ABSTRACT

Foodborne viruses such as norovirus and hepatitis A virus (HAV) are highly transmissible, persistent in the environment, and resistant to many conventional inactivation methods. Foods can become contaminated with these viruses either at the source of harvest or during food handling and processing. Multiple lines of evidence suggest that foodborne viruses can survive desiccation and dry conditions. Several foodborne virus outbreaks have been linked to low-moisture foods (LMFs), indicating that these foods can be vehicles of virus transmission. However, the efficiencies of common virus extraction methodologies have not been examined with LMFs. We adapted the International Organization for Standardization (ISO) 15216-1:2017 method for virus recovery for use with chocolate, pistachios, and cornflakes. We also developed a magnetic bead assay for the recovery of HAV from LMFs and used the porcine gastric mucin–coated magnetic beads (PGM-MBs) to extract norovirus surrogates, feline calicivirus (FCV), and murine norovirus (MNV) from the same LMFs. The efficiency of virus recovery using the bead-based assay was then compared with that of the ISO 15216-1:2017 method. In chocolate and pistachios, the recovery rates with the PGM-MB method were 5.6- and 21.3-fold higher, respectively, for FCV and 1.65- and 18-fold higher, respectively, for MNV than those with the ISO 15216-1:2017 method. However, the PGM-MB method failed to recover MNV and FCV from cornflakes. The recovery rates for HAV in chocolate, pistachios, and corn flakes with the magnetic bead method were 11.5-, 3-, and 5.6-fold higher, respectively, than those with the ISO 15216-1:2017 method. Thus, depending upon the food matrix and the target virus, the bead-based assays can be used to efficiently and rapidly extract viruses from LMFs.

HIGHLIGHTS
• Anti-HAV–coated magnetic beads can be used to efficiently extract HAV from LMFs.

• Extraction efficiency of the PGM-MB method depends on the tested matrix.

• Bead-based assays are faster and have fewer steps than the ISO 15216-1:2017 method.

Foodborne viruses such as human norovirus (NoV) and hepatitis A virus (HAV) are major causes of nonbacterial foodborne illnesses worldwide (25, 36). These viruses are shed at high levels (>108 particles per g) in the feces of infected individuals (52, 55), have low infectious doses (2, 45), and may remain infectious for weeks in the environment and food (29, 47).

NoV is a nonenveloped, positive-sense, single-stranded RNA virus that belongs to the Caliciviridae family. Human NoV is the leading cause of foodborne outbreaks in North America (16) and causes self-limiting, acute gastroenteritis that normally lasts 2 to 3 days but can last longer in immunocompromised individuals (4). To date, the study of NoV survival and inactivation has been hampered by the lack of a routine and robust cell culture system (18); therefore, culturable viruses such as murine NoV (MNV) and feline calicivirus (FCV), which also belong to the Caliciviridae family, have been used as surrogates to examine NoV survival and inactivation in foods and the environment (25, 34).

HAV is a single-stranded RNA virus of the Picornaviridae family and causes acute hepatitis that typically lasts for <2 months, but some patients may have prolonged or relapsing symptoms for up to 6 months (26). Despite the presence of an effective vaccine against HAV, massive outbreaks and sporadic cases of hepatitis A illness from the consumption of contaminated foods continue to be reported worldwide (26).

NoV and HAV were identified as the leading viral agents responsible for numerous gastroenteritis outbreaks related to the consumption of contaminated produce (10, 11, 41). Fresh produce may acquire viral contamination by direct contact with fecally contaminated soil, water, or hands, and in the absence of proper decontamination and inactivation processes these viruses cause illnesses in consumers. Foodborne viruses can survive long-term storage under low-moisture conditions (37).

Although numerous studies have been conducted on decontamination methods and, inactivation and survival of enteric viruses in fresh fruits and produce, limited information is available regarding efficient virus recovery methods and virus survival and transmission in low-moisture foods (LMFs), i.e., foods with a water activity of <0.85 (59). LMFs are generally considered ready-to-eat products and undergo minimal or no pathogen reduction steps (52). Numerous outbreaks of foodborne virus infections associated with LMFs have been reported in recent years, including foodborne HAV infections linked to semidried tomatoes in Europe in 2010 and 2011 (8, 17) and more recently multiple HAV outbreaks in Europe associated with imported dates (1, 21). In 2014, a large NoV outbreak was attributed to contaminated bread in China (23), and NoV GII.17–contaminated dried nori caused a series of foodborne outbreaks in Japan (48) and South Korea (42). In 2019, a recall was issued for chocolate because of concerns about HAV contamination (57).

Previous work has also indicated that HAV and NoV are resistant to desiccation (12). These observations further suggest that foodborne viruses can endure dry conditions for weeks without losing their infectivity. Conventional pathogen inactivation strategies such as pasteurization cannot be applied to most LMFs, and efficient methodologies for isolation and enumeration of foodborne viruses from LMFs are lacking. The goal of this study was to use a bead-based assay for the extraction of foodborne viruses from artificially contaminated LMFs and to compare the efficiency of this method against that of the International Organization for Standardization (ISO) 15216-1:2017 method (24) adapted for specific foods.

Conventional detection of foodborne viruses in foods, such as with the ISO 15216-1:2017 method, is based on viral precipitation followed by RNA extraction and identification of the viral genome by a reverse transcription (RT) PCR assay. However, these methods are time-consuming and tedious, do not provide information on virus infectivity, and are sensitive to the presence of RT-PCR inhibitors (31). In contrast, bead-based assays allow concentration of virus particles with intact viral capsids from complex food matrices. The viral genome can then be released through heat shock of the captured virus. Therefore, bead-based assays offer several advantages: (i) they have short hands-on times compared with other extraction methods; (ii) they have the potential to be used in automated settings; (iii) they can be used to capture undamaged virus particles (15); and (iv) they allow removal of potential RT-PCR inhibitors found in certain food and environmental matrices (53).

NoV and its surrogates bind to histo-blood group antigens on the surface of certain human cells and to chemically similar porcine gastric mucin (PGM) (27, 54). PGM contains 0.5 to 1.5% sialic acid, which is a receptor for MNV (28). Several studies have been conducted to investigate the potential for conjugation of PGM to magnetic beads (PGM-MBs) to predict the infectivity of treated human NoV and its cultivable surrogates. These studies provide confirmatory evidence for the ability of PGM-MB assays to predict infectivity by direct comparison with infectivity assays (31). Compelling evidence indicates that PGM binding is lost in inactivated human NoV (15), and the PGM-MB method is superior for capturing infectious virus particles compared with other techniques that involve monoclonal antibodies and nucleic acid aptamers (35).

