FOODBORNE PATHOGENS AND DISEASE

Volume 7, Number 4, 2010 Review

ª Mary Ann Liebert, Inc.
DOI: 10.1089=fpd.2009.0379

Cronobacter (Enterobacter sakazakii ):

An Opportunistic Foodborne Pathogen

Brendan Healy,1 Shane Cooney,1 Stephen O’Brien,1 Carol Iversen,1 Paul Whyte,1
Jarlath Nally,2 John J. Callanan,2 and Séamus Fanning1

Abstract
Cronobacter spp. (Enterobacter sakazakii) are a recently described genus that is comprised of six genomospecies.
The classification of these organisms was revised based on a detailed polyphasic taxonomic study. Cronobacter
spp. are regarded as ubiquitous organisms having been isolated from a wide variety of foods. These bacteria are
opportunistic pathogens and are linked with life-threatening infections in neonates. Clinical symptoms of Cro-
nobacter infection include necrotizing enterocolitis, bacteremia, and meningitis, with case fatality rates of 50–80%
being reported. Contaminated powdered infant formula has been epidemiologically linked with infections.
Recently, infections among immunocompromised adults, mainly the elderly, have also been reported. A high
tolerance to osmotic stress and elevated temperatures contribute to the survival of Cronobacter spp. in dried foods
such as powdered infant formula. Controlling the organism in the production environment, thereby reducing
dissemination, necessitates the provision of suitable diagnostic tools. Studies demonstrated that a high degree of
variability exists amongst the phenotypic-based methods used to identify Cronobacter spp. However, advances in
molecular detection and subtyping techniques have significantly improved the identification and characteriza-
tion of Cronobacter spp. The dose required to induce infection has yet to be determined. In vitro virulence studies
have shown that Cronobacter spp. may survive in macrophage cells and efficiently attach to and invade epithelial
cell lines. The production of exopolysaccharide may contribute to the formation of biofilm and active efflux
pumps promote resistance to antimicrobial agents such as bile salts and disinfectants. A holistic approach
combining techniques such as comparative genome analysis, proteomics, and in vivo challenges could help
unravel the complex interactions between this pathogen and its host. These data would help identify those
properties in Cronobacter spp. which enable the bacterium to survive in the production environment and infect
vulnerable neonates via the food chain.

Classification E. sakazakii strains, Iversen et al. (2007, 2008b) proposed a

further classification of E. sakazakii with the creation of a new
C ronobacter spp. are a group of Gram-negative,
peritrichous, motile, non–spore-forming, facultative an-
aerobic bacteria, and members of the Enterobacteriaceae.
genus, Cronobacter. This genus proposal was based on detailed
f-amplified fragment length polymorphism (f-AFLP) finger-
printing, ribotyping, characterization of full-length 16S rRNA
These organisms were originally referred to as yellow pig-
gene sequences, DNA–DNA hybridization, and phenotypic
mented Enterobacter cloacae until they were reclassified in 1980
profiles. The genus consists of six genomospecies; C. sakazakii,
as a new species, Enterobacter sakazakii (Farmer et al., 1980).
Cronobacter malonaticus, Cronobacter turicensis, Cronobacter
DNA–DNA hybridization studies indicated that they were
muytjensii, Cronobacter genomospecies 1, and Cronobacter du-
53–54% related to two distinct genera: Enterobacter and Ci-
blinensis which contains three subspecies: dublinensis, lactaridi,
trobacter. However, as the new species were pheno- and ge-
and lausannensis.
notypically closer to E. cloacae than Citrobacter freundii, they
were subsequently included in the Enterobacter genus. Ori-
Reservoirs
ginally 15 biogroups were described (Farmer et al., 1980) with
a 16th group added more recently (Iversen et al., 2006). Fol- The primary reservoir of Cronobacter spp. has yet to be
lowing a polyphasic study of an extensive collection of determined but it is postulated that plant material may be an

1
Centres for Food Safety and Foodborne Zoonomics and 2UCD Veterinary Sciences Centre, School of Agriculture, Food Science, and
Veterinary Medicine, University College Dublin, Belfield, Dublin, Ireland.

339
340 HEALY ET AL.

important source. Cronobacter spp. can be isolated from a wide that an increase in aw or storage temperature accelerated the
variety of foods including milk, cheese, dried foods, meats, rate of death of Cronobacter spp. in dried infant cereals; how-
water, vegetables, rice, bread, tea, herbs, spices, and pow- ever, an initial inoculum of 2 CFU=g could survive in infant
dered infant formula (PIF) (Farber, 2004; Iversen et al., 2004b; cereals for up to 12 months over a 0.30–0.83 aw range. This is
Edelson-Mammel et al., 2005; Gurtler et al., 2005; Beauchamp important as it demonstrates that Cronobacter spp. may sur-
et al., 2006; Estuningsih et al., 2006; Friedemann, 2007; Mullane vive in low aw food (such as powders and cereals) for ex-
et al., 2007a). Surveillance studies have detected Cronobacter tended periods of time.
spp. in households, livestock facilities, food factories, and PIF Riedel and Lehner (2007) used proteomics to identify dif-
production facilities (Arts, 2004; Kandhai et al., 2004; Kilonzo- ferentially expressed proteins in response to two different
Nthenge et al., 2008). Clinically, Cronobacter spp. have been osmotic stresses: desiccation and growth in hyperosmotic
isolated from cerebrospinal fluid, bone marrow, blood, in- media. Results indicated that under high osmolarity condi-
testinal and respiratory tracts, urine, ear and eye swabs, and tions the central metabolic pathways, including amino acid
skin wounds (Adamson and Rodgers, 1981; Gallagher and biosynthesis and transport protein production, were shut-
Ball, 1991; Gurtler et al., 2005). down along with the downregulation of the motility appa-
Contaminated PIF has been epidemiologically linked with ratus. In contrast, an accumulation of structural proteins was
cases of Cronobacter infections in infants (Biering et al., 1989; observed consistent with the need to preserve the proper
Simmons et al., 1989; Lai, 2001; van Acker et al., 2001; Weir, functioning and integrity of the cell.
2002). Plant-derived ingredients added to PIF without any Cronobacter spp. are more thermotolerant compared with
prior heat treatment have been implicated as a potential route other members of the Enterobacteriaceae (Nazarowec-White
of transmission for the bacterium. Thermally sensitive ingre- and Farber, 1997; Breeuwer et al., 2003; Iversen et al., 2004b;
dients used in the production of PIF may also be a potential Edelson-Mammel et al., 2005; Asakura et al., 2007). Iversen et al.
source of contamination. Several studies have investigated the (2004b) reported that Cronobacter spp. can grow over a range
prevalence of Cronobacter spp. in PIF (Muytjens et al., 1988; of temperatures (6–478C) in PIF. Initially, D- and Z-values
Clark et al., 1990; Nazarowec-White and Farber, 1997; Iversen reported from their study correlated closely with data from
et al., 2004b); however, the incidence of Cronobacter spp. re- previous reports (Nazarowec-White and Farber, 1997;
mains low, in the order of between 0.22 and 1.61 CFU=100 g of Breeuwer et al., 2003). Recently, Kim and Park (2007) studied a
powdered sample. O’Brien et al. (2009a) surveyed a total of 39 collection of Cronobacter strains isolated in Korea. The result-
packets of follow-on powdered formula and 8 infant drinks. ing D-values in saline solution were 12–16, 3–5, and 0.9–1 min
Although no Cronobacter spp. were isolated from the follow- at 528C, 568C, and 608C, respectively. An increase in D-value
on formula, it was identified in two cereal-based drinks at was observed for rehydrated PIF and dried baby food. Their
levels of 0.003 and 0.013 most probable number (MPN)=g. data supported the findings of Iversen et al. (2004b). Challenge
Interestingly, both products contained malt extract and cocoa experiments demonstrated that temperature, desiccation, or
ingredients which are added as flavorings. This supports the antimicrobial agent stress significantly lowered the D-values
belief that the addition of unscreened raw ingredients may be for Cronobacter spp.
the source of contamination of PIF in some cases.
Mullane et al. (2008a) surveyed a powdered milk produc-
Clinical Significance
tion facility and identified persistent strains (identical by
molecular subtyping analysis) throughout the manufacturing Cronobacter spp. are considered emerging opportunistic
site. Their study identified the presence of three different pathogens and have been identified as aetiological agents of
clonal populations distributed throughout. A study by bacteremia, necrotizing enterocolitis, and neonatal meningitis
Baumgartner et al. (2009) identified Cronobacter spp. in ready- (Fig. 1a) (Arseni et al., 1987; Biering et al., 1989; Gallagher and
to-eat foods, with 61% of sprout=fresh herbs and 27% of dried Ball, 1991; Bar-Oz et al., 2001; Caubilla-Barron et al., 2007;
herbs=spices sampled containing the bacteria. It is possible Mullane et al., 2007a; Giovannini et al., 2008). Urmenyi and
that raw materials may be an important source, leading to the Frankin (1961) recorded the first isolation of Cronobacter spp.
entry of Cronobacter spp. into the food chain. Subsequent from a case of neonatal meningitis. Case fatality rates as high
strain selection may lead to persistent Cronobacter types de- as 80% have since been reported (Lai, 2001).
veloping within the production facility. Controlling the mi- Based on data from the Centers for Disease Control and
crobial flora is therefore an important step toward limiting the Prevention, it is estimated that there are approximately six
dissemination of Cronobacter spp. in the environment. new cases of Cronobacter infections reported each year
worldwide. However, this estimate does not take into account
the number of misidentifications (false-negative) that may
Osmotic and Thermal Tolerance
occur for Cronobacter strains.
A wide variety of Enterobacteriacae have been isolated Bowen and Braden (2006) performed a risk analysis on 46
from PIF, including Citrobacter koseri, Salmonella enterica, cases of infection in infants. Thirty-three cases of meningitis,
Klebsiella pneumoniae, Klebsiella oxytoca, Pantoea agglomerans, 12 cases of bacteremia, and 1 urinary tract infection were
E. cloacae, and Escherichia vulneris (FAO=WHO, 2004, 2006). analyzed. Notably, infants of greater birth weight (2500 g),
However, when compared with other members of the En- gestational development (37 weeks), and younger age (6
terobacteriacae family, Cronobacter spp. are reported to be days) presented with meningitis, as opposed to bacteremia
more adapted to dry stress (Caubilla-Barron et al., 2007). Lin which developed in infants of a lower birth weight (850 g),
and Beuchat (2007) investigated the effects of water avail- reduced gestational development (27.8 weeks), and older in
ability (aw) and temperature on Cronobacter spp. recovered age (35 days). The risk analysis postulates that infants born
from infant cereal over a 12-month period. These data showed prematurely are more prone to invasive central nervous sys-
CRONOBACTER: AN OPPORTUNISTIC PATHOGEN 341

