MicroRNAs (miRNAs) are non-coding RNAs (ncRNAs) involved in regulation of gene expression. Intragenic miRNAs, especially those exhibiting a high degree of evolutionary conservation, have been shown to be coordinately regulated and/or expressed with their host genes, either with synergistic or antagonistic correlation patterns. However, the degree of cross-species conservation of miRNA/host gene co-location is not known and co-expression information is incomplete and fragmented among several studies. Using the genomic resources (miRBase and Ensembl) we performed a genome-wide in silico screening (GWISS) for miRNA/host gene pairs in three well-annotated vertebrate species: human, mouse, and chicken. Approximately half of currently annotated miRNA genes resided within host genes: 53.0% (849/1,600) in human, 48.8% (418/855) in mouse, and 42.0% (210/499) in chicken, which we present in a central publicly available Catalog of intragenic miRNAs (http://www.integratomics-time.com/miR-host/catalog). The miRNA genes resided within either protein-coding or ncRNA genes, which include long intergenic ncRNAs (lincRNAs) and small nucleolar RNAs (snoRNAs). Twenty-seven miRNA genes were found to be located within the same host genes in all three species and the data integration from literature and databases showed that most (26/27) have been found to be co-expressed. Particularly interesting are miRNA genes located within genes encoding for miRNA silencing machinery (DGCR8, DICER1, and SND1 in human and Cnot3, Gdcr8, Eif4e, Tnrc6b, and Xpo5 in mouse). We furthermore discuss a potential for phenotype misattribution of miRNA host gene polymorphism or gene modification studies due to possible collateral effects on miRNAs hosted within them. In conclusion, the catalog of intragenic miRNAs and identified 27 miRNA/host gene pairs with cross-species conserved co-location, co-expression, and potential co-regulation, provide excellent candidates for further functional annotation of intragenic miRNAs in health and disease.
Citation: Godnic I, Zorc M, Jevsinek Skok D, Calin GA, Horvat S, Dovc P, et al. (2013) Genome-Wide and Species-Wide In Silico Screening for Intragenic MicroRNAs in Human, Mouse and Chicken. PLoS ONE 8(6): e65165. https://doi.org/10.1371/journal.pone.0065165
Editor: Rolf Müller, Philipps University, Germany
Received: January 8, 2013; Accepted: April 22, 2013; Published: June 6, 2013
Copyright: © 2013 Godnic et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by the Slovenian Research Agency (ARRS) through the Research programme Comparative genomics and genome biodiversity [grant number P4–0220]. GAC is supported as a Fellow at The University of Texas M. D. Anderson Research Trust, as a Fellow of The University of Texas System Regents Research Scholar, and by the CLL Global Research Foundation. Work in Dr. Calin’s laboratory is supported in part by National Institutes of Health, by Department of Defense, by Developmental Research Awards in Breast Cancer, Ovarian Cancer and Leukemia SPOREs, and by 2009 Seena Magowitz - Pancreatic Cancer Action Network - AACR Pilot Grant. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
MicroRNAs (miRNAs) are non-coding RNAs (ncRNAs) that post-transcriptionally regulate gene expression. The standard dogma states that expression of protein-coding genes is repressed by binding the target gene's complementary sequence in the 3′ untranslated region (3′-UTR) with the miRNA’s seed region: 2–7 or 2–8 consecutive nucleotides from the 5′-end of the miRNA, which are crucial for target recognition , . This earlier postulated dogma has now been extended with new discoveries. MicroRNAs have also been shown to increase or decrease expression of protein-coding genes by targeting different genomic regions (3′-UTR, 5′-UTR, promoter, and coding sequences) and interact with proteins. Additionally, they have been shown to function in various subcellular compartments, and developmental and metabolic processes . Several components of the miRNA processing machinery are included in miRNA biogenesis, which first take place in the nucleus. Primary miRNA transcripts (pri-miRNAs) are processed by the complex Drosha-DGCR8 (DiGeorge syndrome critical region gene-8), a component of the miRNA processing machinery , . Thereafter precursor miRNAs (pre-miRNAs) are transported to the cytoplasm where they are further cleaved by RNase III Dicer, a key enzyme in miRNA maturation, to form functional mature miRNAs . They are incorporated into the RNA-induced silencing complex (RISC) composed of many associated proteins . Disruption of the miRNA processing machinery core components, miRNA genes and their targets affects overall efficiency of silencing . Indeed, polymorphisms as well as aberrant miRNA expression patterns have previously been shown to be involved in disease development, including several cancer types –.
Approximately half of vertebrate miRNAs are processed from introns of protein-coding genes or genes encoding for other ncRNA classes (e.g. snoRNAs, miRNAs, lincRNAs) , whereas miRNA genes can also be encoded in intergenic regions of DNA, therefore referred to as intergenic miRNAs. In some cases, a miRNA gene can have a “mixed” location, i.e. can be located either in an exon or an intron of the same or different host gene transcripts which depends on their alternative splicing .
A single host gene can comprise multiple and overlapping resident miRNA genes, called a cluster, which are processed from the same polycistronic primary transcript , . It has been observed that miRNA genes which are located in a polycistron and co-expressed in the clusters are pivotal in coordinately regulating multiple processes, including embryonic development, cell cycle and cell differentiation . It was also observed that miRNA genes are more frequently hosted within the short genes than expected by chance, which was hypothesized as a favorable evolutionary feature due to the gene’s interaction with the pre-miRNA splicing mechanism .
Host genes and resident ncRNAs have been considered to have a synergistic effect with important implications for fine-tuning gene expression patterns in the genome , . Expression profiles of intronic miRNAs were in many cases found to coincide with the transcription of their host genes, which raised a question as to how these miRNAs were processed . Intronic miRNAs, like most ncRNAs, are released from the excised host introns in the post-splicing process , . However, it was later indicated that intronic miRNAs might also be processed from unspliced intronic regions prior to splicing catalysis . A class of miRNA precursors, named mirtrons, are processed in an alternative miRNA biogenesis pathway where certain debranched introns mimic the structural features of pre-miRNAs and enter the miRNA-processing pathway, however without the Drosha-mediated cleavage .
Highly correlated expression patterns have been found in closely clustered miRNA genes (50 kb of each other), which coincides with the idea of a polycistronic primary transcript , . He et al.  additionally showed that evolutionary conserved miRNA genes tend to be co-expressed with their host genes: even though the non-conserved miRNAs dominate in the human genome, the majority of intragenic miRNAs exhibiting co-expression with their host gene are phylogenetically old. A high conservation between orthologous intronic miRNAs has been demonstrated in several species , . In addition to co-expression and proposed co-regulation of miRNA and host genes, several studies have described a functional link between them , , . Interestingly, genes highly correlated in expression with a resident miRNA gene were found to be more likely predicted as miRNA targets . The expression of miRNA/host genes and that of predicted miRNA targets tend to be positively or negatively correlated, suggesting that the coordinated transcriptional regulation of a miRNA and its target is an abundant motif in gene networks .
The proportion of miRNA genes located within the same host genes among different species remains unknown, whether their coordinated expression is conserved, and to what degree. The miRNA/host gene co-expression has been analyzed in several studies, yet the data remains fragmented and incomplete. However, based on the report by He et al.  that evolutionary conserved (“old”) miRNA genes tend to be co-expressed with their host genes, but, in contrast, non-conserved (“young”) ones rarely do so, it might be reasonable to predict the same co-expression patterns of miRNA/host gene pairs with conserved cross-species co-location. The conserved pairs would present candidate genes whose matching expression profiles would be of assistance for further annotation and functional analysis.