In our previous work, we found that the PGM-MB method can be used to efficiently and rapidly extract human NoV from fresh herbs and leafy vegetables (51). In the present study, we used the PGM-MB assay for extraction of cultivable surrogates of human NoV (FCV and MNV) from select LMFs (Fig. 1). Breakfast cereals and snack foods are important LMFs that are eaten by a large segment of the population, including high-risk individuals such as young children and older persons. In the present study, we used corn flakes (a breakfast cereal) and two popular snack foods (chocolate and pistachios) to examine the efficiency of the virus isolation methods. We also developed a magnetic bead assay for the isolation of HAV by using an antibody against the viral capsid (Supplemental Fig. S1). The efficacy of the bead-based assays was compared with that of the ISO 15216-1:2017 method using a droplet-digital RT-PCR (ddRT-PCR) assay for quantification of the viral genome (40, 46).

FIGURE 1

An overview of the experimental methods used in this study. The average time that each step takes is shown.

FIGURE 1

An overview of the experimental methods used in this study. The average time that each step takes is shown.

## MATERIALS AND METHODS

### Cells and viruses

Crandell Rees feline kidney (CrFK) cells (CCL-94) were obtained from the American Type Culture Collection (ATCC, Manassas, VA) and maintained as previously described (5). FCV ATCC VR-782 was used to inoculate LMFs.

Murine BV-2 cells (from Dr. C. E. Wobus, University of Michigan, Ann Arbor) were grown in Dulbecco's modified Eagle's medium (Invitrogen, Burlington, Ontario, Canada) supplemented with 10% fetal bovine serum, 0.1 mg/mL l-glutamine, and 0.1 mg of penicillin-streptomycin. The cells were incubated at 37°C with 5% CO2 and maintained as previously described (14). MNV (from Dr. H. W. Virgin, School of Medicine, Washington University, St. Louis, MO) also was used to inoculate LMFs.

HAV strain HM-175 (ATCC VR-1402) and seed cultures of fetal rhesus monkey kidney cells (FRhK-4) were provided by Dr. S. A. Sattar (University of Ottawa, Ottawa, Ontario, Canada). FRhK-4 cells were grown, maintained, and prepared as described previously (33).

### Plaque assay

Virus titers were quantified with a plaque assay in 12-well plates (Z707775, MilliporeSigma, St. Louis, MO). For quantification of FCV, MNV, and HAV, the cell lines CrFK, BV-2, and FRhk-4, respectively, were grown, maintained, and infected with the different LMF samples. The cell line monolayers were grown at 37°C overnight, and each of three wells was inoculated with 100 μL of each sample (pistachios, chocolate, or cornflakes) and sample dilution. Samples were diluted with 1× Earle's balanced salt solution (catalogue no. 24010-043, ThermoFisher, Waltham, MA). Samples were adsorbed for 60 min at 37°C with gentle rocking of the plates every 10 min to evenly distribute the sample on the cell monolayer. A 2-mL mixture of agarose and medium was overlaid onto monolayers, and the plates were incubated at 37°C with 5% CO2 for 2 days (FCV and MNV) or 8 days (HAV). The monolayers were fixed with 3.7% paraformaldehyde (F1635, MilliporeSigma) for a minimum of 4 h. The monolayer was stained with 0.1% crystal violet, plaques were counted manually, and counts were converted to PFU per milliliter.

### Inoculation

Four-gram samples of chocolate liquor (Cargill, Minneapolis, MN) were placed into the wells of a six-well plate and melted at 43°C in a bead bath. After the chocolate solidified, it was stored at room temperature until samples were ready for virus inoculation. Chocolate samples were inoculated with 100 μL of FCV at 5 × 106 PFU/mL (6.2 × 105 genome copies per μL), MNV at 8 × 107 PFU/mL (1.05 × 106 genome copies per μL), or HAV at 1 × 105 PFU/mL (1.2 × 105 genome copies per μL) by pipetting small volumes at a time to ensure even distribution of the virus. Virus on the chocolate was dried at room temperature in a biosafety level 2 cabinet for 1 h. Samples were inoculated in triplicate (three wells of 4 g of chocolate per time point) in two sets.

Twenty-five grams of unsalted dry-roasted, shelled pistachios (pasteurized) from the American Pistachio Growers (Fresno, CA) were weighed in petri dishes in two sets, each in triplicate. The pistachio samples were inoculated with 100 μL of FCV at 5 × 106 PFU/mL (6.2 × 105 genome copies per μL), MNV at 8 × 107 PFU/mL (1.05 × 106 genome copies per μL), or HAV at 1 × 105 PFU/mL (1.2 × 105 genome copies per μL). Each pistachio was inoculated with a small droplet of virus stock until all 25 g of pistachios were covered with at least one virus droplet. Inoculated pistachios were dried in a biosafety level 2 cabinet at room temperature for 1 h.

Corn flakes (Kellogg's, Battle Creek, MI) were weighed in 25-g amounts into petri dishes in two sets, each in triplicate. A 100-μL volume of FCV at 5 × 106 PFU/mL (6.2 × 105 genome copies per μL), MNV at 8 × 107 PFU/mL (1.05 × 106 genome copies per μL), or HAV at 1 × 105 PFU/mL (1.2 × 105 genome copies per μL) was used to inoculate the corn flakes, which were then dried in a biosafety level 2 cabinet at room temperature for 30 min.

### ISO 15216-1:2017 method

To recover FCV, MNV, and HAV from the inoculated chocolate, the entire surface of the chocolate was swabbed five times with a sterile cotton swab premoistened with 1× phosphate-buffered saline (PBS; pH 7.2). The swab was then immersed in lysis buffer (AVL, Qiagen, Valencia, CA), and viral RNA was extracted immediately using the QIAmp viral RNA mini kit (Qiagen) according to the manufacturer's instructions.

