Molecular serotype and evolutionary lineage of Listeria monocytogenes isolated from different Nigerian food items

The molecular serotypes and the evolutionary lineage of Listeria monocytogenes isolated from various foods in Nigeria are yet to be documented. Consequently, popular uncooked food items known locally as Okazi Utazi, Onugbu, Ogbono, Garri and Egusi obtained from plants botanically known as Gnetum africanum, Gongronema latifolium, Vernonia amygdalina, Irvingia gabonensis, Manihot esculanta Crantz and Colocynthis Citrullus, respectively were analyzed. Molecular serotype of three chosen isolates was determined using multiplex polymerase chain reaction (PCR) serotyping before analysis based on prfA virulence gene cluster of L. monocytogenes was carried out to establish the evolutionary lineage. There was no L. monocytogenes detected in foods from I. gabonensis, M. esculanta Crantz and C. citrullus. However, the vegetables from G. africanum, G. latifolium and V. amygdalina showed the presence of the organism and chromogenic tests carried out on the three strains chosen from oxford formulation media indicated that they were not other non-pathogenic strains of Listeria . The V3 region of 16S rRNA gene of one strain showed that a close relative of the isolate is a strain implicated in an outbreak of listeriosis. Leafy vegetables could be a major vehicle for transmission of L. monocytogenes in Nigeria since this pathogenic bacterium occured in different vegetables analyzed.

solute concentrations, resists desiccation and therefore can overcome mild food preservation techniques (McLauchlin et al., 2014).Despite the world-wide reports of outbreaks of food-borne listeriosis, the occurrence of L. monocytogenes is still not widely reported in Nigeria due to lack of a comprehensive surveillance system for foodborne pathogens (Nwaiwu, 2015).
Common popular Nigerian foods found in South eastern region namely Okazi (also called Afang in another region), Utazi, Onugbu, Ogbono, Garri and Egusi can be sourced from plants with botanical names Gnetum africanum, Gongronema latifolium, Vernonia amygdalina, Irvingia gabonensis, Manihot esculanta Crantz and Colocynthis Citrullus, respectively.Leaves of G. africanum are eaten as a vegetable and also used widely as an ingredient in soups and stews and are important for their nutritional and therapeutic properties (Ali et al., 2011).Also popular is G. latifolium, a tropical rain forest plant used as a spice and vegetable and generally believed to possess many medicinal properties (Ugochukwu and Babady, 2002).Equally regarded to have medicinal properties are leaves of V. amygdalina which has several health benefits and also used for preparing local bitter leaf soup (Farombi and Owoeye, 2011).In Nigeria I. gabonensis is known as the wild mango seeds and it is usually used as a conventional soup thickner (Ekundayo et al., 2013).Another popular soup condiment highly rich in protein is Egusi made from melon seeds of C. citrullus (Bankole et al., 2006).Most soups are normally eaten with Garri, a popular food in Nigeria rich in carbohydrate and derived from the fermentation of mash obtained from the enlarged root of the cassava plant M. esculanta Crantz (Okafor et al., 1999).
Reports regarding foodborne listeriosis have increased due to the ability of L. monocytogenes to survive in harsh conditions (Hernandez-Milian and Payeras-Cifre, 2014).In order to reduce occurrence, Gillespie et al. (2006) suggested that dietary advice on avoiding high-risk foods should be provided routinely for the elderly and immunecompromised, not just to the pregnant women, these groups are associated with 15% out of 3088 cases of listeriosis reported in England and Wales between 1990-2010 (Awofisayo et al., 2015).In a serological study in Nigeria among different animal species (Oni et al., 1989), the prevalence of agglutinins to 5 serotypes of L. monocytogenes (1/2a, 1/2b, 1/2c, 3a and 4b) has been reported but molecular serotypes and evolutionary phylogeny of L. monocytogenes from Nigerian foods, animal or environment are still unknown.Although serological (agglutinins) typing correlates well with molecular lineage assignments (Nadon et al., 2001), molecular serotyping methods are now acceptable replacements for sero-grouping procedures (Nightingale et al., 2007).
Many different approaches are now used to evaluate occurrence and evolution of L. monocytogenes Nwaiwu 697 serotypes.Yin et al. (2015) performed multilocus sequence typing (MLST) and multi virulence locus sequence typing (MVLST) for 86 L. monocytogenes strains derived from 8 countries from 1926 to 2012 in order to understand the molecular evolution and genetic characteristics of the organism.The scientists identified a total of 13 clonal complexes and noted that polymorphism of housekeeping genes of isolates belonging to one of the clones increased rapidly over time.Furthermore, the phylogenetic analysis carried out showed that serotype 1/2b and 4b strains had an interval-type evolution pattern, while serotype 1/2a and 1/2c strains had a progressivetype evolution pattern.In another study, Pászti et al. (2014) used the molecular serotyping method of Doumith et al. (2004) and pulsed-field gel electrophoresis (PFGE) to serotype and characterize L. monocytogenes from human infections and found 2 serogroups and many pulsotypes.Other workers have used ribotyping and automated ribotyping (Matloob and Griffiths, 2014), Fourier transform infrared spectroscopy and chemometrics (Nyarko et al., 2014) and fluorescent amplified fragment length polymorphism (Amar, 2014) to differentiate L. Monocytogenes strains.Ferreira et al. (2014) has highlighted that the current available data clearly indicate that L. monocytogenes persistence at various stages of the food chain contributes to contamination of finished products and continued efforts to integrate data on L. monocytogenes persistence is still needed to advance our understanding on the persistence of this important pathogenic bacterium.To this direction, the aim of this study was to establish the molecular serotypes and lineages of L. monocytogenes that may be found in common food items in south eastern Nigeria.