The aim of this study was to create a central Catalog of intragenic miRNAs in three well-annotated vertebrate species (human, mouse, and chicken) serving as a framework for researchers working in the field of intragenic miRNAs. The supplemented information regarding the miRNA/host gene pair’s conserved cross-species co-location, expression data, and disease associations provides a list of high priority intragenic miRNAs for further functional analyses. These include identification and annotation of genes based on cross-species conservation, functional analyses and studies to re-examine potential misattribution of phenotype previously ascribed to host genes or hosted miRNA genes only.
Materials and Methods
Datasets of miRNA/host gene pairs were downloaded from genomic resources: the coordinates of miRNA genes and their host genes in human, mouse, and chicken were downloaded from miRBase, release 19 (http://www.mirbase.org/)  and Ensembl, release 69 (http://www.ensembl.org/index.html), using the latest matching assemblies: GRCh37 for human, GRCm38 for mouse, and WASHUC2 for chicken. The catalog is accessible through a web application written in PHP language, which allows retrieving miRNA/host gene pairs (http://www.integratomics-time.com/miR-host/catalog). The nomenclature of miRNA and host genes was unified according to The HUGO Gene Nomenclature Committee (HGNC) (http://www.genenames.org/) and Mouse Genome Informatics (MGI) (http://www.informatics.jax.org/). The list of miRNA host genes was manually inspected; cases with doubtful gene nomenclature after automatic annotation (e.g. overwriting of a miRNA record with an overlapping snoRNA and lincRNA record) were reported to the source database (Ensembl) and solved case by case. Genomic distribution of miRNA/host gene pairs in human, mouse, and chicken was presented in a genomic view format using Flash GViewer web tool (http://gmod.org/wiki/Flashgviewer/). MicroRNA and host gene expression profiles, their functional links and diseases associated with dysregulated expression were retrieved from: 1) literature using PubMed (http://www.ncbi.nlm.nih.gov/pubmed), Web of Science (http://apps.webofknowledge.com/), and 2) databases Gene Expression Atlas (GEA), release 22.214.171.124 (http://www.ebi.ac.uk/gxa/). Small RNA expression data was obtained from University of California Santa Cruz (UCSC) Genome Bioinformatics (http://genome.ucsc.edu/) based on the ENCODE project . Genetic variability of miRNA genes residing within host genes (protein-coding and non-coding) was determined using miRNA SNiPer tool 3.0 (http://www.integratomics-time.com/miRNA-SNiPer) . Predicted and experimentally validated miRNA targets were obtained using TargetScan (http://www.targetscan.org/), miRecords (http://mirecords.biolead.org/), and miRTarBase (http://mirtarbase.mbc.nctu.edu.tw/). The list of components of the miRNA silencing machinery was obtained from Patrocles database (http://www.patrocles.org) . Pathway enrichment analysis for miRNA host genes was performed using the Ingenuity Pathway Analysis (IPA), release 8.8 (Ingenuity® Systems, http://www.ingenuity.com/) . Multispecies sequence alignments were performed using Ensembl, option Comparative genomics - Alignments (text).
Results and Discussion
We developed a central Catalog of intragenic miRNAs in three well-annotated vertebrate genomes (human, mouse, and chicken) by performing a genome-wide in silico screening (GWISS) of genomic resource databases (Figures 1 and 2). The miRNAs were hosted by protein-coding genes or genes encoding for other ncRNA classes. Further species-wide in silico screening (SWISS) revealed 27 miRNA/host gene pairs with conserved co-location in all three species, most of which have been found to be co-expressed. Coordinately expressed miRNA/host gene pairs with cross-species conserved co-location are considered prioritized candidate genes for future functional analysis.
* - microRNA genes overlapping protein-coding and ncRNA genes; mixed - microRNA genes overlapping intron, exon or UTR, depending on overlapping host gene transcripts. For details see online table: http://www.integratomics-time.com/miR-host/catalog.
1. Genome-wide in silico Screening (GWISS) for Sense-oriented miRNA/host Gene Pairs in Human, Mouse and Chicken
Intragenic miRNAs (Figure 3) have become a topic of increasing research interest. We performed a genome-wide in silico screening (GWISS) of the latest genome assemblies of three well-annotated vertebrate genomes (human, mouse, and chicken) to define how many miRNA genes are located within host genes. The Catalog of intragenic miRNAs is available through a web application (http://www.integratomics-time.com/miR-host/catalog), which allows users to retrieve single or multiple miRNA/host gene pairs, based on 1) selection of species, biotype of host genes, and genomic position of resident miRNAs (exon, intron, 3′ and 5′-UTR), or 2) by querying individual miRNA or their host genes. In all three species approximately half of currently annotated miRNAs are intragenic, residing within protein-coding and/or ncRNA genes: 53.0% (849/1,600) in human, 48.8% (418/855) in mouse, and 42.0% (210/499) in chicken (Figure 2). This percentage however should be considered as an estimate that will change with time as both miRNA and host genes (protein-coding and ncRNA genes) are still being annotated and added to database upgrades. Manual inspection of host genes revealed examples with doubtful annotation in regions with two or three overlapping genes, for which we contacted the source database (Ensembl) and solved ambiguous annotations case by case. Namely, it was observed that in cases where two ncRNA genes (miRNA and snoRNA) overlapped in the same region, the automatic annotation pipeline favored the longer RNA; for example, the record of snoRNA gene SNORA36B overwrote the record of the overlapping miRNA gene hsa-mir-664a. One of the reasons for annotation error may also be the use of non-official and inconsistent nomenclature of genes. For example, a miRNA host transcript with a lincRNA biotype (ENSG00000253522) was merged between the Ensembl automatic pipeline and the Havana manual curation and was found to be given two names, CTC-231O11.1 or hsa-mir-146a. Any updates of the catalog of miRNA/host gene pairs should therefore take into consideration the importance of nomenclature when searching for single or overlapping miRNA genes.
A) Protein-coding gene HTR2C with four resident miRNA genes, two of which form a cluster. B) A miRNA gene cluster located within lincRNA gene FTX. C) MicroRNA gene hsa-mir-10a located within two overlapping protein-coding genes. D) Overlapping miRNA gene (hsa-mir-664b) comprising a miR-seed-SNP, and snoRNA gene (SNORA36A) residing within protein-coding DKC1. E) Gene DGCR8, associated with miRNA biogenesis, hosts two miRNA genes, one of which comprises a miR-seed-SNP.
MicroRNA genes that do not share the same strand orientation as their host genes (i.e. are antisense-oriented) have been shown to have independent transcription mechanisms , whereas sense transcriptional orientation suggests that miRNA and host genes can be transcribed from shared promoters . Additionally, it was found that a majority of predicted promoter regions of intronic miRNA genes (94.2%; 49/52) overlapped with their host gene promoters . In addition to protein-coding host genes, ncRNA genes comprised snoRNAs, lincRNAs, and other unspecified ncRNAs (Figure 2). Long ncRNAs were found to also host clusters of miRNA genes and therefore encode polycistronic primary transcripts that can yield several miRNAs; for example lincRNA FTX (FTX transcript, XIST regulator (non-protein coding)) comprises two miRNA genes: hsa-mir-374a and hsa-mir-545 (Figure 3B). Because miRNA clusters can also overlap with a single protein-coding host gene (Figure 3A), the total number of host genes is lower than the number of intragenic miRNAs: we identified 687 protein-coding host genes in human (with 752 resident miRNA genes), 288 in mouse (with 386 miRNA genes), and 192 in chicken (with 208 miRNA genes). In all three species intragenic miRNA clusters most frequently comprise two miRNAs per host gene, as shown in the online table: http://www.integratomics-time.com/miR-host/catalog. The mouse host gene Sfmbt2 (Scm-like with four mbt domains 2), located on MMU2, was found to comprise the largest number of resident miRNA genes (n = 70) belonging to the mir-297, mir-466, and mir-467 gene families. Our study revealed that around one tenth of miRNA genes formed clusters in protein-coding host genes: 8.8% (141/1,600) in human, 14.5% (124/855) in mouse, and 8.2% (41/499) in chicken. It was also proposed that human miRNAs that share a host gene or are organized in clusters might also, due to clustering propensity, share a significant biological role , . Accordingly, miRNA genes that formed clusters were also found to be coordinately expressed with their host genes, which will be described in section 3.
