Journal of Food and Nutrition Research Vol. 50, 2011, No. 2, pp.

118–124

A method for the detection of Cronobacter strains

in powdered milk-based foods using enrichment and real-time PCR
EVA KACLÍKOVÁ – IMRICH TURCOVSKÝ

Summary
Cronobacter spp. are emerging food-borne pathogens implicated in fatal infections in neonates and infants. As the
contaminated powdered infant formula (PIF) has been identified as the major source of infection, a zero tolerance of
Cronobacter spp. is required for these products. The aim of our study was to evaluate real-time PCR-based method for
Cronobacter detection in milk powder-based products. TaqMan real-time PCR targeting the dnaG gene of MMS-operon
was used. The system was evaluated as 100% specific for Cronobacter spp., which was determined using 97 Cronobacter
strains and 85 non-Cronobacter strains with a PCR detection limit of 1  102 CFU·ml-1. Shortened two-step enrichment
was used. Out of 50 milk powder-based samples analysed, six were positive by the PCR-based method and five by the
standard method. Out of 15 artificially contaminated PIF with Cronobacter at 100 CFU per 10 g, 15 and 13 were positive
by respective methods. Low detection limit of the complete method was not compromised either by Enterobacteriaceae
background up to 107 CFU per 10 g, or by dryness and cold stress of Cronobacter cells used for artificial contamination.
This method facilitated a reliable detection of Cronobacter spp. alternative to the currently available method, providing
a considerable time reduction in comparison to the microbiological standard detection method.

Keywords
pathogenic Cronobacter spp.; real-time PCR; detection; powdered milk food

Cronobacter spp. (formerly Enterobacter saka- rily in low birth weight and immunocompromised
zakii) is a bacterial genus within the Enterobacte- neonates with high mortality rate of 50–80% [5, 6].
riaceae family. The bacterium was called „yellow Whereas Cronobacter spp. is widespread in the
pigmented Enterobacter cloacae“ until 1980 [1], environment and has been isolated from various
when it was renamed to Enterobacter sakazakii. food and environmental samples [7, 8], the prima-
The World Health Organization and the Food and ry source of the organism and the main vehicle for
Agriculture Organization categorized E. sakaza- its transmission in neonatal infections was found
kii together with Salmonella enterica as potencially to be the rehydrated powdered infant formula
dangerous contaminants in powdered infant for- [9]. Cronobacter spp. have been found in samples
mula in 2004 [2]. Recently, E. sakazakii has been of infant formula powders at very low levels and
re-classified to six species within the Cronobacter it was demonstrated that these low numbers (less
genus containing C. sakazakii, C. malonaticus, than 1 CFU per 100 g of the infant formula) may
C. muytjensii, C. dublinensis, C. turicensis and be responsible for infection [10]. Post-process con-
Cronobacter genomospecies I [3]. All species have tamination of powdered infant formula is consid-
been linked retrospectively to clinical cases of ered a likely source of Cronobacter spp., since the
infection in either infants or adults and therefore usual milk pasteurization treatment in the produc-
all Cronobacter spp. should be considered patho- tion process is adequate to destroy the bacterium.
genic [4]. However, some infant products with added ingre-
Cronobacter spp. cause a rare but fatal infec- dients may be produced via dry blending, without
tion manifested as meningitis, meningoencepha- heat treatment, while Cronobacter spp. has been
litis, sepsis and necrotizing enterocolitis prima- isolated from wide variety of production facilities

Eva Kaclíková Eva, Imrich Turcovský, Department of Microbiology and Molecular Biology, VÚP Food Research Institute,
Priemyselná 4, P. O. Box 25, SK – 82475 Bratislava 26, Slovakia.
Correspondence author:
Eva Kaclíková, e-mail: [email protected], tel.: +421-2-50237159, fax: +421-2-55571417

