3M™ Molecular Detection system versus MALDI-TOF mass spectrometry and molecular techniques for the identification of Escherichia coli 0157:H7, Salmonella spp. & Listeria spp.

Journal of Microbiological Methods 101 (2014) 33–43

Contents lists available at ScienceDirect

Journal of Microbiological Methods journal homepage: www.elsevier.com/locate/jmicmeth

3M™ Molecular Detection system versus MALDI-TOF mass spectrometry and molecular techniques for the identification of Escherichia coli 0157:H7, Salmonella spp. & Listeria spp. Marché Loff a, Louise Mare b, Michele de Kwaadsteniet a, Wesaal Khan a,⁎ a b

Department of Microbiology, Faculty of Science, Stellenbosch University, Private Bag X1, Matieland, Stellenbosch, 7602 South Africa 3M™ South Africa (Pty) Ltd., Private Bag X926, Rivonia, 2128 South Africa

a r t i c l e

i n f o

Article history: Received 12 February 2014 Received in revised form 27 March 2014 Accepted 27 March 2014 Available online 12 April 2014 Keywords: 3M™ Molecular Detection system MALDI-TOF-MS Escherichia coli 0157:H7 Listeria spp. Salmonella spp.

a b s t r a c t The aim of this study was to compare standard selective plating, conventional PCR (16S rRNA and species specific primers), MALDI-TOF MS and the 3M™ Molecular Detection System for the routine detection of the pathogens Listeria, Salmonella and Escherichia coli 0157:H7 in wastewater and river water samples. MALDI-TOF MS was able to positively identify 20/21 (95%) of the E. coli isolates obtained at genus and species level, while 16S rRNA sequencing only correctly identified 6/21 (28%) as E. coli strains. None of the presumptive positive Listeria spp. and Salmonella spp. isolates obtained by culturing on selective media were positively identified by MALDITOF and 16S rRNA analysis. The species-specific E. coli 0157:H7 PCR described in this present study, was not able to detect any E. coli 0157:H7 strains in the wastewater and river water samples analysed. However, E. coli strains, Listeria spp., L. monocytogenes and Salmonella spp. were detected using species specific PCR. Escherichia coli 0157:H7, Listeria spp. and Salmonella spp. were also sporadically detected throughout the sampling period in the wastewater and river water samples analysed by the 3M™ Molecular Detection System. MALDI-TOF MS, which is a simple, accurate and cost-effective detection method, efficiently identified the culturable organisms, while in the current study both species specific PCR (Listeria spp. and Salmonella spp.) and 3M™ Molecular Detection System could be utilised for the direct routine analysis of pathogens in water sources. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Water treatment plants routinely utilise indicator organisms such as total and faecal coliforms, enterococci and Escherichia coli (E. coli) to evaluate the quality or level of contamination in wastewater and to determine whether the water meets water quality standards (Ulrich et al., 2005; Shannon et al., 2007). However, research has shown that indicator organisms may not provide a precise indication of the number of pathogens present in water sources (Shannon et al., 2007). Moreover, the culturing methods utilised to monitor for the presence of indicator organisms are extremely time consuming and labour intensive (Lemarchand et al., 2005; Shannon et al., 2007). The standard media methods are also only able to cultivate approximately 0.001–0.1% of the entire population of microorganisms (Basu et al., 2013). In addition, the viable but non-culturable state (Alexandrino et al., 2004) of microorganisms then leads to an underestimation of pathogen numbers or a failure to isolate these pathogens (Toze, 1999). There is thus a need for detection methods that are fast, sensitive, easy to perform, costeffective and culture-independent, which will allow for the screening of wastewater samples for the presence of, not only indicator ⁎ Corresponding author. Tel.: +27 218085804; fax: +27 218085846. E-mail address: [email protected] (W. Khan).

http://dx.doi.org/10.1016/j.mimet.2014.03.015 0167-7012/© 2014 Elsevier B.V. All rights reserved.

organisms, but also specific pathogens (Toze, 1999; Straub and Chandler, 2003; Shannon et al., 2007). Molecular-based detection methods, such as the polymerase chain reaction (PCR), amplifies specific target DNA of the desired organism being investigated, and have exhibited huge potential as routine analysis tools for the detection of specific micro-organisms in different water sources (Shannon et al., 2007). The PCR technique in particular, has resolved several problems that are experienced with conventional detection methods (Jeong et al., 2007; Shannon et al., 2007; Liu, 2008; Moreno et al., 2011). A few limitations, such as low throughput and reduced sensitivity and specificity due to post-PCR analysis, have however been recorded (Toze, 1999; Call et al., 2001). The PCR technique is also unable to distinguish between viable or dead cells, which could lead to false positive results (Okoh et al., 2007; Moreno et al., 2011). In addition, when PCR based methods are applied to environmental samples, for instance wastewater, complications may arise due to the presence of inhibitory substances such as humic acids, fats and metal ions (Thompson et al., 2006) that have a negative effect on the activity of polymerase enzymes (Lemarchand et al., 2005; Moreno et al., 2011). A promising rapid detection method that is extensively used in clinical microbiology laboratories is the Matrix-assisted laser desorption/ ionisation time of flight mass spectrometry (MALDI-TOF MS) (Biswas and Rolain, 2013). While the initial capital cost for the purchasing of

34

M. Loff et al. / Journal of Microbiological Methods 101 (2014) 33–43

the equipment is high, this technique proved inexpensive when utilised for the routine biotyping of bacteria (Biswas and Rolain, 2013; Clark et al., 2013) and proteins. The MALDI-TOF MS is able to yield results in less than ten minutes by measuring the masses of proteins (Daltons) obtained from whole bacterial extracts (Lay, 2001; Prod'Hom et al., 2010) and is able to categorise bacterial organisms to the genus, species, strain as well as sub-strain level (Clark et al., 2013). Research has also shown that the MALDI-TOF MS technique can be utilised to provide information about bacterial organisms, based on their response to environmental influences, their bacterial biology and their chemistry (Lay, 2001). In addition as this method can detect a large spectrum of proteins, it is able to differentiate between species that are very closely related (Biswas and Rolain, 2013). However, when the MALDI-TOF MS is used for biotyping, the isolates are cultured on general media and non-culturable organisms present in the sample are not considered (Lay, 2001; Clark et al., 2013). Therefore the ideal solution would be to develop approaches that do not have prior culture based steps as this would definitely benefit environmental water analysis laboratories. The 3M™ Molecular Detection system is a new technology being extensively used in food analysis for the detection of Listeria spp., Salmonella spp. and E. coli 0157:H7. The detection system makes use of a loop-mediated isothermal amplification (LAMP) method for the detection of the foodborne pathogens. It then combines isothermal DNA amplification with bioluminescence detection (Crowley et al., 2012), which allows for specificity and sensitivity in the testing for pathogens. The isothermal DNA amplification uses multiple primers to distinguish different regions of the entire genome and the Bst DNA polymerase, provides fast, efficient and constant amplification of the target DNA (3M ™ Food Safety/3M Molecular Detection system Brochure, 2013). The amplification and detection occur simultaneously and continuously during the exponential phase, thus improving the specificity of the system. In a previous study conducted by David et al. (2012), it was shown that LAMP (used in 3M™ Molecular Detection system) was less susceptible than PCR to the inhibition that is caused by culture media and certain biological substances. Correspondingly a study conducted by Liu et al. (2012) showed that LAMP is more sensitive than PCR to detect DNA that is present at low concentrations. However, a prior enrichment step of approximately 24 h is required before further analysis can proceed (David et al., 2012). Numerous reports have been published on the inefficiency of Wastewater Treatment Plants (WWTP) to deal with increased influent loads. There is a thus a need to acquire detection methods that will deliver results on the level of contamination in water sources, in a timeand cost-effective manner. The primary aim of this study was thus to compare standard selective plating, PCR (16S rRNA and species specific primers), the 3M™ Molecular Detection System and MALDI-TOF MS, in order to routinely monitor the quality of wastewater and surface water. The study investigated for the presence of Listeria spp., Salmonella spp. and E. coli 0157:H7, as the 3M™ Molecular Detection technique only detects these organisms. Co-occurrence and non-co-occurrence agreements between the methods were also conducted using Statistica™ Ver. 10.0.

