Download What do microRNAs mean for rheumatoid arthritis?

ARTHRITIS & RHEUMATISM
Vol. 64, No. 1, January 2012, pp 11–20
DOI 10.1002/art.30651
© 2012, American College of Rheumatology
REVIEW
What Do MicroRNAs Mean for Rheumatoid Arthritis?
Isabelle Duroux-Richard,1 Christian Jorgensen,2 and Florence Apparailly2
from worms to mammals and function as negative
regulators of gene expression. In animals, miRNAs are
involved in the regulation of multiple biologic pathways
by posttranscriptionally repressing the expression of
protein-encoding genes. The miRNAs achieve this by
base-pairing to target messenger RNAs (mRNAs), thus
influencing either their stability or rate of translation.
Human miRNA genes are nonrandomly located
throughout all chromosomes except the Y chromosome
(1), are predominantly observed in introns (70%), and
are frequently observed in cancer-associated genomic
regions or fragile sites (52.5%) (2). Half of the known
human miRNAs are found in clusters that may be
functionally related by their targeting of the same gene
or different genes that are involved in the same metabolic pathway. Interestingly, evolutionarily conserved
miRNAs are thought to be conserved elements of a gene
regulatory network involved in key biologic processes
such as development and apoptosis. Thus, studying
miRNAs in lower organisms should shed much light on
evolutionary developmental biology (3) and human disorders.
In 2009, Hervé Seitz (Université de Toulouse,
Laboratoire de Biologie Moléculaire Eucaryote, Toulouse, France) proposed a hypothesis that could reconcile several contradicting observations regarding the
biologic role of miRNAs in animals (4). Recently, when
Dr. Seitz was asked, “How can your work on model
organisms help us to decide whether miRNAs are true
mega regulators of cellular functions or just mini tuners
with little impact on physiological conditions in humans,” he responded as follows:
“miRNA-mediated repression is usually
modest (!2-fold), which is hard to reconcile with
the robustness of biological pathways to fluctuations in gene expression (e.g., most genes in animals are haplo-sufficient: their physiological activity is not affected by a 2-fold repression in gene
expression). Gene expression is also very variable
between individuals in natural populations—it is
hard to understand how miRNAs could do any-
Introduction
MicroRNAs (miRNAs) are small noncoding
RNA molecules that modulate the expression of multiple protein-encoding genes at the posttranscription
level. They likely participate in nearly every developmental and physiologic process. Although the function
of most mammalian miRNAs has yet to be determined,
it appears that their aberrant expression may play a role
in the pathogenesis of several pathologic conditions,
including immune-mediated inflammatory disorders.
Over the past 3 years, dysregulated expression of a
dozen miRNAs has been reported in patients with
rheumatoid arthritis (RA), in both the circulation and
inflamed synovium; however, the pathogenic role of only
a few of these has been investigated in experimental
mouse models. Here, we provide an overview of the
current understanding of miRNA biogenesis and of the
activities identified for the miRNAs shown to be dysregulated in RA. We also discuss potential benefits to
patients and clinicians, including identification of novel
and much-required molecular biomarkers of RA, as well
as the possible development of miRNA-based biologic
agents.
Current understanding of miRNA biogenesis
MiRNAs are noncoding single-stranded RNAs of
!19–23 nucleotides (nt) in length that are conserved
Supported by INSERM, the University of Montpellier 1, the
Arthritis Foundation Courtin (grant AO 2010), the French Society of
Rheumatology (SFR project 1419), and the European Union Sixth
Framework Programme (project AutoCure; LSHB-CT-2006-018661)
and Seventh Framework Programme (project EuroTraps; HEALTHF2-2008-200923).
1
Isabelle Duroux-Richard, PhD: INSERM U844 and Université Montpellier 1, Montpellier, France; 2Christian Jorgensen, MD,
PhD, Florence Apparailly, PhD: INSERM U844, Université Montpellier 1, and CHU Lapeyronie, Montpellier, France.