FIG. 1. Colonizations, symptoms, and incidence of occurrence as a percentage of total Cronobacter infections: (a) in children
and infants and (b) in adults.

tem disease; however, it is also proposed that infants born and Farber (1997) modified this method to survey the inci-
prematurely are more likely to receive sterile infant formula. dence of Cronobacter spp. in PIF. In 2002, the U.S. Food and
Conversely, infants born with normal gestational develop- Drug Administration recommended a protocol (Fig. 2a) for
ment and birth weight are likely to receive nonsterile PIF, the detection of Cronobacter spp. which was based on these
potentially exposing them to Cronobacter infections. two methods. In this protocol, samples were directly streaked
Although Cronobacter spp. have been primarily associated onto violet red bile glucose agar following enrichment. Five
with infections in infants, recent reports have highlighted the presumptive-positive colonies were then subcultured onto
risk posed to immunocompromised adults, particularly the tryptic soy agar and incubated for 48–72 h at 258C. Yellow-
elderly. Jimenez and Gimenez (1982) reported the first Cro- pigmented colonies were selected and confirmed using API
nobacter spp. isolated from an adult with bacteremia. An ad- 20E biochemical identification galleries. The detection limits
ditional 19 cases of Cronobacter infections in adults have now for Cronobacter spp. reported (0.36 CFU=100 g) were similar to
been reported (personal communication); clinical symptoms those described by Muytjens et al. (1988). The level of sensi-
include pneumoniae, sepsis, foot ulcers, wound infections, tivity of this assay exceeds those recommended by the Food
osteomyletis, and splenic abscesses (Fig. 1b). Gosney et al. and Agriculture Organization (3.0 CFU=g PIF). However, this
(2006) examined 203 stroke patients and isolated Cronobacter method has since been revised to incorporate both culture on
spp. from the oral cavity in 7 cases. See et al. (2007) reported chromogenic media and real-time polymerase chain reaction
the first case of Cronobacter infection in a nonimmunocom- (PCR) for the detection of Cronobacter spp. in PIF (Chen et al.,
promised adult, wherein a 75-year-old woman presented with 2009). A study of 360 test portions has shown the revised
a splenic abscess. These reports suggest that older individu- method to be significantly better ( p < 0.05) than the original
als may also be more susceptible to Cronobacter infections. protocol for the detection of Cronobacter spp.
The International Organization for Standardization (ISO)
and the International Dairy Federation devised a protocol (Fig.
Microbiological Detection
2b) for the isolation of Cronobacter spp., from milk-based
Muytjens et al. (1988) described the first quantitative powdered formula, ISO=TS 22964 (Anonymous, 2006). Buf-
method for the isolation of Cronobacter spp. Nazarowec-White fered peptone water is used to preenrich PIF samples at 378C.
 a Pre-enrichment b Pre-enrichment c Enrichment
3 x 1/10/100 g Make 1:10 (w/v) of sample Weigh out 50/100g sample into
in BPW 450 ml CEB and homogenize
Incubate 37°C for 18h ± 2h
Incubate 37/41.5°C for 18h ± 2h
1:10 dilution (w/v) in distilled water
Incubate 36 °C for 24h Selective Enrichment
Transfer 100 µl pre-enrichment to 10 ml mLST / Isolation/Enumeration
Original FDA Method Revised FDA Method vancomycin media Streak 10 µl onto
Enrichment Centrifuge 40 ml samples chromagenic agar
Incubate 44°C for 24h ± 2h
Transfer 10 ml pre-enrichment to 3,000 x g, 10 min Incubate 37°C for 12-24h
90 ml EE broth Resuspend pellet in 200 µl PBS
Incubate 36 °C for 24h Isolation
Streak 10µl onto chromagenic agar Confirmation
Streak typical colony onto TSA/Blood agar
Spread Plate Real-time Incubate 44°C for 24h ± 2h
PCR
Selection 100 µl onto chromagenic
media
Directly spread 0.1 ml; or streak
(10 µl) on VRBGA Confirmation Identification
Incubate 36 °C for 24h
Streak 5 typical colonies onto TSA Select colony for biochemical ID
Incubate 36 °C for 24h
Identification
Selection
Subculture presumptive-pos. on TSA
Identification
Two typical colonies from each chromagenic
plate confirmed by real-time PCR, API 20E Select 1 colony from each
Select Yellow Colonies
and Rapid ID 32E TSA plate for biochemical ID

API 20E

FIG. 2. Microbiological protocols for the detection of Cronobacter spp. (a) Original and revised FDA; (b) ISO=TS 22964; (c) adapted CEB. FDA, U.S. Food and Drug Adminis-
tration; ISO, International Organization for Standardization; CEB, Cronobacter enrichment broth.
CRONOBACTER: AN OPPORTUNISTIC PATHOGEN 343

Table 1. Biochemical Differentiation of the Cronobacter Genus (Adapted from Iversen et al., 2008b)

Characteristic CS CM CT CG CMY CDD CDL CDLS

a
Indole production



þ þ þ V
Dulcitolb

þ þ þ



Lactulose þ þ þ þ þ þ þ

Malonatec
þ þ V þ þ


Maltitol þ þ þ þ
þ þ

Palatinose þ þ þ þ V þ þ þ
Putrescine þ V þ V þ þ þ V
Melezitose

þ

þ


Turanose þ þ þ V V þ V

myo-Inositolb V V þ þ þ þ þ

cis-Aconitate þ þ þ þ V þ þ þ
trans-Aconitate
þ
þ V þ þ þ
1-0-Methyl a-d-glucopyranoside þ þ þ þ
þ þ þ
4-Aminobutyrate þ þ þ V þ þ þ þ
a
Kovac’s reagent added.
b
Biolog phenotype array.
c
Malonate phenylalanine broth.
þ, >90% positive;
, <10% positive; V, 20–80% positive.
CS, Cronobacter sakazakii; CM, Cronobacter malonaticus; CT, Cronobacter turicensis; CG, Cronobacter genomospecies 1; CMY, Cronobacter
muytjensii; CDD, Cronobacter dublinensis subspp. dublinensis; CDL, Cronobacter dublinensis subspp. lactaridi; CDLS, Cronobacter dublinensis
subspp. lausannensis.

