A proposed harmonized LPS molecular-subtyping scheme for Cronobacter species

A proposed harmonized LPS molecular-subtyping scheme for Cronobacter species

Food Microbiology 50 (2015) 38e43 Contents lists available at ScienceDirect Food Microbiology journal homepage: www.elsevier.com/locate/fm Short co...

605KB Sizes 1 Downloads 27 Views

Food Microbiology 50 (2015) 38e43

Contents lists available at ScienceDirect

Food Microbiology journal homepage: www.elsevier.com/locate/fm

Short communication

A proposed harmonized LPS molecular-subtyping scheme for Cronobacter species bert f, Larisa H. Trach b, Qiongqiong Yan a, Karen G. Jarvis b, Hannah R. Chase b, Karine He Chloe Lee b, Jennifer Sadowski b, Boram Lee b, Seongeun Hwang b, Venugopal Sathyamoorthy b, Niall Mullane c, Monica Pava-Ripoll d, Carol Iversen e, amus Fanning a, Ben D. Tall b, * Franco Pagotto f, Se a

UCD Centre for Food Safety, WHO Collaborating Centre for Research, Reference, and Training on Cronobacter, University College Dublin, Belfield, Dublin 4, Ireland b CFSAN, FDA, Laurel, MD, USA c Mead Johnson Nutrition, Evansville, IN, USA d CFSAN, FDA, College Park, MD, USA e College of Life Sciences, University of Dundee, Scotland, UK f Food Directorate, Bureau of Microbial Hazards/Health Canada, Ottawa, ON, Canada

a r t i c l e i n f o

a b s t r a c t

Article history: Received 18 November 2014 Received in revised form 6 March 2015 Accepted 9 March 2015 Available online 17 March 2015

Cronobacter are opportunistic pathogens, which cause infections in all age groups. To aid the characterization of Cronobacter in foods and environments a harmonized LPS identification scheme for molecular serotyping is needed. To this end, we studied 409 Cronobacter isolates representing the seven Cronobacter species using two previously reported molecular serotyping schemes, described here as MullaneeJarvis (MeJ) and Sun schemes. PCR analysis revealed many overlapping results that were obtained when independently applying the two serotyping schemes. There were complete agreements between the two PCR schemes for Cronobacter sakazakii (Csak) O:1, Csak O:3, and Csak O:7 serotypes. However, only thirty-five of 41 Csak O:4 strains, identified using the MeJ scheme, were PCR-positive with the Sun scheme primers. Also the Sun scheme Csak O:5 primers failed to identify this serotype in any of the C. sakazakii strains tested, but did recognize seven Cronobacter turicensis strains, which were identified as Ctur O:3 using the MeJ scheme. Similarly, the Sun scheme Csak O:6 primers recognized 30 Cronobacter malonaticus O:2 strains identified with the MeJ scheme, but failed to identify this serotype in any C. sakazakii strain investigated. In this report, these findings are summarized and a harmonized molecular-serotyping scheme is proposed which is predicated on the correct identification of Cronobacter species, prior to serotype determination. In summary, fourteen serotypes were identified using the combined protocol, which consists of Csak O:1-O:4, and Csak O:7; Cmal O:1-O:2; Cdub O:1-O:2, Cmuy O:1-O:2, Cuni O:1, as well as Ctur O:1 and Ctur O:3. Published by Elsevier Ltd.

Keywords: Cronobacter LPS Molecular serotyping SDS-PAGE

1. Introduction Cronobacter species, primarily Cronobacter sakazakii, Cronobacter malonaticus and Cronobacter turicensis, cause life threatening infections such as septicemia, pneumonia, meningitis, and

* Corresponding author. Lab #: 3408, MOD 1 Facility, Virulence Mechanisms Branch, (HFS-025), Division of Virulence Assessment, OARSA, Center for Food Safety and Applied Nutrition, U. S. Food and Drug Administration, 8301 Muirkirk Road, Laurel, MD 20708, USA. Tel.: þ1 301 210 5171. E-mail address: [email protected] (B.D. Tall). http://dx.doi.org/10.1016/j.fm.2015.03.003 0740-0020/Published by Elsevier Ltd.

necrotizing enterocolitis in neonates (infants of less than 28 days of age) and infants (Jaradat et al., 2014; Tall et al., 2014). In addition, the genus also contains four other species: Cronobacter muytjensii, Cronobacter dublinensis, Cronobacter universalis and Cronobacter condimenti (Iversen et al., 2007, 2008; Joseph et al., 2012a). Neonates are most susceptible to infections, which can lead to longterm sequelae for those that survive, including delayed neurological development, hydrocephalus and permanent neurological damage and mental disabilities; estimated mortality rates for this age group are as high as 80% (Bowen and Braden, 2006). Based on a USA-based surveillance study, during 2003e2009 and across all age