Address correspondence to Florence Apparailly, PhD,
INSERM U844, CHU Saint Eloi, Bâtiment INM, 80 Rue Augustin
Fliche, 34295 Montpellier Cedex 5, France. E-mail: florence.apparailly
@inserm.fr.
Submitted for publication April 1, 2011; accepted in revised
form August 23, 2011.
11
12
thing useful if they repress their targets less than
these inter-individual variations. Yet, thousands of
genes exhibit phylogenetically-conserved complementarity to miRNAs in many species, including
humans: that number exceeds, by far, the estimations of the number of dose-sensitive genes. These
observations can be reconciled if you consider that
most of the predicted targets are actually not
functionally targeted by miRNAs, but rather act as
competitive inhibitors; they would be miRNA
modulators, repression miRNAs by titrating them.
The only difference between ”real targets“ and
such competitive inhibitors would be their sensitivity to small changes in gene expression, that depends on the architecture of their biological pathways (negative or positive feedback loops,
stoichiometry of molecular complexes, etc.). Exploring these pathways with systems biology methodologies (assessing quantitatively the consequences of miRNA activity on integrated
phenomena) requires genetic manipulation, which
is only achievable in model organisms.”
In mammals, miRNA genes are transcribed by
RNA polymerase II as monocistronic or polycistronic,
long, primary transcripts named primary miRNA or
pri-miRNA (5,6) that range from !200 nt to several
kilobases in length and are folded into hairpin structures
containing imperfectly base-paired stems. In the nucleus, these pri-miRNAs are processed by the RNase III
type endonuclease Drosha, along with its partner protein DGCR8 (DiGeorge syndrome critical region gene
8), into 70–100-nt–long pre-miRNA (7). The premiRNAs are then transported to the cytoplasm by
exportin 5 (8), where they are cleaved by the RNase III
endonuclease Dicer, along with its partner protein
TRBP (human immunodeficiency virus transactivating
response RNA-binding protein), to produce a miRNA/
miRNA* duplex of 20–22 nt (9). This duplex is unwound
by an helicase, and the miRNA “guide” strand is then
selected for incorporation into the multiproteic RNAinduced silencing complex (RISC) to function as a
so-called mature miRNA, while the “passenger” strand
(miRNA*) is released and degraded (10,11).
High-throughput sequencing studies have provided evidence for both arms of some pre-miRNAs
generating 2 different regulatory small RNAs, suggesting that miRNA* strands may also sometimes regulate
gene expression (12,13). Currently, much debate still
surrounds how the guide strand is selected. The silencing
action of the mature miRNA is mediated by the catalytic
component of the miRISC (RISC with incorporated
DUROUX-RICHARD ET AL
miRNA) complex and a member of the Argonaute
family Ago, together with a number of accessory factors.
It is believed that most miRISC complexes silence gene
expression by mechanisms of either mRNA cleavage or
translational repression, depending on the perfect or
incomplete complementarity between the miRNA guide
strand and its mRNA target. Only Ago2 has been shown
to cleave the targeted mRNA by endonuclease activity,
while other Ago proteins are thought to mediate translational repression (12). In that case, changes in the
expression of targeted genes are evidenced only at the
protein level, without any obvious decrease in mRNA
levels. The miRNA binding sites were originally thought
to be principally located in the 3"-untranslated regions
(3"-UTRs) of target mRNAs; however, recent studies
have suggested that miRNAs might also interact with
complementary sequences residing in the 5"-UTR or the
promoter and even the coding sequence (CDS) regions
of target genes (14–16).
To date, efforts have mainly been focused on the
regulatory function of miRNAs, with little knowledge
being sought on the regulation of expression of the
miRNA genes themselves. Moreover, our current understanding of the precise processing of miRNAs and how
they act to silence target mRNAs remains incomplete
(17), and many exceptions to the rules likely exist.