2. Materials and methods 2.1. Sampling sites Samples were collected for eight weeks (May to July 2013) from the influent point, aeration tank one and effluent point at the Stellenbosch Wastewater Treatment plant (WWTP) (Fig. 1) as well as from the Eerste River at a wine farm (South Africa), which is situated downstream from the WWTP. Sterile Schott Duran bottles (1 l), that were rinsed with tap water and sterilised with 70% ethanol, were used to collect the water samples. The sample bottles were transported to the laboratory on ice to maintain a low temperature.

2.2. Culturing techniques and identification of E. coli, Salmonella spp. and Listeria spp. All the samples were diluted (10−1 till 10−3) by adding 0.5 ml of each sample into 4.5 ml of sterile saline (0.9% NaCl) test tubes. Subsequently 100 μl of the undiluted sample as well as the dilution series was spread plated in duplicate onto various selective media that were used to isolate and preliminary identify the respective pathogens. These included Salmonella Shigella agar (Merck, Biolab, Wadeville, Gauteng) for the culturing of Salmonella species; University of Vermont's (UVM) modified Listeria enrichment broth base (Oxoid, Hampshire, England) supplemented with Listeria Primary Selective Enrichment Supplement for culturing Listeria spp. and mEndo Agar (Merck) for culturing E. coli. The respective plates were then incubated at 37 °C for 24–48 h. The E. coli colonies selected from m-Endo agar were re-streaked onto Chromocult™ Coliform agar for further selection. The respective plates were incubated at 37 °C for 24 h. The colonies exhibiting all the respective organism's characteristics were then restreaked onto Nutrient agar at least three times. The Nutrient agar plates were incubated at 37 °C for 24–48 h. 2.2.1. IMViC test for the confirmation of E. coli Isolates E. coli is a member of the Enterobacteriaceae family and its presence can be analysed for by using the IMViC (indole, methyl red, VogesProskauer and citrate) test. Presumptive E. coli strains selected from Chromocult™ Coliform Agar and mEndo Agar, based on colour reactions and morphologies, were then preliminary identified using the IMViC test (Harley and Prescott, 1993). 2.3. Genomic DNA extraction from isolates, wastewater and river water The isolates obtained from the respective selective media by selecting single colonies, were incubated in 5 ml Luria–Bertani broth at 37 °C overnight. Genomic DNA extractions were then performed using the ZR™ Soil microbe DNA Miniprep Kit according to manufacturer's instructions. The DNA was visualised on a 0.8% agarose gel stained with 0.5 μg/ml ethidium bromide. Electrophoresis was performed at 80 V for approximately 1 h in 1 × Tris/Borate/EDTA (TBE) buffer. For the isolation of total DNA, a total volume of 250 ml of the respective wastewater and river water samples collected throughout the sampling period were centrifuged at 8670 ×g for 20 min. The pellets obtained were incubated in 2 ml Luria–Bertani broth at 37 °C for 5 h (Kong et al., 2002). Genomic DNA extractions were then performed using the ZR™ Soil microbe DNA Miniprep Kit according to manufacturer's instructions. The DNA was visualised on a 0.8% agarose gel stained with 0.5 μg/ml ethidium bromide. Electrophoresis was performed at 80 V for approximately 1 h in 1× TBE buffer. 2.4. Conventional polymerase chain reaction of bacterial isolates Once genomic DNA had been isolated from the various isolates (obtained from the selective media previously mentioned), Polymerase Chain Reactions (PCR) were used to amplify the 16S rRNA conserved sequence of each organism (Rawlings, 1995). Each PCR mixture consisted of a final volume of 50 μl and contained 1× PCR buffer (20 mM Tris hydrochloride pH 8.4, 50 mM KCl), 4 μl MgCl2 (2.0 mM), 0.5 μl of each dNTP (0.1 mM) (Thermo Scientific), 2.5 μl of the PCR primers (0.5 μM) mentioned in Table 1 and 0.3 μl of GoTaq® Flexi DNA Polymerase (1.5 U) (Promega). The PCR amplification was performed in a Gene Cycler (Bio-Rad, USA) utilising the conditions as indicated in Table 1. The PCR products were then visualised on a 0.8% agarose gel stained with 0.5 μg/ml ethidium bromide. Electrophoresis was conducted at 80 V for approximately 1 h in 1 × TBE buffer. Once the amplification of the PCR products had been confirmed, the products were cleaned and concentrated using the DNA Clean & Concentrator™-5 Kit (Zymo Research)

M. Loff et al. / Journal of Microbiological Methods 101 (2014) 33–43

35

Fig. 1. Sampling sites at the Wastewater Treatment Plant, Stellenbosch.

as per manufacturer's instructions. The cleaned products were sent to the Central Analytical Facility (CAF) at Stellenbosch University for sequencing. Chromatograms of each sequence were examined using Finch TV v. 1.4.0 software and were aligned using DNAman™ version 4.1.2.1 software. Sequence identification was completed using the National Centre for Biotechnology Information (NCBI) and The Basic Local Alignment Search Tool (BLAST) (http://blast.ncbi.nlm.nih.gov/ Blast.cgi) to find the closest match of local similarity between isolates and the international database in GenBank, EMBL, DDBJ and PDB sequence data (Altschul et al., 1990). The sequences of representative isolates, that yielded N 97% similarity (b3% diversity), was noted as a positive result for the particular genus. Table 1 also lists the species specific primers and PCR conditions utilised for the detection of E. coli, E. coli O157:H7, Listeria spp., Listeria monocytogenes and Salmonella spp., respectively, in the wastewater and river water samples. E. coli O157:H7 (obtained from CPUT, Bellville campus), Salmonella typhimurium ATCC 14028 and Listeria monocytogenes ATCC 13932 were used as the positive controls in the respective PCRs. The specificity of each primer set was confirmed by using non target DNA extracted from all the above mentioned positive controls. After whole DNA was extracted (ZR™ Soil microbe DNA Miniprep Kit) from the wastewater and river water samples, it was used as the template DNA for the detection of the respective pathogens. The negative controls contained sterile distilled water in place of template DNA. The PCR conditions and reagents utilised for the detection of the uidA gene in E. coli O157:H7 are outlined in Tables 1 and 2. As described in Cebula et al. (1995), template DNA and GoTaq® Flexi DNA polymerase were only added after a pre-incubation step of 94 °C for 5 min which decreases the chance of non-specific binding with wild-type E. coli species (Cebula et al., 1995). After this pre-incubation step, another preincubation step of 95 °C for 5 min was added to the cycling programme for the detection of E. coli 0157:H7, with the rest of the cycling

programme then initiated as described in Cebula et al. (1995). The uidA gene of E. coli 0157:H7 has a conserved base change, a G residue in the uidA allele in comparison to the wild-type E. coli uidA that has a T residue (Cebula et al., 1995). The uidA primers were designed to have two mismatched bases, to ensure that the primers will not bind to the wild-type E. coli uidA gene. The primers for detection of E. coli 0157:H7 as outlined in Table 1 thus have the conserved G (instead of a T) at the 3′ end as well as a G (instead of an A) in the 19th position (Cebula et al., 1995). For the detection of the uidA gene in wild-type E. coli strains in the wastewater and river water samples, the reagent concentrations and procedure as outlined by Bej et al. (1991) were utilised, with an elongation step of 72 °C for 30 s (Tables 1 and 2) added to the cycling programme. For the detection of the prs gene in Listeria species in wastewater and river samples the volumes of the reagent concentrations and procedure (Tables 1 and 2) as outlined by Ryu et al. (2013) were utilised. However, for the detection of the prfA gene in L. monocytogenes in the wastewater and river samples, the volumes of the reagent concentrations were adjusted as shown in Tables 1 and 2, while the cycling programme outlined in Germini et al. (2009) was utilised. Similarly, for the detection of IpaB gene in Salmonella spp. in wastewater and river samples, the volumes of the reagent mixtures were adjusted as shown in Tables 1 and 2 and the cycling programme outlined by Kong et al. (2002) was utilised. All the PCR products were visualised on a 1.2% agarose gel stained with 0.5 μg/ml ethidium bromide. Electrophoresis was conducted at 80 V for approximately one hour in 1 × TBE buffer. The selected PCR products that gave a clear single band of the correct size were sent to CAF for purification and sequencing. When non-specific binding occurred, the expected band sizes of the respective pathogens were cut out of the 1.2% agarose gel and cleaned using the QIAquick® Gel Extraction kit. Samples were subsequently sent to CAF for sequencing. Chromatograms of each sequence were examined using Finch TV v. 1.4.0

