standard
deviations in these experiments were large, and not all
di¡erences appeared signi¢cant. In a di¡erent experiment,
other E. coli strains lacking OmpC or both OmpC and
OmpF (strain SG477 and SG480, respectively) were pregrown to stationary phase in minimal medium. No e¡ect
on survival of either mutation was observed during a
7-day seawater exposure (Table 1).
Another group of genes shown to be important for seawater survival when previously grown in a low phosphate
(0.1 mM Na2 HPO4 ) medium is connected with phosphate
regulation. Mutation in the porin phoE, in the regulatory
gene phoR, and in genes related to the high-a¤nity phosphate-speci¢c transport system pstA, phoS and phoT, decreased survival 10^1500-fold after 6 days in ¢ltered sterilized seawater [100]. Furthermore, E. coli CLG 302, totally
deprived of alkaline phosphatase activity due to a phoA
mutation, was found to have increased seawater sensitivity
when previously grown in a low phosphate (0.1 mM
K2 HPO4 ) medium [75].
A few other genes were also implicated in the ability of
E. coli to form colonies on solid media, but only in conjunction with additional stress factors or following speci¢c
growth conditions. Numerous genes involved in protection
against reactive oxygen species (rpoS, katE katG, dps,
sodA sodB and oxyR) proved to be important when the
cells were exposed to visible light [39,41] (see Section 4.1).
The absence of two genes encoding for glycine betaine
transport, proP and proU (E. coli EF047), was found to
increase E. coli MC4100 seawater sensitivity only when
previously exposed for 1 h to high osmolarity and glycine
betaine (0.1 mM) [47].
Similarly, mutations in two potassium transport systems, kdpABC (osmotically inducible high-a¤nity transport, E. coli TK2240) and trkA (constitutive low-a¤nity
transport, E. coli TK2205) were found to have a deleterious e¡ect on E. coli (MC4100) survival in seawater only
when previously grown in a minimal medium at high osmolarity (0.5 M NaCl) [47].
The absence of one gene, induced by high osmolarity,
but with an unknown function ^ osmB (strain E. coli JJ2)
FEMSRE 725 5-12-01
524
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
^ was found to have a surprisingly protective e¡ect on
seawater survival compared to its parental strain [47].
No explanation for this phenomenon was provided. The
same e¡ect was observed for E. coli strain SASX41B, mutated in hemA, when incubated in seawater exposed to
visible light. In this case, a decrease in intracellular photosensitizer concentrations was suggested [39].
Table 1 summarizes previously published data concerning the e¡ect of speci¢c gene mutations on colony formation ability of E. coli. It also includes results of further
gene screening in our laboratory, in which additional gene
mutations were tested for their e¡ect; the survival of none
of these was signi¢cantly di¡erent from that of its respective wild-type.
Another potential approach to discovering genes the
products of which may be important is not only to expose
the cells to seawater, but also to force them to attempt to
grow under these conditions. This can be carried out by
amending the seawater with su¤cient nutrients. Table 1
also includes the results of such growth experiments carried out under nutrient-rich conditions for all the genes
listed, each mutant compared to its own wild-type parental
strain. An e¡ect was noted when the deleterious in£uence
of the mutation on growth in seawater-based Luria^Bertani broth (LB) was signi¢cantly higher than its e¡ect on
growth in regular LB. Out of 34 mutations screened, there
were several that did not a¡ect survival in unamended
seawater but had a marked e¡ect either on the growth
rate or the duration of the lag phase that preceded it.
Among those were strains with mutations in genes involved in osmoregulatory processes (kdpABC trkA,
kdpABC trkE, ompF ompC), defenses against oxidative
stress (sodA sodB), the c factor rpoE, the gene encoding
the histone-like protein H-NS (hns), and the sodium proton antiporter (nhaA). The e¡ect of the latter mutation
was the most dramatic: no growth at all was observed
in the mutant. The mutation in rpoS had the expected
e¡ect in both incubation conditions.
6.2. Genes induced upon seawater exposure
Another possible strategy in the search for genes playing
a role in seawater survival is to look for those activated in
response to seawater exposure. Such a study was conducted by Rozen et al. [139] who screened a 687-member
E. coli promoter : :luxCDABE fusion library [135]. Twentytwo promoters were found to be activated in seawater
amended by su¤cient nutrients to allow bioluminescent
expression but not growth. The luminescence driven by
these promoters (osmY, ldcC, treA, otsA, poxB, yciG,
ybhP, yohF, yohC, ybdK, yeaG, yhdW, prpR, ytrF, ykgC,
phrB, yjhT, ykgD, cysI, pstS, ygjL, yjbG) increased at least
2-fold ; only nine of the corresponding genes have previously been assigned a function. The most prominent characteristic of the induced genes was that most of them (17
out of 22) were under rpoS positive control (osmY, ldcC,
treA, otsA, poxB, yciG, ybhP, yohF, yohC, ybdK, yeaG,
yhdW, prpR, ytrF, ykgC, phrB, yjhT) and one was under
rpoS negative control (yjbG). The most signi¢cant environmental factors responsible for inducing the majority of
these promoters appeared to be nutrient limitation. Salinity or osmolarity were instrumental in only four cases
(osmY, yciG, ybhP and yjbG), and in three promoters
(cysI, ygjL and yjbG) increased pH also appeared to
play a role [139].
To our knowledge, no data were published regarding de
novo synthesis of proteins in response to seawater incubation, though there are clear evidences that low-level protein synthesis occurs under such conditions for at least
3 weeks (Rozen and Belkin, unpublished).