Food items and detection of L. monocytogenes
Ground food items of I. gabonensis, M. esculanta Crantz and C. citrullus and sundried vegetables from G. africanum, G. latifolium and V. amygdalina (Figure 1) were purchased from a market in Owerri City, south eastern Nigeria after which standard presumptive identification as described by HPA (2007) was carried out.In this study, all media were prepared according to manufacturer's instructions and for primary detection, 25 g of each food item were weighed out into a stomacher bag and made up with 225 ml of half Fraser broth (Oxoid, Basingstoke, United Kingdom; CM895 base plus SR166E supplement) and then homogenized and incubated for 24 h at 30°C.
The incubated half Fraser broth was used for the secondary enrichment by inoculating 0.1 ml into 10 ml of Fraser broth (Oxoid, CM895 base plus SR156 supplement) in a universal bottle.Incubation was carried out at 37°C for 48 h after which 0.1 ml of broth was spread-plated onto Listeria oxford formulation (Oxoid; CM856 base and SR140 supplement) agar plates.Colonies that showed typical L. monocytogenes morphology were sub-cultured on brain heart infusion (BHI) agar (Oxoid, CM1135) to get pure cultures after which standard Gram stain, motility (Vatanyoopaisarn et al., 2000) and catalase tests were carried out.The isolates were stored at -85°C for future use.

Separating L. monocytogenes from non-pathogenic Listeria species
To confirm that selected strains were not non-pathogenic Listeria, tests were performed with SwabSURE-ListeriaP kit (TSC, Heywood, United Kingdom).According to the manufacturers, the protocol works by sampling a pre-moistened swab dosed in the kit's TSC neutralising buffer and then followed by incubation for 24-48 h.The appearance of a turquoise blue colour indicates a positive presence for pathogen specific Phospholipase C enzyme.In order to satisfy manufacturer's requirement for tests to be carried out only on surfaces, a test surface to mimic a food contact surface was prepared with a microscope slide as previously described (Mafu et al., 2011) with modification.Instead of cutting glass coupons from the slide, the whole glass microscope slide with plain cut edges measuring 76 × 26 mm (Fisher Scientific) was used.After inoculating a loop of cells from a colony of the test strain to 5 ml of BHI broth in a sterile test tube, the test tube was incubated in a test tube shaker (New Brunswick Scientific) at 30°C for 18 h after which a 100 µl of the broth culture was placed in the middle of the microscope slide and then spread out with an inoculating loop.The slide was air dried for 3 h after which recovery of L. monocytogenes from the surface of the slide was carried out according to manufacturer's instructions.The kit's pre-moistened swab with foam tip was removed from its holding tube making sure that it did not touch any surface before swabbing the surface of the microscope glass slide in 3 planes (horizontal, vertical and diagonal) whilst rotating swab between forefinger and thumb.After swabbing the surface of the slide, the swab was inserted into the media tube placed on a test tube rack and then incubated horizontally at 37°C for 48 h after which the tube was observed for change of colour.