36

M. Loff et al. / Journal of Microbiological Methods 101 (2014) 33–43

Table 1 Primer sequences and PCR cycling parameters for the detection of bacterial pathogens. Micro-organism

Primer sequence (5′–3′)

Primer name

Target gene

Product size

Polymerase chain reaction cycling conditions

Reference

Universal primers

F: CCGGATCCGTCGACAGAGTTTGAT CITGGCTCAG R: CCAAGCTTCTAGACGG ITACCTTGTTACGACTT

fDD2 rPP2

16S rRNA

1600 bp

Rawlings (1995)

E. coli strains

F: AAAACGGCAAGAAAAAGCAG R: ACGCGTGGTTACAGTCTTGCG

uidA: F uidA: R

uidA gene

147 bp

E. coli 0157:H7

F: GCGAAAACTGTGGAATTGGG R: TGATGCTCCATAACTTCCTG

UAL-754 UAR-900

uidA gene

252 bp

Listeria spp.

F: GCTGAAGAGATTGCGAAAGAAG R: CAAAGAAACCTTGGATTTGCGG

prs: F prs: R

prs gene

370 bp

Listeria F: TCATCGACGGCAACCTCGG monocytogenes R: TGAGCAACGTATCCTCCAGAGT

prfA: F prfA: R

prfA gene

217 bp

Salmonella spp.

IpaB: F IpaB: R

IpaB gene

314 bp

Initial denaturation at 94 °C for 3 min, followed by 30 cycles that consisted of denaturing at 94 °C for 30 s, primer annealing at 59 °C for 30 s and elongation at 72 °C for 1.5 min. The final elongation step was performed at 72 °C for 5 min with a holding temperature at 4 °C. Initial step of 94 °C for 1 min; 25 cycles of denaturing at 94 °C for 1 min, annealing at 94 °C for 1 min, elongation at 72 °C for 30 s with an holding temperature of 4 °C Initial step of 94 °C for 5 min, followed by denaturing at 95 °C for 5 min; 35 cycles of denaturing at 94 °C for 1.5 min, annealing at 64 °C for 1 min, elongation at 72 °C for 1.5 min with an holding temperature of 4 °C Initial denaturing at 94 °C for 5 min; 35 cycles of denaturing at 94 °C for 30 s, annealing at 60 °C for 30 s, elongation at 72 °C for 30 s, final elongation at 72 °C for 5 min, with an holding temperature of 4 °C Initial denaturing at 95 °C for 4 min; 35 cycles of denaturing at 95 °C for 1 min, annealing at 52 °C for 45 s and elongation at 72 °C for 2 min and a final elongation at 72 °C for 8 min with an holding temperature of 4 °C Initial denaturing for 2 min at 94 °C; 35 cycles of denaturing at 94 °C for 1 min, annealing at 62 °C for 1 min, elongation at 72 °C for 2.5 min with an holding temperature of 4 °C

F: GGACTTTTTAAAAGCGGCGG R: GCCTCTCCCAGAGCCGTCTGG

software and were aligned using DNAman™ version 4.1.2.1 software. Sequence identification was completed using the National Centre for Biotechnology Information (NCBI) and The Basic Local Alignment Search Tool (BLAST) (http://blast.ncbi.nlm.nih.gov/Blast.cgi) to find the closest match of local similarity between isolates and the international database in GenBank, EMBL, DDBJ and PDB sequence data (Altschul et al., 1990). 2.5. 3M™ molecular detection system The 3M™ Molecular Detection System supplied by 3M™ South Africa was used for the detection of Listeria spp., E. coli 0157:H7 and Salmonella spp. in the wastewater and river samples collected during the sampling period, as per manufacturer′s instructions. Enrichment was performed by adding 12.5 ml of sample into 225 ml of pre-warmed 3M™ buffered peptone water (BPW-ISO) for Salmonella spp. and E. coli 0157:H7 detection, and 3M™ Modified Listeria recovery broth for Listeria spp. All the enrichment media were incubated for 24 h at 37 °C (± 1 °C). After the enrichment step, 20 μl of each sample was transferred to a lysis tube and heated for 15 min at 100 °C (± 1 °C), cooled for 10 min on a cooling block and then incubated for 5 min at room temperature. Twenty microliter lysate of each sample was then transferred to a reagent tube, which is colour coded for the detection of each pathogen, i.e. pink for E. coli 0157:H7, blue for Listeria spp. and green for Salmonella spp., containing lyophilised pellets with the closed tubes transferred to a speed loader tray. This speed loader tray was placed into the 3M™ Molecular Detection System and the sample run was started (3M™ Food Safety/3M Molecular Detection system Brochure, 2013). The chemistry in the 3M™ Molecular Detection Assays is based on loop-mediated isothermal amplification (LAMP) which was developed by Notomi et al. (2000) and has subsequently been widely used for rapid and efficient DNA amplification under isothermal conditions (Parida et al., 2008; Fu et al., 2011). Multiple specially designed primers are utilised in single-temperature LAMP to recognise distinct regions of the target gene. Bioluminescence technology is then utilised to report the DNA amplification of the target organism in real-time. In this two-step enzymatic process pyrophosphate molecules, produced as a byproduct of the DNA amplification, are used to generate light.

Bej et al. (1991)

Cebula et al. (1995)

Ryu et al. (2013)

Germini et al. (2009)

Kong et al. (2002)

This light emission is then easily read by the 3M™ Molecular Detection System Instrument, which signals the detection of the target organism. 2.6. MALDI-TOF MS The selected isolates were streaked onto Nutrient agar and were incubated at 37 °C, 24 h prior to analysis by MALDI-TOF MS at the Proteomic Unit of the University of the Western Cape. Forty-seven freshlygrown culture suspensions were then directly transferred from the agar plate in triplicate as very thin layers/smears onto a MTP 384 polished-steel target (Bruker Daltonics, Bremen, Germany) using 20 μl Eppendorf tips. After air drying, each spot was further over-laid with 1.0 μl of freshly prepared CHCA matrix solution (10 mg/ml alphacyano-4-hydroxycinnamic acid, 50% acetonitrile, 2.5% trifluoroacetic acid). Thereafter, the spots were processed in the UltrafleXtreme™ MALDI-TOF mass spectrometer (Bruker Daltonics, Bremen, Germany) using the Bacterial Test Standard (BTS) (E. coli DH5α protein extract; Bruker Daltonics ref 255343) as a positive control, as well as a m/z calibrant for the MBT_fc.par method on flexControl, which was executed for linear MALDI-TOF MS analyses over m/z 2000 to 20,000 units. The analyses were performed via the MALDI Biotyper RealTime Classification (RTC) Module, which visualises AutoXecute operation of flexControl with colour coded classifications. In total, 1200 laser shots were employed using the Smartbeam-II Technology (1 kHz repo rate, 7_MBT spot parameter, 337 nm) to eject sample material from each spot at a rate of 100 per raster area (limited to b2000 μm) in a random walk mode. Positive ions generated from the MALDI-TOF were extracted with an accelerating voltage of 20 kV, and thereafter 40 kB spectra were collected at a 0.5 GS/s rate on a MCP detector with a 29× detector gain (3 kV). The RTC Module automated the matching of deionised, peak processed, smoothed and normalised mass spectra against the reference spectra in the Bruker Taxonomy/Listeria database (4110 cellular organisms, Bruker Daltonics, Bremen, Germany) (Cherkaoui et al., 2010). 2.7. Statistical analysis Agreement on the co-occurrence and non-co-occurrence among species specific PCR, the 3M™ Molecular Detection system, MALDI-

