RNA BIOLOGY

2016, VOL. 13, NO. 1, 34–42

http://dx.doi.org/10.1080/15476286.2015.1128065

TECHNICAL REPORT

CircInteractome: A web tool for exploring circular RNAs and their interacting proteins
and microRNAs
Dawood B. Dudekulay, Amaresh C. Panda,y, Ioannis Grammatikakis, Supriyo De, Kotb Abdelmohsen, and
Myriam Gorospe
Laboratory of Genetics, National Institute on Aging-Intramural Research Program, National Institutes of Health, Baltimore, Maryland 21224, USA

ABSTRACT ARTICLE HISTORY

Circular RNAs (circRNAs) are widely expressed in animal cells, but their biogenesis and functions are poorly Received 19 October 2015
understood. CircRNAs have been shown to act as sponges for miRNAs and may also potentially sponge Revised 17 November 2015
RNA-binding proteins (RBPs) and are thus predicted to function as robust posttranscriptional regulators of Accepted 1 December 2015
gene expression. The joint analysis of large-scale transcriptome data coupled with computational analyses KEYWORDS
represents a powerful approach to elucidate possible biological roles of ribonucleoprotein (RNP) CircRNA-miRNA; CLIP-Seq;
complexes. Here, we present a new web tool, CircInteractome (circRNA interactome), for mapping RBP- circRNA siRNA; circRNA IRES;
and miRNA-binding sites on human circRNAs. CircInteractome searches public circRNA, miRNA, and RBP divergent primer design;
databases to provide bioinformatic analyses of binding sites on circRNAs and additionally analyzes miRNA RNA-binding proteins;
and RBP sites on junction and junction-flanking sequences. CircInteractome also allows the user the ability sponge circRNAs;
to (1) identify potential circRNAs which can act as RBP sponges, (2) design junction-spanning primers for transcriptome
specific detection of circRNAs of interest, (3) design siRNAs for circRNA silencing, and (4) identify potential
internal ribosomal entry sites (IRES). In sum, the web tool CircInteractome, freely accessible at http://
circinteractome.nia.nih.gov, facilitates the analysis of circRNAs and circRNP biology.

Introduction the cellular responses to stresses, mitogens, and immune trig-

gers.10,11 Consequently, RBPs have been implicated in a wide
Circular RNAs (circRNAs) are widely expressed RNAs lacking range of human diseases such as cancer, muscle pathologies,
50 and 30 ends, forming instead covalently closed RNA loops. In and neurodegenerative conditions.12-16 Recent developments in
eukaryotes, circRNAs arise most often through backsplicing, a the analysis in RNA-protein (RNP) interactions using cross-
process in which the 5' and 3' ends of a spliced RNA are cova- linking techniques such as CLIP (cross-linking immunoprecipi-
lently linked to form a closed RNA molecule that contains exon tation), PAR-CLIP (photoactivatable-ribonucleoside-enhanced
and/or intron sequences.1-3 Although a few examples of circR- CLIP), HITS-CLIP (high-throughput sequencing CLIP), and
NAs have been known for several decades, only with the recent iCLIP (individual-nucleotide-resolution CLIP) have identified
widespread use of high-throughput RNA sequencing and bioin- with unprecedented precision the binding sequences of RBPs
formatics have we learned that circRNAs constitute a vast class on target RNAs.17-20
of stable RNAs expressed endogenously in cells, often with tis- MicroRNAs (miRNAs) comprise another major class of
sue-specific patterns.1-5 The functional roles of circRNAs are posttranscriptional regulators. They bind thousands of human
not as well established as those of other noncoding (nc)RNAs transcripts through partial complementarity and generally
such as microRNAs (miRNAs). However, one prominent reduce their expression by suppressing mRNA translation and/
mechanism whereby circRNAs are believed to function is by or stability.21-23 In some cases miRNAs may enhance mRNA
sponging miRNAs, sequestering them away from protein-cod- translation.24,25 MicroRNAs are initially synthesized as primary
ing mRNAs.2 It has also been postulated that circRNAs could (Pri)-microRNA transcripts, which are processed by the micro-
serve as sponges for RNA-binding proteins (RBPs), platforms processor complex into microRNA precursors (Pre)-micro-
for assembly of RBPs, and protein-coding templates for RNAs.26-29 Pre-microRNAs are then exported to the cytoplasm
translation.6 where they are further processed by DICER1 and loaded onto
RBPs control all stages of post-transcriptional gene expres- the RNA-inducible silencing complex (RISC) to be directed to
sion, including the splicing, export, turnover, translation, and target mRNAs.28-31
localization of mRNAs.7-9 By modulating gene expression, RBPs and miRNAs can bind the same mRNA in a variety of func-
RBPs play key roles in virtually all cellular processes – prolifera- tional manners; for example, they can compete, cooperate, or bind
tion, differentiation, motility, senescence, apoptosis, as well as sequentially to a given mRNA.32-34 Such corregulatory effects by

