miR-634 is a Pol III-dependent intronic microRNA regulating alternative-polyadenylated isoforms of its host gene PRKCA

miR-634 is a Pol III-dependent intronic microRNA regulating alternative-polyadenylated isoforms of its host gene PRKCA

    miR-634 is a Pol III-dependent intronic microRNA regulating alternativepolyadenylated isoforms of its host gene PRKCA Elvezia Maria P...

894KB Sizes 0 Downloads 9 Views

    miR-634 is a Pol III-dependent intronic microRNA regulating alternativepolyadenylated isoforms of its host gene PRKCA Elvezia Maria Paraboschi, Giulia Cardamone, Valeria Rimoldi, Stefano Duga, Giulia Sold`a, Rosanna Asselta PII: DOI: Reference:

S0304-4165(17)30064-8 doi:10.1016/j.bbagen.2017.02.016 BBAGEN 28778

To appear in:

BBA - General Subjects

Received date: Revised date: Accepted date:

15 June 2016 2 February 2017 13 February 2017

Please cite this article as: Elvezia Maria Paraboschi, Giulia Cardamone, Valeria Rimoldi, Stefano Duga, Giulia Sold`a, Rosanna Asselta, miR-634 is a Pol III-dependent intronic microRNA regulating alternative-polyadenylated isoforms of its host gene PRKCA, BBA - General Subjects (2017), doi:10.1016/j.bbagen.2017.02.016

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT miR-634 is a Pol III-dependent intronic microRNA regulating alternativepolyadenylated isoforms of its host gene PRKCA

IP

T

Elvezia Maria Paraboschia, Giulia Cardamonea, Valeria Rimoldia,b, Stefano Dugaa,b, Giulia Soldàa,b, Rosanna Asseltaa,b a

SC R

Department of Biomedical Sciences, Humanitas University, Via Manzoni 113, 20089 Rozzano, Milan, Italy; b Humanitas Clinical and Research Center, Via Manzoni 56, 20089 Rozzano, Milan, Italy.

NU

Running title: miR-634 down-regulates its host gene.

MA

Key words: miR-634, PRKCA, promoter, target gene, multiple sclerosis.

CE P

TE

D

Corresponding author: Prof. Rosanna Asselta, PhD Associate Professor of Medical Genetics Department of Biomedical Sciences Humanitas University Via A. Manzoni 113 - 20089, Rozzano (Milano), Italy phone: +39 02 82245215; fax: +39 02 82245290 email: [email protected]

AC

Conflict of interest statement: The Authors declare no conflicts of interest.

1

ACCEPTED MANUSCRIPT ABSTRACT Background: The protein kinase C alpha (PRKCA) gene, coding for a Th17-cell-selective kinase, shows a complex splicing pattern, with at least 2 stable alternative transcripts

associated

with

several

conditions,

including

multiple

sclerosis,

asthma,

IP

were

T

characterized by an alternative upstream polyadenylation site. Polymorphisms in this gene

schizophrenia, and cancer. The presence of a microRNA (miRNA), i.e. miR-634, within

a role in the susceptibility to these pathologies.

SC R

intron 15 of the PRKCA gene, suggests the intriguing possibility that this miRNA might play

Methods: Here, we characterized miR-634 expression profile and searched for its putative

NU

targets using a combination of RT-PCR and gene reporter assays. Results: The quantitative analysis of PRKCA and miR-634 transcripts in a panel of human

MA

tissues and cell lines revealed discordant expression profiles, suggesting the presence of an independent miR-634 promoter and/or a possible direct role of miR-634 in modulating PRKCA expression. Functional studies demonstrated the existence of a miRNA-specific

D

promoter, which was shown to be Pol-III-dependent. Furthermore, transfection

TE

experiments showed that miR-634 is able to target its host gene by specifically downregulating the shorter alternative-polyadenylated isoforms.

CE P

Conclusions: MiR-634 is a Pol III-dependent intronic miRNA, which could target its host gene through a “first-order” negative feedback. General significance: MiR-634 is one of the few characterized examples of Pol-IIIdependent intronic miRNAs. Its independent transcription from the host gene suggests

AC

caution in using expression profiles of host genes as proxies for the expression of the corresponding intronic miRNAs.

2

ACCEPTED MANUSCRIPT INTRODUCTION MicroRNAs (miRNAs) are short (~22 nucleotides in length) non-coding RNAs acting as post-transcriptional regulators of gene expression [Bartel, 2004; Cech and Steitz, 2014].

T

They modulate the expression of multiple target transcripts by inducing either mRNA

IP

translational repression or degradation, and thus represent key players in many, if not all, biological processes. Hence, it is not surprising that alterations in miRNA homeostasis can

SC R

contribute to the pathogenesis of conditions such as cancer, neurologic, cardiovascular, and autoimmune diseases [Brennecke et al., 2003; Xu et al., 2003; Bartel, 2004; Harfe, 2005; Gurtan and Sharp, 2013].

NU

MiRNAs can be encoded in independent transcription units, in polycistronic clusters, or within introns of protein-coding genes. They are transcribed, mostly by RNA Pol II, as

MA

capped and polyadenylated primary miRNAs (pri-miRNAs), which contain extended hairpin structures. The pri-miRNA is subjected to a maturation process that results in the incorporation of the mature miRNA into a miRNA-induced silencing protein complex

D

(miRISC), enabling each miRNA to inhibit the expression of hundreds of target mRNAs

TE

[Esquela-Kerscher and Slack, 2006; Lim et al., 2005]. Target mRNA recognition is based on imperfect complementary binding between miRNAs and their target sites, usually

CE P

located within the 3’ untranslated regions (3’ UTR). Until some years ago, intronic miRNAs were generally thought to be processed from the host-gene transcript, with the intronic miRNA and its host gene showing concordant expression levels, since driven by the same promoter. However, subsequent studies

AC

showed many examples of poor correlation in expression levels between miRNAs and their host genes, a phenomenon that could be easily explained by the presence of specific promoters driving the expression of intronic miRNAs [Monteys et al., 2010; Baskerville and Bartel, 2004; Liang et al., 2007]. Supporting this hypothesis, bioinformatics analyses showed that 30% of intronic miRNAs can be transcribed from RNA Pol II and 5% from Pol III intron-resident promoters. Moreover, it was demonstrated that the expression of several intragenic miRNAs occurs independently from the host-gene transcription unit [Ozsolak et al., 2008; Monteys et al., 2010]. In this work, we focused on the characterization of the human miR-634, a miRNA initially identified by Cummins and colleagues in human colorectal cells [Cummins et al., 2006]. The gene encoding miR-634 (MIR634) is located on chromosome 17q24.2, and maps within intron 15 of the Protein Kinase C alpha (PRKCA) gene (Figure 1), a locus repeatedly associated with several disorders, including multiple sclerosis (MS), 3

ACCEPTED MANUSCRIPT schizophrenia, cancer, and asthma [Paraboschi et al., 2014; Carroll et al., 2010; Lahn et al., 2004; Murphy et al., 2009]. PRKCA codes for a member of the protein kinase C (PKC) family, which comprises related serine/threonine kinases involved in many signal

T

transduction pathways and biological processes [Dempsey et al., 2000]. The PRKCA

IP

transcript is characterized by a complex splicing profile, with at least 4 stable alternative isoforms, two of which resulting from an alternative-polyadenylation event, and hence

SC R

showing a different 3’ UTR (Figure 1) [Paraboschi et al., 2014].