Samples are subsequently enriched in modified lauryl sulfate media, Cronobacter screening broth, has the potential to in-
broth (20.0 g tryptose, 5.0 g lactose, 5.0 g sodium chloride, 2.75 g crease the specificity and selectivity for Cronobacter spp. de-
dipotassium hydrogen phosphate, 2.75 g potassium dihydro- tection (Iversen et al., 2008a). Other differential media have
gen phosphate, 0.1 g sodium lauryl sulfate, and 10 mg=L van- also been developed for the detection of Cronobacter spp. based
comycin) at 448C. Chromogenic agar (E. sakazakii isolation on a similar principle targeting a 4-methylumbelliferyl-a,
agar) was then employed to aid identification. However, this d-glucopyranoside substrate (MUaGlc) (Leuschner and Bew,
method has been shown to be unreliable as some isolates 2004; Oh and Kang, 2004) and a 5-bromo-4-chloro-3-indoxyl-
of Cronobacter do not grow in modified lauryl sulfate broth b-d-cellobioside substrate (Restaino et al., 2006).
or Enterobacteriaceae enrichment broth (10.0 g peptone, Iversen et al. (2007) performed a comprehensive study
5.0 g glucose 6.45 g disodium hydrogen phosphate, 2.0 g po- on a collection of 210 Cronobacter strains in which the phe-
tassium dihydrogen phosphate, 20.0 g purified ox bile, and notypic profiles for each of the strains were recorded. Their
0.01 g brilliant green) (Guillaume-Gentil et al., 2005; Lehner investigations identified key biochemical biomarkers for
et al., 2006; Iversen and Forsythe, 2007; Fox and Jordan, the differentiation of Cronobacter spp. Table 1 outlines the
2008). main biochemical tests for the differentiation of Cronobacter
More recently, O’Brien et al. (2009b) evaluated a new one- strains.
step protocol consisting of a combined preenrichment= Commercially available biochemical identification kits
enrichment broth (Cronobacter enrichment broth [CEB]) used have been routinely employed in food and clinical microbi-
in conjunction with selective-differential agar for the detection ology to identify bacteria from a wide variety of sources.
of Cronobacter spp. in PIF. Cronobacter spp. were recovered However, the accuracy and reliability of these tests have been
from all artificially inoculated powders at a significantly questioned of late, with reports that some resulted in false-
higher final concentration when compared with other en- negative and -positive identifications (Restaino et al., 2006;
richment broths ( p < 0.01). There was no significant difference Iversen et al., 2007). Recently, Gen III Identification plates
in bacterial cell concentration for cultures grown in CEB at (Biolog, Hayward, CA) have been developed for the identi-
378C or 41.58C. This protocol removes the need for separate fication of either Gram-negative or Gram-positive organisms.
preenrichment and enrichment steps (Fig. 2c), thus facilitating These are currently the only commercially available bio-
a speedier detection of Cronobacter spp. in PIF. chemical identification kits to include the updated taxonomy
Muytjens et al. (1984) originally identified an a-glucosidase of the Cronobacter genus, with all six genomospecies being
enzyme in all Cronobacter strains tested. Iversen et al. (2004a) identified (Healy, unpublished data).
utilized this important feature to develop a medium
(Brilliance
E. sakazakii agar, DFI (Druggan, Fosythe &
Molecular Detection and Characterization
Iversen) formulation, CM1055; Oxoid, Basingstoke, UK) for
the specific detection of Cronobacter spp. A basal medium was Seo and Brackett (2005) developed a quantitative real-time
supplemented with the chromogenic substrate 5-bromo-4- PCR assay for the detection of Cronobacter spp. The principle
chloro-3-indolyl-a, d-glucopyranoside (X-aGlc). a-Glucosidase of this technique was based on targeting an internal segment
in Cronobacter spp. hydrolyses X-aGlc to form a bromo-chloro- of the dnaG gene, within the macromolecular synthesis op-
indigo pigment and the colonies appear blue-green on DFI eron. To increase the specificity of detection a TaqMan probe
agar plates. This agar in combination with a novel screening was also included, which specifically hybridizes to the prox-
344 HEALY ET AL.