CONTACT Kotb Abdelmohsen [email protected]

y
These authors equally contributed to this work.
Color versions of one or more of the figures in the article can be found online at www.tandfonline.com/krnb.
Supplemental material data for this article can be accessed on the publishers website.
 RNA BIOLOGY 35

RBPs and miRNAs influence physiologic processes as well as dis- targeting the body of mature circRNAs (Fig. S1B). This suggests that
ease.35,36 Thus,itisimportanttounderstandindetailtheirinteraction EIF4A3 could have a preference for binding to circRNA junctions com-
withmRNAs.Severaldatabasesareavailablewhichfacilitateresearch pared to other RBPs (Fig. S1B). An example of this type of search for
on RBPs (e.g., starBase) and miRNAs (e.g., TargetScan) interacting hsa_circ_0000020 is shown in Figure 2B and Figure S4A.
with specific mRNAs. Here, we have developed a new computational
resource named ‘CircInteractome’ which enables researchers to
searchsystematicallyforpossibleinteractionsofcircRNAswithRBPs Mapping binding sites of RBPs on pre-circRNA
and miRNAs. This tool will greatly facilitate the identification of The splicing machinery can generate circRNAs through non-
circRNAs that can sponge miRNAs and/or RBPs and will enhance linear back-splicing, which joins 50 and 30 ends covalently to
the search for circRNAs with potential function as platforms for the make a circRNA.38 Splicing events are tightly regulated by
assemblyofRBPs.Additionalfeaturessuchastheabilitytodisplaythe RBPs and snRNAs binding near the splice sites.39 Thus, we
mRNA counterpart and the genomic and mature sequences of the used CircInteractome to search all datasets (Table S1) in order
circRNA, as well as the ability to design specific circRNA divergent to identify the binding sites of RBPs in the flanking sequences
primers and circRNA-directed siRNAs, are aimed at further facilitat- upstream and downstream of the mature circRNA. For
ingother aspects of circRNA research. The CircInteractome tool cur- instance, the flanking sequences on both sides of the hsa_-
rently searches against Homo sapiens databases but will be expanded circ_0000020 junction indicated 4 binding sites for EIF4A3
toincludeotherspeciesinthefuture. (Fig. 2C, bottom; Fig. S4B), suggesting a possible role for
EIF4A3 in the biogenesis of hsa_circ_0000020. Our analysis
also uncovered a relatively higher frequency of putative binding
Results and discussion sites for splicing factors like TDP43 at the flanking sequences of
CircRNA-wide mapping of RBPs and RBP ‘super-sponges’ circRNAs, which point to a possible role for these RBPs in
circRNA splicing/biogenesis (Fig. S1C).
Published CLIP datasets do not specify if the RBP binding sites are
present in linear or circular RNAs (unless the CLIP hit spans a
junctional sequence). Thus, we utilized CLIP datasets as indicated Potential circRNA translation though IRES
in the workflow (Fig. 1) to create a comprehensive binding map of
RBPs to circRNAs (Fig. 2). We integrated 93 independently CircRNAs are believed to not be translated. While most circR-
NAs do not appear to be associated with polyribosomes, the
reported CLIP datasets from various RBPs (Table S1) obtained
possibility exists that some circRNAs might be translated into
from different tissues and cell lines.37 Computational analyses
protein products.3 In this regard, linear long noncoding (lnc)
revealed that for select RBPs there were large numbers of binding
RNAs are not generally translated into proteins, but a subset of
sites in circRNA sequences (Table S2); for instance, we identified
them appear to be a source of functional small or micropeptides
~117,000 circRNAs that could potentially associate with the RBP
due to the presence of short open reading frames (ORFs).40 For
EIF4A3 (Fig. S1A). Analysis of other RBPs using CircInteractome
example, a micropeptide encoded by a noncoding RNA was
indicated that they could also potentially interact with numerous
circRNAs (Fig. S1A). For example, the mature circRNA hsa_- recently found to regulate muscle performance.41 In addition,
the viral circRNA CCC (covalently closed circular, 220 nt long)
circ_0000020 hosts multiple binding sites for several RBPs like
was recently found to be fully translated into a 16-kDa highly
HuR (6 sites) and FMRP (10 sites) (Fig. 2A, B). Thus, we hypothe-
basic protein in infected rice plants.42 Bioinformatic analysis
sized that circRNAs with relatively high density of binding sites for
revealed the presence of IRES (internal ribosome entry sites)
any single RBP could potentially act as a ‘sponge’ or a ‘decoy’ for
and sites for RBPs that regulate IRES-mediated translation
that RBP. This sponging function would be enhanced by the long
(IRES trans-acting factors or ITAFs) in several circRNA
half-lives of circRNAs. By extension, circRNAs with exceptionally
sequences (Table S3).43 For example, we found that hsa_-
high density of binding sites for a given RBP might be considered
to be ‘super-sponges’; for example, hsa_circ_0024707 (428 nt long) circ_0041407 contains an IRES (underlined sequence) and par-
tial coding sequences of MAX network transcriptional
could function as a super-sponge for AGO2, since AGO2 can
repressor (MNT; Fig. 3A, B) and postulated that hsa_-
potentially bind this relatively short circRNA at 85 predicted posi-
circ_0041407 could give rise to a small chimeric protein of
tions (Table S2). In sum, this tool facilitates the search for potential
»31 kDa (highlighted sequence, Fig. 3B). To our surprise, IRES
RBPs interacting with circRNAs, and can identify possible RBP
regions of circRNAs are predicted sites for many RBPs, includ-
sponges as indicated in Figure 2 and Figure S3.
ing HuR and PTB, 2 proteins reported to modulate IRES-
driven translation.43 Together, these findings suggest that
RBPs on circRNA junctions circRNAs could be translated through IRES sequences, and
thus IRES-bearing circRNAs detected in association with poly-
RBPs may also interact with circRNA junctions and play a role in somes warrant future exploration.
circRNA splicing, processing, folding, stabilization, and localization. To
test this possibility, we queried CircInteractome for possibly binding sites
for RBPs at select sequences spanning 100 nt upstream and downstream miRNA ‘super-sponge’ circRNAs
from the junction site (Fig. 2B). Computational analysis of the RBP bind-
ing sites from various datasets (Table S1) revealed that RBPs may indeed The main function described for circRNAs thus far in the liter-
interact with circRNA junctions. For instance, EIF4A3 targets junction ature is that of miRNA sponges, as shown for the circRNA
sequences with much higher frequency than the frequency seen for ciRS-7, which has multiple binding sites for miR-7.5 To
36 D. B. DUDEKULA ET. AL.