No data are available so far on miR-634 physiological expression profile and functions, and only few targets have been identified (AR, androgen receptor; CYR61, cysteine-rich

NU

angiogenic inducer 61; and PIK3R1, Phosphoinositide-3-Kinase, Regulatory Subunit 1, Alpha) [Östling et al., 2011; Jeansonne et al., 2013; Cui et al., 2016]. As for pathologic

MA

conditions, miR-634 was demonstrated to be down-regulated in chondrocytes of osteoarthritis patients as well as in prostate cancer cell lines; its over-expression was shown to have anti-proliferative effects both against prostate cancer and glioma [Díaz-

D

Prado et al., 2012; Östling et al., 2011; Jeansonne et al., 2013].

TE

In this work, we profiled miR-634 expression pattern and demonstrated that its expression is driven by an intronic Pol III-dependent promoter. Moreover, we found that the

CE P

alternative-polyadenylated PRKCA isoforms could be a target of miR-634, thus suggesting

AC

an important role for this miRNA in the PRKCA post-transcriptional regulation.

4

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Figure 1: Genomic organization of the PRKCA locus. a) Screenshots from the UCSC Genome Browser (http://genome.ucsc.edu/; release Feb. 2009, GRCh37/hg19) of the genomic region corresponding to the PRKCA gene. The upper panel shows, in order: i) the ruler with the scale at the genomic level; ii) chromosome 17 nucleotide numbering; iii) UCSC RefSeqs track; iv) a custom track showing a representation of the PRKCA alternative transcripts. Isoforms characterized by the presence of the canonical UTR are indicated as VL transcripts; those deriving from an alternative polyadenylation event are named VS. Both VL and VS isoforms may include in their mature form an alternative exon 3 (3* or 3* with five additional nucleotides) [Paraboschi et al., 2015]. The isoforms carrying a premature termination codon (PTC), leading to nonsense-mediate mRNA degradation (NMD), are colored in red. Below the main panel, a close-up view of the genomic region surrounding the MIR634 gene is shown (genomic coordinates: chr17:64,781,892-64,783,316). The panel shows, in order: i) the ruler with the scale at the genomic level; ii) chromosome 17 nucleotide numbering; iii) UCSC reference genes; iv) the conservation track; and v) the repetitive element track. Among repetitive elements, those present upstream and downstream of the MIR634 gene are the SINE AluY and the LINE L1ME3A, respectively. b) Predicted secondary structure of the miR-634 precursor obtained by miRBase (http://www.mirbase.org/). The sequence of the mature miRNA is bolded.

5

ACCEPTED MANUSCRIPT MATERIALS AND METHODS

Plasmid preparation

T

The promoter expression vector pGL2-basic (containing the luciferase reporter gene), the

IP

pRL-TK (encoding renilla luciferase), the psiCHECK2 (encoding both the firefly and the renilla luciferase), and the pTARGET vectors were purchased from Promega (Madison,

SC R

WI, USA). The psiUx plasmid containing the miR-183 precursor under the control of the U1 promoter (psiUx-MIRN183) [Soldà et al., 2013] and an empty pTB plasmid (kind gift of Dr. E. Buratti, International Centre for Genetic Engineering and Biotechnology, Trieste, Italy)

NU

were used as controls to normalize transfection experiments.

For promoter studies, 982 bp of the putative miR-634 promoter region were cloned into the

MA

pGL2-basic vector, upstream of the firefly luciferase gene. In addition, a 2983-bp-long fragment comprising the putative promoter and the miR-634 hairpin precursor, or the miR634 hairpin alone, were cloned into the pTARGET vector, previously deprived of its own

D

promoter. This same vector was used for over-expressing miR-634 in different cell lines.

TE

As control for miR-634 over-expression experiments, the miR-22-3p precursor was inserted into the psiUx expression vector.

CE P

All the 3’ UTRs putatively targeted by miR-634 (PRKCA VS-3’UTR versions 1 and 2, PRKCA VL-3’UTR, AR, ADRB2, BRWD1, IL11RA, JAG1, MTCP1, TRAF5), as well as the entire coding sequence of the ADRB2 gene, were directionally cloned downstream of the renilla luciferase gene in the psiCHECK2 plasmid.

AC

All constructs were produced by PCR amplifying the relevant genomic region from the DNA of a healthy subject (DNA extracted by standard protocols), using an appropriate PCR primer couple (Supplementary Table 1), and subsequently by cutting the amplified products with the proper restriction enzyme. Restricted products were ligated into the relevant plasmid. The construct carrying the PRKCA 3’UTR deleted of the miRNA binding site (ΔMRE, miRNA recognition element) was obtained by site-directed mutagenesis, by means of the QuikChange kit (Stratagene, La Jolla, CA, USA) and following the manufacturer’s instructions. All recombinant/mutagenized vectors were verified by Sanger sequencing, as described [Paraboschi et al., 2014]. All plasmids were purified using the PureYield™ Plasmid Miniprep System kit (Promega) according to the manufacturer’s protocol.

6

ACCEPTED MANUSCRIPT Cell cultures and transfection experiments HeLa and HEK293 cells were cultured in Dulbecco's modified Eagle medium containing 2 mM L-glutamine, 10% fetal bovine serum and antibiotics (100 U/ml penicillin and 100

T

μg/ml streptomycin; Euroclone, Wetherby, UK) and grown at 37°C in a humidified

IP

atmosphere of 5% CO2 and 95% air, according to the standard procedures. For promoter studies, HeLa cells were cotransfected using: i) 3.9 µg of the pGL2-basic

SC R

recombinant vector together with 100 ng of the control plasmid pRL-TK; or ii) 3.5 µg of pTARGET recombinant plasmid together with 0.5 µg of the control psiUx-MIRN183 vector. For the miRNA-target interaction analysis, HeLa and HEK293 cells were cotransfected

NU

using: i) 3.5 µg of the pTARGET plasmid expressing miR-634 together with 0.5 µg of the pTB plasmid; or ii) 300 ng of the pTARGET vector expressing miR-634 together with 700 ng of the psiCHECK2 plasmid containing the relevant target 3’ UTR.

MA

In each experiment, an equal number of cells (2.5*105 for HeLa and 3*105 for HEK293) were transfected with the FuGENE 6 reagent (Roche, Basel, Switzerland) in 6-well plates,

D

as described by the manufacturer. All transfection experiments (promoter or miRNA

TE

analysis) included three biological replicates, each performed at least with two technical replicates. On average, we obtained a transfection efficiency of 30%. Depending on the

CE P

measurement test performed at the end of the experiment, we obtained either total RNA, or cell lysates from transfected cells (see below).

Luciferase assays

AC

The activities of firefly/renilla luciferase were measured 48h after transfection in cell lysates by using the Dual-Luciferase Reporter Assay System (Promega) and the Wallac 1420 VICTOR3 V reader (PerkinElmer, Waltham, MA, USA). For promoter studies, the values of firefly luciferase were normalized against the corresponding values of renilla luciferase (all values expressed as relative luminescent units) prior to making comparisons between test groups. For miRNA-target interaction assays, the values of renilla luciferase were normalized against the corresponding values of firefly luciferase.

