Abstract
Human noroviruses (HNoV) have been implicated in gastrointestinal outbreaks associated with fresh produce, juices, and ready-to-eat foods. In order to determine the risk of HNoV transmission by contaminated blueberry juice, survival characteristics of cultivable HNoV surrogates (murine norovirus, MNV-1; feline calicivirus, FCV-F9; and bacteriophage MS2) in blueberry juice (pH=2.77) after 0, 1, 2, 7, 14, and 21 days at refrigeration temperatures (4°C) were studied. High-pressure homogenization (HPH) was studied as a novel processing method for noroviral surrogate inactivation in blueberry juice. Blueberry juice or phosphate-buffered saline (PBS; pH 7.2 as control) was inoculated with each virus, stored over 21 days at 4°C or subjected to HPH, and plaque assayed. FCV-F9 (∼5 log10 PFU/mL) was undetectable after 1 day in blueberry juice at 4°C. MNV-1 (∼4 log10 PFU/ml) showed minimal reduction (1 log10 PFU/mL) after 14 days, with greater reduction (1.95 log10 PFU/mL; p<0.05) after 21 days in blueberry juice at 4°C. Bacteriophage MS2 (∼6 log10 PFU/mL) showed significant reduction (1.93 log10 PFU/mL; p<0.05) after 2 days and was undetectable after 7 days in blueberry juice at 4°C. FCV-F9 remained viable in PBS for up to 21 days (2.28 log10 PFU/mL reduction), while MNV-1 and MS2 survived after 21 days (1.08 and 0.56 log10 PFU/mL reduction, respectively). Intriguingly, FCV-F9 and bacteriophage MS2 showed reduction after minimal homogenization pressures in blueberry juice (pH=2.77), possibly due to the combination of juice pH, juice components, and mechanical effects. MNV-1 in blueberry juice was only slightly reduced at 250 (0.33 log10 PFU/mL) and 300 MPa (0.71 log10 PFU/mL). Virus surrogate survival in blueberry juice at 4°C correlates well with the ease of HNoV transmission via juices. HPH for viral inactivation in juices is dependent on virus type, and higher homogenization pressures may be needed for MNV-1 inactivation.
Introduction
I
It is well recognized that foodborne viruses are transmitted primarily via the fecal-oral route through contaminated water, food (that does not undergo further processing), infectious vomitus, and via infected individuals through person-to-person contact or person-to-fomite-to-person contact (Patel et al., 2009; Sair et al., 2002; Wadl et al., 2010). Studies have shown that fruit juices contain natural antimicrobials (Su et al., 2010a,b; Haidari et al., 2009; Howell et al., 1998), yet juice beverages, especially when served in self-service containers, are a common vehicle of foodborne illness. An outbreak in February 2009, caused by the consumption of juice from large self-service containers with a tap, sickened 38 persons on two Dutch airline coach trips who visited the same hotel in Germany (Visser et al., 2010). A major hepatitis A outbreak among tourists returning from Egypt in 2004 involved 351 infected individuals from nine European countries where orange juice was implicated as the most likely infection vehicle (Frank et al., 2007). One of the largest norovirus outbreaks occurred on domestic airline flights in 1991, involving 3000 individuals who had consumed orange juice served to passengers on these flights (Fleet et al., 2000). These outbreaks show the need to have better hygienic practices and control measures to prevent juice-related outbreaks.
Blueberries contain structurally related polyphenols, including proanthocyanidins (PAC) associated with health benefits and antimicrobial properties (Smith et al., 2000; Bomser et al., 1996; Howell et al., 1998; Ofek et al., 1991). Citrobacter freundii and Enterococcus faecalis were found to be sensitive to anthocyanin extracts of blueberry (Vaccinium corymbosum L.) fruits (Burdulis et al., 2009). Blueberry PAC was shown to inhibit the expression and replication of hepatitis C virus (Takeshita et al., 2009). However, the survival characteristics of human enteric viral surrogates in blueberry juice are currently unknown.