To recover viruses from pistachios, the 25 g of inoculated pistachios was added to mesh filter bags (11216-904, VWR, Radnor, PA), and 40 mL of 100 mM Tris, 50 mM glycine, and 1% (w/v) beef extract (TGBE) buffer (pH 9.5) was added to each sample. The sample was mixed and incubated at room temperature on a rocking plate for 20 min. The eluate was transferred to a 50-mL centrifuge tube and centrifuged at 10,000 × g for 30 min at 4°C. The supernatant was decanted into a new tube, and the pH was adjusted to 7 ± 0.5 with HCl. A 5× solution of polyethylene glycol (PEG) 6000 (500 g/L; MilliporeSigma) and NaCl (1.5 mol/L; ThermoFisher) of quarter volumes of the weight of each sample was added to each tube, which was then incubated on ice on a rocking plate for 1 h. Samples were centrifuged at 10,000 × g for 30 min at 4°C. The supernatant was discarded, and the pellet was resuspended in 500 μL of PBS and stored at −80°C. Viral RNA was extracted using with the QIAmp viral RNA mini kit according to the manufacturer's instructions.

To recover viruses from corn flakes, the 25 g of inoculated corn flakes was added to mesh filter bags, and 175 mL of TGBE buffer was added to each sample. The sample was mixed and incubated at room temperature on a rocking plate for 20 min. The eluate was transferred to a 50-mL centrifuge tube and centrifuged at 10,000 × g for 30 min at 4°C. The supernatant was decanted into a new tube, and the pH was adjusted to 7 ± 0.5 with HCl. Samples were frozen at −80°C overnight and thawed the next day, and 5× PEG plus NaCl of quarter volumes of the weight of each sample was added to each tube, which was incubated on ice on a rocking plate for 1 h. Samples were centrifuged at 10,000 × g for 30 min at 4°C. The supernatant was decanted and discarded, and the pellet was resuspended in 1,000 to 2,000 μL of PBS, depending on the size of the pellet. The viral RNA was extracted with the QIAmp viral RNA mini kit according to the manufacturer's instructions. RNA was stored at −80°C.

A 1:10 bead-to-sample ratio was used for all the bead-based extractions (100 μL of beads for 1 mL of eluate). PGM-MBs were prepared as described previously (15, 51). To recover FCV or MNV from inoculated chocolate using the bead-based method, the surface of the chocolate was swabbed with a premoistened cotton swab five times, and the virus was released in 800 μL of PBS and extracted using 80 μL of PGM-MBs.

To recover FCV or MNV from inoculated pistachios using the PGM-MB method, 25 g of virus-inoculated pistachios was placed in mesh filter bags with 40 mL of PBS buffer. The samples were incubated at room temperature on a rocking plate for 20 min. The eluate was transferred to a 50-mL centrifuge tube, and 1 mL of eluate was extracted using 100 μL of PGM-MBs.

To recover FCV or MNV from inoculated corn flakes using the PGM-MB method, 25 g of virus-inoculated corn flakes was placed in mesh filter bags with 175 mL of PBS. The samples were incubated at room temperature on a rocking plate for 20 min. The eluate was transferred to a 50-mL centrifuge tube, and 1 mL of eluate was extracted using 100 μL of PGM-MBs.

Samples were incubated with PGM-MBs for 30 min on a rotary shaker. The beads were washed with PBS three times and resuspended in 50 μL of ultrapure water, heated at 100°C for 10 min, and then quickly chilled on ice. RNA was stored at −80°C.

### HAV antibody coupling to epoxy Dynabeads

HAV antibody–coupled beads were prepared using the magnetic Dynabeads antibody coupling kit (14311D, ThermoFisher) according to the manufacturer's instructions. Dynabeads M-270 Epoxy (10 mg) was added to 1 mL of C1 solution (ThermoFisher). Samples were incubated on a rotary shaker for 5 min. In a separate tube, 450 μL of C1 was combined with 50 μL of HAV antibody (DPA0224, Creative Diagnostics, Shirley, NY) and added to the beads. A 500-μL volume of C2 solution (ThermoFisher) was added and incubated on a roller at 37°C for 16 to 24 h. The beads were washed once with HB and LB solutions and washed twice with SB solution (all from ThermoFisher). The final mixture was incubated on a roller for 15 min at room temperature. The supernatant was removed, and the beads were resuspended in 1 mL of SB and stored at 4°C.

### Virus recovery using anti-HAV antibody–conjugated beads

To recover HAV from inoculated chocolate using the bead-based method, the surface of the chocolate was swabbed with a premoistened cotton swab five times, and the virus was released in 800 μL of PBS and extracted using 80 μL of anti-HAV antibody–conjugated beads.

To recover HAV from pistachios using the bead method, 25 g of virus-inoculated pistachios was placed in mesh filter bags with 40 mL of PBS. The samples were incubated at room temperature on a rocking plate for 20 min. The eluate was transferred to a 50-mL centrifuge tube, and 1 mL of eluate was extracted using 100 μL of HAV antibody–conjugated beads.

To recover HAV from corn flakes using the bead method, 25 g of virus-inoculated corn flakes was placed in mesh filter bags with 175 mL of PBS. The samples were incubated at room temperature on a rocking plate for 20 min. The eluate was transferred to a 50-mL centrifuge tube, and 1 mL of eluate was extracted using 100 μL of HAV antibody–conjugated beads.

The anti-HAV antibody–conjugated beads were washed with PBS plus 0.1% bovine serum albumin (A2153, MilliporeSigma). Eluates containing viruses were added to the beads and incubated at room temperature for 1 h on a rotary shaker. The beads were washed with PBS four times, resuspended in 50 μL of ultrapure water, heated at 100°C for 10 min, and then quickly chilled on ice. RNA was stored at −80°C.

### ddRT-PCR

FCV, MNV, and HAV RNA recovered by the ISO 15216-1:2017 and PGM-MB or HAV antibody methods was quantified using the One-Step ddRT-PCR advanced kit for probes (Bio-Rad, Hercules, CA) according to the manufacturer's instructions and as described previously (39, 44). Primers and probes used to quantify FCV, MNV, and HAV are listed in Table 1.

TABLE 1

List of primers and probes used in this study

The probes for the detection and quantification of the viruses in this study were made with the Iowa Black quencher (Integrated DNA Technologies, Coralville, IA) because this quencher produced the least background. The FAM reporter dye was used for probes to detect FCV and MNV, and the HEX reporter dye was used for HAV. All probes were made with an internal ZEN quencher to reduce background noise (Integrated DNA Technologies).

The thermocycling conditions were 50°C for 60 min, 95°C for 10 min, and 40 cycles of 95°C for 30 s (ramp = 2°C/s) with an annealing temperature of 55°C for FCV and 53°C for MNV and HAV for 1 min (ramp = 2°C/s), and 98°C for 10 min. The ddRT-PCR results were analyzed using the QX200 Droplet Digital system (Bio-Rad).