For all three species (human, mouse, and chicken) we presented online genomic-views of intragenic miRNAs genes, connected to miRBase and host genes connected to Ensembl, with an outgoing link (http://www.integratomics-time.com/miR-host/GViews). The human genomic-view is presented in Figure S1. Intragenic miRNAs were found distributed among all chromosomes, however some, e.g. HSA14, HSA19, and HSAX, were found to comprise less intragenic miRNA genes compared to other chromosomes (Figure S2). In most cases miRNA genes resided within a single host gene. For example, human hsa-mir-1307 gene overlaps with a single host gene USMG5 (up-regulated during skeletal muscle growth 5 homolog (mouse)) gene. On the other hand, ten miRNA genes were found to overlap with two protein-coding host genes in human (http://www.integratomics-time.com/miR-host/human_coding). For example hsa-mir-10a overlapped with both, HOXB3 (homeobox B3) and HOXB4 (homeobox B4) (Figure 3C). Regarding the location of miRNA genes, we found that in accordance with previous publications , ,  a majority of intragenic miRNA genes were located within introns of their protein-coding host genes: 86.4% (650/752) in human, 84.4% (326/386) in mouse, and 97.1% (202/208) in chicken (Figure 2). Intronic miRNAs were also most frequently found to be coordinately expressed with their host genes among species, which will be further discussed in results section 2 and 3.
1.1. Co-location of miRNA with other ncRNA genes.
Besides the half of miRNAs located within protein-coding genes, we found that around 4% were positioned within genes encoding for other ncRNA classes. These include lincRNAs, snoRNAs, or other ncRNAs: 6.4% (103/1,600) in human, 4.8% (41/855) in mouse, and 1% (5/499) in chicken, which can be accessed at http://www.integratomics-time.com/miR-host/catalog. Nomenclature conflicts of miRNA and ncRNA names may occur due to annotation difficulties: information merged from the Ensembl automatic pipeline and the Havana manual curation, which assign gene names according to miRBase and the HUGO Gene Nomenclature Committee. Six human miRNA genes were found located in both, protein-coding and ncRNA genes: hsa-mir-600, -664a, -664b, -1248, -1291, and -3651 (online table http://www.integratomics-time.com/miR-host/human_table). MicroRNA gene hsa-mir-664b, its overlapping protein-coding host gene DKC1 (dyskeratosis congenita 1, dyskerin) and snoRNA SNORA36A gene are shown in Figure 3D. Some miRNA genes were found to form clusters within hosting ncRNA genes: for example the miRNA gene cluster, comprising hsa-mir-374a and hsa-mir-545, is located within lincRNA gene FTX (Figure 3B). Additionally, lincRNAs have also been found to be the most frequent type of ncRNA host genes (97/103) as shown in the online table: http://www.integratomics-time.com/miR-host/human_table. In some cases the designated lincRNAs have been found to be the primary transcripts and not actual lincRNA genes, for example MIR155HG (also known as BIC) and DLEU2 (deleted in lymphocytic leukemia 2 (non-protein coding), previously known as LEU2, are primary transcripts of their resident miRNA genes hsa-mir-155 and hsa-mir-15a/16-1, respectively. Besides miRNAs themselves being regulators of gene expression participating in a wide regulatory network , , their long ncRNA genes have likewise been found associated with human diseases. For example, lincRNA H19 (H19, imprinted maternally expressed transcript (non-protein coding)), which hosts hsa-mir-675, was implicated in human tumor growth  in esophageal  and breast cancer , and different carcinomas and hepatic metastases . Another study demonstrated that H19 and hsa-mir-675 were upregulated in human colon cancer cell lines and primary colorectal cancer tissues . Long intergenic ncRNA MEG3 (maternally expressed gene 3) could act as a tumor suppressor , while both the miRNA gene hsa-mir-155 and BIC RNA (MIR155HG) from which it is processed, were overexpressed in human B-cell lymphomas . Similarly, it was shown that the deletion of the 13q14 region, which encodes both, lincRNA DLEU2 and its resident miRNA cluster hsa-mir-15a/16-1, led to chronic lymphocytic leukemia in both human  and mouse .
1.2. Genetic variability of intragenic miRNA genes.
The intragenic miRNAs were also analyzed for genetic variability within the miRNA seed region (miR-seed-SNPs). By analyzing variation databases we found that 14.2% of intragenic miRNAs had polymorphic seed regions in human (121/849), 2.1% in mouse (9/418), and 1.4% in chicken (3/210) (Table S1). According to the NCBI database 18 out of 121 miRNA genes in human and two murine miRNA genes have not yet had validated miRNA seed polymorphisms. The actual proportion of polymorphic miRNA genes cannot yet be determined because miRNAs and polymorphisms, most of which are experimentally unvalidated, are still being discovered and added to the databases. That is why the results from previous studies tend to differ: Saunders et al.  found that less than 1% (3/474) of human miRNA genes miR-seed-SNPs, whereas in our previous study, Zorc et al. , we reported that 5.9% of miRNA genes comprised miR-seed-SNPs. Polymorphic miRNA genes are an interesting feature to include in the host gene analysis because they have previously been found to have functional associations. For example, we found a link between two independent studies: human MYH7B gene (myosin, heavy chain 7B, cardiac muscle, beta) hosts hsa-mir-499a, a miRNA upregulated in human and murine cardiac hypertrophy and cardiomyopathy , which comprises miR-seed-SNP rs3746444 linked with increased risk of dilated cardiomyopathy . A similar overlap was demonstrated previously comprising a mouse miRNA gene mmu-mir-717, a miR-seed-SNP identified in the lean mouse strain 129/Sv, a body mass associated host gene Gpc3 (glypican 3), as well as a growth associated quantitative trait locus (QTL) . Our catalog provides the basis for a more targeted selection of SNPs and functional connections with the miRNA and host genes.
1.3. MicroRNA/host gene pairs in miRNA biogenesis and regulation.
By considering the host gene’s function our study revealed an interesting observation that miRNAs are also located within genes encoding for components of the miRNA processing machinery. There were four miRNAs in human located within genes encoding for components of miRNA biogenesis: DGCR8, DICER1, and SND1 (Figure 4). Similarly, five miRNA genes in mouse were located within Cnot3, Dgcr8, Eif4e, Tnrc6b, and Xpo5 (Figure S3). Two miRNA genes (hsa-mir-1306 and hsa-mir-3618) reside within gene DGCR8, whose protein product is essential for miRNA biogenesis (Figure 3E). Human miRNA gene hsa-mir-3173, was found located within an intron of host gene DICER1, encoding a protein that functions as a ribonuclease required to produce active RNAs. MicroRNA gene hsa-mir-593 resided within an intron of SND1 (staphylococcal nuclease and tudor domain containing 1), a component of RISC. By performing a target gene analysis we found that each of the residing miRNAs was predicted to target genes which also host other miRNA genes (Figure 4). According to previous experimental studies, DICER1 was found targeted by nine miRNAs: hsa-let-7a, -7b, -7c, and -7d, hsa-mir-18a, -103, -107, -374a, and -519a –. Additionally, hsa-mir-3618 and hsa-mir-593 were found to comprise a miR-seed-SNPs (rs12159555 and rs73721294, respectively), however both SNPs still need to be validated. Where miRNA molecule targets a gene from a miRNA processing machinery this could indicate a negative regulatory loop and a multi-layer regulatory cross-point, possibly associated with the disrupted processing of miRNAs. Also, alterations in gene regulation could have pathologic implications, as all three miRNA silencing machinery genes have previously been linked to certain diseases: DICER1 with cancer , , DGCR8 with DiGeorge syndrome , and SND1 was found frequently up-regulated in human and mouse cancers, as well as in aberrant crypt foci . To summarize, this miRNA-related genomic cross-points consists of: 1) intragenic miRNAs, 2) miRNA gene polymorphisms, 3) miRNA host genes encoding for proteins involved in miRNA biogenesis and silencing, 4) miRNA target sites within miRNA host genes, and 5) their resident miRNAs targeting other host genes. Polymorphisms and aberrations in this miRNA-related and disease-associated genomic cross-point could therefore have a significant effect on phenotypic variation, including disease susceptibility and deserve further analysis.
