International Journal of Food Microbiology 141 (2010) S147S155

Contents lists available at ScienceDirect

International Journal of Food Microbiology

j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / i j f o o d m i c r o

Culture-independent techniques applied to food industry water surveillance A case study

Jessica Varela Villarreal , Thomas Schwartz, Ursula Obst
Karlsruhe Institute of Technology (former: Forschungszentrum Karlsruhe), Institute of Functional Interfaces (IFG), Microbiology of Natural and Technical Interfaces Department, P.O. Box 3640, 76021 Karlsruhe, Germany

a r t i c l e
Keywords: Drinking water Food industry Monitoring Pathogen PCR-DGGE Real-time PCR

i n f o

a b s t r a c t
Culture-independent techniques were used for the detection of pathogenic bacteria in drinking water at potentially critical control points along the production lines at a German dairy company and a Spanish dry cured ham company. Denaturing gradient gel electrophoresis (DGGE) was used to describe bacterial population shifts indicating biological instability in the drinking water samples. Autochthonous bacteria were identied by sequencing the excised DGGE DNA bands. More specically, real-time PCR was applied to detect a number of pathogenic bacteria, i.e. Listeria monocytogenes, Mycobacterium avium subsp. paratuberculosis, Campylobacter jejuni, Enterococcus spp., Salmonella spp, Escherichia coli, and Pseudomonas aeruginosa. Due to the detection limits of the real-time PCR method, a specic protocol was established in order to meet the technical detection requirements and to avoid unwanted polymerase inhibitions. Autochthonous bacterial populations were found to be highly stable at most of the sampling points. Only one sampling point exhibited population shifts at the German dairy company. Enterococci and P. aeruginosa were detected in some water samples from these companies by molecular biology detection methods, but not by conventional culturing methods. Some opportunistic bacteria as Enterobacter sp., Acinetobacter, Sphingomonas sp. and non-pathogenic Bacillus, were also detected after DNA sequencing of DGGE bands. 2010 Elsevier B.V. All rights reserved.

1. Introduction Drinking water coming from public suppliers is not sterile, but contains a number of autochthonous and mostly harmless bacteria (Szewzyk et al., 2000; WHO, 2004a). Drinking water distribution systems are an enormous heterogeneous reactor in which the different zones behave almost independently, especially regarding the density and diversity of bacterial populations (Leclerc, 2003). Pathogenic or opportunistic bacteria may enter drinking water facilities under irregular operating conditions. In this case, some of these bacteria are able to persist and distribute across the production lines at food industries (Allen et al., 2004; USEPA, 1992). Besides pathogens, also opportunistic bacteria may cause diseases in immunocompromised people. In this case, the dose response is an important issue. It will vary depending on the pathogen and the host as well as on many other factors (Szewzyk et al., 2000; Leclerc et al., 2002). Pathogenic bacteria may lose their viability and pathogenicity after leaving their natural host or enter a physiological state called viable but not culturable (VBNC) (Oliver, 2000, 2005). VBNC bacteria enter this state in response to one or more natural stresses, i.e. nutrient deprivation, shift in the optimal growth temperature, oxygen concentration, elevated osmotic concen-

Corresponding author. Tel.: + 49 7247 825148; fax: + 49 7247 826858. E-mail address: jessica.varela-villa@ifg.fzk.de (J. Varela Villarreal). 0168-1605/$ see front matter 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.ijfoodmicro.2010.03.001

trations, and exposure to white light. They may be reanimated when the stress factor that induced this state is removed (Oliver, 2005). Most of the pathogenic bacteria are not expected to stay infectious in water over a long term and some will disappear with time, since they are unable to multiply under these conditions. Some species, such as Pseudomonas, Aeromonas, and Mycobacterium avium, however, may even multiply in drinking water (Legnani et al., 1999; Grobe et al., 2001; Leclerc et al., 2002). It is important to note that some waterborne bacteria are capable of multiplying rapidly when contained in foodstuff. This can enormously increase their inoculum's potential and make even initially low and noninfectious doses of bacterial pathogens a hazard in food production. If pathogens nd their optimal growth conditions (e.g., nutrient, humidity, and temperature), a proliferation in food and subsequent transfer to humans becomes a threat. Drinking water is used in food industry for many purposes. It can be in direct contact with foodstuff or in indirect contact with the food product during cleaning and rinsing processes (Casani and Knochel, 2002). According to the Drinking Water Directive 98/83/EC (EU, 1998), water for human consumption should full the highest drinking water standards imposed by the local authorities. Consequently, public drinking water is controlled by the local municipal suppliers, but surveillance by the suppliers ceases when the water enters food production facilities. Various scenarios may inuence the microbial drinking water quality, e.g. rupture of pipelines, water stagnation, pipeline material, etc. Culture-independent methods were applied to identify potential water-derived critical points

