Access the full text.
Sign up today, get DeepDyve free for 14 days.
Seroprevalence data for pig herds suggested that there must be a relevant reservoir for hepatitis E virus (HEV) in Switzer- land. To know more about the viral presence in ready-to-eat meat products, we screened pork liver sausages and raw meat sausages from the Swiss retail market for the presence of HEV. Testing was performed with a detection method where the virus extraction step was optimized. As for the performance of the improved method, the mean recovery rate for the mengo- virus process control was 24.4%, whereas for HEV-inoculated sample matrices between 10.4 and 100% were achieved. The 3 2 limit of detection was about 1.56 × 10 and 1.56 × 10 genome copies per gram for liver sausages and raw meat sausages, respectively. In the screening programme, HEV-RNA was detected in 10 of total 90 (11.1%) meat products, 7 of 37 (18.9%) liver sausages, and 3 of 53 (5.7%) raw meat sausages. Virus loads of up to 5.54 log HEV genome copies per gram were measured. All sequences retrieved from positive samples belonged to HEV genotype 3. The significance of the presented work was a current overview of the HEV prevalence in ready-to-eat meat products on the Swiss retail marked and an improvement of the extraction efficiency of the HEV detection method. Keywords Hepatitis E virus · Virus extraction · RT-qPCR · Raw meat sausage · Pork liver sausage · Risk-assessment Introduction (Worm et al. 2002) and in Germany, it was shown that this zoonotic reservoir is responsible for autochthonous sporadic In Europe, the most important enteric viruses are hepatitis cases (Wichmann et al. 2008). Another investigation showed A virus (HAV), norovirus, enterovirus, rotavirus, and astro- a rather high HEV-seroprevalence of 16.8% indicating that virus (Le Guyader et al. 2000). Moreover, hepatitis E virus HEV is endemic in Germany (Faber et al. 2012). HEV-borne (HEV) increasingly received attention in recent past. This outbreaks, mainly due to contaminated water, are common in agent occurs worldwide with an estimated number of 20 mil- countries with poor hygienic conditions. Nevertheless out- lion cases and 56,000 fatalities per year. Ae ff cted are mainly breaks in developed countries are reported, but these were developing countries where hepatitis E is a waterborne linked to the consumption of food containing pork liver. In infection and HEV genotypes 1 and 2 cause large outbreaks France for example, a group of wedding participants got (Blasco-Perrin et al. 2016). In developed countries hepati- infected due to the consumption of an undercooked pork tis E occurred relatively rarely and sporadically, but with a liver-based stuffing (Guillois et al. 2016) and in Australia, tendency of increase (SurvStat@RKI 2.0). It was assumed pork liver pâté served in a restaurant caused an outbreak that the registered cases were mainly persons infected on (Yapa et al. 2016). trips to Asia or Africa (Federal Institute for Risk Assess- From the aforementioned six viral agents, only HAV has ment 2010). However, in recent years HEV was isolated to be mandatorily reported in Switzerland. The recorded from swine suggesting that hepatitis E is a zoonotic disease cases of hepatitis A are mainly due to travelling activities and the only food-borne outbreak with HAV was registered in the year 2000 where a shedder working in a bakery * Dominik Moor shop was identified as the source of infections (Schmid email@example.com and Baumgartner 2012). In contrast to Germany, hepatitis Risk Assessment Division, Federal Food Safety E is currently not a reportable disease. Other than sporadic and Veterinary Office FSVO, Schwarzenburgstrasse 155, cases, outbreaks have to be reported, but to date such an Bern 3003, Switzerland Vol.:(0123456789) 1 3 264 Food and Environmental Virology (2018) 10:263–271 incident was never registered in Switzerland. The lack of Preparation of Food Samples for Virus Extraction data from the official reporting system is a drawback but at least, a study with blood donors is available showing a Extraction of virus particles from sample material was per- low seroprevalence for HEV of 4.9% in Switzerland (Kauf- formed according to a published method (Szabo et al. 2015). mann et al. 2011). This prevalence was considerably below The sample preparation step was slightly modified. For each the values of similar investigations in England (13.5%), food sample, a single extraction was performed. Samples France (16.6%), and Denmark (20.6%). In contrast, a high of 2 g liver sausage or 5 g raw meat sausage were manually prevalence of 60% for antibodies against HEV in pigs at chopped using a surgical blade and transferred into a sterile slaughter was found (Wacheck et al. 2012). Another study 400-ml blender bag with side-filter (BagFilter P). Then 7 ml carried out in the same period of time revealed a sero- TRI Reagent Solution (Ambion), 5 ml sterile phosphate buff- prevalence in domestic pigs of 58.1% and in wild boars ered saline (PBS, pH 7.4, Sigma) and 50 µl of a process control of 12.5% (Burri et al. 2014). In a recent publication, two solution consisting of bacteriophage MS2 (DSM 13767) and HEV isolates from a patient