imal region of the amplified sequence. This method success- subtype a collection of geno- and phenotypically diverse
fully differentiated Cronobacter spp. from Enterobacter strains Cronobacter isolates. This study successfully discriminated
and other members of the Enterobacteriaceae. The sensitivity between the 112 isolates, and the discriminatory power was
of this method is approximately 100 CFU=mL of rehydrated comparable with PFGE. Healy et al. (2008) described the ap-
PIF. Coupled with a 24-h enrichment step the sensitivity im- plication of a semiautomated repetitive sequence-based PCR
proved to the detection of 0.6 CFU=g of PIF. analysis of a collection of Cronobacter strains. Although PFGE
Liu et al. (2006) evaluated a real-time PCR detection method analysis exhibited greater discrimination between individual
for Cronobacter spp. in PIF based on the amplification of an strains, repetitive sequence-based PCR analysis successfully
internal transcribed spacer sequence of the 16S–23S rDNA. grouped C. sakazakii, C. malonaticus, and C. dublinensis strains
Their method tested both a TaqMan fluorogenic probe and a into distinct clusters.
SYBR Green fluorescent dye. SYBR Green binds to the minor Significant reductions in the time and costs associated
groove of amplified DNA during PCR. The level of detection with genome sequencing have led to an increase in available
using the probe and SYBR Green was 1.1 CFU=100 g PIF. genome data. Ziegler (2003) identified certain genes that
Other target genes have also been investigated for PCR-based could be employed to estimate whole-genome resemblances
detection of Cronobacter spp.; these include the a-glucosidase based on the similarity of these genes. Kuhnert et al. (in press)
gene (gluA) (Lehner et al., 2006), outer membrane protein A described the genomic similarity of a collection of Cronobacter
(OmpA) (Mohan Nair and Venkitanarayanan, 2006), and strains based on recN, thdF, and rpoA gene sequence analy-
zinc-containing metalloprotease (zpx) (Kothary et al., 2007). sis. El-Sharoud et al. (2009) utilized recN gene sequence anal-
Cawthorn and Witthuhn (2008) used propidium monoazide ysis to speciate a collection of Cronobacter strains obtained
to inhibit PCR amplification of nonviable Cronobacter spp. from dried milk and related products. This approach suc-
while causing no inhibition to viable cells. Another PCR- cessfully characterized the collection of strains, including the
based method was developed recently by Stoop et al. (2009): identification of a novel phenotypic profile for a C. sakazakii
The rpoB gene (b-subunit of RNA polymerase) was targeted to isolate.
differentiate between the various species within the Crono-
bacter genus. This approach is based on the detection of single-
Pathogenicity and Virulence
nucleotide polymorphisms in the rpoB gene using mismatch
PCR. The mechanisms that contribute to the pathogenicity of
Fluorescence in situ hybridization is a technique that can be Cronobacter spp., particularly in relation to neonatal menin-
employed for the detection of bacteria. It is based upon the gitis, remain to be defined. It was originally speculated that
binding of specific probes to nucleic acid target regions. Al- one possible route of infection is the translocation of the
meida et al. (2009) recently described the application of a novel bacterium from blood to cerebrospinal fluid (CSF) via the
peptide nucleic acid probe for the detection of Cronobacter spp. choroid plexus, followed by invasion of the bacterium into
in PIF. Their study demonstrated that the method could detect nutrient-rich cerebral matter (Iversen and Forsythe, 2004).
Cronobacter spp. at an initial inoculum of 110
2 CFU=mL Numerous studies since have suggested that OmpA contrib-
following an 8-h preenrichment step. They also observed that utes significantly to the virulence potential of Cronobacter spp.
peptide nucleic acid fluorescence in situ hybridization facili- (Singamsetty et al., 2008; Mittal et al., 2009; Mohan Nair et al.,
tated the detection of Cronobacter spp. in PIF when tested in 2009). Mohan Nair and Venkitanarayanan (2007) described
mixed bacterial populations and in a lower concentration to how OmpA binds fibronectin, facilitating the invasion of
other bacteria present. brain endothelial cells. Mittal et al. (2009) demonstrated that
Molecular subtyping techniques have long been regarded ompA expression affected the onset of meningitis in newborn
as useful tools to aid our understanding of microbial ecology. rats. OmpA-positive Cronobacter isolates successfully crossed
Pulsed-field gel electrophoresis (PFGE) is well established and the intestinal barrier, multiplied in the blood, and were able to
widely used as a gold-standard method for molecular typing subsequently transverse the blood brain barrier, whereas
of bacteria including Cronobacter spp. (Nazarowec-White and OmpA-negative isolates could not bind to intestinal epithelial
Farber, 1999; Mullane et al., 2007b; Kim et al., 2008; Proudy cells. Further, a 100% mortality rate was observed in OmpA-
et al., 2008). Mullane et al. (2007b) applied PFGE to monitor the positive Cronobacter spp.-infected newborn rats compared
prevalence of Cronobacter spp. in a PIF-processing facility. with no pathological manifestations being observed in
PFGE was performed using an XbaI restriction enzyme digest OmpA-negative Cronobacter spp.-infected newborn rats.
which yielded between 9 and 18 DNA fragments ranging Several in vitro studies using mammalian cell lines have
from 48.5 to 1000 kbp. Eighty Cronobacter isolates were re- investigated the attachment and invasion properties of Cro-
covered from the PIF production facility, and these were nobacter strains. Mange et al. (2006) investigated the adherence
subsequently mapped to the layout of the site. The study to two epithelial cell lines, HEp-2 and Caco-2, and to the
highlighted points of contamination and also identified per- human-derived brain microvascular endothelial cell line.
sistent clonal isolates in the facility. Data showed that adherence was greatest during the late ex-
Recently, other molecular subtyping techniques have been ponential phase of bacterial growth with a 10-fold increase in
developed and compared with PFGE for the analysis of Cro- adherent cells during this phase. Similarly, Townsend et al.
nobacter strains. Variable-number tandem-repeat motifs are (2008) investigated the attachment and invasion properties of
short repeat sequences dispersed throughout bacterial ge- seven Cronobacter strains that were associated with an out-
nomes that are highly polymorphic. These regions have been break of necrotizing enterocolitis, bacteremia, and meningitis
used for subtyping bacteria (Lindstedt et al., 2004; Torpdahl in a neonatal intensive care unit. All were found to attach to
et al., 2006; van Belkum, 2007). Mullane et al. (2008b) applied and invade Caco-2 cells after a period of 3 h. These isolates
multiple-locus variable-number tandem-repeat analysis to could replicate and persist in macrophage cells (U937) for a
CRONOBACTER: AN OPPORTUNISTIC PATHOGEN 345

period of 48 h. Invasion studies using rat brain capillary en- salts (natural substrates), antibiotics, disinfectants=sanitizers,
dothelial cells (rBCEC4) showed that a strain from a fatal and dyes (Touze et al., 2004). Cronobacter spp. possess a variety
meningitis case was considerably more invasive when com- of resistance–nodulation–cell division efflux pumps includ-
pared with other nonmeningitis-associated strains. ing AcrAB-TolC (Healy, unpublished data). The AcrAB-TolC
Attempts to bridge the knowledge gap between in vitro as- efflux system is composed of the innermembrane resistance
says and mechanistic models in vivo remain in their infancy. nodulation division transporter AcrB, the outer membrane
Investigations have shown that Cronobacter spp. may success- channel TolC, and the periplasmic membrane fusion protein
fully adhere to and invade endothelial cells (Mange et al., 2006; AcrA. This efflux mechanism expels a broad range of antibi-
Townsend et al., 2007), though these observations are based otic classes, detergents, biocides, and dyes (Piddock, 2006).
on culturing monolayers of endothelial cells. An endothelial= The polyspecificity of efflux transporters confers a general
astrocyte coculture assay may prove a practical solution to resistance phenotype that can result in the acquisition of ad-
more accurately represent a blood brain barrier in vitro. ditional antimicrobial resistance. Studies are underway to
The infectious dose associated with Cronobacter spp. has yet explore the contribution of these transporters to virulence of
to be elucidated. Iversen and Forsythe (2004) initially esti- Cronobacter isolates.
mated that the infectious dose may be approximately
1000 CFU. The FAO=WHO (2004) proposed an infectious
Efficacy of Antimicrobial Agents
dose in the order of approximately 10,000 CFU. However,
despite these preliminary estimates, it is widely accepted that Cronobacter spp. have been described as sensitive, or at least
the dose–response relationship is also host dependent. Pa- displaying intermediate sensitivity, to many classes of anti-
gotto et al. (2003) first described the production of enterotoxin microbials including acylureidopenicillins, aminoglycosides,
in Cronobacter strains. A suckling mouse assay was used to ampicillin, antifolates, aztreonam, carbapenems, cephalo-
determine the pathogenicity of a small number of strains. All sporins, chloramphenicol, nitrofurantoin, quinolones, tetra-
were lethal at a level of 1108 CFU. Raghav and Aggarwal cyclines, ticarcillin, and several b-lactams (Farmer et al., 1980;
(2007) also reported the production of enterotoxin in Crono- Muytjens and van der Ros-van de Rede, 1986; Willis and
bacter strains. The activity of the toxin was found to be highest Robinson, 1988; Hawkins et al., 1991). Cronobacter spp. are
at pH 6 and showed stability at 908C for 30 min. more sensitive, compared with other members of the En-
The attachment of bacteria to environmental surfaces and terobacteriaceae, to many antibiotics including aminoglyco-
the formation of biofilms are known to contribute to survival sides and carboxy-penicillins (Stock and Wiedemann, 2002).
and increased resistance to antimicrobial treatments. Nu- Nearly all species of the Enterobacteriaceae exhibit a nat-
trient availability and temperature are significant factors ural resistance to glycopeptides, rifampicin, linocosamides,
affecting biofilm formation by Cronobacter spp. (Kim et al., fusidic acid, and streptogramins. This resistance phenotype
2006, 2007b, 2008). Iversen et al. (2004b) investigated the may be attributed to their outer membrane which acts as a
formation of biofilms on various surfaces that are commonly barrier preventing passage of these agents (Stock and Wei-
associated with PIF-feeding equipment and surfaces. Isolates demann, 2002). Cronobacter spp. display reduced susceptibil-
adhered to silicon, latex, polycarbonate, and stainless steel, ity to oxacillin, benzylpenicillin, clindamycin, and some
with apparently greater attachment occurring with bacteria macrolides. Some strains also exhibit a reduced sensitivity to
that produced exopolysaccharide capsules. Dancer et al. fosfomycin (Stock and Weidemann, 2002).
(2009) investigated the influences of milk components on The effectiveness of disinfectants commonly used in both
biofilm formation by Cronobacter spp. They proposed that clinical and food environments to promote clean surfaces and
whey protein and casein are more important than lactose prevent Cronobacter spp. contamination was investigated
with respect to biofilm formation in skim milk. The authors (Kim et al., 2007b). All the agents studied were effective
also suggested that nitrogen source is an important deter- against planktonic cells. However, it was observed that when
minant of biofilm formation by Cronobacter spp. compared bacterial cells were dried onto a solid surface in the presence
with carbohydrate. More recently, Hartmann et al. (2009) of an organic matrix the efficacy of the disinfectant was re-
adopted a molecular approach to characterize biofilm for- duced. Populations of cells were not significantly reduced
mation by Cronobacter spp. Their study identified that ex- when present as part of a biofilm. These data clearly dem-
tracellular DNA plays an important role in the early stages onstrate that the choice of sanitizing agent and its point of
of biofilm formation. It is believed that extracellular DNA application may not act to limit the dissemination of Crono-
helps to maintain a stable structure before other exopoly- bacter spp. Extrinsic contamination of materials and surfaces is
saccharide matrix components carry on this role. Colanic believed to be a significant determining factor regarding
acid (CA) is an exopolysaccharide component previously Cronobacter infections. Recently, the efficacy of recommended
identified in Cronobacter spp. (Scheepe-Leberkühne and methods for the decontamination of PIF-feeding bottles was
Wagner, 1986). CA may contribute to adherence of some assessed (Redmond and Griffith, 2009). The findings of this
surfaces and resistance to environmental stresses. The au- study indicated that good hygiene combined with hypo-
thors considered that CA may be a significant factor con- chlorite chemical or heat treatments, which adhered strictly to
tributing to biofilm formation and increased resistance to the suggested guidelines, resulted in effective decontamina-
environmental stresses (desiccation, heat, and pH) in Cro- tion of PIF-feeding apparatus.
nobacter spp.
Active efflux is a recognized virulence mechanism in En-
Alternative Control Measures
terobacteriaceae, contributing to survival in the host’s gastro-
intestinal tract. These membrane-associated pumps extrude a Alternative prevention strategies aimed at limiting the
range of xenobiotic compounds from the cell, including bile growth of Cronobacter spp. have been investigated. Biocontrol
346 HEALY ET AL.