7. Concluding remarks
A review of the scienti¢c literature on the survival of
enteric bacteria in the marine environment reveals a set of
attempts, many of them highly empirical in nature, to
describe the actual fate of wastewater or enteric bacteria
upon exposure to seawater. Only in recent years have
these been expanded to include also some molecular aspects of the studied phenomena. In this review we attempted to clarify the extent of our knowledge in the ¢eld,
and derive from it a de¢nition of desirable future research
directions.
While it is generally accepted that one of the earlier
phenomena observed in enteric bacteria exposed to seawater is the loss of the ability to form colonies on solid
media, there is a controversy in regard to the physiological
state of the nonculturable cells. Bogosian et al. [16,17]
claim that VBNC cells are either dead or of no signi¢cance
as they cannot be resuscitated if experimental procedures
are carefully carried out. In contrast, other evidence suggests that they are not only viable but that the pathogenic
among them may still be infective [6]. Such an observation
has a major signi¢cance in view of the worldwide practice
of releasing non-disinfected wastewaters into the sea and
its potential public health consequences. The seawater-related VBNC controversy is a part of the broader issue
[136^138] of the true meaning of di¡erent viability criteria
and of the molecular and biochemical mechanisms controlling the shift from colony forming to VBNC and vice
versa. These regulatory systems are only partially understood, and probably vary according to the stimuli imposed
on the cells.
When enteric bacteria are exposed to seawater they are
simultaneously challenged by a combination of stress factors, including pH, temperature, salinity, nutrient availability, light radiation and its associated oxidative stress.
It is claimed [30^33] that the latter two may probably be
the most signi¢cant components of this hostile combination, though relevant only for shallow waters of coastal
areas. Salinity in itself appears to be less signi¢cant; in-
FEMSRE 725 5-12-01
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
deed, when supplied with su¤cient organic nutrients, E.
coli can grow in seawater almost as well as it does in rich
laboratory media and out-compete marine strains [59].
Generally, lower pH values and temperatures contribute
to increased survival, as does implementation into marine
sediments. Increased hydrostatic pressure, relevant in cases
of deep-sea discharge, was shown to have a variable e¡ect.
Among possible biotic factors that may play a role in
determining survival, predation by protozoa was the
only one shown to be potentially signi¢cant. The contribution of other e¡ectors, such as bacterial predation, viral
infections or unexplained antibiotic-like e¡ects, was either
insigni¢cant or not demonstrated under natural conditions.
A recurring point arising from the data collated is that
in addition to actual seawater incubation conditions, previous growth history also has a major in£uence on subsequent survival in the hostile marine environment. This
clearly indicates the existence of an adaptation potential:
at least some of the risks inherent in seawater exposure
can be handled better if the cells can ¢rst mobilize the
necessary defense circuits.
The molecular data available to date clearly point at
possibly the most signi¢cant among such adaptive systems : the rpoS regulon. At least 50 di¡erent genes were
previously shown to be under rpoS control; they are induced upon a shift to a stationary growth phase, as well as
by diverse stresses including salinity and starvation. Not
all rpoS-controlled genes were assigned a clear function,
but among those that were, many were involved in combating the e¡ects of such stresses. The dominance of this
regulatory circuit was observed both in the signi¢cant negative e¡ect that rpoS mutations had on E. coli survival
[42,43] and in the observation that rpoS-dominated genes
accounted for 18 out of 22 shown to be induced by seawater exposure [139].
The impetus for many of the studies quoted in this review was the desire to protect near-shore areas from fecal
pollution and its ultimate e¡ects on human health. In the
design of marine sewage e¥uent outlets, engineers often
make use of bacterial (mostly E. coli) `die o¡' constants.
Such values may be based on both laboratory and ¢eld
experiments, and it is of importance to recognize the limitation of both approaches. The signi¢cance of many of
the ¢eld experiments is often site-speci¢c, and they tend to
ignore previous growth history of the monitored strains.
Thus, mathematical models based on such results cannot
replace the need for routine bacterial pollution monitoring. Conversely, laboratory experiments can be accurately
designed to test the e¡ects of speci¢c parameters, but cannot simulate or imitate the complexity of the real marine
environment. Nevertheless, a true understanding of responses of enteric bacteria to the imposed conditions
and their molecular controls can only be obtained in the
laboratory.
Another front in which basic molecular studies will be
525
able to markedly a¡ect wastewater pollution monitoring
practices are the detection methodologies of either pathogenic or indicator microorganisms. Present approaches are
based upon a few simple though highly informative indicator assays. In recent years, several molecular approaches
have been proposed that allow fast and sensitive detection
of bacteria, viruses and other organisms. Such methods
may lead to a more accurate and rapid wastewater pollution detection, including bacteria in the VBNC state.
These methods will provide a broader understanding of
the survival of these organisms, thus meeting the challenge
of future health management of marine waters.
In order to continue to unravel the dominant mechanisms in seawater survival of enteric bacteria, future research will have to combine traditional ¢eld and laboratory viability experiments with the molecular insight
allowed by genetic approaches. In addition to enhancing
our understanding of the basic mechanisms involved in the
defenses against multiple stress conditions, such studies
will surely also allow a more knowledgeable forecast of
bacterial decay rates in the sea.
One of the directions these studies should take in order
to be e¡ective will be towards a broadening of the spectrum of the studied responses. This should include a
screening of all E. coli genes, as is now possible with modern DNA array technologies, and be expanded to comprehensive expression assays, speci¢c mutation e¡ects and
protein identi¢cation. The speci¢c roles played by individual genes and proteins identi¢ed in this manner should be
studied in greater detail. In parallel, additional enteric
bacteria should be studied as well, with a greater focus
on infectivity bioassays in addition to the various viability
parameters. In this context, the signi¢cance of the `viable
but nonculturable' issue in seawater calls for a clearer resolution than it has been awarded to date.