118 © 2011 VÚP Food Research Institute, Bratislava

 Cronobacter PCR detection in powdered foods

[11]. Moreover, the presence in dried foods and cially contaminated powdered milk-based infant
powdered ingredients may be due to the increased foods and evaluated by comparing real-time PCR-
desiccation resistance of Cronobacter strains [12]. based detection with the standard one.
According to the currently valid European
Commission Regulation No. 2073 [13], a zero
tolerance of Cronobacter presence is required for MATERIALS AND METHODS
powdered infant formula and powdered dietary
foods for infants up to 6 months of age. Therefore, Bacterial strains and culture conditions
the reliable detection method is necessary to con- Cronobacter spp. (Enterobacter sakazakii)
trol the microbiological quality of final products strains used in this study were obtained from
and ingredients used in powdered infant formula American Type Culture Collection (ATCC,
(PIF) production as well. Currently available Manassas, Virginia, USA), Belgian Co-ordinated
standard method ISO/TS 22964 [14] for the de- Collections of Microorganisms in Gent (BCCM/
tection of Cronobacter spp. (E. sakazakii) is based LMG Bacteria Collection) and from Czech Col-
on two-step enrichment including non-selective lection of Microorganisms (CCM, Brno, Czech
enrichment for 18 h ± 2 h followed by selective Republic). Five strains isolated from fruit pow-
enrichment for 24 h, isolation of presumptive co- der were kindly provided by Prof. Stephan from
lonies on chromogenic medium and several con- the Institute of Food Safety and Hygiene (ILS),
firmation tests. The procedure is time-consuming Vetsuisse Faculty University of Zurich, Switzer-
and labour-intensive, and the biochemical confir- land. Most of used Cronobacter strains were iso-
mation need not always provide reliable results. lated from various food samples from the market
The confirmation of presumptive Cronobacter in Slovakia and identified in our laboratory [23].
colonies selected from chromogenic medium uti- Numbers of strains belonging to different Crono-
lizes the detection of the production of a yellow bacter species used in this study are summarized
pigment and several biochemical tests. This iden- in Tab. 1. Bacterial strains other than Cronobacter
tification may not be sufficiently reliable as not all were obtained from culture collections or from
Cronobacter strains produce the yellow pigment
[15]. This procedure is considerably labourious
and moreover typical colonies on the chromogenic Tab. 1. Results of dnaG-targeted real-time PCR
media are formed also by other bacterial species in inclusivity on a panel of 97 Cronobacter strains.
addition to Cronobacter spp. [16]. This procedure
Strain used
requires up to six days to confirm positive results. Number PCR
Cronobacter spp. for artificial
of strains result
Therefore, there is a strong need for more contamination
rapid and specific methods for detection and C. sakazakii 73 + ATCC 29544
identification of Cronobacter spp. During the last
C. malonaticus 10 + LMG 23826
few years, several PCR-based methods have been
C. dublinensis 6 + LMG 23823
developed that enable identification of Crono-
bacter spp. to the genus and species level. Real- C. muytjensii 5 + ATCC 51329
time PCR-based methods provide the powerful C. turicensis 3 + LMG 23827
tool for highly specific and sensitive identification
and are considered reliable alternatives to conven-
tional methods for food control purposes. Real- Tab. 2. Results of dnaG real-time PCR
time PCR methods for Cronobacter spp. (E. saka- of 85 non-Cronobacter strains for exclusivity testing.
zakii) detection utilizing various targets have been Genus Number of strains PCR result
developed recently [17–20] and subsequently Enterobacter 13 –
evaluated as having various level of specificity [7].
Campylobacter 5 –
PCR-based detection of Cronobacter strains tar-
Citrobacter 16 –
geting dnaG gene of MMS-operon developed by
SEO and BRACKETT [17] has been widely evaluated Edwardsiella 1 –
and confirmed to be efficient, the molecular target Escherichia 10 –
being found to be reliable [21, 22]. Klebsiella 1 –
In this study, we describe the rapid and sen- Proteus 1 –
sitive Cronobater spp. detection method using
Salmonella 27 –
TaqMan real-time PCR targeted to dnaG gene,
suitable for routine application. The developed Serratia 1 –
method has been applied to naturally and artifi- Yersinia 10 –