Resuscitation of control strains
Four strains of L. monocytogenes with known serotype and lineage were used as control strains for this study, among them two reference strains were included (Table 1).The strains were maintained under cryo preservation at -85°C and these were obtained from the culture collection of Division of Food Science, University of Nottingham, UK.Cryo preservation beads containing the strains were removed from the -85°C freezer and streaked on to BHI media.Colonies that emerged after 48-72 h incubation at 37°C were used.To ensure that there was no contamination during recovery of the strains, the organisms were Gram stained and also grown on Listeria selective Oxford medium to confirm typical colonies of L. monocytogenes.

Extraction of genomic DNA
This was carried out as described by Pitcher et al. (1989).To the sterile test tubes, 5 ml of BHI broth was added and inoculated with a colony of of L. monocytogenes.The test tubes were placed in a test tube shaker (New Brunswick Scientific) and incubated at 37°C for 24 h.A slight modification to the protocol was introduced by pelleting 1.5 ml of the incubated culture in an Eppendorf at 13000 xg for 1 min with a microfuge (Biofuge 13; Heraeus Sepatech) instead of centrifuging at 1000 xg for 15 min to obtain a suitable cell pellet.The process was repeated to obtain a bigger pellet following which 1 ml of ice cold lysis buffer (25 mM Tris-HCl 8.0, 10 mM EDTA, 50 mM sucrose, Fisher Scientific, UK) was added.The mixture was then treated with 50 mgml -1 lysozyme (Sigma) and incubated for 1 h at 37°C after which 0.5 ml lysis solution (5 M guanidium thiocynate, 0.1 M EDTA, Fisher Scientific; 0.5% Sakosyl, Sigma) was added, mixed well and left at room temperature for 5 min.The lysate was cooled on ice for 30 min and 0.25 ml of ammonium acetate was added, vortexed and incubated again on ice for another 10 min.Chloroform : isoamyl alcohol (24:1 ratio, 0.5ml, Fisher Scientific, UK) was added, vortexed and spun for 10 min at 13000 xg.The upper phase 850 µl was removed and introduced into a clean Eppendorf tube following which exactly 0.54 volumes of cold isopropanol were added.The mixture was centrifuged at 13000 xg for 20 s and the visible pellet was washed 3 times in 70% ethanol.The pellet was resuspended in 50 µl TE (10mM Tris, 1mM EDTA pH7.0).To remove RNA, 1 µl (50 mgml -1 ) of lyophilized RNAse was added to the suspension and incubated at 37°C for 30 min.The integrity of the DNA was checked by horizontal electrophoresis in 0.7% agarose in TAE buffer (40 mM Tris base, 20 mM glacial acetic acid 1 mM EDTA, pH 8, Fisher Scientific) containing ethidium bromide (0.5 mgml -1 ) and viewed under ultraviolet light (ImageMaster® VDS system (Amersham Biosciences,No.80-6254-80).The concentration (ng/µl) was measured by placing 1 μl of DNA sample onto NanoDrop spectrophotometer (Thermo Fisher Scientific Inc.,USA) after which the DNA was stored at -20°C till its subsequent use in PCR reactions.

Molecular serotyping
Serotyping was carried out using the multiplex PCR protocol described by Doumith et al. (2004) using the primer pairs No. 5-9 (Table 2).A final volume of 50 µl containing 1 U of Taq polymerase (ABgene), 10 ng DNA template, 0.2 mM deoxyribonucleoside triphosphate (Promega), 2 mM MgCl 2 (ABgene) and 5 µl of 1X PCR buffer (ABgene) was used.Primers were added at the following concentration: 1 µM for lmo0737, ORF2819 and ORF2110; 1.5 µM for lmo1118 and 0.2 µM for prs.PCR was performed with an initial denaturation step at 94°C for 3 min followed by 35 cycles of 94°C for 0.40 min, 53°C for 1.15 min and 72°C for 1.15 min and one final cycle of 72°C for 7 min in a thermocycler (Techne 312).Following PCR reaction, the product (5 µl) was mixed with 2 µl of loading buffer and separated on a 2% agarose gel in TAE buffer.Staining with ethidium bromide enabled visualization of PCR fragments under UV light.After scoring the PCR products against a 100bp ladder, serotypes were assigned to the test strains based on genes amplified by the corresponding reference strain.

