Neda Nasheri

Human noroviruses (HuNoVs) are the most common cause of non-bacterial foodborne illnesses worldwide. HuNoVs are notoriously difficult to culture, and current cell culture systems, such as human intestinal enteroids have limitations in efficiency and scalability, which has limited their applications in food safety studies. Recently, it has been shown that transformed salivary gland cell lines (including NS-SV-TT-DC) allow for replication and passaging of norovirus GII.4. Herein, we examined whether we can independently observe viral replication in salivary gland cells and if other strains of human norovirus can replicate in these cells. For this purpose, we first evaluated the replication of several norovirus strains in NS-SV-TT-DC cells by either infecting them with vesicle-cloaked viruses or total viruses in stool filtrates. Low levels of viral replication was observed and vesicle isolation did not improve viral replication. Next, we screened for replication of 50 norovirus samples from 9 genotypes, and 14 isolates demonstrated over a two-fold viral replication. We observed that the initial viral load in the samples does not have a significant effect on replication efficiency. We were also able to successfully passage several isolates. Subsequently, we examined whether interferon inhibitors such as Ruxolitinib, TPCA-1, and BX795 would improve viral replication and observed negligible effects. Finally we tested multiple supplements including MgCl ₂ , bile salt, ceramide, and TAK-779. Most supplements had little or no effect on HuNoV replication in NS-SV-TT-DC cells. Bile salt and TAK-779 showed limited, genotype-specific benefits. Altogether these results indicate that salivary gland cells allow limited replication of certain HuNoV isolates. Importance: The lack of robust and accessible cell culture systems continues to hinder human norovirus research. Salivary gland cell lines have recently emerged as a promising alternative to human intestinal enteroids because they are easy to maintain and relatively unexpensive. Here, we provide a large-scale independent evaluation of the NS-SV-TT-DC salivary gland model using 50 clinical isolates from multiple human norovirus genotypes. Although several isolates replicated and could be serially passaged, replication was typically modest. Furthermore, multiple approaches commonly used to enhance viral replication in cell culture had little or no effect on system efficiency. These findings establish realistic expectations for the use of this model and provide a framework for future optimization of human norovirus cultivation systems.

Human noroviruses (HuNoVs) are the leading cause of non-bacterial gastroenteritis worldwide, exhibiting extensive genetic diversity driven by recombination and mutation. The norovirus RNA-dependent RNA polymerase (RdRp), encoded by the NS7 gene, is the main driver of virus evolution by determining the mutation and recombination rates, and impacting viral fitness. Herein, we conducted spatiotemporal phylodynamic analysis on 1,094 complete RdRp amino acid sequences from genogroups GI and GII collected between 1972 and 2024 to investigate molecular evolution, amino acid variation, evolutionary rates, and selection pressures. Time-scaled phylogenetic analysis revealed that GI polymerases diverged into three major lineages since the year ~ 1630, while GII polymerases formed four lineages, including a distinct cluster for P16. Genotype distribution was dominated by GII.P16 (43%) and GII.P31 (20%), both historically associated with GII.4 capsids. Amino acid substitutions were rare to non-existent within conserved motifs (A–G), indicating strong structural and functional constraints. Evolutionary rate estimates showed GII polymerases evolve faster (mean 1.5 × 10⁻³ substitutions/site/year) than GI (mean 3.8 × 10⁻⁴ substitutions/site/year), though overall RdRp evolves more slowly than the major capsid protein VP1. Selection analyses demonstrated pervasive purifying selection, with limited evidence of positive selection at codons outside of the conserved motifs. These findings highlight the high conservation of RdRp, its slower evolutionary dynamics compared to capsid gene, and its role in norovirus diversification, supporting its relevance as a target for antiviral development.

