Article pubs.acs.org/est
Temperature and Humidity Influences on Inactivation Kinetics of Enteric Viruses on Surfaces Su Jung Kim,† Jiyeon Si,† Jung Eun Lee, and GwangPyo Ko* Department of Environmental Health, Graduate School of Public Health, Seoul National University, 1 Gwanak-ro, Gwanak-gu, Seoul 151-742, South Korea S Supporting Information *
ABSTRACT: Norovirus (NoV) and hepatitis A virus (HAV) are pathogenic enteric viruses responsible for public health concerns worldwide. The viral transmission occurs through fecally contaminated food, water, fomites, or direct contact. However, the difficulty in cultivating these viruses makes it a challenge to characterize the resistance to various environmental stresses. In this study, we characterized the inactivation rates of murine norovirus (MNV), MS2, and HAV on either lacquer coating rubber tree wood or stainless steel under different temperature and relative humidity (RH) conditions. The viruses were analyzed at temperatures of 15 °C, 25 °C, 32 °C, and 40 °C and at RHs of 30%, 50%, and 70% for 30 days. Overall, they survived significantly longer on wood than on steel at lower temperature (P < 0.05). The inactivation rate of MS2 and MNV increased at higher RH levels, whereas HAV survived the best at a medium RH level (50%). The effect of RH was significant only for MS2 (P < 0.05). MS2 persisted longest under all of the environmental conditions examined. Both a linear and a nonlinear Weibull model were used to describe the viral inactivation data in this study. The data produced a better fit to the survival curves that were predicted by the Weibull model.
■
INTRODUCTION Both norovirus (NoV) and hepatitis A virus (HAV) are among the most important food-borne viruses in terms of the frequency and severity of enteric viral diseases.1,2 NoV is highly infectious and requires as little as 10−100 virions to cause viral gastroenteritis.3 NoV outbreaks have also frequently been reported in various settings including schools, restaurants, tourist resorts, nursing homes, and cruise ships worldwide.3−7 These outbreaks have mostly occurred through the fecal/oral route, person-to-person contact, or indirectly through contaminated food, water, or fomites allowing the virus to contaminate the public through multiple routes. However, viral surrogates such as murine norovirus (MNV) or MS2 are typically used in studies instead of NoV as it is not possible to cultivate the virus with conventional cell culture techniques.8,9 MNV is a recently discovered surrogate that can be cultured in mammalian cells.10 Its biological similarity to NoV makes it the most appropriate surrogate in studies on NoV infectivity.8 MS2 is an F-specific RNA phage and has also been extensively used in many previous studies as a suitable and convenient representative of the nonenveloped viruses.11−14 HAV causes acute hepatitis and is another major concern for public health.15 In a similar manner to NoV, HAV transmission occurs through the consumption of fecally contaminated water or food. In addition to documented outbreaks in public places, its persistence in the environment increases the chance of viral transmission.16 The survival characteristics of these two viruses under various combinations of environmental conditions have not been well characterized. Transmission of enteric viruses is influenced by © 2012 American Chemical Society
environmental factors such as temperature, relative humidity (RH), and the type of surface contaminated.15,17 As NoV cannot be cultured in mammalian cells, inactivation of the virus is commonly monitored with cultivable feline calcivirus (FCV) as both belong to the Caliciviridae family.18−21 However, FCV is an animal virus that causes respiratory diseases in cats.22 A recent study has reported that MNV is more applicable as an NoV surrogate than FCV due to its environmental stability and genetic similarity.21 When the cultivation of HAV has been achieved, the survival characteristics have been observed under a very limited range of temperature and RH conditions15,23 and are not yet well characterized. In this study, we investigated the persistence and survival of food-borne viruses under a wide range of temperature and relative humidity (RH) conditions over a relatively long period of time (>1 month). Two commonly used surfaces for food preparation (i.e., wood and stainless steel) were chosen for direct comparison of the surface effects on virus survival. We assessed the effects of temperature, RH, and surface characteristics on the survival of MNV, MS2, and HAV, and quantitatively evaluated the survival characteristics of these viruses using different mathematical models.
