Phenotypic and Genotypic Characterization of Cronobacter isolated from Powdered Infant Formula Retailed in Nigeria

Cronobacter is a genus with emerging pathogens that has been associated with life threatening diseases in neonates, infants and immunocompromised adults. Three Cronobacter species were isolated from powdered infant formula retailed in Nigeria. Different methods of phenotypic and genotypic characterization were carried out. All the isolates were identified biochemically by Microscan identification analysis as Enterobacter sakazakii (98.87%). The Vitek MALDI-TOF system identified the isolates as Cronobacter sakazakii. 16S rRNA sequencing identified the isolates as C. sakazakii. In contrast the use of species-specific PCR assays targeting rpoB, and cgcA, helped to identify two of the three strains as C. sakazakii and the last strain was identified as C. malonaticus. Multi locus sequence typing (MLST) analysis was used to identify each strain’s sequence type and the results identified three new sequence types: 303, 304 and 296. C. sakazakii BAA 894 served as a positive control for all the experiments. Biochemical methods and commercial identification systems are not sensitive enough to identify Cronobacter strains to the species level. Molecular methods are needed to confirm the species identity of strains.


Introduction
Cronobacter, (former Enterobacter sakazakii) is an emerging, opportunistic pathogen that causes infections such as sepsis, meningitis and necrotizing enterocolitis in neonates and infants, and can sometimes lead to death.It is ubiquitous in various food products such as dairy based products, adult and infant cereals, and spices.It has been associated with the ingestion of contaminated reconstituted powdered infant formula (PIF), and can be found in various environments, in particular PIF production facilities.Cronobacter spp have shown high resistance to osmotic stress and this contributes to its persistence in PIF factories, dried products and environments (Osaili & Forsythe, 2009).The frequency of disease caused by Cronobacter is very low but the mortality rate has been reported to be as high as 80% with surviving patients often suffering severe neurological sequelae (Alsonosi et al., 2015), hydrocephalus and permanent mental damage.The World Health Organization (WHO) has recognized all Cronobacter species as microorganisms pathogenic for human beings of all ages (FAO/ WHO, 2008) although this organism is also part of the normal human flora (Holy et al., 2014).Current international microbiological standards require the absence of all Cronobacter species in PIF (test volume 10 g) (Jackson et al., 2014).Of all Cronobacter species only C. sakazakii, C. malonaticus and C. turicensis have been linked to infantile infections (Alsonosi et al., 2015).
Cronobacter spp are Gram-negative motile rods of the family Enterobacteriaceae.They were formerly known as the yellow pigmented Enterobacter sakazakii.Reclassification based on the results of independent molecular methods and biochemical markers resulted in a new genus with seven species being described: Cronobacter sakazakii, C. malonaticus, C. muytjensii, C. dublinensis, C. turicensis, C. condimenti and C. universalis (Joseph et al., 2012).Target genes for PCR probe-based methods for Cronobacter identification include cgcA, gyrB, ompA, rpoB, gluA, dnaG, zpx, iron acquisition genes, the macromolecular synthesis operon, the 16S rDNA gene, and the 16S-23S intergenic transcribed spacer region (Carter et al., 2013, Grim et al., 2013, Lehner et al., 2012, Stoop et al., 2009).Molecular based techniques such as Multi locus sequence typing (MLST), random amplification of polymorphic DNA, pulsed-field gel electrophoresis (PFGE) have also been successfully applied to the characterization of Cronobacter spp (Fields et al., 2011).Pulsed field gel electrophoresis (PFGE) is considered the 'gold standard' method for subtyping foodborne bacteria and the most discriminatory technique for genotyping, and in 2012, a PulseNet protocol was validated for subtyping Cronobacter spp.(Brengi et al., 2012).
The aim of this study was to identify Cronobacter spp isolated from powdered infant formula products retailed in Nigeria using various phenotypic and genotypic methodologies.

Samples Collection
A total of 154 different samples of PIF were purchased from local markets and super markets in different geopolitical zones of Nigeria.