The spillover of highly pathogenic avian influenza (HPAI) to dairy cattle and its presence at high titers in raw milk has created a concern that consumption of dairy products made from unpasteurized milk could result in human exposure. We examined whether H1N1, as a surrogate for HPAI viruses, and murine norovirus (MNV), as a surrogate for unenveloped enteric viruses, can survive the cheese-making process by making cream cheese-style and feta-style cheeses from artificially inoculated raw milk. We used the modified ISO 15216 method for viral extraction, which has a limit of detection of 3.0–3.7 log for infectious H1N1 in soft cheeses. Infectious H1N1 virus was detected in the curd from both types of cheeses and in the whey from cream cheese but not in the whey from feta cheese. Infectious MNV particles were detected in the curd and in the whey for both types of cheeses. To study the effect of salt, pH, aging, and microbiome of various unpasteurized milk cheeses on viral survival, we artificially inoculated different commercial unpasteurized milk cheeses with 7 log PFU of H1N1 and 6 log PFU of MNV and examined viral survival over an 8 week period at 4°C. We observed that H1N1 survived for 8 weeks, with an average of 2.6, 2.75, and 4.0 log reductions on cheddar, washed rind firm cheese, and semisoft brie-like cheese, respectively ( P < 0.001). The level of infectious H1N1 fell below the limit of detection in cream cheese after 2 weeks of inoculation. It is noteworthy that the sensory qualities of soft and semisoft cheeses deteriorated in 2 weeks, but the sensory qualities of firm cheeses did not change drastically, demonstrating that, as expected, they endure the aging process better than the soft and semisoft cheeses. Furthermore, the limit of detection for soft cheeses is approximately 1 log lower compared to firm cheeses, and the extraction method has a lower efficiency for soft cheeses compared to firm cheeses. MNV survived in firm cheeses for 8 weeks with only 1 log reduction on cheddar cheese and 2 log reductions on the washed rind firm cheese ( P < 0.05). The data obtained in this study could help with the risk assessment of dairy products made from unpasteurized milk. IMPORTANCE: The rapid spread of H5N1 viruses among dairy cattle in the United States and the viral shedding at high titers raised concern regarding the safety of dairy products, specifically cheeses made with unpasteurized milk. In this study, we demonstrated that murine norovirus and influenza virus can survive the cheese-making and aging process. Therefore, it is recommended that milk contaminated with viruses be heat-treated to ensure safety.

Human norovirus (HuNoV) is the leading cause of acute gastroenteritis globally, with frozen berries frequently implicated in foodborne outbreaks. Current surveillance relies on quantitative reverse transcription PCR (qRT-PCR), which cannot differentiate between infectious and non-infectious viral particles, complicating risk assessment. This study is aimed to establish the minimum viral load on frozen berries detectable by qRT-PCR that corresponds to infectious virus, using murine norovirus (MNV) as a surrogate for HuNoV. Frozen raspberries were artificially inoculated with serial dilutions of MNV (7.1–1.0 log PFU/25 g) and processed using the ISO 15216:2017 method. Infectious virus was quantified by plaque assay, and viral RNA was detected by qRT-PCR. The limit of detection (LOD) for cell culture was 3.1 log PFU/25 g, whereas qRT-PCR extended sensitivity to 1.0 log PFU/25 g (Ct value at 36.7 ± 0.6), representing a 2-log difference. Recovery rates for infectious virus exceeded the ISO 15,216 minimum threshold (1%), and PCR inhibition was negligible. We next examined the extraction efficiency for both infectious MNV and its genetic material from frozen strawberries at inoculation levels higher than the LOD, and observed that the viral recovery from frozen strawberries is very similar to viral recovery from frozen raspberries with no significant differences between them. The disparity between LODs indicates that a substantial proportion of MNV genomes detected by qRT-PCR do not represent infectious particles, aligning with previous findings that one PFU may correspond to multiple genome copies. Given that many surveillance studies report high Ct values (> 35), our data suggest that such detections may not indicate viable virus, underscoring the importance of contextualizing qRT-PCR results with epidemiological evidence. These findings highlight the need for cautious interpretation of surveillance data, particularly for public health decision-making.