Q. Yan et al. / Food Microbiology 50 (2015) 38e43

groups, Patrick et al. (2014) recently reported that Cronobacter infections are more common in adults, than in infants, that had originally been recognized. Those infections reported among adults included nosocomial and community-acquired urinary tract infections, septicemia, pneumonia, and wound infections. However, the highest incidence rate of invasive disease (case definition: confirmed isolate obtained from blood or cerebrospinal fluid) was associated with infants (Patrick et al., 2014). These findings also corroborate those of a single hospital-based microbiological survey and similar findings for elderly stroke patients as reported previously (Gosney et al., 2006; Holý et al., 2014). Traditionally, serology identifies differences among O-antigens associated with Gram-negative pathogens of human health importance, based on antibodyeantigen interactions. However, due to the high cost of producing epitope-specific antibodies, the laborintensive nature, and lack of specificity of the procedures, antibodybased methods are now being replaced with molecular-based PCR assays (Jarvis et al., 2011, 2013; Mullane et al., 2008). In most Enterobacteriaceae, the lipopolysaccharide (LPS) O-antigen biosynthesis encoding gene clusters (located at the rfb-encoding loci) are flanked by the galF and gnd genes, contain from 6 to 19 genes, and are generally conserved within a species (Samuel and Reeves, 2003). Typically O-antigen gene clusters contain three types of genes; those specific to nucleotide sugar biosynthesis pathways, glycosyltransferase genes that provide unique linkages between sugar residues, and O-antigen processing genes required for assembly of the polysaccharide repeating units and their transport through the periplasm and into the outer membrane (Samuel and Reeves, 2003). The most common O-antigen processing pathway found in members of the Enterobacteriaceae is the Wzx/Wzy-dependent transport process, which utilizes two proteins, an O-antigen flippase, Wzx and an O-antigen polymerase, Wzy. However, in some instances O-antigens are processed via an ABC transporter mechanism using the Wzm protein, which is responsible for polysaccharide export, in conjunction with an ATP binding protein designated as Wzt (Cuthbertson et al., 2010; Lewis et al., 2012; Jarvis et al., 2013). Previous work in our laboratories and others have identified up to 17 Cronobacter molecular serogroups among the seven species using PCR methods that were derived from sequence analysis of polymorphic markers of the corresponding rfb-encoding loci (Jarvis et al., 2011, 2013; Mullane et al., 2008; Sun et al., 2012, 2011). Based on this approach, two independent schemes for the molecular serotyping of Cronobacter have been described. The first scheme was described by Mullane et al. and reported the molecular identification of the first two serotypes within Cronobacter (Csak O:1 and O:2) (Mullane et al., 2008). This scheme was collaboratively extended by Jarvis et al. to include two more C. sakazakii serotypes (Csak O:3 and Csak O:4) and nine more serotypes that were associated with other Cronobacter species (Jarvis et al., 2011, 2013). The other reported molecular serotyping scheme was described by Sun et al. and consists of a multiplex PCR assay designed to identify seven serotypes associated with C. sakazakii (Sun et al., 2011), including serotypes Csak O:1 and Csak O:2, which were described previously by Mullane et al. (2008). Additionally, a single PCR reaction which identifies C. turicensis serotype O:2 was also reported by Sun et al. (2012). In this report both of these schemes, designated as the MullaneeJarvis (MeJ) scheme and Sun scheme, are comparatively assessed for their abilities to identify serotypes designated for C. sakazakii serotypes O:1 through O:7 (Csak O:1-Csak O:7); C. malonaticus serotypes O:1 and O:2 (Cmal O:1, Cmal O:2); C. dublinensis serotypes O:1 and O:2 (Cdub O:1, Cdub O:2); C. muytjensii serotypes O:1 and O:2 (Cmuy O:1, Cmuy O:2); C. universalis O:1 (Cuni O:1); and C. turicensis serotypes O:1-O:3 (Ctur O:1-O:3). In an effort to

