Influenza virus
In addition to HIV, Influenza virus A is another classic example that the viruses have zoonotic origins with associated public health concern of host adaptation, permanently establishing in the human population[64]. Varble and colleagues[24]first barcoded influenza A virus, with which the authors attempted to understand the probability and stochastic nature resulting from different transmission routes. Varble and colleagues[24]employed molecular barcodes on the genome of a model influenza A virus to track viral transmission. The challenge to insert molecular barcodes in the genome of influenza A virus is because the eight single-stranded RNA segments are all required for virion production; therefore there is little space for molecular manipulation. Previously Varble and colleagues[65]have chosen to engineer segment 8, encoding the nonstructural protein 1 (NS1) and the nuclear export protein (NEP, also referred to as NS2) because of the shorter of two viral transcripts, facilitating to give any genetic material. Since NS1 and NEP use different reading frames with the overlap between the C-terminal transcript of NS1 and the N-terminal transcript of NEP, the authors thus disrupted the endogenous splice acceptor site and placed it after the stop codon of NS1[65](Figure 4A). This modified segment 8 thus harbored a noncoding intergenic region where a string of 22 nucleotides of a molecular barcode was inserted (Figure 4A). Barcodes were amplified from the 30 arm of a shRNA library[66], along with 100 base pairs of common flanking sequence (Figure 4A). The authors confirmed that inserted molecular barcodes did not affect viral replication in human lung epithelial cells.
Russel and colleagues[67]utilized three different types of barcodes, cell barcodes, UMI and viral barcodes to distinguish every single mRNA transcript based on the single-cell droplet-based platform[67,68](Figure 4B). Cell barcodes were used to distinguish one cell from another; UMIs incorporated in primers were designed to separate each mRNA molecule during the process of reverse transcription, allowing to quantify the number of molecules of each mRNA that have been captured for each cell. The challenge here is how to separate viral- and cellular mRNA generated in the same cell. The authors thus engineered the viral genomic sequence where two synonymous mutations were given in the vicinity of the 3’ end of each transcript (Figure 4B) for this discrimination. The transcription of the viral hemagglutinin from the viruses carrying these synonymous mutations was compatible with the wild-type viruses[67], indicating that synonymous mutations did not affect viral fitness. The authors successfully used these viral barcodes to distinguish the viral transcripts carrying mutations from the transcripts produced by the wild-type viruses[67]. By employing this three-barcodes approach, the authors established a threshold determining truly infected cells and quantitatively investigated stochastic gene expression of influenza viruses among individual infected cells[67].