### Recovery rate calculation

The recovery rates for both methods were calculated by comparison of the absolute genome copy number of the virus recovered from the LMFs with the absolute genome copy number of the virus aliquot used to inoculate the LMFs:
$$\def\upalpha{\unicode[Times]{x3B1}}$$$$\def\upbeta{\unicode[Times]{x3B2}}$$$$\def\upgamma{\unicode[Times]{x3B3}}$$$$\def\updelta{\unicode[Times]{x3B4}}$$$$\def\upvarepsilon{\unicode[Times]{x3B5}}$$$$\def\upzeta{\unicode[Times]{x3B6}}$$$$\def\upeta{\unicode[Times]{x3B7}}$$$$\def\uptheta{\unicode[Times]{x3B8}}$$$$\def\upiota{\unicode[Times]{x3B9}}$$$$\def\upkappa{\unicode[Times]{x3BA}}$$$$\def\uplambda{\unicode[Times]{x3BB}}$$$$\def\upmu{\unicode[Times]{x3BC}}$$$$\def\upnu{\unicode[Times]{x3BD}}$$$$\def\upxi{\unicode[Times]{x3BE}}$$$$\def\upomicron{\unicode[Times]{x3BF}}$$$$\def\uppi{\unicode[Times]{x3C0}}$$$$\def\uprho{\unicode[Times]{x3C1}}$$$$\def\upsigma{\unicode[Times]{x3C3}}$$$$\def\uptau{\unicode[Times]{x3C4}}$$$$\def\upupsilon{\unicode[Times]{x3C5}}$$$$\def\upphi{\unicode[Times]{x3C6}}$$$$\def\upchi{\unicode[Times]{x3C7}}$$$$\def\uppsy{\unicode[Times]{x3C8}}$$$$\def\upomega{\unicode[Times]{x3C9}}$$$$\def\bialpha{\boldsymbol{\alpha}}$$$$\def\bibeta{\boldsymbol{\beta}}$$$$\def\bigamma{\boldsymbol{\gamma}}$$$$\def\bidelta{\boldsymbol{\delta}}$$$$\def\bivarepsilon{\boldsymbol{\varepsilon}}$$$$\def\bizeta{\boldsymbol{\zeta}}$$$$\def\bieta{\boldsymbol{\eta}}$$$$\def\bitheta{\boldsymbol{\theta}}$$$$\def\biiota{\\boldsymbol{\iota}}$$$$\def\bikappa{\boldsymbol{\kappa}}$$$$\def\bilambda{\boldsymbol{\lambda}}$$$$\def\\bimu{\boldsymbol{\mu}}$$$$\def\binu{\boldsymbol{\nu}}$$$$\def\bixi{\boldsymbol{\xi}}$$$$\def\biomicron{\boldsymbol{\micron}}$$$$\def\bipi{\boldsymbol{\pi}}$$$$\def\birho{\boldsymbol{\rho}}$$$$\def\bisigma{\boldsymbol{\sigma}}$$$$\def\bitau{\boldsymbol{\\tau}}$$$$\def\biupsilon{\boldsymbol{\upsilon}}$$$$\def\biphi{\boldsymbol{\phi}}$$$$\def\bichi{\boldsymbol{\chi}}$$$$\def\bipsy{\boldsymbol{\psy}}$$$$\def\biomega{\boldsymbol{\omega}}$$$$\def\bupalpha{\bf{\alpha}}$$$$\def\bupbeta{\bf{\beta}}$$$$\def\bupgamma{\bf{\gamma}}$$$$\def\bupdelta{\bf{\delta}}$$$$\def\bupvarepsilon{\bf{\varepsilon}}$$$$\def\bupzeta{\bf{\zeta}}$$$$\def\bupeta{\bf{\eta}}$$$$\def\buptheta{\bf{\theta}}$$$$\def\bupiota{\bf{\iota}}$$$$\def\bupkappa{\bf{\kappa}}$$$$\def\\buplambda{\bf{\lambda}}$$$$\def\bupmu{\bf{\mu}}$$$$\def\bupnu{\bf{\nu}}$$$$\def\bupxi{\bf{\xi}}$$$$\def\bupomicron{\bf{\micron}}$$$$\def\buppi{\bf{\pi}}$$$$\def\buprho{\bf{\rho}}$$$$\def\bupsigma{\bf{\sigma}}$$$$\def\buptau{\bf{\tau}}$$$$\def\bupupsilon{\bf{\upsilon}}$$$$\def\bupphi{\bf{\phi}}$$$$\def\bupchi{\bf{\chi}}$$$$\def\buppsy{\bf{\psy}}$$$$\def\bupomega{\bf{\omega}}$$$$\def\bGamma{\bf{\Gamma}}$$$$\def\bDelta{\bf{\Delta}}$$$$\def\bTheta{\bf{\Theta}}$$$$\def\bLambda{\bf{\Lambda}}$$$$\def\bXi{\bf{\Xi}}$$$$\def\bPi{\bf{\Pi}}$$$$\def\bSigma{\bf{\Sigma}}$$$$\def\bPhi{\bf{\Phi}}$$$$\def\bPsi{\bf{\Psi}}$$$$\def\bOmega{\bf{\Omega}}$$$${\rm{Recovery\ rate}\ (\%) = {{{\rm{obtained\ genome\ copy\ number}}} \over {{\rm{inoculated\ genome\ copy\ number}}}} \times 100}$$

### Statistical analysis

Statistical analysis was performed with Excel 2016 (Microsoft, Redmond, WA). The significance of differences in the recovery rates obtained by both methods was determined using a paired t test. The P values are provided in Table 2.

TABLE 2

Recovery ratio for the bead-based method compared with the ISO 15216-1:2017 method and P values for the higher recovery rates obtained by the bead-based assays

## RESULTS AND DISCUSSION

### Developing and implementing bead-based assays for isolating foodborne viruses

In the absence of a robust and readily available cell culture system for the human NoV, we chose to test the efficiency of a bead-based method for extraction of the cultivable surrogates of human NoV. This new method will allow future survival and inactivation studies on LMFs to be harnessed to infectivity assays. We extended the PGM-MB method to capture FCV and MNV from corn flakes, pistachios, and chocolate using the protocol that has been described previously (51).