39

better understand the applicability of the two typing schemes, 409 strains including the seven species were serotyped using both schemes. The results showed that there are useful attributes of each scheme that could be developed into an integrated standardized molecular serotyping protocol. In this report, we discuss these findings and propose a unified molecular serotyping scheme to the global food safety and public health communities so that they can move forward in a harmonized way to support characterization of the genus. 2. Material and methods 2.1. Bacterial strains: identification, and molecular characterization Four-hundred and nine Cronobacter isolates were characterized biochemically and identified according to the classification scheme described by Iversen et al. (2006, 2007, 2008), and their species identities were confirmed using both the rpoB and cgcA speciesspecific PCR assays as described by Stoop et al. (2009), Lehner et al. (2012) and Carter et al. (2013). These isolates were obtained from clinical, food, environmental, and unknown sources from diverse global geographical locations. Furthermore, the isolates were subjected to repFIB plasmid typing to identify genus- and species-specific plasmid-borne virulence factors as described by Franco et al. (2011), Grim et al. (2012), Gopinath et al. (2013), and Jarvis et al. (2011, 2013), the results of which support each strain's species identity (data not shown). For the comparison of the two serotyping schemes, the molecular serotype of each strain was characterized using PCR primers and reaction conditions as described by Mullane et al. (2008), Jarvis et al. (2011, 2013) and Sun et al. (2011, 2012). The primers and PCR assay amplification parameters used in this combined approach are described in Table 1. 2.2. SDS-polyacrylamide gel electrophoresis (SDS-PAGE) analysis of Cronobacter LPS To characterize the structural diversity among Cronobacter serotypes, LPS from a representative group of strains were prepared following sub culturing of overnight cultures into 100 ml of fresh pre-warmed tryptone soy broth (TSB, Difco, Becton Dickinson, Sharps, MD) at 37  C with shaking at 150 rpm. These were then grown to late exponential phase (to an optical density at 610 nm of 0.8 units to give approximately 108 cfu/ml) and the bacterial pellet was harvested by centrifugation. Following three washes with phosphate-buffered saline (PBS), the bacterial pellets were solubilized in 200 mg/ml proteinase K (Roche, Mannheim, Germany) and incubated overnight at 37  C. Following the addition of sample buffer containing 4% sodium dodecyl sulfate (SDS), the digests were boiled at 95  C for 10 min and subsequently cooled. Preparation of these digests was subjected to SDS-polyacrylamide gel electrophoresis (SDS-PAGE) using a 4% (wt/vol) stacking and a 15% (wt/ vol) separating gel. Forty microliters of the samples were loaded for electrophoresis at 160 V for 70 min until the dye ran out of the gel, followed by silver staining (Mullane et al., 2008). The images were visualized and photographed with a Kodak Gel Logic 1500 Imaging System (Carestream Health, Inc., NY, USA). 3. Results and discussion The isolates serotyped in this study included 303 C. sakazakii, 42 C. malonaticus, 15 C. muytjensii, 32 C. dublinensis, 14 C. turicensis, two C. universalis strains, and the single known C. condimenti strain. Table 2 summarizes the serotype comparisons of these strains using both the MeJ and Sun schemes, of which 270 strains were screened and reported previously using the MeJ scheme during the

40

Q. Yan et al. / Food Microbiology 50 (2015) 38e43

Table 1 Primers used in this study to identify Cronobacter serotypes using both the Mullane-Jarvis and Sun schemes. Serotype

Primer name

Primer sequence

Amplicon size (bp)

PCR conditions

References

Csak O:1

EsLPS1F EsLPS1R EsLPS2F EsLPS2R 2156-wzxF1 2156-wzxR1 E764-wzxF1 E764-wzxR1 wl-40039 wl-40040 z3032-wzxF5 z3032-wzxR4 E825-wzxF3 E825-wzxR2 z3032-wzxF5 z3032-wzxR4 WL-Ctur O:2F WL-Ctur O:2R E609-wzxF1 E609-wzxR1 E464-wzxF1 E464-wzxR1 E515-wzmF1 E515-wzmR1 2156-wzxF1 2156-wzxR1 51329-wzxF1 51329-wzxR1 E797-wzxF1 E797-wzxR1

CACGTTCGCCCTGCAAAAATa GCAAGCGGCCAGACTGGATAb TCCTGCATTTGTGGATTTTGCc AACGCATTGCGCTTGAGAAAd TGGCTGTCATGGTTTTCTTGCe TAGTTGGCACCATCAACGCCf CAGTAGCGGGGACATGGCTTg CCTGCCAGTAACCTGCATCCTCh CATTTCCAGATTATTACCTTTC ACACTGGCGATTCTACCC AGGGGCACGGCTTAGTTCTGG CCCGCTTGCCCTTCACCTAAC TGGCCCTTGTTAGCAAGACGTTTC ATCCACATGCCGTCCTTCATCTGT AGGGGCACGGCTTAGTTCTGG CCCGCTTGCCCTTCACCTAAC TTTCTTGTTATTGCCTGTGT AACAAAATCAGCGAGACTAA GCATCCCTTCAGAGTAGCGCA ACCACCTGCCATTGTCCTACTG TCGTTTTGATGCTCTCGCTGCG ACAAATCGCGTGCTGGCTTGAA CTCGGTTCATGGATTTGCGGC CAGCGTGAAAACAGCCAGGT TGGCTGTCATGGTTTTCTTGC TAGTTGGCACCATCAACGCC CGCTGCGATTATGGTAGTGGGT TTCCCAGCTCAGCTCGTTTGC CATTCTCGCTTCCGCAGTTGC CCCAACCATCATTAGGGCCGAG

341

(1)l

(Mullane et al., 2008)

329

(1)

(Mullane et al., 2008)

258

(1)

(Jarvis et al., 2011, 2013)

216

(1)

(Jarvis et al., 2013)

615

(2)m

(Sun et al., 2011)

323

(1)

(Jarvis et al., 2011, 2013)

394

(1)

(Jarvis et al., 2011, 2013)

323

(1)

(Jarvis et al., 2011, 2013)

418j

(3)n

(Sun et al., 2012)

236

(1)

(Jarvis et al., 2011, 2013)

435

(1)

(Jarvis et al., 2013)

227

(1)

(Jarvis et al., 2013)

258

(1)

(Jarvis et al., 2011, 2013)

475

(1)

(Jarvis et al., 2013)

145

(1)

(Jarvis et al., 2013)

