Herpes simplex virus type 1 (HSV-1) diversity was greater between newborns than within the same newborn at the consensus-genome level, while minor-variant counts did not differ significantly by disease severity, according to a study published in The Journal of Infectious Diseases.
Researchers analyzed 30 residual diagnostic samples from 10 newborns who presented for diagnosis and treatment at a single pediatric hospital in the northeastern United States between 2009 and 2016. Four newborns had skin, eye and mouth disease, while six had severe invasive disease involving the central nervous system, disseminated disease or both. Two patients with severe disease died. The cohort was predominantly male, and all patients with a documented gestational age were full term.
Samples were collected from day 6 through day 40 of life from multiple body sites, including the skin, blood and eyes, as well as other internal and external sites. Fourteen of the 30 samples from nine patients yielded viral cultures. Six cultures were from patients with mild disease and eight cultures were from patients with severe disease. Samples that could be cultured had higher viral genome copy numbers than samples that could not be cultured (P < .0001).
Researchers performed deep sequencing and assembled consensus genomes from all 44 neonatal HSV-1 samples, including residual diagnostic samples and their cultured isolates. Genomic analysis showed that samples from the same newborn were generally grouped together, while samples from different newborns reflected the broader known genetic diversity of HSV-1.
Researchers also quantified minor variants occurring at frequencies between 2% and 50%. The average number of minor variants did not differ significantly between samples from newborns with mild and severe disease. Minor-variant counts differed between skin and eye samples (P = .0193) and between skin samples and samples categorized as other sites (P = .0316). Direct patient samples that could not be cultured had significantly more minor variants than either culturable direct samples or the resulting cultures.
Six of the 10 patients had samples with more than 900 minor variants, which the researchers reported was above levels typically detected in adult samples. Polymerase chain reaction amplification during sequencing could introduce bias in minor-variant detection, so researchers used read deduplication to reduce this potential effect.
The 14 cultured isolates produced a range of plaque sizes and morphologies. In an analysis of 100 plaques per isolate, researchers found no statistically significant pattern in plaque size by disease diagnosis or sample source. Among six representative isolates evaluated in multistep growth assays, one isolate from a patient with severe disease produced significantly higher titers than all other isolates at 24 and 48 hours in Vero cells (P < .05). No isolate differed significantly from all other isolates in the human fibroblast or keratinocyte cell models.
Viral protein expression also differed among isolates from different newborns. Researchers observed differences in several viral proteins at early and late time points, but those differences did not correlate with disease severity or in vitro growth characteristics. Among three newborns with multiple cultured isolates, samples collected from different sites or times showed highly similar protein-expression profiles.
"In both the viral consensus genomes and the protein expression assays, between-host differences exceeded within-host differences," wrote Holly A. Passetti of Pennsylvania State University, and colleagues.
Researchers further assessed neurovirulence by intracranially inoculating mice with 1,000 plaque-forming units of HSV-1 and monitoring them for 14 days. Survival ranged from 0% to 50% depending on the isolate. Survival was greater than 25% among mice inoculated with most isolates from newborns with mild disease and below 17% among mice inoculated with most isolates from newborns with severe disease. However, researchers found no statistically significant correlation between human disease diagnosis and murine neurovirulence as measured by survival or brain viral titers.
In a differentiated human neuron-like cell line, isolates from patients with mild disease tended to replicate less than isolates from patients with severe disease. Two isolates showed patterns in the neuron-like cells similar to those observed in the murine model. The researchers said additional evaluation of genotype-phenotype associations will require comparisons of viral recombinants differing at key loci and/or testing strains through a peripheral infection route.
The researchers noted that the number of clinical source cases was limited and that immune measurements were unavailable for the residual samples. They also noted the potential for polymerase chain reaction amplification to affect minor-variant detection and said additional studies are needed to determine whether these samples reflect coinfection or rapid viral evolution in an immune-naive host.
The study received funding from the National Institutes of Health through the National Institute of Allergy and Infectious Diseases and the National Institute of Neurologic Disorders and Stroke. The researchers reported no conflicts of interest.