M. Loff et al. / Journal of Microbiological Methods 101 (2014) 33–43

37

Table 2 Conventional polymerase chain reaction reagent mixtures for the detection of E. coli 0157:H7, E. coli strains, Listeria spp., L. monocytogenes and Salmonella spp. in wastewater and river water samples. PCR reagent

E. coli 0157:H7 (μl)

E. coli strains (μl)

Listeria spp. (μl)

L. monocytogenes (μl)

Salmonella spp. (μl)

Final concentration

Buffer (5×) MgCl2 (25 mM) uidA F uidA R UAL-754 UAR-900 prs F prs R prfA F prfA R IpaB F IpaB R dNTP (10 mM) GoTaq polymerase (5 U/μl) Template DNA dH2O Final volume

10 4 2.5 2.5

10 4

5 2

10 4

10 4

1× 2 mM 0.5 μM 0.5 μM 0.25 μM 0.25 μM 0.2 μM 0.2 μM 0.5 μM 0.5 μM 0.1 μM 0.1 μM 0.1 mM, 0.2 mMa 1.5 U, 1 U, 1.25 Ub

a b

1.25 1.25 0.5 0.5 2.5 2.5

0.5 0.3

1 0.25

0.25 0.2

0.5 0.3

0.5 0.5 0.5 0.3

10 20.2 50

10 22.25 50

5 6.55 25

10 20.2 50

10 32.2 50

The dNTP final concentration for E. coli 0157:H7, Listeria spp., L. monocytogenes and Salmonella spp. was 0.1 mM, while for E. coli strains it was 0.2 mM. The GoTaq polymerase final concentration was 1.5 U for E. coli 0157:H7, L. monocytogenes and Salmonella spp. and 1 U and 1.25 U for E. coli strains and for Listeria spp. respectively.

TOF MS and 16S rRNA for the detection of E. coli 0157:H7, E. coli strains, Listeria spp., L. monocytogenes and Salmonella spp. in wastewater and river water samples was conducted using the statistical software package Statistica™ Ver. 10.0. 3. Results and discussion 3.1. Universal 16S rRNA analysis of E. coli, Listeria spp. and Salmonella spp. isolates E. coli, Listeria and Salmonella isolates were obtained by spread plating samples, collected from various points (May to July 2013) in the Stellenbosch Wastewater Treatment Plant (WWTP) and river water samples from the Eerste river collected at a wine farm, onto selective media. Forty-nine presumptive positive E. coli isolates were obtained from the selective media. Thereafter these isolates were subjected to the IMViC test for further selection (Harley and Prescott, 1993). Of these, 49 isolates, 34% (17 isolates), then tested positive for E. coli based on the IMViC results. It should however be noted that the identity of the isolates, that yielded positive results by the IMViC test, as well as the E. coli isolates that were negative for only one IMViC criteria, yet produced the characteristic violet blue morphology on the Chromocult® Coliform Agar, were then confirmed through universal 16S rRNA PCR with sequencing. Given the criteria above, a total of 21 presumptive E. coli isolates were analysed using 16S rRNA amplification and sequencing, with six (28%) isolates positively identified as E. coli after blast analysis. Enteric species, from the genera Enterobacter, Erwinia and Shigella, were also identified. The Shigella genus is closely related to E. coli and certain Shigella species share the same IMViC profile as E. coli (Cowan and Steel, 1961). In addition, a presumptive E. coli isolate that produced a green colour on the Chromocult® Coliform Agar and tested positive for the IMViC test, was positively identified as E. coli by 16S rRNA analysis. This indicates that the selective media utilised was not always efficient in identifying presumptive E. coli isolates. To obtain Listeria isolates, the undiluted sample and serial dilutions were spread plated onto Listeria enrichment broth base with enrichment supplement. The seventeen presumptive Listeria spp. isolates obtained were analysed using 16S rRNA amplification and sequencing. After blast analysis, no positive results were obtained for the presumptive Listeria spp. isolates. However, 16S rRNA analysis identified the isolates as E. coli, Serratia spp., Pantoea agglomerans, Pantoea rodasii,

Pseudoalteromonas, and Enterobacteriacea, amongst others. Research has however shown that Listeria spp. can enter a viable, but nonculturable state (Moreno et al., 2011) when environmental conditions are not favourable. The culturing and subsequent identification of Listeria are further complicated, due to their slow growth and the ability of other bacteria to outgrow them (Bruhn et al., 2005). A suggestion would be to use PALCAM (polymyxin–acriflavine–LiCl–ceftazidime– aesculin–mannitol) agar base and Listeria selective agar for the detection of Listeria spp., since the efficiencies of these media were 94% and 96.3%, respectively, when these selective plates were incubated for 48 h in a study conducted by Jamali et al. (2013). The presence of false negatives during the isolation of Listeria spp. were also decreased when these selective plates were used (Jamali et al., 2013). To obtain the Salmonella isolates, the undiluted sample and serial dilutions were spread plated onto Salmonella Shigella agar. Nine presumptive positive Salmonella isolates were obtained. However, after 16S rRNA analysis, none of the isolates were positively identified as Salmonella spp. The results obtained for Salmonella spp. 16S rRNA analysis were uncultured Klebsiella and Citrobacter freudii, amongst others. While the Salmonella Shigella medium can be utilised effectively to analyse clinical samples, it is hypothesised that its efficiency is reduced when environmental samples are analysed (Cox and Berrang, 2000). A solution could be to use more than one selective media in order to reduce false negative results (Cox and Berrang, 2000), but the labour and time utilised to analyse the samples will then increase (Lim et al., 1980). In a study performed by Smith et al. (2012), Rappaport Vassiliadis medium semisolid (RPVA) and Bismuth Sulfite agar (BSA) were successfully utilised for the detection as well as isolation of Salmonella spp. and these media could thus be utilised instead of Salmonella Shigella agar for environmental sample analysis. 3.2. MALDI-TOF MS The MALDI-TOF MS detection method is used in microbiology to analyse specific peptides or proteins that are desorbed from bacteria, spores and viruses (Marvin et al., 2003). This detection method is designed to generate profile spectra of small proteins and peptides that are mostly derived from the bacterial ribosome (Sparbier et al., 2012). Forty-seven isolates (presumptive E. coli strains, Salmonella spp. and Listeria spp.), obtained from the selective media were subjected to MALDI-TOF MS to identify the respective strains.