Acknowledgements
The use of many E. coli strains from the collection of
T.K. Van Dyk and R.A. LaRossa (Dupont, Wilmington,
DE, USA), S. Dukan (CNRS/INSU UMR, Marseille,
France) and M. Berlyn at CGSC (E. coli Genetic Stock
Center, Yale University, New Haven, CT, USA) is gratefully acknowledged. M. Gouremelon is warmly thanked
for providing her PhD thesis. This study was supported
by the Israel Science foundation Grant No. 113-00-1 and
by the Israeli Ministry of Science and the Arts. Y.R. is a
Rieger-JNF fellow.
References
[1] Carlucci, A.F. and Pramer, D. (1959) Microbiological process report.
Factors a¡ecting the survival of bacteria in sea water. Appl. Microbiol. 7, 388^392.
[2] Grimes, D.J., Atwell, R.W., Brayton, P.R., Palmer, L.M., Rollins,
FEMSRE 725 5-12-01
526
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
D.M., Roszak, D.B., Singleton, F.L., Tamplin, M.L. and Colwell,
R.R. (1986) The fate of enteric pathogenic bacteria in estuarine
and marine environments. Microbiol. Sci. 3, 324^329.
Roszak, D.B. and Colwell, R.R. (1987) Survival strategies of bacteria
in natural environment. Microbiol. Rev. 51, 365^379.
Barcina, I., Lebaron, P. and Vives-Rego, J. (1997) Survival of allochthonous bacteria in aquatic systems: a biological approach.
FEMS Microbiol. Ecol. 23, 1^9.
Joux, F., Lebaron, P. and Troussellier, M. (1997) Succession of cellular states in a Salmonella typhimurium population during starvation
in arti¢cial seawater microcosms. FEMS Microbiol. Ecol. 22, 65^76.
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. Appl. Environ. Microbiol. 62, 4621^4626.
Davies, C.M., Long, J.A., Donald, M. and Ashbolt, N.J. (1995)
Survival of fecal microorganisms in marine and freshwater sediments.
Appl. Environ. Microbiol. 61, 1888^1896.
Caro, A., Got, P., Lesne, J., Binard, S. and Baleux, B. (1999) Viability and virulence of experimentally stressed nonculturable Salmonella typhimurium. Appl. Environ. Microbiol. 67, 3229^3232.
Troussellier, M. et al. (1998) Responses of enteric bacteria to environmental stresses in seawater. Oceanol. Acta 21, 965^981.
Grimes, D.J. and Colwell, R.R. (1986) Viability and virulence of
Escherichia coli suspended by membrane chamber in semitropical
ocean water. FEMS Microbiol. Lett. 34, 161^165.
Gourmelon, M., Cillard, J. and Pommepuy, M. (1994) Visible light
damage to Escherichia coli in sea water: oxidative stress hypothesis.
J. Appl. Bacteriol. 77, 105^112.
Davies, C.M. and Evison, L.M. (1991) Sunlight and the survival of
enteric bacteria in natural waters. J. Appl. Bacteriol. 70, 265^274.
Gauthier, M.J. (2000) Environmental parameters associated with the
viable but nonculturable state. In: Nonculturable Microorganisms in
the Environment (Colwell, R.R. and Grimes, D.J., Eds.), pp. 87^112.
ASM Press, Washington, DC.
Roth, W.G., Leckie, M.P. and Dietzler, D.N. (1988) Restoration of
colony forming activity in osmotically stressed Escherichia coli by
betaine. Appl. Environ. Microbiol. 54, 3142^3146.
Roszak, D.B., Grimes, D.J. and Colwell, R.R. (1984) Viable but
nonrecoverable stage of Salmonella enteritidis in aquatic systems.
Can. J. Microbiol. 30, 334^338.
Bogosian, G., Sammons, L.E., Morris, P.J., O'Neil, J.P., Heitkamp,
M.A. and Weber, D.B. (1996) Death of the Escherichia coli K-12
strain W3110 in soil and water. Appl. Environ. Microbiol. 62,
4114^4120.
Bogosian, G., Morris, P.J.L. and O'Neil, J.P. (1998) A mix culture
recovery method indicates that enteric bacteria do not enter the viable but nonculturable state. Appl. Environ. Microbiol. 64, 1736^1742.
Garcia-Lara, J., Martinez, J., Vilamu, M. and Vives-Rego, J. (1993)
E¡ect of previous growth conditions on the starvation-survival of
Escherichia coli in seawater. J. Gen. Microbiol. 139, 1425^1431.
Fiksdal, L., Pommepuy, M., Derrien, A. and Cormier, M. (1989)
Production of 4-methylumbelliferyl heptanoate hydrolase by Escherichia coli exposed to seawater. Appl. Environ. Microbiol. 55, 2424^
2427.
Galdiereo, E., Donnarumma, G., de-Martino, L., Marcatili, A., del'Ero, G.C. and Merone, A. (1994) E¡ect of low-nutrient seawater on
morphology, chemical composition, and virulence of Salmonella
typhimurium. Arch. Microbiol. 162, 41^47.
Smith, J.J., Howington, J.P. and McFeters, G.A. (1994) Survival,
physiological response, and recovery on enteric bacteria exposed to
polar marine environment. Appl. Environ. Microbiol. 60, 2977^
2984.