119
Kaclíková, E. – Turcovský, I. J. Food Nutr. Res., 50, 2011, pp. 118–124

reference laboratories (Tab. 2). Five selected of C. sakazakii ATCC 29544 and C. muytjensii
strains, one of each species (Tab. 1), were used ATCC 51329 overnight cultures were performed,
for artificial contamination of food samples. All respectively.
strains were maintained at –18 °C in 20% glycerol
solution or freeze-dried for long-period storage. Enrichment
Working cultures were prepared by inoculation Two-step enrichment of 10 g sample consisting
in Brain Heart (BH) broth (Merck, Darmstadt, of the pre-enrichment in 90 ml of Buffered Pep-
Germany) and incubation for 20 h ± 1 h at 37 °C. tone Water (Merck) at 37 °C for 18 h ± 1 h and
Decimal dilutions of the culture were prepared in selective enrichment (1/100) at 45 °C in modified
0.85% NaCl and the cell concentrations were de- Lauryl Sulfate Tryptose (mLST) broth (Merck)
termined by the plate-count technique on BH agar with additional NaCl (Merck) to the final con-
(Merck) incubated overnight at 37 °C. centration of 0.5 mol·l-1 and vancomycin 10 mg·l-1
(Fluka, Buchs, Switzerland) was used. The se-
Preparation of DNA samples lective culture was followed by the PCR and the
and 5’-nuclease real-time PCR standard Cronobacter detection after 5 h and 24 h
DNA samples were prepared from overnight of enrichment.
cultures in BH broth (Merck) using lysis by boil-
ing, as described previously [24]. DNA from Food samples and artificial contamination
Cronobacter sakazakii type strain ATCC 29544 to Samples of milk powder-based foods and in-
be used as PCR positive control was extracted us- fant formula used in this study were obtained from
ing QiaAmp DNA Mini Kit (Qiagen, Hilden, Ger- retail markets and drug stores in Slovakia. All
many). Each PCR reaction contained 200 nmol·l-1 samples were analysed for the presence of Crono-
of primers, 200 nmol·l-1 of the TaqMan probe bacter spp. after enrichment using microbiological
(all from Qiagen Operon), 500 mmol·l-1 of each detection according to ISO/TS 22964 and the real-
dNTP (Applied Biosystems, Foster City, Califor- time PCR-based detection method. For artificial
nia, USA), 1.5 U of HotStar Taq DNA polymer- contamination, five different Cronobacter species
ase (Qiagen), 1 × concentrated PCR buffer, listed in Tab. 1 were inoculated at two initial con-
4.5 mmol·l-1 magnesium chloride, TaqMan Exo- centration levels of 100 and 101 CFU per 10 g sam-
genous Internal Positive control VIC (Applied ple and 90 ml of pre-enrichment medium, respec-
Biosystems), 2.5 ml of the DNA sample and water tively. Simultaneously, stressed cells at the same
to make the total volume up to 25 ml. Real-time concentration levels were inoculated. The stressed
PCR was performed in a PTC-200 thermal cycler cells were prepared by dessication of culture dilu-
coupled to a Chromo 4 continuous fluorescence tions applied on the filter paper in air-ventilated
detector (MJ Research, Waltham, Massachusetts, oven at 45 °C followed by 6 h freezing at –18 °C
USA) using a thermal programme consisting of under aseptic conditions. Model Enterobacteria-
the initial denaturation of 15 min at 95 °C, and ceae microflora consisting of Enterobacter cloacae
45 cycles of 15 s at 95 °C and 60 s at 60 °C. Two CCM 1903, Citrobacter braakii CCM 3393, Es-
positive and three negative controls were included cherichia vulneris CCM 3681, Salmonella Enteri-
in each experiment. tidis CCM 4420 strains at levels of 108, 107 and 106
CFU per 100 ml of medium was used as the back-
Determination of PCR detection limit ground microflora. It was prepared from decimally
and detection probability diluted mixture of overnight cultures of individual
In order to determine the DNA-based PCR de- strains at proportional concentrations.
tection limit, DNA from the overnight culture of
type strain C. sakazakii ATCC 29544 was isolated Reference method
using QiaAmp DNA Mini Kit. Concentration The method according to ISO/TS 22964 Milk
of total extracted DNA was determined by using and milk products – Detection of Enterobacter
Quant-iT PicoGreen Assay (Invitrogen, Carlsbad, sakazakii [14] was used as the reference one.
California, USA) with fluorescence measurement Cronobacter strains were detected in natural and
in a Tecan Saphire2 plate reader (Salzburg, Aus- artificialy contaminated food samples using the
tria). Two parallel PCR analyses of each diluted two-step enrichment procedure described above.
sample DNA as a template were performed. Presumptive Cronobacter strains were selected on
For the determination of the practical PCR de- Chromogenic Enterobacter sakazakii agar (DFI
tection limit and PCR detection probability, two formulation; Oxoid, Basingstoke, United King-
independent sets of parallel PCR analyses of suit- dom) as green-blue colonies. Selected colonies
ably diluted (ranging from 104 to 100 CFU·ml-1) were confirmed on the basis of the yellow pigment