38

M. Loff et al. / Journal of Microbiological Methods 101 (2014) 33–43

The MALDI-TOF MS spectra analysis reveals a characteristic peak pattern for each bacterial isolate (Sparbier et al., 2012). The molecular masses and charge densities of the components present in the biological sample are represented by the mass spectral peaks (Cherkaoui et al., 2010) (Fig. 2). The acquired spectrum is compared to a library database and this allows for the identification of the bacterial species (Sparbier et al., 2012). Illustrated in Fig. 2 is the characteristic mass spectrum for one of the presumptive E. coli isolates that was positively identified as an E. coli strain. For the MALDI-TOF MS, identification at genus level occurs when the identification score is ≥1.7 and ≤2 and at species level when the identification score is ≥ 2 (Prod'Hom et al., 2010; Sparbier et al., 2012). The peaks illustrated in Fig. 2 represent the mass-tocharge (m/z) values that indicate the molecular masses and charge densities of the proteins present in the E. coli isolate (Cherkaoui et al., 2010; Clark et al., 2013). The higher the peak weights, the greater the chance that the isolate will be identified at species level (Jadhav et al., 2014). Of the 21 presumptive E. coli isolates, MALDI-TOF MS identified 20 (95%) at genus and species level as E. coli strains, with one isolate identified as Citrobacter freundii. Similarly, a study conducted by Siegrist et al. (2007) showed that MALDI-TOF MS was successful in identifying environmental E. coli isolates. In comparison, 16S rRNA only identified six of the 21 (29%) presumptive E. coli isolates as E. coli strains. However, of the 17 presumptive Listeria isolates that were tested by MALDI-TOF MS, none were identified as Listeria strains. The 16S rRNA sequencing results also indicate that no Listeria spp. could be isolated and identified. The MALDI-TOF MS identified the presumptive Listeria isolates as Citrobacter freundii, E. coli and Enterobacter spp., respectively. In a study conducted by Barbuddhe et al. (2008) and Jadhav et al., 2014 Listeria spp. as well as L. monocytogenes was identified, respectively, indicating that the MALDI-TOF MS is able to identify Listeria at genus and species level. It should thus be noted that the selective media utilised in the current study, could have limited the selection of Listeria isolates. Presumptive positive Listeria colonies produce a straw-like colour on Listeria enrichment broth base with enrichment supplement (Oxoid, Hampshire, England). However, the variation of colour for the isolates growing on the media range from pale yellow to cream, which could have complicated the selection of the organisms. For the nine presumptive Salmonella isolates that were subjected to MALDI-TOF MS, none were identified as Salmonella spp. These isolates were identified as Citrobacter freundii, Citrobacter braakii, Citrobacter youngae and Providencia alcalifaciens, which is similar to the 16S rRNA sequencing results where no Salmonella species were identified. It should however, be noted, that many pathogenic organisms (e.g. E. coli, Listeria monocytogenes, Helicobacter pylori and Salmonella typhimurium, among others) may enter a viable but non culturable state (VBNC), making them difficult to culture (Kong et al., 2002; Oliver, 2005, 2010; van Frankenhuyzen et al., 2013). In a study conducted by Sparbier et al. (2012), Salmonella spp. were however, successfully identified utilising the MALDI-TOF MS technique. Comparison of the 16S rRNA sequencing analysis of the isolates obtained from culturing on selective media to the MALDI-TOF MS analysis of the same strains, shows that MALDI-TOF MS was more accurate in identifying presumptive E. coli isolates (Siegrist et al., 2007). Listeria and Salmonella spp. were not identified using the either of these techniques. However it is assumed that the selective media utilised in the study were not ideal for the selection of Listeria and Salmonella spp. from wastewater and river water samples, as false positives were obtained. 3.3. 3M™ molecular detection system The 3M™ Molecular Detection System was utilised to identify the respective pathogens, Listeria spp., Salmonella spp. and E. coli 0157:H7, directly from the water samples collected from the various points in the WWTP and Eerste River during the sampling period. This system yields qualitative results, and merely indicates the presence or absence of the

particular pathogen. Positive results are obtained in real time in 75 min with negative results presented at the end of the run (3M™ Food Safety/ 3M Molecular Detection system Brochure, 2013). In the current study however, enrichment media was utilised to improve the isolation rate in an attempt to recover the bacterial strains before detection utilising the 3M™ Molecular Detection System. The 3M™ Modified Listeria recovery broth provided as part of the 3M™ Molecular Detection System is specific for the enrichment of Listeria strains. However, the media (Buffered Peptone water) that is used for E. coli and Salmonella is not specific for only these two bacteria and will allow for the growth of a wide range of organisms. In general enrichment media were thus utilised in the pre-incubation step to allow healthy and injured strains to grow to detectable limits, followed by strain specific typing (Ryser et al., 1996; Baylis et al., 2000). For each sampling time a sample was then collected from the influent point, aeration tank one and effluent point of the WWTP and from the Eerste River, resulting in a total of four samples per sampling session and 32 samples over the 8 week period. This is the first time that this system was utilised to analyse river and wastewater samples, as it has been specifically designed for the detection of Listeria spp., Salmonella spp. and E. coli 0157:H7 in food samples (3M™ Food Safety/3M Molecular Detection system Brochure, 2013). Fig. 3 (results for sampling three) represents an organogram which is generated after a sampling session. Colour coded reagent tubes, placed into the tray of the 3M™ Molecular Detection system, are used to differentiate between the three detection assays, as can be seen in Fig. 3, with green representing the Salmonella spp. assay (lane 1), pink representing the E. coli O157:H7 assay (lane 3) and blue representing the Listeria spp. assay (lane 5) (3M™ Food Safety/3M Molecular Detection system Brochure, 2013). The reagent control, which represents the positive controls, and negative controls, were loaded in lanes A and B (Fig 3.), respectively, for each assay. The matrix controls for the BPW-ISO broth (Salmonella spp. and E. coli 0157:H7 detection) and the 3M™ Modified Listeria recovery broth (Listeria spp. detection) were loaded into lanes 7 and 9, respectively. The matrix control is required to determine if any inhibition takes place in the respective samples, as inhibition interferes with the amplification step during analysis of the three step process (step one: enrichment, step two, lysis step, step three: amplification step) utilised in the 3M™ Molecular Detection system (3M Confidential). The influent samples were then loaded in row C, aeration samples loaded in row D, effluent samples were loaded in row E and river water samples were loaded in row F. As indicated in Fig. 3, the matrix control is visible as a blue colour, indicating that no inhibition occurred. In the test samples of the respective detection assays (lanes 1, 3 and 5), the plus (red colour) indicates that the respective pathogen was present and a negative (green colour) implies that the pathogen was not present in the sample. As mentioned previously, lanes 1, 3 and 5 are the results of the Salmonella spp., E. coli O157:H7 and Listeria spp. detection assays, respectively, with the influent, aeration, effluent and river results of the different assays represented in rows C, D, E and F respectively, for each corresponding lane. This loading order was maintained for all the samples throughout the sampling period (sampling one to eight). Of the 32 water samples analysed using the 3M™ Molecular Detection System, E. coli 0157:H7 was detected 11 times (34%) throughout the sampling period (sampling one to sampling eight) (Table 3). For sampling one and two, E. coli 0157:H7 was not detected in the influent and aeration tank samples, however it was present in the effluent sample for both sampling periods and varying results for the river water samples (Table 3). E. coli 0157:H7 was also observed in the influent sample for samplings three, four and six, with sporadic detection in the aeration tank, effluent and river water samples for these sampling periods (Table 3). No E. coli 0157:H7 was detected in sampling sessions five and eight. In contrast, Listeria spp. were detected in 27 (84%) of the 32 water samples collected in sampling one to eight (Table 3). For four of the

M. Loff et al. / Journal of Microbiological Methods 101 (2014) 33–43

39

Fig. 2. Mass spectrum (MS) profile of a presumptive E. coli isolate positively identified as an E. coli strain.

sampling periods [sampling two, three (Fig. 3), six and eight] Listeria spp. were detected in all the sampling points (influent, aeration tank, effluent and river water samples). However, no Listeria spp. were identified in the effluent sample for sampling sessions one, four, five and seven, with no Listeria spp. also detected by 3M™ analysis in the river water sample of session seven (Table 3). Salmonella spp., were detected in 8 (25%) of the 32 water samples collected in sampling one to eight (Table 3) using the 3M™ Molecular Detection system. For sampling one to three (Fig. 3), Salmonella spp. were consistently detected in the influent sample, while results for the aeration, effluent and river samples varied. From sampling 4 to 5, Salmonella spp. were only detected in the influent sample (Table 3). For sampling six to eight (Table 3), no Salmonella spp. were identified in the sampling points (influent, aeration, effluent and river water samples). 3.4. Species specific PCR 3.4.1. E. coli strains and E. coli 0157:H7 Species specific primers with PCR were used to screen for the wildtype uidA gene (Bej et al., 1991) and the uidA gene with the conserved base modification (Cebula et al., 1995), present in E. coli strains and

E. coli O157:H7, respectively. Analysis was performed on all the water samples collected from the WWTP and the Eerste River throughout the sampling period. Based on species specific analysis, E. coli 0157:H7 was not detected in any of the samples analysed for sampling one to eight. In contrast, after BLAST analysis (Table 4), the wild-type uidA gene, present in E. coli strains, was detected in 31 of the 32 water samples (96%) collected during the sampling period (sampling one to eight). Only one sampling point (aeration tank), in sampling four, did not yield the expected E. coli PCR product band size of 147 bp. Furthermore the results indicate that the wastewater treatment was not effective in removing the E. coli strains from the wastewater and that E. coli strains was, at the time of sampling, being released into the environment via the river systems. As previously mentioned, the 3M™ Molecular Detection System was able to detect E. coli 0157:H7 in 34% of the samples, in comparison to the species specific PCR utilising the uidA gene, where no E. coli 0157:H7 was identified. This indicates that the 3M™ Molecular Detection system was more effective in detecting the presence of E. coli 0157:H7. A study performed by David et al. (2012), also indicated the efficiency of the 3M™ Molecular detection system for E. coli 0157:H7 detection, where this virulent strain was positively identified in 66 food samples tested.