Overlapping miRNA genes (hsa-mir-3618 and mir-1306, mir-3173, and mir-593), miRNA polymorphisms (miR-seed-SNPs (rs12159555 and rs73721294), host genes encoding for miRNA processing machinery components (DGCR8, DICER1, and SND1), miRNA target sites within host genes, and miRNAs targeting other host genes. Arrow with solid line: experimentally validated miRNA targets; arrow with dashed line: predicted miRNA targets.
2. Cross-species Conservation of miRNA/host Gene Co-location
In order to determine how many intragenic miRNAs are located within the same host genes in human, mouse, and chicken, we performed a species-wide in silico screening (SWISS) of their co-location. We found that 27 miRNA genes had conserved co-location within the same 23 host genes in all three species (Table 1, Figure S4). In some cases the host genes (NFYC, SMC4, and C9orf3) encompassed more than one resident miRNA, explaining the co-location of the 27 miRNAs within 23 host genes. Moreover, additional 93 miRNA/host gene pairs were found to have conserved co-location in human and mouse (online table: http://www.integratomics-time.com/miR-host/species_cons). Most of the intragenic miRNAs were found to reside within introns of their host genes (25/27) (Table 1). MicroRNA/host gene pairs with conserved co-location offer a foundation for structural annotation of novel miRNA genes in other species. Using this approach, we proposed a novel miRNA gene in chicken (mir-3064) based on its pre-miRNA region that was found conserved in human and mouse (Figure S5). Similarly, 15 potential miRNA genes in human have been suggested by comparing the annotated murine miRNA genes with the human genome. Sequences of potential human miRNAs were examined for small RNA expression data using the UCSC database. Four of the human sequences (complementary to mouse mmu-mir-677, -1839, -1897, and -1949) had available expression data (Figure S5), which further confirms that these sequences encode miRNAs. The proposed novel miRNA genes present candidates for further experimental validation, annotation and expression analysis. In this manuscript the proposed miRNAs (one in chicken and 15 in human) have been given temporary names and will be submitted to the miRBase upon acceptance of this manuscript by the peer review process.
3. Coordinated Expression and Functional Association of miRNA/host Gene Pairs
To find out whether miRNA/host gene pairs with conserved cross-species co-location are also co-expressed, we integrated experimental data from two different sources: published studies that experimentally confirmed miRNA/host gene co-expression and databases providing gene expression data for miRNA and host genes separately.
3.1. Co-expression of miRNA/host gene pairs with conserved cross-species co-location.
For the first step in determining if the 27 miRNA/host gene pairs with conserved cross-species co-location (in human, mouse, and chicken) (Table 1) are also co-expressed, we analyzed data from 28 studies that experimentally confirmed their coordinated expression , , , –. The data integration revealed that most miRNA/host gene pairs (26/27) have previously been found to be coordinately expressed (either both up- or down-regulated) in human and/or mouse (online table: http://www.integratomics-time.com/miR-host/co-exp). Co-expression of only one miRNA/host gene pair, mir-1306/DGCR8, has not yet been experimentally demonstrated. We also found opposing results regarding the expression of two miRNA/host gene pairs, murine mmu-mir-103/Pank3 and mmu-mir-107/Pank1– these have previously been demonstrated to have coordinate  as well as anti-correlative (or discordant) expression patterns . Out of the 26 miRNA/host gene pairs with coordinated expression, 11 have been found to be coordinately expressed in both, human and mouse , , , –, –, , –: mir-103/PANK3, mir-107/PANK1, mir-126/EGFL7, mir-128-1/R3HDM1, mir-140/WWP2, mir-211/TRPM1, mir-218-1/SLIT2, mir-218-2/SLIT3, mir-27b/C9orf3, mir-33/SREBF2, and mir-499/MYH7B. Moreover, two miRNA/host gene pairs have been found to have expression patterns associated with the same phenotype in both species: mir-499/MYH7B with heart development  and mir-33/SREBF2 with cholesterol homeostasis , , . Several independent studies in chicken have similarly indicated that gga-mir-33 and its host gene SREBF2 are highly expressed in the liver, suggesting involvement in expression upregulation of genes related to cholesterol biosynthesis , .
To further test the hypothesis that miRNA/host gene pairs with cross-species conserved co-location are coordinately expressed, we integrated expression data for 27 miRNA and their host genes using the GEA database. By comparing the gene expression data, we found that 24 miRNAs and their host genes had matching expression patterns in at least one disease (either over- or under-expression) (Table S2). Because of the same expression patterns and similar functions, the miRNA/host gene pairs are likely to be controlled by the same regulatory mechanisms. The miRNA/host gene pairs with conserved cross-species co-location, co-expression, and potential co-regulation provide a starting point for researchers investigating the involvement of intragenic miRNAs with disease development or control of production traits.
To better determine the role of the miRNA host genes from the pairs with conserved cross-species co-location, we performed a pathway enrichment analysis, using the IPA software . Pathway analysis performed on the 23 host genes (Table 1) revealed networks associated with cancer, dermatological diseases and conditions, and hematological diseases (Figure S6A). Most significant biological functions included cancer, in addition to reproductive system diseases and infectious diseases. A molecular network diagram was constructed involving 14 miRNA host genes (CTDSPL, C9orf3, COL27A1, EGFL7, HNRNPK, NFYC, PANK1, SLIT2, SLIT3, SMC4, SREBF2, TLN2, TRPM1, and WWP2) which were found related to cancer, dermatologic and hematological diseases (Figure S6B). Within this network, several hubs were found encoding transcription factors, the largest two of which were MYC (v-myc myelocytomatosis viral oncogene homolog (avian)) and TP53 (tumor protein p53), previously also linked with regulation of miRNA gene expression , .
3.2. Epigenetically silenced miRNA genes located within host genes.
Silenced expression of co-located miRNA and host genes might also be a subject of epigenetic regulation . Namely, the proximal CpG islands located within their promoter or 5′UTR regions could epigenetically silence gene expression through DNA hypermethylation. In a recent study, 81.2% of protein-coding genes harboring miRNA genes in their 5′-end have been found located 500 bp downstream of CpG islands . By performing a cross-section of 133 miRNA genes that have previously been found to be epigenetically regulated in cancer , we found that 30 are located within protein-coding, and 13 within ncRNA host genes, i.e. genes encoding for lincRNAs (Figure 1, Table 2). However, in order to determine the exact proportion of epigenetically regulated miRNA/host gene pairs a systematic genome-wide epigenetic analysis should be performed. Previous studies revealed that five miRNA genes as well as their host genes (hsa-mir-10a/HOXB4, hsa-mir-126/EGFL7, hsa-mir-152/COPZ2, hsa-mir-191/DALRD3, and hsa-mir-342/EVL) were found to be epigenetically downregulated, either by histone modification and/or CpG island hypermethylation in the promoter region in cancer cells , – (Table 2). Additionally, several host genes have, independently of miRNA studies, been found to be silenced through DNA hypermethylation: DALRD3 , HOXA9 –, HOXB4 , HOXB7 , HOXC4 , HOXD3 , HTR2C , and IGF2 . The identified epigenetically regulated intragenic miRNA genes can now be analyzed together with their host genes in order to study their potential epigenetic co-regulation. We found that around half (20/43) of the epigenetically silenced miRNA genes were located within the 5′-UTR or in the first intron or exon of their host genes, suggesting the possibility of shared promoter regions that comprise CpG islands. Further studies on epigenetic regulation of miRNA/genes may reveal novel approaches for prevention or treatment of human cancer.