120
 Cronobacter PCR detection in powdered foods

production when exposed to light during incuba- one genome copy calculated based on C. sakazakii
tion on Tryptone Soya Agar (Oxoid) with bile salts genome size of 4.37 Mbp  4.8 fg, (database Gen-
(Fluka). Identification was confirmed by profile Bank, accession No. NC 009778, National Center
determination using API 20E system (bioMérieux, for Biotechnology Information, Bethesda, Mary-
Marcy l’Etoile, France). land, USA).
Practical PCR detection limit was estimated
using lysed cells from decimally diluted overnight
cultures of two Cronobacter strains (C. sakazakii
RESULTS AND DISCUSSION ATCC 29544 and C. muytjensii ATCC 51329). The
results of PCR analyses showed that the detec-
Evaluation of dnaG real-time PCR specificity tion probability of the cell suspension was 100%
The specificity of the real-time PCR detection at a concentration of (1.0  0.2) 102 CFU·ml-1
system targeting dnaG gene of MMS operon de- (equal to 2.5 CFU per reaction) and at higher con-
veloped by SEO and BRACKETT [17] was evaluated centrations for both strains, and 80% for C. saka-
using 97 Cronobacter spp. including 70 well-identi- zakii and 70% for C. mutjensii at a concentration
fied isolates from our laboratory [23] resulting in of (0.5  0.1) 102 CFU·ml-1 (equal to 1.25 CFU
the determination of the value of 100% inclusiv- per reaction; data not shown).
ity, and 85 non-Cronobacter strains of Enterobacte- Comparable or higher detection limits using
riaceae family resulting in in the determination of the same PCR system were obtained by other re-
the value of 100% exclusivity. The results of dnaG search groups. Detection limit from 102 CFU·ml-1
real-time PCR for all strains are summarized in to 103 CFU·ml-1 was determined using real-time
Tab. 1 and Tab. 2. The oligonucleotide sequences PCR and different target sequences for Crono-
of the primers and probe used in the study are pre- bacter detection [18, 20]. Conventional PCR pro-
sented in Tab. 3. vided the detection at a higher level of more than
Several PCR methods based on different tar- 103 CFU·ml-1 [26].
get sequences were developed and subsequently For simultaneous detection of Cronobacter us-
validated for their specificity by other researchers. ing PCR according to SEO and BRACKETT [17] and
CAWTHORN et al. [25] valuated the PCR amplifica- Salmonella spp., a detection limit of 103 CFU·ml-1
tions using six different genus-specific primer pairs was determined for both targets without enrich-
targeting 16S rRNA gene, gene responsible for ment [27].
-glucosidase activity and ITS sequence between Based on the results obtained in this study, the
16S and 23S rDNA published previously. The spe- Cronobacter concentration of at least 102 CFU·ml-1
cificity was determined to range from 8% to 92%. should be reached by the enrichment procedure to
JARADAT et al. [8] used eight sets of previously ensure the reliable detection by downstream real-
published Cronobacter spp.-specific primers tar- time PCR.
geting ITS sequences, 16S rRNA, ompA, zpx, gluA
and gluB genes and none of the methods proved Evaluation of the enrichment procedure
to be a reliable method for the identification of with a shortened second step
Cronobacter isolates being free of false positives or The time reduction of the enrichment proce-
false negatives. dure was evaluated in an attempt to obtain results
On the other hand, the sequence of dnaG on the next day after the sample reception. Five
gene of MMS operon used in our study was con- different Cronobacter species (Tab. 1) at initial
sidered 100% specific for Cronobacter strains by concentrations of 101 and 100 CFU were inoculat-
other researchers as well [21, 22], without obtain- ed to two-step enrichment described above. Selec-
ing any false-positive or false-negative results. The tive enrichment of 5 h and 24 h, respectively, was
described PCR method provided reliable Crono- followed by real-time PCR detection and standard
bacter identification. detection according to ISO/TS 22964 was used as
the reference method. For both levels of contami-
PCR detection limit nation, the selective enrichment of 5 h was found
DNA-based detection limit was determined to be sufficient for Cronobacter detection using
using DNA isolated from C. sakazakii ATCC real-time PCR (data not shown).
29544 diluted to 100 fg, 50 fg, 10 fg, 7.5 fg, 5 fg, Similar evaluation was performed with added
2,5 fg, 1 fg, 0.5 fg and 0.1 fg. The DNA amount of artificial background microflora of Enterobacte-
 5 fg in PCR reaction was detected in 10 paral- riaceae strains. PCR provided more sensitive de-
lel analyses with 100% detection probability (data tection in the case of model Enterobacteriaceae
not shown). This is equivalent to approximately background microflora up to 107 CFU added

