Evaluation of the use of alternative sample types for mosquito-borne flavivirus surveillance: Using Usutu virus as a model

Wild birds are reservoirs of several zoonotic arboviruses including West Nile virus (WNV) and Usutu virus (USUV), and are often monitored as indicators for virus introduction and spread. To optimize the bird surveillance for arboviruses in the Netherlands and to explore the possibilities for citizen science in surveillance, we investigated the suitability of using alternative sample types from live and dead birds. The sensitivity of molecular detection via RT-PCR of viral RNA in feather, heart, lung, throat and cloaca swabs from dead birds, and serum, dried blood spots (DBS) and throat and cloaca swabs from live birds were compared. IgY antibody detection was also assessed from DBS relative to serum on protein-microarray and virus neutralization test. Feathers showed a high detection sensitivity for USUV RNA in both live and dead birds, and no significant decrease was observed in the RNA loads in the feathers after being stored dry at room temperature for 43 days. Additionally, viral RNAs extracted from feathers of day 0 and 43 were successfully sequenced. The results indicated no statistical significant difference in sensitivity and viral loads detection in heart, spleen, and lung relative to corresponding brain samples in dead birds. In live birds, viral RNA loads did not differ between throat and cloaca swabs. This study identified less-invasive sample types that allows involvement of citizens in collecting samples from wild birds for arbovirus surveillance. Sensitivity and specificity of DBS-based antibody detections were significantly lower and therefore need optimization.