Csak O:2 Csak O:3 Csak O:4 Csak O:7 Cmal O:1 Cmal O:2 Ctur O:1 Ctur O:2 Ctur O:3 Cdub O:1 Cdub O:2 Cmuy O:1 Cmuy O:2 Cuni O:1

i

k

Csak O:1 forward primer wl-35646 (50 -CCCGCTTGTATGGATGTT-30 ) from the Sun scheme can be used as an alternative. Csak O:1 reverse primer wl-35647 (50 -CTTTGGGAGCGTTAGGTT-30 ) from the Sun scheme can be used as an alternative. c Csak O:2 forward primer wl-37256 (50 -ATTGTTTGCGATGGTGAG-30 ) from the Sun scheme can be used as an alternative and any PCR-negative strain need to be assayed by the MeJ scheme Csak O:2 primers. d Csak O:2 reverse primer wl-37257 (50 -AAAACAATCCAGCAGCAA-30 ) from the Sun scheme can be used as an alternative and any PCR-negative strain need to be assayed by the MeJ scheme Csak O:2 primers. e Csak O:3 forward primer wl-37258 (50 -CTCTGTTACTCTCCATAGTGTTC-30 ) from the Sun scheme can be used as an alternative. f Csak O:3 reverse primer wl-37259 (50 -GATTAGACCACCATAGCCA-30 ) from the Sun scheme can be used as an alternative. g Csak O:4 forward primer wl-39105 (50 -ACTATGGTTTGGCTATACTCCT-30 ) from the Sun scheme can be used as an alternative and any PCR-negative strain need to be assayed by the MeJ scheme Csak O:4 primers. h Csak O:4 reverse primer wl-39106 (50 -ATTCATATCCTGCGTGGC-30 ) from the Sun scheme can be used as an alternative and any PCR-negative strain need to be assayed by the MeJ scheme Csak O:4 primers. i Strains, which were identified as Cmal O:2 by the MeJ scheme were also amplified by Csak O:6 primers (forward primer wl-40041: 50 -ATGGTGAAGGGAACGACT-30 and reverse primer wl-40042: 50 -ATCCCCGTGCTATGAGAC-30 ) in the Sun scheme. j The amplicon size for Ctur O:2 was determined by aligning primer sequences described by Sun et al. (2012) to the sequence of JQ354993 using BLAST. k Strains, which were identified as Ctur O:3 by the MeJ scheme were also amplified by Csak O:5 primers (forward primer wl-39873: 50 -GATGATTTTGTAAGCGGTCT-30 and reverse primer wl-39874: 50 -ACCTACTGGCATAGAGGATAA-30 ) in the Sun scheme. l PCR reaction (1): Initial denaturation step at 95  C for 2 min, 25 cycles of 95  C for 30 s, 55  C for 30 s, 72  C for 1 min, with a final extension step at 72  C for 5 min (Jarvis et al., 2011, 2013). m PCR reaction (2): Initial denaturation step at 95  C for 5 min, 30 cycles of 94  C for 30 s, 53  C for 30 s, 72  C for 1 min, with a final extension step at 72  C for 5 min (Sun et al., 2012). n PCR reaction (3): Initial denaturation step at 95  C for 5 min, 30 cycles of 94  C for 30 s, 50  C for 30 s, 72  C for 1 min, with a final extension step at 72  C for 5 min (Sun et al., 2012). a

b

development of new serotype-specific PCR assays (Jarvis et al., 2011, 2013; Mullane et al., 2008). Five serotypes within C. sakazakii were shared among the strain collections, including 106 Csak O:1, 103 Csak O:2, 37 Csak O:3, 41 Csak O:4 and 11 Csak O:7 strains. In the case of Csak O:1, Csak O:3, and Csak O:7, there were complete agreements between the two schemes. However, only thirty-five of 41 C. sakazakii strains, identified as Csak O:4 using the MeJ scheme, were PCR-positive with the Sun Csak O:4 primers. Additionally, two of the 103 Csak O:2 strains were not identified by the Sun scheme. Interestingly, the Sun Csak O:5 primers were negative with all of the C. sakazakii strains tested here. However, these primers were positive with seven C. turicensis O:3 strains (Table 2). Similarly, the Csak O:6 Sun primers recognized 28 of 30 C. malonaticus strains, which were positive with the MeJ Cmal O:2 LPS primers. The Sun primers were also negative for 32 C. dublinensis, 14 C. muytjensii, 13

C. malonaticus, two C. universalis strains and the single C. condimenti strain (Table 2). Nine of 14 C. muytjensii strains were identified using the MeJ scheme Cmuy O:1 and Cmuy O:2 primers. Ten C. turicensis strains were found to possess the Ctur O:1 and Ctur O:3 LPS molecular determinants, respectively. However, the molecular LPS determinants of four of the C. turicensis strains remain unknown by either scheme. None of the Sun et al. primers reacted with the C. mal O:1, Ctur O:1, Cdub O:2, Cmuy O:1 and O:2 or Cuni strains. Isolates positive for the Csak O:5, Csak O:6, and Ctur O:2 serotypes as described by Sun et al. (2012) were not identified in this study. It could be that these serotypes are endemic only within China. The formation of a harmonized Cronobacter LPS identification scheme will support the global food safety and public health communities in monitoring for the presence of Cronobacter in