is one possible measure that could ensure foods and surfaces In this study the phenolic compounds malic, tartaric, and
are free from contamination. The use of bacteriophages to tannic acids exhibited antimicrobial activity against Crono-
prevent the growth of organism has been investigated (Kim bacter spp. Amalaradjou et al. (2009) described the antimi-
et al., 2007a; Zuber et al., 2008). It is proposed that phages may crobial activity of trans-cinnamaldehyde (a component of
be specific, exhibiting lytic activity only with Cronobacter cells, bark extract obtained from cinnamon) on Cronobacter spp. in
thus eradicating the bacteria from PIF without damaging the reconstituted PIF. Trans-cinnamaldehyde reduced Crono-
normal flora when consumed. Hayes et al. (2006) assessed the bacter spp. to an undetectable level at a concentration of 0.5%
activities of antimicrobial peptides produced by Lactobacillus after 4-h incubation at 378C. Lee and Jin (2008) determined the
acidophilus against pathogenic bacteria-like Cronobacter spp. minimum inhibitory concentration values for five different
Two peptides were identified as having a similar antimicro- organic compounds. Carvacrol and thymol showed the
bial activity as isracidin. These were caseicins A and B. As strongest inhibition of Cronobacter isolates, with a minimum
these peptides are derived from milk proteins, they could inhibitory concentration of 1.25 mM. Studies such as these
possibly serve a bioprotective role in dairy-based food prod- highlight the use of natural organic compounds that can be
ucts such as PIF. More recently, the activity of a sodium ca- used as food additives to inhibit the growth of Cronobacter
seinate fermentate for reducing the numbers of Cronobacter spp. in foods.
spp. in reconstituted PIF was tested (Hayes et al., 2009). At
higher final concentrations (3.33%, wt=vol), numbers were
Future Direction
reduced from approximately 6 logs to 0 log CFU=mL over a
period of 60 min. Cronobacter spp. are a recently classified genus and much
Human milk has been found to inhibit the growth of some work has yet to be completed to better understand this unique
of the aetiological agents of neonatal infections, including group of organisms. Reliable detection and accurate identifi-
protozoans, viruses, and bacteria such as Staphylococcus, cation are paramount with respect to food and clinical en-
Group B Streptococcus, Escherichia coli and Cronobacter spp. vironments. Although molecular methods of detection are
(Chan, 2003). Previous studies have shown that fewer gas- generally faster compared with conventional microbiological=
trointestinal, respiratory, and meningeal infections occur in phenotype-based methods, the requirement for specialized
infants who are breast-fed compared with those who are equipments and operator training deters some sectors from
formula-fed (Cunningham et al., 1991; McGuire and An- implementing these protocols. Nonetheless, in the clinical
thony, 2003). Chan (2003) compared the antimicrobial ac- environment, rapid diagnostic protocols are essential. Al-
tivity of two commercially available milk fortifiers when though efforts have been made to update databases for
added to human milk. The results showed that fortifiers commercially available detection systems, improvements in
with high iron content inhibit the antimicrobial properties of the identification and differentiation of Cronobacter spp. could
human milk and that this was due to its effect on lactoferrin. be made.
Lactoferrin is an iron-binding protein found in human milk, The application of techniques such as comparative genome
which deprives the infectious organism of an essential analysis, transcriptomics, and proteomics, combined with
growth nutrient. It has a high binding affinity for iron but bioinformatics, are essential to unravel the complex interac-
only in an iron-deprived environment. Saturation of iron tions between pathogen and host and to identify new and
reduces the antimicrobial activity of lactoferrin. For this emerging properties in Cronobacter spp. Further, little is
reason it would be advisable to further investigate the ad- known about the in vivo virulence factors and pathogenicity of
dition of human milk fortifiers to inhibit the growth of Cronobacter spp., which are crucial in the design of therapies to
Cronobacter spp. treat and control infections. A better understanding of the
Collado et al. (2008) recently examined the possible ap- progression and pathogenesis of Cronobacter spp.-related
plication of probiotic strains to reduce the risk of a Crono- diseases, particularly using in vitro cell-based assays com-
bacter infection. Specifically, their study targeted the bined with animal model studies, is needed.
interaction between Cronobacter spp. and human intestinal As with many pathogenic organisms, Cronobacter spp.
mucus. The authors suggest that the addition of specific have emerged from a wider bacterial population through its
probiotics to PIF may be used to help prevent infection. ability to survive in a specific set of adverse environmental
Competitive exclusion is the primary mechanism for pre- conditions such as those encountered in PIF-processing fa-
venting pathogen attachment to the intestinal mucus. cilities. As the emergence of new pathogenic organisms in
However, results showed that the degree of adhesion of this way appears to be inevitable, it is paramount that, in
probiotic strains to intestinal mucus was not directly pro- future, improved food safety systems are based on the sci-
portional to the degree of Cronobacter spp. displacement. entific risk assessment to identify appropriate control and
Clearly other factors are therefore impacting on the estab- management solutions. Such systems can only be effective
lishment of Cronobacter spp. attachment and an under- when underpinned by data generated through the use of the
standing of these is critical in future studies to understand most up-to-date and sensitive microbiological methods. For
the interactions between pathogen and probiotic strains. this reason, much emphasis should be placed on the com-
The antimicrobial activities of several natural organic munication and dissemination of new technologies, proto-
compounds against Cronobacter spp. have been investigated. cols, and food safety issues to industry, policy makers,
Kim et al. (2009) examined the possible inactivation of Cro- researchers, and consumers.
nobacter spp. with water-soluble muscadine seed extracts. The
high phenolic content of the seed extracts may be increased
Disclosure Statement
further by heat treatment. Phenolic compounds have previ-
ously been shown to have antimicrobial activity against E. coli. No competing financial interests exist.
CRONOBACTER: AN OPPORTUNISTIC PATHOGEN 347