121
Kaclíková, E. – Turcovský, I. J. Food Nutr. Res., 50, 2011, pp. 118–124

Tab. 3. Oligonucleotides used in the study (5’-3’).

Designation Sequence and labelling
EsMMSf GGGATATTGTCCCCTGAAACAG
EsMMSr CGAGAATAAGCCGCGCATT
EsMMSp FAM-AGAGTAGTTGTAGAGGCCGTGCTTCCGAAAG-TAMRA

Tab. 4. Results of real-time PCR detection of C. sakazakii ATCC 29544 at two contamination levels
in combination with three concentration levels of background microflora.
C. sakazakii ATCC 29544 (CFU per 100 ml)
Enterobacteriaceae
background microflora Result of PCR detection Result of standard detection
(CFU per 100 ml)
101 100 101 100
– + + + +
106 + + + +
107 + + + –
108 + – + –

to the sample, in comparison to 106 CFU using Detection of Cronobacter spp. in food samples
standard method. However, at the extreme load of The two step enrichment with reduced selec-
108 CFU, the detection of 100 CFU of C. sakazakii tive enrichment of 5 h-duration and real-time
per sample completely failed using both methods. PCR detection was compared with the reference
The results for the detection of C. sakazakii detection method. Fifty samples of milk powder-
ATCC 25944 at contamination levels of 100 CFU based foods, including 20 samples of PIF, were an-
and 101 CFU per 100 ml culture, in combination alysed. Six samples were found positive for Crono-
with the background microflora of 106, 107 and 108 bacter spp. by the real-time PCR-based method
CFU of Enterobacteriaceae mixture, are presented and five samples by the standard detection method
in Tab. 4. (Tab. 5). Compared to real-time PCR-based detec-
Although new alternative selective media were tion, one sample of the infant food was false-nega-
developed for better recovery of Cronobacter tive using the standard detection method.
strains isolated from foods, we used the two-step The efficiency of the developed real-time
enrichment according to the currently valid refer- PCR-based detection method was further evalu-
ence method in order to compare the efficiency of ated using 15 samples artificially contaminated
microbiological and real-time PCR-based detec- with C. sakazakii ATCC 25944 at two concentra-
tion of Cronobacter spp. in milk powder-based in- tion levels. From 15 samples of PIF artificially
fant foods. contaminated at a level of 101 CFU per 10 g, all