S148

J. Varela Villarreal et al. / International Journal of Food Microbiology 141 (2010) S147S155

according to the HACCP concept (EU, 2005; FAO/WHO, 2006). Water samples taken at different points of the production lines of a German dairy company and a Spanish dry cured ham company were subjected to an integral evaluation. The denaturing gradient gel electrophoresis (DGGE) technique has been applied in this work to compare the autochthonous bacterial population of drinking water at different points at food industries in order to analyse the stability of the water within the industry. Real-time PCR, and DNA sequencing were used as additional tools for the identication of Listeria monocytogenes, Mycobacterium avium subsp. paratuberculosis, Campylobacter jejuni, Enterococcus spp, Salmonella spp, Escherichia coli, and Pseudomonas aeruginosa in the water samples. Cultivation experiments complemented the DNA-based experimental techniques. 2. Materials and methods 2.1. Cultivation methods and extraction of genomic DNA Genomic DNA was extracted in order to carry out standard curves and to determine the detection limits of the real-time PCR assays. DNA of L. monocytogenes ATCC 19112 (American Type Culture Collection, Rockville, MD.USA) was provided by the Max Rubner Institute in Karlsruhe, Germany. M. avium subsp. paratuberculosis DSM 44133 (German Collection of Microorganisms and Cell Cultures, Braunschweig, Germany) was grown in two different media: Middlebrook 7H10 agar (DifcoTM, BD, Le Pont de Claix, France) with Middlebrook OADC growth supplement (BBLTM, BD, Maryland, USA) and Mycobactine J (Synbiotics Europe, Lyon, France), and Harrold's egg yolk agar slants with Mycobactine J and ANV (BD, Le Pont de Claix, France) at 37 C for 1 month. C. jejuni DSM 4688 was plated on Campylosel agar (bioMrieux, Nrtingen, Germany) and Columbia agar (bioMrieux) and incubated at 37 C for 48 h. Enterococcus faecium DSM 20477 and Enterococcus faecalis DSM 2981 were plated on Chromocult Enterococci agar (Merck, Darmstadt, Germany) and Slanletz-Bartley agar (Oxoid, Hampshire, England) and incubated at 37 C for 48 h. E. coli DSM 1103 and P. aeruginosa DSM 1117 were grown in tripticase soya broth and nutrient broth at 37 C for 24 h. Salmonella enterica subsp. enterica DSMZ 9274 was grown in selective agar Salmonella (Merck) at 37 C for 48 h. Single colonies of each strain were transferred to rich nutrient media, i.e. tripticase soya broth, TGA-medium or brain heart infusion. Cells were harvested by centrifugation at 5000 rpm for 5 min and supernatant decant off. Reference strains were stored in 25% glycerine at 80 C until use. Total genomic DNA was puried from each bacterium starting with a colony or a cell suspension of the isolate. DNA was puried using PrepMan Ultra Sample preparation (Applied Biosystems, Darmstadt, Germany) in accordance with the manufacturer's guidelines. Concentration of each puried DNA template was determined by spectrophotometry (NanoDrop 1000, peqlab, Erlangen, Germany). Genomic DNA aliquots were stored at 20 C until use. The number of viable culturable bacteria in the water samples was quantied by plating methods. 100 ml water sample was ltered, as indicated by most of the drinking water guidelines, placed on each specic agar, and subjected to the required cultivation conditions. Enterococci, C. jejuni, and Salmonella sp. were cultured using the same agar media as described above. E. coli were grown in two different media, Mac Conkey agar (Merck) and Lactose TTC agar (Merck), at 37 C for 48 h. P. aeruginosa were grown on Cetrimide agar (Merck, Darmstadt, Germany) at 37 C for 48 h. 2.2. Water samples and extraction of total DNA Five and six sampling points were selected at the German and Spanish food companies, respectively. The rst sampling point at both companies was the entry of conditioned public drinking water at the plants. Downstream sampling points differed from one company to

another depending on the production processes with drinking water (Table 1). Planktonic bacteria from water samples were concentrated by ltration using 0.2 m mixed cellulose ester membrane lters (Whatman, Dassel, Germany). A volume of 2000 ml water from each sampling point at the German company was reduced to 1 ml by ltration. In this way, the bacterial concentration in the water samples was increased by a factor of 2000 for the rst sampling period. For the second sampling period at the German company, the bacterial concentration of each water sample was increased by a factor of 10000. The water samples were taken at the same sampling points as in the rst sampling period. At the Spanish company, 5000 ml water was ltered and then resuspended in 1.5 ml water, increasing the bacterial concentration of every water sample by a factor of 3700. These samples were transported frozen to Germany. The bacteria on the lter were resuspended by thorough vortexing in sterile water, and the lter was thrown away. Due to the low number of bacteria expected in drinking water samples, cells in the suspension were disrupted by the commonly used freezingthaw method (Muldrew and McGann, 1994) and kept at 20 C until use. 2.3. Detection of PCR inhibitors and prevention of PCR inhibition Due to the ltration of the water samples and to the different origins of the water (groundwater and surface water), the presence of PCR inhibitors was examined by a PCR efciency assay. This step was required to avoid false negative results. Eubacterial ribosomal primer systems targeting 16S rDNA were applied to perform the PCR. Forward primers were modied by adding a GC clamp at the 5 end for subsequent DGGE analysis. The primer GC27F 5 -CAGAGTTTGATCCTGGCTCAG-3 with 517R 5-ATTACCGCGGCTGCTGG-3 (Muyzer et al., 1993; Emtiazi et al., 2004) and the primer GC341F 5CTACGGGAGGCAGCAG-3 with 907R 5-CCGTCAATTCTTTGAGTTT-3 (Green and Minz, 2005) were used to obtain 490 bp and 566 bp PCR products, respectively. 25 l PCR nal reaction mixture contained 2.5 U HotStar Taq-DNA polymerase (Qiagen, Hilden, Germany), 10 pmol of each primer, 10 PCR buffer, 200 mmol/l dNTPs, and 10 l template. A GeneAmp PCR System 9700 (Applied Biosystems) was used for the amplication. To control PCR efciency, 9 l of each template was spiked with 1 l enterococcal DNA (10 ng/l). In parallel, the standard DNA was used exclusively. The temperature prole consisted in a treatment of 15 min at 95 C, followed by 35 cycles of 0:30 min at 94 C, 0:30 min at 54 C and 1:30 min at 72 C, and a nal step of 7 min at 72 C. Aliquots of 10 l PCR products were subjected to electrophoresis on 1% agarose gel to verify their sizes and estimated amounts. If PCR inhibitors were present, 0.5 l sterile bovine serum albumin (BSA) (Sigma, Munich, Germany) solution (5 mg/ml) was added to the PCR reaction mix according to Kreader (1996). In case of stronger inhibitions, a polyvinylpolypyrrolidone (PVPP) (Sigma) treatment of the samples was performed according to Sutlovic et al. (2007). 2.4. Denaturing gradient gel electrophoresis and sequencing The above described eubacterial ribosomal primer systems targeting 16S rDNA were subsequently used for the DGGE analyses.
Table 1 Sampling points at food companies. German dairy company 1. Entry of public conditioned drinking water 2. Lactic acid tank 3. Portioner 4. Hand washbasin 5. Maturation room 6. Feta packaging Spanish dry cured ham company 1. Entry of public conditioned drinking water 2. Hygienic sluice 3. Salt wash-off 4. Hand washbasin of deboning room 5. Hand washbasin of packaging room