Human enteric viruses are quite resistant to some traditional thermal and chemical inactivation methods (Sair et al., 2002). High-pressure homogenization (HPH) is a novel inactivation process used for fluid foods (Taylor et al., 2007). D'Souza et al. (2011) reported that homogenization pressures of 300 MPa showed >4.95 log PFU/mL FCV-F9 reduction. MS2 and MNV-1 in phosphate-buffered saline (PBS) at homogenization pressures of 300 MPa showed ∼3 log PFU and ∼0.8 log PFU reduction, respectively (D'Souza et al., 2009). The objective of this study was to evaluate the survival of human enteric viral surrogates (FCV-F9, MNV-1, and MS2) in blueberry juice at refrigeration temperatures over 21 days (to simulate household storage conditions of juices). HPH as a novel alternative method for human enteric virus inactivation in blueberry juice was also investigated.
Materials and Methods
Viruses and cell lines
As described earlier, FCV-F9 and its host Crandell Reese Feline Kidney (CRFK) cells, and bacteriophage MS2 and its host E. coli B-15597 were obtained from American Type Culture Collection (ATCC; Manassas, VA), while MNV-1 was kindly provided as a gift by Dr. Skip Virgin of Washington University, St. Louis, Missouri (Horm and D'Souza, 2011; Su and D'Souza, 2011). MNV-1 host cells (RAW 264.7) were obtained from the University of Tennessee cell culture collection.
FCV-F9 stocks were prepared as previously described (D'Souza et al., 2009; D'Souza and Su, 2010; Su et al., 2010a). Briefly, FCV-F9 was infected onto confluent CRFK cells in 175-cm2 flasks and incubated at 37°C under 5% CO2 for 24 h until 90–100% cell lysis was observed (D'Souza and Su, 2010; Su et al., 2010a). This flask was freeze-thawed thrice and centrifuged at room temperature at 5000×g for 10 min. The supernatant was filtered through 0.2-μm filters, aliquoted, and stored at −80°C, and plaque assayed to determine the titer and used as viral stocks.
Similarly, MNV-1 stock was prepared by infecting confluent RAW 264.7 cells in 175-cm2 flasks, followed by incubation at 37°C under 5% CO2 for 4–6 days until 90–100% cell lysis was observed, and the procedures described above were followed for viral stock preparations (Su et al., 2010a).
Following previously described procedures, bacteriophage MS2 was propagated in its host E. coli B-15597 at 37°C for ∼18 h in trypticase soy broth (TSB) containing 0.1% glucose, 2 mM CaCl2, and 10 μg/mL thiamine (D'Souza et al., 2009). Viruses were recovered as described above for FCV-F9 and plaque assayed as described below.
FCV-F9 plaque assay
Using Dulbecco's modified Eagle's medium (DMEM; Lonza, Walkersville, MD) containing 2% fetal bovine serum (FBS) and 1% Gibco® antibiotic-antimycotic (Invitrogen™, Grand Island, NY), CRFK cells were cultivated to confluency and plaque assays were performed as described before (D'Souza and Su, 2010). Briefly, confluent CRFK cells were infected with 0.5 mL of 0–21-day stored FCV-F9 as well as HPH-treated or untreated FCV-F9 that were serially diluted in DMEM containing 2% FBS, incubated for 2 h at 37°C under 5% CO2, and overlaid with 2 mL DMEM containing 2% FBS, 0.75% agarose, and 1% penicillin-streptomycin (HyClone®, Logan, UT). After 48 h of incubation, cells were stained with 0.01% neutral red and the plaques were counted after incubation for 24 h at 37°C.
MNV-1 plaque assay
RAW 264.7 cells were cultivated under 5% CO2 at 37°C until 80–90% confluency and infected with 0.5 mL of 10-fold serial viral dilutions of the stored and HPH-treated viruses and untreated controls and plaque assayed as described earlier (Su et al., 2010a).
MS2 plaque assay
The double agar layer method of Bae and Schwab (2008) was used for the MS2 infectivity plaque assay with some modifications as described before (Bae and Schwab, 2008; D'Souza et al., 2009). Plaques were counted after incubation at 37°C overnight.
