A basic notion of modern systems biology is that biological functions are performed by groups of genes that act in an interdependent and synergic way. This is particularly true for regulatory processes for which it is by now mandatory to assume a “network” point of view.
Among the various important consequences of this approach, a prominent role is played by the notion of “network motifs”. The idea is that a complex network (say a regulatory network) can be divided into simpler, distinct regulatory patterns called network motifs, typically composed of three or four interacting components that are able to perform elementary signal processing functions. Network motifs can be thought of as the smallest functional modules of the network and, by suitably combining them, the whole complexity of the original network can be recovered.
In this paper we shall be interested in “mixed” network motifs involving both transcriptional (T) and post-transcriptional (PT) regulatory interactions, and in particular we shall especially focus our attention on the mixed feed-forward loops. Feed-forward loops (FFLs) have been shown to be one of the most important classes of transcriptional network motifs.1,2
The major goal of our work is to extend them to those also including post-transcriptional regulatory interactions.
Indeed, in the last few years it has become more and more evident that post-transcriptional processes play a much more important role than previously expected in the regulation of gene expression.
Among the various mechanisms of post-transcriptional regulation, a prominent role is played by a class of small RNAs called microRNAs (miRNAs), reviewed in refs. 3 and 4
. miRNAs are a family of ~22 nt small non-coding RNAs, which negatively regulate gene expression at the post-transcriptional level in a wide range of organisms. They are involved in different biological functions, including developmental timing, pattern formation, embryogenesis, differentiation, organogenesis, growth control and cell death. They certainly play a major role in human diseases as well.5,6
Mature miRNAs are produced from longer precursors, which in some cases cluster together in so-called miRNA “transcriptional units” (TU),7
and their expression is regulated by the same molecular mechanisms that control protein-coding gene expression. Even though the precise mechanism of action of the miRNAs is not well understood, the current paradigm is that in animals, miRNAs are able to repress the translation of target genes by binding, in general, in a Watson–Crick complementary manner to 7 nucleotides (nts) long sequences present at the 3′-untranslated region (3′-UTR) of the regulated genes. The binding usually involves nts 2–8 of the miRNA, the so-called “seed”. Often, the miRNA binding sites at the 3′-UTR of the target genes are over-represented.8–14
All these findings, in addition to the large amount of work related to the discovery of transcription factor binding sites (for a recent review, see for instance ref. 15
), suggest that both transcriptional and post-transcriptional regulatory interactions could be predicted in silico
by searching over-represented short sequences of nts present in promoters or 3′-UTRs, and by filtering the results with suitable evolutionary or functional constraints.
Stemming from these observations, the aim of our work was to use computational tools to generate a list of feed-forward loops in which a master transcription factor (TF) regulated a miRNA, and together with it, a set of target genes (see ). We performed a genome wide “ab initio” search, and we found in this way a total of 638 putative (merged) FFLs. In order to investigate their biological relevance, we then filtered these circuits using three selection criteria: (I) GeneOntology enrichment among the joint targets of the FFL, (II) independent computational evidence for the regolatory interactions of the FFL, extracted from the ECRbase, miRBase, PicTar end TargetScan databases, and (III) relevance to cancer of the FFL as deduced from their intersection with the Oncomir and Cancer gene census databases.
Feed-forward loops. (a) Representation of a typical mixed feed-forward loop (FFL) analyzed in this work. In the square box, TF is the master transcription factor; in the diamond-shaped box miR represents the microRNA involved in the circuit, while in (more ...)
In a few cases some (or all) of the regulatory interactions that composed the feed-forward loop were found to be already known in the literature, with their interplay in a closed regulatory circuit not noticed, thus representing an important validation of our approach. However, for several loops we predicted new regulatory interactions, which represent reliable targets for experimental validation.
Let us finally notice that in this work we only discuss the simplest non-trivial regulatory circuits (feed-forward loops). However, our raw data could be easily used to construct more complex network motifs. For this reason, we make them accessible to the interested investigators as ESI.†