Bacterial Strains and Cultivation
Cronobacter strains used in this study were isolated from the PIF using the FDA method in combination with the method described by Iversen et al. (2004).Briefly, 25g of powdered infant formula was added to 225ml of buffered peptone water (BPW; pH 7.2+ 0.2).The suspended powdered infant formula was then incubated at 37 0 C for 24h without shaking.Four aliquots of 40ml each were removed from the solution and placed into 50ml centrifuge tubes.The tubes were subjected to centrifugation at 3000 x g for 10 min.The supernatants of each tube were discarded and the resultant pellets were suspended in 200μL of phosphate-buffered saline (PBS; pH 7.2+ 0.2).The aliquots were cultured on Druggan-Forsythe-Iversen (DFI) agar.The control strain, C. sakazakii ATCC BAA 894 was a gift from the Division of Virulence Assessment, Office of Applied Research and Safety Assessment, Center for Food Safety and Applied Nutrition, U. S. Food and Drug Administration (MD, USA).

Phenotyping
Cronobacter isolates were phenotyped using the MicroScan WalkAway identification panel (Beckman Coulter Inc., CA, USA) according to the manufacturer's instructions.A sterile applicator stick was used to touch the surface of 4-5 morphologically similar, well isolated colonies from an 18-24 h Brain Heart Infusion (BHI) (Difco, New Jersey) agar plate.This was emulsified in 3 mL of sterile deionized water and the suspension was vortexed for 2-3 s to achieve a final turbidity similar to the 0.5 McFarland Turbidity Standard scale.One-hundred microliters (100 μL) of the standardized suspension was pipetted into 25 mL of inoculum buffer, capped tightly and inverted 8-10 times to mix.An oxidase test was performed using tetramethyl-p-phenylenediaminedihydrochloride (Sigma-Aldrich, USA) prior to inoculating the panels.The panel was rehydrated and inoculated using the RENOK system with Inoculators-D (Siemens, Frimley, Camberley, UK).A dropper bottle was used to overlay the GLU, URE, H 2 S, LYS, ARG, ORN and DCB (Glucose, Urea, Hydrogen sulphide, Lysine, Arginine, Ornithine, and Decarboxylase Base) with 3 drops of sterile mineral oil.The panel was incubated in the Walkaway System for 16-20 h after which the results were read.

Species Identification Using the VITEK®MS MALDI-TOF
Identification of the Cronobacter isolates was carried out using VITEK®MS MALDI-TOF (Matrix Assisted Laser Desorption Ionization Time-of-Flight) technology (bioMerieux, France), according to the manufacturer's instructions.Two milliliters of bacterial broth was added to 1.0 ml of lysis buffer (0.6% polyoxyethylene 10 oleoyl ether (Brij 97) in 0.4 M (3-(cyclohexylamino)-1-propane sulfonic acid) (CAPS) filtered through a 0.2-µm-pore-size filter, pH 11.7), vortexed for 5 s, and allowed to incubate for 2 to 4 min at room temperature.The resulting lysate was filtered through a 25-mm 0.45-µm-pore-size filter (catalog no.HPWP02500; Millipore Express PLUS, Billerica, USA), MA and the microbial cells remaining on the filter were removed and washed three times with wash buffer (20 mM Na 3 PO 4 , 0.05% Brij 97, and 0.45% NaCl), using a 0.2-µm-pore-size filter, pH 7.2); washed three times with deionized water; and removed from the surface by scraping the filter with a micro-swab (Texwipe CleanTips swabs; catalog no.TX754B; Kernersville, NC).Sample processing time was approximately 10-15 min for up to three samples.Microorganisms recovered from the filter were directly applied to VITEK MS target plates and covered with 1 μL of CHCA (α-cyna-4-hydroxycinnamic acid) matrix.If the VITEK MS was unable to identify an isolate on the first attempt, the sample was repeated using double the volume of culture broth and corresponding buffers.A sample was considered to have a valid VITEK MS ID if at least one spot on the target slide gave a confidence level of ≥75% without conflicting identifications from replicate spots of the same sample.Samples that did not generate an ID on the first attempt were repeated only once.

PCR Probe Assays of Cronobacter spp
PCR was carried out using the 16S rDNA gene (Sigma-Aldrich, USA), β-subunit of RNA Polymerase gene (rpoB) (Sigma-Aldrich, USA), and Diguanylate Cyclase-Encoding gene (cgcA) (Sigma-Aldrich, USA).The PCR conditions and primers used are stated on Table 1.PCR amplicons were subjected to agarose gel electrophoresis using 0.8% Tris-borate-EDTA buffer (TBE; Sigma-Aldrich, USA) in a Bio-Rad sub cell GT (Bio-Rad, Belgium) horizontal electrophoresis unit and were photographed with transilluminated UV light using an Alpha Imager system (Alpha Innotech Corp, San Leandro, CA, USA).The PCR products were purified using Wizard SV Gel and PCR clean up system (Promega, USA) and sequenced by GENEWIZ (New Jersey, USA) in a Sanger DNA Sequencer System.The obtained nucleotide sequences were compared with the corresponding sequences of C. sakazakii strain ATCC BAA 894 (NCBI accession no.290339) and with other Enterobacteriaceae using DNASTAR-MegAlign.
Initial denaturation step at 94 0 C for 3 min, 30 cycles of 94 0 C for 60 s, 67 0 C for 30 s, 72 0 C for 1 min, with a final extension step at 72 0 C for 10 min.