Tab. 5. Results of Cronobacter spp. detection in milk powder-based foods.

Number Number of positive samples

Food sample
of samples PCR detection Reference method
Powdered infant formula (up to 6 months) 20 0 0
Infant food with rice 5 0 0
Infant food with fruit 5 0 0
Infant food with cocoa 5 1 0
Hot chocolate 5 2 2
Milk shake 5 2 2
Milk powder 5 1 1

122
 Cronobacter PCR detection in powdered foods

Tab. 6. Results of Cronobacter spp. detection in powdered infant formulae

artificially contaminated at two levels of 101 and 100 CFU per 10 g.
Number of positive samples
Number
Artificial contamination by PCR by standard method
of samples
101 100 101 100
Unstresed cells 15 15 15 15 13
Stressed cells 15 15 15 15 11

samples were detected by both methods. From based method may be more sensitive than the con-
15 samples contaminated at a level of 100 CFU per ventional culture method. The developed method
10 g, 15 were detected by real time PCR and 13 by can be considered as a faster and reliable alterna-
standard method. When stressed Cronobacter cells tive for Cronobacter detection.
were used for artificial contamination of 15 PIF
samples, 15 were detected by real-time PCR-based ACKNOWLEDGEMENTS
method and 10 by the standard one (Tab. 6). This work was supported by Slovak Research and
Insufficient detection ability using the standard Development Agency under the contract No. APVV-
method was observed also by other investigators. 27-009705 and by Slovakian Department of Agriculture
in frames of the project “Development of progres-
DERZELLE and DILASSER [20] analysed 41 sam-
sive methods and procedures to support quality and
ples of infant formulae and samples from pro- safety of food production and control”, contract
duction environment for the presence of E. saka- No. 2006UO27/08W0301/08W0309.
zakii. Using the standard enrichment procedure,
23 samples were positive by real-time PCR and
22 by the standard method, giving the concord-
ance of 97.5%. REFERENCES

1. Farmer, J. J. III – Asbury, M. A. – Hickman, F.

W. – Brenner, D. J. and the Enterobacteriaceae
CONCLUSION study group: Enterobacter sakazakii: A new spe-
cies of „Enterobacteriaceae“ isolated from clini-
Currently, the standard method available for cal specimens. International Journal of Systematic
Cronobacter spp. (Enterobacter sakazakii) de- Bacteriology, 30, 1980, pp. 569–584.
tection in milk and milk products according to 2. Enterobacter sakazakii and other microorganisms
ISO/IDF 22964 is based on two-step enrichment in powdered infant formula : Meeting report.
followed by microbiological and biochemical iden- Microbiological Risk Assessment Series No. 6.
tification. This method is laborious as well as time- Rome : Food and Agriculture Organization, 2004.
consuming, it takes up to six days to confirm posi- 80 pp. ISBN 978-92-5-105164-1. Available also at
tive results. Generally, molecular-based methods <http://www.fao.org/docrep/007/y5502e/y5502e00.
htm>
have a well-grounded potential to overcome the
3. Iversen, C. – Mullane, N. – McCardell, B. –
insufficiences of identification procedures asso- Tall, B. D. – Lehner, A. – Fanning, S. – Stephan, R. –
ciated with the results based on the biochemical Joosten, H.: Cronobacter gen. nov., a new genus to
characteristics. The aim of our study was to de- accommodate the biogroups of Enterobacter saka-
velop and optimize sensitive and reliable method zakii, and proposal of Cronobacter sakazakii gen
for the rapid and definite detection of Crono- nov., comb. nov., Cronobacter malonaticus sp.
bacter spp. in foods, particularly milk-powder nov., Cronobacter turicensis sp. nov., Cronobacter
based infant foods. The developed real-time PCR- muytjensii sp. nov., Cronobacter dublinensis sp. nov.,
based method, which involves an optimized short- Cronobacter genomospecies 1 and three subspe-
cies, Cronobacter dublinensis subsp. dublinensis
ened selective enrichment, facilitated the detec-
subsp. nov., Cronobacter dublinensis subsp. lausan-
tion of Cronobacter spp. on the next day after the nensis subsp. nov. and Cronobacter dublinensis
sample reception. The method is fully effective subsp. lactaridi subsp. nov. International Journal of
also in the presence of background Enterobacte- Systematic and Evolutionary Microbiology, 58, 2008,
riaceae strains in concentrations up to 107 CFU pp. 1442–1447.
per sample. Moreover, our results suggest that, 4. Enterobacter sakazakii (Cronobacter spp.) in
besides being significantly faster, the molecular- powdered follow-up formula : Meeting report.

