Use of sequence analysis of the P2 domain for characterization of norovirus strains causing a large multistate outbreak of norovirus gastroenteritis in Germany 2012

Abstract

Human norovirus is the main cause of non-bacterial gastroenteritis worldwide. It is transmitted from person to person, by fecally contaminated food or water or through virus containing aerosols originating during vomiting of infected persons. In September and October 2012, the largest foodborne norovirus outbreak in Germany so far spread over 5 Federal States (Berlin, Brandenburg, Saxony, Saxony-Anhalt, and Thuringia) affecting nearly 11,000 people mainly in schools and child care facilities. Epidemiological and trace-back investigations supported the assumption that a batch of frozen strawberries imported from China was the likely source of the outbreak. Sequence analysis of the capsid region encoding the P2 domain was used successfully for identification of transmission routes and epidemiologic relationship but was hampered by a lack of universal primers for all known genotypes so far. In the present study, a molecular approach was designed to track outbreak-related samples from the affected states of the large foodborne outbreak in Germany. Therefore, sequence analysis within the highly variable P2 domain of the capsid gene using newly developed universal P2 primers for genogroup I and genogroup II strains in combination with sequencing of the polymerase gene (region A) and the orf1/orf2 junction (region c) was used. The sequence analysis of 138 norovirus positive stool samples suspected to be outbreak-related revealed a considerable genomic diversity. At least 3 strains of genogroup I (I.3, I.4, and I.9) and 5 strains of genogroup II (II.6, II.7, II. 8, and recombinants II.P7_II.6, and II.P16_II.13) as well as 19 samples containing mixtures of these strains were detected. Six samples were considered as not linked to the outbreak. The most prevalent genotype was GI.4 (48/132; 36%). Genotype I.9 and the recombinant strain II.P16_II.13 were detected for the first time in Germany. Notably, the genotype II.P16_II.13 could also be determined in one of the samples of the frozen strawberry lot suspected as infection source. Especially, due to the good concordance of the P2 sequences from infected patients of 5 Federal States the outbreak-relation of the strains could be demonstrated. The high diversity of virus strains and the occurrence of sub-clusters within genotypes I.3, II.8, II.P16_II.13, and II.7 revealed the complex mixture of the outbreak source suggesting a possible waterborne fecal contamination of the strawberries. The typing system described here is in general useful for analysis of outbreaks caused by mixed infection sources. Extensive sequence analysis of different gene regions including the highly variable P2 domain in a sufficient number of cases is required to confirm the epidemiological relation of samples from outbreaks with high diversity of strains spreading over several geographic locations.

Keywords: Foodborne outbreak; Genotyping; Molecular epidemiology; Norovirus; P2 domain.