Morinigo, M.A., Cornax, R., Castro, D., Martinez-Manzanares, E.
and Borrego, J.J. (1990) Viability of Salmonella spp and indicator
microorganisms in seawater using membrane di¡usion chambers. Antonie Van Leeuwenhoek 57, 109^117.
[23] Dawe, L.L. and Penrose, W.E. (1978) `Bactericidal' property of seawater: death or debilitation? Appl. Environ. Microbiol. 35, 829^833.
[24] Gauthier, M.J., Munro, P.M. and Mohajer, S. (1987) In£uence of
salts and sodium chloride on the recovery of Escherichia coli from
seawater. Curr. Microbiol. 15, 5^10.
[25] Apte, S.C. and Batley, G.E. (1994) Rapid detection of sewage contamination in marine waters using £uorimetric assay of L-D-galactosidase activity. Sci. Total Environ. 141, 175^180.
[26] Fiksdal, L., Pommepuy, M., Caprais, M.P. and Midttun, I. (1994)
Monitoring of fecal pollution in coastal waters by use of rapid enzymatic techniques. Appl. Environ. Microbiol. 60, 1581^1584.
[27] Davies, C.M., Apte, S.C. and Peterson, S.M. (1995) L-D-Galactosidase activity of viable, non-culturable coliform bacteria in marine
waters. Lett. Appl. Microbiol. 21, 99^102.
[28] Pommepuy, M., Fiksdal, L., Gourmelon, M., Melikechi, H., Caprais,
M.P., Cormier, M. and Colwell, R.R. (1996) E¡ect of seawater on
Escherichia coli L-galactosidase activity. J. Appl. Bacteriol. 81, 174^
180.
[29] Byrd, J.J. and Colwell, R.R. (1990) Maintenance of plasmid pBR322
and pUC8 in Escherichia coli nonculturable in marine environment.
Appl. Environ. Microbiol. 56, 2104^2107.
[30] Gameson, A.L.H. and Gould, D.J. (1975) E¡ects of solar radiation
on the mortality of some terrestrial bacteria in sea water. In: Proceedings of the International Symposium on Discharge of Sewage
from Sea Outfalls (Gameson, A.L.H., Ed.), pp. 209^219. Pergamon
Press, London.
[31] Chamberlin, C.E. and Mitchell, R. (1978) A decay model for enteric
bacteria in natural waters. In: Water Pollution Microbiology, Vol. 2
(Mitchell, R., Ed.), pp. 325^348. John Wiley and Sons, New York.
[32] Fujioka, R.S., Hashimoto, H.H., Siwak, E.B. and Young, R.H.
(1981) E¡ect of sunlight on survival of indicator bacteria in seawater.
Appl. Environ. Microbiol. 41, 690^696.
[33] Sinton, L.W., Davies-Colley, R.J. and Bell, R.G. (1994) Inactivation
of enterococci and fecal coliforms from sewage and meatworks e¥uents in seawater chambers. Appl. Environ. Microbiol. 60, 2040^2048.
[34] Davies-Colley, R.J., Bell, R.G. and Donnison, A.M. (1994) Sunlight
inactivation of enterococci and fecal coliforms in sewage e¥uent
diluted in seawater. Appl. Environ. Microbiol. 60, 2049^2058.
[35] Sinton, L.W., Finalay, R.K. and Lynch, P.A. (1999) Sunlight inactivation of fecal bacteriophages and bacteria in sewage-polluted seawater. Appl. Environ. Microbiol. 65, 3605^3613.
[36] Barcina, I., Gonzalez, J.M., Iriberri, J. and Egea, L. (1990) Survival
strategy of Escherichia coli and Enterococcus faecalis in illuminated
fresh and marine systems. J. Appl. Bacteriol. 68, 189^198.
[37] Calkins, S. and Barcelo, J.A. (1982) Action spectra. In: The Role of
Solar Ultraviolet Radiation in Marine Ecosystems (Calkins, J., Ed.),
pp. 143^150. Plenum Press, New York.
[38] Lessard, E.J. and Sieburth, J.M. (1983) Survival of natural sewage
populations of enteric bacteria in di¡usion and batch chambers in the
marine environment. Appl. Environ. Microbiol. 45, 950^959.
[39] Gourmelon, M. (1995) Etude de la Lumie©re Visible comme Facteur
Limitant de la Survie de Escherichia coli en Milieu Marin, 144 pp.
Univ. Rennes I.
[40] Kapuscinski, R.B. and Mitchell, R. (1981) Solar radiation induces
sublethal injury in Escherichia coli in seawater. Appl. Environ. Microbiol. 41, 670^674.
[41] Gourmelon, M., Touati, D., Pommepuy, M. and Cormier, M. (1997)
Survival of Escherichia coli exposed to visible light in seawater: analysis of rpoS-dependent e¡ects. Can. J. Microbiol. 43, 1036^1043.
[42] Munro, P.M., Clement, R.L., Flatau, G.N. and Gauthier, M.J.
(1994) E¡ect of thermal, oxidative, acidic, osmotic, or nutritional
stresses on subsequent culturability of Escherichia coli in seawater.
Microb. Ecol. 27, 57^63.
[43] Munro, P.M., Flatau, G.N., Clement, R.L. and Gauthier, M.J.
(1995) In£uence of the RpoS (KatF) sigma factor on maintenance
of viability and culturability of Escherichia coli and Salmonella typhimurium in seawater. Appl. Environ. Microbiol. 61, 1853^1858.
FEMSRE 725 5-12-01
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
[44] Calkins, J. (1982) A method for the estimation of the penetration of
biologically injurious solar ultraviolet radiation into natural water.
In: The Role of Solar Ultraviolet Radiation in Marine Ecosystems
(Calkins, J., Ed.), pp. 247^261. Plenum, New York.