Our current knowledge on miRNAs and RA
Rheumatoid arthritis is a chronic inflammatory
autoimmune disorder affecting millions of people worldwide. It commonly affects the joints, with progressive
systemic inflammatory manifestations that can then affect many organs throughout the body and lead to
substantial morbidity. Although a better understanding
of the pathophysiology of RA has enabled the development of numerous efficient targeted therapies, RA
pathogenesis remains largely unsolved. The discovery of
a potential link between miRNAs and the pathophysiology of RA has stimulated research toward obtaining a
more integrated view of the networks that are dysregulated in RA and perhaps accessing a key pivotal mediator of pathogenesis.
The first evidence for miRNAs playing a role in
RA emerged in 2007 with the identification, in the
serum of patients with RA, of autoantibodies directed
against GW bodies (18), which are cytoplasmic structures for mRNA storage and/or degradation, including
the RNA interference machinery (19). Three studies
over the following year demonstrated the up-regulation
of specific miRNAs either in the circulation (20) or
MicroRNAs IN RHEUMATOID ARTHRITIS
13
Figure 1. Tissue distribution of microRNAs that are dysregulated in rheumatoid arthritis. CD4! " T cell lymphocytes positive for the CD4
phenotypic marker; RA-FLS $ rheumatoid arthritis fibroblast-like synoviocytes; PBMC $ peripheral blood mononuclear cell; DC $ dendritic cell;
NK $ natural killer cell. Figure 1 was produced using Servier Medical Art.
within the inflamed joints of patients with RA (21,22);
these specific miRNAs are miR-16, miR-132, miR-146a,
and miR-155. Since then, the abnormal expression of a
dozen miRNAs has been documented in patients with
RA (23); most of these miRNAs are up-regulated
(Figure 1).
Not surprisingly, among the first miRNAs investigated in RA samples were miR-146a and miR-155,
both of which are involved in the development of innate
and adaptive immune cells, are up-regulated under
inflammatory conditions and in response to a variety of
microbial components, and are overexpressed in several
immune-mediated inflammatory disorders (24–26).
Both miRNAs are thought to finely tune the immune
response and the inflammatory response through negative feedback loops on molecules downstream of Toll/
interleukin-1 (IL-1) receptor (TIR) signaling (27).
Since 2008, the majority of studies investigating
the expression and role of miRNAs in RA have focused
on miR-146a (20–22,28–30). These studies showed that
miR-146a is strongly up-regulated in synovial tissue,
fibroblast-like synoviocytes (FLS), macrophages, B cells,
and CD3# T cells along the superficial and sublining
layers, as well as in CD4# T cells from the synovial fluid
of patients with RA compared with normal donors and
patients with osteoarthritis (OA). The expression of
miR-146a is also increased in plasma, peripheral blood
mononuclear cells (PBMCs), and CD4# T cells isolated
from patients with RA compared with healthy donors,
but not compared with patients with OA (20,30). Both
locally and systemically, monocyte/macrophages more
than lymphocytes seem to represent the major cellular
source contributing to the increased miR-146a expression levels in patients with RA. Although Niimoto and
colleagues (30) did demonstrate that miR-146a expression colocalizes with IL-17 staining in RA synovium, this
demonstration does not formally prove that miR-146a is
expressed by IL-17–producing T cells (30). Other cell
types, including macrophages, also express IL-17.
Interestingly, using comparative analysis of
miRNA concentrations in various fluids and cell types
from patients with RA and control subjects, Murata et al
showed that miR-146a expression levels in synovial fluid
mimic those in synovial tissue/fibroblasts, and that high
plasma concentrations are inversely correlated with tender joint counts (29). Similarly, Pauley et al observed a
correlation between high miR-146a expression levels
and disease activity based on C-reactive protein and
erythrocyte sedimentation rate values, providing further
support for the potential usefulness of miR-146a as a
biomarker of RA disease activity (20).
