Abstract
The anucleate platelets play a critical role in the formation of thrombi and prevention of bleeding. In recent years, next-generation RNA sequencing (RNA-seq) has proven very useful in shedding light on the specifics of the platelet transcriptome. For example, RNA-seq of the long RNAs in platelets has revealed many non-coding RNAs (ncRNAs) as well as a diverse set of protein-coding genes whose mRNAs are highly correlated amongst individuals but only weakly linked to the currently available platelet proteome. By comparison, the short RNA transcriptome has not been as thoroughly characterized. As a matter of fact, these studies have so far focused on the 100’s of microRNAs (miRNAs) that are present in platelets leaving large swaths of the short RNA-ome uncharacterized.
To gauge the complexity of the platelet short RNA-ome we performed short RNA-seq of leukocyte-depleted platelets from 10 healthy males (5 white and 5 black). The sequencing was done on the SOLiD 5500 XL platform and generated over 1.5 billion sequenced reads.
To comprehensively characterize the complete short RNA-ome we only considered sequence reads that mapped on the genome without any mismatches but allowed a read to map to as many as 10,000 locations within the genome. This approach gave us the ability to simultaneously examine both the uniquely-present and the repeat-derived expressed elements of the genome. Using this approach, we were able to map ~50% of the sequenced reads.
We found that for ~55% of the mapped reads their sequences are present at multiple genomic locations whereas the remaining ~45% originated from unique locations. Of the RNAs with unique genomic origins: ~50% correspond to miRNAs (with miR-223-3p being the most abundant miRNA across all 10 individuals), ~20% originate from various classes of repeat elements, and, the remaining 30% correspond to non-annotated regions of the genome that were non-annotated a of Release 75 of the ENSEMBL database. By comparison, of the RNAs with ambiguous genomic origins: ~20% belong to miRNAs (with miR-103a-3p, a miRNA present in two locations in the genome, being the most abundant miRNA across all 10 individuals) and ~60% correspond to various classes of repeat elements (with members of the HY4 scRNA ncRNAs accounting for nearly a third of all sequence reads). These findings make it evident that the platelet transcriptome has a considerable richness in short RNAs that arise from repetitive elements.
To further characterize those RNAs that map to regions of the genome that are not currently annotated, we considered the possibility that they may be novel miRNAs. Using the miRDeep2 algorithm, we sought novel miRNAs among the uncharacterized transcripts and identified 47 of them; the sequences for 18 of these 47 appear at multiple genomic locations in analogy to miR-103/107, miR-19a/19b, etc.
Lastly, as our ten samples represented two races, we hypothesized that a subset of the identified sequences would be differentially expressed between the two groups. Using DESeq2, we identified over 157 sequences to be differentially expressed. The most highly differentially expressed sequences corresponded to a miRNA and a repeat element.
In summary, our RNA-seq analyses have revealed a very diverse spectrum of platelet short RNAs that transcends the miRNA category. Indeed, we find that short transcripts that have their source in genomic loci that have not been previously discussed or analyzed in the platelet context represent a very significant portion of all short RNAs in platelets. This in turn highlights an unanticipated richness, and presumably commensurate complexity, for the platelet transcriptome. While the role of these novel non-protein coding short RNAs is currently unknown it is expected that at least some of them may be of functional significance. Consequently, they could contribute to processes beyond thrombosis and hemostasis and may permit a better understanding of the molecular mechanisms that regulate platelet physiology.
Disclosures
No relevant conflicts of interest to declare.
Expression levels of the tetratricopeptide repeat protein gene
ttc39b
covary with carotenoid-based skin colour in cichlid fish