C. malonaticus:
Initial denaturation step at 94 0 C for 3 min, 25 cycles of 94 0 C for 60 s, 60 0 C for 30 s, 72 0 C for 30 s, with a final extension step at 72 0 C for 10 min 1500 Cronobacter spp Kuhnert et al., 1996 rpoB Initial denaturation step at 94 0 C for 3 min, 30 cycles of 94 0 C for 60 s, 67 0 C for 30 s, 72 0 C for 1 min, with a final extension step at 72 0 C for 5 min.

Pulsed Field Gel Electrophoresis (PFGE)
PFGE analysis of Cronobacter isolates was performed as previously described by Ribot et al. ( 2006) with some modification.The restriction enzyme Xbal was used for DNA digestion.Bands were separated using a CHEF-DR III System (BIO-RAD, Belgium) at 14 -17 o C. Electrophoretic conditions of initial switch time 1.8 s to a final switch time of 25 s, at 6 volts/cm for a run time of 17-18 h were used.Gels were stained for 30 min in 500 mL dH2O containing 25 μl ethidium bromide (10 mg/mL) and visualized under UV light using an Alpha Imager system (Alpha Innotech Corp., San Leandro, CA, USA).

Multi Locus Sequence Typing
MLST was performed as previously described by Baldwin et al. (2009)  The following PCR conditions were used for amplification: an initial denaturation step at 95°C for 5 min, followed by 30 amplification cycles at 94°C for 30 s, 53°C for 30 s, and 72°C for 1 min, with a final extension step at 72°C for 5 min.Samples (5 μl) of the PCR products were subjected to agarose gel electrophoresis for examination.

Results
Out of the 154 different samples of powdered infant formula analyzed, only 3 (1.95%)were positive for Cronobacter spp.

Biochemical Identification of the Cronobacter Strains Using Phenotypic Assays
MicroScan WalkAway analysis identified all the isolates as Enterobacter sakazakii with 98.87% probability.Supplemental Table 1.