4. MicroRNA/host Gene Pairs – Potential for Misattribution of Phenotype?
In our study we demonstrated that a very large proportion of miRNAs are located within the host genes (Figure 2) in human (1,131/1,600), mouse (518/855), and chicken (240/499) and that miRNA/host gene pairs have important conservation and co-expression issues. Our study can be used as a platform for researchers to re-examine questions related to earlier or planned studies correlating genetic variation or modification of the miRNA/host gene pairs with diseases or trait control. Namely, it is prudent to ask if some of the gene variation-phenotype association studies targeted at the miRNA host genes, spontaneous, radiation or chemically induced mutations, knockout and overexpression models need reinterpretation to take into account collateral effects on miRNAs. MicroRNA genes harbored within another host gene, as shown by many examples in our study, may have several target genes and functions unrelated to their host genes. The host gene mutations or modifications may also collaterally affect the level, time or tissue specificity of miRNA expression thereby leading to several pleiotropic effects in the phenotype that could not be causally ascribed to the host gene only. Many types of spontaneous and induced mutations within the host gene locus (e.g. promoter, splicing mutations, or mRNA stability mutations) may affect the transcript quantity, temporal and/or spatial expression pattern of hosted miRNA.
In addition to aforementioned effects, transgenic overexpression and knockout host gene models may alter hosted miRNA function through exogenous sequences left in the locus such as selection marker genes (e.g. neomycin resistance, NeoR), plasmid vector and other sequences (e.g. strong phosphoglycerate kinase (pgk) gene promotor). We note that among the knockout mice of relevance in Tables 1A and B, most models retained the NeoR marker and also other exogenous sequences that can potentially affect expression and function of hosted miRNA gene in addition to the target host gene itself. Many targeting constructs are designed to delete large portions of the target gene in order to ensure loss of function of the host locus. The weakness of this strategy is that some of the deleted sequence may contain miRNAs or regulatory sequences affecting neighboring genes. Significantly for this discussion, inadvertent deletion of mmu-mir-126 has led to the misattribution of phenotype - angiogenesis defects previously reported in a knockout of the Egfl7 locus were subsequently shown to have arisen due to deletion of the mmu-mir-126 .
A degree of common sense can be applied to assessing the level of confidence attributed to specific phenotypes of the miRNA/host gene pairs. Where the phenotype is consistent with what was expected from knowledge of gene expression and biochemistry for the host gene and hosted miRNA gene, one can be reasonably comfortable in attributing a phenotype to the host target gene function. However, where the phenotype is unexpected, or where multiple genotype-phenotype or multiple gene modification models show disparate effects, then one is justified in being more cautious and to proceed by further experimentation to differentiate the host gene from hosted miRNA gene phenotypic effects. In the future gene modification experiments many concerns raised above can be minimized by using recent technology of Zinc finger  and Tal nucleases . These methods generate minimal targeted modifications (i.e. point mutation generating premature stop codon) and do not leave exogenous sequence in the genome thereby providing excellent transgenic in vitro and in vivo models for miRNA/host gene pairs studies.
Our web site (http://www.integratomics-time.com/miR-host/) provides an efficient tool to check which host genes contain miRNAs while other tables list important functional and literature information to aid researchers in re-examining potential misattribution of phenotype previously ascribed to host genes or hosted miRNA genes only.
5. Future Perspective
Our assembled and supplemented catalog of miRNA/host gene pairs available via the web application will provide researchers with a data mining tool for investigating miRNA/host gene pair involvement of their coordinated expression, shared regulation, and function in diseases: 1) structural annotation - miRNA/host gene pairs with conserved cross-species co-location in the examined species present candidate genes for future annotation in other species. 2) Functional annotation - miRNA/host gene pairs with matching expression patterns integrated from databases are high priority candidates for experimental validation of their potential co-expression and co-regulation. 3) MicroRNAs overlapping with protein-coding and other ncRNA host genes (lincRNA and snoRNA) present candidates for evaluating molecular mechanisms underlying previously shown functional links. 4) MicroRNAs residing within genes encoding for miRNA silencing machinery present important miRNA-related regulatory cross-talk needing additional mechanistic experimentation to elucidate targeting interplay in which miRNAs target genes for miRNA processing components and, in a feedback loop, influences the production of miRNAs. 5) Identification and validation of polymorphisms located within miRNA genes, their host genes, and genes encoding for and processing machinery components may also reveal whether they contribute to phenotypic variation, including disease susceptibility. 6) Epigenetic silencing of both, miRNA and their host genes, offers insights into their shared regulation and their re-expression may be used to contribute to the effects of epigenetic therapy. The assembled epigenetically regulated intragenic miRNAs represent candidate genes for the study of miRNA/host gene pair epigenetic co-regulation. 7) Our web site also provides an efficient tool to identify certain miRNA/host gene pairs where previous studies show inconsistencies of the effects of natural or induced mutations on the phenotype. We point to examples where such phenotype misinterpretations could arise due to attribution collateral effects of such mutations on hosted miRNAs. Our catalog can therefore direct researchers to critically examine designs and interpretation of such miRNA/host gene cases.
In conclusion, the assembled catalog is, to our knowledge, the most comprehensive integrated assembly of intragenic miRNAs and their host genes in human, mouse, and chicken. The systematically integrated physical (genomic location and cross-species conserved co-location) and functional characterization (co-expression data) of miRNA/host gene pairs provides a starting point for researchers investigating involvement of intragenic miRNAs with human and animal health, and animal production traits. Using this approach we found that miRNA/host gene pairs with cross-species conserved co-location are very likely to be co-expressed. The expanding field of miRNA research requires a consideration of interplay of interconnecting regulatory mechanisms and their function into an intricate network, in which miRNA genes and their co-expressed host genes also play a role.
Print-screen of genomic view of intragenic miRNAs in human. Enlarged chromosome 22 showing hsa-mir-1306 and its host gene DGCR8 with databases linked through outgoing links.
Distribution of intragenic miRNA genes according to chromosome in A) human, B) mouse, and C) chicken.
MicroRNA genes located within genes encoding for the miRNA processing machinery in mouse.
Venn diagram of the number of miRNA/host gene pairs with cross-species conserved co-location.
Alignment of orthologous miRNA genes. A) Human (hsa-mir-3064) and mouse (mmu-mir-3064) miRNA genes matching the sequence in chicken. Mature miRNA regions are marked with a square. B) Murine miRNA genes (mmu-mir-677, -686, -717, -763, -1839, -1893, -1896, -1897, -1898, -1902, -1907, -1949, -2139, -3059, and -5125) aligned with human sequences. C) Fifteen potential human miRNA genes acquired based on alignment with 15 murine miRNA genes. D) Small RNA expression data for sequences matching the four potential new miRNA genes in human (hsa-mir-677, -1839, -1897, and -1949).