[45] Baker, K.S. and Smith, R.C. (1982) Spectral irradiance penetration in
natural waters. In: The Role of Solar Ultraviolet Radiation in Marine Ecosystems (Calkins, J., Ed.), pp. 233^246. Plenum, New York.
[46] Jerlov, N.G. (1976) Marine Optics, pp. 47^63. Elsevier, New York.
[47] Munro, P.M., Gauthier, M.J., Breittmayer, V.A. and Bongiovanni, J.
(1989) In£uence of osmoregulation processes on starvation survival
of Escherichia coli in seawater. Appl. Environ. Microbiol. 55, 2017^
2024.
[48] Carlucci, A.F. and Pramer, D. (1960) An evaluation of factors e¡ecting the survival of Escherichia coli in seawater II: salinity, pH, and
nutrients. Appl. Environ. Microbiol. 8, 247^250.
[49] Anderson, I.C., Rhodes, M.W. and Kator, H.I. (1979) Sublethal
stress in Escherichia coli: a function of salinity. Appl. Environ. Microbiol. 38, 1147^1152.
[50] Flatau, G.N., Brunt, D. and Gauthier, M.G. (1994) E¡ect of longterm incubation in seawater on the ability of Escherichia coli cells to
incorporate histidine and maltose. Lett. Appl. Microbiol. 18, 257^
259.
[51] Csonka, L.N. and Epstein, W. (1996) Osmoregulation. In: Escherichia coli and Salmonella: Cellular and Molecular Biology, Vol. 1
(Neidhardt, F.C., Curtiss, R., Ingraham, J.L., Lin, E.C.C., Low,
K.B., Magasanik, B., Rezniko¡, W.S., Riley, M., Schaechter, M.
and Umbarger, H.E., Eds.), pp. 1210^1223. ASM Press, Washington,
DC.
[52] Dupray, E. and Derrien, A. (1995) In£uence of the previous stay of
Escherichia coli and Salmonella spp. in waste waters on their survival
in seawater. Water Res. 29, 1005^1011.
[53] Gauthier, M.J., Fatau, G.N., Munro, P.M. and Clement, R.L. (1993)
Glutamate uptake and synthesis by Escherichia coli cells in seawater:
e¡ects on culturability loss and glycine betaine transport. Microb.
Releases 2, 53^59.
[54] Munro, P.M., Laumond, F. and Gauthier, M.J. (1987) A previous
growth of enteric bacteria on salted medium increased their survival
in seawater. Lett. Appl. Microbiol. 4, 121^124.
[55] Gauthier, M.J., Flatau, G.N., Le Rudulier, D., Clement, R.L. and
Combarro, M.P. (1991) Intracellular accumulation of potassium and
glutamate speci¢cally enhances survival of Escherichia coli in seawater. Appl. Environ. Microbiol. 57, 272^276.
[56] Gauthier, M.J. and Le Rudulier, D. (1990) Survival in seawater of
Escherichia coli cells grown in marine sediments containing glycine
betaine. Appl. Environ. Microbiol. 56, 2915^2918.
[57] Combarro, M.P., Gauthier, M.J., Flatau, G.N. and Clement, R.L.
(1992) E¡ect of transient incubation in waste water on the sensitivity
to sea water of Escherichia coli cells grown under intestinal-like conditions. Biomed. Lett. 47, 185^190.
[58] Harvey, H.W. (1955) In: The Chemistry and Fertility of Sea Waters.
Cambridge University Press, London.
[59] Jannasch, H.W. (1968) Competitive elimination of Enterobacteriaceae from seawater. Appl. Microbiol. 16, 1616^1618.
[60] Lopez-Torres, A.J., Prieto, L. and Hazen, T.C. (1988) Comparison of
the in situ survival and activity of Klebsiella pneumoniae and Escherichia coli in tropical marine environment. Microb. Ecol. 15, 41^
57.
[61] Ingraham, J.L. and Marr, A.G. (1996) E¡ect of temperature, pressure, pH, and osmotic stress on growth. In: Escherichia coli and
Salmonella : Cellular and Molecular Biology, Vol. 2 (Neidhardt,
F.C., Curtiss, R., Ingraham, J.L., Lin, E.C.C., Low, K.B., Magasanik, B., Rezniko¡, W.S., Riley, M., Schaechter, M. and Umbarger,
H.E., Eds.), pp. 1570^1578. ASM Press, Washington, DC.
[62] Carlucci, A.F. and Pramer, D. (1960) An evaluation of factors e¡ecting the survival of Escherichia coli in seawater. I. Experimental procedures. Appl. Environ. Microbiol. 8, 243^247.
[63] Vasconcelos, G.J. and Swartz, R.G. (1976) Survival of bacteria in
[64]
[65]
[66]
[67]
[68]
[69]
[70]
[71]
[72]
[73]
[74]
[75]
[76]
[77]
[78]
[79]
[80]
[81]
[82]
[83]
[84]
[85]
527
seawater using a di¡usion chamber apparatus in situ. Appl. Environ.
Microbiol. 31, 913^920.
Welch, T.J., Farewell, A., Neidhardt, F.C. and Bartlett, D.H. (1993)
Stress response of Escherichia coli to elevated hydrostatic pressure.
J. Bacteriol. 175, 7170^7177.
Baross, J.A., Hanus, F.J. and Morita, R.Y. (1975) Survival of human
enteric and other sewage microorganisms under simulated deep-sea
conditions. Appl. Microbiol. 30, 309^318.
