Nucleic Acids Research
Web Servers Issue 2010
What are IREs?
Iron is an essential nutrient required by almost every organism. Its capacity to exchange electrons makes it essential for fundamental cell functions, like DNA synthesis, transport of oxygen and the respiratory chain. However, it is also a potential catalyst for chemical reactions involving free-radical formation and subsequent cell damage and death. Therefore, cellular iron levels have to be carefully regulated inside the cells.
Intracellular iron homeostasis is mainly regulated post-transcriptionally by the IRE/IRP regulatory system . The iron regulatory proteins (IRP1 and IRP2) can recognize a cis-regulatory mRNA motif termed iron responsive element (IRE), a conserved RNA element located in the untranslated regions (UTR) of mRNAs that encode proteins involved in iron metabolism.
A canonical iron responsive element structure is composed of a 6-nucleotide apical loop (5’-CAGWGH-3’; whereby W stands for A or U and H for A, C or U) on a stem of five paired nucleotides, a small asymmetrical bulge with an unpaired cytosine on the 5’strand of the stem, and an additional lower stem of variable length .
Iron regulatory proteins/iron responsive element interactions regulate the expression of the mRNAs encoding proteins for iron acquisition (transferrin receptor 1, Tfrc; divalent metal transporter 1, Dmt1), iron storage (ferritin H, Fth1; ferritin L, Ftl), iron utilization (erythroid 5-aminolevulinic acid synthase, e-Alas; mitochondrial aconitase, Aco2; Drosophila succinate dehydrogenase, Sdh), iron export (ferroportin, Fpn), oxygen availability (hypoxia-inducible factor 2 alpha, Hif2aplha) and cell cycle (CDC14A) [5,6,9]. The mRNAs of Fth1, Ftl, e-Alas, Aco2, Sdh, Fpn and Hif2alpha contain one single IRE in their 5´UTRs. The mRNA of Slc11a2 and CDC14A contain also one single IRE but in its 3´UTR and Tfrc mRNA is so far the only known mRNAs with multiple (five) IREs, all of them located in its 3´UTR (Fig. 1).
Figure 1. Human and mouse sequences of iron-responsive elements.
In response to fluctuations in the level of the labile iron pool, IRPs bind to IREs. Depending on the IRE location, they regulate gene expression by different mechanisms. Both IRPs inhibit translation initiation when bound to 5’UTR IREs, whereas their association with the 3’UTR IREs of the Tfrc mRNA mediates mRNA stabilization by preventing an endonucleolytic cleavage [11,12]. Therefore, the IRPs act as key regulators of cellular iron homoeostasis as a result of the expression control of a number of iron metabolism-related genes (Fig. 2).
Figure 2. The iron responsive-element/iron-regulatory protein (IRE/IRP) network.[from ref. 9]
How are IREs predicted?
We used an in-house Perl program to identify IRE-like sequences in gene transcripts. First, the sequence is screened to find the apical loop sequence of the IRE corresponding to 18 different motifs and the n8 nucleotide (C in 17 cases and G in one case, motif 18). Then the base pairing sequences of the upper stem and at position n07-n25 are checked allowing the presence of one single mismatch or one. The detection of one bulge nucleotide at the right side of the upper stem or a mismatch in the upper stem or at position n07-n25 is mutually exclusive.
Is there any Score or something I can trust?
According to our experience we have set 3 levels of stringency (high, medium and low) for the prediction of IREs. High and medium stringency are based on previous known and in vivo and/or in vitro well characterized IREs, i.e. the IREs present in ferritin L, ferritin H, TfR1, e-Alas, Aco2, Fpn, dSDH, CDC14A, Dmt1, Gox and Hif2aplha. Low stringent IRE predictions includes IREs with SELEX motifs 3 to 18 with or without one single mismatch or bulge in the upper stem; such IREs have been predicted in novel IRP-interacting mRNAs detected in a recent genome-wide study; however, although some of them have been validated in vitro, the in vivo relevance of those IREs has not yet been studied (Sanchez et al, manuscript in preparation). To validate the predicted IREs reported by our program, we strongly recommend studying the in vitro functionality of the predicted IRE by competitive EMSA experiments as previously reported [5,6]. The minimum free energy of the predicted IRE motif calculated by RNAfold webserver is also an indicator of IRE reliability. Predicted IREs by SIREs that fail to fold with a negative free energy most probably represent non bona fidea IREs.
Which version of the server should I use: the interactive or the batch one?
That depends on the number of sequences and the level of detail you need in your results. If you have more that 50,000 residues in your sequences you will have to use the batch server. That will provide you with the list of predicted IREs, but not the graphical representation of the ires or their predicted folding. From there, you will be able to submit up to 10 sequences to a more detailed analysis. You can see an example output here. Even if you have a short number of sequences you may want to use the batch server, since it is about 10 time faster than the interactive one.