Survival Study
Survival of MNV-1 (4 log10 PFU/mL), FCV-F9 (5 log10 PFU/mL), and bacteriophage MS2 (6 log10 PFU/mL) was determined in blueberry juice (Oceanspray® Diet Blueberry Juice, Lakeville-Middleboro, MA) or PBS (pH=7.2, control; and pH=2.77 for comparison to the similar pH of blueberry juice). Two hundred microliters (0.2 mL) of each virus was added to 1.8 mL of juice or PBS, and stored for 0, 1, 2, 7, 14, and 21 days at 4°C. Each virus was serially diluted in DMEM containing 10% FBS and plaque assayed in duplicate, as described above, after each time point. Each experiment was replicated four times. At the dilution levels tested, blueberry juice did not adversely affect the cell cultures for use in the plaque assay. In addition, to determine the effects of juice components alone, the survival study of virus in PBS (at a pH similar to blueberry juice of 2.77 as mentioned above) over 21 days was also undertaken. Oxalic acid was added to PBS until a pH of 2.77 was attained, and each virus was added to the solution (0.2 mL of virus to 1.8 mL of PBS), serially diluted, and plaque assayed either immediately or over 21 day storage at 4°C.
HPH unit and treatments
One milliliter of each virus was added to 500 mL of PBS (control) or blueberry juice to obtain final titers of FCV-F9 at 4 log10 PFU/mL, MNV-1 at 4 log10 PFU/mL, or MS2 at 5 log10 PFU/mL, and subjected to homogenization pressures of 0–300 MPa. Each treatment (500 mL) was passed through the high-pressure homogenizer as described previously (D'Souza et al., 2009). Briefly, all processes were carried out in a FPG 12500 bench top high–pressure homogenizer (Stansted Fluid Power, Ltd., Essex, UK), with the homogenizer valve temperature held at 2°C under ice using a water bath. Pressure and temperature were recorded continuously during treatments (Lookout, v. 5.1; National Instruments, Austin, TX) at 0, 100, 200, 250, and 300 MPa with recorded exposure temperatures of 24°C, 46°C, 63°C, 70°C, and 75°C, respectively, for <2 s; 15 mL of sample were collected at each pressure. Each experiment was performed thrice using three separate runs on different days. These samples were immediately placed on ice, aliquoted, and stored at −80°C prior to diluting and assaying.
Statistical analysis
FCV-F9, MNV-1, and MS2 infectivity assay data were statistically analyzed using analysis of variance (ANOVA) and Tukey's test on a completely randomized design with the SAS system (Windows v. 9.2; Statistical Analysis Institute [SAS], Cary, NC). Data were analyzed for differences in survivability over time in PBS and blueberry juice with four duplicate treatments of each virus. Data were also analyzed for differences in pressure inactivation with blueberry juice for three replication treatments of each virus.
Results
Survival trends of FCV-F9, MNV-1, and MS2 in PBS (pH=7.2 and pH=2.77) and blueberry juice over 21 days at 4°C
FCV-F9 in PBS at pH 7.2 at refrigeration showed slight reduction in titer after 7 days (∼1 log10 PFU/mL) with reduction of 2.28 log10 PFU/mL after 21 days of storage (Table 1). MNV-1 was less affected by storage in PBS at 4°C and showed only a slight reduction of ∼l log10 PFU/mL after 21 days. Similarly, MS2 in PBS was reduced <1 log10 PFU/mL after 21 days at 4°C.
All experiments were carried out four times in duplicates. Different letters denote significant differences when compared within each column alone (p<0.05).
FCV-F9 at an initial count of ∼5 log10 PFU/mL in blueberry juice was completely reduced after 1 day at 4°C (Table 1). MNV-1 at ∼4 log10 PFU/mL showed minimal reduction (∼1 log10 PFU/mL) after 14 days in blueberry juice but significant reduction (1.95 log10 PFU/mL; p<0.05) after 21 days at 4°C. Bacteriophage MS2 at an initial count of ∼6 log10 PFU/mL showed significant reduction (1.93 log10 PFU/mL) after 2 days and complete reduction after 7 days in blueberry juice at 4°C (p<0.05). To determine the influence of the pH of blueberry juice on viral reduction, the viruses were exposed to pH 2.77 in PBS over the same time period. FCV-F9 in PBS at pH 2.77 showed significant reduction of 2.27 log10 PFU/mL after 2 days and complete reduction after 7 days at 4°C. MNV-1 showed only slight reduction (0.44 log10 PFU/mL) after 21 days, and MS2 demonstrated significant reduction after 7 and 14 days (2.74 and 3.37 log10 PFU/mL, respectively) and complete reduction (based on the assay detection limit and was not detectable by the plaque assays) after 21 days at 4°C (Table 1).