Gerba, C.P. and McLeod, J.S. (1976) E¡ect of sediments on the
survival of Escherichia coli in marine waters. Appl. Environ. Microbiol. 32, 114^120.
Shiaris, M.P., Rex, A.C., Pettibone, G.W., Keay, K., McManus, P.,
Rex, M.A., Ebersole, J. and Gallagher, E. (1987) Distribution of
indicator bacteria and Vibrio parahaemolyticus in sewage-polluted
intertidal sediments. Appl. Environ. Microbiol. 53, 1756^1761.
Hood, M.A. and Ness, G.E. (1982) Survival of Vibrio cholerae and
Escherichia coli in estuarine waters and sediments. Appl. Environ.
Microbiol. 43, 578^584.
Van Donsel, D.J. and Geldreich, E.E. (1971) Relationships of Salmonella to fecal coliforms in bottom sediments. Water Res. 5, 1079^
1087.
Ghoul, M., Bernard, T. and Cormier, M. (1990) Evidence that Escherichia coli accumulates glycine betaine from marine sediments.
Appl. Environ. Microbiol. 56, 551^554.
King, G.M. (1984) Metabolism of trimethylamine, choline, and glycine betaine by sulfate-reducing and methanogenic bacteria in marine
sediments. Appl. Environ. Microbiol. 48, 719^725.
Munro, P.M., Gauthier, M.J. and Laumond, F.M. (1987) Changes in
Escherichia coli cells starved in seawater or grown in seawater-wastewater mixtures. Appl. Environ. Microbiol. 53, 1476^1481.
Gauthier, M.J., Munro, P.M. and Breittmayer, V.A. (1989) In£uence
of prior growth conditions on low nutrient response of Escherichia
coli in seawater. Can. J. Microbiol. 35, 379^383.
Gauthier, M.J., Thomas, P. and Munro, P.M. (1989) Modi¢cation of
the envelope and the protein content of Escherichia coli surviving in
sea water. Can. J. Microbiol. 35, 843^849.
Gauthier, M.J., Flatau, G.N. and Clement, R.L. (1990) In£uence of
phosphate ions and alkaline phosphatase activity on survival of Escherichia coli cells in seawater. Microb. Ecol. 20, 245^251.
Gauthier, M.J., Flatau, G.N., Clement, R.L. and Munro, P.M.
(1992) Sensitivity of Escherichia coli cells to seawater closely depends
on their growth stage. J. Appl. Bacteriol. 73, 257^262.
Gonzalez, J.M., Iriberri, J., Egea, L. and Barcina, I. (1992) Characterization of culturability, protistan grazing, and death of enteric
bacteria in aquatic ecosystems. Appl. Environ. Microbiol. 58, 998^
1004.
Le Guyader, F., Pommepuy, M. and Cormier, M. (1991) Implantation of Escherichia coli in pilot experiments and the in£uence of
competition on the £ora. Can. J. Microbiol. 37, 116^121.
Greenberg, A.E. (1956) Survival of enteric organisms in sea water.
Public Health Rep. 71, 77^86.
Mitchell, R. and Nevo, Z. (1965) Decomposition of structural polysaccharides of bacteria by marine micro-organisms. Nature 205,
1007^1008.
Mitchell, R., Yankofsky, S. and Jannasch, H.W. (1967) Lysis of Escherichia coli by marine micro-organisms. Nature 215, 891^893.
Guelin, A., Lepine, P. and Lamblin, D. (1967) Pouvoir bactëricide
des eaux polluëes et roªle de Bdellovibrio bacteriovorus. Ann. Inst.
Pasteur Paris 113, 660^665.
Mitchell, R. and Morris, J.C. (1969) The fate of intestinal bacteria in
the sea. In: Advances in Water Pollution Research. Proceedings,
Fourth International Conference, Prague (Jenkins, S.H., Ed.), pp.
811^817. Pergamon Press, New York.
Enzinger, R.M. and Cooper, R.C. (1976) Role of bacteria and protozoa in the removal of Escherichia coli from estuarine waters. Appl.
Environ. Microbiol. 31, 758^763.
Barcina, I., Gonzalez, J.M., Iriberri, J. and Egea, L. (1992) Role of
FEMSRE 725 5-12-01
528
[86]
[87]
[88]
[89]
[90]
[91]
[92]
[93]
[94]
[95]
[96]
[97]
[98]
[99]
[100]
[101]
[102]
[103]
[104]
[105]
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
protozoa in the regulation of enteric bacteria populations in seawater. Marine Microb. Food Webs 5, 179^188.
Mitchell, R. (1968) Factors a¡ecting the decline of non-marine organisms in sea water. Water Res. 2, 535^543.
Carlucci, A.F. and Pramer, D. (1960) An evaluation of factors effecting the survival of Escherichia coli in seawater. IV. Bacteriophages. Appl. Environ. Microbiol. 8, 254^256.
Sogaard, H. (1983) Relationship between the occurrence of coliphages and E. coli in Danish marine bathing areas. Zentralbl. Bakteriol. Mikrobiol. Hyg. [B] 177, 394^401.
Paul, J.H., Rose, J.B., Jiang, S.C., Kellogg, C.A. and Dickson, L.
(1993) Distribution of viral abundance in the reef environment of
Key Largo, Florida. Appl. Environ. Microbiol. 59, 718^724.
Paul, J.H., Rose, J.B., Jiang, S.C., London, P., Xhou, X. and Kellogg, C. (1997) Coliphage and indigenous phage in Mamala Bay,
Oahu, Hawaii. Appl. Environ. Microbiol. 63, 133^138.