Modulation of RNApol III-mediated transcription For RNA polymerase III inhibition assays, HEK293 cells were plated at a density of 3x105 cells in six-well plates and treated with ML-60218 (Calbiochem, San Diego, CA, USA) at

7

ACCEPTED MANUSCRIPT the final concentration of 50 µM. Untreated samples were incubated with the drug solvent (DMSO). RNA extraction was performed 24 hours after the treatment. To induce increased expression levels of Alu RNA, HEK293 cells were submitted to heat-

T

shock or cycloheximide treatments [Liu et al., 1995; Gu et al., 2009]. For the heat-shock

IP

treatment, cells were heated by putting a 25-cm2-flask, containing 5 mL of medium, in a water bath at 45°C for 30 minutes. The heat shock was followed by an incubation period of

SC R

1 hour at 37°C, before performing RNA extraction. As for the cycloheximide treatment, cells were plated at a density of 5.5x106 per 10-cm dish and, after 72h, treated for 8 hours with cycloheximide (100 µg/mL). Untreated samples were incubated with the drug solvent.

NU

After the treatment, cells were washed twice with phosphate buffered saline and total RNA

MA

was extracted.

RNA samples

Expression profiles of miR-634 and PRKCA isoforms were determined using RNA from

D

peripheral blood mononuclear cells (PBMCs) derived from a healthy individual, a panel of

TE

21 human tissues (Ambion, Austin, USA), and 9 cell lines. Differential expression of miR-634 was tested in PBMCs derived from a cohort of 26 MS

CE P

relapsing-remitting patients (i.e., suffering from the most common form of MS) [Milo and Miller, 2014] and 26 healthy subjects. Patients were all in remitting phase, they had not received any immunomodulatory therapy within the month prior to blood withdrawal. Controls were age- and sex-matched with patient. PBMCs were isolated immediately after

AC

phlebotomy from heparinized blood, by centrifugation on a Lympholyte Cell separation media (Cederlane Laboratories Limited, Hornby, Canada) gradient. RNA from both PBMCs and cell lines was isolated using the Eurozol kit (Euroclone). RNA concentration/quality was assessed using the Nanodrop ND-1000 (Thermo Fisher Scientific, Waltham, MA, USA). This study was approved by local Ethical Committees and was performed according to the Declaration of Helsinki and to the Italian legislation on sensible data recording. Signed informed consent was obtained from all participants.

Semi-quantitative real-time RT-PCR For the evaluation of the expression of specific transcripts, random hexamers and the Superscript-III Reverse Transcriptase (Invitrogen, Carlsbad, CA, USA) were used to perform first-strand cDNA synthesis starting from 1 µg of RNA extracted from transfected 8

ACCEPTED MANUSCRIPT cells, or RNA derived from a panel of 22 human tissues (Ambion Inc, Austin, TX, USA). From a total of 20 µL of the reverse-transcription (RT) reaction, 1 µL was used as template for amplifications using the FastStart SYBR Green Master Mix (Roche) on a LightCycler

T

480 (Roche), following a touchdown thermal protocol. Expression levels were normalized

IP

using HMBS (hydroxymethylbilane synthase gene) and ACTB (β-actin) as housekeeping genes.

SC R

For the comparative analysis of the two PRKCA isoforms VS and VL, the amplification efficiency of the corresponding real-time RT-PCR assays was tested by generating standard curves, using serial dilutions of the cDNA prepared from PBMCs of a control

NU

individual.

Commercial stem-loop qRT-PCR assays from Life Technologies (Foster City, CA, USA) used

to

quantitate:

i)

mature

miR-634

(TaqMan

microRNA

Assay

ID

MA

were

Hs04273340_s1); ii) snoRNA U43 (used as housekeeping gene; TaqMan microRNA Assay ID 001095); iii) mature miR-183 (used as transfection control; Assay ID

D

Hs04331526_s); iv) U6 snRNA (used as positive control in Pol III-inhibition assays; Assay

TE

ID 001973); v) mature miR-9 (used as negative control in Pol III-inhibition assays, as well as in cycloheximide and heat-shock treatments; Assay ID 000583); and vi) mature miR-

CE P

517a (used as positive control in cycloheximide and heat-shock treatments; Assay ID 002402) Assays were performed by reverse-transcribing 600 ng of total RNA in a 20-μL reaction volume, using the ImProm-II Reverse Transcriptase (Promega). Up to a total of 4 μL of the RT reaction were used for subsequent amplifications, using the FastStart

AC

TaqMan Probe Master (Roche) and a thermal protocol (including 40 amplification cycles), according to the manufacturer. In all cases, real-time qRT-PCR assays were performed with at least three technical replicates on a LightCycler 480, and expression levels were analyzed by the GeNorm software [Vandesompele et al., 2002]. Primer couples used in these assays are listed in Supplementary Table 1. Thermal profiles can be provided on request.

Computational and statistical analyses To predict potential miR-634 targets, we used microRNA.org [Betel et al., 2008] and PITA [Kertesz et al., 2007] programs, as well as the miRWalk2 suite [Dweep et al., 2011]. Among the programs implemented in miRWalk2, we specifically used: miRWalk2.0, RNA22, miRanda, miRDB, TargetScan, PICTAR, and Diana-microT. The RNAhybrid software [Rehmsmeier et al., 2004] was used to visualize the miRNA-mRNA interactions. 9

ACCEPTED MANUSCRIPT Statistical analysis was performed using the R software (http://www.r-project.org/). Values were expressed as mean ± standard deviation (SD) or standard error (SEM) for parametric data. Difference between groups was evaluated using student’s t-test. Correlation between

T

miR-634 and PRKCA expression profiles was calculated using the Pearson’s correlation.

IP

Pearson’s coefficients <-0.5 and >0.5 were considered as anti-correlation and positive

AC

CE P

TE

D

MA

NU

SC R

correlation, respectively. P values <0.05 were considered as statistically significant.

10

ACCEPTED MANUSCRIPT RESULTS

miR-634 and PRKCA display discordant expression profiles

T

The expression profiles of miR-634 and its host gene were evaluated in a panel of 22

IP

human tissues and 9 cell lines by semi-quantitative real-time RT-PCR assays (Figure 2; Supplementary figure 1). Due to the presence of multiple PRKCA transcripts, showing two

SC R

different 3’ UTRs (Figure 1), specific RT-PCR assays were designed to detect: i) only transcripts using the canonical polyadenylation site (long isoforms, VL); and ii) only transcripts containing the alternative 3’UTR (short isoforms, VS). This analysis showed that

NU

PRKCA is ubiquitously expressed (with the VS isorform representing 1-3% of the transcripts in most tissues), whereas miR-634 displays a marked tissue-specific

MA

expression, being expressed at highest levels in small intestine, prostate, and heart, and being barely detectable in 5 of the 22 analyzed tissues. No significant correlation between the VL and VS PRKCA isoforms and miR-634 expression levels was evident (Pearson’s

D

correlation coefficients <-0.07). These data suggested that: i) miR-634 expression could

TE

be driven by a specific promoter, independent from the PRKCA one; ii) PRKCA and miR634 are transcribed from the same promoter, but tissue-specific factors may influence

CE P

miRNA processing post-transcriptionally; iii) at least in some tissues, miR-634 could act as a negative regulator of PRKCA, by targeting one or both its alternative 3’UTRs: and/or iv) a

AC

combination of the above-mentioned hypotheses.

11

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 2: Expression profiles of miR-634 in human tissues and cell lines. MiR-634 expression levels were evaluated by semi-quantitative real-time RT-PCRs in a panel of commercially available human tissues (upper panel) and in human cell lines (lower panel) using a TaqMan commercial stem-loop assay. In all cases, expression levels are shown as normalized rescaled values, setting as 1 the value measured in the ovary and HeLa cells, respectively (indicated in black). Bars represent means + SD of three technical replicates.