J. Varela Villarreal et al. / International Journal of Food Microbiology 141 (2010) S147S155

S149

10 l of the ltered sample was used as template for the PCR. DGGE analysis of PCR products was performed by means of the D-CodeSystem (BioRad Laboratories GmbH, Munich, Germany) using polyacrylamide gels containing a 4070% denaturing gradient of formamide-urea. DGGE gels were run in 1 TAE buffer (40 mmol/l Tris, 20 mmol/l acetate, 1 mmol/l EDTA) at 70 V and 60 C for 16 h. The gels were stained with SYBR Gold (Invitrogen, Karlsruhe, Germany). The stained gels were immediately analysed using the Lumi-Imager Working Station (Roche Diagnostics, Mannheim, Germany). DGGE ngerprints were scored manually by the presence or absence of DNA bands. Pattern similarities were calculated using the Dice coefcient Cs = 2j(a + b) 1, where j is the number of bands common to samples A and B, and a and b are the total numbers of bands in samples A and B, respectively. This index ranges from 0 (no common bands) to 1 (100% similarity of band patterns) (Murray et al., 1996). In all experiments the main entrance point of public drinking water at the food company facilities was used as reference for population shifts within the downstream drinking water facilities. Intensively stained bands were excised from DGGE and the gel slices were equilibrated in 15 l sterile water over night at room temperature. The DNA extract was re-amplied by PCR and subjected to a DGGE again to verify the purity of the PCR re-amplication product. PCR products were puried with the ExoSap kit (usb, Staufen, Germany), the sequence reaction was done with the BigDye Terminator v1.1 Cycle Sequencing Kit (Applied Biosystems), and the sequence detection was accomplished using the ABI Prism 310 genetic analyser (Applied Biosystems) according to the manufacturer's protocol. Bacteria identication was achieved by comparing the nucleic acid sequences with GenBank sequences using the BLAST program (http://www.ncbi.nlm.nih.gov). 2.5. Real-time PCR for specic pathogen detection TaqMan primers and FAM/TAMRA probes were provided by Sigma Aldrich Chemie (Taufkirchen, Germany) and Biomers.net (Ulm, Germany). Sequences are listed in Table 2. Quantitative real-time PCR was accomplished by amplifying aliquots of 10 l template in 25 l reaction volumes containing 300 nM of each primer, 200 nM FAM/TAMRA-labelled probe, and 12.5 l TaqMan Universal Master Mix (Applied Biosystems). Duplicates or triplicates of each sample were run. Sterile water was used as No Template Control (NTC). The

temperature prole was standardised for all detection systems and comprised 2 min at 50 C, 10 min at 95 C, 45 cycles of 15 s at 95 C and 1 min at 60 C. Results were analysed with the ABI Prism 7000 SDS software 1.1 (Applied Biosystems). To determine the sensitivity of the different specic detection systems, serial dilutions of the DNA from the reference strains were applied. Average Ct values were calculated from triplicates or duplicates. The amounts of bacteria used for measuring standard parameters were calculated from their genome sizes (Suess et al., 2006). To calculate the nal amounts of bacteria in the samples, the initial volume of each sample was considered. This calculation is based on the assumption of the average weight of a base pair (bp) being 650 Da. This means that 1 mol of a bp weighs 650 g and that the molecular weight of any doublestranded DNA template can be estimated by multiplying its length (in bp) by 650. The inverse of the molecular weight is the number of moles of template present in 1 g of material. Using the Avogadro number 6.002 1023 molecules/mol, the number of molecules of the template per gram can be calculated as: mol=g molecules=mol = molecules=g: Finally, the number of bacteria or number of copies of template in the sample can be estimated by multiplying by 1 109 for conversion to ng and then multiplying by the amount of template (in ng). The genome sizes are listed in Fogel et al. (1999). 3. Results 3.1. Protocol developed for drinking water surveillance The strategy developed for the molecular biology detection of pathogens in drinking water and subsequent identication of potentially critical control points at the food companies is shown in Fig. 1. Selection of the sampling points together with the person responsible for quality control at the food company was of great importance. Water samples have to be taken strategically at those points, where the water might endanger food hygiene. Due to the low amounts of bacteria expected to be present in drinking water, the samples had to be subjected to ltration. During ltration, PCR inhibitors might be enriched and subsequently interfere with the PCR methods. When no DNA amplication was observed after the eubacterial 16S rDNA PCR

Table 2 TaqMan primers and FAM/TAMRA probes used for real-time PCR assays. Primers and probes hlyQF hlyQR hlyQP Ecst784F Enc854R Gpl813TQ Pa23F Pa23Rb Pa23P VS1F VS1R VS1P mycF2 mycR2 mycP InvA139 F InvA141 R InvAP ECOuidAF ECOuidAR ECOuidAP
a

Sequences (53) CATGGCACCACCAGCATCT ATCCGCGTGTTTCTTTTCGA FAM-CGCCTGCAAGTCCTAAGACGCCA-TAMRA AGAAATTCCAAACGAACTTG CAGTGCTCTACCTCCATCATT FAM- TGGTTCTCTCCGAAATAGCTTTAGGGCTA-TAMRA TCCAAGTTTAAGGTGGTAGGCTG ACCACTTCGTCATCTAAAAGACGAC FAM-AGGTAAATCCGGGGTTTCAAGGCC-TAMRA ATTAGGTCTTAATACTAAAGATCAGCAAGGT CGTCCTTTGTCTTATGGTTTGAATT FAM-TGGCGTATTTGATGAATGTTT-TAMRA AATGACGGTTACGGAGGTGGT GCAGTAATGGTCGGCCTTACC FAM-TCCACGCCCGCCCAGACAGGTTG-TAMRA GTGAAATAATCGCCACGTCGGGCAA TCATCGCACCGTCAAAGGAACC FAM-TTATTGGCGATAGCCTGGCGGTGGGTTTTGT TG-TAMRA GTGTGATATCTACCCGCTTCGC AGAACGGTTTGTGGTTAATCAGGA FAM-TCGGCATCCGGTCAGTGGCAGT-TAMRA

Microorganism Listeria monocytogenes

Gene hly

Gene function Hemolysin

Product size (bp) 64

Literature source Rodrguez-Lzaro et al. (2004) Frahm et al. (1998)

Enterococcus sp.

23S rDNA

92

Pseudomonas aeruginosa

23S rDNA

93

Volkmann et al. (2007)

Campylobacter jejuni

VS

Variable sequence

115

This worka

Mycobacterium avium subsp paratuberculosis Salmonella spp.

IS900

invA

Insertion sequenceIS900-like transposase Membrane spanning protein Glucuronidase

76

Cook and Britt (2007)

284

Malorny et al. (2001) and Hein et al. (2006) Frahm and Obst (2003)

Escherichia coli

uidA

87

Designed by Dr. H. Volkmann.