Although no foodborne viral outbreaks have been linked to the tested LMFs, multiple outbreaks caused by other foodborne pathogens have been associated with breakfast cereals, chocolate, and pistachios, and massive recalls have been issued (6, 7, 9, 5658). Because foodborne viral illnesses and outbreaks are overwhelmingly underreported (11, 41), contamination of the tested LMFs with enteric viruses is likely to occur, and the methods utilized in this study can be extended to other LMFs.

Capsid amino acid sequences are highly conserved among strains of HAV; to date, only one serotype for this virus has been identified (49). Thus, affinity concentration of HAV can be an efficient approach for isolation of HAV from various food commodities. Immunomagnetic separation has successfully been used to concentrate HAV particles from green onions (60). In the present study, we used a different approach for the conjugation of anti-HAV antibody to epoxy Dynabeads for capturing HAV from artificially contaminated LMFs (Fig. S1) because these beads have less nonspecific binding (data not shown). The viral RNA genome was released by on-bead heat shock treatment of the captured virus. The extracted RNA was then subjected to quantification by ddRT-PCR assay, and the recovery rate for each virus, commodity, and method was calculated by comparing the recovered genomes to the viral load used to inoculate the sample. Overall, viral isolation using the bead-based assays takes approximately 90 min (Fig. 1).

### Evaluation of the efficiency of the bead-based methods

The efficiency of the ISO 15216-1:2017 and the bead-based methods for detection and quantification of viral RNA in select LMFs was compared in parallel. In replicated samples over two independent experiments, viral RNA was extracted either by heat denaturation of magnetic-bead–captured virus or by PEG precipitation of virus particles followed by RNA isolation according to the ISO 15216-1:2017 protocol (Fig. 1). The extracted RNA from both methods was quantified by ddRT-PCR assay.

The recovery rates obtained with the ISO 15216-1:2017 and the bead-based method in chocolate, pistachios, and corn flakes for FCV, MNV, and HAV are compared in Figure 2A, 2B, and 2C, respectively. The recovery ratios between the bead-based assays and the ISO 15216-1:2017 method are listed in Table 2. Although virus recovery with the bead-based method was significantly higher than that with the ISO 15216-1:2017 method for all three viruses in chocolate and pistachios, the bead-based method failed to recover FCV and MNV from corn flakes (Figs. 2A and 3B and Table 2). However, in corn flakes HAV recovery with the bead-based method was significantly higher than that with the ISO 15216-1:2017 method (Fig. 2C and Table 2). The highest recovery rates with the bead-based methods were for FCV and MNV in pistachios and HAV in corn flakes, at 38.4% ± 9.3%, 46.7% ± 0.5%, and 36.8% ± 7.5%, respectively (Fig. 2). The highest recoveries with the ISO 15216-1:2017 method were obtained for FCV in corn flakes and MNV in chocolate and corn flakes, at 24.05% ± 7.1%, 21.6% ± 7.9%, and 8.5% ± 3.1%, respectively. The lowest recoveries were obtained for HAV in chocolate (0.8% ± 0.2%) and pistachios (1.6% ± 0.7%) with the ISO 15216-1:2017 methods, and collectively the recovery rates for HAV were lower than those for FCV and MNV (Fig. 2).

FIGURE 2

The mean percent extraction efficiency (obtained genome copy number ÷ inoculated genome copy number × 100) obtained from FCV (A), MNV (B), and HAV (C). Error bars represent the standard deviations. * P < 0.05; *** P < 0.001.

FIGURE 2

The mean percent extraction efficiency (obtained genome copy number ÷ inoculated genome copy number × 100) obtained from FCV (A), MNV (B), and HAV (C). Error bars represent the standard deviations. * P < 0.05; *** P < 0.001.

We used the ddRT-PCR assay for quantification of the viral load recovered from LMFs because this approach is more precise for quantification and more tolerant to matrix-associated RT-PCR inhibitors (20, 43, 51). To examine the level of inhibition for each tested matrix, 1:10 and 1:100 dilutions of the extracted RNA were tested with the ddRT-PCR assay (51). The quantification results were similar to those for the undiluted samples (data not shown); thus, we concluded that the level of inhibition in the tested matrices was negligible.

The difference in efficacy of the bead-based methods for recovery of virus from corn flakes can be explained by the difference in the capturing mechanisms for these viruses. PGM is a heavily glycosylated glycoprotein (13), and NoV binding to carbohydrates has been comprehensively studied (18, 19, 30, 38). The high carbohydrate concentration of corn flakes may have outcompeted the PGM-coated beads for recovery of FCV and MNV. Histo-blood group antigens and sialic acid–like components in certain produce also compete with PGM binding (19, 51). This hypothesis is supported by the fact that anti-HAV antibody–coated beads, which do not rely on carbohydrate interaction for capturing HAV, efficiently recovered this virus from corn flakes (Figs. 2 and S1). Other than FCV and MNV recovery from corn flakes, the bead-based assays were more efficient, with recovery rates 3- to 21-fold higher than those of the ISO 15216-1:2017 method (Table 2). Therefore, the PGM-MB assay could potentially be used to extract NoV and its surrogates in LMFs with low carbohydrate components, whereas the carbohydrate component does not make a significant difference in extraction of HAV by the bead-based assay.

We did not compare the limit of detection and limit of quantification of the two methods because these limits often vary depending on the tested matrix (19, 24). Thus, we cannot comment on the sensitivity of the compared methods. However, because the bead-based assays have higher recovery rates and fewer steps, which decreases the risk of virus loss, we anticipate that these assay also have higher sensitivity. This hypothesis should be tested by recovering low levels of virus from other food commodities.

One disadvantage of virus recovery using the bead-based assay is that because of the absence of a virus concentration step only a portion of the sample is subjected to bead extraction, whereas the PEG precipitation step in the ISO 15216-1:2017 method allows for analysis of the viral load in the whole sample.

In conclusion, comparison of recovery methods is a critical first step to improving the detection of foodborne viruses and identifying the most efficient methods for isolation of these viruses from various food commodities. In this study, for the first time the efficiency of common virus extraction methods was examined with LMFs. The results indicate that the extraction efficiency of the PGM-MB method for recovery of NoV surrogates depends on the composition of the commodity.

Although virus extraction with MB-based methods relies on the integrity of the viral capsid, the infectivity of the captured viruses can be determined only with infectivity assays because many factors are involved in viral replication and infection. To date, no method has been developed that allows for assessing the infectivity of the viral particles captured by MBs. Therefore, some noninfectious viruses may still bind to PGM or anti-HAV antibody and be further quantified with the ddRT-PCR assay. However, bead-based isolation has fewer steps, which makes these assays competitive for rapid and efficient virus detection and quantification in many food commodities. Thus, the coated MB assay can be used for surveillance and evaluation of long-term survival of and potential inactivation strategies for foodborne viruses in LMFs.