12

ACCEPTED MANUSCRIPT miR-634 is transcribed from its own promoter To demonstrate the existence of a specific promoter driving the expression of miR-634, we cloned about 1 kb of the putative promoter region in the pGL2-basic reporter vector,

T

upstream of the firefly luciferase gene (Figure 3a). After transfection in HeLa cells, we

IP

performed a luciferase reporter gene assay. This experiment showed that the 1-kb-long promoter is able to drive a significantly higher luciferase expression (about 35 folds) than

SC R

the empty vector, used as control (p<0.05) (Figure 3a).

To confirm these results, we cloned a 3-kb region containing the miR-634 putative promoter together with the precursor hairpin into the mammalian expression vector

NU

pTARGET. As control, the miR-634 precursor alone (97-bp long) was inserted into a promoterless pTARGET vector as well (Figure 3b). After transfection in HeLa cells, we

MA

performed a real-time RT-PCR assay to assess miR-634 expression levels. Data were normalized by cotransfecting in each well a vector expressing the MIRN183 gene [Soldà et al., 2012]. The mature miR-634 was highly expressed in cells transfected with the vector

D

containing the 3-kb promoter upstream of the miR-634 hairpin (P=0.0047), whereas only

TE

basal levels of miR-634 were measured in cells transfected with the miR-634 precursor, compatible with miR-634 levels endogenously expressed in HeLa cells (Figure 3b; Figure

AC

CE P

2).

13

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 3: Mir-634 expression is driven by its own promoter. a) Luciferase-based reporter assays. On the left, a schematic representation of the vectors used in transfection experiments is shown: the pGL2-basic plasmid does not contain any promoter sequence, whereas the pGL2-(1-kb-long) vector contains 982 bp of the putative promoter region of MIR634 upstream of the luciferase reporter gene. Transfection experiments were performed in HeLa cells; 48 hours after transfection, cells were harvested and lysates prepared to perform the reporter assays. The right panel shows the 14

ACCEPTED MANUSCRIPT

D

MA

NU

SC R

IP

T

normalized luciferase activity measured in extracts of HeLa cells transfected with either plasmid. As control, the activity values of the PRKCA promoter (the one characterized by 14 repeats) [Paraboschi et al., 2014] are also shown. Bars represent means + SEM of three independent experiments (biological replicates), each performed in triplicate (technical replicates). b) Semi-quantitative real-time RT-PCR assays. On the top, the plasmids used in transfection experiments are presented: the pTargeT-MIR634 vector contains only the premiRNA sequence, whereas the pTargeT-(3-kb-long) vector contains the pre-miRNA sequence proceeded by 2983 bp of the putative promoter region. As control, we used the psiUx plasmid containing the MIRN22 gene under the control of the U1 promoter. Transfection experiments were performed in HeLa cells; 24 hours after transfection, cells were harvested and total RNA extracted. The right panel shows miR-634 expression levels (measured by using a Taqman commercial stem-loop RT-PCR assay) as normalized rescaled values, setting as 1 the value measured in the cells transfected with the precursor miR-634 alone (not expressing the miRNA). As control, we used the psiUx empty vector (not expressing any miRNA), and the psiUx construct expressing miR-22 under the control of the U1 promoter. Below each histogram, the mean threshold cycles (Ct) are indicated. Bars represent means + SEM of three independent experiments (biological replicates), each performed in triplicate (technical replicates). Significance levels of t-tests are shown. *: P<0.05; **: P<0.01.

TE

miR-634 is transcribed by Pol III

Having demonstrated the existence of a miR-634 specific promoter, we used the

CE P

Encyclopedia of DNA Elements (ENCODE) data [ENCODE Project Consortium, 2012], available through UCSC Genome Browser (http://genome.ucsc.edu/), to search in silico for the presence of putative miR-634 promoter features, i.e.: i) CpG islands; ii) H3K4Me1,

AC

H3K4Me3, and H3K27Ac histone modification marks; iii) chromatin immunoprecipitation and sequencing (ChIP-seq) data in HeLa cells indicating the interaction with specific transcription factors; and iv) presence of Pol-II transcription start sites (TSS). Except for weak H3K4Me1, H3K4Me3, and H3K27Ac signals, indicating the presence of an actively transcribed region, we were not able to evidence the presence of any typical element characterizing Pol-II promoters (see Supplementary figure 2, where a comparison with the Pol-II PRKCA promoter is present). We hence searched for typical elements characterizing Pol-III promoters [Schramm and Hernandez, 2002] in a 400-bp-long region upstream of the pre-miR-634 sequence. We found all the main features of a type-2 Pol-III promoter, i.e., the absence of a TATA sequence, the presence of an A box (consensus: TRGCNNARYNGG) and of a B box (consensus: GGTTCRANNCC) (Figure 4a and b) [Geiduschek and Tocchini-Valentini, 1988]. 15

ACCEPTED MANUSCRIPT To confirm these in-silico predictions, we performed a Pol-III inhibition assay, by treating HEK293 cells with the ML-60218 inhibitor. We observed a significant reduction in the expression levels of both miR-634 and U6 snRNA (this last one used as positive control).

T

In particular, miR-634 expression level was reduced to 51% (P=0.0008) respect to the

IP

untreated sample. Conversely, no effects of the Pol-III inhibitor were observed for the PolII dependent miR-9 and PRKCA transcripts (Figure 4c, left panel).

SC R

It has been suggested that miRNA genes interspersed among Pol-III transcribed SINE repeats (e.g. Alu and MIR sequences) could “take advantage” of Pol-III recruitment to the same regions [Schanen and Li, 2011]. As a further support of a Pol-III-mediated

NU

transcription, miR-634 lies in a genomic context enriched in SINE elements (Figure 1). Hence, we either treated HEK293 cells with cycloheximide or submitted them to a heat-

MA

shock treatment (both procedures are known to induce Pol-III mediated transcriptions of SINE elements) [Gu et al., 2009]. These treatments demonstrated a significant upregulation of miR-634 (2.98 and 2.43 fold, respectively; P<0.05), but not of its Pol-II

D

dependent host gene (Figure 4c, middle and right panels). As expected, the Pol-II-

TE

dependent miR-9 expression remained unchanged upon both treatments, whereas we observed an increment in the expression levels of miR-517a, which has been

AC

CE P

demonstrated to be transcribed through its upstream Alu [Borchert et al., 2006].