S150

J. Varela Villarreal et al. / International Journal of Food Microbiology 141 (2010) S147S155

(Fig. 2, panel A), a PCR efciency assay was carried out. A known quantity of enterococcal genomic DNA was added to the samples before this PCR. An inhibition was indicated clearly by the absence of an amplication product (Fig. 2, panel B). PCR inhibitors were not removed from the samples with BSA only (results not shown). When the samples were treated with PVPP, weak PCR products were observed (Fig. 2, panel C). To conrm that the intensity of these bands corresponded to a low DNA concentration in the samples and not to the presence of PCR inhibitors, a PCR efciency assay was performed again. The bands observed after this PCR efciency assay (Fig. 2, panel D) exhibited the same or even higher intensities than the added genomic DNA (Fig. 2, lane P), indicating that no PCR inhibitors were present in the water samples after the PVPP treatment. 3.2. Detection of pathogens using real-time PCR assays Real-time PCR assays were developed or optimised to detect bacteria in drinking water, which are hygienically relevant to food industry. Real-time PCR primers and probes that target specic virulence or taxon-specic genes are listed in Table 2. Genomic DNA dilutions were used instead of bacterial suspensions for sensitivity assays due to the retarded growth of some bacterial species, such as M. avium subsp. paratuberculosis. The sensitivities of the real-time PCR assays shown in Table 3 were obtained when the standard curves were done, after amplifying genomic DNA serial dilutions of each target bacteria. Average Ct values were calculated from triple reactions. Considering that the DNA of the samples would be detected

Fig. 2. PCR efciency assay. Lanes 15 correspond to the Spanish dry cured ham company's water sampling points. 10 l of the respective 16S rDNA amplicons was separated in 1% agarose gel (amplicon size: 566 bp). Panel A, original water templates; panel B, original water templates spiked with enterococcal genomic DNA; panel C, original water templates after PVPP treatment; panel D, original water templates spiked with enterococcal genomic DNA after PVPP treatment. N: negative template control, P: positive control, and M: 100 bp DNA marker.

by real-time PCR in a volume of 10 l template and that the bacteria present in this template would be concentrated 10000 times by ltration of the original water sample, the detection limits calculated

Fig. 1. Final protocol used for molecular biology detection of pathogens in drinking water.

J. Varela Villarreal et al. / International Journal of Food Microbiology 141 (2010) S147S155 Table 3 Detection limits of real-time PCR systems. Bacteria Enterococcus faecium Salmonella enterica Campylobacter jejuni M. avium subsp paratuberculosis Listeria monocytogenes Pseudomonas aeruginosa Escherichia coli
a

S151

Target gene 23S rDNA invA VS1 IS900 hly 23S rDNA uidA

Genome size (kb) 555 4746 765 5838 3150 1637 4639

Detection limit (cell/100 ml)a 1 1 4 1090 3 1 2

Detection limits were calculated according to the nal protocol.

for E. faecium, S. enterica, and P. aeruginosa were similar to those of the standard plating methods (1 cell/100 ml). The real-time PCR detection limits calculated for C. jejuni, L. monocytogenes, and E. coli were 2 to 4 cells/100 ml. In the case of M. avium subsp. paratuberculosis, the real-time PCR detection limit was 1090 cell/100 ml. The equations of the standard curve of each pathogen given in Fig. 3 were estimated by linear regression. These equations were used to determine the bacterial concentration present in the water samples from their genome sizes, as described previously. High PCR efciencies were obtained for these assays (Fig. 3) The correlation coefcients (between 0.9958 and 0.9995) showed a high precision of the assays and a strong correlation between template DNA concentrations and Ct values. These parameters indicated that they are useful for quantitative measurements. In the case of M. avium subsp. paratuberculosis, the standard curve reected a high correlation coefcient, but the calculated detection limit minimised the application of this assay. False-positive results were obtained by real-time PCR using the uidA gene targeting E. coli. The commonly used polymerase appeared to be a contamination source of E. coli DNA, because this enzyme was expressed as a recombinant protein in E. coli (Shannon et al., 2007). In order to avoid this, the real-time PCR used for the detection of E. coli was done with the TaqMan Gene Expression Master Mix (Applied Biosystems). This kit uses the AmpliTaq Gold DNA Polymerase Ultra Pure enzyme that is identical to the AmpliTaq Gold DNA Polymerase, but further puried to reduce bacterial DNA introduced from the host organism. The purication process ensures that non-specic, falsepositive DNA products due to bacterial DNA contamination are minimised during PCR (protocol AmpliTaq Gold DNA Polymerase Ultra Pure enzyme, Applied Biosystems). 3.3. German dairy company The German dairy company was supplied with conditioned groundwater exclusively and no further disinfection was performed onsite. The pipeline system was made of stainless steel, hoses were used at sampling points 3 (portioner) and 2 (lactic acid tank), and warm water was used at points 2 (lactic acid tank) and 4 (hand washbasin). The drinking water at the entrance point met all requirements of the German drinking water regulations. None of the indicated pathogenic bacteria was detected after ltering 100 ml of each water sample and carrying out the plating methods on the specic selective media. In some cases, unspecic bacterial growth was observed on agar plates, but these colonies were identied as false-positive isolates after sequencing of 16S ribosomal DNA. No PCR inhibition was detected after performing the PCR efciency assay, as already described above. Real-time PCR results of the rst sampling period are shown in Table 4. The sample from point 2 (lactic acid tank), where hoses were involved in the process, was the only sample that exhibited positive results for P. aeruginosa and enterococci after real-time PCR analysis. An average Ct value of 33.21 was found for P. aeruginosa. By transpolating this value to the standard curve, a value of 2.45 fg P. aeruginosa DNA per l was obtained. Knowing that