Legnani, P., Leoni, E., Lev, D., Rossi, R., Villa, G.C. and Bisbini,
P. (1998) Distribution of indicator bacteria and bacteriophages in
shell¢sh and shell¢sh-growing waters. J. Appl. Microbiol. 85, 790^
798.
Marino, F.J., Martinez-Manazanares, E., Morinigo, M.A. and Borrego, J.J. (1995) Applicability of recreational water quality standard
guidelines. Water Sci. Technol. 31, 27^31.
Borrego, J.J., Morinigo, M.A., Vicente, A.d., Cornax, R. and Romero, P. (1987) Coliphages as an indicator of fecal pollution in
water. Its relationship with indicator and pathogen micro-organisms. Water Res. 21, 1473^1480.
Dutka, B.J., Shaarawi, A.E. and Martins, M.T. (1987) North and
south American studies on the potential of coliphages as a water
quality indicator. Water Res. 21, 1127^1135.
Penon, F.J., Martinez, J., Vives-Rego, J. and Garcia-Lara, J. (1991)
Mortality of marine bacterial strains in seawater. Antonie Van
Leeuwenhoek 59, 207^213.
Rosenfeld, W.D. and ZoBell, C.E. (1947) Antibiotics production by
marine microorganisms. J. Bacteriol. 54, 393^398.
Vaccaro, R.F., Briggs, M.P., Carey, C.L. and Ketchum, B.H. (1950)
Viability of Escherichia coli in seawater. Am. J. Public Health 40,
1257^1266.
Richou, R., Neant, M. and Richou, H. (1955) Sur le pouvoir bactëricide de l'eau de mer a l'ëgard du staphylocoque. Rev. Immunol.
19, 64^68.
Gauthier, M.J., Benson, S.A., Flatau, G.N., Clement, R.L., Breittmayer, V.A. and Munro, P.M. (1992) OmpC and OmpF porins
in£uence viability and culturability of Escherichia coli cells incubated in seawater. Microb. Releases 1, 47^50.
Gauthier, M.J., Flatau, G.N., Clement, R.L. and Munro, P.M.
(1993) The loss of culturability by Escherichia coli cells in seawater
depends on availability of phosphate ions and phosphate transport
systems. Microb. Ecol. 26, 29^35.
Loewen, P.C., Hu, B., Strutinsky, J. and Sparling, R. (1998) Regulation in the rpoS regulon of Escherichia coli. Can. J. Microbiol. 44,
707^717.
Cashel, M., Gentry, D.R., Hernandez, V.J. and Vinella, D. (1996)
The stringent response. In: Escherichia coli and Salmonella: Cellular
and Molecular Biology, Vol. 1 (Neidhardt, F.C., Curtiss, R., Ingraham, J.L., Lin, E.C.C., Low, K.B., Magasanik, B., Rezniko¡, W.S.,
Riley, M., Schaechter, M. and Umbarger, H.E., Eds.), pp. 1458^
1496. ASM Press, Washington, DC.
Gentry, D.R., Hernandez, V.J., Nguyen, L.H., Jensen, D.B. and
Cashel, M. (1993) Synthesis of the stationary-phase sigma factor
sigma s is positively regulated by ppGpp. J. Bacteriol. 175, 7982^
7989.
Giaever, H.M., Styrvold, O.B., Kaasen, I. and Strom, A.R. (1988)
Biochemical and genetic characterization of osmoregulatory trehalose synthesis in Escherichia coli. J. Bacteriol. 170, 2841^2849.
Hengge-Aronis, R., Klein, W., Lange, R., Rimmele, M. and Boos,
W. (1991) Trehalose synthesis genes are controlled by the putative
[106]
[107]
[108]
[109]
[110]
[111]
[112]
[113]
[114]
[115]
[116]
[117]
[118]
[119]
[120]
[121]
[122]
[123]
[124]
[125]
sigma factor encoded by rpoS and are involved in stationary-phase
thermotolerance in Escherichia coli. J. Bacteriol. 173, 7918^7924.
Kaasen, I., Falkenberg, P., Styrvold, O.B. and Strom, A.R. (1992)
Molecular cloning and physical mapping of the otsBA genes, which
encode the osmoregulatory trehalose pathway of Escherichia coli:
evidence that transcription is activated by katF. J. Bacteriol. 174,
889^898.
Epstein, W. and Kim, B.S. (1971) Potassium transport loci in Escherichia coli K-12. J. Bacteriol. 108, 639^644.
Benson, S.A. and Decloux, A. (1985) Isolation and characterization
of outer membrane permeability mutants in Escherichia coli K-12.
J. Bacteriol. 161, 361^367.
Garrett, S., Taylor, R.K. and Silhavy, T.J. (1983) Isolation and
characterization of chain-terminating nonsense mutations in a porin
regulator gene, envZ. J. Bacteriol. 156, 62^69.
Boos, W., Ehmann, U., Bremer, E., Middendorf, A. and Postma, P.
(1987) Trehalase of Escherichia coli. Mapping and cloning of its
structural gene and identi¢cation of the enzyme as a periplasmic
protein induced under high osmolarity growth conditions. J. Biol.
Chem. 262, 13212^13218.
Karpel, R., Alon, T., Glaser, G., Schuldiner, S. and Padan, E.
(1991) Expression of a sodium proton antiporter (NhaA) in Escherichia coli is induced by Na‡ and Li‡ ions. J. Biol. Chem. 266,
21753^21759.
Dover, N. and Padan, E. (2001) Transcription of nhaA, the main
Na(+)/H(+) antiporter of Escherichia coli, is regulated by Na(+) and
growth phase. J. Bacteriol. 183, 644^653.
