Ebola virus (EBOV) is a virulent pathogen, notorious for inducing life-threatening hemorrhagic fever, that has been responsible for several outbreaks in Africa and remains a public health threat. Yet, its pathogenesis is still not completely understood. Although there have been numerous studies on host transcriptional response to EBOV, with an emphasis on the clinical features, the impact of EBOV infection on post-transcriptional regulatory elements, such as microRNAs (miRNAs), remains largely unexplored. MiRNAs are involved in inflammation and immunity and are believed to be important modulators of the host response to viral infection. Here, we have used small RNA sequencing (sRNA-Seq), qPCR and functional analyses to obtain the first comparative miRNA transcriptome (miRNome) of a human liver cell line (Huh7) infected with one of the following three EBOV strains: Mayinga (responsible for the first Zaire outbreak in 1976), Makona (responsible for the West Africa outbreak in 2013–2016) and the epizootic Reston (presumably innocuous to humans). Our results highlight specific miRNA-based immunity pathways and substantial differences between the strains beyond their clinical manifestation and pathogenicity. These analyses shed new light into the molecular signature of liver cells upon EBOV infection and reveal new insights into miRNA-based virus attack and host defense strategy.

MicroRNAs (miRNAs) are important gene regulatory molecules involved in a broad range of cellular activities. Although the existence and functions of miRNAs are clearly defined and well established in eukaryotes, this is not always the case for those of viral origin. Indeed, the existence of viral miRNAs is the subject of intense controversy, especially those of RNA viruses. Here, we characterized the miRNA transcriptome of cultured human liver cells infected or not with either of the two Ebola virus (EBOV) variants: Mayinga or Makona; or with Reston virus (RESTV). Bioinformatic analyses revealed the presence of two EBOV-encoded miRNAs, miR-MAY-251 and miR-MAK-403, originating from the EBOV Mayinga and Makona variants, respectively. From the miRDB database, miR-MAY-251 and miR-MAK-403 displayed on average more than 700 potential human host target candidates, 25% of which had a confidence score higher than 80%. By RT-qPCR and dual luciferase assays, we assessed the potential regulatory effect of these two EBOV miRNAs on selected host mRNA targets. Further analysis of Panther pathways unveiled that these two EBOV miRNAs, in addition to general regulatory functions, can potentially target genes involved in the hemorrhagic phenotype, regulation of viral replication and modulation of host immune defense.

Small RNA sequencing (sRNA-Seq) approaches unveiled sequences derived from longer non-coding RNAs, such as transfer RNA (tRNA) and ribosomal RNA (rRNA) fragments, known as tRFs and rRFs, respectively. However, rRNAs and RNAs shorter than 16 nt are often depleted from library preparations/sequencing analyses, although they may be functional. Here, we sought to obtain a complete repertoire of small RNAs by sequencing the total RNA from 11 samples of 6 different eukaryotic organisms, from yeasts to human, in an extended 8- to 30-nt window of RNA length. The 8- to 15-nt window essentially contained fragments of longer non-coding RNAs, such as microRNAs, PIWI-associated RNAs (piRNAs), small nucleolar RNAs (snoRNAs), tRNAs and rRNAs. Notably, unusually short RNAs < 16 nt were more abundant than those >16 nt in bilaterian organisms. A new RT-qPCR method confirmed that two unusually short rRFs of 12 and 13 nt were more overly abundant (~3-log difference) than two microRNAs. We propose to not deplete rRNA and to reduce the lower threshold of RNA length to include unusually short RNAs in sRNA-Seq analyses and datasets, as their abundance and diversity support their potential role and importance as biomarkers of disease and/or mediators of cellular function.