Network analysis of host genes from 27 conserved miRNA/host gene pairs, in human and mouse. A) Top network and biological functions associated miRNA host genes. B) Diagram of a top molecular network showing 14 miRNA host genes (gray-filled shapes) associated with cancer, dermatological diseases and conditions, and hematological diseases. White-filled shapes indicate connecting elements in between host genes in the network.
Intragenic miRNAs with polymorphic seed regions in human, mouse, and chicken.
Conceived and designed the experiments: TK. Analyzed the data: IG MZ DJS GAC SH TK. Contributed reagents/materials/analysis tools: MZ GAC. Wrote the paper: IG SH TK. Reviewed and evaluated the article: GAC SH PD MK TK. Designed the software: MZ.
- 1. Bartel DP (2004) MicroRNAs: Genomics, Biogenesis, Mechanism, and Function. Cell 116: 281–297.
- 2. Sun G, Yan J, Noltner K, Feng J, Li H, et al. (2009) SNPs in human miRNA genes affect biogenesis and function. RNA 15: 1640–1651.
- 3. Kunej T, Godnic I, Horvat S, Zorc M, Calin GA (2012) Cross Talk Between MicroRNA and Coding Cancer Genes. Cancer J 18: 223–231.
- 4. Lee Y, Ahn C, Han J, Choi H, Kim J, et al. (2003) The nuclear RNase III Drosha initiates microRNA processing. Nature 425: 415–419.
- 5. Han J, Lee Y, Yeom KH, Kim YK, Jin H, et al. (2004) The Drosha-DGCR8 complex in primary microRNA processing. Genes Dev 18: 3016–3027.
- 6. Bernstein E, Caudy AA, Hammond SM, Hannon GJ (2001) Role for a bidentate ribonuclease in the initiation step of RNA interference. Nature 409: 363–366.
- 7. Gregory RI, Chendrimada TP, Cooch N, Shiekhattar R (2005) Human RISC couples microRNA biogenesis and posttranscriptional gene silencing. Cell 123: 631–640.
- 8. Georges M, Coppieters W, Charlier C (2007) Polymorphic miRNA-mediated gene regulation: contribution to phenotypic variation and disease. Current Opinion in Genetics & Development 17: 166–176.
- 9. Ferdin J, Kunej T, Calin G (2010) Non-coding RNAs: Identification of Cancer-Associated microRNAs by Gene Profiling. Technology in Cancer Research & Treatment: 123–138.
- 10. Nicoloso MS, Sun H, Spizzo R, Kim H, Wickramasinghe P, et al. (2010) Single-nucleotide polymorphisms inside microRNA target sites influence tumor susceptibility. Cancer Res 70: 2789–2798.
- 11. Sand M, Gambichler T, Skrygan M, Sand D, Scola N, et al. (2010) Expression levels of the microRNA processing enzymes Drosha and dicer in epithelial skin cancer. Cancer Invest 28: 649–653.
- 12. Sand M, Skrygan M, Georgas D, Arenz C, Gambichler T, et al.. (2011) Expression levels of the microRNA maturing microprocessor complex component DGCR8 and the RNA-induced silencing complex (RISC) components Argonaute-1, Argonaute-2, PACT, TARBP1, and TARBP2 in epithelial skin cancer. Mol Carcinog.
- 13. Rodriguez A, Griffiths-Jones S, Ashurst JL, Bradley A (2004) Identification of Mammalian microRNA Host Genes and Transcription Units. Genome Research 14: 1902–1910.
- 14. Ambros V (2004) The functions of animal microRNAs. Nature 431: 350–355.
- 15. Zhang Y, Zhang R, Su B (2009) Diversity and evolution of MicroRNA gene clusters. Sci China C Life Sci 52: 261–266.
- 16. Golan D, Levy C, Friedman B, Shomron N (2010) Biased hosting of intronic microRNA genes. Bioinformatics 26: 992–995.
- 17. Rearick D, Prakash A, McSweeny A, Shepard SS, Fedorova L, et al. (2011) Critical association of ncRNA with introns. Nucleic Acids Res 39: 2357–2366.
- 18. Lutter D, Marr C, Krumsiek J, Lang EW, Theis FJ (2010) Intronic microRNAs support their host genes by mediating synergistic and antagonistic regulatory effects. BMC Genomics 11: 224.
- 19. Baskerville S, Bartel D (2005) Microarray profiling of microRNAs reveals frequent coexpression with neighboring miRNAs and host genes. Rna-a Publication of the Rna Society: 241–247.
- 20. Kim YK, Kim VN (2007) Processing of intronic microRNAs. EMBO J 26: 775–783.
- 21. Ruby JG, Jan CH, Bartel DP (2007) Intronic microRNA precursors that bypass Drosha processing. Nature 448: 83–86.
- 22. Sempere LF, Freemantle S, Pitha-Rowe I, Moss E, Dmitrovsky E, et al. (2004) Expression profiling of mammalian microRNAs uncovers a subset of brain-expressed microRNAs with possible roles in murine and human neuronal differentiation. Genome Biol 5: R13.
- 23. He C, Li Z, Chen P, Huang H, Hurst LD, et al.. (2012) Young intragenic miRNAs are less coexpressed with host genes than old ones: implications of miRNA-host gene coevolution. Nucleic Acids Res.
- 24. Saini HK, Enright AJ, Griffiths-Jones S (2008) Annotation of mammalian primary microRNAs. BMC Genomics 9: 564.
- 25. Ying SY, Lin SL (2005) Intronic microRNAs. Biochem Biophys Res Commun 326: 515–520.
- 26. Wang S, Aurora AB, Johnson BA, Qi X, McAnally J, et al. (2008) The endothelial-specific microRNA miR-126 governs vascular integrity and angiogenesis. Dev Cell 15: 261–271.
- 27. Saito Y, Friedman JM, Chihara Y, Egger G, Chuang JC, et al. (2009) Epigenetic therapy upregulates the tumor suppressor microRNA-126 and its host gene EGFL7 in human cancer cells. Biochemical and Biophysical Research Communications 379: 726–731.
- 28. Tsang J, Zhu J, van Oudenaarden A (2007) MicroRNA-mediated feedback and feedforward loops are recurrent network motifs in mammals. Mol Cell 26: 753–767.
- 29. Kozomara A, Griffiths-Jones S (2011) miRBase: integrating microRNA annotation and deep-sequencing data. Nucleic Acids Research 39: D152–D157.
- 30. Rosenbloom KR, Sloan CA, Malladi VS, Dreszer TR, Learned K, et al. (2013) ENCODE Data in the UCSC Genome Browser: year 5 update. Nucleic Acids Res 41: D56–63.
- 31. Zorc M, Jevsinek Skok D, Godnic I, Calin GA, Horvat S, et al. (2012) Catalog of MicroRNA Seed Polymorphisms in Vertebrates. PLoS One 7: e30737.
- 32. Hiard S, Charlier C, Coppieters W, Georges M, Baurain D (2010) Patrocles: a database of polymorphic miRNA-mediated gene regulation in vertebrates. Nucleic Acids Research 38: D640–D651.
- 33. Ingenuity Pathway Analysis system.
- 34. Li S-C, Tang P, Lin W-C (2007) Intronic MicroRNA: Discovery and Biological Implications. DNA and Cell Biology 26: 195–207.
- 35. Wang G, Wang Y, Shen C, Huang YW, Huang K, et al. (2010) RNA polymerase II binding patterns reveal genomic regions involved in microRNA gene regulation. PLoS One 5: e13798.
- 36. Sikand K, Slane SD, Shukla GC (2009) Intrinsic expression of host genes and intronic miRNAs in prostate carcinoma cells. Cancer Cell Int 9: 21.