Q. Yan et al. / Food Microbiology 50 (2015) 38e43

41

Table 2 PCR results for 409 isolates obtained from the comparison of primers described for the MullaneeJarvis (MeJ) and Sun schemes. Species

Proposed approach

MeJ schemea

Sun schemeb Csak O:1

Cronobacter sakazakii

Cronobacter malonaticus Cronobacter turicensis

Cronobacter dublinensis Cronobacter muytjensii Cronobacter universalis Not determined Total

MeJ-Csak O:1 MeJ-Csak O:2 MeJ-Csak O:3 MeJ-Csak O:4 Sun-Csak O:7 MeJ-Cmal O:1 MeJ-Cmal O:2 MeJ-Ctur O:1 Sun-Ctur O:2 MeJ-Ctur O:3 MeJ-Cdub O:1 MeJ-Cdub O:2 MeJ-Cmuy O:1 MeJ-Cmuy O:2 MeJ-Cuni O:1

106 103 37 41 11 11 30 3 0c 7 15 10 4 5 1 25 409

Csak O:2

Csak O:3

Csak O:4

Csak O:5

Csak O:6

Csak O:7

Ctur O:2

NDd

106 101

2 37 35

6 0 11 2 3

11 28 0 7

15 10 4 5 1 59

a

In the MeJ scheme, Csak O:5eO:7 and Ctur O:2 primers were adapted from the Sun scheme (Sun et al., 2011, 2012). In the Sun scheme, Csak O:1eO:7 primers were designed independently (Sun et al., 2011, 2012), where Csak O:1eO:6 primers are based on sequences of the wzy gene or wzx gene (Csak O:7) compared to wehC, wehI, and wzx genes as described by Mullane et al. (2008) and Jarvis et al. (2011, 2013) for Csak O:1eO:4. c Ctur O:2 was not identified using both schemes, however, the authors think that it's appropriate at this time to keep Ctur O:2 on the list for community usage. d ND, means not determined. LPS molecular determinants for five C. sakazakii, four C. turicensis, seven C. dublinensis, six C. muytjensii strains, and one strain each for C. malonaticus, C. universalis, and C. condimenti remain to be characterized. b

various food matrices, their associated production environments, and clinical samples. In an effort to understand the differences observed in this study between these two molecular serotyping schemes, we noted that the strains used to develop the Sun et al. serotyping primers were identified using only 16S rRNA gene sequence analysis which according to ours and other observations is not entirely reliable for Cronobacter identification (Joseph et al., 2012a,b). Further in this study we observed that the Sun et al. primers for Csak O:5 and Csak O:6 identified C. turicensis and C. malonaticus isolates, respectively, that were reliably identified using two proven methods namely the rpoB- and cgcA-based PCR methods (Stoop et al., 2009; Lehner et al., 2012; Carter et al., 2013). Speculatively, these findings suggest that the strains originally used to design the Sun-based serotype primers may have been misidentified as C. sakazakii, which subsequently led to an incorrect identification of the corresponding serotypes reported. The latter arises as a result of the minimal sequence diversity observed among 16S rRNA genes, particularly when attempting to differentiate between C. sakazakii and C. malonaticus, which possess a 99.7% similarity in their respective 16S rDNA gene sequences (Joseph et al., 2012b). Therefore, surveillance studies, where species identification was performed using 16S rDNA sequence analysis alone, may be open to question, and this approach should be revised. Nonetheless and based on the legal requirements mandated in international protocols at this time, there are no policies against its use as a genus-level target. SDS-PAGE analysis of silver stained LPS samples isolated from 10 strains of various serotypes representing six of the seven Cronobacter species are shown in Fig. 1. This analysis showed the typical LPS migration pattern for Gram-negative organisms, that included the lipid A-core moieties (located towards the bottom of the gel, in Fig. 1) and the variable-sized O-antigen side chains migrating above the lipid A region. These banding patterns also reflect the structural diversities observed among many of the Cronobacter serotypes that have been reported previously (Arbatsky et al., 2012, 2011, 2010a, 2010b; MacLean et al., 2009a, 2009b, 2011, 2012; Shashkov et al., 2011). The PCR data reported here support the view that a unified species identification and serotyping scheme would greatly

improve our ability to understand the prevalence of Cronobacter in foods, production facilities and in clinical samples. It is proposed that identification of isolates be accomplished using the rpoBspecies-specific PCR-based assay that has been widely used and validated for species identification of Cronobacter in a number of surveillance studies (El-Sharoud et al., 2008; Molloy et al., 2009;  et al., 2013). The authors propose that the global food Mozrova safety and public health communities apply the rpoB speciesspecific PCR assay to identify the Cronobacter species of interest, as suggested by Stoop et al. (2009) and Lehner et al. (2012) prior to determination of a strain's serotype. Although the cgcA speciesspecific multiplex PCR assay has not been fully validated, the sole laboratory validation study (Carter et al., 2013) along with the 100% correlation reported in this study shows that this approach may be a suitable alternative for species-specific identification due to its