Figure 1. Workflow of the web tool Circular RNA Interactome or ‘CircInteractome’.

characterize miRNA-circRNA interactions (Fig. 4, Fig. S5), we numerous target sites for a specific miRNA. In cases of excep-
incorporated into CircInteractome the ability to search using tionally high hit numbers, a circRNA could be considered a
the TargetScan algorithm, which predicts miRNAs that target ‘super-sponge’ for microRNAs. Strikingly, over 3,000 circRNAs
circRNA by surveying for 7-mer or 8-mer complementarity to were found to have at least 20 miRNA target sites in a single
the seed region as well as the 30 end of each miRNA.46 A survey circRNA, and most of them had AGO2 binding sites
of miRNA target sites in circRNAs revealed the presence of (Table S4). For example, hsa_circ_0139850, which is only
 RNA BIOLOGY 37

Figure 2. View of CircInteractome input and output pages. A. Example the ‘Circular RNA’ page exhibiting the input parameters needed for a CircInteractome run. B.
Screenshot of ‘Circular RNA search’ for has_circ_0000020, showing RBPs binding in different regions of this circRNA. C. Schematic representation of the potential binding
sites of different RBPs to mature and pre-hsa_circ_0000020.

437 nt long, has 22 sites for miR-7 and thus might act a super- specific to circRNAs. Thus, we incorporated primer design
sponge for miR-7 (Fig. 4). The general finding that circRNAs tools into CircInteractome (Primer345 or NCBI primer design
often have more miRNA target sites than those predicted by tool) and used as template the sequence around the circRNA
chance lends further support to the idea that binding micro- junction to ensure that the circRNA was amplified specifically
RNAs could be one of the key functions of circRNAs. by reverse transcription (RT) followed by real-time quantitative
(q)PCR analysis (Fig. 5A). To illustrate this feature of CircIn-
teractome, we show the design of divergent primers at the junc-
Divergent primer design
tion sequence of hsa_circ_0000020 (Fig. 5B; Fig. S6A,B).
Designing specific primers for quantification of circRNA using
qPCR amplification can be challenging and prone to errors,
siRNA design
since the mature circRNA sequences after splicing are not read-
ily available in many cases and the primers must be divergent Designing siRNAs selectively directed at circRNAs is
and must span the junction. At present, there is no software or challenging due to the limited knowledge of the junc-
web server that enables direct design of divergent primers tional sequences and the lack of appropriate software
38 D. B. DUDEKULA ET. AL.