Fig. 3. A representation of a run report for Sampling 3, with RC, NC and MC representing the reaction control, negative control and matrix control, respectively. Positive results are shown in red (+) and negative results in green (−).

40

M. Loff et al. / Journal of Microbiological Methods 101 (2014) 33–43

Species specific primers and PCR were used to screen for the prs gene (band size of 370 bp) present in Listeria spp. in the collected wastewater and river water samples collected throughout the sampling period (Germini et al., 2009). After sequencing and BLAST analysis, this gene was detected in 11 of the 32 water samples (34%) collected during the sampling period (sampling one to eight, Table 4). For Listeria spp. positive results were obtained for the river sample of sampling one and for the influent and aeration samples of sampling three and six. All the samples in sampling seven were positive for the detection of Listeria spp., while Listeria spp. were only detected in the aeration and river samples of sampling eight (Table 4). When the results of the species specific PCR of Listeria spp. are compared to the 3M™ Molecular Detection System, a higher percentage of positive results [27 out of the 32 (84%)] was obtained with the 3M™ Molecular Detection System, compared to the Listeria species specific PCR, where only 11/32 (34%) positive results were obtained. These results indicate that the 3M™ Molecular Detection System was more sensitive in detecting Listeria spp. in wastewater and river water samples. For L. monocytogenes detection, primers targeting specifically the prfA gene (band size of 217 bp) were utilised (Germini et al., 2009). Sequencing and BLAST analysis of PCR products yielding the band size of 217 bp, showed that L. monocytogenes was detected in 15 of the 32 water samples (46%), collected during the sampling period (sampling one to eight) (Table 4). This primer set was thus more effective in identifying the specific Listeria strain, while overall the 3M™ Molecular Detection System was more effective in the preliminary identification of Listeria spp.

were always detected in the influent sample, while the detection results for the aeration, effluent and river samples varied. According to Kong et al. (2002) the primer sets utilised in the current study can detect the majority of universal Salmonella spp. Salmonella spp. were also detected more often in the wastewater and river water samples when utilising species specific PCR, than when utilising the 3M™ Molecular Detection System (8/32 samples, 25%). This could be due to false negatives, which occur when you transfer less than the amount of lysate to the sample reagent tube, when you transfer resin to the reagent tube or when cross-contamination occurs (3M™ Confidential). 3.5. Statistical analyses In Table 5 the occurrences among species specific PCR, 3M™ Molecular Detection System, MALDI-TOF MS and 16S rRNA for the detection of E. coli strains, Listeria spp. and Salmonella spp. were compared pairwise using the statistical software package Statistica™ Ver. 10.0. The percentage of total agreement was calculated by adding the percentage of co-occurrence and non-co-occurrence for each of the pairwise comparisons (Ahmed et al., 2013) as indicated in Table 5. For the pairwise comparison of the 3M™ Molecular Detection system and species specific PCR for L. monocytogenes, a co-occurrence agreement of 40% was obtained. However, the 3M™ Molecular Detection System and the species specific PCR for Salmonella spp. had the lowest percentage (21%) co-occurrence agreement. In addition, the co-occurrence values for the pairwise comparisons of 3M™ Molecular Detection System and species specific PCR for E. coli strains and Listeria spp. did not vary significantly (34% and 37%, respectively) (Table 5). For the pairwise comparisons of MALDI-TOF MS and the 16S rRNA PCR for E. coli strains, a high total agreement percentage of 99% was obtained, with a co-occurrence agreement of 61% recorded (Table 5). No pairwise comparisons were conducted for the Listeria spp. and Salmonella spp., as no positive isolates were obtained for these pathogens.

3.4.2. Salmonella spp. Species specific primers and PCR were used to screen for the IpaB gene (band size of 314 bp) present in Salmonella spp. in the collected wastewater and river water samples (Kong et al., 2002). After sequencing and BLAST analysis, the IpaB gene present in Salmonella spp., was detected in 24 of the 32 water samples (75%) collected during the sampling period (sampling one to eight, Table 4). Salmonella spp., with Salmonella enterica identified most often using blast analysis, was detected in all the water samples collected during sampling one and three, while for sampling two and four Salmonella spp. were present in the influent, effluent and river water samples, and not in the samples collected in the aeration tanks. For sampling five to eight Salmonella spp.

4. Conclusion When MALDI-TOF MS was compared to 16S rRNA analysis, for the identification of the E. coli isolates cultured on the selective media, MALDI-TOF MS was able to identify 95% of the presumptive E. coli

Table 3 Analysis of E. coli 0157:H7, Listeria spp., and Salmonella spp. in wastewater and river water samples by 3M™ Molecular Detection analysis for sampling 1 to sampling 8.

Microorganism

No. of isolates

E. coli o157:H7

32

Listeria

32

Salmonella

Sampling period

32

1

2

3

4

5

6

7

8

I

A

E

R

I

A

E

R

I

A

E

R

I

A

E

R

I

A

E

R

I

A

E

R

I

A

E

R

I

A

E

R

-

-

+

-

-

-

+

+

+

-

+

+

+

-

-

-

-

-

-

-

+

-

+

+

-

-

-

+

-

-

-

-

+

+

-

+

+

+

+

+

+

+

+

+

+

+

-

+

+

+

-

+

+

+

+

+

+

+

-

-

+

+

+

+

+

-

-

-

+

+

-

-

+

+

-

+

+

-

-

-

+

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

I, A, E, R, indicate Influent, aeration, effluent and river, respectively. (+) indicate positive result, (−) indicate negative result.

M. Loff et al. / Journal of Microbiological Methods 101 (2014) 33–43

41

Table 4 Analysis of E. coli 0157:H7, E. coli strains, Listeria spp., Listeria monocytogenes and Salmonella spp. in wastewater and river water samples by species specific PCR for sampling 1 to sampling 8.

Microorganism

No. of isolates

I

E. coli

32

+

E. coli O157:H7

32

Listeria

32

Listeria monocytogenes

32

Salmonella

Sampling period

32

1

2

3

4

5

6

7

8

A

E

R

I

A

E

R

I

A

E

R

I

A

E

R

I

A

E

R

I

A

E

R

I

A

E

R

I

A

E

R

+

+

+

+

+

+

+

+

+

+

+

+

-

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

-

+

-

-

-

-

+

+

-

-

-

-

-

-

-

-

-

-

+

+

-

-

+

+

+

+

-

+

-

+

+

+

-

-

+

-

-

-

+

+

-

+

+

+

-

-

+

-

-

-

+

+

-

-

+

+

+

-

+

-

-

-

+

+

+

+

+

-

+

+

+

+

+

+

+

-

+

+

+

-

+

-

+

+

-

+

+

+

+

-

+

-

+

-

I, A, E, R, indicate influent, aeration, effluent and river, respectively. (+) indicate positive result, (−) indicate negative result.