Fig. 1. SDS-PAGE analysis of silver-stained purified LPS samples from 10 of the 14 Cronobacter serotypes previously classified (Mullane et al., 2008; Jarvis et al., 2011; Jarvis et al., 2013; Sun et al., 2011; Sun et al., 2012). The relative positions of the Oantigen side chains and lipid A core moieties are indicated by the right side brackets. Lane M, EZ-RunTM Pre-Stained Rec Protein Ladder (ThermoFisher Scientific Company, Waltham, MA). Values at left are in kilodaltons, kDa. Lane 1, ATCC ® BAA 894 (Csak O:1); Lane 2, E 830 (Csak O:2); Lane 3, E 615 (Cmal O:1); Lane 4, E 769 (Cmuy O:1); Lane 5, E618 (Cmal O:2), Lane 6, E 464 (Cdub O:1); Lane 7, E764 (Csak O:4); Lane 8, ATCC ® 51329 (Cmuy O:2); Lane 9, E 797 (Cuni O:1); Lane 10, E 515 (Cdub O:2).

42

Q. Yan et al. / Food Microbiology 50 (2015) 38e43

Fig. 2. A schematic outline of the proposed integrated Cronobacter species identification and LPS molecular serotyping scheme showing the detailed steps involved.

multiplex design. The disadvantage is that the cgcA protocol does not currently enable the identification of C. condimenti. Presently, there is only one strain of C. condimenti known to exist. However, species identification assessed by the analysis of individual multilocus sequence typing housekeeping genes has also been reported (Joseph et al., 2012b); but currently the usefulness of this tool for speciation is still open for validation, and therefore this approach is not recommended without additional information. A harmonized LPS molecular-based serotyping scheme is proposed based on the accurate identification of species followed by a serotyping PCR method as described in Table 1 and Fig. 2. In summary Csak O:1 and Csak O:3, primers from both the MeJ and Sun schemes are equally successful in identifying these serotypes. However, the MeJ primers for Csak O:2, Csak O:4, Cmal O:1-O:2, Ctur O:1, Ctur O:3, Cdub O:1-O:2, Cmuy O:1-O:2, and Cuni O:1 in the MeJ scheme are more discriminatory than those of the Sun scheme, therefore these primers should be selected for use in the proposed approach. Similarly, primers of Csak O:7 were more selective in the Sun scheme. An overall approach along with its alternative primers are presented in Table 1 and Fig. 2. Twenty-five strains, representing the seven species of Cronobacter gave negative PCR results using both the MeJ and Sun schemes suggesting that additional serotypes remain to be identified and characterized. The authors advocate that the global food safety community should consider adopting the unified typing scheme outlined in this study for future Cronobacter surveillance studies.

4. Conclusion This study shows the common and diverse attributes of the two previously reported Cronobacter molecular serotyping schemes. Overall, fourteen serotypes were identified using protocols and primers from both schemes, and these consisted of Csak O:1, Csak O:2, Csak O:3, Csak O:4, and Csak O:7; Cmal O:1 and Cmal O:2; Cdub O:1 and Cdub O:2, Cmuy O:1 and Cmuy O:2, Cuni O:1, as well

as Ctur O:1, and Ctur O:3. It is necessary to ensure that reliable and robust methods for identification are used prior to LPS serotyping so that the control of these organisms during the food manufacturing process or “farm to fork continuum” is achieved. Acknowledgments The authors report no financial or other conflicts of interest relevant to the subject of this article. References Arbatsky, N.P., Sun, Y., Shashkov, A.S., Wang, M., Liu, B., Daeva, E.D., Wang, L., Knirel, Y.A., 2012. Structure and genetics of the O-antigen of Cronobacter sakazakii G2726 (serotype O3) closely related to the O-antigen of C. muytjensii 3270. Carbohydr. Res. 355, 50e55. Arbatsky, N.P., Wang, M., Daeva, E.D., Shashkov, A.S., Feng, L., Knirel, Y.A., Wang, L., 2011. Elucidation of the structure and characterization of the gene cluster of the O-antigen of Cronobacter sakazakii G2592, the reference strain of C. sakazakii O7 serotype. Carbohydr. Res. 346, 1169e1172. Arbatsky, N.P., Wang, M., Shashkov, A.S., Chizhov, A.O., Feng, L., Knirel, Y.A., Wang, L., 2010a. Structure of the O-polysaccharide of Cronobacter sakazakii O2 with a randomly O-acetylated L-rhamnose residue. Carbohydr. Res. 345, 2090e2094. Arbatsky, N.P., Wang, M., Shashkov, A.S., Feng, L., Knirel, Y.A., Wang, L., 2010b. Structure of the O-polysaccharide of Cronobacter sakazakii O1 containing 3-(Nacetyl-L-alanyl) amino-3,6-dideoxy-D-glucose. Carbohydr. Res. 345, 2095e2098. Bowen, A.B., Braden, C.R., 2006. Invasive Enterobacter sakazakii disease in infants. Emerg. Infect. Dis. 12, 1185e1189. Carter, L., Lindsey, L.A., Grim, C.J., Sathyamoorthy, V., Jarvis, K.G., Gopinath, G., Lee, C., Sadowski, J.A., Trach, L., Pava-Ripoll, M., McCardell, B.A., Tall, B.D., Hu, L., 2013. Multiplex PCR assay targeting a diguanylate cyclase-encoding gene, cgcA, to differentiate species within the genus Cronobacter. Appl. Environ. Microbiol. 79, 734e737. Cuthbertson, L., Kos, V., Whitfield, C., 2010. ABC transporters involved in export of cell surface glycoconjugates. Microbiol. Mol. Biol. Rev. 74, 341e362. El-Sharoud, W.M., El-Din, M.Z., Ziada, D.M., Ahmed, S.F., Klena, J.D., 2008. Surveillance and genotyping of Enterobacter sakazakii suggest its potential transmission from milk powder into imitation recombined soft cheese. J. Appl. Microbiol. 105, 559e566. Franco, A.A., Hu, L., Grim, C.J., Gopinath, G., Sathyamoorthy, V., Jarvis, K.G., Lee, C., Sadowski, J., Kim, J., Kothary, M.H., McCardell, B.A., Tall, B.D., 2011. Characterization of putative virulence genes on the related RepFIB plasmids harbored by Cronobacter spp. Appl. Environ. Microbiol. 77, 3255e3267.