hospitalized in Switzerland mengovirus (MeV) strain vMC (ATCC VR-1597) diluted in with acute hepatitis and from a raw sausage containing pig PBS were added. The amount of both process control viruses 5 6 liver were sequenced in full length. The analysis implied added to each sample corresponded to about 10 and 10 that the two isolates belong to the same virus strain and genome copies for the screening study of ready-to-eat meat that they may form a Swiss-specific HEV genotype 3 sub- products and the method optimization experiments, respec- cluster (Kubacki et al. 2017). These studies suggested that tively. Samples were homogenized for 2 min at highest velocity there could be a relevant reservoir for HEV in Switzerland using a Stomacher laboratory blender (Seward). The liquid and it was therefore hypothesized by us that meat prod- was removed by pipetting from the filter partition, transferred ucts from domestic pigs and wild boars are contaminated into a graduated 50-ml tube and the volume was determined. with HEV. To find out to what extent such contaminations Then the tube was centrifuged at 10,000×g for 20 min at 4 °C might occur, domestic and imported meat products con- (Heraeus Biofuge Stratos), the resulting supernatant was pipet- taining raw pork or pork liver were screened for the pres- ted into a new 50-ml tube and 1.4 ml chloroform was added. ence of HEV with a molecular detection method. Testing After thoroughly mixing for 15 s, the tube was incubated at foods for the presence of viruses is time-consuming and room temperature for 10 min, then centrifuged at 10,000×g for laborious. A critical point is often the limited virus extrac- 15 min at 4 °C. The upper (aqueous) phase was transferred into tion efficiency of the methods. These difficulties result in a new tube and kept at 4 °C until RNA extraction. the unsatisfactory situation that at least half of viral food- To determine the efficiency of virus recovery from matrix borne outbreaks are not recognized (Stals et al. 2012). samples, a process control sample was prepared by adding We addressed this critical point to find a way to improve 50 µl of the MS2/MeV process control solution to a volume the sensitivity of a recognized detection method published of PBS, equal to the volume of liquid recovered after sample earlier (Szabo et al. 2015). Finally, considerations about homogenization (8–10 ml). The solution was kept at 4 °C until the risk of HEV-contaminated foods were made and pos- RNA extraction. For the HEV screening study of ready-to- sible consequences for authorities of food control and con- eat meat products, one process control sample of an average sumers were discussed. volume of 9 ml was prepared for up to 12 matrix samples processed in parallel. The extraction recovery rate of HEV was determined alike in spike experiments using 50 µl of diluted supernatant of cell Materials and Methods culture infected with HEV genotype 3 strain 47832c (Johne et al. 2014). The amount of HEV corresponded to about Collection of Samples 5 × 10 genome copies. This HEV inoculum was added to previously HEV-negative tested matrix sample homogenates A total of 90 ready-to-eat food products of domestic and containing MS2/MeV process control solution, as well as to imported pork liver sausages and raw meat sausages were the corresponding volume of PBS (8-10 ml), also containing purchased at retail shops in Switzerland between February MS2/MeV process control solution. The solution was kept at and April 2016. Liver sausages should have passed a heating 4 °C until RNA extraction. step. Local sausage specialties (“Mortadella cruda”, “Leber- salsiz”, “Salametti”) containing raw pork liver or raw game Determination of the Detection Limit meat of deer, wild boar, chamois and ibex were provided by Agroscope, Research Division Food Microbial Systems. To determine the detection limit of HEV in matrix sam- Samples were stored at 4 °C for short term or at − 20 °C for ples, cell culture supernatant infected with HEV genotype 3 long term until virus extraction. 1 3 Food and Environmental Virology (2018) 10:263–271 265 strain 47832c (Johne et al. 2014) was serially diluted in PBS repeatedly resulted in a positive outcome, the sample was and an aliquot of 50 µl was added to HEV-negative matrix judged as HEV positive. sample homogenates, together with 50 µl of the MS2/MeV process control solution. The virus extraction was performed Calculation of Extraction Efficiency as previously described. The virus extraction efficiency or the recovery rate, Isolation of Viral RNA expressed as a percentage (%), of the MS2 and MeV process controls or of HEV spiked to matrix samples, was calculated Total RNA was extracted from 1 ml of sample using the according to the equation: NucliSens magnetic extraction system (BioMérieux). −ΔCq E = 2 × 100, Extraction was done manually with a magnetic rack, accord- whereas ΔCq is the difference in the RT-qPCR quantifica - ing to the user manual. RNA was eluted in 60 µl elution tion cycle (Cq) of the matrix sample and the corresponding buffer and immediately used for further analysis or stored process control sample. The Cq of the process control sam- at − 80 °C. ple represented 100% extraction