- 37. Zhang Y, Zhang R, Su B (2009) Diversity and evolution of MicroRNA gene clusters. Sci China C Life Sci 52: 261–266.
- 38. Griffiths-Jones S, Grocock RJ, van Dongen S, Bateman A, Enright AJ (2006) miRBase: microRNA sequences, targets and gene nomenclature. Nucleic Acids Res 34: D140–144.
- 39. Matouk IJ, DeGroot N, Mezan S, Ayesh S, Abu-lail R, et al. (2007) The H19 non-coding RNA is essential for human tumor growth. PLoS One 2: e845.
- 40. Hibi K, Nakamura H, Hirai A, Fujikake Y, Kasai Y, et al. (1996) Loss of H19 imprinting in esophageal cancer. Cancer Res 56: 480–482.
- 41. Berteaux N, Lottin S, Monté D, Pinte S, Quatannens B, et al. (2005) H19 mRNA-like noncoding RNA promotes breast cancer cell proliferation through positive control by E2F1. J Biol Chem 280: 29625–29636.
- 42. Fellig Y, Ariel I, Ohana P, Schachter P, Sinelnikov I, et al. (2005) H19 expression in hepatic metastases from a range of human carcinomas. J Clin Pathol 58: 1064–1068.
- 43. Tsang WP, Ng EK, Ng SS, Jin H, Yu J, et al. (2010) Oncofetal H19-derived miR-675 regulates tumor suppressor RB in human colorectal cancer. Carcinogenesis 31: 350–358.
- 44. Zhang X, Zhou Y, Mehta KR, Danila DC, Scolavino S, et al. (2003) A pituitary-derived MEG3 isoform functions as a growth suppressor in tumor cells. J Clin Endocrinol Metab 88: 5119–5126.
- 45. Eis PS, Tam W, Sun L, Chadburn A, Li Z, et al. (2005) Accumulation of miR-155 and BIC RNA in human B cell lymphomas. Proc Natl Acad Sci U S A 102: 3627–3632.
- 46. Calin GA, Dumitru CD, Shimizu M, Bichi R, Zupo S, et al. (2002) Frequent deletions and down-regulation of micro- RNA genes miR15 and miR16 at 13q14 in chronic lymphocytic leukemia. Proc Natl Acad Sci U S A 99: 15524–15529.
- 47. Klein U, Lia M, Crespo M, Siegel R, Shen Q, et al. (2010) The DLEU2/miR-15a/16–1 cluster controls B cell proliferation and its deletion leads to chronic lymphocytic leukemia. Cancer Cell 17: 28–40.
- 48. Saunders MA, Liang H, Li W-H (2007) Human polymorphism at microRNAs and microRNA target sites. Proceedings of the National Academy of Sciences 104: 3300–3305.
- 49. Matkovich SJ, Hu Y, Eschenbacher WH, Dorn LE, Dorn GW (2012) Direct and indirect involvement of microRNA-499 in clinical and experimental cardiomyopathy. Circ Res 111: 521–531.
- 50. Zhou B, Rao L, Peng Y, Wang Y, Chen Y, et al. (2010) Common genetic polymorphisms in pre-microRNAs were associated with increased risk of dilated cardiomyopathy. Clin Chim Acta 411: 1287–1290.
- 51. Kunej T, Jevsinek Skok D, Horvat S, Dovc P, Jiang Z (2010) The Glypican 3-Hosted Murine Mir717 Gene: Sequence Conservation, Seed Region Polymorphisms and Putative Targets. International Journal of Biological Sciences: 769–772.
- 52. Tokumaru S, Suzuki M, Yamada H, Nagino M, Takahashi T (2008) let-7 regulates Dicer expression and constitutes a negative feedback loop. Carcinogenesis 29: 2073–2077.
- 53. Forman JJ, Legesse-Miller A, Coller HA (2008) A search for conserved sequences in coding regions reveals that the let-7 microRNA targets Dicer within its coding sequence. Proc Natl Acad Sci U S A 105: 14879–14884.
- 54. Tao J, Wu D, Li P, Xu B, Lu Q, et al. (2012) microRNA-18a, a member of the oncogenic miR-17–92 cluster, targets Dicer and suppresses cell proliferation in bladder cancer T24 cells. Mol Med Report 5: 167–172.
- 55. Huang Y, Chuang A, Hao H, Talbot C, Sen T, et al. (2011) Phospho-ΔNp63α is a key regulator of the cisplatin-induced microRNAome in cancer cells. Cell Death Differ 18: 1220–1230.
- 56. Yan M, Huang HY, Wang T, Wan Y, Cui SD, et al.. (2011) Dysregulated Expression of Dicer and Drosha in Breast Cancer. Pathol Oncol Res.
- 57. Shiohama A, Sasaki T, Noda S, Minoshima S, Shimizu N (2003) Molecular cloning and expression analysis of a novel gene DGCR8 located in the DiGeorge syndrome chromosomal region. Biochem Biophys Res Commun 304: 184–190.
- 58. Tsuchiya N, Ochiai M, Nakashima K, Ubagai T, Sugimura T, et al. (2007) SND1, a component of RNA-induced silencing complex, is up-regulated in human colon cancers and implicated in early stage colon carcinogenesis. Cancer Res 67: 9568–9576.
- 59. Liang Y, Ridzon D, Wong L, Chen C (2007) Characterization of microRNA expression profiles in normal human tissues. BMC Genomics 8: 166.
- 60. Böhlig L, Friedrich M, Engeland K (2011) p53 activates the PANK1/miRNA-107 gene leading to downregulation of CDK6 and p130 cell cycle proteins. Nucleic Acids Res 39: 440–453.
- 61. Musiyenko A, Bitko V, Barik S (2008) Ectopic expression of miR-126*, an intronic product of the vascular endothelial EGF-like 7 gene, regulates prostein translation and invasiveness of prostate cancer LNCaP cells. J Mol Med (Berl) 86: 313–322.
- 62. Watanabe K, Emoto N, Hamano E, Sunohara M, Kawakami M, et al. (2012) Genome structure-based screening identified epigenetically silenced microRNA associated with invasiveness in non-small-cell lung cancer. Int J Cancer 130: 2580–2590.
- 63. Wang YP, Li KB (2009) Correlation of expression profiles between microRNAs and mRNA targets using NCI-60 data. BMC Genomics 10: 218.
- 64. Lages E, Guttin A, El Atifi M, Ramus C, Ipas H, et al. (2011) MicroRNA and target protein patterns reveal physiopathological features of glioma subtypes. PLoS One 6: e20600.
- 65. Donzelli S, Fontemaggi G, Fazi F, Di Agostino S, Padula F, et al.. (2011) MicroRNA-128–2 targets the transcriptional repressor E2F5 enhancing mutant p53 gain of function. Cell Death Differ.
- 66. Beezhold K, Liu J, Kan H, Meighan T, Castranova V, et al. (2011) miR-190-mediated downregulation of PHLPP contributes to arsenic-induced Akt activation and carcinogenesis. Toxicol Sci 123: 411–420.
- 67. Tie J, Pan Y, Zhao L, Wu K, Liu J, et al. (2010) MiR-218 inhibits invasion and metastasis of gastric cancer by targeting the Robo1 receptor. PLoS Genet 6: e1000879.
- 68. Bak M, Silahtaroglu A, Møller M, Christensen M, Rath MF, et al. (2008) MicroRNA expression in the adult mouse central nervous system. RNA 14: 432–444.
- 69. Hackler L, Wan J, Swaroop A, Qian J, Zack DJ (2010) MicroRNA profile of the developing mouse retina. Invest Ophthalmol Vis Sci 51: 1823–1831.