Evolutionary lineage analysis
Lineage group identification based on prfA virulence gene cluster of L. monocytogenes developed by Ward et al. (2004) was used by amplifying allelic specific oligonucleotides (ASO) using primer pair No. 2-4 (Table 2).Amplifications were performed in 50 µl volumes with 0.5 µM concentrations of each primer, 2 mM MgCl 2 , 0.2 mM concentrations of each deoxynucleoside triphosphate, 0.5 U of Taq polymerase (Abgene) and 100 ng of DNA.Amplifications consisted of 25 cycles of 15 s at 94°C, 10s at 56°C and 10 s at 72°C following which amplification products were resolved in 2% (wt/vol) agarose gel and scored relative to 100 bp DNA size ladder before evolutionary lineages were assigned to the test strains based on genes amplified by the corresponding reference strain.

Sequencing of the V3 region of 16S rDNA
Sequencing for an isolate of interest was carried out by PCR amplification of the V3 region of the 16s rDNA (Muyzer et al., 1993;Coppola et al., 2001) using primer pairs No. 1 (Table 2).A 1 µl of DNA template (100 ng/µl) was used and the reaction mixture (50 μl) for amplification contained 1.25 U of Taq DNA polymerase (Fisher Scientific) 5 μl of 10X PCR buffer (Fisher Scientific,), 0.2 mM of the deoxynucleotide triphosphates, 0.2 pmol/μl of forward and reverse primers and 2.5 mM of magnesium chloride.
DNA was denatured for 5 min at 94°C followed by initial annealing temperature of 66°C, which was decreased 1°C every cycle for 10 cycles after which another 20 cycles were performed at 56°C.Elongation was performed at 72°C for 3 min before a final extension at 72°C for 10 min was carried out.The PCR fragments generated were cut out from the gel and cleaned with a purification kit (Qiagen) and then sequenced (MGW Eurofins, Germany).Sequence generated was used for a search on the databases of National Center for Biotechnology Information (http://ncbi.nlm.nih.gov) using the BLAST program after which the sequence was submitted to EMBL archives and given accession number LN832629.

Detection and identification of L. monocytogenes
It is known that the ability of L monocytogenes to hydrolyze esculin to esculetin (Qadri et al., 1980) and the  subsequent reaction of esculetin with ferric ions which results in blackening of media is used for the identification of the organism.After secondary enrichment, there was complete blackening of the Fraser broth for food samples from G. latifolium, G. africanum, V. Amygdalina whereas broth samples of food from I. gabonensis, M. esculanta Crantz and C. citrullus did not blacken (Figure 2).
All broths were subcultured on Oxford agar before discarding, irrespective of colour change.
After incubation at 30°C for 48 h colony types that showed black colouration of media were selected and Gram stained.Colonies (Colony III, Figure 3) consistent with L. monocytogenes morphology were found to be Gram positive rods, had a doughnut like appearance on Oxford media and were about 0.5 µm in width and 1-1.5 mm in length whereas staining of aesculin producing colonies with non-conforming L. monocytogenes morphology (Colony I, Figure 3) showed Gram positive cocci shaped organisms similar to Enterococcus species that are known to be also capable of aesculin hydrolysis (HPA, 2010).
This confirms that relying on black coloration of Oxford media for selection and enumeration purposes as shown in some reports (Nwaiwu, 2015), may result in false positives (Fraser and Sperber, 1988).No isolate fitting Listeria morphology was found in food items obtained from I. gabonensis, M. esculanta Crantz and C. Citrullus.One potential L. monocytogenes isolate which was catalase positive and showed tumbling motility at 25°C (Vatanyoopaisarn et al., 2000) was selected for each food sample from the vegetables G. latifolium, G. africanum, V. amygdalina and designated LmNG1, LmNG2 and LmNG3, respectively and stored at -85°C for future use.Overall, only these 3 isolates were presumptively identified as L. monocytogenes and selected for molecular processing.

