The latest study by the Functional Annotation of the Mammalian genome (FANTOM) group, an international consortium led by Japan's RIKEN Institute, has compiled the first extensive atlas of microRNA expression in human primary cells, which could help development of new cancer treatments.

MicroRNAs are small non-coding RNA molecules containing around 22 nucleotides. MicroRNA is responsible for RNA silencing of gene expression and regulation of post-transcriptional gene expression.

"It is now recognized that most of the human genome [is] transcribed, with most RNAs generated being non-coding," study leader Michiel De Hoon, a unit leader at the RIKEN Center for Life Science Technologies (CLST) in Yokohama, told BioWorld.

"For most non-coding RNAs, especially the long ones, the functions remain unknown," he noted. "We have also found thousands of genomic loci that generate short RNAs, which may be microRNAs but could very well be a different class of RNAs."

The human body comprises multiple different cell types, each of which has very different functions and behaviors, despite the genome sequence of almost all of an individual's cells being identical.

This variation in cells' functional roles is achieved through an intricate regulatory network comprising regulatory proteins and RNAs such as microRNAs. Dysregulation of such networks plays a major role in disease development, in particular, cancer.

"The most obvious causes of dysregulation are amplification of DNA loci encoding microRNAs or loss of such loci; mutations in the microRNA sequence or promoter region; or amplification, deletion or mutations in transcription factors regulating microRNA expression by binding to the promoter region, causing changes in expression level," explained De Hoon.

"We were primarily interested in the microRNA promoter regions and have compiled the largest and most accurate microRNA promoter region atlas to date," in the new study reported Aug. 21 in Nature Biotechnology.

Using RNA samples collected for the 5th edition of FANTOM (FANTOM5), which was led by FANTOM5's principal organizer Yoshihide Hayashizaki, the group sequenced microRNA libraries of hundreds of human samples, including cell types in which the microRNA had not been previously investigated.

"In total we looked at 109 different cell types, some of which had been profiled before, such as blood cell types," said project co-leader Alistair Forrest, professor and head, Systems Biology and Genomics Laboratory, Harry Perkins Institute of Medical Research, Perth, Australia.

"However, through the FANTOM consortium we obtained access to rare cell types which are usually not profiled, because they are difficult to handle in the laboratory. Importantly, most of these were primary cells, not cell lines," said Forrest.

In the earlier stages of FANTOM, the samples had been profiled using Cap Analysis Gene Expression (CAGE), a technology similar to RNA sequencing that was developed at RIKEN, to discover the precise starting site of RNA transcriptions. (See BioWorld Today, Nov. 18, 2015.)

"Mapping CAGE reads accurately identifies the 5' end of the original RNA transcript and therefore its transcription start site," said De Hoon. "In FANTOM5, we have thousands of CAGE libraries and by combining these we could see the 5' end of the primary microRNA transcript, which has low expression levels, so had previously been difficult to map accurately."

Combining these CAGE data with microRNA data allowed creation of an integrated atlas of microRNA expression, as well as mapping the genomic regions controlling microRNA expression in different cell types.

Together, these datasets provide an insight into how these regulators contribute to establishing the unique identity of each cell type in the human body.

The researchers also discovered thousands of new genomic loci producing short RNAs, which may prove to constitute a novel class of regulatory short RNAs.

"We have to keep an open mind about the function of these. Some may be novel microRNAs, but they could also have other functions. For example, we previously found that short RNAs are produced as part of the DNA damage response pathway, but they could be a different class altogether," said De Hoon.

"Because we now know the microRNA promoter regions, we can analyze how microRNA expression is regulated, for example by investigating which transcription factors bind to the promoter. In addition, since microRNA and CAGE expression are correlated, we can use CAGE expression as a proxy for microRNA expression, allowing us to extend the microRNA expression atlas to the thousands of samples in FANTOM5."

On the relevance of these findings to cancer treatments, "some microRNAs, called oncomirs, have clear implications in cancer," said De Hoon. Oncomirs are associated with carcinogenesis, malignant transformation and metastasis.

"For example, microRNA-21 is highly expressed in most cancers. Knowing the microRNA promoter regions can help to understand their dysregulation in cancer, and may provide a pathway to modulate their expression," he said.

Inhibiting microRNA-21 activity in vivo may be a therapeutic strategy for cervical cancer and an anti-microRNA-21 inhibitor has been developed at Tokyo's Keio University. MicroRNAs may also have effects on drug resistance and be useful in combination with other cancer drugs.

"We have made the expression atlas available online [http://fantom.gsc.riken.jp/5/suppl/De_Rie_et_al_2017] and expect to have thousands of users. We believe it will be an essential resource for understanding microRNA regulation and its role in human disease," De Hoon said. "The ultimate goal of FANTOM is to annotate functionally the entire mammalian genome. In FANTOM6, which is currently underway under Piero Carninci, director of Genomic Technologies at RIKEN CLST, we are now studying the function of long non-coding RNAs, which constitute the majority of the human transcriptome and may have a regulatory role."