Identification of the Cronobacter Strains Using 16S rDNA Gene Sequencing
Amplification of the 16S rDNA gene from the Cronobacter isolates gave PCR products of about 1,500 bp.The PCR products were sequenced and identified using BLAST analysis (http://www.ncbi.nlm.nih.gov/).All amplification products were identified as the 16S rDNA gene of Cronobacter and had a 97-98% identity with Cronobacter spp.(Carter et al., 2013) CS 14 which was previously characterized as C. malonaticus using the cgcA and rpoB species-specific PCR assays was closely clustered to C. sakazakii ATCC BAA 894.This is highly misleading, thus showing the limitation of the 16s rDNA gene.
The rpoB gene sequence analysis has been proposed as a method for inferring relationships among very closely related species (Adekambi et al., 2009).Li et al. (2012) reported that phylogenetic analysis based on partial rpoB gene sequence analysis cannot distinguish between C. sakazakii and C. malonaticus even though it can differentiate these two species from other Cronobacter species.All isolates including BAA 894 showed similar band patterns using the C. sakazakii rpoB primers while CS 14 and CS 124 showed a similar pattern using the C. malonaticus rpoB primers.This result confirms the reported by Li et al. (2012) because CS 124 was identified as C. sakazakii using the multiplex cgcA primers.
Cyclic diguanylate (c-di-GMP) is a bacterial second messenger signal transduction molecule recognized for its involvement in the regulation of a number of complex physiological processes, including bacterial virulence, biofilm formation, and persistence (long-term survival) (Sondermann et al., 2012).Carter et al., (2013) reported the use of Cronobacter multiplex cgcA PCR assay to identify Cronobacter strains in single reaction.This PCR assay was found to be 100% specific (n=305) and 100% sensitive (n=20).The multiplex cgcA primers were used to identify sample CS14 as C. malonaticus, and CS17 and CS124 as C. sakazakii.The control strain BAA 894 was correctly identified as C. sakazakii.
Applying discriminatory molecular subtyping methods to characterize foodborne pathogens facilitates the detection of outbreaks, sources of infection, and transmission pathways (Fields et al., 2011).Epidemiologically related isolates share the same profile (Kuhnert et al., 1996).All Cronobacter isolates and the control strain BAA 894 demonstrated distinct PFGE profiles, indicating different sources.
The set of seven housekeeping genes in MLST has greater sequence diversity than the 16S rRNA gene and has been applied to identify many bacteria (Urwin & Maiden, 2003).New sequence types were created for each of the isolates based on the allelic combination of the seven loci.None of the fusA profiles were shared between any of the isolates, thus proving that all the isolates are different.This locus has been recommended for use with two PCR primer sets to define species of Cronobacter without the ambiguity of 16S rRNA gene sequence analysis.MLST has proven to be an effective and robust typing scheme for the Cronobacter genus and has exhibited a high level of discrimination between the isolates (Joseph et al., 2012).The new sequence types (303, 304 and 296) are not clustered with any other known Cronobacter pathovars but further work will be needed to analyze the genes that are present in these strains.
The O-antigen is a highly variable component of the lipo-polysaccharide of Gram-negative bacteria and is used for the development of both serological and molecular typing methods (Jarvis et al., 2013).C. sakazakii O-antigen gene clusters of all seven serotypes are located on the chromosome between the housekeeping genes galF and gnd.All the isolates in this study proved to be of different serogroups.Sun et al. (2012) reported the high-level identity (99.3%) of C. sakazakii O6 and C. malonaticus O2, implying the recent lateral transfer of the respective O-antigen gene cluster between these two species.However, Yan et al. (2015) reported 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.So, C. sakazakii O6 serotype should be corrected to C. malonaticus O2.Interestingly, C. sakazakii O2, to which CS 17 belongs, was isolated between 2010 and 2011 from infant clinical cases of Cronobacter infections in the USA (FDA/CFSAN, 2011).

Conclusion
The emerging pathogenic genus Cronobacter was identified as a contaminant in powdered infant formula retailed in Nigeria.The use of PCR assays, biochemical identification tests, and gene sequenced-based methods gave a reliable identification and profiling of all the Cronobacter isolates thus limiting the misidentification of false positive and negative results.Our results suggest that more severe measures must to be taken in order to improve the quality control on powdered infant formula production in order to protect neonates and infants from diseases caused by Cronobacter spp., which sometimes can be fatal.rendered in the course of this research.We thank Dr. Randall J. Olsen, M.D. for his assistance with the MALDI-TOF analysis.We also thank Dr. Audrey Wanger for her assistance with the Microscan Walkaway.

Figure 1 .
Figure 1.Gel image of the PCR products obtained from the amplification of the 16S rDNA gene from Cronobacter samples (CS) isolated from PIF. Lane M, DNA ladder (Invitrogen); lanes 1-3, CS 124; lanes 4 and 5, CS 14; lanes 6 and 7, CS 17; lanes 8 and 9, BAA 894 (positive control).The PCR products were about 1500 bp.Phylogenetic tree derived from the 16S rDNA sequence showing the relationship of CS 14, CS 17 and CS 124 to other Cronobacter spp.

Table 1 .
Primers and PCR conditions used for allelic profiling of Cronobacter spp

Table 2 .
(Baldwin et al., 2009) primer sequences for the amplification and sequencing of genes from C. sakazakii and C. malonaticus, with gene number and location corresponds to C. sakazakii strain ATCC BAA-894 genome(Baldwin et al., 2009)Primers designed based on the wzx and wzy gene (O antigen) sequences(Sun et al., 2012)were used.Multiplex PCR was performed by mixing all primers in a final volume of 50 μL containing the following components: 1× Taq Reaction buffer; 2.5 mM MgCl 2 ; 400 μM (each) of dATP, dCTP, dGTP, and dTTP; 0.06 to 0.10 μM primer sets listed in Table3, 2.5 U of Taq DNA polymerase and 50 to 100 ng of template DNA.

Table 3 .
Primers used for serotyping(Sun et al., 2012) closely related organisms such as C. sakazakii and C. malonaticus because of minimal sequence diversity or the presence of multiple copies of 16S rDNA gene loci