## ACKNOWLEDGMENTS

This study was supported in part by a grant from International Life Sciences Institute (ILSI) North America and the Microbiology Research Division of the Bureau of Microbial Hazards, Health Canada. ILSI North America is a public, nonprofit science foundation that provides a forum for advancing understanding of scientific issues related to the nutritional quality and safety of the food supply. ILSI North America receives support primarily from its industry membership. ILSI North America had no role in the design, analysis, interpretation, or presentation of the data and results. We thank Dr. Jeffrey Farber (University of Guelph, Guelph, Ontario, Canada) for his insightful comments throughout the project.

## SUPPLEMENTAL MATERIAL

Supplemental material associated with this article can be found online at: https://doi.org/10.4315/0362-028X.JFP-19-345.s1

## REFERENCES

REFERENCES
1.
Anonymous
.
2019
.
Dates from Iran linked to hepatitis A outbreak for second time in 2 years
.
Food Saf. News
.
2.
Atmar,
R. L.,
Opekun
A. R.,
Gilger
M. A.,
Estes
M. K.,
Crawford
S. E.,
Neill
F. H.,
Ramani
S.,
Hill
H.,
Ferreira
J.,
and
Graham
D. Y.
2014
.
Determination of the 50% human infectious dose for Norwalk virus
.
J. Infect. Dis
.
209
:
1016
1022
.
3.
Bae,
J.,
and
Schwab
K. J.
2008
.
Evaluation of murine norovirus, feline calicivirus, poliovirus, and MS2 as surrogates for human norovirus in a model of viral persistence in surface water and groundwater
.
Appl. Environ. Microbiol
.
74
:
477
484
.
4.
Banyai,
K.,
Estes
M. K.,
Martella
V.,
and
Parashar
U. D.
2018
.
Viral gastroenteritis
.
Lancet
392
:
175
186
.
5.
Bidawid,
S.,
Malik
N.,
O.,
Sattar
S. A.,
and
Farber
J. M.
2003
.
A feline kidney cell line–based plaque assay for feline calicivirus, a surrogate for Norwalk virus
.
J. Virol. Methods
107
:
163
167
.
6.
Brar,
P. K.,
and
Danyluk
M. D.
2018
.
Nuts and grains: microbiology and preharvest contamination risks
.
Microbiol. Spectr.
6
(2)
.
7.
.
2019
.
Food recall warning—Wonderful brand pistachios recalled due to Salmonella
.
8.
Carvalho,
C.,
Thomas
H.,
Balogun
K.,
Tedder
R.,
Pebody
R.,
Ramsay
M.,
and
Ngui
S.
2012
.
A possible outbreak of hepatitis A associated with semi-dried tomatoes, England, July–November 2011
.
Euro Surveill
.
17
(6)
:
pii=20083.
9.
Centers for Disease Control and Prevention
.
2018
.
Multistate outbreak of Salmonella Mbandaka infections linked to Kellogg's Honey Smacks Cereal (final update)
.
Available at: https://www.cdc.gov/salmonella/mbandaka-06-18/index.html. Accessed 16 October 2019.
10.
Centers for Disease Control and Prevention
.
2018
.
Surveillance for foodborne disease outbreaks, United States, 2016, annual report
.
Available at: https://www.cdc.gov/fdoss/pdf/2016_FoodBorneOutbreaks_508.pdf. Accessed 6 November 2019.
11.
Chatziprodromidou,
I. P.,
Bellou
M.,
Vantarakis
G.,
and
Vantarakis
A.
2018
.
Viral outbreaks linked to fresh produce consumption: a systematic review
.
J. Appl. Microbiol
.
124
:
932
942
.
12.
Cook,
N.,
Bertrand
I.,
Gantzer
C.,
Pinto
R. M.,
and
Bosch
A.
2018
.
Persistence of hepatitis A virus in fresh produce and production environments, and the effect of disinfection procedures: a review
.
Food Environ. Virol
.
10
:
253
262
.
13.
Corfield,
A. P.,
Carroll
D.,
Myerscough
N.,
and
Probert
C. S.
2001
.
Mucins in the gastrointestinal tract in health and disease
.
Front. Biosci
.
6
:
d1321
d1357
.
14.
Cox,
C.,
Cao
S.,
and
Lu
Y.
2009
.
Enhanced detection and study of murine norovirus-1 using a more efficient microglial cell line
.
Virol. J.
6
:
196
.
15.
Dancho,
B. A.,
Chen
H.,
and
Kingsley
D. H.
2012
.
Discrimination between infectious and non-infectious human norovirus using porcine gastric mucin
.
Int. J. Food Microbiol
.
155
:
222
226
.
16.
Dewey-Mattia,
D.,
Manikonda
K.,
Hall
A. J.,
Wise
M. E.,
and
Crowe
S. J.
2018
.
Surveillance for foodborne disease outbreaks—United States, 2009–2015
.
Morb. Mortal. Wkly. Rep. Surveill. Summ
.
67
(10)
:
1
11
.
17.
Donnan,
E. J.,
Fielding
J. E.,
Gregory
J. E.,
Lalor
K.,
Rowe
S.,
Goldsmith
P.,
Antoniou
M.,
Fullerton
K. E.,
Knope
K.,
Copland
J. G.,
Bowden
D. S.,
Tracy
S. L.,
Hogg
G. G.,
Tan
A.,
J.,
Gaston
J.,
and
Vally
H.
2012
.
A multistate outbreak of hepatitis A associated with semidried tomatoes in Australia, 2009
.
Clin. Infect. Dis
.
54
:
775
781
.
18.
Esseili,
M. A.,
Saif
L. J.,
Farkas
T.,
and
Wang
Q.
2015
.
Feline calicivirus, murine norovirus, porcine sapovirus, and Tulane virus survival on postharvest lettuce
.
Appl. Environ. Microbiol
.
81
:
5085
5092
.
19.
Esseili,
M. A.,
Wang
Q.,
and
Saif
L. J.
2012
.
Binding of human GII.4 norovirus virus-like particles to carbohydrates of romaine lettuce leaf cell wall materials