efficiency. Recovery rates of matrix samples slightly exceeding 100% due to variation RT‑qPCR for HEV, MS2 and MeV RNA Detection of Cq values were rounded to 100%. Quantitative reverse transcription PCR (RT-qPCR) was per- formed on a Rotor-Gene Q thermocycler (Qiagen) in dupli- Calculation of HEV Copy Numbers cate reactions. The QuantiTect Probe RT-PCR Kit (Qiagen) was used in a total reaction volume of 25 µl, with 5 µl RNA The HEV genome copy number per gram of positive food extract or PCR-grade water as no template control (NTC). samples was determined by HEV RT-qPCR in triplicate Primer and probe concentrations were 0.4 and 0.2 µM, reactions, using a standard curve generated of a serially respectively. The oligonucleotide sequences are outlined diluted HEV PCR product of known concentration. This in Table 1. Cycling conditions were as follows: 50 °C for HEV standard was previously amplified using extracted 30 min, 95 °C for 15 min, then 45 cycles each with 94 °C RNA (QIAamp Viral RNA Mini Kit, Qiagen) from super- for 10 s, 55 °C for 20 s and 72 °C for 60 s. Data acquisition natant of a cell culture infected with HEV genotype 3 strain was done after the extension step at 72 °C. For data analy- 47832c (Johne et al. 2014). This RNA was further used as sis, the Rotor-Gene software version 2.2.3 was used. RNA positive template control (PTC) in HEV RT-qPCR. The samples that were tested positive for HEV (positive ampli- PCR product standard was purified (NucleoSpin Gel and fication signal for both duplicate reactions) were confirmed PCR Clean-up Kit, Macherey–Nagel) and measured on a by a second RT-qPCR measurement. If duplicate reactions ND-1000 Spectrophotometer (NanoDrop). The copy number Table 1 Oligonucleotides used for RT-qPCR detection of HEV, MS2 and MeV RNA, and for genotyping of HEV PCR system Name Sequence (5′–3′) and modifications Amplicon References length (bp) HEV RT-qPCR JVHEVFGGT GGT TTC TGG GGT GAC 70 Jothikumar et al. (2006) JVHEVRAGG GGT TGG TTG GAT GAA JVHEVPFAM-TGA TTC TCA GCC CTT CGC -BHQ1 MS2 RT-qPCR MS2-TM2-FTGC TCG CGG ATA CCCG 61 Dreier et al. (2005) MS2-TM2-RAAC TTG CGT TCT CGA GCG AT MS2-TM2JOEJOE-ACC TCG GGT TTC CGT CTT GCT CGT -BHQ1 MeV RT-qPCR Mengo110GCG GGT CCT GCC GAA AGT 100 Pinto et al. (2009) Mengo209GAA GTA ACA TAT AGA CAG ACG CAC AC Mengo147Cy5-ATC ACA TTA CTG GCC GAA GC-MGB NFQ HEV outer nested PCR HEV-csTCG CGC ATCACMTTY TTC CARAA 469–472 Johne et al. (2010) HEV-casGCC ATG TTC CAG ACDGTR TTC CA HEV inner nested PCR HEV-csnTGT GCT CTG TTT GGCCCNTGG TTY CG 331–334 HEV-casnCCA GGC TCA CCR GAR TGY TTC TTC CA FAM 6-carboxyfluorescein, BHQ1 black hole quencher 1, JOE 6-carboxy-4′,5′-dichloro-2′,7′-dimethoxyfluorescein, Cy5 cyanine dye 5, MGB minor groove binder, NFQ non-fluorescent quencher, bp basepairs 1 3 266 Food and Environmental Virology (2018) 10:263–271 was calculated on the basis of the concentration in ng/µl and inhibitory effect of the liver sausage matrix on the virus the molecular mass of the PCR product. and/or RNA extraction procedure and the RT-qPCR assay was suspected, but not further investigated in detail. Several Genotyping of HEV‑Positive Samples measures for optimizing were tested. By adding 2 ml and 5 ml PBS buffer to the liver sausage samples in the homog- RNA extracts tested positive in the HEV RT-qPCR were enization step, the MS2 recovery rate increased from 0.001 further subjected to genotyping, which was performed to 0.01 and 0.2%, respectively (Table 2). Depending on the following a published nested-PCR approach (Johne et al. sample matrix, the volume of homogenate averaged 3–5 and 2010) with slight modifications. Synthesis of cDNA was 8–10 ml, for undiluted samples and samples with 5 ml PBS done using the High-Capacity cDNA Reverse Transcrip- addition, respectively. Although the addition of 5 ml PBS tion Kit (Applied Biosystems), in a total reaction volume of increased the MS2 Cq value from 24.46 to 25.73, as was 20 µl with 10 µl RNA extract. Both outer and inner nested observed for the process control sample, the matrix sample PCRs were done in a total reaction volume of 25 µl, with 5 µl MS2 Cq value decreased about more than 6, from 40.57 cDNA for the outer nested PCR, and 5 µl reaction mix of to 34.46 (Table 2). In spiking experiments where sample the outer nested PCR for the inner nested PCR, respectively. homogenates were artificially inoculated with HEV, it was One unit of Red Diamond Taq DNA polymerase (Eurogen- shown that both MS2 and HEV recovery rates were con- tec) was used with 1x reaction buffer, 2 mM MgCl and comitantly increased in a similar extent by the addition of 0.2 mM dNTP each. Primer concentrations were 0.5 µM. PBS. A balance between overcoming the inhibitory effect The oligonucleotide sequences are outlined in Table 1. and loss of analytical sensitivity due to sample dilution was Cycling conditions were as follows: 95 °C for 3 min, then achieved by adding 5 ml of PBS at most. The addition of 40 cycles each with 94 °C for 30 s, 50 °C for 30 s and 72 °C PBS increased the recovery rates for MS2 process control for 60 s, with a final extension at 72 °C for 5 min. PCR for liver sausages and raw meat sausages from 0.001 to 0.1% products were separated on 1.5% agarose