123
Kaclíková, E. – Turcovský, I. J. Food Nutr. Res., 50, 2011, pp. 118–124

Microbiological Risk Assessment Series No. 15. Environmental Microbiology, 73, 2007, pp. 48–52.
Rome : Food and Agriculture Organization,, 2008. 17. Seo, K. H. – Brackett, R. E.: Rapid, specific detec-
104 pp. ISBN 978-92-5-106119-0. Available also at tion of Enterobacter sakazakii in infant formula using
<ftp://ftp.fao.org/docrep/fao/011/i0453e/i0453e00. a real-time PCR. Journal of Food Protection, 68,
pdf>. 2005, pp. 59–63.
5. Lai, K. K.: Enterobacter sakazakii infections among 18. Malorny, B. – Wagner, M.: Detection of Enterobacter
neonates, infants, children, and adults: case reports sakazakii strains by real-time PCR. Journal of Food
and a review of the literature. Medicine, 80, 2001, Protection, 68, 2005, pp. 1623–1627.
pp. 113–122. 19. Liu, Y. – Cai, X. – Zhang, X. – Gao, Q. – Yang, X. –
6. Mullane, N. R. – Iversen, C. – Healy, B. – Walsh, Zheng, Z. – Luo, M. – Huang, X.: Real-time PCR
C. – Whyte, P. – Wall, P. G. – Quinn, T. – Fanning, S.: using TaqMan and SYBR Green for detection of
Enterobacter sakazakii an emerging bacterial patho- Enterobacter sakazakii in infant formula. Journal of
gen with implications for infant health. Minerva Microbiological Methods, 65, 2006, pp. 21–31.
Pediatrica, 59, 2007, pp. 137–148. 20. Derzelle, S. – Dilasser, F.: A robotic DNA purifi-
7. Drudy, D. – O’Rourke, M. – Murphy, M. – cation protocol and real-time PCR for the detec-
Mullane, M. R. – O’Mahony, R. – Kelly, R. – Fischer, M. – tion of Enterobacter sakazakii in powdered infant
Sanjaq, S. – Shannon, P. – Wall, P. – O’Mahony, M. – formulae. BMC Microbiology, 6, 2006, No. 100.
Whyte, P. – Fanning, S.: Characterization of a collec- doi:10.1186/1471-2180-6-100.
tion of Enterobacter sakazakii isolates from environ- 21. Lampel, K. A. – Chen, Y.: Method for the isolation
mental and food sources. International Journal of and detection of Enterobacter sakazakii (Cronobacter)
Food Microbiology, 110, 2006, p. 127–134. from powdered infant formula. International Journal
8. Jaradat, Z. W. – Ababneh, Q. O. – Saa- of Food Microbiology, 136, 2009, pp. 179–184.
doun, I. M. – Samara, N. A. – Rashdan, A. M.: 22. El-Sharoud, W. M. – O’Brien, S. – Negredo, C. –
Isolation of Cronobacter spp. (formerly Enterobacter Iversen, C. – Fanning, S. – Healy, B.: Characterization
sakazakii) from infant food, herbs and environ- of Cronobacter recovered from dried milk and relat-
mental samples and the subsequent identifica- ed products. BMC Microbiology, 9, 2009, No. 24.
tion and confirmation of the isolates using bio- doi:10.1186/1471-2180-9-24.
chemical, chromogenic assays, PCR and 16S rRNA 23. Turcovský, I. – Kuniková, K. – Drahovská. H. –