Figure 3. Predicted IRES possibly mediating circRNA translation. A. Schematic representation of hsa_circ_0041407, which contains the MNT 36-160 IRES that may
potentially drive translation. B. The sequence of hsa_circ_0041407 highlighted can potentially be translated via the IRES underlined.

tools to help design them. Thus, we incorporated into role in sequestering RBPs and/or miRNAs and thereby reduc-
CircInteractome the ability to generate 21-nt siRNAs tar- ing their availability for mRNAs. CircInteractome also facili-
geting junctional sequences spanning 5–16 nt on either tates the design of primers for studying circRNA by RT-qPCR
side of the junction. Using standard siRNA design crite- analysis. CircInteractome can be used to predict RBP binding
ria, we filtered junctional sequences to find the best to upstream and downstream sequences of the pre-spliced tran-
siRNA target sequence 46,47; users need to avoid siRNAs script, thus potentially shedding light into the biogenesis of
with long stretches rich in G/C, and confirm using circRNAs. The interaction of translation regulators (miRNAs,
BLAST that the siRNAs are specific. For better silencing RBPs, ITAFs) and the presence of IRESs in circRNAs can help
effects, the output siRNAs may be ordered with an addi- to decipher whether circRNAs have protein/peptide-coding
tional 2 nucleotides (dTdT) as 30 DNA overhangs capabilities.
(Fig. S7). Users may then purchase them from any CircInteractome incorporates several features from
source; links to forms to complete siRNA orders using other freely available web resources, such as circBase,
IDT or Dharmacon are provided. Examples of the 10 best StareBase 2.0, TargetScan 7.0, and Primer3. By integrating
possible siRNA target sequences for hsa_circ_0000020 are these resources, CircInteractome enables the user to find
shown (Fig. 6). In short, CircInteractome can be used to out the genomic and mature circRNA sequences (Fig. S2),
predict siRNA sequences targeting the circRNA junctions. circRNA-binding partners (RBPs, miRNAs), siRNAs, and
primers to study circRNA levels, localization, and
function.
Conclusions and future directions
Although CircInteractome has many user-friendly features
This freely available web service can facilitate the study of cir- for circRNA researchers, it is limited in its ability to predict
cular RNAs and their interactions with other binding factors, RBP and miRNA interactions when circRNAs form secondary
mainly RBPs and miRNAs. CircInteractome provides research- or tertiary structures. Given that all of the data provided here
ers with valuable detail about circular RNAs and their possible are predicted based on sequence matches and the presence of
 RNA BIOLOGY 39

Figure 4. Potential miRNA-circRNA interactions. For a given circRNA ID entered, the miRNAs potentially targeting that circRNA are identified.

secondary or tertiary structures in circRNA cannot be consid- Methods

ered systematically, experimental validation is essential to verify
Acquisition of circular RNA sequences
RBP and miRNA functional sites.
We plan to maintain, update and curate CircInterac- CircBase (http://www.circbase.org) is a public database that was
tome in the foreseeable future, and will include additional developed to gather unified datasets of circRNA IDs, genomic
RBPs, microRNAs, and circRNAs as they become avail- coordinates and best transcripts.48 The mature sequences of all
able. Since RBP and miRNA sites may be masked or of the reported circRNAs were downloaded from the UCSC
revealed depending on whether the RNA is single- or browser mirror (http://genome.mdc-berlin.de) at the Max-
double-stranded, we will also include information on sec- Delbr€uck-Centrum f€ ur Molekulare Medizin (MDC), Berlin,
ondary structure as it is reported. In addition, experimen- Germany.49A complete list of all human circRNAs used in this
tal validation of circRNAs interacting with RBPs and study is available in the CircInteractome website. We also
microRNAs will also be included in CircInteractome. We retrieved 1000 bp upstream and downstream of circRNA from
will also integrate circRNAs from other species such as the UCSC genome browser mirror as circRNA flanking regions.
mouse and monkey, and will include predictions of RNA We used 100 nucleotides from the circRNA 30 and joined them
hybrids including circRNA:mRNA and circRNA:lncRNA. to 100 nucleotides at the 50 to identify the junctional sequence.
Collectively, this resource will accelerate our efforts to For circRNAs shorter than 200 nt, we divided the RNA
understand the roles of circRNAs in biological processes sequence in to two equal halves and the first half was joined at
relevant to health and disease. the 30 of the second half to generate the template for primer
40 D. B. DUDEKULA ET. AL.