isolates tested, while 16S rRNA sequencing only detected 28%. However, according to statistical analysis, there is no significant difference (p N 0.05) between the two methods. For the Salmonella spp. and Listeria spp., the two methods could not be compared, due to the inability to culture these organisms on the selected media used in this study. Alternative media such PALCAM agar and Listeria selective agar (Jamali et al., 2013) could be used for the culturing of Listeria spp. and Rappaport Vassiliadis medium semisolid (RPVA) and Bismuth Sulfite agar (BSA) for the culturing of Salmonella spp. (Smith et al., 2012). While MALDI-TOF MS is able to identify presumptive isolates at genus and species level, the disadvantage of both MALDI-TOF MS and 16S rRNA is that these methods are dependent on the culturing of the pathogens on selective media and therefore only culturable organisms can be detected and monitored. Micro-organisms that enter a viable but not culturable state (VBNC) can therefore not be detected by these two methods (Kong et al., 2002; Oliver, 2005, 2010; van Frankenhuyzen et al., 2013). The VBNC state in micro-organisms usually occurs under stressful conditions such as changing oxygen and temperature fluctuations, which generally occurs during wastewater treatment (Oliver, 2010). When species specific PCR is compared to the 3M™ Molecular Detection System, for E. coli 0157:H7, the 3M™ Molecular Detection System was able to detect E. coli 0157:H7 in 11 samples throughout the sampling period, while no E. coli 0157:H7 were detected using species specific PCR. When E. coli strain PCR is compared to the 3M™ Molecular Detection System, results showed that even though low numbers of E. coli 0157:H7 was present, E. coli strains were still abundantly present

in the wastewater and river water samples. For Listeria spp. and L. monocytogenes, 3M™ analysis detected 84% Listeria spp. in comparison to Listeria spp. and L. monocytogenes species specific PCR that only detected 46% and 34%, respectively. This indicates that the 3M™ Molecular Detection System is not only efficient for the detection of E. coli 0157:H7, but also in detecting Listeria spp. However, for Salmonella spp., the 3M™ analysis was less effective than species specific PCR in detecting Salmonella spp. The 3M™ Molecular Detection System is a very simple detection method and with its colour coded tubes (3M™ Food Safety/3M Molecular Detection system Brochure, 2013) for the respective pathogens that are used during analysis, the chances of human error are decreased, while for species specific PCR human errors can occur more frequently. This system also reveals real-time results in 75 min (3M™ Food Safety/3M Molecular Detection system Brochure, 2013), while species specific PCR takes between 4 and 5 h, including the postanalysis step, to visualise the results. However, a 24 h pre-incubation step is required for 3M™ analysis. Thus from the results of the 3M™ Molecular Detection System used in this study on river and wastewater samples and previous studies (Crowley et al., 2012; David et al., 2012) it is evident that this system is effective in routinely detecting the pathogens E. coli 0157: H7 and Listeria spp. Acknowledgements The authors would like to acknowledge the following persons and institutes for their contributions to this project: Mr Vivian Kloppers, at

42

M. Loff et al. / Journal of Microbiological Methods 101 (2014) 33–43

Table 5 Agreement on the co-occurrence and non-co-occurrence among species specific PCR, the 3M™ Molecular Detection System, MALDI-TOF MS and 16S rRNA for the detection of E. coli 0157: H7, E. coli strains, Listeria spp., L. monocytogenes and Salmonella spp. in wastewater and river water samples. Pairwise comparison

Co-occurrence agreement (%)

Non-co-occurrence agreement (%)

Total agreement (%)

Total disagreement (%)

3M vs species specific PCR (E. coli strains) 3M vs species specific PCR (Listeria spp.) 3M vs species specific PCR (L. monocytogenes) 3M vs species specific PCR (Salmonella spp.) ^MALDI-TOF MS vs 16S rRNA (E. coli strains)

34% 37% 40% 21% 61%

62% 46% 43% 53% 38%

96% 83% 83% 74% 99%

4% 17% 17% 26% 1%

3M — abbreviation for the 3M™ Molecular Detection system. ^Results for E. coli only are presented and Listeria spp. and Salmonella spp. were not detected using the MALDI-TOF MS and 16S rRNA analysis.

the Stellenbosch Wastewater Treatment Plant and the wine farm for allowing us to sample at the sites; Cape Peninsula University of Technology for supplying the E. coli O157:H7 control strain. References 3M Food Safety/3M Molecular Detection system Brochure, 2013. [S.a] [Online] Available: http://www.slideshare.net/3MFoodSafety/3m-food-safety-3m-molecular-detectionsystem-brochure (2013, April 13). Ahmed, W., Sritharan, T., Palmer, A., Sidhu, J.P.S., Tozea, S., 2013. Evaluation of bovine feces-associated microbial source tracking markers and their correlations with fecal indicators and zoonotic pathogens in a Brisbane, Australia, reservoir. Appl. Environ. Microbiol. 79, 2682–2691. Alexandrino, M., Grohmann, E., Szewzyk, U., 2004. Optimization of PCR-based methods for rapid detection of Campylobacter jejuni, Campylobacter coli and Yersinia enterocolitica serovar 0:3 in wastewater samples. Water Res. 38, 1340–1346. Altschul, S.F., Gish, W., Miller, W., Myers, E.W., Lipman, D.J., 1990. Basic local alignment search tool. J. Mol. Biol. 215, 403–410. Barbuddhe, S.B., Maier, T., Schwarz, G., Kostrzewa, M., Hof, H., Domann, E., Chakraborty, T., Hain, T., 2008. Rapid identification and typing of Listeria species by matrix-assisted laser desorption ionization-time of flight mass spectrometry. Appl. Environ. Microbiol. 74, 5402–5407. Basu, S., Chatterjee, S., Bandyopadhyay, A., Sarkar, K., 2013. Potential application of superparamagnetic nanoparticles for extraction of bacterial genomic DNA from contaminated food and environmental samples. J. Sci. Food Agric. 93, 788–793. Baylis, C.L., MacPhee, S., Marin, K.W., Humphrey, T.J., Betts, R.P., 2000. Comparison of three enrichment media for the isolation of Campylobacter spp. from foods. J. Appl. Microbiol. 89, 884–891. Bej, A.K., DiCesare, J.L., Haff, L., Atlas, R.M., 1991. Erratum: detection of Escherichia coli and Shigella spp. in water by using the polymerase chain reaction and gene probes for uid. Appl. Environ. Microbiol. 57, 1013–1017. Biswas, S., Rolain, J., 2013. Use of MALDI-TOF mass spectrometry for identification of bacteria that are difficult to culture. J. Microbiol. Methods 92, 14–24. Bruhn, J.B., Vogel, B.F., Gram, L., 2005. Bias in the Listeria monocytogenes enrichment procedure: lineage 2 strains outcompete lineage 1 strains in University of Vermont selective enrichments. Appl. Environ. Microbiol. 71, 961–967. Call, D.R., Brockman, F.J., Chandler, D.P., 2001. Detecting and genotyping Escherichia coli O157:H7 using multiplexed PCR and nucleic acid microarrays. Int. J. Food Microbiol. 67, 71–80. Cebula, A.T., Payne, W.L., Feng, P., 1995. Simultaneous identification of strains of Escherichia coli serotype O157:H7 and their shiga-like toxin type by mismatch amplification mutation assay-multiplex PCR. J. Clin. Microbiol. 33, 248–250. Cherkaoui, A., Hibbs, J., Emonet, S., Tangomo, M., Girard, M., Francois, P., Schrenzel, J., 2010. Comparison of two matrix-assisted laser desorption ionization-time of flight mass spectrometry methods with conventional phenotypic identification for routine identification of bacteria to the species level. J. Clin. Microbiol. 48, 1169–1175. Clark, C.G., Kruczkiewicz, P., Guan, C., McCorrister, S.J., Chong, P., Wylie, J., van Caeseele, P., Tabor, H.A., Snarr, P., Gilmour, M.W., Taboada, E.N., Westmacott, G.R., 2013. Evaluation of MALDI-TOF mass spectroscopy methods for determination of Escherichia coli pathotypes. J. Microbiol. Methods 94, 180–191. Cowan, S.T., Steel, K.J., 1961. Diagnostic tables for the common medical bacteria. J. Hyg. 59, 357–373. Cox, N.A., Berrang, M.E., 2000. Inadequacy of selective plating media in field detection of Salmonella. J. Appl. Poult. Res. 9, 403–406. Crowley, E., Bird, P., Fisher, K., Juenger, M., Huffman, T., Boyle, M., Benzinger, M.J., Agin, J., Goins, D., Zook, C., David, J., 2012. P1-102: comparative evaluation of the 3M™ molecular detection assay Escherichia coli O157:H7 for the detection of Escherichia coli O157:H7 in foods. [S.a] [Online] Available: http://solutions. 3mdeutschland.de/3MContentRetrievalAPI/BlobServlet?lmd=1354703207000&locale=de_DE&assetType=MMM_Image&assetId=1319243049213&blobAttribute= ImageFile (15 April 2013). David, J.M., Rosauer, M.L., Zook, C.D., 2012. Performance of the 3M™ molecular detection system for the detection of Salmonella and Escherichia coli O157: H7. [S.a] [Online] Available: http://solutions.3mdeutschland.de/3MContentRetrievalAPI/BlobServlet? lmd=1354703207000&locale=de_DE&assetType=MMM_Image&assetId= 1319243049213&blobAttribute=ThumbnailImage (15 April 2013). Fu, S., Qu, G., Guo, S., Ma, L., Zhang, N., Zhang, S., Gao, S., Shen, Z., 2011. Applications of loopmediated isothermal DNA amplification. Appl. Biochem. Biotechnol. 163 (7), 845–850.