gel and visualized and 0.4 to 2.2%, respectively (Table 3). For HEV, the recov- with HD Green Plus DNA stain (Intas) under UV-light. The ery rates increased for both food matrices from 0.1 to 62.0% inner nested PCR products were purified (NucleoSpin Gel and 33.7 to 92.7%, respectively. Other diluents like sodium and PCR Clean-up Kit, Macherey–Nagel) and DNA strands dodecyl sulphate (SDS) buffer commonly used for genomic were sequenced (Microsynth) in separate forward- and DNA extraction of meat products [10 mM Tris–HCl (pH reverse-primer reactions. The edited nucleotide sequences 8.0), 150 mM NaCl, 2 mM EDTA, 1% (w/v) SDS, 0.5 M were analysed for phylogenetic relationship using the web guanidine-HCl] (Meyer et al. 1996) could not improve the service Phylogeny.fr (Dereeper et al. 2010, 2008). A set of recovery rates. previously described sequences with HEV genotypes 1–4 were used as reference (Johne et al. 2010). Table 3 Influence of PBS and SDS buffers on the MS2 and HEV virus extraction recovery rates of liver sausage and raw meat sausage Results Undiluted +5 ml PBS +5 ml SDS Optimizing of Virus Extraction RR (%) RR (%) RR (%) Liver sausage (2 g) MS2 0.001 0.1 0.1 Our first attempts to establish a published method for HEV HEV 0.1 62.0 49.7 detection (Szabo et al. 2015) from 2-g samples of a liver Raw meat sausage MS2 0.4 2.2 2.1 sausage containing 18% pork liver resulted in low extrac- (5 g) HEV 33.7 92.7 72.2 tion recovery rates of the MS2 process control of 0.001% (Tables 2, 3). The recovery rate for 5-g samples of raw meat Results were obtained from a single experiment sausages was 0.4% (Table 3) which is still below 1%. An RR recovery rate Table 2 Influence of PBS buffer Undiluted (5 ml) +2 ml PBS (7 ml) +5 ml PBS (9 ml) on Cq values of MS2 and HEV RT-qPCR and virus extraction Cq Cq RR (%) Cq Cq RR (%) Cq Cq RR (%) ms pcs ms pcs ms pcs recovery rates measured for 2 g MS2 40.57 24.46 0.001 37.70 24.94 0.01 34.46 25.73 0.2 samples of liver sausage HEV 37.78 31.20 1.0 34.81 31.88 13.1 31.54 31.96 100 The total lysate volume is indicated in brackets. Results were obtained from a single experiment Cq Cq value of matrix sample, Cq Cq value of process control sample, RR recovery rate ms pcs 1 3 Food and Environmental Virology (2018) 10:263–271 267 The recovery rates observed for HEV were higher than sausages and 3 raw meat sausages. This represents a HEV- for MS2. To test the influence of a particular process control positive rate of 18.9% for liver sausages and 5.7% for raw virus on the extraction efficiency, MeV was evaluated as meat sausages. Three of the positive samples were imported process control, in the same way that MS2. For both food products from Germany. 22 of the 53 analysed raw meat matrices, the experiment revealed higher recovery rates for sausages (13 “Mortadella cruda” and 9 “Lebersalsiz”) repre- MeV (18.8, 51.4%), compared to MS2 (0.6, 27.6%), and sented high-risk products to be contaminated with infectious HEV showing the highest values of 55.5 and 97.3%, respec- HEV particles since they contained raw pork liver. HEV- tively (Table 4). To assess the performance of both process RNA was detected in 3 of these samples (13.6%), twice controls on a broader database, MS2 as well as MeV were “Mortadella cruda” and once “Lebersalsiz”. No HEV-RNA used for testing of ready-to-eat meat products containing was detected in the other 31 raw meat sausages that did not pork liver or raw meat on the Swiss retail market. contain liver, including 15 products made of game meat like The detection limit for HEV in matrix samples was about deer (7), wild boar (5), chamois (2) and ibex (1). The range 3 2 1.56 x10 and 1.56 x10 genome copies per gram of liver of virus load was between 1.72 and 5.54 log HEV genome sausage and raw meat sausage, respectively (Table 5). Over copies per gram (Table 7). the whole HEV dilution range high recovery rates between For all samples, the determined average virus extraction 10.4 and 33.0% were measured. recovery rates for the MS2 and MeV process controls were 10.0 and 24.4%, respectively (Table 8). Screening of Ready‑to‑Eat Meat Products from the Swiss Retail Market Genotyping of HEV‑Positive Samples A total of 90 samples of pork liver sausages (37) and raw Nine of the 10 HEV-positive tested samples were success- meat sausages (53) were analysed for HEV using the pre- fully genotyped. The mean Cq values of the HEV RT-qPCR viously optimized method (Tables 6, 7). HEV-RNA was ranged from 26.97 to 38.17. In one sample of liver sausage detected in 10 samples (11.1%), 7 of which were liver Table 6 Results of the HEV screening of meat products from the Swiss retail market Table 4 Virus extraction recovery rates of MS2 and MeV process controls and HEV, for liver sausage and raw meat sausage, with addi- Samples HEV-positive HEV- tion of 5 ml PBS buffer tested samples positive rate (%) MS2 MeV HEV RR (%) RR (%) RR (%) Liver sausages 37 7 18.9 Liver sausage (2 g) 0.6 18.8 55.5 Raw meat sausages 53 3 5.7 Raw meat sausage (5 g) 27.6 51.4 97.3 With raw liver 22 3 13.6 With game meat 15 0 0 Results were obtained from a single experiment Total 90 10 11.1 RR recovery rate Table 5 Detection limit of HEV inoculation level [log genome