Figure 5. CircRNA primer design. A. Schematic representation of the design of divergent primers for circRNA detection by RT-qPCR analysis. B. Input webpage for diver-
gent primer design. C. Screenshot of the output page showing the junction sequence of circRNA and links of primer design tools (Primer3 and NCBI).

design (Fig. 5A). This dataset was used for sequence alignment homology and maximum of 2 mismatches to find their binding
with RBP binding sites. sites (Fig. 1). When more than one tag was predicted at the
same site for an RBP, only one tag was considered as the bind-
ing sequence.
Acquisition of binding sites of RBPs on circRNAs
Datasets including the binding sites of 35 RBPs (Table S1) iden-
IRES dataset
tified by PAR-CLIP, HITS-CLIP, or iCLIP were retrieved from
starBase v2.0 (http://starbase.sysu.edu.cn/),37 UCSC browser, The sequences of reported IRES were downloaded from the
and published datasets. A complete list of all datasets used in IRESite (http://iresite.org/), which contains information about
the study is shown in Table S2. Data were previously analyzed experimentally validated IRES sequences.50 These IRES sequen-
and thus we used these datasets for our analysis without modi- ces were analyzed against mature circRNA sequences to find
fication. The sequences for the RBP CLIP clusters were down- out the presence of IRES in circRNAs; 99% similarity between
loaded from the UCSC browser mirror (http://genome.mdc- the IRES sequence and the mature circRNA sequence was
berlin.de)49 and were analyzed against circRNA mature, flank- allowed (Fig. 1). IRES sequences present in circRNAs were fur-
ing, and junction sequences of circRNAs with 95% sequence ther analyzed against the RBP CLIP cluster sequences to
 RNA BIOLOGY 41

Acknowledgements
This research was supported in full by the National Institute on Aging
Intramural Research Program of the National Institutes of Health.