Other studies have shown increased expression of
miR-155 in RA synovium compared with samples of
both noninflamed synovial fluid (29) and noninflamed
tissue (22) from patients with OA, which is mainly
attributable to the constitutive expression of miR-155 in
RA FLS and in CD14# cells from synovial fluid. Obtaining a clear picture from studies assessing miR-155
14
expression in the circulation is, however, more cumbersome. Indeed, although miR-155 is detectable in plasma,
the concentrations are comparable between patients
with RA, patients with OA, and healthy donors (29).
However, miR-155 expression levels in PBMCs isolated
from patients with RA are higher than those in patients
with OA and healthy control subjects (20,30). As observed with miR-146a, Pauley et al also showed that the
adherent cells in blood from patients with RA (monocyte/macrophages) expressed higher levels of miR-155
than did the nonadherent fraction (lymphocytes) (20).
The abnormal abundance of miR-16 in blood and
RA joints has also been reported. Similar to the expression of miR-155, the expression levels of miR-16 in
plasma are comparable among patients with RA, patients with OA, and healthy individuals (29), but miR-16
expression levels are higher in PBMCs isolated from
patients with RA versus healthy control subjects (20).
The expression levels of miR-16 in synovial fluid are
higher in patients with RA than in patients with OA.
Because miR-16 is ubiquitously expressed, this observation might reflect higher cellularity within inflamed
tissues and the circulation of patients with RA compared
with control subjects. Pauley et al observed a correlation
between miR-16 expression levels in PBMCs and RA
disease activity (based on the C-reactive protein level
and the erythrocyte sedimentation rate), while Murata
et al showed that plasma miR-16 concentrations were
inversely correlated with the Disease Activity Score in 28
joints (DAS28) (31).
Marked overexpression of another miRNA, miR223, in patients with RA, both systemically and within
inflamed joints, has been shown by 3 different groups of
investigators (29,32,33). Hematopoietically specific, this
miRNA is abundantly expressed in blood and serum.
Although no detectable variations of plasma concentrations were measured in RA samples compared with OA
and control samples (29), Fulci and coworkers showed
that miR-223 is strikingly up-regulated in naive CD4# T
lymphocytes from the blood of patients with RA compared with that from healthy donors (32); however, no
correlation between miR-223 levels in CD4# T cells and
clinical parameters was observed. Nakamachi et al observed stronger expression of miR-223 in RA FLS than
in OA FLS (33), which was then confirmed at the
synovial fluid level by Murata et al (29). Interestingly,
this comparative study by Murata et al on secreted
miRNAs by different cell types also showed that, in
addition to synovial tissue that appears to be the main
source of the miRNAs detected in synovial fluid, mononuclear cells infiltrating the synovial fluid also strongly
DUROUX-RICHARD ET AL
contribute to the levels of miR-223 in synovial fluid.
Finally, Murata et al observed an inverse correlation
between the plasma miR-223 level and the tender joint
count (29).
Results from studies on miR-132 expression levels appear to be conflicting, with levels being increased
in PBMCs isolated from the blood of patients with RA
(20) while being significantly decreased in plasma samples obtained from patients with RA (29) compared with
healthy controls. This apparent discrepancy between
expression levels in plasma and PBMCs isolated from
blood, also found with miR-155 (see above), might be
explained by the fact that not all the miRNAs expressed
by blood cells are necessarily retrieved in plasma.
The dysregulation of other miRNAs has been
shown in RA synovium: miR-124a expression is decreased, while miR-203, miR-133a, miR-142-3p, and
miR-142-5p are up-regulated in RA FLS compared with
OA samples (33,34). The expression of miR-363 and
miR-498a is decreased in CD4# T cells isolated from
the blood and synovial fluid of patients with RA (28).
Importantly, none of the miRNAs identified as abnormally expressed in RA tissues are specific for RA.
According to an abundance of reports in the
literature, the expression of some miRNAs is altered in
human pathologic conditions including immunemediated inflammatory disorders other than RA, but
the main questions remaining are as follows: what does
the altered expression of these miRNAs in RA mean in
terms of the impact on patients, and what is its significance to clinicians and researchers? Based on our current knowledge, we have attempted in the following
sections to provide readers with initial answers to these
fundamental questions.