copies] HEV in matrix samples, the corresponding virus extraction 4.70 4.10 3.49 2.89 2.29 NTC recovery rates of MS2 and MeV Liver sausage (2 g) MS2 RR (%) 15.3 9.4 7.3 5.0 25.5 16.9 process controls and HEV, and the Cq values of HEV RT-qPCR MeV RR (%) 39.0 31.5 29.0 21.6 33.8 60.7 HEV RR (%) 24.0 28.1 26.8 – – – HEV Cq 32.70 35.21 38.26 – – – ms HEV Cq 30.64 33.38 36.36 38.37 – – pcs Raw meat sausage (5 g) MS2 RR (%) 0.3 1.1 2.2 3.1 0.9 0.1 MeV RR (%) 19.3 35.4 43.4 32.1 11.5 6.8 HEV RR (%) 14.8 28.1 33.0 10.4 – – HEV Cq 33.56 34.97 37.34 39.77 – – ms HEV Cq 30.80 33.14 35.74 36.51 – – pcs Results were obtained from a single experiment RR recovery rate, Cq Cq value of matrix sample, Cq Cq value of process control sample, NTC no tem- ms pcs plate control (matrix sample without HEV inoculation) 1 3 268 Food and Environmental Virology (2018) 10:263–271 Table 7 Product type and Product type Country MS2 MeV HEV RT-HEV (log HEV genotype origin, virus extraction recovery of origin RR (%) RR (%) qPCR (Cq genome rates of the MS2 and MeV value) copies/g) process controls, mean Cq value of HEV RT-qPCR, and Liver sausage CH 0.2 n.d. 31.97 4.01 3 genotype of HEV-positive Liver sausage CH 6.1 n.d. 36.86 2.51 n.d. samples Liver sausage CH 2.8 n.d. 28.38 5.10 3 Liver sausage DE 1.2 25.3 26.97 5.54 3 Liver sausage DE 1.3 25.8 31.19 4.25 3 Liver sausage DE 1.6 11.5 29.97 4.62 3 Liver sausage CH 3.2 34.3 34.42 3.26 3 Raw meat sausage CH 0.01 0.2 38.17 1.72 3 Raw meat sausage CH 0.01 0.7 32.35 3.49 3 Raw meat sausage CH 0.5 6.0 34.34 2.89 3 RR recovery rate, CH Switzerland, DE Germany, n.d. not determined Local sausage specialty “Lebesalsiz” Local sausage specialty “Mortadella cruda” with a Cq value of 36.86 genotyping was not successful. The the mean extraction efficiency of MS2 for raw sausages analysis of the phylogenetic relationship of the inner nested raised to 11.2%, but for liver sausages it did not exceed PCR product sequences revealed an unambiguous assign- 1%. Based on this experience, we addressed the issue of ment to HEV genotype 3 (Table 7). improving the virus recovery rate in our study, especially for difficult food matrices like liver sausage. The addi - tion of PBS buffer to the sample in the homogenization Discussion step resulted in an increased extraction efficiency for MS2. Even higher recovery rates were obtained by using MeV Optimizing of Virus Extraction as process control instead of MS2. For liver sausages, the range of recovery rates for MeV was from 0.8 to 100%, For reliable detection and quantification of viruses in food with a mean 41.2%, showing a high variability in recovery samples, a method featuring a high extraction efficiency rates (Table 8). The range of recovery rates for HEV was is vital. The optimization of such methods is therefore between 10.4 and 100% (Tables 2, 5) for both inoculated crucial to achieve a low detection limit and for accu- sample matrices. rate measurement of the actual virus load. In a recently If required, our method can easily be scaled up for conducted study (Szabo et al. 2015), different published homogenization of higher amounts of sample material, just extraction methods (Baert et al. 2008; Stals et al. 2011) by keeping the proportion of sample and liquids, e.g. PBS were assessed for their recovery rates by testing of a raw and TRI Reagent Solution. The procedure of virus extrac- sausage matrix using MS2 as process control virus, as tion is simple and can be done with standard equipment and a surrogate for HEV. The range of recovery rates was consumables that are available in a microbiological labo- between 0.04 and 1.92%, whereas for the same sample ratory. Another method for HEV extraction of increased matrix inoculated with HEV a recovery rate of 4.9% was sample size of 25-g pork meat products has been published achieved. Their T rizol -based protocol was further opti- recently (Mykytczuk et al. 2017). They reached an aver- mized with regard to the homogenization technique. Thus, age extraction efficiency of 5.45% for the feline calicivirus Table 8 Virus extraction N MS2 RR (%) MeV RR (%) recovery rates (mean, min, max, and median values) of MS2 Mean Min Max Median Mean Min Max Median and MeV process controls for All samples 90 10.0 0.01 100 1.2 24.4 0.1 100 14.3 different sample subsets Liver sausage 37 22.3 0.01 100 3.2 41.2 0.8 100 32.4 Raw meat sausage 53 1.4 0.01 15.6 0.4 12.0 0.1 57.4 4.7 HEV-positive samples 10 1.7 0.01 6.1 1.3 14.8 0.2 34.3 11.5 n number of samples, RR recovery rate 1 3 Food and Environmental Virology (2018) 10:263–271 269 (FCV) process control and 5.63% for HAV, but their method of 12.5 and 29.9–41.3% were measured for Switzerland and is laborious, time-consuming and uses an in-house con- Germany, respectively (Adlhoch et al. 2009; Burri et al. structed filter device. 