.
Appl. Environ. Microbiol
.
78
:
786
794
.
20.
Fraisse,
A.,
Coudray-Meunier
C.,
Martin-Latil
S.,
Hennechart-Collette
C.,
Delannoy
S.,
Fach
P.,
and
Perelle
S.
2017
.
Digital RT-PCR method for hepatitis A virus and norovirus quantification in soft berries
.
Int. J. Food Microbiol
.
243
:
36
45
.
21.
Gassowski,
M.,
Michaelis
K.,
Wenzel
J. J.,
Faber
M.,
Figoni
J.,
Mouna
L.,
Friesema
I. H.,
Vennema
H.,
Avellon
A.,
Varela
C.,
Sundqvist
L.,
Lundberg Ederth
J.,
Plunkett
J.,
Balogun
K.,
Ngui
S. L.,
Midgley
S. E.,
Gillesberg Lassen
S.,
and
Muller
L.
2018
.
Two concurrent outbreaks of hepatitis A highlight the risk of infection for non-immune travellers to Morocco, January to June 2018
.
Euro Surveill
.
23
(27)
:
1800329.
22.
Guevremont,
E.,
Brassard
J.,
Houde
A.,
Simard
C.,
and
Trottier
Y. L.
2006
.
Development of an extraction and concentration procedure and comparison of RT-PCR primer systems for the detection of hepatitis A virus and norovirus GII in green onions
.
J. Virol. Methods
134
:
130
135
.
23.
Guo,
Z.,
Huang
J.,
Shi
G.,
Su
C. H.,
and
Niu
J. J.
2014
.
A food-borne outbreak of gastroenteritis caused by norovirus GII in a university located in Xiamen City, China
.
Int. J. Infect. Dis
.
28
:
101
106
.
24.
International Organization for Standardization
.
2017
.
Microbiology of the food chain—horizontal method for determination of hepatitis A virus and norovirus using real-time RT-PCR. Part 1. Method for quantification
.
ISO 15216-1:2017
.
International Organization for Standardization
,
Geneva
.
25.
Knight,
A.,
Haines
J.,
Stals
A.,
Li
D.,
Uyttendaele
M.,
Knight
A.,
and
Jaykus
L. A.
2016
.
A systematic review of human norovirus survival reveals a greater persistence of human norovirus RT-qPCR signals compared to those of cultivable surrogate viruses
.
Int. J. Food Microbiol
.
216
:
40
49
.
26.
Lemon,
S. M.,
and
Walker
C. M.
2018
.
Hepatitis A virus and hepatitis E virus: emerging and re-emerging enterically transmitted hepatitis viruses
.
Cold Spring Harb. Perspect. Med.
9
(6)
:
pii=a031823.
27.
Le Pendu,
J.,
Nystrom
K.,
and
Ruvoen-Clouet
N.
2014
.
Host-pathogen co-evolution and glycan interactions
.
Curr. Opin. Virol
.
7
:
88
94
.
28.
Li,
X.,
and
Chen
H.
2015
.
Evaluation of the porcine gastric mucin binding assay for high-pressure-inactivation studies using murine norovirus and Tulane virus
.
Appl. Environ. Microbiol
.
81
:
515
521
.
29.
Lopman,
B.,
P.,
Park
G. W.,
Hall
A. J.,
Parashar
U. D.,
and
Vinje
J.
2012
.
Environmental transmission of norovirus gastroenteritis
.
Curr. Opin. Virol
.
2
:
96
102
.
30.
Mallagaray,
A.,
Creutznacher
R.,
Dulfer
J.,
Mayer
P. H. O.,
Grimm
L. L.,
Orduna
J. M.,
Trabjerg
E.,
Stehle
T.,
Rand
K. D.,
Blaum
B. S.,
Uetrecht
C.,
and
Peters
T.
2019
.
A post-translational modification of human norovirus capsid protein attenuates glycan binding
.
Nat. Commun.
10
.
31.
Manuel,
C. S.,
Moore
M. D.,
and
Jaykus
L. A.
2018
.
Predicting human norovirus infectivity—recent advances and continued challenges
.
Food Microbiol
.
76
:
337
345
.
32.
Mattison,
K.,
Brassard
J.,
Gagne
M. J.,
Ward
P.,
Houde
A.,
Lessard
L.,
Simard
C.,
Shukla
A.,
Pagotto
F.,
Jones
T. H.,
and
Trottier
Y. L.
2009
.
The feline calicivirus as a sample process control for the detection of food and waterborne RNA viruses
.
Int. J. Food Microbiol
.
132
:
73
77
.
33.
Mbithi,
J. N.,
Springthorpe
V. S.,
and
Sattar
S. A.
1991
.
Effect of relative humidity and air temperature on survival of hepatitis A virus on environmental surfaces
.
Appl. Environ. Microbiol
.
57
:
1394
1399
.
34.
Montazeri,
N.,
Manuel
C.,
Moorman
E.,
J. R.,
Williams
L. L.,
and
Jaykus
L. A.
2017
.
Virucidal activity of fogged chlorine dioxide– and hydrogen peroxide–based disinfectants against human norovirus and its surrogate, feline calicivirus, on hard-to-reach surfaces
.
Front. Microbiol
.
8
:
1031
.
35.
Moore,
M. D.,
Bobay
B. G.,
Mertens
B.,
and
Jaykus
L. A.
2016
.
Human norovirus aptamer exhibits high degree of target conformation-dependent binding similar to that of receptors and discriminates particle functionality
.
mSphere
1
(6)
:
pii=e00298-16.
36.
Moore,
M. D.,
Goulter
R. M.,
and
Jaykus
L. A.
2015
.
Human norovirus as a foodborne pathogen: challenges and developments
.
Annu. Rev. Food Sci. Technol
.
6
:
411
433
.
37.
Mormann,
S.,
Heissenberg
C.,
Pfannebecker
J.,
and
Becker
B.
2015
.
Tenacity of human norovirus and the surrogates feline calicivirus and murine norovirus during long-term storage on common nonporous food contact surfaces
.
J. Food Prot
.
78
:
224
229
.
38.
Morozov,
V.,
Borkowski
J.,
and
Hanisch
F. G.
2018
.
The double face of mucin-type O-glycans in lectin-mediated infection and immunity
.
Molecules
23
(5)
:
pii=E1151.
39.