What hopes do miRNAs carry for RA?
A novel class of biomarkers fulfilling unmet
medical needs in RA? RA is a heterogeneous disorder
with a fluctuating clinical course and an unpredictable
prognosis. Although a large panel of very effective
biotherapies is now available to clinicians, the most
challenging issue remains: the identification of biomarkers for early disease diagnosis and to predict therapeutic
outcome that would enable clinicians to treat RA patients as early as possible with the most optimal biologic
therapy. Among molecules that are able to fulfill this
requirement, miRNAs certainly represent a strong candidate. Numerous studies in cancer demonstrate that
miRNAs represent highly specific and sensitive noninvasive biomarkers that are able to provide miRNA-
MicroRNAs IN RHEUMATOID ARTHRITIS
based fingerprints for disease diagnosis or to predict
drug response. Correlations between miRNA expression
levels and the development of malignancies, disease
severity and aggressiveness, metastatic potential, therapeutic response, and survival rate have all been reported
in various cancer types (35). In 2008, concomitant to the
first evidence revealing the abnormal expression of
specific miRNAs in RA tissue, the first reports demonstrating the presence and good stability of miRNAs in
body fluids were published (36,37). Thereafter, the
search for unique expression patterns of circulating
miRNAs in RA patients for diagnostic/prognostic purposes became clinically realistic. However, very few
groups of investigators are addressing this issue.
In only one study was investigation of the concentrations of miRNAs in plasma and synovial fluid as
diagnostic markers for RA reported (29). Murata and
colleagues showed that, among the 5 miRNAs investigated (miR-16, miR-132, miR-146a, miR-155, and miR223), only low plasma levels of miR-132 can differentiate
patients with RA from healthy control subjects. However, plasma miR-132 levels cannot discriminate between RA and OA, which means that none of the 5
miRNAs quantified is able to differentiate between RA
and OA.
So far, no plasma miRNA has been formally
identified as being specific for RA and thus useful for
diagnosis. Murata et al did, however, reveal that plasma
levels of miR-16, miR-146a, miR-155, and miR-223 are
inversely correlated with disease activity (tender joint
count and/or DAS28) and thus represent biomarkers for
disease activity that might be useful for treatment followup. Those investigators showed that synovial fluid
levels of miR-16, miR-146a, miR-155, and miR-223 can
discriminate between patients with RA and those with
OA. Although they suggested that the detection of these
4 miRNAs might be useful for RA diagnosis, further
studies comparing these results with those for patients
with other rheumatic diseases and healthy donors should
first be conducted. Pauley et al suggested that analyzing
miRNA expression in purified PBMCs may also be
useful as a biomarker of disease activity, because low
expression levels of miR-16 and miR-146a in PBMCs
correlate with disease inactivity (DAS28 and tender joint
counts) and vice versa, independently of age, ethnicity,
or treatment (20).
High concentrations of cell-free miRNAs originating from the primary tumor have been observed in
the plasma of cancer patients, and tumor-associated
miRNomes appear highly tissue-specific. In RA, one
could thus speculate that abnormal miRNA expression
15
in plasma might reflect rheumatoid synovium, and that
miRNAs specific for pathologic synovial tissue might be
retrievable in the circulation. However, according to the
in-depth study performed by Murata et al, there is no
correlation between plasma miRNA concentrations and
synovial fluid miRNA levels. Those investigators showed
that miRNAs detected in synovial fluid and plasma have
different origins, with the expression pattern of synovial
fluid miRNAs, but not that of plasma, being similar to
that of the miRNAs secreted by the synovial tissue.
Further investigations are required to identify the jointassociated miRNome in the circulation of patients with
RA. Yet, for prognostic purposes, a biomarker does not
necessarily need to be disease specific. In addition,
because RA is a systemic chronic inflammatory disorder
in which organs other than joints may be affected, a
blood-based miRNA signature is of great interest.