References
1. Salzman J, Chen RE, Olsen MN, Wang PL, Brown PO. Cell type-spe-
cific features of circular RNA expression. PLoS Genet 2013; 9:
e1003777; PMID:24039610; http://dx.doi.org/10.1371/journal.
pgen.1003777
2. Memczak S, Jens M, Elefsinioti A, Torti F, Krueger J, Rybak A, Maier
L, Mackowiak SD, Gregersen LH, Munschauer M, et al. Circular
RNAs are a large class of animal RNAs with regulatory potency.
Nature 2013; 495:333-8; PMID:23446348; http://dx.doi.org/10.1038/
nature11928
3. Jeck WR, Sorrentino JA, Wang K, Slevin MK, Burd CE, Liu J, Marzluff
WF, Sharpless NE. Circular RNAs are abundant, conserved, and asso-
ciated with ALU repeats. RNA 2013; 19:141-57; PMID:23249747;
http://dx.doi.org/10.1261/rna.035667.112
Figure 6. CircRNA siRNA design. Screenshot of the CircInteractome siRNA design
page, including an example of output siRNAs targeting the junction sequence of a 4. Salzman J, Gawad C, Wang PL, Lacayo N, Brown PO. Circular RNAs
given circRNA. are the predominant transcript isoform from hundreds of human
genes in diverse cell types. PLoS One 2012; 7:e30733;
PMID:22319583; http://dx.doi.org/10.1371/journal.pone.0030733
5. Hansen TB, Jensen TI, Clausen BH, Bramsen JB, Finsen B, Damgaard
identify binding sites of different RBPs in the IRES sequences of CK, Kjems J. Natural RNA circles function as efficient microRNA
circRNAs (Fig. 1). sponges. Nature 2013; 495:384-8; PMID:23446346; http://dx.doi.org/
10.1038/nature11993
6. Hentze MW, Preiss T. Circular RNAs: splicing’s enigma variations.
miRNA Target ciRNAs EMBO J 2013; 32:923-5; PMID:23463100; http://dx.doi.org/10.1038/
emboj.2013.53
Mature sequences of circRNAs were used in the TargetScan 7. Grigull J, Mnaimneh S, Pootoolal J, Robinson MD, Hughes TR.
Perl Script to predict the miRNAs which have sequence com- Genome-wide analysis of mRNA stability using transcription inhibi-
tors and microarrays reveals posttranscriptional control of ribosome
plementarity with circRNA.44 The complete miRNA list and
biogenesis factors. Mol Cell Biol 2004; 24:5534-47; PMID:15169913;
sequences were taken from the microRNA database (http:// http://dx.doi.org/10.1128/MCB.24.12.5534-5547.2004
www.mirbase.org/).51 8. Lecuyer E, Yoshida H, Parthasarathy N, Alm C, Babak T, Cerovina T,
Hughes TR, Tomancak P, Krause HM. Global analysis of mRNA
localization reveals a prominent role in organizing cellular architec-
Divergent primer design ture and function. Cell 2007; 131:174-87; PMID:17923096; http://dx.
doi.org/10.1016/j.cell.2007.08.003
The circRNA junction sequences were retrieved from the 9. Blencowe BJ. Alternative splicing: New insights from global analyses.
mature circRNA sequences and exported to primer design tools Cell 2006; 126:37-47; PMID:16839875; http://dx.doi.org/10.1016/j.
(Primer3 or NCBI) to give maximum of 5 sets of primer pairs cell.2006.06.023
10. Glisovic T, Bachorik JL, Yong J, Dreyfuss G. RNA-binding proteins
for PCR products of sizes ranging from 120–200 bp and span- and post-transcriptional gene regulation. FEBS Lett 2008; 582:1977-
ning the junction (Fig. S6A, B). For circRNAs smaller than 86; PMID:18342629; http://dx.doi.org/10.1016/j.febslet.2008.03.004
200 nt, users are given a choice for the length of PCR amplicon 11. Abdelmohsen K, Kuwano Y, Kim HH, Gorospe M. Posttranscrip-
based on circRNA length. tional gene regulation by RNA-binding proteins during oxidative
stress: implications for cellular senescence. Biol Chem 2008; 389:243-
55; PMID:18177264; http://dx.doi.org/10.1515/BC.2008.022
CircRNA siRNA design 12. Abdelmohsen K, Gorospe M. Posttranscriptional regulation of cancer
traits by HuR. WIRES RNA 2010; 1:214-29; PMID:21935886; http://
siRNAs spanning circRNA junction were designed using crite- dx.doi.org/10.1002/wrna.4
ria for 21-nt siRNAs described previously.46,47 To design the 13. Gao FB, Taylor JP. RNA-binding proteins in neurological disease.
Brain Res 2012; 1462:1-2; PMID:22682432; http://dx.doi.org/10.1016/
21-nt siRNA target sequence spanning the junction, we used a j.brainres.2012.05.038
minimum of 5 nt on either side of the junction. 14. Gerstberger S, Hafner M, Ascano M, Tuschl T. Evolutionary conserva-
tion and expression of human RNA-binding proteins and their role in
human genetic disease. Syst Biol RNA Binding Proteins 2014; 825:1-
Availability 55; PMID:25201102; http://dx.doi.org/10.1007/978-1-4939-1221-6_1
15. Castello A, Fischer B, Hentze MW, Preiss T. RNA-binding proteins in
The database is freely accessible through the web server at Mendelian disease. Trends Genet 2013; 29:318-27; PMID:23415593;
http://circinteractome.nia.nih.gov. http://dx.doi.org/10.1016/j.tig.2013.01.004
16. Abdelmohsen K, Gorospe M. RNA-binding protein nucleolin in dis-
ease. RNA Biol 2012; 9:799-808; PMID:22617883; http://dx.doi.org/
Disclosure of potential conflicts of interest 10.4161/rna.19718
17. Ule J, Jensen KB, Ruggiu M, Mele A, Ule A, Darnell RB. CLIP identi-
No potential conflicts of interest were disclosed. fies Nova-regulated RNA networks in the brain. Science 2003;
42 D. B. DUDEKULA ET. AL.