sequencing. BMC Microbiology, 9, 2009, No. 225. Kaclíková, E.: Biochemical and molecular charac-
doi:10.1186/1471-2180-9-225. terization of Cronobacter spp (formerly Enterobacter
9. Bowen, A. B. – Braden, C. R.: Invazive Enterobacter sakazakii) isolated from foods. Antonie van
sakazakii disease in infants. Emerging Infectious Leeuwenhoek, 99, 2011, pp. 257–269.
Diseases, 12, 2006, pp. 1185–1189. 24. Krascsenicsová, K. – Trnčíková, T. – Kaclíková, E.:
10. Iversen, C. – Forsythe, S. Isolation of Enterobacter Detection and quantification of Enterobacter sakaza-
sakazakii and other Enterobacteriaceae from pow- kii by 5‘-nuclease polymerase chain reaction target-
dered infant formula milk and related products. ing the palE gene. Food Analytical methods, 1, 2008,
Food Microbiology, 21, 2004, pp. 771–777. pp. 85–94.
11. Kandhai, M. C. – Reij, M. W. – Gorris, L. G. – 25. Cawthorn, D. M. – Botha, S. – Witthuhn, R. C.:
Guillaume-Gentil, O. – Schothorst, M.: Occurrence Evaluation of different methods for the detection
of Enterobacter sakazakii in food production environ- and identification of Enterobacter sakazakii isolated
ments and households. Lancet, 363, 2004, pp. 39–40. from South African infant formula milks and the
12. Osaili, T. – Forsythe, S.: Desiccation resistance and processing environment. International Journal of
persistence of Cronobacter species in infant formula. Food Microbiology, 127, 2008, pp. 129–138.
International Journal of Food Microbiology, 136, 26. Mohan Nair, K. M. – Venkitanarayanan, K. S.: Cloning
2009, pp. 214–220. and sequencing of the ompA gene of Enterobacter
13. Commission regulation (EC) No. 2073/2005 of sakazakii and development of an ompA-targeted
15 November 2005 on microbiological criteria for PCR for rapid detection of Enterobacter sakaza-
foodstuffs. Official Journal of the European Union, kii in infant formula. Applied and Environmental
L 338, 22.12.2005, pp. 1–23. Microbiology, 72, 2006, pp. 2539–2546.
14. ISO/TS 22964 : 2006. Milk and milk products – 27. Hyeon, J. Y. – Park, C. – Choi, I. S. – Holt, P. S. –
Detection of Enterobacter sakazakii, . Geneva : Seo, K. H.: Development of multiplex real-time PCR
International Standard Organization, 2006. 13 pp. with internal amplification control for simultaneous
15. Iversen, C. – Druggan, P. – Forsythe, S.: A selec- detection of Salmonella and Cronobacter in pow-
tive differential medium for Enterobacter sakazakii dered infant formula. International Journal of Food
a preliminary study. International Journal of Food Microbiology, 144, 2010, pp. 177–181.
Microbiology, 96, 2004, pp. 133–139.
16. Iversen, C. – Forsythe S. J.: Comparison of media for Received 28 February 2011; revised 28 March 2011; accept-
the isolation of Enterobacter sakazakii. Applied and ed 4 April 2011.

124