Abstract
This article serves as a formal correction to the 2016 research paper titled "Middle East Respiratory Syndrome Coronavirus Intra-Host Populations Are Characterized by Numerous High Frequency Variants" (DOI: 10.1371/journal.pone.0146251). While the fundamental conclusions regarding the high genetic diversity of MERS-CoV within individual hosts remain valid, this update provides essential corrections to genomic coordinates, residue numbering, and mutation frequency data for specific viral variants.
Introduction
In the study of viral evolution, the concept of a "quasispecies" is central. RNA viruses like the Middle East Respiratory Syndrome Coronavirus (MERS-CoV) do not exist as a single, static genetic sequence within an infected host. Instead, they form a dynamic population of closely related variants. Identifying these intra-host single nucleotide variants (iSNVs) is critical for understanding how the virus adapts to host immune pressures, shifts between species, and potentially develops resistance to therapeutic interventions.
This correction addresses specific errors in the mapping of these variants, ensuring that the genomic data provided to the scientific community is accurate for future structural modeling and evolutionary analysis.
Correcting the Genomic Map
The most significant revision in this correction concerns the ORF5 protein region. In the original publication, a key stop codon was incorrectly identified at residue 109. The corrected data confirms that this stop codon actually occurs at residue 107 (nucleotide position 27162).
This variant was detected at a frequency of 12% in the C2D5 sample (a dromedary camel host model at five days post-infection). Notably, this was the only sample with sufficient sequencing depth (greater than 100x coverage) at this specific site to reliably identify the mutation, highlighting the technical challenges inherent in deep-sequencing viral populations.
Seed Stock Heterogeneity
A critical insight clarified by this correction is the nature of the initial viral "Seed" stock used for the infections. The original analysis suggested a higher degree of genetic purity than was actually present. The corrected data reveals that the Seed stock already contained iSNVs at nucleotides 307–313 within the nsp1 open reading frame at a 4% frequency. This indicates that a secondary genotype was present from the start of the experiments, which must be accounted for when tracking the emergence of new variants during the course of infection.
Widespread Nonsynonymous Mutations
The corrected data provides a comprehensive landscape of nonsynonymous mutations—those that change the amino acid sequence of viral proteins—across the MERS-CoV genome. These variants are not localized but are distributed across several key genes:
- Non-structural proteins (nsps): For instance, the I631L mutation in nsp2 was found at high frequencies (87% to 100%) across multiple host samples, suggesting rapid selection or dominance of this variant.
- Structural Proteins: Significant variation was confirmed in the Spike (S), Membrane (M), and Nucleocapsid (N) genes.
- Spike Protein Diversity: Mutations such as E565D and A1007T in the Spike protein are of particular interest, as this protein is the primary target for neutralizing antibodies and vaccine development.
Conclusion and Technical Considerations
The corrections provided here underscore the importance of high-depth sequencing in viral genomics. Detecting low-frequency variants (those occurring at >1% but <50% frequency) requires rigorous quality control and extensive coverage. As the authors note, gaps in sequencing coverage can lead to an underestimation of true genetic diversity.
By refining the "map" of MERS-CoV intra-host variants, this correction ensures that researchers can accurately model the structural impacts of these mutations and better predict the evolutionary trajectory of this significant human pathogen.
Summary updated