Veterinary World, EISSN: 2231-0916 Available at www.veterinaryworld.org/Vol.6/Oct-2013/13.pdf REVIEW ARTICLE Open Access MicroRNA: biogenesis and computational target identification: a review

(2013) MicroRNA: biogenesis and computational target identification: a review, Veterinary World 6(10): 761-765.


Introduction
Multiplicity indicates that a single miRNA has many target sites, whereas cooperativity denotes that many MicroRNAs (miRNA) are endogenously produced, miRNAs have common target sites [8].Data generated small noncoding RNAs, play important roles in gene by new generation sequencing technologies along with expression, and have multiple target binding sites in the miRNA biology enhances our understanding of mRNAs resulting in their silencing [1].miRNA evolution of miRNA and their targets [9].expression profiling is now popularized as they are key Biogenesis of miRNA regulators in gene expression networks, have impact on many biological processes and are now proved to be They were originally discovered in Caenorhabditis biomarkers for disease [2].miRNA change the gene elegans as post transcriptional generegulators [4,10].

expression by attaching to complementary base pairs in
In C. elegans, the translation of lin-14 mRNA is the mRNAs which then lead to translational repression regulated by a small non-coding RNA which has many or transcript degradation [3].Translational repression target binding sites in 3'UTR of lin-4 gene [11].All may be the primary mode of action in animals because eukaryotic and several viruses express miRNAs [4,12] of an imperfect duplex formation between miRNA and but no RNA virus except retrovirus has been found to mRNA, whereas in plants they have perfect binding produce miRNA which was confirmed by computasites lead to degradation of mRNA [4].Despite the tional and experimental analysis on the genome of actual underlying mechanism being still poorly hepatitis C virus and yellow-fever virus [13].They understood, miRNAs seem to be major factors for regulate a variety of biological processes like cell regulation of different pathways or processes like growth, tissue differentiation, apoptosis, viral infection development, differentiation, proliferation and cell and some reports also show their role in the death [4].
inflammatory process [4,14].miRNAs also emerged as important regulators of They bind to the 3'UTR region of transcript and metabolism [5].At the 5' end of the miRNA, a 7-8 inhibit translation or degrade mRNA [15].They also nucleotide long region is the primary component of bind to other regions like coding sequence and 5' UTR miRNA target specificity which is termed as the seed but these are less efficient sites of interaction when region [6].There are two types of miRNA i.e. intronic compared to 3'UTR.A single miRNA can regulate the and exonic, the difference in both being only of expression of many genes and modulate many polymerases and RNA splicing components required biological pathways [16].In animals, miRNAs were during their biogenesis [7].miRNAs show phenomena thought to reduce translation of their targets with very of multiplicity or redundancy and cooperativity.less or no influence on mRNA quantity, whereas opposite effects were thought to be true in plants [17].Some miRNA genes show very similar expression profile which are clustered and separated by only a few nucleotides, suggesting that they are monitored by involve a different mechanism [26].However, instead common promoters [18].
of mRNA degradation, it may start a cascade for The biogenesis of miRNA is a well conserved blocking initiation of proteins from binding to the 5´ process (Figure -1).The microRNA producing gene is cap of the mRNA [27].An additional mechanism of first transcribed into primary miRNA, which is further translational repression is by translocation of the processed by a nuclear RNase type III enzyme, Drosha, miRNA:mRNA complex to cytoplasmic foci in the into precursor miRNA [19].The pre-miRNA is then cell, known as processing bodies (P-bodies), after the transported to cytoplasm by the nuclear export factor miRNA:RISC complex binds to the mRNA target [28].Exportin 5 and a cofactor Ran-GTP.In cytoplasm, the Principles of miRNA targets pre-miRNA is again cleaved by Dicer into miRNA: The miRNA target prediction principles used by miRNA (In heteroduplex miRNA is less stable) duplex most of the approaches are almost similar [29] such as [19].From miRNA: miRNA (In heteroduplex miRNA seed sequence complementarity [30], thermodynamic is less stable) duplex, one strand is degraded and other stability, conserved sites etc. is identified by the RNA-induced silencing complex (RISC) [20] which targets the mRNA transcript.
1. Seed sequence complementarity: At the 5'end of the miRNA, a 2-7 nucleotide region called as 'seed sequence' binds within mRNA target site.This Watson-Crick paring of miRNA and mRNA is important factor of target gene prediction [31].Along with this seed pairing, there is pairing at 3' end of miRNA from 13-16 nucleotide (3-4 nucleotides) called as '3'-supplementary' pairing which leads to increased specificity and affinity.Furthermore for compensating a mismatch in seed region, Watson-Crick paring of miRNA and mRNA at 3' end of miRNA is called as '3'-compensatory sites' [31]. in the alignment of the 3'UTR sequences [32].
miRNA nomenclature 3. Thermodynamics: Thermodynamic stability is another filter criteria for target gene prediction.The The miRNA has 3 or 4 prefix letter for a particular more negative free energy of miRNA:target duplex, the species like 'gga' for chicken.The processed final higher energy is required to disrupt this duplex miRNA or mature sequences are nominated as 'miR' in formation.Thus, a RNA duplex is in a thermothe database, whereas the precursor hairpins for mature dynamically more stable state.miRNAs are labeled as 'mir' [21].Orthologous miRNAs differ only in the first letter prefixes like hsa-4.Accessibility of target site: The accessibility of the miR-101 in human and mmu-miR-101 in mouse.
target sites were evaluated based on the secondary However, in case of paralogous sequences whose structure of the target mRNA.The secondary structure of mature microRNAs vary at only one or two positions the mRNA appears also to play an important role.For are given lettered suffixes like gga-miR-10a and ggabinding to the miRNA the target site has to be accessible, miR-10b in chicken genome.
which means it has to be opened and must not interact with other sites within the mRNA [33,34].Two types of