References neonatal nosocomial outbreaks. Diagn Microbiol Infect Dis

1990;13:467–472.
Adamson DH and Rodgers JR. Enterobacter sakazakii meningitis Collado MC, Isolauri E, and Salminen S. Specific probiotic
with sepsis. Clin Microbiol Newsl 1981;3:19–20. strains and their combinations counteract adhesion of En-
Almeida C, Azevedo NF, Iversen C, Fanning S, Keevil CW, terobacter sakazakii to intestinal mucus. FEMS Microbiol Lett
Vieira MJ. Development and application of a novel peptide 2008;285:58–64.
nucleic acid probe for the specific detection of Cronobacter Cunningham AS, Jelliffe DB, and Jelliffe EF. Breast-feeding and
(Enterobacter sakazakii) in powdered infant formula. Appl En- health in the 1980s: a global epidemiologic review. J Pediatr
viron Microbiol 2009;75:2925–2930. 1991;118:659–666.
Amalaradjou MA, Hoagland TA, and Venkitanarayanan K. In- Dancer GI, Mah JH, and Kang DH. Influences of milk compo-
activation of Enterobacter sakazakii in reconstituted infant for- nents on biofilm formation of Cronobacter spp. (Enterobacter
mula by trans-cinnamaldehyde. Int J Food Microbiol 2009;129: sakazakii). Lett Appl Microbiol 2009;48:718–725.
146–149. Edelson-Mammel SG, Porteous MK, and Buchanan RL. Survival
Anonymous. Milk and Milk Products—Detection of Enterobac- of Enterobacter sakazakii in a dehydrated powdered infant
ter sakazakii. Technical Specification ISO=TS 22964. ISO=TS formula. J Food Prot 2005;68:1900–1902.
22964:2006(E) and IDF=RM 210:2006(E), 1st edition. Geneva, El-Sharoud WM, O’Brien S, Negredo C, Iversen C, Fanning S, and
Switzerland: International Organization for Standardization, Healy B. Characterization of Cronobacter recovered from dried
2006. milk and related products. BMC Microbiol 2009;9:24.
Arseni A, Malamou-Ladas E, Koutsia C, Xanthou M, and Trikka Estuningsih S, Kress C, Hassan AA, Akineden O, Schneider E,
E. Outbreak of colonization of neonates with Enterobacter sa- and Usleber E. Enterobacteriaceae in dehydrated powdered
kazakii. J Hosp Infect 1987;9:143–150. infant formula manufactured in Indonesia and Malaysia.
Arts M. Enterobacter sakazakii in factories and households. Lancet J Food Prot 2006;69:3013–3017.
2004;364:414. [FAO=WHO] Food and Agriculture Organization=World
Asakura H, Morita-Ishihara T, Yamamoto S, and Igimi S. Ge- Health Organization. Enterobacter sakazakii and other micro-
netic characterization of thermal tolerance in Enterobacter sa- organisms in powdered infant formula: meeting report. MRA
kazakii. Microbiol Immunol 2007;51:671–677. Series 6. 2004. Available at www.who.int=foodsafety=micro=
Bar-Oz B, Preminger A, Peleg O, Block C, and Arad I. En- meetings=feb2004=en=, accessed February, 25, 2009. (Online.)
terobacter sakazakii infection in the newborn. Acta Paediatr [FAO=WHO] Food and Agriculture Organization=World Health
2001;90:356–358. Organization. Enterobacter sakazakii and Salmonella in pow-
Baumgartner A, Grand M, Liniger M, and Iversen C. Detection dered infant Formula. Second Risk Assessment Workshop:
and frequency of Cronobacter (Enterobacter sakazakii) in differ- Enterobacter sakazakii and Salmonella in powdered infant for-
ent categories of ready-to-eat foods other than infant formula. mula. Second Risk Assessment Workshop. Rome, Italy, 2006.
Int J Food Microbiol 2009 (in press). Available at www.fao.org=ag=agn=jemra=enterobacter_en.stm,
Beauchamp CJ, Simao-Beaunoir AM, Beaulieu C, and Chalifour accessed February 25, 2009. (Online.)
FP. Confirmation of E. coli among other thermotolerant coli- Farber JM. Enterobacter sakazakii-new foods for thought? Lancet
form bacteria in paper mill effluents, wood chips screening 2004;363:5–6.
rejects and paper sludges. Water Res 2006;40:2452–2462. Farmer JJIII, Hickmann AM, and Brenner DJ. Enterobacter saka-
Biering G, Karlsson S, Clark NC, Jonsdottir KE, Ludvigsson P, zakii: a new species of ‘‘Enterobacteriaceae’’ isolated from
and Steingrimsson O. Three cases of neonatal meningitis clinical specimens. Int J Syst Bacteriol 1980;30:569–584.
caused by Enterobacter sakazakii in powdered milk. J Clin Mi- Fox EM and Jordan KN. Towards a one-step Enterobacter saka-
crobiol 1989;27:2054–2056. zakii enrichment. J Appl Microbiol 2008;105:1091–1097.
Bowen AB and Braden CR. Invasive Enterobacter sakazakii disease Friedemann M. Enterobacter sakazakii in food and beverages
in infants. Emerg Infect Dis 2006;12:1185–1189. (other than infant formula and milk powder). Int J Food Mi-
Breeuwer P, Lardeau A, Peterz M, and Joosten HM. Desiccation crobiol 2007;16:1–10.
and heat tolerance of Enterobacter sakazakii. J Appl Microbiol Gallagher PG and Ball WS. Cerebral infarctions due to CNS
2003;95:967–973. infection with Enterobacter sakazakii. Pediatr Radiol 1991;21:
Caubilla-Barron J, Hurrell E, Townsend S, Cheetham P, Loc- 135–136.
Carrillo C, Fayet O, Prère MF, and Forsythe SJ. Genotypic and Giovannini M, Verduci E, Ghisleni D, Salvatici E, Riva E, and
phenotypic analysis of Enterobacter sakazakii strains from an Agostoni C. Enterobacter sakazakii: an emerging problem in
outbreak resulting in fatalities in a neonatal intensive care unit paediatric nutrition. J Int Med Res 2008;36:394–399.
in France. J Clin Microbiol 2007;45:3979–3985. Gosney MA, Martin MV, Wright AE, and Gallagher M. En-
Cawthorn DM and Witthuhn RC. Selective PCR detection of terobacter sakazakii in the mouths of stroke patients and its
viable Enterobacter sakazakii cells utilizing propidium mono- association with aspiration pneumonia. Eur J Int Med 2006;
azide or ethidium bromide monoazide. J Appl Microbiol 2008; 17:185–188.
105:1178–1185. Guillaume-Gentil O, Sonnard V, Kandhai MC, Marugg JD, and
Chan GM. Effects of powdered human milk fortifiers on the Joosten H. A simple and rapid cultural method for detection
antibacterial actions of human milk. J Perinatol 2003;23:620– of Enterobacter sakazakii in environmental samples. J Food Prot
623. 2005;68:64–69.
Chen Y, Hammack TS, Song KY, and Lampel KA. Evaluation Gurtler JB, Kornacki JL, and Beuchat LR. Enterobacter sakazakii: a
of a revised U.S. Food and Drug Administration for the detection coliform of increased concern to infant health. Int J Food Mi-
and isolation of Enterobacter sakazakii in powdered infant for- crobiol 2005;104:1–34.
mula: precollaborative study. J AOAC Int 2009;92:862–872. Hartmann I, Carranza P, Lehner A, Stephan R, Eberl L, and
Clark NC, Hill BC, O’Hara CM, Steingrimsson O, and Cooksey Riedel K. Genes involved in Cronobacter sakazakii biofilm for-
RC. Epidemiologic typing of Enterobacter sakazakii in two mation. Appl Environ Microbiol 2009.
348 HEALY ET AL.