Assessing pathogenicity potential of isolated strains
To indicate if isolates are pathogenic L monocytogenes, a test was carried out using SwabSure ListeriaP kit, a chromogenic kit described by the manufacturers as being capable of differentiating pathogenic L monocytogenes and L. ivanovii from other commensal microorganisms including L. innocua.Test with the kit after swabbing the surface of a microscope glass slide containing dried cultures produced the turquoise blue colour for positive identification indicating the presence of pathogen specific phospholipase C enzyme (Notermans et al., 1991).

Molecular analysis of selected strains
According to Gasanov et al. (2005) early typing methods differentiated isolates based on phenotypic markers (motility, hemolysis, sugar fermentation) but these phenotypic typing methods are being replaced by molecular tests, which reflect genetic relationships between isolates and are more accurate.Further confirmation and determination of the serotypes and lineage of the isolated strains were ascertained by PCR multiplex serotyping and lineage analysis.These methods are still widely used (Warren et al., 2015) by other investigators.Strains representing the four major serotypes 1/2a, 1/2b, 1/2c and 4b of L. monocytogenes (Liu and Busse, 2009) and strains representing Lineage I and II were used as control (Table 1).
Genomic DNA template which had approximately 2.0 (260/280 nm) absorption ratio after Nano-Drop nucleic acid measurements was used for all PCR reactions.The result of the multiplex PCR serotyping carried out is shown in Figure 4.The control and test strains amplified prs gene (approximately 370 bp).In addition to prs, the control strain that is 4b amplified ORF2110 and ORF2819 while strains that were serotype 1/2a amplified lmo 0733.The strain that is 1/2c also amplified lmo 0733 in addition to lmo1118 and strains of 1/2b serotype amplified ORF2819.A summary table of the result gained is in Table 3.When all the amplification patterns were considered using the grouping scheme shown by Doumith et al. (2004), it was found that L. monocytogenes isolated from food item of G. latifolium and that of G. africanum were Group 3 (1/2b, 3b,7) whereas the strain from V. amygdalina food sample was grouped into Group 1 (1/2a, 3a).
There was no prevalence found in groups 2 (1/2c, 3c) or 4 (4b, 4d and 4e).The amplification patterns corresponded to the control strains Lm4 (1/2b) and Lm10403S (1/2a) which indicated that the new strains LmNG1 and LmNG2 are1/2b serotype whereas LmNG3 is serotype 1/2a.Table 3. Summary of serotype assignment based on multiplex PCR serotyping of L. monocytogenes strains isolated from Nigerian vegetables.Serotypes were assigned to the strains based on the pattern of genes amplified.Strains of known serotypes were used as control.

