NEW YORK (GenomeWeb) – Recent research has shown that an enzyme involved in adenosine-to-inosine (A-to-I) RNA editing can affect microRNA biogenesis and processing, yet its potential role in disease has been unclear. But new research out of the MD Anderson Cancer Center suggests that this enzyme is manipulated in one kind of cancer, boosting the expression of an oncogenic miRNA to promote tumor progression and metastasis.
Though the findings were limited to melanoma, researchers involved in the study believe this mechanism is likely to be found in other tumor types. The data, which appeared in Nature Cell Biology, may also point to a new opportunity for treating cancer.
A-to-I editing is the most common form of RNA editing and is mediated by the activity of adenosine deaminases acting on RNA (ADAR) enzymes. Such RNA editing has been shown to occur in regulatory RNAs such as miRNAs, and in 2013 a team from the Wistar Institute reported that one specific ADAR enzyme, ADAR1, interacts directly with Dicer in an RNA-binding-independent manner, increasing miRNA processing, RISC loading of miRNAs, and silencing of target RNAs.
That same year, MD Anderson's Menashe Bar-Eli and collaborators published a report showing that ADAR1 is often reduced in metastatic melanomas and that knocking down the enzyme increased tumorigenicity in vivo.
In that work, they also demonstrated that ADAR1 controls the expression of over 100 miRNAs associated with tumor growth and that melanoma cells specifically can inhibit ADAR1 by overexpressing two miRNAs — miR-17 and miR-432 — that directly target the ADAR1 transcript.
Previously, Bar-Eli has reported that the transcription factor cyclic AMP-responsive element-binding protein, or CREB, promotes the malignant progression of melanoma by inhibiting expression of the tumor suppressor AP-2alpha.
As part of an effort to identify other downstream targets of CREB, he and his team silenced the transcription factor in highly metastatic melanoma cells, which were then subjected to gene-expression profiling.
As detailed in Nature Cell Biology, the researchers found that when CREB was inhibited, ADAR1 expression increased. They also discovered that ADAR1 expression, as well as its RNA-editing activity, decreased with melanoma progression.
The MD Anderson group delivered ADAR1-silenced melanoma cells subcutaneously into mice and observed a significant increase in tumor growth, as well as lung metastases, compared with cancer cells in which ADAR1 activity was not suppressed. Meanwhile, overexpression of ADAR1 was associated with decreased growth of melanoma cells and the inhibition of spontaneous metastasis.
Hypothesizing that the effects of ADAR1 suppression might be linked to its influence on miRNAs, the team ran miRNA expression arrays in two different sets of metastatic melanoma cells: ones with low metastatic potential in which ADAR1 was suppressed, and highly metastatic ones in which CREB was silenced. The scientists identified changes in the expression of hundreds of miRNAs between the two arrays, with 12 being altered in both. Of these, the team discovered that A-to-I editing takes place in three: miR-378-3p, miR-324-5p, and miR-455-5p.
The investigators focused on miR-455-5p as it was found to contain two ADAR1-mediated A-to-I RNA editing sites, Bar-Eli told GenomeWeb, and then sought to identify what differences there were between the edited and wild-type forms of the miRNA.
As expected, the edited form, as would be found in non-metastatic melanoma cells, and the non-edited form, which is found in metastatic cells, targeted completely different sets of genes, he explained.
Among those genes targeted by wild-type miR-455-5p was CPEB1, a known tumor suppressor
The MD Anderson researchers next compared the activity of wild-type and edited miR-455 in vitro and found that their biological functions were different, with each recognizing a different set of genes. One of the genes identified as a target of wild-type miR-455-5p was CPEB1, a known tumor suppressor.
Further experimentation demonstrated that when ADAR1 was overexpressed in melanoma cells, thereby lowering levels of wild-type miR-455 and raising levels of edited miR-455, CPEB1 expression increased two-fold at the mRNA level and 1.5-fold at the protein level. When ADAR1 was silenced, CPEB1 expression dropped. Additionally, CPEB1 levels decreased with overexpression of wild-type miR-455, but increased with expression of the edited form of the miRNA.
To explore the therapeutic potential of this finding, Bar-Eli, along with MD Anderson's Anil Sood and others, tested the effects of delivering wild-type and edited versions of miR-455-5p, as well as miR-455-5p antagonists, to mouse models of melanoma. The unedited miRNA enhanced melanoma growth and metastasis, while the edited version and antagomir inhibited these features.
To Bar-Eli, the findings not only suggest miR-455-5p's potential as a therapeutic target in melanoma, but in other cancers, as well.
Noting that other groups have linked RNA editing of miRNAs to cancer aggressiveness, including one team that reported such an association in glioblastoma in 2012, "I don't think it's going to be confined to melanoma," he said.
Sood agreed, stating that he'd be "shocked if it didn't occur in other tumors." Should that be the case, he added, miR-455 could prove to be a valuable drug target for slowing or even preventing tumor growth across a variety of cancers.
Still, such a therapeutic approach would require significant additional testing to better understand the global role of miR-455, Sood cautioned.
"You don't want to antagonize a microRNA in one place and have good effect, and you end up finding it has deleterious effects in other sites," he told GenomeWeb.