2014; Schielke et al. 2015). The detection limit for HEV in liver sausages and All sequences retrieved from HEV-positive samples raw meat sausages was determined to be 5.34 × 10 and belonged to HEV genotype 3, which further supports the 2.93 × 10 genome equivalents per gram, respectively conclusive presumption that pigs are a reservoir of this (Szabo et al. 2015). Similar results were obtained in a pre- zoonotic genotype. The sample for which genotyping vious study investigating swine organs and tissues at slaugh- failed, a rather high Cq value of 36.86 was measured. Since terhouse (Leblanc et al. 2010). The limit of detection of our outer and inner nested PCR products of the HEV genotyp- optimized method was tenfold lower, about 1.56 × 10 and ing approach are longer than the HEV RT-qPCR product 1.56 × 10 genome copies per gram, for liver sausages and (about 470 bp and 330 bp versus 70 bp), it is possible that raw meat sausages, respectively (Table 5). A higher sensitiv- for strongly processed matrices with reduced RNA integrity ity could for example be achieved by HEV replication prior the amplification of larger PCR templates will fail. to RT-qPCR detection, e.g. in a cell culture-based in vitro assay or by other alternative techniques. Public Health Considerations Screening of Ready‑to‑Eat Meat Products The HEV detection method developed in this study was applied to test certain categories of meat products for HEV, Since no data are available on the frequency of contami- to find out frequencies of contamination, to measure viral nation and viral loads of HEV in meat products from the loads and to genotype isolated HEV sequences. The obtained Swiss retail market, this study aimed to close this gap at least data showed us that HEV occurs rather frequently in certain for ready-to-eat products containing pork liver or raw meat, pork meat and pork liver products and that in some samples which are suspected to be a source for HEV infections of high HEV-RNA loads of up to 5.54 log genome copies per humans. Our findings for the prevalence of HEV in liver sau- gram can occur. Our findings have to be seen as elements sages (18.9%) and raw meat sausages (5.7%) revealed similar for future risk-assessments. For a full and complete risk- results as presented in other studies conducted in Europe, assessment, much more information would be needed. For in which positive rates of 4–31% were reported (Berto example, it would be necessary to exactly know the produc- et al. 2012; Di Bartolo et al. 2012, 2017; Martin-Latil et al. tion processes of analysed foods. Since we tested products 2014; Szabo et al. 2015; Pavio et al. 2014). Based on HEV purchased at retail level, this information was not available. prevalence in pork livers, a study assessing the risk of food- There are also gaps of knowledge concerning the survival of borne transmission of hepatitis E in Switzerland estimated HEV in food matrices such as ready-to-eat raw meat prod- that < 6% of the products containing pork liver and < 3% ucts containing pork or pork liver. Furthermore, there is a of the products containing pork could be contaminated with lack of information with regard to the infectious dose of HEV genomic material (Muller et al. 2017). However, our HEV or the dose–response relationship, respectively (Sarno experimental data revealed a higher prevalence. Regard- et al. 2017). To clarify that particular question, attempts of ing the viral load in pork products at retail, their estimates cases-source-attributions might be helpful (Muller et al. ranged between 1.6 and 3.5 log HEV genome copies per 2017). However, precondition for such studies would be the gram. Another study reported a similar value of 3.7 log mandatory reporting of human hepatitis E infections what is viral units per gram for the estimated prevalence of HEV- currently not the case in Switzerland. Future studies should RNA in porcine-derived food in Switzerland and Germany also try to quantitatively differentiate between inactivated (Sarno et al. 2017). Our experiments have given slightly and infectious HEV particles in contaminated foods which higher values between 1.72 and 5.54 lo g HEV genome would imply the availability of appropriate cell culture sys- copies per gram. In other studies, similar results for HEV tems. Although there are still several questions to answer, loads of up to 6 log genome copies per gram were meas- the data in this study and already published epidemiological ured for dried and fresh liver sausages (Colson et al. 2010; findings justify certain precautionary measures. For the time Pavio et al. 2014). These data may even underestimate the being, our recommendation is that raw pork liver has to be true HEV load in such products since they have been col- thoroughly heated prior to consumption and that the fabrica- lected with virus extraction methods showing lower recovery tion processes of meat products containing pork liver must rates. contain a virus-inactivating hurdle. Interestingly we did not find HEV in raw meat sausages made of game meat like wild boar or deer, even though posi- Acknowledgements The authors would like to thank Prof. Dr. Rei- tive rates of 10% were reported for wild boar sausages in mar Johne and Dr. Eva Trojnar (Federal Institute for Risk Assessment, Germany (Szabo et al. 2015) and high seroprevalence rates BfR, Berlin, Germany) for providing supernatant of HEV-infected cell 1 3 270 Food and Environmental Virology (2018) 10:263–271 culture. We also thank Livia Schwendimann and Thomas Berger (Agro- Dreier, J., Stormer, M., & Kleesiek, K. (2005). Use of bacteriophage