Q. Yan et al. / Food Microbiology 50 (2015) 38e43 Gopinath, G., Hari, K., Jain, R., Kothary, M.H., Jarvis, K.G., Franco, A.A., Grim, C.J., Hu, L., Sathyamoorthy, V., Mammel, M.K., Datta, A.R., Patel, I.R., Jackson, S.A., Kotewicz, M.L., LeClerc, J.E., Wekell, M., McCardell, B.A., Solomotis, M.D., Tall, B.D., 2013. The Pathogen-annotated Tracking Resource Network (PATRN) System: a web-based resource to aid regulatory science and investigations of foodborne pathogens and disease. Food Microbiol. 34, 303e318. Gosney, M.A., Martin, M.V., Wright, A.E., Gallagher, M., 2006. Enterobacter sakazakii in the mouths of stroke patients and its association with aspiration pneumonia. Eur. J. Intern. Med. 17, 185e188. Grim, C.J., Kothary, M.H., Gopinath, G., Jarvis, K.G., Beaubrun, J.G., McClelland, M., Tall, B.D., Franco, A.A., 2012. Identification and characterization of Cronobacter iron acquisition systems. Appl. Environ. Microbiol. 78, 6035e6050. , J., Hanulík, V., Chroma , M., Matouskov Holý, O., Petr zelova a, I., Forsythe, S.J., 2014. Epidemiology of Cronobacter spp. isolates from patients admitted to the Olomouc University Hospital (Czech Republic). Epidemiol. Mikrobiol. Imunol. 63, 69e72. Iversen, C., Waddington, M., Farmer 3rd, J.J., Forsythe, S.J., 2006. The biochemical differentiation of Enterobacter sakazakii genotypes. BMC Microbiol. 6, 94. Iversen, C., Lehner, A., Mullane, N., Marugg, J., Fanning, S., Stephan, R., Joosten, H., 2007. Identification of “Cronobacter” spp. (Enterobacter sakazakii). J. Clin. Microbiol. 45, 3814e3816. Iversen, C., Mullane, N., McCardell, B., Tall, B.D., Lehner, A., Fanning, S., Stephan, R., Joosten, H., 2008. Cronobacter gen. nov., a new genus to accommodate the biogroups of Enterobacter sakazakii, and proposal of Cronobacter sakazakii gen. nov., comb. nov., Cronobacter malonaticus sp. nov., Cronobacter turicensis sp. nov., Cronobacter muytjensii sp. nov., Cronobacter dublinensis sp. nov., Cronobacter genomospecies 1, and of three subspecies, Cronobacter dublinensis subsp. dublinensis subsp. nov., Cronobacter dublinensis subsp. lausannensis subsp. nov. and Cronobacter dublinensis subsp. lactaridi subsp. nov. Int. J. Syst. Evol. Microbiol. 58, 1442e1447. Jaradat, Z.W., Al Mousa, W., Elbetieha, A., Al Nabulsi, A., Tall, B.D., 2014. Cronobacter, an opportunistic food borne pathogen; a review of its virulence and environmental adaptive traits. J. Med. Microbiol. 63, 1023e1037. Jarvis, K.G., Grim, C.J., Franco, A.A., Gopinath, G., Sathyamoorthy, V., Hu, L., Sadowski, J.A., Lee, C.S., Tall, B.D., 2011. Molecular characterization of Cronobacter lipopolysaccharide O-antigen gene clusters and development of serotype-specific PCR assays. Appllied Environ. Microbiol. 77, 4017e4026. Jarvis, K.G., Yan, Q.Q., Grim, C.J., Power, K.A., Franco, A.A., Hu, L., Gopinath, G., Sathyamoorthy, V., Kotewicz, M.L., Kothary, M.H., Lee, C., Sadowski, J., Fanning, S., Tall, B.D., 2013. Identification and characterization of five new molecular serogroups of Cronobacter spp. Foodborne Pathogens Dis. 10, 343e352. Joseph, S., Cetinkaya, E., Drahovska, H., Levican, A., Figueras, M.J., Forsythe, S.J., 2012a. Cronobacter condimenti sp. nov., isolated from spiced meat, and Cronobacter universalis sp. nov., a species designation for Cronobacter sp. genomospecies 1, recovered from a leg infection, water and food ingredients. Int. J. Syst. Evol. Microbiol. 62, 1277e1283. Joseph, S., Sonbol, H., Hariri, S., Desai, P., McClelland, M., Forsythe, S.J., 2012b. Diversity of the Cronobacter genus as revealed by multilocus sequence typing. J. Clin. Microbiol. 50, 3031e3039. Lehner, A., Fricker-Feer, C., Stephan, R., 2012. Identification of the recently described Cronobacter condimenti by a rpoB based PCR system. J. Med. Microbiol. 61, 1034e1035.