■
MATERIALS AND METHODS Preparation of Virus Stocks. MNV was cultivated in RAW 264.7 cells in Dulbecco’s modified Eagle’s medium Received: Revised: Accepted: Published: 13303
August 7, 2012 November 15, 2012 November 15, 2012 November 15, 2012 dx.doi.org/10.1021/es3032105 | Environ. Sci. Technol. 2012, 46, 13303−13310
Environmental Science & Technology
Article
predetermined times: day 1, day 3, day 7, day 10, day 20, and day 30. Each fragment was immersed in 10 mL PBS solution, sonicated for 30 min at 4 °C, and analyzed by a plaque assay. When the viruses were fully inactivated, in approximately 1 day, a repeated test was performed with shorter time intervals: MS2 and MNV at 9 h, 18 h, and 24 h, and HAV at 6 h, 12 h, and 24 h. Elution efficiency was tested by comparing HAV deposited on the surfaces with the virus directly inoculated into PBS as a control. Approximately 100% and 98% of the inoculated virus was eluted from stainless steel and wood, respectively. Plaque Assay. The titer of MNV was determined by a plaque assay as previously described.25 The assay for HAV followed a similar process. FRhK-4 cells were seeded at 1 × 106 cells/well for 3−6 h. The cells were infected with the virus and covered with an agar layer. On the seventh day postinfection, the cells were stained with a second agar overlay mixed with 0.2% of neutral red solution. The plates were incubated for another 3 days and counted. MS2 was assayed by the single agar layer (SAL) method according to EPA standard method 1602.26 Briefly, each of the eluted MS2 solutions was serially diluted in PBS. These MS2 suspensions were mixed with 0.3 mL of log phased E. coli C3000 and 30 mL of tryptic soy broth (TSB) with 0.8% agar in 50-mL polypropylene tubes. These suspensions were poured into sterile 150-mm Petri dishes and mixed by swaying. When the agar layer was hardened, the plates were incubated at 37 °C for 24 h before being counted. Countable plates were those having 3−300 plaques per plate. Data Analysis and Modeling of Survivability. The number of plaques at each time point was analyzed by an exponential inactivation model. The D-value was estimated from the slope of log percent survival and was equivalent to the time required for l-log10 or a 90% viral reduction:27
(DMEM; Gibco, Grand Island, NY, U.S.) adding 10% fetal bovine serum (FBS; Gibco), 10 mM hydroxyethyl piperazineethanesulfonic acid (HEPES; Gibco), 10 mM sodium bicarbonate (Gibco), gentamicin 50 g/L (Gibco), and 10 mM Earle’s minimal essential/nonessential amino acid (Gibco).10 For the preparation of virus stocks, RAW 264.7 cells were cultured confluently (80−90%) in 175-cm2 flasks at 37 °C in 5% CO2 and infected with MNV. After 3−4 days of incubation, the infected cells were exposed to three freezing and thawing cycles and then filtered using a 0.22-μm pore-sized syringe filter (Millipore, Billerica, MA, U.S.). Supernatants were centrifuged every 10 min at 2000 × g and 4 °C 25−30 times to enrich the virus and stored at −80 °C until use. HAV was kindly obtained from Dr. Sobsey’s laboratory at the University of North Carolina at Chapel Hill and cultivated as previously described.24 HAV stock was prepared by infecting FRhK-4 cells (ATCC CRL-1688) with 0.1 multiplicity of infection. The infected cells were incubated at 37 °C for 7−8 days until cytopathogenic effects (CPEs) were observed. After three cycles of freezing and thawing, the infected cell lysates were purified by chloroform extraction and centrifuged for 20 min at 4650 × g, 4 °C and then at 2000 × g every 10 min 25−30 times. MS2 bacteriophage was propagated in Escherichia coli C3000. A mixture of MS2, suspensions of the host, and tryptic soy agar (TSA) was incubated for 18−24 h at 37 °C. The surface of the agar plates was washed with phosphate-buffered saline (PBS) and mixed with an equal volume of chloroform in 50-mL polypropylene tubes. Supernatants were