scope, Bern, Switzerland) for providing food samples. MS2 as an internal control in viral reverse transcription-PCR assays. Journal of Clinical Microbiology, 43(9), 4551–4557. https ://doi.org/10.1128/JCM.43.9.4551-4557.2005. Funding This study was commissioned and supported by the Federal Di Bartolo, I., Ponterio, E., Angeloni, G., Morandi, F., Ostanello, F., Food Safety and Veterinary Office FSVO, Bern, Switzerland. Nicoloso, S., et al. (2017). Presence of hepatitis E virus in a RED deer (Cervus elaphus) population in Central Italy. Transboundary Compliance with Ethical Standards and Emerging Diseases, 64(1), 137–143. https://doi.or g/10.1111/ tbed.12353 . Conflict of interest The authors declare that they have no conflict of Faber, M. S., Wenzel, J. J., Jilg, W., Thamm, M., Hohle, M., & Stark, interest. K. (2012). Hepatitis E virus seroprevalence among adults, Ger- many. Emerging Infectious Diseases, 18(10), 1654–1657. https Ethical Approval This article does not contain any studies with human ://doi.org/10.3201/eid18 10.11175 6. participants or animals performed by any of the authors. Federal Institute for Risk Assessment, B. (2010). Hepatitis E-virus in deutschen Wildschweinen (pp. 1–3). Federal Institute for Risk Assessment, BfR. Open Access This article is distributed under the terms of the Crea- Guillois, Y., Abravanel, F., Miura, T., Pavio, N., Vaillant, V., tive Commons Attribution 4.0 International License (http://creat iveco Lhomme, S., et al. (2016). High proportion of asymptomatic mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- infections in an outbreak of hepatitis E associated with a spit- tion, and reproduction in any medium, provided you give appropriate roasted piglet, France, 2013. Clinical Infectious Diseases, 62(3), credit to the original author(s) and the source, provide a link to the 351–357. https ://doi.org/10.1093/cid/civ86 2. Creative Commons license, and indicate if changes were made. Johne, R., Plenge-Bonig, A., Hess, M., Ulrich, R. G., Reetz, J., & Schielke, A. (2010). Detection of a novel hepatitis E-like virus in faeces of wild rats using a nested broad-spectrum RT-PCR. Journal of General Virology, 91(Pt 3), 750–758. https ://doi. org/10.1099/vir.0.01658 4-0. References Johne, R., Reetz, J., Ulrich, R. G., Machnowska, P., Sachsenroder, J., Nickel, P., et al. (2014). An ORF1-rearranged hepatitis E virus Adlhoch, C., Wolf, A., Meisel, H., Kaiser, M., Ellerbrok, H., & Pauli, derived from a chronically infected patient efficiently replicates G. (2009). High HEV presence in four different wild boar popu- in cell culture. Journal of Viral Hepatitis, 21(6), 447–456. https lations in East and West Germany. Veterinary Microbiology, ://doi.org/10.1111/jvh.12157 . 139(3–4), 270–278. https://doi.or g/10.1016/j.vetmic.2009.06.032 . Jothikumar, N., Cromeans, T. L., Robertson, B. H., Meng, X. J., & Baert, L., Uyttendaele, M., & Debevere, J. (2008). Evaluation of viral Hill, V. R. (2006). A broadly reactive one-step real-time RT-PCR extraction methods on a broad range of Ready-To-Eat foods with assay for rapid and sensitive detection of hepatitis E virus. Journal conventional and real-time RT-PCR for Norovirus GII detection. of Virological Methods, 131(1), 65–71. https ://doi.org/10.1016/j. International Journal of Food Microbiology, 123(1–2), 101–108. jviro met.2005.07.004. https ://doi.org/10.1016/j.ijfoo dmicr o.2007.12.020. Kaufmann, A., Kenfak-Foguena, A., Andre, C., Canellini, G., Bur- Berto, A., Martelli, F., Grierson, S., & Banks, M. (2012). Hepatitis E gisser, P., Moradpour, D., et al. (2011). Hepatitis E virus sero- virus in pork food chain, United Kingdom, 2009-2010. Emerging prevalence among blood donors in southwest Switzerland. PLoS Infectious Diseases, 18(8), 1358–1360. https ://doi.org/10.3201/ ONE, 6(6), e21150. https://doi.or g/10.1371/journal.pone.00211 eid18 08.11164 7. Blasco-Perrin, H., Abravanel, F., Blasco-Baque, V., & Peron, J. M. Kubacki, J., Fraefel, C., Jermini, M., Giannini, P., Martinetti, G., Ripel- (2016). Hepatitis E, the neglected one. Liver International, lino, P., et al. (2017). Complete genome sequences of two swiss 36(Suppl 1), 130–134. https ://doi.org/10.1111/liv.13014 . hepatitis E virus isolates from human stool and raw pork sausage. Burri, C., Vial, F., Ryser-Degiorgis, M. P., Schwermer, H., Darling, Genome Announcements. https: //doi.org/10.1128/genome A.00888 K., Reist, M., et al. (2014). Seroprevalence of hepatitis E virus -17. in domestic pigs and wild boars in Switzerland. Zoonoses Public Le Guyader, F., Haugarreau, L., Miossec, L., Dubois, E., & Pomme- Health, 61(8), 537–544. https ://doi.org/10.1111/zph.12103 . puy, M. (2000). Three-year study to assess human enteric viruses Colson, P., Borentain, P., Queyriaux, B., Kaba, M., Moal, V., Gallian, in shellfish. Applied and Environment Microbiology, 66(8), P., et al. (2010). Pig liver sausage as a source of hepatitis E virus 3241–3248. transmission to humans. Journal of Infectious Diseases, 202(6), Leblanc, D., Poitras, E., Gagne, M. J., Ward, P., & Houde, A. (2010). 