Risk assessment of norovirus on shellfish from Indonesian fish markets

Norovirus (NoV) infection is estimated to cause almost 20% of acute gastroenteritis cases worldwide. Infants, the elderly and the immunocompromised are those most susceptible to NoV infection. NoV is known to be persistent in the environment for long periods (60-80 days at 25°C), is infectious at low doses (at 8 – 2,800 viral particles), can be shed at high concentration (up to 10\(^9\)-10\(^{11}\) viral copies per gram faeces of infected person), and is mainly transmitted through the faecal-oral route. Therefore, a small amount of NoV contamination in the environment, water or food can cause large outbreaks. Shellfish, in particular, are susceptible to NoV contamination because they filter large amounts of water and accumulate different types of suspended particles including bacteria and viruses when grown or harvested from contaminated areas. In Indonesia, some shellfish growing and harvesting areas are located close to estuaries which can be contaminated by untreated domestic sewage effluent, especially during flood incidents. Even though shellfish in Indonesia are mostly consumed cooked, inadequate cooking and cross-contamination during food preparation steps can lead to NoV contamination in the prepared meal. Risk assessment of NoV, especially in shellfish from Indonesian markets, remains challenging due to the lack of prevalence data, no recorded NoV outbreaks caused by shellfish consumption, and the lack of knowledge of the efficacy of post-processing steps including handling and cooking based on consumer behaviour in Indonesia. Boiling, stir-frying and steaming are the most common cooking practise of shellfish in Indonesia which can reduce the NoV contamination. In case the shellfish is being consumed as a raw or fresh product, the use of disinfectant such as Chlorine Dioxide (ClO\(_2\)) to reduce the viral contamination or to prevent cross-contamination during post harvesting or handling is a potential risk management strategy. In addition, standard quantification assays for NoV based on the cell-culture system are as yet unavailable. Therefore, NoV studies rely on molecular based methods such as Reverse Transcription Polymerase Chain Reaction (RT-PCR). This project optimised a Reverse Transcription Quantitative Polymerase Chain Reaction (RT-qPCR) method to obtain prevalence data on NoV in shellfish from Indonesian markets and further utilised a NoV surrogate (MS2 bacteriophage, ‘MS2’) for inactivation studies, to fill those data gaps. The results provided better understanding of NoV prevalence and survival and could be used to predict the risk of NoV contamination in shellfish from Indonesian markets. The first aim of this thesis was to evaluate the application of RT-qPCR after pre-treatment with enzymes because current methods quantify both infectious and non-infectious viral particles and may over-estimate the risk of viral infections, especially in the inactivation and prevalence studies. Therefore, sample pre-treatments are required to differentiate the infectious from non-infectious viral RNA. MS2, a cultivable NoV surrogate was used in this study. RT-qPCR after pre-treatment with RNase followed by RNasin showed better performance than RNase alone or TaqI in the elimination of the RNA from inactivated MS2 and produced a comparable result to the plaque assay. This modified RTqPCR method was shown to be applicable for the quantification of infectious MS2 after inactivation treatment by heat or ClO\(_2\), producing comparable results to plaque assays. The next aim of this thesis was to compare the inactivation kinetics of NoV and MS2 treated by heating and ClO\(_2\) in buffered media (PBS solution) and the shellfish matrix, as the NoV surrogates may have different inactivation kinetics compared to NoV. The efficacy of both inactivation methods was also determined. To provide artificial contamination of NoV and MS2 in the mussel for heat treatment studies, bioaccumulation process of the viruses in Tasmanian Blue Mussel (Mytilus galloprovincialis) was done to mimic the actual virus transmission routes in shellfish. While for ClO\(_2\) treatment studies, the mussels were artificially contaminated by dipping the tissue in solutions of NoV and MS2 for 30 min to represent the cross-contamination process. NoV and MS2 in buffered media and bioaccumulated mussel were heated at 60, 72 and 90° C at various times. The evaluation of NoV and MS2 inactivation kinetics showed that the Weibull model performed better in estimating the survival of NoV and MS2 in buffered media, while the Biphasic model provided better estimation of virus survival in mussel matrix. The D values of NoV were generally higher than MS2 in both buffered medium and mussel matrix, showing a higher resistance of NoV towards heat treatment. Furthermore, for all temperatures, inactivation of both viruses in mussel matrix required a longer time to achieve 1 log\(_{10}\) reduction compared to inactivation in buffered media. The efficacy of chlorine dioxide (ClO\(_2\)) to inactivate NoV and MS2 in buffered media and artificially contaminated mussel was studied using ClO at 10, 20 and 40 ppm with various exposure times at 25°C. The result showed that 40 ppm ClO\(_2\) treatment reduced NoV and MS2 more significantly in both buffered media and mussel matrix than 10 and 20 ppm treatments. In general, the virus reduction was higher in buffered media than in mussel matrix for all ClO\(_2\) treatments. For example, the reduction of MS2 in buffered media treated with 40 ppm ClO\(_2\) for 200 min resulted in > 6 log\(_{10}\) PFU/ml reduction, while in mussel matrix the same treatment only reduced MS2 by < 3 log\(_{10}\) PFU/g. At the same treatment, NoV in buffered media were reduced for more than 3 log\(_{10}\) copies/ml, while only 2.36 log\(_{10}\) copies/g reduction was observed in mussel matrix. The inactivation of ClO\(_2\) of both viruses in buffered media and mussel matrix was equally well described using the quasi-mechanistic Hom model or the Weibull model. The first prevalence data for NoV GII in shellfish in Indonesia are presented in this thesis. The data are for three shellfish species i.e. Green Mussel (Perna viridis), Blood Cockle (Anadara granosa) and Oriental Hard Clam (Meretrix lusoria), that are commonly consumed in Indonesia. Shellfish were sampled from four fish markets in Jakarta and Panimbang, Indonesia, in July 2016 and 2017. The NoV from extracted digestive tissue (DT) of shellfish was enumerated using the enzymatic pretreated RT-qPCR developed in this study. NoV GII was detected in 11 out of 171 samples with contamination levels from 1.43 to 3.55 log\(_{10}\) copies/g DT. The NoV GII prevalence in Green Mussels was 10%, which was higher than the prevalence in Oriental Hard Clam (7.14%) and Blood Cockle (2.9%). All NoV-contaminated shellfish were collected from traditional fish markets (Muara Kamal and Cilincing) harvested from Jakarta Bay. Due to the paucity of relevant data, a deterministic approach was used to estimate the risk of illness due to the consumption of NoV contaminated shellfish from Indonesian markets. In the worst-case scenario where the level of contamination is 8.98 x 10\(^3\)log\(_{10}\) copies/g DT, boiling for more than 30 min during cooking step can significantly reduce the estimated NoV outbreaks due to shellfish consumption. Based on the results from the inactivation studies, both inactivation treatments (heat and ClO\(_2\)) can be used as control measures to reduce NoV contamination in shellfish. Even though MS2 was more susceptible to heat treatment than NoV, the use of this surrogate in those studies has provided a better understanding on inactivation kinetics and tailing phenomenon in both treatments. Together with the data of NoV exposure or prevalence in shellfish from the markets, the result from the inactivation studies was used to develop a risk assessment that can assist in risk management These data provided scientific evidence which can be applied to improve the quality and safety of shellfish production and provide consumer protection from NoV infection in Indonesia. The findings from this study also emphasised the need for regular surveillance in the polluted growing or harvesting areas such as Jakarta Bay, and the application of proper cooking or disinfection to reduce the risks of NoV gastroenteritis from consumption of the contaminated shellfish