Regulation of translation by miRNA
folds, namely the 'global' fold which was the secondary It is confirmed by many researchers that miRNA structure of the complete coding sequence and the reduces translation of protein-coding genes [22].Some 'regional' fold which was the secondary structure obtained by folding about 220 bases with 100 bases on either sides evidences have also emerged that contradicts the of the target site sequence can be used for target site current dogma of decreased gene translation [23].
accessibility.The MFE of regional folding with These regulatory mechanisms include mRNA degradaconstraint of unpairing at the target site (? G ) tion by removing the cap at 5' end and deadenylation of unpaired compared with that of the native fold (? G ) to assess the poly-A tail of mRNA result in repression of free the free energy of unfolding (? G ). translation [24,25]

Algorithms for miRNA target prediction Types of miRNA target sites
Computational target prediction algorithm which miRNAs down-regulate gene expression mostly is a rule-based algorithm came to existence since by imperfect binding to complementary sites within TargetScan [39] was proposed in 2003 and is still mRNA which lead to reduction of translation, mRNA among the most popular algorithms.It is based on degradation or induce their cleavage.There are three types of miRNA target sites.This includes Canonical, simple discriminative rules derived from important Marginal, and Atypical target sites [31].
features of target recognition observed from experiments.
The rule-based algorithms include TargetScan [39], Canonical target sites: It has a complete pairing within miRanda [40], PITA [33], etc.In recent years, new the seed region which determines the certainty of the data-driven prediction algorithms emerge along with interaction and account for the majority of validated the improving knowledge of miRNA target recognition conserved targets [35].These sites include 7mer-A1 and the increasing availability of various types of site, 7mer-m8 site and 8mer site.The 7mer-A1 site has relevant data sets.Data driven algorithms rely on 2-7 seed match plus nucleotide A at 1 position in important discriminative features learned from data mRNA, 7mer-m8 site has 2-7 seed match plus a match using sophisticated models.It includes MirTarget [41], at position 8 in mRNA and 8mer site has combination PicTar [42], miTarget [43], etc. Features of different of above two sites [31].
target prediction algorithms are depicted in Validation of miRNA target genes: Because of seed region, where an offset 6mer site has matching at insufficient efficacy and specificity of different target 3-8 positions.These 6mer sites typically have reduced prediction tools, there is a need to biological validation efficacy and are conserved by chance more frequently of predicted target genes.The reporter gene assay is than the larger sites.In stringent seed pairing criteria used most commonly for assessing miRNA:mRNA for site efficacy and prediction efficiency, 6mer sites interactions [45].In this method, the 3' UTR of the are ignored [31].
mRNA is cloned downstream of a luciferase ORF.Atypical sites: It includes 3'-supplementary and 3'-When this vector is transfected into cells expressing the compensatory sites.For 3'-supplementary sites, Watsontargeting miRNA, luciferase activity should be lower Crick pairing usually centering miRNA nucleotides for the empty vector.Microarray analysis can be used 13-16 supplements the seed matching which increase to observe miRNA:mRNA interactions [46] which the efficacy [36].For 3'-compensatory sites, Watson measure changes of mRNA levels.MicroRNA -Crick pairing usually centering on miRNA nucleotides profiling data obtained from sequencing of small 13-16 can compensate for a seed mismatch and RNAs can also be validated by qRT-PCR.Another thereby create a functional site.Now, it is known that approach to verify a miRNA-target interaction would simple base pairing is not sufficient for miRNA target be to knockout the miRNA gene and examine the prediction [37] and the secondary structure of miRNAeffects on protein changes [31].Northern blotting can mRNA duplexes is a factor that should be taken into be useful, although time-consuming, especially in consideration [38].