16

AC

CE P

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

Figure 4: MiR-634 is transcribed by Pol III. a) The PRKCA/MIR634 sequence (genomic coordinates: chr17:64,782,790-64,783,370; genome assembly GRCh37/hg19, Feb. 2009) is reported. It comprises 403 bp upstream of the pre-miR-634 gene (i.e., the putative Pol-III promoter), 97 bp of MIR634, and 11 bp downstream of the pre-miR-634 gene. PRKCA exon 15 is highlighted in dark grey, whereas the pre-miR-634 region is indicated in light grey. The predicted Box A and Box B are bolded and underlined (asterisks indicate the mismatches respect to the consensus sequences, both reported below the PRKCA/MIR634 sequence). Stretches of TTTT (terminators) are also indicated. b) Schematic representations of the typical type-2 Pol-III promoter (tRNA) and of the 17

ACCEPTED MANUSCRIPT

NU

SC R

IP

T

putative miR-634 promoter are reported. The main elements (boxes A and B, terminators, as well as the hairpin corresponding to pre-miR-634) are indicated together with their relative positions. The two schemes are approximately to scale up to the first terminator. c) Expression levels of PRKCA (all transcripts) and MIR634 were measured in HEK293 cells untreated (NT), treated for 24 hours with the RNA Pol III inhibitor ML-60218, for 8 hours with cycloheximide, or submitted to a heat-shock treatment for 30 minutes. In all assays positive and negative controls were also analyzed (U6 has a Pol III promoter; miR9 and PRKCA have a Pol II promoter; miR-517a is transcribed through its upstream Alu). Endogenous levels of all the selected genes were measured by semi-quantitative real-time RT-PCRs. Bars represent means + SEM of three independent experiments (biological replicates), each performed in duplicate (technical replicates). Significance levels of t-tests are shown. *: P<0.05; **: P<0.01; ***: P<0.001.

miR-634 directly downregulates its host gene

MA

Beside the fact that miR-634 is transcribed from an independent promoter, another possible explanation for the poor correlation between PRKCA and miR-634 expression levels could be related to a direct action of miR-634 in targeting PRKCA isoforms. To verify

D

this hypothesis, we first in-silico searched for miR-634 target sites in both the 6707-bp-long

TE

3’UTR characterizing the major transcript (VL) and the 4301-bp-long alternative 3’UTR characterizing the PRKCA shorter isoforms (VS) (Figure 1). Multiple miR-634 target sites were predicted for both UTRs (for instance, using the PITA program and a cut-off value for

CE P

G duplex<12, we predicted 4 sites for VL and 5 for VS, respectively). Among predicted sites, those showing the lowest free energy of miRNA-target interactions were located at chromosome positions chr17:64,803,508-64,803,530 and chr17:64,772,057-64,772,076

AC

for the VL- and VS-3’UTR, respectively (Figure 5a). We hence evaluated the effect of the miR-634 over-expression on the regulation of PRKCA isoforms. To this aim, we analyzed the expression levels of PRKCA VL and VS isoforms upon miR-634 over-expression by specific semi-quantitative real-time RT-PCR assays in HeLa cells. Cells were transfected with either the miR-634 hairpin precursor or the miR-634 precursor preceded by its promoter region. The pTB plasmid was cotransfected and used as internal control to normalize for transfection efficiency. We obtained an over-expression of miR-634 of at least 200 fold respect to its endogenous basal levels. In these experiments, a significant reduction of the PRKCA VS expression levels was observed (32% reduction, P=0.0027), whereas the VL levels resulted unaffected by miR-634 over-expression (Figure 5b). As control, we repeated the same experiments by over-expressing miR-22-3p: no down-regulation was observed for both the VL and VS transcripts, which do not contain any predicted miR-22-3p responsive element (Figure 5b). 18

ACCEPTED MANUSCRIPT To confirm these results, both PRKCA 3’UTRs were cloned downstream of the luciferase gene in the psiCHECK2 vector. We cotransfected in HeLa cells each of these plasmids together with the pTARGET vector expressing the MIR634 gene. Since the full-length VS-

T

3’UTR resulted highly unstable in our analysis (no luciferase activity measured irrespective

IP

of miR-634 over-expression; data not shown), we decided to clone in the psiCHECK2 vector a smaller region (169 bp), containing the best putative recognition site for miR-634

SC R

(Supplementary figure 3). The results of transfection experiments substantially confirmed that miR-634 is able to target only the PRKCA VS-3’UTR (21.5% reduction; P=0.024) (Figure 4c). The direct regulation of PRKCA VS-3’UTR was confirmed by the deletion of

NU

the best-predicted target site (in the 169-bp insert) by site-directed mutagenesis, which

AC

CE P

TE

D

MA

resulted in the complete abrogation of the response to the miRNA (Figure 5c).

19

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Figure 5: Mir-634 targets PRKCA VS isoforms. a) Predicted miR-634 target sites within the 3’UTR of PRKCA VL and VS isoforms. Duplexes are shown with the miR-634 sequence above the target sequence. The RNAhybrid software was used to visualize the miRNA-mRNA interactions. The calculated mfe (minimum free energy) is also reported. b) The pTargeT-MIR634 vector, containing only the pre-miR-634 sequence (mock experiment), as well as the pTargeT-(3-kb-long) vector, containing the pre-miRNA sequence proceeded by about 3000 bp of the miR-634 promoter region (miR-634 overexpression condition), were independently transfected in HeLa cells; 24 hours after transfection, cells were collected and total RNA extracted. Expression levels of PRKCA VL, and VS isoforms were measured by semi-quantitative real-time RT-PCRs (a scheme of the PCR-amplified regions is displayed, together with the positions of primers, on the right). For control, experiments were repeated by over-expressing the miR-22 precursor, which is not predicted to target VL and VS. In all cases, expression levels are shown as normalized rescaled values, setting as 1 the value measured in cell transfected with the precursor miR-634 alone. Bars represent means + SEM of three independent experiments (biological replicates), each performed in triplicate (technical replicates). c) On the left, a schematic representation of the vectors used in transfection experiments is shown: the VL and VS 3’UTRs were independently cloned in the psiCHECK vector, downstream of the luciferase reporter gene. The MRE-VS version of the plasmid does not contain the miR-634 recognition element (MRE). Each of these plasmids was cotransfected with the pTargeT-(3-kb-long) vector (in order to obtain miR-634 overexpression) in HeLa cells; 48 hours after transfection, cells were collected and lysates prepared to perform the reporter assays. The right panel shows the normalized luciferase activity measured in extracts of HeLa cells transfected with each plasmid. Bars represent 20

ACCEPTED MANUSCRIPT

T

means + SEM of three independent experiments (biological replicates), each performed in triplicate (technical replicates). Significance levels of t-tests are shown. *: P<0.05; **: P<0.01; ns: not significant.

IP

Searching for novel miR-634 targets

In the attempt to find additional targets of miR-634, we selected for functional validation

SC R

seven candidates that were predicted as potential miR-634 targets by at least 4 bioinformatics programs, and, based on literature data, were demonstrated to be involved in immunity, inflammation processes, or have been associated with MS or other complex

NU

diseases with an immune/inflammatory component. The list of candidate genes is presented in Supplementary Table 2.

MA

We hence evaluated in HEK293 the effect of miR-634 over-expression on the regulation of the selected candidates (all endogenously expressed in HEK293) by means of semiquantitative real-time RT-PCR assays. As control, we tested the AR transcript, a known

D

target of miR-634 [Östling et al., 2011]. Besides AR, a significant reduction in expression

TE

levels was observed only in the case of the ADRB2 transcript (80% reduction, P<0.0001) (Figure 6a). Subsequently, we cloned in the psiCHECK2 vector the whole-length 3’UTR of

CE P

all selected genes. The obtained plasmids were independently transfected in HEK293 cells under miR-634 over-expression. The subsequent reporter assays did not reveal any significant change in the luciferase activity levels for the selected genes, including the ADRB2 gene (Figure 6b).

AC

The inconsistency observed for ADRB2 could be explained by an “indirect effect” of miR634, which may target other transcripts that, in turn, exert their functions on ADRB2. A second explanation could stem on the presence of miR-634 target sites in the coding sequence (CDS) of the ADRB2 gene. Bioinformatics analyses evidenced the presence of three putative MREs for miR-634 in the ADRB2 CDS. We hence cloned in the psiCHECK2 vector the whole ADRB2 CDS, and repeated miR-634 over-expression experiments in HEK293 cells. Once again, we did not observe any significant variation in luciferase activity (Figure 6b).