Hawkins RE, Lissner CR, and Sanford JP. Enterobacter sakazakii Kim H, Bang J, Beuchat LR, and Ryu JH. Fate of Enterobacter
bacteremia in an adult. South Med J 1991;84:793–795. sakazakii attached to or in biofilms on stainless steel upon ex-
Hayes M, Barrett E, Ross RP, Fitzerald GF, Hill C, and Stanton C. posure to various temperatures or relative humidities. J Food
Evaluation of an antimicrobial ingredient prepared from a Prot 2008;71:940–945.
Lactobacillus acidophilus casein fermentate against Enterobacter Kim K, Klumpp J, and Loessner MJ. Enterobacter sakazakii bac-
sakazakii. J Food Prot 2009;72:340–346. teriophages can prevent bacterial growth in reconstituted in-
Hayes M, Ross RP, Fitzgerald GF, Hill C, and Stanton C. Casein- fant formula. Int J Food Microbiol 2007a;115:195–203.
derived antimicrobial peptides generated by Lactobacillus Kim SH and Park JH. Thermal resistance and inactivation of
acidophilus DPC6026. Appl Environ Microbiol 2006;72:2260– Enterobacter sakazakii isolates during rehydration of powdered
2264. infant formula. J Microbiol Biotechnol 2007;17:364–368.
Healy B, Mullane N, Collin V, Mailler S, Iversen C, Chatellier S, Kim H, Ryu JH, and Beuchat LR. Attachment of and biofilm
Storrs M, and Fanning S. Evaluation of an automated repeti- formation by Enterobacter sakazakii on stainless steel and en-
tive sequence-based PCR system for subtyping Enterobacter teral feeding tubes. Appl Environ Microbiol 2006;72:5846–
sakazakii. J Food Prot 2008;71:1273–1378. 5856.
Iversen C, Druggan P, and Forsythe S. A selective differential Kim H, Ryu JH, and Beuchat LR. Effectiveness of disinfectants in
medium for Enterobacter sakazakii, a preliminary study. Int J killing Enterobacter sakazakii in suspension, dried on the sur-
Food Microbiol 2004a;96:133–139. face of stainless steel, and in a biofilm. Appl Environ Micro-
Iversen C, Druggan P, Schumacher S, Lehner A, Feer C, biol 2007b;73:1256–1265.
Gschwend K, Joosten H, and Stephan R. Development of a Kim TJ, Silva JL, Weng WL, Chen WW, Corbitt M, Jung YS, and
novel screening method for the isolation of ‘‘Cronobacter’’ spp. Chen YS. Inactivation of Enterobacter sakazakii by water-soluble
(Enterobacter sakazakii). Appl Environ Microbiol 2008a;74:2550– muscadine seed extracts. Int J Food Microbiol 2009;129:295–
2553. 299.
Iversen C and Forsythe SJ. Comparison of media for the isola- Kothary MH, McCardell BA, Frazar CD, Deer D, and Tall BD.
tion of Enterobacter sakazakii. Appl Environ Microbiol 2007;73: Characterization of the zinc-containing metalloprotease en-
48–52. coded by zpx and development of a species-specific detection
Iversen C and Forsythe S. Isolation of Enterobacter sakazakii and method for Enterobacter sakazakii. Appl Environ Microbiol
other Enterobacteriaceae from powdered infant formula milk 2007;73:4142–4151.
and related products. Food Microbiol 2004;21:771–777. Kuhnert P, Korczak BM, Stephan R, Joosten H, and Iversen C.
Iversen C, Lane M, and Forsythe SJ. The growth profile, ther- Phylogeny and whole genome DNA-DNA similarity of En-
motolerance and biofilm formation of Enterobacter sakazakii terobacter and related taxa by multilocus sequence analysis
grown in infant formula milk. Lett Appl Microbiol 2004b; (MLSA). Int J Food Microbiol (in press).
38:378–382. Lai KK. Enterobacter sakazakii infections among neonates, infants,
Iversen C, Lehner A, Mullane N, Bidlas E, Cleenwerck I, Marugg J, children, and adults. Case reports and a review of the litera-
Fanning S, Stephan R, and Joosten H. The taxonomy of ture. Medicine (Baltimore) 2001;80:113–122.
Enterobacter sakazakii: proposal of a new genus Cronobacter gen. Lee SY and Jin HH. Inhibitory activity of natural antimicrobial
nov. and descriptions of Cronobacter sakazakii comb. nov. Cro- compounds alone or in combination with nisin against En-
nobacter sakazakii subsp. sakazakii, comb. nov., Cronobacter saka- terobacter sakazakii. Lett Appl Microbiol 2008;47:315–321.
zakii subsp. malonaticus subsp. nov., Cronobacter turicensis sp. Lehner A, Nitzsche S, Breeuwer P, Deip B, Thelen K, and
nov., Cronobacter muytjensii sp. nov., Cronobacter dublinensis sp. Stephan R. Comparison of two chromogenic media and
nov. and Cronobacter genomospecies 1. BMC Evol Biol 2007;7:64. evaluation of two molecular based identification systems for
Iversen C, Mullane N, McCardell B, Tall BD, Lehner A, Fanning Enterobacter sakazakii detection. BMC Microbiol 2006;6:15.
S, Stephan R, and Joosten H. Cronobacter gen. nov., a new Leuschner RG and Bew J. A medium for the presumptive de-
genus to accommodate the biogroups of Enterobacter sakazakii, tection of Enterobacter sakazakii in infant formula: inter-
and proposal of Cronobacter sakazakii gen. nov., comb. nov., laboratory study. J AOAC Int 2004;87:604–613.
Cronobacter malonaticus sp. nov., Cronobacter turicensis sp. nov., Lin LC and Beuchat LR. Survival of Enterobacter sakazakii in in-
Cronobacter muytjensii sp. nov., Cronobacter dublinensis sp. nov., fant cereal as affected by composition, water activity, and
Cronobacter genomospecies 1, and of three subspecies, Crono- temperature. Food Microbiol 2007;24:767–777.
bacter dublinensis subsp. dublinensis subsp. nov., Cronobacter Lindstedt BA, Varund T, and Kapperud G. Multiple-locus
dublinensis subsp. lausannensis subsp. nov. and Cronobacter variable-number tandem-repeats analysis of Escherichia coli
dublinensis subsp. lactaridi subsp. nov. Int J Syst Evol Microbiol O157 using PCR multiplexing and multi-colored capillary
2008b;58:1442–1447. electrophoresis. J Microbiol Methods 2004;58:213–222.
Iversen C, Waddington M, Farmer JJIII, and Forsythe SJ. The Liu Y, Cai X, Zhang X, Gao Q, Yang X, Zheng Z, Luo M, and
biochemical differentiation of Enterobacter sakazakii genotypes. Huang X. Real time PCR using TaqMan and SYBR Green for
BMC Microbiol 2006;6:94. detection of Enterobacter sakazakii in infant formula. J Microbiol
Jimenez E and Gimenez C. Septic shock due to Enterobacter sa- Methods 2006;65:21–31.
kazakii. Clin Microbiol Newsl 1982;4:30. Mange JP, Stephan R, Borel N, Wild P, Kim KS, Pospischil A,
Kandhai MC, Reij MW, Gorris LG, Guillaume-Gentil O, and van and Lehner A. Adhesive properties of Enterobacter sakazakii to
Schothorst M. Occurrence of Enterobacter sakazakii in food human epithelial and brain microvascular endothelial cells.
production environments and households. Lancet 2004;363: BMC Microbiol 2006;6:58.
39–40. McGuire W and Anthony MY. Donor human milk versus for-
Kilonzo-Nthenge A, Chen FC, and Godwin SL. Occurrence mula for preventing necrotizing enterocolitis in preterm in-
of Listeria and Enterobacteriaceae in domestic refrigerators. fants: systematic review. Arch Dis Child Fetal Neonatal Ed
J Food Prot 2008;71:608–612. 2003;88:11–14.
CRONOBACTER: AN OPPORTUNISTIC PATHOGEN 349