The circulation of miRNAs within microparticles
confers high stability due to their resistance to drastic
conditions and protection from endogenous RNase activity and allows their detection thanks to the expression
of tissue-specific markers. This means that one can track
the cellular origin of a miRNA-containing vesicle found
in body fluids (38). Because blood is easily accessible,
plasma or serum miRNA–based diagnostics is of great
interest (39). Despite this, regardless of the disease
studied, very few reports describe the use of miRNAs in
serum or plasma as biomarkers. This is certainly attributable to technical hurdles relating to the extraction,
detection, and quantification of the miRNAs in these
fluids. Indeed, few miRNAs are detectable in serum or
plasma, and they are often recovered at low concentrations, thus limiting their use for the establishment of
solid profiling. Mainly for these technical reasons,
whole-blood profiling should be preferred for future
biomarker discovery programs, as opposed to plasma- or
serum-based signatures alone.
Another way to unravel RA pathophysiology?
The pathogenesis of RA has not yet been fully elucidated. Because the abnormal expression of a dozen
miRNAs has been reported in patients with RA, and
because miRNAs are known to modulate several processes, their dysregulation would be expected to reflect
their pathologic role, as for instance has been shown in
cancer. As such, increasing interest surrounds investigating miRNAs as a novel pathway to shedding light on the
molecular mechanisms involved in RA pathophysiologic
processes. Investigating the role of miRNA dysregulation in RA is, however, not so simple, as each miRNA is
thought to regulate multiple genes, and multiple
16
miRNAs may regulate a single mRNA through multiple
target sites.
Following their discovery in Caenorhabditis elegans, miRNAs were thought to form part of a primitive
immune system resulting from the race between host
and pathogens (40). In accordance with this, plant and
animal cells direct RNA interference against invading
viruses to prevent or slow down viral replication. Conversely, some viruses have evolved to produce viral
proteins or miRNAs that interfere with the host gene
silencing machinery to attenuate antiviral defenses
(41,42). More studies followed showing the contribution
of miRNAs in development (embryology, tissue and
cell differentiation), and diseases such as cancer, with
most of the miRNAs studied being involved in the
control of proliferation, apoptosis, or cell fate. To date,
it is clear that miRNAs are involved in a broad range of
biologic functions, and it is believed that those that are
philogenetically conserved are likely to be of regulatory
importance in key cellular functions.
Some miRNAs are widely expressed, but others
can exhibit tissue-, lineage-, or developmental stage–
specific expression patterns. Because of the impact of
miRNAs on the expression of protein-encoding genes,
their abnormal expression in specific cell types during
disease development should indicate certain gene pathways, or even cells, playing a detrimental role in specific
stages of the disease of interest.
Monsef Benkirane (Institut de Génétique Humaine, Laboratoire de Virologie Moléculaire, Montpellier, France), an expert in immunology and virology,
heads a laboratory aiming at understanding cellular
factors involved in the regulation of human immunodeficiency virus 1 (HIV-1) gene expression (43). Recently,
when Dr. Benkirane was asked, “Knowing that
pathogen-associated molecular patterns are activated in
inflamed RA tissues and that target prediction software
predicts components of the Toll-like receptor pathways
for miRNAs abnormally expressed in RA tissues, how
could the intertwined race existing between host and
pathogens that you showed for HIV-1 replication and
the RNA interference [RNAi] pathway help us to understand RA etiology,” he responded as follows:
“Recent advances have revealed an implication of innate antiviral immunity in many autoimmune disorders. Antiviral immunity is triggered
by innate sensors that induce inflammatory cytokine production required for the activation of
protective innate and adaptive immunity. This
response must be tightly regulated to avoid excessive inflammation. Many susceptibility genes are
DUROUX-RICHARD ET AL
regulators of pathogen-sensing pathways and sensor activation is observed in many autoimmune
disorders including RA, suggesting a role of antiviral innate responses in autoimmunity. However
the implication of viral infection in the etiology of
these disorders is unclear. While viral infection in
a susceptible genetic background has been [shown]
to induce inflammatory disease, aberrant activation of this innate sensor and inflammatory cytokine production in the absence of viral infection is
also observed and results in autoimmune disorders
that resemble viral infection.
