Mistakes during protein synthesis, including transcriptional errors, frequently produce non-inheritable amino acid substitutions in proteins, referred to here as phenotypic mutations. Using a consensus approach applied to high-throughput sequencing, the authors mapped the error landscape of influenza hemagglutinin (HA) mRNA. They report that single-site transcription errors occur with markedly different frequencies across the HA transcript. The most prevalent error encodes an amino acid change that reduces binding of broadly neutralizing antibodies to HA. That error is present in roughly 0.2–0.5% of HA mRNA molecules, meaning many virions are likely to expose HA variants bearing the encoded substitution on their surface. The authors propose that transcription-driven phenotypic variation can act as a mechanism of antibody evasion akin to programmed recoding, facilitated by inheritable RNA sequence or structural motifs that promote errors.
The study applied a consensus strategy within a high-throughput sequencing framework to determine the transcription error profile of influenza HA mRNA. The consensus approach is intended to distinguish true transcription-derived sequence variants from sequencing artifacts by requiring reproducible evidence across reads or technical replicates. Details reported in the source note the use of this sequencing-based method to quantify single-site error frequencies along the HA transcript. Specific experimental parameters, sample sources, sequencing depth, validation assays, and exact bioinformatic thresholds used for consensus calling were not reported in the source summary.
Single-site transcription errors were not uniform: different positions in HA mRNA exhibited widely different error frequencies. The authors identified one transcription error as the most prevalent across the dataset. The reported prevalence for this dominant error is approximately 0.2–0.5% of HA mRNA molecules. Because transcription errors are non-heritable (they do not alter the genomic sequence), they generate phenotypic heterogeneity at the level of expressed HA protein within infected cells and in virion surface proteins assembled from those translation products.
Functional assessment reported in the source indicates that the amino acid replacement encoded by the most prevalent transcription error impairs binding of broadly neutralizing antibodies that target HA. The impairment in antibody binding implies that virions presenting the altered HA form may be less susceptible to neutralization by those antibodies. The source describes this impairment qualitatively; quantitative binding change metrics (for example, fold-reduction in affinity or neutralization titer), the specific antibody clones tested, the exact residue change in HA, and assays used to measure binding were not detailed in the provided summary.
The central implication the authors draw is that frequent transcription errors can create a pool of HA variants at the protein level that escape recognition by broadly neutralizing antibodies. Because the error frequency at the dominant site is on the order of tenths of a percent per mRNA, a substantial fraction of newly synthesized HA molecules — and therefore a measurable proportion of assembled virions — could display the altered epitope. The authors frame this as a mechanism of antibody evasion analogous to programmed recoding: the evading amino acid is not fixed in the viral genome but is reproducibly generated at the mRNA/protein level due to sequence- or structure-dependent error propensity. This model highlights a potential role for transcriptional fidelity and RNA sequence/structure in modulating antigenic presentation and host immune recognition.
The source material is an abstract and metadata from a preprint and therefore lacks many experimental details. The following items were not reported in the provided text and cannot be inferred: the exact nucleotide and amino acid substitution involved; which broadly neutralizing antibodies were tested (names, specificities, or epitopes); quantitative measures of reduced binding or neutralization; experimental systems used for binding assays (e.g., purified protein, pseudovirus, live virus); replication of findings across strains or isolates; the sample size and sequencing depth; and controls used to exclude artefactual sequencing or reverse transcription errors. The preprint status is explicitly noted; peer review may alter interpretation or add methodological details.
The work declares funding from Instituto de Salud Carlos III (IHRC22/00004). The authors declared no competing interests. This report is available as a preprint on bioRxiv and has not been certified by peer review. Readers should consider the preliminary nature of the findings and consult the full preprint for complete methods and data.
Notes
All facts and numerical values in this summary are taken from the source abstract and article metadata. Where the source did not report specific experimental parameters or quantitative outcomes, this summary states that such details were not reported in the source.