825–834. https ://doi.org/10.1086/65589 8. Hepatitis E virus load in swine organs and tissues at slaughter- Dereeper, A., Audic, S., Claverie, J. M., & Blanc, G. (2010). house determined by real-time RT-PCR. International Journal of BLAST-EXPLORER helps you building datasets for phyloge- Food Microbiology, 139(3), 206–209. https:/ /doi.org/10.1016/j. netic analysis. BMC Evolutionary Biology, 10, 8. https ://doi. ijfoo dmicr o.2010.02.016. org/10.1186/1471-2148-10-8. Martin-Latil, S., Hennechart-Collette, C., Guillier, L., & Perelle, S. Dereeper, A., Guignon, V., Blanc, G., Audic, S., Buffet, S., Chevenet, (2014). Method for HEV detection in raw pig liver products and F., et al. (2008). Phylogeny.fr: robust phylogenetic analysis for its implementation for naturally contaminated food. International the non-specialist. Nucleic Acids Research, 36(Web Server issue), Journal of Food Microbiology, 176, 1–8. https://doi.or g/10.1016/j. W465–W469. https ://doi.org/10.1093/nar/gkn18 0. ijfoo dmicr o.2014.01.016. Di Bartolo, I., Diez-Valcarce, M., Vasickova, P., Kralik, P., Hernandez, Meyer, R., Chardonnens, F., Hubner, P., & Luthy, J. (1996). Polymer- M., Angeloni, G., et al. (2012). Hepatitis E virus in pork produc- ase chain reaction (PCR) in the quality and safety assurance of tion chain in Czech Republic, Italy, and Spain, 2010. Emerging food: detection of soya in processed meat products. Zeitschrift Infectious Diseases, 18(8), 1282–1289. https ://doi.org/10.3201/ für Lebensmittel-Untersuchung und Forschung, 203(4), 339–344. eid18 08.11178 3. Muller, A., Collineau, L., Stephan, R., Muller, A., & Stark, K. D. (2017). Assessment of the risk of foodborne transmission and 1 3 Food and Environmental Virology (2018) 10:263–271 271 burden of hepatitis E in Switzerland. International Journal of for ready-to-eat foods. International Journal of Food Microbi- Food Microbiology, 242, 107–115. https://doi.or g/10.1016/j.ijfoo ology, 145(2–3), 420–425. https ://doi.org/10.1016/j.ijfoo dmicr dmicr o.2016.11.018. o.2011.01.013. Mykytczuk, O., Harlow, J., Bidawid, S., Corneau, N., & Nasheri, N. Stals, A., Baert, L., Van Coillie, E., & Uyttendaele, M. (2012). Extrac- (2017). Prevalence and molecular characterization of the hepatitis tion of food-borne viruses from food samples: a review. Interna- E virus in retail pork products marketed in Canada. Food and tional Journal of Food Microbiology, 153(1–2), 1–9. https ://doi. Environmental Virology, 9(2), 208–218. https ://doi.org/10.1007/org/10.1016/j.ijfoo dmicr o.2011.10.014. s1256 0-017-9281-9. SurvStat@RKI 2.0. Robert Koch Institute. https ://survs tat.rki.de. Pavio, N., Merbah, T., & Thebault, A. (2014). Frequent hepatitis E Szabo, K., Trojnar, E., Anheyer-Behmenburg, H., Binder, A., Schotte, virus contamination in food containing raw pork liver, France. U., Ellerbroek, L., et al. (2015). Detection of hepatitis E virus Emerging Infectious Diseases, 20(11), 1925–1927. https ://doi. RNA in raw sausages and liver sausages from retail in Ger- org/10.3201/eid20 11.14089 1. many using an optimized method. International Journal of Food Pinto, R. M., Costafreda, M. I., & Bosch, A. (2009). Risk assessment Microbiology, 215, 149–156. https://doi.or g/10.1016/j.ijfoodmicr in shellfish-borne outbreaks of hepatitis A. Applied and Environ- o.2015.09.013. mental Microbiology, 75(23), 7350–7355. https: //doi.org/10.1128/ Wacheck, S., Sarno, E., Martlbauer, E., Zweifel, C., & Stephan, R. AEM.01177 -09. (2012). Seroprevalence of anti-hepatitis E virus and anti-Salmo- Sarno, E., Martin, A., McFarland, S., Johne, R., Stephan, R., & Greiner, nella antibodies in pigs at slaughter in Switzerland. Journal of M. (2017). Estimated exposure to hepatitis E virus through con- Food Protection, 75(8), 1483–1485. https: //doi.org/10.4315/0362- sumption of swine liver and liver sausages. Food Control, 73, 028X.JFP-12-058. 821–828. https ://doi.org/10.1016/j.foodc ont.2016.09.030. Wichmann, O., Schimanski, S., Koch, J., Kohler, M., Rothe, C., Plentz, Schielke, A., Ibrahim, V., Czogiel, I., Faber, M., Schrader, C., Drem- A., et al. (2008). Phylogenetic and case-control study on hepati- sek, P., et al. (2015). Hepatitis E virus antibody prevalence in tis E virus infection in Germany. Journal of Infectious Diseases, hunters from a district in Central Germany, 2013: a cross-sectional 198(12), 1732–1741. https ://doi.org/10.1086/59321 1. study providing evidence for the benefit of protective gloves dur - Worm, H. C., Schlauder, G. G., & Brandstatter, G. (2002). Hepati- ing disembowelling of wild boars. BMC Infectious Diseases, 15, tis E and its emergence in non-endemic areas. Wiener Klinische 440. https ://doi.org/10.1186/s1287 9-015-1199-y. Wochenschrift, 114(15–16), 663–670. Schmid, H., Baumgartner, A. (2012). Lebensmittelbedingte Gruppener- Yapa, C. M., Furlong, C., Rosewell, A., Ward, K. A., Adamson, S., krankungen in der Schweiz (pp. 1–89). Federal Office of Public Shadbolt, C., et al. (2016). First reported outbreak of locally Health FOPH. acquired hepatitis E virus infection in Australia. Medical Journal Stals, A., Baert, L., De Keuckelaere, A., Van Coillie, E., & Uyttend- of Australia, 204(7), 274. aele, M. (2011). Evaluation of a norovirus detection methodology 1 3
Food and Environmental Virology – Springer Journals
Published: Feb 28, 2018
Access the full text.
Sign up today, get DeepDyve free for 14 days.