Effect of HPH on FCV-F9, MNV-1, and MS2 in blueberry juice
In PBS, the minimal homogenization pressures to inactivate >3 log10 FCV-F9, >1.9 log10 MNV-1, and >2.37 log10 MS2 were 250, 300, and 300 MPa, respectively (Table 2). In contrast, FCV-F9 and MS2 were almost completely inactivated (4–5 log10 reduction) at the lowest homogenization processing pressure of 100 MPa in blueberry juice. MNV-1 in blueberry juice was less susceptible to the HPH and was only slightly reduced at 300 MPa (0.71 log log10 PFU/mL) (Table 2).
All experiments were replicated three times. The recorded exposure temperatures were 24°C, 46°C, 63°C, 70°C, and 75°C at 0, 100, 200, 250, and 300 MPa, respectively. Different letters denote significant differences when compared within each column alone (p<0.05).
Discussion
While FCV-F9 and MNV-1 are used as surrogates for human NoV studies, Guan et al. (2006) suggested that MS2 was a suitable surrogate for HAV and also perhaps enteroviruses, due to its persistence in adverse environmental conditions (Guan et al., 2006). Studies indicate that the human noroviral incubation period is 1–2 days (McCarthy et al., 2000; Sair et al., 2002), but can be shed in the stool before symptoms occur and up to 3 weeks after symptoms have abated (Parashar et al., 1998; Koopman and Duizer, 2004; Barrabeig et al., 2010; Atmar et al., 2008). Foods commonly associated with viral contamination are shellfish, produce items, and ready-to-eat (RTE) foods (Cannon et al., 2006; Anderson et al., 2001; Berg et al., 2000; Ponka et al., 1999; Grove et al., 2008). Fruit juices are also of concern, since major outbreaks associated with these beverages have been documented (Frank et al., 2007).
Foodborne viruses have been shown to survive on various foods and surfaces for extended periods of time. FCV-F9 was found to be detectable on environmental surfaces up to 7 days at room temperature (D'Souza et al., 2006). On brass and computer keyboards at 25°C, FCV-F9 was found to be reduced by 90% after less than 4 h (Clay et al., 2006). A 2 log10 PFU/mL reduction of FCV-F9 was observed on lettuce after 7 days, with >2.5 log10 reduction on strawberries after 6 days at refrigeration (4°C) (Mattison et al., 2007). MS2 bacteriophage was shown to survive on several fresh produce items with minimal reduction (<1 log10 PFU/mL) after 50 days at 4°C and 8°C (Dawson et al., 2005). Studies on environmental persistence of human NoV surrogates should help in understanding their transmission routes.
MNV-1 is known to be more resistant than FCV-F9 to low pH values (Cannon et al., 2006). FCV-F9 was reduced by 5 logs at a pH of ≤2 for 30 min at 37°C (Duizer et al., 2004). When exposed to a pH of 2 at 37°C for 30 min, MNV-1 titer was reported to be reduced by <1 log (Cannon et al., 2006). Our control experiments in PBS (pH=7.2) showed survival of FCV-F9, MNV-1, and MS2 over 21 days at 4°C. FCV-F9 and MS2 showed complete reduction after 7 days and 21 days at 4°C, respectively, in PBS at pH 2.77, while MNV-1 showed minimal reduction (<1 log10 PFU/mL) after 21 days, as expected based on previous studies (Cannon et al., 2006). Bacteriophage MS2 was found to be slightly more resistant than FCV-9 in blueberry juice, but was completely reduced after 7 days at 4°C, which correlates with earlier pH studies (Cannon et al., 2006; Hewitt and Greening, 2004).