Mykytczuk,
O.,
Harlow
J.,
Bidawid
S.,
Corneau
N.,
and
Nasheri
N.
2017
.
Prevalence and molecular characterization of the hepatitis E virus in retail pork products marketed in Canada
.
Food Environ. Virol
.
9
:
208
218
.
40.
Nasheri,
N.,
Petronella
N.,
Ronholm
J.,
Bidawid
S.,
and
Corneau
N.
2017
.
Characterization of the genomic diversity of norovirus in linked patients using a metagenomic deep sequencing approach
.
Front. Microbiol
.
8
:
73
.
41.
Nasheri,
N.,
Vester
A.,
and
Petronella
N.
2019
.
Foodborne viral outbreaks associated with frozen produce
.
Epidemiol. Infect
.
147
:
e291
.
42.
Park,
J. H.,
Jeong
H. S.,
Lee
J. S.,
Lee
S. W.,
Choi
Y. H.,
Choi
S. J.,
Joo
I. S.,
Kim
Y. R.,
Park
Y. K.,
and
Youn
S. K.
2015
.
First norovirus outbreaks associated with consumption of green seaweed (Enteromorpha spp.) in South Korea
.
Epidemiol. Infect
.
143
:
515
521
.
43.
S.,
Eriksson
R.,
Lowther
J.,
Ellstrom
P.,
and
Simonsson
M.
2018
.
Comparison between RT droplet digital PCR and RT real-time PCR for quantification of noroviruses in oysters
.
Int. J. Food Microbiol
.
284
:
73
83
.
44.
Petronella,
N.,
Ronholm
J.,
Suresh
M.,
Harlow
J.,
Mykytczuk
O.,
Corneau
N.,
Bidawid
S.,
and
Nasheri
N.
2018
.
Genetic characterization of norovirus GII.4 variants circulating in Canada using a metagenomic technique
.
BMC Infect. Dis
.
18
(1)
:
521
.
45.
Purcell,
R. H.,
Wong
D. C.,
and
Shapiro
M.
2002
.
Relative infectivity of hepatitis A virus by the oral and intravenous routes in 2 species of nonhuman primates
.
J. Infect. Dis
.
185
:
1668
1671
.
46.
Racki,
N.,
Morisset
D.,
Gutierrez-Aguirre
I.,
and
Ravnikar
M.
2014
.
One-step RT-droplet digital PCR: a breakthrough in the quantification of waterborne RNA viruses
.
Anal. Bioanal. Chem
.
406
:
661
667
.
47.
Rzezutka,
A.,
and
Cook
N.
2004
.
Survival of human enteric viruses in the environment and food
.
FEMS Microbiol. Rev
.
28
:
441
453
.
48.
Sakon,
N.,
K.,
Shinkai
T.,
Hamajima
Y.,
Yoshitomi
H.,
Matsushima
Y.,
R.,
Terasoma
F.,
Nakamura
A.,
Komano
J.,
Nagasawa
K.,
Shimizu
H.,
Katayama
K.,
and
Kimura
H.
2018
.
Foodborne outbreaks caused by human norovirus GII.P17-GII.17–contaminated nori, Japan, 2017
.
Emerg. Infect. Dis
.
24
:
920
923
.
49.
Sanchez,
G.,
Bosch
A.,
and
Pinto
R. M.
2003
.
Genome variability and capsid structural constraints of hepatitis A virus
.
J. Virol
.
77
:
452
459
.
50.
A. F.,
Lee
A.,
and
Farber
J. M.
2018
.
Methods for the control of foodborne pathogens in low-moisture foods
.
Annu. Rev. Food Sci. Technol
.
9
:
177
208
.
51.
Suresh,
M.,
Harlow
J.,
and
Nasheri
N.
2019
.
Evaluation of porcine gastric mucin assay for detection and quantification of human norovirus in fresh herbs and leafy vegetables
.
Food Microbiol
.
84
:
103254
.
52.
Teunis,
P. F.,
Sukhrie
F. H.,
Vennema
H.,
Bogerman
J.,
Beersma
M. F.,
and
Koopmans
M. P.
2015
.
Shedding of norovirus in symptomatic and asymptomatic infections
.
Epidemiol. Infect
.
143
:
1710
1717
.
53.
Tian,
P.,
Engelbrektson
A.,
and
Mandrell
R.
2008
.
Two-log increase in sensitivity for detection of norovirus in complex samples by concentration with porcine gastric mucin conjugated to magnetic beads
.
Appl. Environ. Microbiol
.
74
:
4271
4276
.
54.
Tian,
P.,
Jiang
X.,
Zhong
W.,
Jensen
H. M.,
Brandl
M.,
Bates
A. H.,
Engelbrektson
A. L.,
and
Mandrell
R.
2007
.
Binding of recombinant norovirus like particle to histo-blood group antigen on cells in the lumen of pig duodenum
.
Res. Vet. Sci
.
83
:
410
418
.
55.
Tjon,
G. M.,
Coutinho
R. A.,
van den Hoek
A.,
Esman
S.,
Wijkmans
C. J.,
Hoebe
C. J.,
Wolters
B.,
Swaan
C.,
Geskus
R. B.,
Dukers
N.,
and
Bruisten
S. M.
2006
.
High and persistent excretion of hepatitis A virus in immunocompetent patients
.
J. Med. Virol
.
78
:
1398
1405
.
56.
.
2019
.
FDA investigated multistate outbreak of Salmonella infections linked to Wonderful Pistachio
.
57.
.
2019
.
Public health alert concerning a possible hepatitis A contamination of Bauer's Candies' Modjeskas
.
58.
Werber,
D.,
Dreesman
J.,
Feil
F.,
van Treeck
U.,
Fell
G.,
Ethelberg
S.,
Hauri
A. M.,
Roggentin
P.,
Prager
R.,
Fisher
I. S.,
Behnke
S. C.,
Bartelt
E.,
Weise
E.,
Ellis
A.,
Siitonen
A.,
Y.,
Tschape
H.,
Kramer
M. H.,
and
Ammon
A.
2005
.
International outbreak of Salmonella Oranienburg due to German chocolate
.
BMC Infect. Dis.
5
.
59.
Young,
I.,
L.,
Cahill
S.,
Kojima
M.,
Clarke
R.,
and
Rajic
A.
2015
.
Application of a rapid knowledge synthesis and transfer approach to assess the microbial safety of low-moisture foods
.
J. Food Prot
.
78
:
2264
2278
.
60.
Zheng,
Y.,
and
Hu
Y.
2017
.
Development of a fast and efficient method for hepatitis A virus concentration from green onion
.
J. Virol. Methods
249
:
161
164
.
This is an open access article