one P. aeruginosa bacterial cell DNA weighs 3.99 fg, that 10 l template was used for the real-time PCR, and that the bacteria present in the sample were concentrated by a factor of 2000 by ltration, the calculated number of P. aeruginosa for this sample was 31 cells/100 ml water sample. Enterococci-specic positive signals at this point were also detected, but the values were lower as the calculated detection limits and reached an average Ct value of 37.91. None of the other water samples taken at this company exhibited positive real-time PCR results for any of the specic targeted pathogens (Table 4). Some hygienic recommendations, such as a more frequent exchange of hoses, were made before the second sampling period. During this period, higher volumes were ltered in order to achieve detection limits similar to those of the standard plating techniques. Monitoring of pathogens during the second sampling period did not produce any positive results, no matter whether traditional plating methods or culture-independent methods were applied. Analysis of the autochthonous bacterial population of water samples during the rst sampling period (Fig. 4) revealed a total number of 9 DGGE DNA bands in the reference sample (Fig. 4, lane 1). Each DNA band was assumed to represent one bacteria species. In the subsequent samples the number of bands did not differ or increased only slightly by 1 to 3 bands when compared to the reference sample. Using the above Dice coefcients, only sampling point 6 (feta packaging) was found to exhibit a decreased similarity value of 30%. All the other points presented high bacterial population similarities ranging from 44 to 60%. Consequently, point 6 was considered a potentially critical point. A total number of 13 bands were sliced from the DGGE gel for sequencing. Most of these bacteria were - or -Proteobacteria. None of the targeted pathogens were identied by sequencing, but some opportunistic bacteria as Sphingomonas and Acinetobacter were aligned (Table 5). Although one potentially critical point was identied after analysing the autochthonous bacterial population, no technical problems or irregular operation during food production were encountered during the evaluation. When the bacterial populations of the water samples during the second sampling period were analysed, the similarity values between the different sampling points and the reference point were between 53 and 86%. No sampling point presented a similarity value below 40%. 3.4. Spanish dry cured ham company The water supplied by the Spanish public distribution network was chlorine-treated conditioned groundwater having a residual chlorine content of 0.4 mg/l. No additional treatment was done at the company. It was not possible to apply traditional plating methods due to the lack of equipment at the sampling place. The water samples were ltered in Spain and then transported to Germany, where culture-independent methods exclusively were applied for their analysis. Initially, no DNA amplication was observed but, as explained above, this changed when the samples were treated with PVPP (Fig. 2). Real-time PCR results are shown in Table 4. Some weak positive signals became obvious after P. aeruginosa-specic real-time PCR analysis. At points 2 (salt wash-off), 3 (hand washbasin of bone removal room), and 5 (hand washbasin of packaging room), average Ct values of 38.27, 37.21, and 39.05 were obtained. All these values were close to the detection limit of the real-time PCR detection system. None of the other water samples of this company showed positive real-time PCR results for any of the specic targeted pathogens (Table 4). A total number of 7 DGGE DNA bands were observed in the reference sample (point 1) when the autochthonous bacterial population of the water samples was analysed. The downstream water samples exhibited 5 to 9 bands. When comparing the bacterial populations of the water sample and the incoming water using the already described Dice coefcient, no signicant difference was found.

S152

J. Varela Villarreal et al. / International Journal of Food Microbiology 141 (2010) S147S155

Fig. 3. Real-time PCR standard analysis curves. Serial dilutions of reference strain genomic DNA were used as template. Cycle threshold values (Ct) are plotted against log10 copies of bacterial DNA. Linear regression, PCR efciency (E) and regression coefcients (R2) for each bacterial detection system are shown. (A) Campylobacter jejuni, (B) Escherichia coli, (C) Enterococcus faecium, (D) Listeria monocytogenes, (E) Mycobacterium avium subsp. paratuberculosis, (F) Pseudomonas aeruginosa, and (G) Salmonella enterica subsp. enterica. In parallel, sterile water was used for NTCs.

The similarities of the samples with the reference sample were quite high. They ranged between 63% and 77%, indicating a biological stability of the analysed water samples. 11 DNA bands were sliced from the DGGE gel for sequencing. Most of the sequenced DNA fragments belonged to the -Proteobacteria subclass and were mostly aligned with uncultured bacteria. Non-pathogenic Bacillus sp. and some opportunistic bacteria, as Sphingomonas sp., Enterobacter sp., and, Pseudomonas sp. were also identied after sequencing the DNA of DGGE bands. Hence, the presence of Pseudomonas found by the previous real-time PCR was conrmed.

Although some positive pathogenic bacteria results were seen after the use of culture-independent methods, it was not possible to distinguish DNA from live or dead cells. 4. Discussion Molecular biology techniques have been used for several years for the examination of water for multiple purposes (Frahm et al., 1998; Frahm and Obst, 2003; Grobe et al., 2001; Schwartz et al., 1998, 2003). The work reported here was focused on the testing and optimisation

J. Varela Villarreal et al. / International Journal of Food Microbiology 141 (2010) S147S155

S153

Table 4 Conventional plating and real-time PCR results of water samples of the German dairy company (rst sampling period) and the Spanish dry cured ham company. Duplicates or triplicates of each sample were run. Sampling point German dairy company 1 Plating methods Real-time PCR Enterococcus spp. Salmonella spp. Campylobacter jejuni M. avium subsp. paratuberculosis Listeria monocytogenes Pseudomonas aeruginosa Escherichia coli
a

Spanish dry cured ham company 3 4 5 6 1 2 3 4 5

Negative for all pathogens +a +a

Not determined +a +a +a

Positive results are described in more detail in the text.

of culture-independent techniques to analyse the bacterial drinking water quality at two food companies. According to the Drinking Water Directive 98/83/EC (EU, 1998) of the European Union, indicator microorganisms should be routinely monitored in drinking water in order to control microbial water quality of public distribution systems. The German Drinking Water Ordinance (TrinkwV 2001) and the Spanish Drinking Water Guidelines (Real Decreto 140/2003, 2003)

based on the above EU directive stipulate that no E. coli, enterococci, and coliform bacteria should be present in 100 ml drinking water of public distribution systems. The standard detection method described in these guidelines is the conventional plating on dened media. It is commonly accepted that culture-dependent methods do not reect the different physiological states of bacteria that inuence their culturability (Oliver, 2000). Consequently, culture-independent methods were applied as an alternative approach to monitor the most important food-borne pathogens in drinking water. DNA ngerprinting was used to characterise the autochthonous bacterial population of drinking water at the food companies. Nowadays, the use of molecular biology methods in routine drinking water surveillance is still limited, as these new methods have not yet been accepted by the authorities. According to the EU guidelines (EU, 1998), such methods can be used for the monitoring of indicator bacteria only when it can be demonstrated that the results obtained are at least as reliable as those produced by the specied methods. Hence, the detection limits of the assays play a critical role for bacterial quantication in drinking water samples. The detection limits of the real-time PCR systems were not always optimal to reach the parameters established by the water authorities, especially those obtained for the detection of M. avium subsp. paratuberculosis. In order to reach detection limits of at least 1 bacterium per 100 ml without an additional enrichment step, a protocol with higher sample ltration volumes was developed. In the case of M. avium subsp. paratuberculosis, even higher bacterial concentration rates should be achieved. No pathogenic bacteria were cultivated from the water samples using standard plating methods. However, some positive results were obtained by using culture-independent techniques. This could be due to the higher sensitivity of PCR that leads to a greater number of