those cases where different isoforms of the target are Many existing algorithms based on conservation expressed.The effect of miRNA on target gene can also analyses, also include the binding energy of miRNAbe tested at protein level by western blotting, ELISA etc.Some miRNA target prediction algorithms: By using derived from a comprehensive sequencing project of a computational or bioinformatics tools one can predict large set of mammalian tissues and cell lines of normal and disease origin.In this prediction tool users can the targets and then go for validation by experimenexplore the set of genes that are potentially regulated by tation.Some of the miRNAs target prediction a particular microRNA.algorithms are discussed below: PicTar: PicTar is a computational method for identifying  [32].The system which plays an important role in modulating validated targets component of the miRecords contains virtually all the biological processes.Identification of validated miRNA-target interactions with systematic novel miRNAs and their targets may lead to better documentation of experimental support for such understanding of many regulatory processes inside the interactions.The Predicted Targets component of living systems.This is an emerging science with high miRecords is an integration of predicted miRNA futuristic impact.Computationally, target prediction is targets produced by 11 established miRNA target an attractive tool in studying the target genes which prediction programs.miRecords is expected to serve as will be biologically validated later on.These a useful resource not only for experimental miRNA bioinformatics algorithms help the researchers to researchers, but also for informatics scientists developing identify the target genes and further study in relation the next-generation miRNA target prediction programs.
with their biological function.
In conclusion, there is need to improve the miRanda: It finds potential binding sites by searching accuracy of computational or bioinformatics tools for for high complementarity between miRNA and mRNA target gene identification, reduction of false positive [40].The algorithm favors complementarity between targets and inclusion of experimentally validated 5' end of the miRNA and 3' end of mRNA and then targets.evaluate thermodynamically, using the Vienna RNA folding package.mirSVR, a regression model that is

2 .Figure- 1 .
Figure-1.Biogenesis of microRNA . Another study dealing with the open miRNA mechanism of translational repressionThe accessibility energy of the miRNA target sites revealed that mRNA abundance did not always were calculated as: decrease with gene translation, indicating that it may Accessibility energy (? ?G) = ?G -? G ,(where

Table - 1
Marginal sites: It includes 6 nucleotides Watson-Crick while some useful databases related to miRNA target pairing in the seed region.These are 6mer site and prediction are summarized in Table-2.offset 6mer sites.The 6mer sites have matching at 2-7
It is an online database system for miRNA RNAhybrid: It can identify multiple potential binding target gene prediction in five species: human, mouse, sites of miRNAs in target RNA sequence.RNAhybrid rat, dog and chicken.miRDB is different from existing avoids intramolecular hybridizations, that is base miRNA functional databases mainly in the following: It is an integrated resource for animal miRDB: miRecords: