While there have been almost 2,000 human microRNAs discovered to date, new research out of Spain's Center for Genomic Regulation (CRG) suggests that there are substantially more yet to be found, some of which may be human- and cell-specific.
Since the first miRNAs were identified in the early 2000s, improvements in sequencing technology have increased the rate at which the small, non-coding RNAs are identified. Still, there have been no attempts to pin down just how many miRNAs exist in the human genome since 2003.
As such, investigators from CRG, along with collaborators from a number of other institutions, set out to see they could address this knowledge gap, reporting their findings this month in a study appearing in Genome Biology.
According to Marc Friedlander, a postdoc in the lab of CRG investigator Xavier Estivill and lead author of the paper, he has long been interested in the miRNA field and, in analyzing sequencing data from different human tissues, would routinely come across new miRNAs.
At the same time, "there was very limited overlap between the microRNAs we were discovering, for instance, in liver and in brain," he told Gene Silencing News. "This led us to think that the actual number of human microRNA genes is much bigger than what is currently annotated in the databases."
In order to confirm this hypothesis, Friedlander — who is starting his own lab at the Science for Life Laboratory in Stockholm later this year — and his colleagues developed software to comb through nearly 100 human small RNA sequencing datasets representing about 700 million sequencing reads and representing primary tissues and cell cultures.
The datasets were pooled and analyzed. Specifically, reads were mapped to the genome, and the researchers considered only perfect matches while discarding reads that mapped to more than five genomic locations.
Sequences not detected in at least two distinct datasets were discarded, as were sequences overlapping known miRNA, ribosomal RNA, or transfer RNA genes. "Only sequences flanked by RNA hairpin structures similar to miRNA precursors were retained, and reads from the 94 datasets were mapped to the RNA hairpin structures, and those with mappings inconsistent with Dicer processing were discarded," they added.
The result was 2,469 candidate novel miRNA hairpins, of which 1,098 were validated by in-house experimentation and published data.
Notably, nearly 300 of the novel miRNAs were "robustly expressed" in a neuronal cell system and were regulated during induced cell differentiation, which suggests that they are not "chance side-products of housekeeping RNAs or spurious transcripts, but are in fact linked to core regulatory processes of the cells," the team wrote in its paper. These miRNAs were also up- or downregulated when the miRNA biogenesis factors Dicer, Drosha, DGCR8, or Ago2 were silenced.
Meantime, in a kidney cell system, 400 of the novel miRNAs were found to interact with DGCR8 at transcript positions that "suggest miRNA hairpin recognition, and 1,000 of the new miRNA candidates interact with Ago1 or Ago2, indicating that they are directly bound by miRNA effector proteins."
Interestingly, the newly identified miRNAs were found to be lowly, but specifically, expressed. While this might indicate that many of the small RNAs are not functional, Friedlander pointed out that this could also mean that they are highly functional but only in specific cells.
Such a situation has already been observed in nematodes, where the miRNA lsy-6 is specifically expressed in a single neuron and plays a critical role in controlling left/right neuronal asymmetry.
"It is entirely feasible that the same might hold true for microRNA expression in human neurons," Friedlander said.
The majority of the newly discovered miRNAs were also found to be evolutionarily young and overrepresented in the human brain — findings that could indicate their importance in speciation and defining cellular identity.
In not being deeply conserved, it may be that the miRNAs are simply not functional, with hairpins entering the miRNA biogenesis pathways but not impacting the transcriptome in a meaningful way.
But while deeply conserved miRNAs are likely to be function, the reverse isn't necessarily true, the scientists wrote in their paper. Indeed, examples exist of species-specific miRNAs that have well-defined functions such as miR-941, which is human-specific and has been linked to cellular differentiation.
"It is very interesting to speculate that some of [the young miRNAs] might actually have roles," Friedlander said, adding that his team believes some may be involved in speciation.
As for the new miRNAs' overrepresentation in human brain tissue, Friedlander suggested that this could indicate that the non-coding RNAs have important cell-specific functions or may influence cell identity.
"The human brain has an incredible diversity of neuronal cell types," he said. "It is interesting to speculate that some of the new microRNAs are important in actually defining the identity of specific brain cells.
Answering such questions, however, is challenging given the absence of technologies for sequencing small RNAs in single cells, he added. But such methods are under development, and in anticipation of their availability, it continues to be important to build a saturated catalog of annotated miRNAs to enable future research, he said.