Batch testing for noroviruses in frozen raspberries

Abstract

Berries, in particular raspberries, have been associated with multiple recalls due to norovirus contamination and were linked to a number of norovirus (NoV) outbreaks. In the present study a total of 130 samples of frozen raspberries were collected from 26 batches in four different raspberry processing companies. In two companies the samples consisted of bulk frozen raspberries serving as raw material for the production of raspberry puree (an intermediate food product in a business to business setting). In two other companies, the samples consisted of bulk individually quick frozen (IQF) raspberries serving as raw material for the production of frozen fruit mixes (as a final food product for consumer). Enumeration of Escherichia coli and coliforms was performed as well as real-time reverse transcription PCR (RT-qPCR) detection of GI and GII NoV (in 2×10g). In addition, in cases where positive NoV GI or GII RT-qPCR signals were obtained, an attempt to sequence the amplicons was undertaken.Six out of 70 samples taken from the 14 batches of frozen raspberries serving raspberry puree production provided a NoV RT-qPCR signal confirmed by sequencing. Four of these six positive samples clustered in one batch whereas the other two positive samples clustered in another batch from the same company. All six positive samples showed NoV RT-qPCR signals above the limit of quantification of the RT-qPCR assay. These two positive batches of frozen raspberries can be classified as being of insufficient sanitary quality. The mean NoV level in 20g of these raspberry samples was 4.3 log genomic copies NoV GI/20g. The concern for public health is uncertain as NoV RT-qPCR detection is unable to discriminate between infectious and non-infectious virus particles. For the IQF raspberries, one batch out of 12 tested NoV positive, but only 1 out of the 5 samples analyzed in this batch showed a positive RT-qPCR GI NoV signal confirmed by sequencing. The RT-qPCR signal was below the limit of quantification of the assay used (<3.7 log genomic copies/20g). It was shown that the applied protocol for sequencing of the amplicon to confirm the specificity of the RT-qPCR signal was successful for GI NoV amplicons but often failed and provided an inconclusive result for GII NoV amplicons.