Jung, J.U., Gutierrez, C. and Villarejo, M.R. (1989) Sequence of an
osmotically inducible lipoprotein gene. J. Bacteriol. 171, 511^520.
May, G., Faatz, E., Villarejo, M. and Bremer, E. (1986) Binding
protein dependent transport of glycine betaine and its osmotic regulation in Escherichia coli K-12. Mol. Gen. Genet. 205, 225^233.
Lange, R. and Hengge-Aronis, R. (1991) Identi¢cation of a central
regulator of stationary-phase gene expression in Escherichia coli.
Mol. Microbiol. 5, 49^59.
Dukan, S. and Touati, D. (1996) Hypochlorous acid stress in Escherichia coli : resistance, DNA damage, and comparison with hydrogen peroxide stress. J. Bacteriol. 178, 6145^6150.
Loewen, P.C. and Triggs, B.L. (1984) Genetic mapping of katF, a
locus that with katE a¡ects the synthesis of a second catalase species
in Escherichia coli. J. Bacteriol. 160, 668^675.
Kumar, S. (1976) Properties of adenyl cyclase and cyclic adenosine
3P,5P-monophosphate receptor protein-de¢cient mutants of Escherichia coli. J. Bacteriol. 125, 545^555.
Sabourin, D. and Beckwith, J. (1975) Deletion of the Escherichia
coli crp gene. J. Bacteriol. 122, 338^340.
Xiao, H., Kalman, M., Ikehara, K., Zemel, S., Glaser, G. and
Cashel, M. (1991) Residual guanosine 3P,5P-bispyrophosphate synthetic activity of relA null mutants can be eliminated by spoT null
mutations. J. Biol. Chem. 266, 5980^5990.
Touati, D., Jacques, M., Tardat, B., Bouchard, L. and Despied, S.
(1995) Lethal oxidative damage and mutagenesis are generated by
iron in vfur mutants of Escherichia coli : protective role of superoxide dismutase. J. Bacteriol. 177, 2305^2314.
Dukan, S., Belkin, S. and Touati, D. (1999) Reactive oxygen species
are partially involved in the bacteriocidal action of hypochlorous
acid. Arch. Biochem. Biophys. 367, 311^316.
Carlioz, A. and Touati, D. (1986) Isolation of superoxide dismutase
mutants in Escherichia coli is superoxide dismutase necessary for
aerobic life? EMBO J. 3, 623^630.
Mount, D.W., Low, K.B. and Edmiston, S.J. (1972) Dominant mutations (lex) in Escherichia coli K-12 which a¡ect radiation sensitivity and frequency of ultraviolet light-induced mutation. J. Bacteriol.
112, 886^893.
Kusukawa, N. and Yura, T. (1988) Heat shock protein GroE of
Escherichia coli: key protective roles against thermal stress. Genes
Dev. 2, 874^882.
FEMSRE 725 5-12-01
Y. Rozen, S. Belkin / FEMS Microbiology Reviews 25 (2001) 513^529
[126] Howard-Flanders, P., Simson, E. and Theriot, L. (1964) A locus
that controls ¢lament formation and sensitivity to radiation in Escherichia coli K-12. Genetics 49, 237^246.
[127] Rouviere, P.E., De-Las-Penas, A., Mecsas, J., Lu, C.Z., Rudd, K.E.
and Gross, C.A. (1995) rpoE, the gene encoding the second heatshock sigma factor, sigma E, in Escherichia coli. EMBO J. 14, 1032^
1042.
[128] Farewell, A., Diez, A.A., DiRusso, C.C. and Nystrom, T. (1996)
Role of the Escherichia coli FadR regulator in stasis survival and
growth phase-dependent expression of the uspA, fad, and fab genes.
J. Bacteriol. 178, 6443^6450.
[129] Lederberg, J., The Rockefeller University, Suite 400 (Founders
Hall), 1230 York Avenue, New York, NY 0021-6399, USA.
[130] Willsky, G.R., Bennett, R.L. and Malamy, M. (1973) Inorganic
phosphate transport in Escherichia coli: involvement of two genes
which play a role in alkaline phosphatase regulation. J. Bacteriol.
113, 529^539.
[131] Garen, A., Molecular Biology and Biophysics, Yale, New Haven,
CT, USA.
[132] Garen, A. and Garen, S. (1963) Genetic evidence on the nature of
the repressor for alkaline phosphatase in E. coli. J. Mol. Biol. 6,
433^438.
529
[133] Neidhardt, F.C., Bloch, P.L. and Smith, D.F. (1974) Culture medium for enterobacteria. J. Bacteriol. 119, 736^747.
[134] Miller, J.H. (1972) Experiments in Molecular Genetics. Cold Spring
Harbor Laboratory Press, Cold Spring Harbor, NY.
[135] Van Dyk, T.K., Wei, Y., Hanafey, M.K., Dolan, M., Reeve,
M.J.G., Rafalski, J.A., Rothman-Denes, L.B. and LaRossa, R.A.
(2001) A genomic approach to gene fusion technology. Proc. Natl.
Acad. Sci. USA 98, 2555^2560.
[136] 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.
[137] Barer, M.R. and Harwood, C.R. (1999) Bacterial viability and culturability. Adv. Microb. Physiol. 41, 93^137.
[138] Edwards, C. (2000) Problems posed by natural environments for
monitoring microorganisms. Mol. Biotechnol. 15, 211^223.
[139] Rozen, Y., Van Dyk, T.K., LaRossa, R.A. and Belkin, S. Seawater
activation of Escherichia coli gene promoter elements: dominance of
rpoS control. Microb. Ecol., in press.
FEMSRE 725 5-12-01