Strain
Multiplex The lineage classification performed showed that strains that are serotype 1/2b and 4b amplified fragments of actA1 (373bp) expected for lineage I, while strains that are serotypes 1/2a and 1/2c amplified fragments of plcB2 (564bp) expected for lineage II (Figure 5).The lineage classification of all the strains under study is shown in Table 4.There was no amplification of lineage III specific sequence showing that no lineage III strain was found in this study.Isolates LmNG1 and LmNG2 were found to be lineage 1 whereas LmNG3 was found to be lineage II.
Over 90% of the human listeriosis cases are caused by L. monocytogenes serotypes 1/2a,  1/2b and 4b strains (Laksanalamai et al., 2014) and all the serotypes are spread over at least 4 (I-IV) evolutionary lineages (Combrouse et al., 2013).Serotype 1/2a is common in the environment while serotype 1/2b and 4b are isolated from many clinical samples.Previous reports show that isolates from human listeriosis cases are common in lineage I whereas lineage II strains are prevalent in food isolates and lineage III and IV rarely occurs (Orsi et al., 2011).The occurrence of Lineage II isolate (serotype 1/2a) in sample from V. amygdalina is in line with predominance of serotype 1/2a in food and environmental samples but the strains recovered in samples from G. latifolium and G. africanum were least expected because they have the serotype and lineage associated more with clinical samples.However, recovery of Lineage 1 serotypes (mostly associated with clinical samples) from food is no longer unusual because the serotype 1/2b belonging to lineage 1 has now been isolated from other vegetables (Cordano and Jacquet, 2009;Sant'Ana et al., 2012).The isolation of L. monocytogenes from food product of G. latifolium used to prepare local salads was least expected given the anecdotal antimicrobial status of G. latifolium in south eastern Nigeria.Hence, the serotype 1/2b isolate from G. latifolium was analyzed to determine its phylogeny.After using the BLAST program and comparing sequences producing significant alignment, L. monocytogenes strain N2306 (Accession CP011004) involved in an outbreak of listeriosis in Switzerland was the top hit among 100 strains with 99% match (denotes percent identity of sequence query alignment) and the same Expect value (E-value).The full genome sequence of strain N2306 is on the NCBI data base hence a future study to carry out genome sequencing of the new isolate (Accession LN832629) for comparison would be beneficial.Yin et al. (2015) found that L. monocytogenes strains from temporally and geographically unrelated outbreaks in different countries were clustered in the same subgroup of phylogenetic tree, and suggested that they have similar virulence genes and genetic characteristics due to adaptation.Therefore, the food isolate from G. latifolium might be a potential causative agent for an outbreak because the closest relative of the isolate in a geographically unrelated region has been implicated in an outbreak of listeriosis.Also, the isolate is serotype 1/2b and the ability of this serotype to cause outbreaks has been reported, while this was listed among serotypes that caused 24 listeriosis outbreaks in the United States for the period 1998-2008 (Cartwright et al., 2013).
Presence of L. monocytogenes in only sundried vegetables from G. latifolium, G. africanum, V. amygdalina might be due to the fact that sundried leafy vegetables were not ready to use when they were purchased and may still have environmental and other post-harvest debris present which were not readily visible.Also, the vegetables still needed a final wash before it could be cooked for food.Washing reduces the soil present in fresh food and may lower the level of the organisms present.Furthermore, the prevalence of L. monocytogenes might come from post-harvest contamination and moreover, most vegetables sold in Nigeria are not pre-packed under controlled atmosphere.
Fresh vegetables are increasingly recognized as a source of food-borne outbreaks in many parts of the world (Tromp et al., 2010) and prevalence of L. monocytogenes has been severally reported in vegetables.Pre-harvest prevalence of pathogenic L. monocytogenes in selected vegetables from agricultural farm samples has been reported in a previous study (Soni et al., 2014) and many workers have found the post-harvest occurrence of L. monocytogenes in vegetables.Little et al. (2007) found that 4.8%, out of 2686 samples of pre-packed mixed vegetable salads in UK were contaminated with L. monocytogenes and Aparecida de Oliveira et al. ( 2010) used immunoassay Listeria rapid test and 16s rRNAreal-time PCR assay to detect L. monocytogenes in minimally processed leafy vegetables.

Conclusions
The molecular of the three strains isolated from different popular vegetables from Nigeria indicates that the major Listeria serotypes and evolutionary lineages associated with listeriosis could occur in popular Nigerian leafy vegetables.To the best of the knowledge of the author, this is the first time that the major L. monocytogenes molecular serotypes have been linked with popular Nigerian vegetables.The harsh process of obtaining food products from I. gabonensis, M. esculanta Crantz and C. citrullus and the intrinsic and extrinsic factors in play from farm to fork appears to be unfavourable to L. monocytogenes denoting that occurence of the organism is unlikely if good food processing and preservation practices are applied and maintained after harvest.
Popular Nigerian leaf vegetables from I. gabonensis, M. esculanta Crantz and C. citrullus could be a major vehicle for transmission of the pathogenic organism in Nigeria based on the study.Therefore, further genomic investigations and increased surveilance would be a major concern due to the finding that the closest relative of the sequenced isolate has been implicated in an outbreak of listeriosis.

Figure 3 .
Figure 3. Three colony types isolated from leaves of G. latifolium growing on Oxford formulation agar showing aesculin (I and III) and non-aesculin (II) hydrolyzing colonies after 48 h incubation at 37°C.

Table 1 .
Control strains representing the major serotypes of L. monocytogenes used in this study.

Table 2 .
Primers pairs used in this study.

Table 4 .
Lineage classification of Listeria monocytogenes strains.