302:1212-5; PMID:14615540; http://dx.doi.org/10.1126/ 35. Ciafre SA, Galardi S. microRNAs and RNA-binding proteins A com-
science.1090095 plex network of interactions and reciprocal regulations in cancer.
18. Hafner M, Landthaler M, Burger L, Khorshid M, Hausser J, Berninger RNA Biol 2013; 10:935-43; PMID:23696003; http://dx.doi.org/
P, Rothballer A, Ascano M Jr, Jungkamp AC, Munschauer M, et al. 10.4161/rna.24641
Transcriptome-wide identification of RNA-binding protein and 36. Agami R. microRNAs, RNA binding proteins and cancer. Eur J Clin
microRNA target sites by PAR-CLIP. Cell 2010; 141:129-41; Invest 2010; 40:370-4; PMID:20486997; http://dx.doi.org/10.1111/
PMID:20371350; http://dx.doi.org/10.1016/j.cell.2010.03.009 j.1365-2362.2010.02279.x
19. Licatalosi DD, Mele A, Fak JJ, Ule J, Kayikci M, Chi SW, Clark TA, 37. Li JH, Liu S, Zhou H, Qu LH, Yang JH. starBase v2.0: decoding
Schweitzer AC, Blume JE, Wang X, et al. HITS-CLIP yields genome- miRNA-ceRNA, miRNA-ncRNA and protein-RNA interaction net-
wide insights into brain alternative RNA processing. Nature 2008; works from large-scale CLIP-Seq data. Nuc Acids Res 2014; 42:D92-
456:464-U22; PMID:18978773; http://dx.doi.org/10.1038/nature07488 D7; PMID:24297251; http://dx.doi.org/10.1093/nar/gkt1248
20. Konig J, Zarnack K, Rot G, Curk T, Kayikci M, Zupan B, Turner DJ, 38. Lasda E, Parker R. Circular RNAs: diversity of form and function.
Luscombe NM, Ule J. iCLIP reveals the function of hnRNP particles RNA 2014; 20:1829-42; PMID:25404635; http://dx.doi.org/10.1261/
in splicing at individual nucleotide resolution. Nat Struct Mol Biol rna.047126.114
2010; 17:909-U166; PMID:20601959; http://dx.doi.org/10.1038/ 39. Witten JT, Ule J. Understanding splicing regulation through RNA
nsmb.1838 splicing maps. Trends Genet 2011; 27:89-97; PMID:21232811; http://
21. Lewis BP, Burge CB, Bartel DP. Conserved seed pairing, often flanked dx.doi.org/10.1016/j.tig.2010.12.001
by adenosines, indicates that thousands of human genes are micro- 40. Bazzini AA, Johnstone TG, Christiano R, Mackowiak SD, Obermayer
RNA targets. Cell 2005; 120:15-20; PMID:15652477; http://dx.doi.org/ B, Fleming ES, Vejnar CE, Lee MT, Rajewsky N, Walther TC, et al.
10.1016/j.cell.2004.12.035 Identification of small ORFs in vertebrates using ribosome footprint-
22. Yekta S, Shih IH, Bartel DP. MicroRNA-directed cleavage of HOXB8 ing and evolutionary conservation. EMBO J 2014; 33:981-93;
mRNA. Science 2004; 304:594-6; PMID:15105502; http://dx.doi.org/ PMID:24705786; http://dx.doi.org/10.1002/embj.201488411
10.1126/science.1097434 41. Anderson DM, Anderson KM, Chang CL, Makarewich CA, Nelson
23. Abdelmohsen K, Hutchison ER, Lee EK, Kuwano Y, Kim MM, BR, McAnally JR, Kasaragod P, Shelton JM, Liou J, Bassel-Duby R,
Masuda K, Srikantan S, Subaran SS, Marasa BS, Mattson MP, et al. et al. A micropeptide encoded by a putative long noncoding RNA reg-
miR-375 inhibits differentiation of neurites by lowering HuD levels. ulates muscle performance. Cell 2015; 160:595-606; PMID:25640239;
Mol Cell Biol 2010; 30:4197-210; PMID:20584986; http://dx.doi.org/ http://dx.doi.org/10.1016/j.cell.2015.01.009
10.1128/MCB.00316-10 42. AbouHaidar MG, Venkataraman S, Golshani A, Liu BL, Ahmad T.
24. Vasudevan S, Tong YC, Steitz JA. Switching from repression to activa- Novel coding, translation, and gene expression of a replicating cova-
tion: MicroRNAs can up-regulate translation. Science 2007; 318:1931- lently closed circular RNA of 220 nt. Proc Natl Acad Sci USA 2014;
4; PMID:18048652; http://dx.doi.org/10.1126/science.1149460 111:14542-7; PMID:25253891; http://dx.doi.org/10.1073/
25. Panda AC, Sahu I, Kulkarni SD, Martindale JL, Abdelmohsen K, pnas.1402814111