21

TE

D

MA

NU

SC R

IP

T

ACCEPTED MANUSCRIPT

AC

CE P

Figure 6: Functional study of predicted miR-634 targets. a) The pTargeT-MIR634 vector, containing only the pre-miR-634 sequence (mock experiment), as well as the pTargeT-(3-kb-long) vector, containing the pre-miRNA sequence proceeded by about 3000 bp of the miR-634 promoter region (miR-634 overexpression condition), were independently transfected in HEK293 cells; 24 hours after transfection, cells were collected and total RNA extracted. Expression levels of PRKCA VL, VS, as well as nine selected putative miR-634 targets were measured by semi-quantitative real-time RT-PCRs (the specific primer couples used in these assays are listed in Supplementary table 1). The AR transcript was used as control, being a known target of miR-634 [Östling et al., 2011]. In all cases, expression levels are shown as normalized rescaled values, setting as 1 the value measured in cell transfected with the precursor miR-634 alone. Bars represent means + SEM of three independent experiments (biological replicates), each performed in triplicate (technical replicates). b) Left panel: luciferase reporter assay data from HEK293 cells transfected with the psiCHECK2 vector coupled to the 3’UTR regions of selected putative miR-634 targets are shown. Each of these psiCHECK2 recombinant plasmid was cotransfected with the pTargeT-(3-kb-long) vector (in order to obtain miR-634 over-expression); 48 hours after transfection, cells were collected and lysates prepared to perform the reporter assays. Right panel: luciferase reporter assays were repeated by using the the psiCHECK2 vector coupled to the entire ADRB2 coding sequence (CDS). As positive controls, the psiCHECK2 vectors coupled to the 3’UTR of AR and VS UTRs were also transfected. Bars represent means + SEM of three independent experiments (biological replicates), each performed in triplicate (technical replicates). Significance levels of t-tests are shown. *: P<0.05; **: P<0.01; ***: P<0.0001. 22

ACCEPTED MANUSCRIPT MiR-634 is not dysregulated in MS patients Considering that miR-634 regulates PRKCA short isoforms, whose relative expression levels were demonstrated to be significantly different in MS patients [Paraboschi et al.,

T

2014], we verified whether it could be differentially expressed in a cohort of MS patients

IP

with respect tocontrols. To this aim, we evaluated miR-634 expression levels in PBMCs of 26 relapsing-remitting MS patients (free from treatments) and an equal number of healthy

SC R

controls by performing semi-quantitative real-time RT-PCRs. MiR-634 resulted not

AC

CE P

TE

D

MA

NU

differentially expressed between cases and controls (P=0.21). (Supplementary Figure 4).

23

ACCEPTED MANUSCRIPT DISCUSSION Initially identified by Cummins and colleagues in human colorectal cells [2006], the intronresident miR-634 has been so-far poorly studied, though embedded within a gene,

T

PRKCA, that has been associated with several disorders, including MS [Barton et al.,

IP

2004; Saarela et al., 2006; Paraboschi et al., 2014]. Our previous studies on the regulation of expression of PRKCA and its potential role in MS [Paraboschi et al., 2014], led us to

SC R

attempt an in-depth characterization of this hominoid-lineage specific miRNA. In particular, we first profiled miR-634 expression pattern in a panel of human tissues/cell lines (Figure 2). The poor correlation observed between miR-634 and PRKCA expression

NU

levels well fit the notion that “young” intragenic miRNAs are less coexpressed with their host genes than the evolutionary-conserved ones [He et al., 2012]. This observation

MA

suggested the possibility that miR-634 transcription could be driven by an independent promoter. We hence verified this hypothesis by cloning the putative promoter region in appropriate expression vectors, and subsequently by evaluating its transcriptional activity

D

in eukaryotic cells both by real-time RT-PCR and by classic luciferase-based reporter

TE

assays. Our findings strongly support the existence of a specific miR-634 promoter (Figure 3). Interestingly, our data overlap those reported by Ozsolak and colleagues [2008], who,

CE P

by combining nucleosome mapping with chromatin signatures for promoters, managed to identify the proximal promoters of 175 human miRNAs. Among them, miR-634 was predicted to have an independent intronic promoter, but no further characterization was attempted.

AC

The lack of the typical features characterizing a Pol-II promoter (Supplementary figure 2a) was suggestive of a Pol-III-mediated transcription of miR-634. Though not very common, Pol-III miRNA transcription has been observed in several cases [Borchert et al., 2006; Monteys et al., 2010]. We verified the presence of all the main characteristics of a type-2 Pol-III promoter for the MIR634 gene (Figure 4a and b) [Geiduschek and TocchiniValentini, 1988]. Another feature of Pol III genes is the TTTT termination signal, which must be associated with an upstream RNA hairpin structure, formed by the body of the transcript itself, to ensure an efficient elongation complex destruction [Nielsen et al., 2013]. In the case of MIR634, we propose that the pre-miR hairpin might possibly facilitate the termination of its own transcription together with the TTTT stretch located 7 bp downstream (Figure 4b). This mechanism of termination of transcription has been suggested for instance for tRNA transcripts, which are characterized by extensive secondary structures [Zenkin, 2014]. Notably, our in-silico predictions received support by 24

ACCEPTED MANUSCRIPT all the experiments aimed at modulating the Pol-III driven expression of miR-634 (Figure 4c). MiR-634 independent transcription from its host gene suggests carefulness in using expression profiles of host genes as proxies for the expression of the associated intronic

T

miRNAs. This is particularly relevant when expression profiles of host genes are used to

IP

predict miRNA target genes [He et al., 2012].

To date, miR-634 has been experimentally demonstrated to target only three genes (AR,

SC R

CYR61, and PIK3R1) [Östling et al., 2011; Jeansonne et al., 2013; Cui et al., 2016]. However, none of these studies dealt with autoimmune processes or were related to MS. We

hence

bioinformatically

prioritized

a

list

of

seven

genes

involved

in

NU

immune/inflammatory pathways to be tested as targets of miR-634. None of the putative miR-634 target sites were validated by luciferase-based assays, though a preliminary, very

MA

encouraging, significant signal emerged - after miR-634 over-expression and RT-PCR assays - for the ADRB2 gene (80% reduction in HEK293 cells) (Figure 6). We tried to explain this striking inconsistency extending our luciferase-based assays to the coding

D

sequence, having observed the presence of three putative MREs for miR-634 in the

TE

ADRB2 CDS. In addition, the ADRB2 gene is characterized by a 3’UTR of “only” 561 nucleotides (GenBank accession number: NM_000024), thus reinforcing the notion of a

CE P

possible miR-634/ADRB2 interaction at the CDS level. In fact, a computational analysis of CDS target sites, based on mammalian high-throughput immunoprecipitation and sequencing data, revealed that genes with 3'UTRs shorter than 500 nucleotides show a significantly higher CDS target score [Reczko et al., 2012]. Unfortunately, our miR-634

AC

over-expression experiments did not evidence any significant decrease in the luciferase activity levels for ADRB2, sustaining a “indirect effect” of the miRNA in controlling the ADRB2 transcript levels (Figure 6a). For instance, it has been demonstrated in prostate cancer cells that the inhibition of the miR-634-target AR gene can induce repressive epigenetic programs via direct activation of the H3K27 methyltransferase EZH2, which in turn targets genes such as ADRB2 [Yu et al., 2010]. This kind of regulatory circuits have not yet been clarified in the central nervous system, though ADRB2 – known to promote glucose uptake in astrocytes – is potentially implicated in the development of neuronal diseases characterized by a dysregulation of astrocyte glucose metabolism (such as MS and Alzheimer's disease) [Dong et al., 2012]. Emerging evidence indicates that intronic miRNAs can either directly target their host gene transcripts or indirectly down-regulate them by targeting host-gene transcription factors: these phenomena are defined as “first-order” and “second-order” negative feedback [Li et 25