Human noroviruses (HuNov) are the major cause of foodborne illness globally. Several HuNoV outbreaks have been linked to contaminated ready-to-eat seaweed products. Standard protocols such as the ISO 15216 show limited efficiency in extracting foodborne viruses from seaweed products. Therefore, we evaluated the efficiency of an extraction protocol based on porcine gastric mucin conjugated magnetic beads (PGM-MBs) to recover HuNoVs from Wakame seaweed salad. Compared to other HuNoV extraction methods, the PGM-MB method was more efficient. We then aimed to further improve this protocol by modifying several factors such as the buffers, pH, bead concentration, centrifugation and incubation time. The optimized PGM-MB method yielded 19 ± 3% and 17 ± 4% recovery, for HuNoV GI and GII, respectively. The limit of detection (LOD₉₅) for Wakame seaweed salad was 131 and 56 genomic equivalents per 25 g for HuNoV GI and GII. Although some variability in recovery efficiency was observed between the PGM sources, the optimized PGM-MB protocol effectively extracts HuNoVs from Wakame seaweed salads of various brands and other commodities such as dates, green onions, and salted seaweed. These results support the implementation of the optimized PGM-MB method as a viable alternative for HuNoV surveillance in complex food matrices.

The recent spillover events of highly pathogenic avian influenza (HPAI) A(H5N1) clade 2.3.4.4b to dairy cattle, and high viral shedding in the milk from infected animals, has created concern that milk and dairy products could be a route for human infection. It has been demonstrated that pasteurization is effective in inactivating A(H5N1) in milk. However, multiple dairy products are made with unpasteurized but thermised milk. The aim of this study was to examine whether some conditions commonly used for thermisation are effective against inactivation of A(H5N1) in whole milk. For this purpose, we artificially inoculated whole raw cow milk with 6.5 log ₁₀ EID ₅₀ A(H5N1) and heated for 15 seconds at 60°C, 63°C and 66°C, the viral infectivity was tested using embryonated chicken eggs. We observed over 4 and 5 log ₁₀ reduction in viral infectivity at 60°C and 63°C, respectively. The viral infectivity was reduced to below the detection limit at 66°C. We also calculated the D-values, the time required to reduce the viral titer by one log ₁₀ , for each treatment and as expected, we observed a decrease in D-values with increasing thermisation temperature. These data demonstrate that thermisation is effective in reducing the viral load and thus they allow for informed risk assessment of A(H5N1) contaminated dairy products made from thermized milk.

Human noroviruses are the leading cause of non-bacterial shellfish-associated gastroenteritis. In 2022, a multi-jurisdictional norovirus outbreak associated with contaminated oysters occurred that involved hundreds of illnesses. Here, we conducted genetic analysis on 30 clinical samples associated with this oyster outbreak. We first determined the capsid genotypes by Sanger sequencing and viral titers by droplet-digital reverse transcription PCR. Multiple genotypes were identified in this outbreak, which could indicate contamination with wastewaters. The majority of samples belonged to GII.3[P12], followed by GII.2[P16], GII.17[P17], and GII.4 Sydney[P16]. We next proceeded with whole-genome sequencing and obtained full genomes for 19 samples. Phylogenetic analysis revealed that some of the isolates showed high similarity with the sequences isolated from the United States related to the same outbreak. We also analyzed amino acid variations in the sequenced genomes and found that overall the GII.3[P12] isolates have lower variations compared to other genotypes. IMPORTANCE: Norovirus outbreaks associated with contaminated shellfish occur frequently. Whole-genome sequencing (WGS) could play a critical role in understanding and controlling norovirus outbreaks as it allows for source attribution, tracking transmission pathways, and detecting recurrent or linked outbreaks. Here, we described how the data obtained by WGS were employed for understanding transmission patterns and norovirus epidemiology.