Vindu A, Joseph J, Gorospe M, Seshadri V. miR-196b-mediated trans- 43. Martinez-Salas E, Lozano G, Fernandez-Chamorro J, Francisco-Velilla
lation regulation of mouse insulin2 via the 5 ’ UTR. PLoS ONE 2014; R, Galan A, Diaz R. RNA-binding proteins impacting on internal initi-
9:e101084; PMID:25003985 ation of translation. Intl J Mol Sci 2013; 14:21705-26; PMID:24189219;
26. Lee Y, Kim M, Han J, Yeom KH, Lee S, Baek SH, Kim VN. MicroRNA http://dx.doi.org/10.3390/ijms141121705
genes are transcribed by RNA polymerase II. EMBO J 2004; 23:4051- 44. Grimson A, Farh KKH, Johnston WK, Garrett-Engele P, Lim LP, Bar-
60; PMID:15372072; http://dx.doi.org/10.1038/sj.emboj.7600385 tel DP. MicroRNA targeting specificity in mammals: determinants
27. Borchert GM, Lanier W, Davidson BL. RNA polymerase III tran- beyond seed pairing. Mol Cell 2007; 27:91-105; PMID:17612493;
scribes human microRNAs. Nat Struct Mol Biol 2006; 13:1097-101; http://dx.doi.org/10.1016/j.molcel.2007.06.017
PMID:17099701; http://dx.doi.org/10.1038/nsmb1167 45. Untergasser A, Cutcutache I, Koressaar T, Ye J, Faircloth BC,
28. Kim VN, Han J, Siomi MC. Biogenesis of small RNAs in animals. Nat Remm M, Rozen SG. Primer3-new capabilities and interfaces.
Rev Mol Cell Biol 2009; 10:126-39; PMID:19165215; http://dx.doi.org/ Nuc Acids Res 2012; 40; PMID:22730293; http://dx.doi.org/
10.1038/nrm2632 10.1093/nar/gks596
29. Newman MA, Hammond SM. Emerging paradigms of regulated 46. Ngoc BT, Ho TB, Saori K. A Descriptive method for generating siRNA
microRNA processing. Genes Dev 2010; 24:1086-92; PMID:20516194; design rules. Lect Notes Comput Sc 2013; 7803:196-205; http://dx.doi.
http://dx.doi.org/10.1101/gad.1919710 org/10.1007/978-3-642-36543-0_21
30. Fabian MR, Sonenberg N, Filipowicz W. Regulation of mRNA transla- 47. Naito Y, Yamada T, Ui-Tei K, Morishita S, Saigo K. siDirect: highly
tion and stability by microRNAs. Ann Rev Biochem 2010; 79:351-79; effective, target-specific siRNA design software for mammalian RNA
PMID:20533884; http://dx.doi.org/10.1146/annurev-biochem- interference. Nuc Acids Res 2004; 32:W124-W9; PMID:15215364;
060308-103103 http://dx.doi.org/10.1093/nar/gkh442
31. Bartel DP. MicroRNAs: target recognition and regulatory functions. 48. Glazar P, Papavasileiou P, Rajewsky N. circBase: a database for circu-
Cell 2009; 136:215-33; PMID:19167326; http://dx.doi.org/10.1016/j. lar RNAs. RNA 2014; 20:1666-70; PMID:25234927; http://dx.doi.org/
cell.2009.01.002 10.1261/rna.043687.113
32. Srikantan S, Tominaga K, Gorospe M. Functional interplay between 49. Kuhn RM, Karolchik D, Zweig AS, Wang T, Smith KE, Rosenbloom
RNA-binding protein HuR and microRNAs. Curr Prot Pept Sci 2012; KR, Rhead B, Raney BJ, Pohl A, Pheasant M, et al. The UCSC genome
13:372-9; PMID:22708488; http://dx.doi.org/10.2174/ browser database: update 2009. Nuc Acids Res 2009; 37:D755-D61;
138920312801619394 PMID:18996895; http://dx.doi.org/10.1093/nar/gkn875
33. Tominaga K, Srikantan S, Lee EK, Subaran SS, Martindale JL, Abdel- 50. Mokrejs M, Masek T, Vopalensky V, Hlubucek P, Delbos P, Pospisek
mohsen K, Gorospe M. Competitive regulation of nucleolin expres- M. IRESite-a tool for the examination of viral and cellular internal
sion by HuR and miR-494. Mol Cell Biol 2011; 31:4219-31; ribosome entry sites. Nuc Acids Res 2010; 38:D131-D6;
PMID:21859890; http://dx.doi.org/10.1128/MCB.05955-11 PMID:19917642; http://dx.doi.org/10.1093/nar/gkp981
34. Kim HH, Kuwano Y, Srikantan S, Lee EK, Martindale JL, Gorospe M. 51. Kozomara A, Griffiths-Jones S. miRBase: annotating high confidence
HuR recruits let-7/RISC to repress c-Myc expression. Gen Dev 2009; microRNAs using deep sequencing data. Nuc Acids Res 2014; 42:
23:1743-8; PMID:19574298; http://dx.doi.org/10.1101/gad.1812509 D68-D73; PMID:24275495; http://dx.doi.org/10.1093/nar/gkt1181