Germini, A., Masola, A., Carnevali, P., Marchelli, R., 2009. Simultaneous detection of Escherichia coli O175:H7, Salmonella spp., and Listeria monocytogenes by multiplex PCR. Food Control 20, 733–738. Harley, J.P., Prescott, M.L., 1993. Laboratory Exercises in Microbiology. Wm. C. Brown Publishers, United States of America. Jadhav, S., Sevior, D., Bhave, M., Palombo, E.A., 2014. Detection of Listeria monocytogenes rom selective enrichment broth using MALDI-TOF Mass Spectrometry. J. Proteome 97, 100–106. Jamali, H., Chai, L.C., Thong, K.L., 2013. Detection and isolation of Listeria spp. and Listeria monocytogenes in ready-to-eat foods with various selective culture media. Food Control 32, 19–24. Jeong, S.K., Gang, G.L., Jong, S.P., Yong, H.J., Hyo, S.K., Soo, B.K., Young, S.N., Kwon, S., 2007. A novel multiplex PCR assay for rapid and simultaneous detection of five pathogenic bacteria: Escherichia coli O157:H7, Salmonella, Staphylococcus aureus, Listeria monocytogenes, and Vibrio parahaemolyticus. J. Food Protect. 70, 1656–1662. Kong, R.Y.C., Lee, S.K.Y., Law, T.W.F., Law, S.H.W., Wu, R.S.S., 2002. Rapid detection of six types of bacterial pathogens in marine waters by multiplex PCR. Water Res. 36, 2802–2812. Lay, J.O., 2001. MALDI-TOF mass spectrometry of bacteria. Mass Spectrom. Rev. 20, 172–194. Lemarchand, K., Berthiaume, F., Maynard, C., Harel, J., Payment, P., Bayardelle, P., Masson, L., Brousseau, R., 2005. Optimization of microbial DNA extraction and purification from raw wastewater samples for downstream pathogen detection by microarrays. J. Microbiol. Methods 63, 115–126. Lim, Y.S., Jegathesan, M., Ong, M.Y., 1980. Comparison of six selective media for recovering Salmonella. Malays. J. Pathol. 3, 31–33. Liu, D., 2008. Preparation of Listeria monocytogenes specimens for molecular detection and identification. Int. J. Food Microbiol. 122, 229–242. Liu, A., Guan, G., Du, P., Gou, H., Zhang, J., Lui, Z., Ma, M., Ren, Q., Liu, J., Yang, J., Li, Y., Niu, Z., Bai, Q., Yin, H., Luo, J., 2012. Rapid identification and differentiation of Theilera sergenti and Theileria sinsensis using a loop-mediated isothermal amplification (LAMP) assay. Vet. Parasitol. 191, 15–22. Marvin, L.F., Roberts, M.A., Fay, L.B., 2003. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry in clinical chemistry. Clin. Chim. Acta 337, 11–21. Moreno, Y., Ballesteros, L., García-Hernández, J., Santiago, P., González, A., Ferrús, M.A., 2011. Specific detection of viable Listeria monocytogenes in spanish wastewater treatment plants by fluorescent in situ hybridization and PCR. Water Res. 45, 4634–4640. Notomi, T., Okayama, H., Masubuchi, H., Yonekawa, T., Watanabe, K., Amino, N., Hase, T., 2000. Loop-mediated isothermal amplification of DNA. Nucleic Acids Res. 28 (12), E63. Okoh, A.I., Odjadjare, E.E., Igbinosa, E.O., Osode, A.N., 2007. Wastewater treatment plants as a source of microbial pathogens in receiving watersheds. Afr. J. Biotechnol. 6, 2932–2944. Oliver, J.D., 2005. The viable but nonculturable state in bacteria. J. Microbiol. 43, 93–100. Oliver, J.D., 2010. Recent findings on the viable but nonculturable state in pathogenic bacteria. FEMS Microbiol. Rev. 34, 415–425. Parida, M., Sannarangaiah, S., Dash, P.K., Rao, P.V., Morita, K., 2008. Loop mediated isothermal amplification (LAMP): a new generation of innovative gene amplification technique; perspectives in clinical diagnosis of infectious diseases. Rev. Med. Virol. 18 (6), 407–421. Prod'Hom, G., Bizzini, A., Durussel, C., Bille, J., Greub, G., 2010. Matrix-assisted laser desorption ionization-time of flight mass spectrometry for direct bacterial identification from positive blood culture pellets. J. Clin. Microbiol. 48, 1481–1483. Rawlings, D.E., 1995. Restriction enzyme analysis of 16S rRNA genes for the rapid identification of Thiobacillus ferooxidans, Thiobacillus thiooxidans, Leptospirillum ferooxidans strains in leaching environments. In: Jerez, C.A., Vargas, T., Toledo, H., Wiertz, J.V. (Eds.), International biohydrometallurgical symposium. University of Chile Press, Santiago, pp. 9–17. Ryser, E.T., Arimi, S.M., Bunduki, M.-C., Donnelly, C.W., 1996. Recovery of Different Listeria Ribotypes from Naturally Contaminated, Raw Refrigerated Meat and Poultry Products with Two Primary Enrichment Media. Appl. Environ. Microbiol. 62 (5), 1781–1787. Ryu, J., Park, S.H., Yeom, Y.S., Shrivastav, A., Lee, S., Kim, Y., Kim, H., 2013. Simultaneous detection of Listeria species isolated from meat processed foods using multiplex PCR. Food Control 32, 659–664. Shannon, K.E., Lee, D., Trevors, J.T., Beaudette, L.A., 2007. Application of real-time quantitative PCR for the detection of selected bacterial pathogens during municipal wastewater treatment. Sci. Total Environ. 382, 121–129. Siegrist, T.J., Anderson, P.D., Huen, W.H., Kleinheinz, G.T., McDermott, C.M., Sandrin, T.R., 2007. Discrimination and characterization of environmental strains of Escherichia coli by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS). J. Microbiol. Methods 68, 554–562.

M. Loff et al. / Journal of Microbiological Methods 101 (2014) 33–43 Smith, S., Opere, B., Fowora, M., Aderohunmu, A., Ibrahim, R., Omonigbehin, E., Bamidele, M., Adeneye, A., 2012. Molecular characterization of Salmonella spp. directly from snack and food sold in Lagos, Nigeria. Southeast Asian J. Trop. Med. Public Health 43, 718–723. Sparbier, K., Weller, U., Boogen, C., Kostrzewa, M., 2012. Rapid detection of Salmonella spp. by means of a combination of selective enrichment broth and MALDI-TOF MS. Eur. J. Clin. Microbiol. Infect. Dis. 31, 767–773. Straub, T.M., Chandler, D.P., 2003. Towards a unified system for detecting waterborne pathogens. J. Microbiol. Methods 53, 185–197. Thompson, D.E., Rajal, V.B., De Batz, S., Wuertz, S., 2006. Detection of Salmonella spp. in water using magnetic capture hybridization combined with PCR or real-time PCR. J. Water Health 4, 67–75.

43

Toze, S., 1999. PCR and the detection of microbial pathogens in water and wastewater. Water Res. 33, 3545–3556. Ulrich, H., Klaus, D., Irmgard, F., Annette, H., Juan, L., Regine, S., 2005. Microbiological investigations for sanitary assessment of wastewater treated in constructed wetlands. Water Res. 39, 4849–4858. van Frankenhuyzen, J.K., Trevors, J.T., Flemming, C.A., Lee, H., Habash, M.B., 2013. Optimization, validation, and application of a real-time PCR protocol for quantification of viable bacterial cells in municipal sewage sludge and biosolids using reporter genes and Escherichia coli. J. Ind. Microbiol. Biotechnol. 1–11.