centrifuged at 4000 × g for 20 min at 4 °C and stored at −80 °C. The viral titers of MNV and MS2 were determined by plaque assay as described previously25 and estimated to be about 109 and 1011 plaque forming units (PFU)/mL, respectively. HAV stock was also estimated to be 107 PFU/mL by a plaque assay. Measurement of Viral Inactivation on Wood and Stainless Steel under Various Environmental Conditions. A chopping board (made from the wood of a rubber tree with a nitrocellulose clear lacquer finish) was purchased from a commercial company (Daiso A-sung Corp., Seoul, South Korea) and used for wood surfaces. Stainless steel was also purchased from a commercial company (Buil Science, Seoul, South Korea) to represent metal surfaces such as a kitchen knife. The chopping board and stainless steel were cut into identically sized fragments of 2 × 1.5 × 1.5 cm and 1.5 × 1.5 cm, respectively. These fragments were sterilized by autoclaving and dried overnight. To prevent any other microbial contamination, these materials were disinfected by exposure to 254-nm germicidal UV for 10 min prior to the inactivation experiments and used only once for this study. One hundred microliters of the virus to be tested (virus titer was 107 PFU/ mL for MNV, 108 PFU/mL for MS2, and 106 PFU/mL for HAV) was deposited onto three replicate samples of wood and stainless steel were placed in a Petri dish. Lids were placed on top of the Petri dishes to prevent desiccation and contamination, but kept in place with tape, allowing a partial opening. These samples were then stored in a preconditioned experimental chamber (Temperature and Humidity Chamber; JEIO TECH, Daejeon, South Korea), where temperature and RH conditions were set and continuously monitored over a 1month period. Four different temperature conditions were used: 15 °C, 25 °C, 32 °C, and 40 °C. For each temperature, the RH levels were set at 30% ± 0.3%, 50% ± 0.3%, and 70% ± 0.3%. Viruses eluted immediately from the inoculated surfaces were used as a control (time = 0). Other samples were eluted at
log10
Nt t =− N0 D
where Nt = concentration of viruses (PFU/mL) after the elapsed time, N0 = concentration of viruses (PFU/mL) at initial time, t = time (day), and D = D-value for the indicated conditions. In addition to the linear model described above, a nonlinear Weibull model was applied to the experimental data as a predictive survival model.27,28 The Weibull model is as follows: log10
Nt = −bt n N0
where t = exposure time (day), b = scale parameter, and n = shape parameter. These parameters were determined by nonlinear regression analysis using SigmaPlot ver. 10.0 (Chicago, IL, U.S.). The parameter n alters the shape of the survival curve in a concave upward direction if n < 1, a concave downward direction if n > 1, and is linear if n = 1. The value of 1-log10 reduction for the Weibull model, Td, was calculated as follows: Td =
⎛ d ⎞1/ n ⎜ ⎟ ⎝b⎠
where d is the number of decimal reductions. To assess the goodness-of-fit of the model, the regression coefficient (R2) and root-mean-square error (RMSE) were determined.27 For further analysis of the model fitting, correlation plots were prepared for the observed experimental values and fitted values 13304
dx.doi.org/10.1021/es3032105 | Environ. Sci. Technol. 2012, 46, 13303−13310
Environmental Science & Technology
Article
Figure 1. Box-and-whisker plots of D-values under various temperatures, RH conditions, and surfaces for MS2, MNV, and HAV. The boxes represent the 25th percentile, median, and 75th percentile. The whiskers represent the lowest and the highest D-values. Filled circles indicate the outliers.
Table 1. Simple Linear Regression Analysis for the Rates of Inactivation and Environmental Variables D-value MS2
a
MNV
HAV
variables
βa
standard error
p-value
β
standard error
p-value
β
standard error
p-value
temperature RH surface
−3.19 −2.38 4.603
0.497 0.68 1.11