ACCEPTED MANUSCRIPT al., 2007]. Indeed, genome-wide bioinformatics analyses of human intronic miRNAs showed that the first-order regulatory mechanism involves up to 20% of intragenic miRNAs [Hinske et al., 2010]. Our data suggest the existence of a first-order negative regulation

T

mediated by miR-634, which specifically targets the alternative-polyadenylated isoforms of

IP

PRKCA (the so-called VS transcripts) (Figure 1). However, it should be underlined that our analysis did not rule out another possible regulatory mechanism, i.e. the possibility that

SC R

miR-634 could affect PRKCA splicing through a reduction of intron 14 retention (thus reducing VS and increasing VL levels).

Alternative polyadenylation (APA) represents a widespread mechanism used to modulate

NU

gene expression, e.g. by regulating the impact of miRNAs on the expression of their target genes [Di Giammartino et al., 2011]. Notably, Hinske and colleagues [2015] found that

MA

host genes that both contain sites for their intronic miRNAs and present an APA-mediated regulation are often linked to signal transduction pathways (i.e. genes for which a tight control of expression is pivotal). PRKCA reconciles with this observation, being a gene

D

coding for a Th17-cell-selective kinase (PKCα). The APA event characterizing PRKCA is

TE

indeed able to change its coding potential, leading to the production of partially-deleted VS protein isoforms [Paraboschi et al., 2014]. Thus, the interaction between the intron-

CE P

resident miR-634 and the VS 3’UTR could be a mechanism exploited by cells to prevent the formation of potentially dangerous PKCα variants. In this frame, the comparable levels of miR-634 measured in MS patients and controls may indicated that in MS cases, characterized by higher levels of the VS transcript [Paraboschi et al., 2014], miR-634 is not

AC

sufficient to finely tune the regulation of this isoform.

Acknowledgments: This work was supported by FISM – Fondazione Italiana Sclerosi Multipla [grant 2008/R/1 to RA]. Dr. Claudia Dall’Osso and Dr. Alessia Donato are acknowledged for their invaluable work, assistance, and enthusiasm. We would also like to thank Dr. Donato Gemmati and Dr. Marta Spreafico for providing blood samples from MS cases and controls.

26

ACCEPTED MANUSCRIPT REFERENCES

1.

Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell.

Cech TR, Steitz JA. The noncoding RNA revolution-trashing old rules to forge new

IP

2.

T

2004;116:281-97.

ones. Cell. 2014;157:77-94.

Brennecke J, Hipfner DR, Stark A, Russell RB, Cohen SM. bantam encodes a

SC R

3.

developmentally regulated microRNA that controls cell proliferation and regulates the proapoptotic gene hid in Drosophila. Cell. 2003;113:25-36.

Xu P, Vernooy SY, Guo M, Hay BA. The Drosophila microRNA Mir-14 suppresses

NU

4.

cell death and is required for normal fat metabolism. Curr Biol. 2003;13:790-5. Harfe BD. MicroRNAs in vertebrate development. Curr Opin Genet Dev.

MA

5.

2005;15:410-5. 6.

Gurtan AM, Sharp PA. The role of miRNAs in regulating gene expression networks.

Esquela-Kerscher A, Slack FJ. Oncomirs - microRNAs with a role in cancer. Nat Rev

Cancer. 2006;6:259-69.

Lim LP, Lau NC, Garrett-Engele P, Grimson A, Schelter JM, Castle J, Bartel DP,

CE P

8.

TE

7.

D

J Mol Biol. 2013;425:3582-600.

Linsley PS, Johnson JM. Microarray analysis shows that some microRNAs downregulate large numbers of target mRNAs. Nature. 2005;433:769-73. 9.

Monteys AM, Spengler RM, Wan J, Tecedor L, Lennox KA, Xing Y, Davidson BL.

AC

Structure and activity of putative intronic miRNA promoters. RNA. 2010;16:495-505. 10. Baskerville S, Bartel DP. Microarray profiling of microRNAs reveals frequent coexpression with neighboring miRNAs and host genes. RNA. 2005;11:241-7. 11. Liang Y, Ridzon D, Wong L, Chen C. Characterization of microRNA expression profiles in normal human tissues. BMC Genomics. 2007;8:166. 12. Ozsolak F, Poling LL, Wang Z, Liu H, Liu XS, Roeder RG, Zhang X, Song JS, Fisher DE. Chromatin structure analyses identify miRNA promoters. Genes Dev. 2008;22:317283. 13. Cummins JM, He Y, Leary RJ, Pagliarini R, Diaz LA Jr, Sjoblom T, Barad O, Bentwich Z, Szafranska AE, Labourier E, Raymond CK, Roberts BS, Juhl H, Kinzler KW, Vogelstein B, Velculescu VE. The colorectal microRNAome. Proc Natl Acad Sci USA. 2006;103:3687-92.

27

ACCEPTED MANUSCRIPT 14. Paraboschi EM, Rimoldi V, Soldà G, Tabaglio T, Dall'Osso C, Saba E, Vigliano M, Salviati A, Leone M, Benedetti MD, Fornasari D, Saarela J, De Jager PL, Patsopoulos NA, D'Alfonso S, Gemmati D, Duga S, Asselta R. Functional variations modulating PRKCA

T

expression and alternative splicing predispose to multiple sclerosis. Hum Mol Genet.

IP

2014;23:6746-61.

15. Carroll LS, Williams NM, Moskvina V, Russell E, Norton N, Williams HJ, Peirce T,

SC R

Georgieva L, Dwyer S, Grozeva D, Greene E, Farmer A, McGuffin P, Morris DW, Corvin A, Gill M, Rujescu D, Sham P, Holmans P, Jones I, Kirov G, Craddock N, O'Donovan MC, Owen MJ. Evidence for rare and common genetic risk variants for schizophrenia at protein

NU

kinase C, alpha. Mol Psychiatry. 2010;15:1101-11.

16. Lahn M, Sundell K, Gleave M, Ladan F, Su C, Li S, Ma D, Paterson BM, Bumol TF.

MA

Protein kinase C-alpha in prostate cancer. BJU Int. 2004;93:1076-81. 17. Murphy A, Tantisira KG, Soto-Quirós ME, Avila L, Klanderman BJ, Lake S, Weiss ST, Celedón JC. PRKCA: a positional candidate gene for body mass index and asthma. Am J

D

Hum Genet. 2009;85:87-96.