- 70. Careccia S, Mainardi S, Pelosi A, Gurtner A, Diverio D, et al. (2009) A restricted signature of miRNAs distinguishes APL blasts from normal promyelocytes. Oncogene 28: 4034–4040.
- 71. Xie H, Lim B, Lodish HF (2009) MicroRNAs induced during adipogenesis that accelerate fat cell development are downregulated in obesity. Diabetes 58: 1050–1057.
- 72. Polster BJ, Westaway SK, Nguyen TM, Yoon MY, Hayflick SJ (2010) Discordant expression of miR-103/7 and pantothenate kinase host genes in mouse. Mol Genet Metab 101: 292–295.
- 73. Yang J, Qin S, Yi C, Ma G, Zhu H, et al. (2011) MiR-140 is co-expressed with Wwp2-C transcript and activated by Sox9 to target Sp1 in maintaining the chondrocyte proliferation. FEBS Lett 585: 2992–2997.
- 74. Najafi-Shoushtari SH, Kristo F, Li Y, Shioda T, Cohen DE, et al. (2010) MicroRNA-33 and the SREBP host genes cooperate to control cholesterol homeostasis. Science 328: 1566–1569.
- 75. Horie T, Ono K, Horiguchi M, Nishi H, Nakamura T, et al. (2010) MicroRNA-33 encoded by an intron of sterol regulatory element-binding protein 2 (Srebp2) regulates HDL in vivo. Proc Natl Acad Sci U S A 107: 17321–17326.
- 76. Rayner KJ, Suárez Y, Dávalos A, Parathath S, Fitzgerald ML, et al. (2010) MiR-33 contributes to the regulation of cholesterol homeostasis. Science 328: 1570–1573.
- 77. Marquart TJ, Allen RM, Ory DS, Baldán A (2010) miR-33 links SREBP-2 induction to repression of sterol transporters. Proc Natl Acad Sci U S A 107: 12228–12232.
- 78. Dávalos A, Goedeke L, Smibert P, Ramírez CM, Warrier NP, et al. (2011) miR-33a/b contribute to the regulation of fatty acid metabolism and insulin signaling. Proc Natl Acad Sci U S A 108: 9232–9237.
- 79. van Rooij E, Quiat D, Johnson BA, Sutherland LB, Qi X, et al. (2009) A family of microRNAs encoded by myosin genes governs myosin expression and muscle performance. Dev Cell 17: 662–673.
- 80. Hicks JA, Trakooljul N, Liu HC (2010) Discovery of chicken microRNAs associated with lipogenesis and cell proliferation. Physiol Genomics 41: 185–193.
- 81. Sakakura Y, Shimano H, Sone H, Takahashi A, Inoue N, et al. (2001) Sterol regulatory element-binding proteins induce an entire pathway of cholesterol synthesis. Biochem Biophys Res Commun 286: 176–183.
- 82. O'Donnell KA, Wentzel EA, Zeller KI, Dang CV, Mendell JT (2005) c-Myc-regulated microRNAs modulate E2F1 expression. Nature 435: 839–843.
- 83. Tarasov V, Jung P, Verdoodt B, Lodygin D, Epanchintsev A, et al. (2007) Differential regulation of microRNAs by p53 revealed by massively parallel Sequencing - miR-34a is a p53 target that induces apoptosis and G(1)-arrest. Cell Cycle 6: 1586–1593.
- 84. Kozaki K, Inazawa J (2012) Tumor-suppressive microRNA silenced by tumor-specific DNA hypermethylation in cancer cells. Cancer Sci 103: 837–845.
- 85. Kunej T, Godnic I, Ferdin J, Horvat S, Dovc P, et al. (2011) Epigenetic regulation of microRNAs in cancer: an integrated review of literature. Mutat Res 717: 77–84.
- 86. Grady WM, Parkin RK, Mitchell PS, Lee JH, Kim YH, et al. (2008) Epigenetic silencing of the intronic microRNA hsa-miR-342 and its host gene EVL in colorectal cancer. Oncogene 27: 3880–3888.
- 87. Tsuruta T, Kozaki K, Uesugi A, Furuta M, Hirasawa A, et al. (2011) miR-152 is a tumor suppressor microRNA that is silenced by DNA hypermethylation in endometrial cancer. Cancer Res 71: 6450–6462.
- 88. He Y, Cui Y, Wang W, Gu J, Guo S, et al. (2011) Hypomethylation of the hsa-miR-191 locus causes high expression of hsa-mir-191 and promotes the epithelial-to-mesenchymal transition in hepatocellular carcinoma. Neoplasia 13: 841–853.
- 89. Shen J, Wang S, Zhang YJ, Kappil MA, Chen Wu H, et al.. (2012) Genome-wide aberrant DNA methylation of microRNA host genes in hepatocellular carcinoma. Epigenetics 7.
- 90. Hwang SH, Kim KU, Kim JE, Kim HH, Lee MK, et al. (2011) Detection of HOXA9 gene methylation in tumor tissues and induced sputum samples from primary lung cancer patients. Clin Chem Lab Med 49: 699–704.
- 91. Wu Q, Lothe RA, Ahlquist T, Silins I, Tropé CG, et al. (2007) DNA methylation profiling of ovarian carcinomas and their in vitro models identifies HOXA9, HOXB5, SCGB3A1, and CRABP1 as novel targets. Mol Cancer 6: 45.
- 92. Bandyopadhyay S, Harris DP, Adams GN, Lause GE, McHugh A, et al. (2012) HOXA9 methylation by PRMT5 is essential for endothelial cell expression of leukocyte adhesion molecules. Mol Cell Biol 32: 1202–1213.
- 93. Zheng CL, Guo ZX, Han ZC, Zhou YL, Lu SH, et al. (2009) [Analysis on promoter CpG methylation and expression of HOXB4 gene in cord blood CD34(+) cells and peripheral blood mononuclear cells]. Zhongguo Shi Yan Xue Ye Xue Za Zhi 17: 674–678.
- 94. Bennett LB, Schnabel JL, Kelchen JM, Taylor KH, Guo J, et al. (2009) DNA hypermethylation accompanied by transcriptional repression in follicular lymphoma. Genes Chromosomes Cancer 48: 828–841.
- 95. Issa J-P (2009) DNA methylation as an epigenetic factor in the development and progression of polycythemia vera. University of Texas M.D. Anderson Cancer Center Houston, TX 77030.
- 96. Kron KJ, Liu L, Pethe VV, Demetrashvili N, Nesbitt ME, et al. (2010) DNA methylation of HOXD3 as a marker of prostate cancer progression. Lab Invest 90: 1060–1067.
- 97. Anderton JA, Lindsey JC, Lusher ME, Gilbertson RJ, Bailey S, et al. (2008) Global analysis of the medulloblastoma epigenome identifies disease-subgroup-specific inactivation of COL1A2. Neuro Oncol 10: 981–994.
- 98. Dejeux E, Olaso R, Dousset B, Audebourg A, Gut IG, et al. (2009) Hypermethylation of the IGF2 differentially methylated region 2 is a specific event in insulinomas leading to loss-of-imprinting and overexpression. Endocr Relat Cancer 16: 939–952.
- 99. Kuhnert F, Mancuso MR, Hampton J, Stankunas K, Asano T, et al. (2008) Attribution of vascular phenotypes of the murine Egfl7 locus to the microRNA miR-126. Development 135: 3989–3993.
- 100. Urnov FD, Rebar EJ, Holmes MC, Zhang HS, Gregory PD (2010) Genome editing with engineered zinc finger nucleases. Nat Rev Genet 11: 636–646.
- 101. Bogdanove AJ, Voytas DF (2011) TAL effectors: customizable proteins for DNA targeting. Science 333: 1843–1846.