Table 5 Identication of bacteria in water samples from the German dairy company (rst sampling period) after sequencing the DNA bands excised from the DGGE gel shown in Fig. 4. Numbers correspond to the respective DNA bands. Bacterium 1. Rhodoferax sp. 2. Acidovorax 3. Uncultured bacteria 4. Uncultured bacteria 5. Caulobacter crescentis 6. Aquabacterium 7. Aquabacterium 8. Sphingomonas 9. Acinetobacter 10. Aquabacterium 11. Meiothermus 12. Sphingomonas 13. Sphingomonas Proteobacteria subclass Max. ident. (%) 100 99 99 98 98 98 99 95 98 88 94 99 99 Accession number AY788965.1 DQ153906.1 DQ409991.1 DQ664220.1 AE005673.1 EF651436.1 EF651436.1 AY026948.1 EF570077.2 EF179861.1 AY845055.1 AY026948.1 AY026948.1

Fig. 4. DGGE DNA ngerprints of 16S rDNA amplicons from the German dairy company's water samples (rst sampling period). Lanes 1 to 6 correspond to the sampling points named in Table 1, the numbers on the gel correspond to the sequenced DNA bands (see Table 5), and the numbers at the bottom are the total DNA bands of the lane.

S154

J. Varela Villarreal et al. / International Journal of Food Microbiology 141 (2010) S147S155

positive results in comparison to conventional plating methods, which was also described by Sachse and Frey (2003). It is also wellknown that culture-independent techniques based on the analysis of the DNA present in the samples cannot distinguish among viable, viable but not culturable (VBNC), injured, and dead cells. VBNC or injured bacteria fail to grow on the routine bacteriological media, but are alive and metabolically active (Oliver, 2000). False negative results might be obtained when traditional plating methods are used. About 60 bacterial species have been already described to enter the VBNC state. Among these are some relevant food-borne pathogens, e.g. enterococci, C. jejuni, Salmonella spp., Helicobacter pylori, Klebsiella spp., L. monocytogenes, and E. coli (including EHEC) (Oliver, 2005). Therefore, the detection of bacteria, including VBNC bacteria, in food industry's drinking water is essential to ensure the microbiological safety of food. Although positive DNA-based results do not reect the presence of exclusively live bacteria, they give hints of possible irregular operations that might support the transfer of pathogen targets. New techniques based on the use of propidium monoazide (PMA) (Nocker et al., 2007; Rieder et al., 2008), ethidium monoazide (Delgado-Viscogliosi et al., 2009) or DNase I (Nogva et al., 2000) are available and need to be optimised to distinguish the different metabolic states of such bacteria in drinking water. Another critical point that should be considered when using molecular biology techniques is the possible presence of PCR inhibitors. Organic substances like humic acids and other PCR inhibitors are often present in surface waters (Wilson, 1997). Such substances were found in the water samples taken at the Spanish dry cured ham company. The use of PVPP as mentioned by Sutlovic et al. (2007) and Gusbeth et al. (2009) successfully removed the PCR inhibitors in this work. Characterisation of the bacterial populations of water samples is an innovative approach to demonstrate the biological stability of water in an industrial process. Previous studies revealed that Dice coefcients between 0.40 and 1 (i.e. between 40 and 100% similarity) reected a natural range of population diversity in a drinking water distribution system (Emtiazi et al., 2004). Hence, similarities below 40% are discussed to indicate a population shift in the autochthonous bacterial population of drinking water systems. Only one point (feta packaging) during the German dairy company's rst sampling period had a lower similarity when compared to the reference point, indicating that something was affecting the natural microbiological population of water. The similarity values between the different sampling points and the reference point observed during the second sampling period after implementing the hygienic recommendations were high, which demonstrated that the PCR-DGGE method was adequate for the evaluation of drinking water bacterial stability in food industry. The quality of the supplied drinking water is of signicant importance for a good hygienic practice in downstream process lines. Therefore, information from raw water quality is needed in concern of potential contaminations with hygienically relevant bacteria (WHO, 2004b). Groundwater and surface water are frequently conditioned in Germany and many other countries. Usually, groundwater is supposed to have a better biological quality than surface water, but some waterborne diseases have also been transmitted by contaminated groundwater (Craun, 1985; Scandura and Sobsey, 1997; Ritter et al., 2002). Data about the drinking water conditioning at the waterworks is essential for the estimation of the biological stability of the drinking water during its distribution. Disinfection measures are mostly important to inactivate microorganisms. Depending on the drinking water character, sustainability of the disinfection measure is impaired. Chemical (chlorine, chlorine dioxide, and ozone) disinfection and UV irradiation are the most frequently used disinfection techniques at European waterworks. It has been demonstrated that these treatments have various disinfection efciencies (WHO, 2004b). Some hygienically relevant bacteria, such as Pseudomonas spp., are well-known to have a

high capability to survive in chlorinated water and to form biolms (Grobe et al., 2001; Leclerc et al., 2002). It was demonstrated recently that a specic DNA dark repair mechanism of P. aeruginosa was induced at UV exposures of 400 J/m2, which corresponds to the German standard for UV disinfection (Jungfer et al., 2007). Drinking water suppliers control drinking water production and its municipal distribution systems (WHO, 2004a), but not the drinking water distribution in food industry. However, it is important to control drinking water facilities in food industry to avoid irregular operations (inadequate pipeline or connection materials, water stagnation, softening, pipe corrosion, etc.) that might inuence bacterial growth or re-growth (WHO, 2004b, 2006, 2008). Furthermore, irregular operations may result in an increased biolm formation. Biolms are potential habitats of all kinds of bacteria, including pathogens (Emtiazi et al., 2004; Lehtola et al., 2004; Schwartz et al., 1998, 2003) and may be responsible for contaminations of bulk water systems. Old pipes in combination with increased water hardness values may result in pipe incrustations that are also known to support undesired biolm formation (WHO, 2004b, 2006). This might be the reason for the presence of P. aeruginosa at the Spanish company, where the pipelines were 20 years old. The use of accessory facilities like hoses for cleaning processes could be responsible for cross-contaminations during food production. Such hoses should be exchanged regularly, especially when warm water is used, since warm water systems support the growth of hygienically relevant bacteria, such as E. coli, P. aeruginosa, Aeromonas sp, Legionella spp. (Legnani et al., 1999; Leclerc et al., 2002). Our extended investigations of the two food companies demonstrated that they met the drinking water standards. The cultureindependent techniques used cannot distinguish among viable, viable but not culturable, injured, and dead cells. Still, such techniques can be used to identify critical control points in all stages of food production where water is involved.

