TUGging at heartstrings: control of aortic valve calcification by non-coding RNA

This editorial refers to ‘LncRNA TUG1 sponges miR-204-5p to promote osteoblast differentiation through upregulating Runx2 in aortic valve calcification’, by C. Yu et al., pp. 168–179. It is long known that only a tiny portion (2–3%) of the human genome codes for protein. The remaining part of the genome was thought to be non-functional for a long time and was referred to as ‘Junk-DNA’. Since recent years, it is clear that most of this ‘Junk-DNA’ is actually transcribed into RNA and many of these so-called non-coding RNAs have now been shown to be involved in key cellular processes.1 In this issue of Cardiovascular Research, Yu et al.2 show that one of these non-coding RNAs, the long non-coding RNA (lncRNA) TUG1, contributes to aortic valve calcification. Taurine up-regulated 1 (TUG1) does this by promoting the differentiation of valve interstitial cells (VICs) into osteoblast-like cells, which leads to calcifications.3 The mechanism by which TUG1 regulates calcifications is intriguing: TUG1 induces the expression of Runx2, a transcription factor that induces osteoblast differentiation,4 by inhibiting a microRNA (miR-204) that is known to reduce the expression of Runx2.5,6 MicroRNAs (miRs) are also non-coding RNAs, but are much smaller than lncRNAs at about 18–21 nucleotides in length, and inhibit gene expression by binding to mRNAs.7 Importantly, single miRs can have multiple target RNAs, thereby regulating entire networks of genes, but these target RNAs also compete for the binding of miRs. This latter mechanism is used by TUG1 that ‘sponges’ miR-204, preventing it from binding to Runx2. In fact, miR-204 was shown to be ‘sponged’ by another lncRNA, MALAT1, before,8 thereby regulating Smad4 and subsequent valve calcification (Figure 1). These mechanisms apparently act synergistically. Non-coding RNA control of aortic valve calcification. The lncRNAs TUG1 and MALAT1 are induced by calcification induction and repress the levels of miR-204. This in turn leads to an induction of Runx2 and Smad4 that are key transcription factors inducing a calcification-promoting gene expression program. One of the open questions is how miR-204 is degraded by TUG1 (or MALAT1). Many papers showing ‘sponging’ by lncRNAs show an actual reduction in miR levels, which seems to indicate that the ‘sponged’ miR is actively degraded, rather than just sequestered. This seems to be a common observation and there must be a general mechanism behind this, but apart from some viral RNAs inducing miR degradation,9 there is relatively known about this mechanism. Interestingly, lncRNAs are also very often found to act as miR ‘sponge’, giving the impression that this is a very widespread function of lncRNAs in general. However, given the high abundance of miRs and relatively low abundance of lncRNAs, it is unlikely that this is indeed a general function of lncRNAs.10 Moreover, many lncRNAs reside in the nucleus, whereas mature miRs are present in the cytoplasm. Perhaps the apparent overrepresentation in the literature of lncRNAs functioning as ‘sponge’ for miRs is rather the result of non-coding RNA researchers focusing on non-coding RNA mediated mechanisms (IE miR mechanisms). Nonetheless, the lncRNA TUG1 is very interesting, as it is highly expressed in VICs, in comparison to valve endothelial cells (VECs)2 while it is already one of the highest expressed lncRNAs in endothelial cells in culture.11 This likely makes TUG1 one of the highest expressed lncRNAs in VICs and potentially very important in controlling VIC phenotype. How the expression of TUG1 is regulated in VICs was not addressed in the paper by Yu et al.,2 but given its pivotal role in controlling VIC phenotype it is highly relevant as one could exploit it to control TUG1 expression therapeutically. Furthermore, in different cell types TUG1 was shown to induce EMT, proliferation and migration.12 These processes likely contribute to the robust induction of calcification by TUG1. Finally, inhibition of TUG1 as therapy to treat or prevent calcifications seems promising, but should be accompanied by local application of TUG1 inhibitors to avoid TUG1 inhibition elsewhere in the body. Inhibition of lncRNAs can be achieved with RNase H-inducing antisense DNA oligonucleotides, so-called gapmeRs or with conventional siRNAs. Alternatively, one could envision the delivery of miR-204, as this seems to play a more central role in calcifications (Figure 1). Conflict of interest: none declared. The opinions expressed in this article are not necessarily those of the Editors of Cardiovascular Research or of the European Society of Cardiology.