TE

18. Dempsey EC, Newton AC, Mochly-Rosen D, Fields AP, Reyland ME, Insel PA, Messing RO. Protein kinase C isozymes and the regulation of diverse cell responses. Am

CE P

J Physiol Lung Cell Mol Physiol. 2000;279:L429-38. 19. Östling P, Leivonen SK, Aakula A, Kohonen P, Mäkelä R, Hagman Z, Edsjö A, Kangaspeska S, Edgren H, Nicorici D, Bjartell A, Ceder Y, Perälä M, Kallioniemi O. Systematic analysis of microRNAs targeting the androgen receptor in prostate cancer

AC

cells. Cancer Res. 2011;71:1956-67. 20. Jeansonne D, Pacifici M, Lassak A, Reiss K, Russo G, Zabaleta J, Peruzzi F. Differential Effects of MicroRNAs on Glioblastoma Growth and Migration. Genes. 2013;4:46-64. 21. Cui X, Wang S, Cai H, Lin Y, Zheng X, Zhang B, Xia C. Overexpression of microRNA-634 suppresses survival and matrix synthesis of human osteoarthritis chondrocytes by targeting PIK3R1. Sci Rep. 2016;6:23117. 22. Díaz-Prado S, Cicione C, Muiños-López E, Hermida-Gómez T, Oreiro N, FernándezLópez C, Blanco FJ. Characterization of microRNA expression profiles in normal and osteoarthritic human chondrocytes. BMC Musculoskelet Disord. 2012;13:144. 23. Soldà G, Robusto M, Primignani P, Castorina P, Benzoni E, Cesarani A, Ambrosetti U, Asselta R, Duga S. A novel mutation within the MIR96 gene causes non-syndromic

28

ACCEPTED MANUSCRIPT inherited hearing loss in an Italian family by altering pre-miRNA processing. Hum Mol Genet. 2012;21:577-85. 24. Liu WM, Chu WM, Choudary PV, Schmid CW. Cell stress and translational inhibitors

T

transiently increase the abundance of mammalian SINE transcripts. Nucleic Acids Res.

IP

1995;23:1758-65.

25. Gu TJ, Yi X, Zhao XW, Zhao Y, Yin JQ. Alu-directed transcriptional regulation of

SC R

some novel miRNAs. BMC Genomics. 2009 Nov 30;10:563.

26. Milo R, Miller A. Revised diagnostic criteria of multiple sclerosis. Autoimmun Rev. 2014;13:518-24.

NU

27. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De Paepe A, Speleman F. Accurate normalization of real-time quantitative RT-PCR data by geometric

MA

averaging of multiple internal control genes. Genome Biol. 2002;3:RESEARCH0034. 28. Betel D, Wilson M, Gabow A, Marks DS, Sander C. The microRNA.org resource: targets and expression. Nucleic Acids Res. 2008;36:D149-53.

D

29. Kertesz M, Iovino N, Unnerstall U, Gaul U, Segal E. The role of site accessibility in

TE

microRNA target recognition. Nat Genet. 2007;39:1278-84. 30. Dweep H, Sticht C, Pandey P, Gretz N. miRWalk-database: prediction of possible

2011;44:839-47.

CE P

miRNA binding sites by "walking" the genes of three genomes. J Biomed Inform.

31. Rehmsmeier M, Steffen P, Hochsmann M, Giegerich R. Fast and effective prediction of microRNA/target duplexes. RNA. 2004;10:1507-17.

AC

32. ENCODE Project Consortium. An integrated encyclopedia of DNA elements in the human genome. Nature. 2012;489:57-74. 33. Schramm L, Hernandez N. Recruitment of RNA polymerase III to its target promoters. Genes Dev. 2002;16:2593-620. 34. Geiduschek EP, Tocchini-Valentini GP. Transcription by RNA polymerase III. Annu Rev Biochem. 1988;57:873-914. 35. Schanen BC, Li X. Transcriptional regulation of mammalian miRNA genes. Genomics. 2011;97:1-6. 36. Gu TJ, Yi X, Zhao XW, Zhao Y, Yin JQ. Alu-directed transcriptional regulation of some novel miRNAs. BMC Genomics. 2009 Nov 30;10:563 37. Barton A, Woolmore JA, Ward D, Eyre S, Hinks A, Ollier WE, Strange RC, Fryer AA, John S, Hawkins CP, Worthington J. Association of protein kinase C alpha (PRKCA) gene with multiple sclerosis in a UK population. Brain. 2004;127:1717-22. 29

ACCEPTED MANUSCRIPT 38. Saarela J, Kallio SP, Chen D, Montpetit A, Jokiaho A, Choi E, Asselta R, Bronnikov D, Lincoln MR, Sadovnick AD, Tienari PJ, Koivisto K, Palotie A, Ebers GC, Hudson TJ, Peltonen L. PRKCA and multiple sclerosis: association in two independent populations.

T

PLoS Genet. 2006;2:e42.

IP

39. He C, Li Z, Chen P, Huang H, Hurst LD, Chen J. Young intragenic miRNAs are less coexpressed with host genes than old ones: implications of miRNA-host gene coevolution.

SC R

Nucleic Acids Res. 2012;40:4002-12.

40. Ozsolak F, Poling LL, Wang Z, Liu H, Liu XS, Roeder RG, Zhang X, Song JS, Fisher DE. Chromatin structure analyses identify miRNA promoters. Genes Dev. 2008;22:3172-

NU

83.

41. Borchert GM, Lanier W, Davidson BL. RNA polymerase III transcribes human

MA

microRNAs. Nat Struct Mol Biol. 2006;13:1097-101.

42. Nielsen S, Yuzenkova Y, Zenkin N. Mechanism of eukaryotic RNA polymerase III transcription termination. Science. 2013;340:1577-80.

D

43. Zenkin N. RNA secondary structure-dependent termination of transcription. Cell

TE

Cycle. 2014;13:3-4.

44. Reczko M, Maragkakis M, Alexiou P, Grosse I, Hatzigeorgiou AG. Functional

CE P

microRNA targets in protein coding sequences. Bioinformatics. 2012;28:771-6. 45. Yu J, Yu J, Mani RS, Cao Q, Brenner CJ, Cao X, Wang X, Wu L, Li J, Hu M, Gong Y, Cheng H, Laxman B, Vellaichamy A, Shankar S, Li Y, Dhanasekaran SM, Morey R, Barrette T, Lonigro RJ, Tomlins SA, Varambally S, Qin ZS, Chinnaiyan AM. An integrated

AC

network of androgen receptor, polycomb, and TMPRSS2-ERG gene fusions in prostate cancer progression. Cancer Cell. 2010;17:443-54. 46. Dong JH, Chen X, Cui M, Yu X, Pang Q, Sun JP. β2-adrenergic receptor and astrocyte glucose metabolism. J Mol Neurosci. 2012;48:456-63. 47. Li SC, Tang P, Lin WC. Intronic microRNA: discovery and biological implications. DNA Cell Biol. 2007;26:195-207. 48. Hinske LC, Galante PA, Kuo WP, Ohno-Machado L. A potential role for intragenic miRNAs on their hosts' interactome. BMC Genomics. 2010;11:533. 49. Di Giammartino DC, Nishida K, Manley JL. Mechanisms and consequences of alternative polyadenylation. Mol Cell. 2011;43:853-66. 50. Hinske LC, Galante PA, Limbeck E, Möhnle P, Parmigiani RB, Ohno-Machado L, Camargo AA, Kreth S. Alternative polyadenylation allows differential negative feedback of human miRNA miR-579 on its host gene ZFR. PLoS One. 2015;10:e0121507. 30

ACCEPTED MANUSCRIPT Highlights for Review

MiR-634 is an intragenic miRNA that is transcribed independently from its host gene



MiR-634 is one of the few characterized examples of Pol-III-dependent miRNAs



MiR-634 targets, among others, some alternative isoforms of its host gene



ADRB2, promoting glucose uptake in astrocytes, may be a miR-634 indirect target

AC

CE P

TE

D

MA

NU

SC R

IP

T



31