Acknowledgments The authors are grateful for the assistance of Johannes Knoll and for the helpful discussions of Christina Jungfer and Jacqueline S. This study was supported by the European Commission within the EU 6th Framework Program, the PathogenCombat Project, and by Forschungszentrum Karlsruhe GmbH. L. monocytogenes DNA was kindly provided by the Max Rubner Institute, Karlsruhe. Special thanks to Jordi Rovira and his team (Burgos University, Burgos, Spain) for the assistance in sampling in Spain. Cooperation of the German and Spanish food companies is also gratefully acknowledged.

References
Allen, M.J., Edberg, S.C., Reasoner, D.J., 2004. Heterotrophic plate count bacteria what is their signicance in drinking water? International Journal of Food Microbiology 92, 265274. Casani, S., Knochel, S., 2002. Application of HACCP to water reuse in the food industry. Food Control 13, 315327. Cook, K.L., Britt, J.S., 2007. Optimization of methods for detecting Mycobacterium avium subsp paratuberculosis in environmental samples using quantitative, real-time PCR. Journal of Microbiological Methods 69, 154160. Craun, G., 1985. A summary of waterborne illness transmitted through contaminated groundwater. Journal of Environmental Health 48, 122127. Delgado-Viscogliosi, P., Solignac, L., Dellatre, J.-M., 2009. Viability PCR, a cultureindependent method for rapid and selective quantication of viable Legionella pneumophila cells in environmental water samples. Applied and Environmental Microbiology 75 (11), 35023512. Emtiazi, F., Schwartz, T., Marten, S.-M., Krolla-Sidenstein, P., Obst, U., 2004. Investigation of natural biolms formed during the production of drinking water from surface water embankment ltration. Water Research 38, 11971206. EU, 1998. European Union. Council Directive 98/83/EC of 3 November 1998 relating to the quality of water intended for human consumption. Ofcial Journal of the European Communities L 330, 3254. EU, 2005. European Union. Guidance Document. Implementation of procedures based on the HACCP principles, and facilitation of the implementation of the HACCP principles in certain food businesses. Brussels.

J. Varela Villarreal et al. / International Journal of Food Microbiology 141 (2010) S147S155 FAO/WHO, 2006. Food and Agriculture Organisation/World Health Organisation. Guidance to governments on the application of HACCP in small and/or lessdeveloped food businesses. FAO food and nutrition paper 86. ISSN 0254-4725. Fogel, G.B., Collins, C.R., Li, J., Brunk, C.F., 1999. Prokaryotic genome size and SSU rDNA copy number: estimation of microbial relative abundance from a mixed population. Microbial Ecology 38, 93113. Frahm, E., Obst, U., 2003. Application of the ourogenic probe technique (TaqMan PCR) to the detection of Enterococcus spp. and Escherichia coli in water samples. Journal of Microbiological Methods 52, 123131. Frahm, E., Heiber, I., Hoffman, S., Koob, C., Meier, H., Ludwig, W., Amann, R., Schleifer, K.H., Obst, U., 1998. Application of 23S rDNA-targeted oligonucleotide probes specic for enterococci to water hygiene control. Systematic and Applied Microbiology 21, 450453. Green, S.J., Minz, G., 2005. Suicide polymerase endonuclease restriction, a novel technique for enhancing PCR amplication of minor DNA templates. Applied and Environmental Microbiology 71, 47214727. Grobe, S., Wingender, J., Flemming, H.C., 2001. Capability of mucoid Pseudomonas aeruginosa to survive in chlorinated water. International Journal of Hygiene and Environmental Health 204, 139142. Gusbeth, C., Frey, W., Volkmann, H., Schwartz, T., Bluhm, H., 2009. Pulsed electric eld treatment for bacteria reduction and its impact in hospital wastewater. Chemosphere 75 (2), 228233. Hein, I., Flekna, G., Krassnig, M., Wagner, M., 2006. Real-Time PCR for the detection of Salmonella spp. in food: an alternative approach to a conventional PCR system suggested by the FOOD-PCR project. Journal of Microbiological Methods 66, 538547. Jungfer, C., Schwartz, T., Obst, U., 2007. UV-induced dark repair mechanisms in bacteria associated with drinking water. Water Research 41 (1), 188196. Kreader, C.A., 1996. Relief of amplication inhibition in PCR with bovine serum albumin or T4 Gene 32 Protein. Applied and Environmental Microbiology 62, 11021106. Leclerc, H., 2003. Relationships between common water bacteria and pathogens in drinking water. In: Bartram, J., Cotruvo, J., Exner, M., Fricker, C., Glasmacher, A. (Eds.), Heterotrophic Plate Counts and Drinking Water Safety: The Signicance of HPCs for Water Quality and Human Health. : WHO Emerging Issues in Water and Infectious Disease Series, Chapter 6. IWA Publishing, London. Leclerc, H., Schwartzbrod, L., Dei-Cas, E., 2002. Microbial agents associated with waterborne diseases. Critical Reviews in Microbiology 28 (4), 371409. Legnani, P., Rapuano, S., Turin, D., Valenti, C., 1999. Survival and growth of Pseudomonas aeruginosa in natural mineral water: a 5-year study. International Journal of Food Microbiology 53 (23), 153158. Lehtola, M.J., Nissinen, T.K., Miettinen, I.T., MArtikainen, P.J., Vartiainen, T., 2004. Removal of soft deposits from the distribution system improves the drinking water quality. Water Research 38 (3), 601610. Malorny, B., Bunge, C., Helmuth, R., 2001. Evaluation of Salmonella spp.-specic primer sets for the validation within the Food PCR Project. Federal Institute for Health Protection of Consumers and Veterinary Medicine, National Reference Laboratory for Salmonella, Diedersdofer Weg, 1, p. 12277. Berlin, Germany. Muldrew, K., McGann, L.E., 1994. The osmotic rupture hypothesis of intracellular freezing injury. Biophysical Journal 66, 532541. Murray, A.E., Hollibaugh, J.T., Orrego, C., 1996. Phylogenetic compositions of bacterioplankton from two California estuaries compared by denaturing gradient gel electrophoresis of 16S rDNA fragments. Applied and Environmental Microbiology 62, 26762680. Muyzer, G., De Waal, E., Uiterlinden, A.G., 1993. Proling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplied genes coding for 16S rRNA. Applied and Environmental Microbiology 59 (3), 695700. Nocker, A., Sossa-Fernandez, P., Burr, M.D., Camper, A.K., 2007. Use of propidium monoazide for live/dead distinction in microbial ecology. Applied and Environmental Microbiology 73 (16), 51115117. Nogva, H.K., Bergh, A., Holck, A., Rudi, K., 2000. Application of the 5-nuclease PCR assay in evaluation and development of methods for quantitative detection of Campylobacter jejuni. Applied and Environmental Microbiology 66 (9), 40294036. Oliver, J.D., 2000. The public health signicance of viable but nonculturable bacteria. In: Colwell, R.R., Grimes, D.J. (Eds.), Nonculturable Microorganisms in the Environment. ASM Press, Washington, D.C., pp. 277300.