Blueberries, both the lowbush “wild” type (Vaccinium augustifolium Aiton) and the highbush “cultivated” type (Vaccinium corybosum L.), are flowering plants native to North America and sold fresh or in processed foods with properties beneficial to health (Kalt et al., 2001; Takeshita et al., 2009; Arnold and Noble, 1978; Brossaud et al., 2001). Pomegranates (Punica granatum L.) also have antibacterial, antiviral, and health benefits (Viuda-Martos et al., 2010; Haidari et al., 2009; Al-Zoreky, 2009). Recently, MS2, FCV-F9, and MNV-1 titers were reported to be reduced by pomegranate juice, pomegranate polyphenols, cranberry juice, and cranberry proanthocyanidins (Su et al., 2010a,b). It is not surprising that FCV-F9 did persist for shorter periods of time in blueberry juice. However, MNV-1 still did persist long enough over 21 days that it could correlate with the ease of human NoV transmission during contamination. MNV-1 may be more appropriate for use as a surrogate in survival studies than the pH-sensitive FCV-F9. Even though pomegranate and cranberry juice may cause some reduction in human enteric viral surrogate titers, these same viral surrogates can survive for longer periods of time in blueberry juice. Their ability to survive in blueberry juice over a relatively short period of time at refrigeration drives the need to improve the safety of juices to prevent foodborne outbreaks.
Based on the results with pH at 2.7 alone, it appears that additional factors associated with blueberry juice play a role in the decreased survival of the tested surrogates over refrigeration. As seen in Table 1, all three viruses survived longer in PBS at pH 2.7 than in blueberry juice. The reduced survival of surrogates in blueberry juice can partially be attributed to low pH and also potentially to the natural polyphenols found in blueberry juice along with other “unidentified or uncharacterized” components of the juice (Burdulis et al., 2009). Hence, to test this hypothesis, further studies to determine if blueberry juice contains natural antivirals to prevent foodborne gastroenteritis should be carried out.
Heat and chemical disinfectants are commonly used to control microorganisms that may affect food texture, flavor, and nutritive value. Thus, it is important to investigate alternative technologies to control outbreaks. As could be predicted based on previous studies with MNV-1 and HPH, MNV-1 was found to be the most resistant among the tested viruses (D'Souza et al., 2009), being reduced at 300 MPa in blueberry juice by only 0.71 log10 PFU/mL. In this study, FCV-F9 and MS2 were not detectable after exposure to 100 MPa in blueberry juice compared to 250–300 MPa required in PBS. Reductions for FCV-F9 are not surprising, since FCV-F9 is reported to be sensitive to high-pressure treatments. Complete inactivation of 7 log FCV after exposure to ≥275 MPa of hydrostatic pressure was reported (Grove et al., 2006). FCV-F9 was also shown to be completely inactivated after 5 min of ≥275 MPa high hydrostatic pressure (Kingsley et al., 2002; Kingsley et al., 2007).
HPH is a novel, high-pressure, non-thermal, continuous inactivation process that maintains the sensory and nutritional attributes of fluids. The exact mechanism of viral inactivation by HPH is still unclear. It is theorized that the process causes changes or disruption in viral capsid structure and different viruses may have different resistance due to dissimilarity in their coat proteins and morphology (D'Souza et al., 2009; Moroni et al., 2002). Further studies involving transmission electron microscopy are needed to address the mechanism of action. HPH shows promise as one of the novel technologies for the inactivation of foodborne viruses in fluid foods.
Conclusion
These data suggest that FCV-F9 was the most sensitive among the three tested virus surrogates in blueberry juice and after HPH treatment, while MNV-1 was the sturdiest. These studies show the survivability of human enteric virus surrogates in blueberry juice that could account for human enteric viral outbreaks. Hygienic practices are essential during handling and storage of juices. The application of novel-processing methods along with hurdle approaches can potentially reduce juice-related foodborne outbreaks.
Footnotes
Acknowledgments
We gratefully acknowledge funding provided by the Tennessee Agricultural Experiment Station (UT-TEN-HATCH #00391).
Disclosure Statement
No competing financial interests exist.