Mittal R, Wang Y, Hunter CJ, Gonzalez-Gomez I, and Prasa- Oh SW and Kang DH. Fluorogenic selective and differential
darao NV. Brain damage in newborn rat model of meningitis medium for isolation of Enterobacter sakazakii. Appl Environ
by Enterobacter sakazakii: a role for outer membrane protein A. Microbiol 2004;70:5692–5694.
Lab Invest 2009;89:263–277. Pagotto FJ, Nazarowec-White M, Bidawid S, and Farber JM.
Mohan Nair MK and Venkitanarayanan KS. Cloning and se- Enterobacter sakazakii: infectivity and enterotoxin production
quencing of the ompA gene of Enterobacter sakazakii and de- in vitro and in vivo. J Food Prot 2003;66:370–375.
velopment of an ompA-targeted PCR for rapid detection of Piddock LJ. Multidrug-resistance efflux pumps—not just for
Enterobacter sakazakii in infant formula. Appl Environ Micro- resistance. Nat Rev Microbiol 2006;4:629–636.
biol 2006;72:2539–2546. Proudy I, Bougle D, Coton E, Coton M, Leclercq R, and Verg-
Mohan Nair MK and Venkitanarayanan KS. Role of bacterial naud M. Genotypic characterization of Enterobacter sakazakii
OmpA and host cytoskeleton in the invasion of human in- isolates by PFGE, BOX-PCR and sequencing of the fliC gene.
testinal epithelial cells by Enterobacter sakazakii. Pediatr Res J Appl Microbiol 2008;104:26–34.
2007;62:664–669. Raghav M and Aggarwal PK. Purification and characterization
Mohan Nair MK, Venkitanarayanan K, Silbart LK, and Kim KS. of Enterobacter sakazakii enterotoxin. Can J Microbiol 2007;53:
Outer membrane protein A (OmpA) of Enterobacter sakazakii 750–755.
binds fibronectin and contributes to invasion of human brain Redmond E and Griffith CJ. Disinfection methods used in de-
microvascular endothelial cells. Foodborne Pathog Dis 2009;6: contamination of bottles used for feeding powdered infant
495–501. formula. J Fam Health Care 2009;19:26–31.
Mullane N, Healy B, Meade J, Whyte P, Wall P, and Fanning S. Restaino L, Frampton EW, Lionberg WC, and Becker RJ.
Dissemination of Cronobacter (Enterobacter sakazakii) in a A chromogenic plating medium for the isolation and
powdered milk protein manufacturing facility. Appl Environ identification of Enterobacter sakazakii from foods, food ingre-
Microbiol 2008a;74:5913–5917. dients, and environmental sources. J Food Prot 2006;69:315–
Mullane NR, Iversen C, Healy B, Walsh C, Whyte P, Wall PG, 322.
Quinn T, and Fanning S. Enterobacter sakazakii an emerging Riedel K and Lehner A. Identification of proteins involved in
bacterial pathogen with implications for infant health. Mi- osmotic stress response in Enterobacter sakazakii by proteomics.
nerva Pediatr 2007a;59:137–148. Proteomics 2007;7:1217–1231.
Mullane NR, Ryan M, Iversen C, Murphy M, O’Gaora P, Quinn Scheepe-Leberkühne M and Wagner F. Optimization and pre-
T, Whyte P, Wall PG, and Fanning S. Development of liminary characterisation of an exopolysaccharide synthesized
multiple-locus variable-number tandem-repeat analysis for by Enterobacter sakazakii. Biotechnol Lett 1986;8:695–700.
the molecular subtyping of Enterobacter sakazakii. Appl En- See KC, Than HA, and Tang T. Enterobacter sakazakii bacteraemia
viron Microbiol 2008b;74:1223–1231. with multiple splenic abscesses in a 75-year-old woman: a case
Mullane NR, Whyte P, Wall PG, Quinn T, and Fanning S. Ap- report. Age Ageing 2007;36:595–596.
plication of pulsed-field gel electrophoresis to characterise Seo KH and Brackett RE. Rapid, specific detection of Enterobacter
and trace the prevalence of Enterobacter sakazakii in an infant sakazakii in infant formula using a real-time PCR assay. J Food
formula processing facility. Int J Food Microbiol 2007b;116:73– Prot 2005;68:59–63.
81. Simmons BP, Gelfand MS, Haas M, Metts L, and Ferguson J.
Muytjens HL and van der Ros-van de Repe J. Comparative Enterobacter sakazakii infections in neonates associated with
in vitro susceptibilities of eight Enterobacter species, with special intrinsic contamination of a powdered infant formula. Infect
reference to Enterobacter sakazakii. Antimicrob Agents Che- Control Hosp Epidemiol 1989;10:398–401.
mother 1986;29:367–370. Singamsetty VK, Wang Y, Shimada H, and Prasadarao NV.
Muytjens HL, Roelofs-Willemse H, and Jaspar GH. Quality of Outer membrane protein A expression in Enterobacter sakazakii
powdered substitutes for breast milk with regard to members is required to induce microtubule condensation in human
of the family Enterobacteriaceae. J Clin Microbiol 1988;26:743– brain microvascular endothelial cells for invasion. Microb
746. Pathog 2008;45:181–191.
Muytjens HL, van der Ros-van de Repe J, and van Druten HA. Stock I and Wiedemann B. Natural antibiotic susceptibility of
Enzymatic profiles of Enterobacter sakazakii and related species Enterobacter amnigenus, Enterobacter cancerogenus, Enterobacter
with special reference to the alpha-glucosidase reaction and gergoviae and Enterobacter sakazakii strains. Clin Microbiol In-
reproducibility of the test system. J Clin Microbiol 1984;20: fect 2002;8:564–578.
684–686. Stoop B, Lehner A, Iversen C, Fanning S, and Stephan R.
Nazarowec-White M and Farber JM. Phenotypic and genotypic Development and evaluation of rpoB-based PCR systems to
typing of food and clinical isolates of Enterobacter sakazakii. differentiate the six proposed species within the genus Cro-
J Med Microbiol 1999;48:559–567. nobacter. Int J Food Microbiol 2009 (in press).
Nazarowec-White M and Farber JM. Thermal resistance of En- Torpdahl M, Sorensen G, Ethelberg S, Sando G, Gammelgard K,
terobacter sakazakii in reconstituted dried-infant formula. Lett and Porsbo LJ. A regional outbreak of S. Typhimurium in
Appl Microbiol 1997;24:9–13. Denmark and identification of the source using MLVA typing.
O’Brien S, Healy B, Negredo C, Anderson W, Fanning S, and Eur Surveill 2006;11:134–136.
Iversen C. Prevalence of Cronobacter species (Enterobacter sa- Touze T, Eswaran J, Bokma E, Koronakis E, Hughes C, and
kazakii) in follow-on infant formulae and infant drinks. Lett Koronakis V. Interactions underlying assembly of the Escher-
Appl Microbiol 2009a;48:536–541. ichia coli AcrAB-TolC multidrug efflux system. Mol Microbiol
O’Brien S, Healy B, Negredo C, Fanning S, and Iversen C. 2004;53:697–706.
Evaluation of enrichment methods for the detection of Cro- Townsend S, Hurrell E, and Forsythe S. Virulence studies of
nobacter spp. (Enterobacter sakazakii) from powdered infant Enterobacter sakazakii isolates associated with a neonatal in-
formula. J Food Prot 2009b (in press). tensive care unit outbreak. BMC Microbiol 2008;8:64.
350 HEALY ET AL.

Townsend S, Hurrell E, Gonzalez-Gomez I, Lowe J, Frye JG, Ziegler DR. Gene sequences useful for predicting relatedness of
Forsythe S, and Badger JL. Enterobacter sakazakii invades brain whole genome in bacteria. Int J Syst Evol Microbiol 2003;53:
capillary endothelial cells, persists in human macrophages 1893–1900.
influencing cytokine secretion and induces severe brain pa- Zuber S, Boissin-Delaporte C, Michot L, Iversen C, Diep B,
thology in the neonatal rat. Microbiology 2007;153:3538–3547. Brüssow H, Breeuwer P. Decreasing Enterobacter sakazakii
Urmenyi AM and Franklin AW. Neonatal death from pigmented (Cronobacter spp.) food contamination level with bacterio-
coliform infection. Lancet 1961;1:313–315. phages: prospects and problems. Microb Biotechnol 2008;1:
van Acker J, de Smet F, Muyldermans G, Bougatef A, Naessens 532–543.
A, and Lauwers S. Outbreak of necrotizing enterocolitis as-
sociated with Enterobacter sakazakii in powdered milk formula. Address correspondence to:
J Clin Microbiol 2001;39:293–297. Séamus Fanning, B.Sc., Ph.D.
van Belkum A. Tracing isolates of bacterial species by multilocus Centres for Food Safety and Foodborne Zoonomics
variable number of tandem repeat analysis (MLVA). FEMS School of Agriculture, Food Science, and Veterinary Medicine
Immunol Med Microbiol 2007;49:22–27. University College Dublin
Weir E. Powdered infant formula and fatal infection with En- Belfield, Dublin 4
terobacter sakazakii. CMAJ 2002;166:1570. Ireland
Willis J and Robinson JE. Enterobacter sakazakii meningitis in
neonates. Pediatr Infect Dis J 1988;7:196–199. E-mail: [email protected]