43

Lewis, V.G., Ween, M.P., McDevitt, C.A., 2012. The role of ATP-binding cassette transporters in bacterial pathogenicity. Protoplasma 249, 919e942. MacLean, L.L., Pagotto, F., Farber, J.M., Perry, M.B., 2009a. Structure of the antigenic repeating pentasaccharide unit of the LPS O-polysaccharide of Cronobacter sakazakii implicated in the Tennessee outbreak. Biochem. Cell Biol. 87, 459e465. MacLean, L.L., Pagotto, F., Farber, J.M., Perry, M.B., 2009b. The structure of the Oantigen in the endotoxin of the emerging food pathogen Cronobacter (Enterobacter) muytjensii strain 3270. Carbohydr. Res. 344, 667e671. MacLean, L.L., Vinogradov, E., Pagotto, F., Perry, M.B., 2011. Characterization of the lipopolysaccharide O-antigen of Cronobacter turicensis HPB3287 as a polysaccharide containing a 5,7-diacetamido-3,5,7,9-tetradeoxy-D-glycero-D-galacto-non-2-ulosonic acid (legionaminic acid) residue. Carbohydr. Res. 346, 2589e2594. MacLean, L.L., Vinogradov, E., Pagotto, F., Perry, M.B., 2012. Structure of the O-antigen polysaccharide present in the lipopolysaccharide of Cronobacter dublinensis (subspecies lactaridi or lausannensis) HPB 3169. Can. J. Microbiol. 58, 540e546. Molloy, C., Cagney, C., O'Brien, S., Iversen, C., Fanning, S., Duffy, G., 2009. Surveillance and characterisation by pulsed-field gel electrophoresis of Cronobacter spp. in farming and domestic environments, food production animals and retail foods. Int. J. Food Microbiol. 136, 198e203. , V., Bren  ova , N., Mra zek, J., Lukesov Mozrova a, D., Marounek, M., 2013. Surveillance and characterisation of Cronobacter spp. in Czech retail food and environmental samples. Folia Microbiol. 59, 63e68. Mullane, N., O'Gaora, P., Nally, J.E., Iversen, C., Whyte, P., Wall, P.G., Fanning, S., 2008. Molecular analysis of the Enterobacter sakazakii O-antigen gene locus. Appllied Environ. Microbiol. 74, 3783e3794. Patrick, M.E., Mahon, B.E., Greene, S.A., Rounds, J., Cronquist, A., Wymore, K., Boothe, E., Lathrop, S., Palmer, A., Bowen, A., 2014. Incidence of Cronobacter spp. infections, United States, 2003-2009. Emerg. Infect. Dis. 20, 1520e1523. Samuel, G., Reeves, P., 2003. Biosynthesis of O-antigens: genes and pathways involved in nucleotide sugar precursor synthesis and O-antigen assembly. Carbohydr. Res. 338, 2503e2519. Shashkov, A.S., Arbatsky, N.P., Knirel, Y.A., 2011. Structures and genetics of Kdocontaining O-antigens of Cronobacter sakazakii G2706 and G2704, the reference strains of serotypes O5 and O6. Carbohydr. Res. 346, 1924e1929. Stoop, B., Lehner, A., Iversen, C., Fanning, S., Stephan, R., 2009. Development and evaluation of rpoB based PCR systems to differentiate the six proposed species within the genus Cronobacter. Int. J. Food Microbiol. 136, 165e168. Sun, Y., Arbatsky, N.P., Wang, M., Shashkov, A.S., Liu, B., Wang, L., Knirel, Y.A., 2012. Structure and genetics of the O-antigen of Cronobacter turicensis G3882 from a new serotype, C. turicensis O2, and identification of a serotype-specific gene. FEMS Immunol. Med. Microbiol. 66, 323e333. Sun, Y., Wang, M., Liu, H., Wang, J., He, X., Zeng, J., Guo, X., Li, K., Cao, B., Wang, L., 2011. Development of an O-antigen serotyping scheme for Cronobacter sakazakii. Appl. Environ. Microbiol. 77, 2209e2214. Tall, B.D., Chen, Y., Yan, Q.Q., Gopinath, G.R., Grim, C.J., Jarvis, K.G., Fanning, S., Lampel, K.A., 2014. Cronobacter: an emergent pathogen causing meningitis to neonates through their feeds. Sci. Prog. 97, 154e172.