ABSTRACT
The fragmented nature of the influenza A genome allows the exchange of gene segments when two or more influenza viruses infect the same cell, but little is known about the rules underlying this process. Here, we studied genetic reassortment between the A/Moscow/10/99 (H3N2, MO) virus originally isolated from human and the avian A/Finch/England/2051/91 (H5N2, EN) virus and found that this process is strongly biased. Importantly, the avian HA segment never entered the MO genetic background alone but always was accompanied by the avian PA and M fragments. Introduction of the 5' and 3' packaging sequences of HA(MO) into an otherwise HA(EN) backbone allowed efficient incorporation of the chimerical viral RNA (vRNA) into the MO genetic background. Furthermore, forcing the incorporation of the avian M segment or introducing five silent mutations into the human M segment was sufficient to drive coincorporation of the avian HA segment into the MO genetic background. These silent mutations also strongly affected the genotype of reassortant viruses. Taken together, our results indicate that packaging signals are crucial for genetic reassortment and that suboptimal compatibility between the vRNA packaging signals, which are detected only when vRNAs compete for packaging, limit this process.
Subject(s)
Birds/virology , Coinfection/virology , Gene Transfer, Horizontal/genetics , Hemagglutinins, Viral/genetics , Influenza A Virus, H3N2 Subtype/genetics , Influenza A Virus, H5N2 Subtype/genetics , Virus Assembly/genetics , Amino Acid Sequence , Animals , Base Sequence , DNA Primers/genetics , Dogs , Genotype , HEK293 Cells , Humans , Madin Darby Canine Kidney Cells , Molecular Sequence Data , Mutation/genetics , Sequence Analysis, DNA , Signal Transduction/genetics , Species SpecificityABSTRACT
Influenza neuraminidases hydrolyze the ketosidic linkage between N-acetylneuraminic acid and its adjacent galactose residue in sialosides. This enzyme is a tetrameric protein that plays a critical role in the release of progeny virions. Several methods have been described for the determination of neuraminidase activity, usually based on colorimetric, fluorescent, or chemiluminescent detection. However, only a few of these tests allow discrimination of the sialyl-linkage specificity (i.e., α2-3- versus α2-6-linked sialyllactosides) of the neuraminidase. Herein we report a glycoarray-based assay and a MALDI-TOF study for assessing the activity and specificity of two influenza neuraminidases on whole viruses. The human A(H3N2) and avian A(H5N2) neuraminidase activities were investigated. The results from both approaches demonstrated that α2-3 sialyllactoside was a better substrate than α2-6 sialyllactoside for both viruses and that H5N2 virus had a lower hydrolytic activity than H3N2.