S155

Oliver, J.D., 2005. Viable but nonculturable bacteria in food environments. In: Fratamico, P.M., Bhunia, A.K., Smith, J.L. (Eds.), Foodborne Pathogens Microbiology and Molecular Biology. Caister Academic Press, Great Britain, p. 99. Real Decreto 140/2003, 2003. Royal Decree of 7 February, establishment of sanitary criteria of water quality for human consumption. BOE number 45, pp. 72287245. Spain. Rieder, A., Schwartz, T., Schn-Hlz, K., Marten, S.-M., S, J., Gusbeth, C., Kohnen, W., Swoboda, W., Obst, U., Frey, W., 2008. Molecular monitoring of inactivation efciencies of bacteria during pulsed electric eld treatment of clinical wastewater. Journal of Applied Microbiology 105 (6), 20352045. Ritter, l., Solomon, K., Sibley, P., Hall, K., Keen, P., Mattu, G., Linton, B., 2002. Sources, pathways, and relative risks of contaminants in surface water and groundwater: a perspective prepared for the Walkerton inquiry. Journal of Toxicology and Environmental Health 65 (1), 1142. Rodrguez-Lzaro, D., Hernndez, M., Scortti, M., Esteve, T., Vsquez-Boland, J.A., Pla, M., 2004. Quantitative detection of Listeria monocytogenes and Listeria innocua by realtime PCR: assessment of hly, iap, and lin02483 targets and AmpliFluor Technology. Applied and Environmental Microbiology 70, 13661377. Sachse, K., Frey, J., 2003. PCR detection of microbial pathogens. In: Sachse, Konrad, Frey, Joachim (Eds.), Methods in Molecular Biology, vol. 216. Humana Press Inc, otowa, NJ, p. 20. Scandura, J.E., Sobsey, M.D., 1997. Viral and bacterial contamination of groundwater from on-site sewage treatment systems. Water Science and Technology 35 (11), 141146. Schwartz, T., Hoffman, S., Obst, U., 1998. Formation and bacterial composition of young, natural biolms obtained from public bank-ltered drinking water systems. Water Research 32 (9), 27872797. Schwartz, T., Hoffmann, S., Obst, U., 2003. Formation of natural biolms during chlorine dioxide and UV disinfection in a public drinking water distribution system. Journal of Applied Microbiology 95, 591601. Shannon, K.E., Lee, D.Y., Trevors, J.T., Beaudette, L.A., 2007. Application of real-time quantitative PCR for the detection of selected bacterial pathogens during municipal wastewater treatment. Science of the Total Environment 382, 121129. Suess, J., Schubert, K., Sass, H., Cypionka, H., Overmann, J., Engelen, B., 2006. Widespread distribution and high abundance of Rhizobium radiobacter within Mediterranean subsurface sediments. Environmental Microbiology 8, 17531763. Sutlovic, D., Gojanovic, M.D., Andelinovic, S., 2007. Rapid extraction of human DNA containing humic acid. Croatica Chemica Acta 80, 117120. Szewzyk, U., Szewzyk, R., Manz, W., Schleifer, K.-H., 2000. Microbiological safety of drinking water. Annual Review of Microbiology 54, 81127. TrinkwV 2001. Trinkwasserverordnung [Drinking Water Ordinance 2001]. Verordnung ber die Qualitt von Wasser fr den menschlichen Gebrauch [Ordinance on water quality for human use]. Germany. http://bundesrecht.juris.de/bundesrecht/ trinkwv_2001/gesamt.pdf. USEPA, 1992. Seminar publication: control of biolm growth in drinking water distribution systems. EPA/625/R-92/001. Ofce of Research and Development, U.S. Environmental Protection Agency, Washington, DC. http://www.p2pays.org/ref/ 15/14291.pdf. Volkmann, H., Schwartz, T., Kirchen, S., Stofer, C., Obst, U., 2007. Evaluation of inhibition and cross-reaction effects on real-time PCR applied to the total DNA of wastewater samples for the quantication of bacterial antibiotic resistance genes and taxonspecic targets. Molecular and Cellular Probes 21, 125133. WHO, 2004a. In: Ainsworth, R. (Ed.), World Health Organization. Safe piped water: managing microbial water quality in piped distribution systems. IWA Publishing, London, UK. ISBN: 1843390396. WHO, 2004b. In: LeChevallier, M.W., Au, K.K. (Eds.), World Health Organization. Water treatment and pathogen control: process efciency in achieving safe drinking water. IWA Publishing, London, UK. ISBN: 1 84339069 8. WHO, 2006. World Health Organization. Health aspects of plumbing. World Health Organization and World Plumbing Council, Geneva, Switzerland. ISBN: 978 92 4 156318 5. WHO, 2008. World Health Organization. Guidelines for Drinking Water Quality: incorporating 1st and 2nd addenda. In: Volume 1 Recommendations, Third edition. . ISBN: 978 92 4 154761 1. Geneva, Switzerland. Wilson, I.G., 1997. Minireview: inhibition and facilitation of nucleic acid amplication. Applied and Environmental Microbiology 63, 37413751.