Future work will have to define the role of
the RNAi pathway or RNAi effectors in the regulation of endogenous retroelements and the metabolism of endogenous RNA that can trigger
innate sensor activation and inflammation. Recent
work has revealed an important role of the microRNAs in inflammation by targeting pathogens
sensing signaling. Could viral infection-mediated
deregulation of the RNAi pathway result in excessive innate immunity activation that could result in
autoimmune pathology?“
Considering the dozen miRNAs dysregulated in
RA tissues and the few mRNAs validated as their direct
targets, hematopoietic activities clearly predominate,
and most of the miRNAs concerned are involved in
hematopoietic-related cancers (Figure 2). At least half
of them have either high basal expression levels in
tissues (such as miR-16 or miR-223) or play important
roles in the inflammatory response (miR-132, miR-1423p, miR-146a, miR-155, and miR-223), or both. Not
surprisingly, most are also dysregulated in other types of
immune-mediated inflammatory disorders (25,40), thus
appearing to reflect a loss of homeostatic regulation of
inflammatory events rather than an RA-specific pathogenic process. Indeed, several of the miRNAs found to
be dysregulated in RA are common players in the
homeostatic regulation of immune function and inflammation. They are induced by TIR activation in innate
immune cells and target mRNAs encoding components
of the TIR signaling system; thus, they represent components of a negative feedback loop (27).
Typically, miR-146a and miR-155 have been a
particular focus for investigators, because both are induced by proinflammatory stimuli (IL-1!, tumor necrosis factor ", and Toll-like receptors); both are detected
in synovial fibroblasts, T cells and B cells, as well as cells
of myeloid lineage; and both have multiple targets and
represent a link between innate and adaptive immunity.
Numerous publications have demonstrated their critical
MicroRNAs IN RHEUMATOID ARTHRITIS
17
Figure 2. Comprehensive overview of the functional activities targeted by microRNAs that are dysregulated in rheumatoid arthritis. AC9 $
adenylate cyclase 9; LMO2 $ T cell translocation gene 2; E2F1 $ E2F transcription factor 1; CEBP $ cytoplasmic polyadenylation element binding
protein 1; P300 $ E1A binding protein p300; IL13R"1 $ interleukin-13 receptor "1; SOCS1 $ suppressor of cytokine signaling 1; SHIP-1 $ SH2
domain–containing inositol-5"-phosphatase; RIP1 $ receptor-interacting protein 1; ZIC3 $ zinc finger protein 203; HIVEP2 $ human
immunodeficiency virus type I enhancer binding protein 2; ZNF652 $ zinc finger protein 652; ARID2 $ AT-rich interactive domain 2; MCP1 $
chemokine (C-C motif) ligand; CDK-2 $ cyclin-dependent kinase 2; IRAK-1 $ interleukin-1 receptor–associated kinase 1; TRAF6 $ tumor
necrosis factor receptor–associated factor 6; IRF-5 $ interferon regulatory factor 5; FAF-1 $ Fas-associated protein factor 1; FADD $
Fas-associated death domain; PTC1 $ patched homolog 1; Bmi1 $ Bmi1 polycomb ring finger oncogene; CCND1 $ cyclin D1; MYB $ v-myb
myeloblastosis viral oncogene homolog; PDCD4 $ programmed death 4 protein; RASSF5 $ Ras association domain–containing family member 5;
DLK1 $ delta-like 1 homolog; TIA1 $ TIA1 cytotoxic granule-associated RNA binding protein; GSTP1 $ glutathione S-transferase P1;
TAGLN2 $ transgelin 2; CAV1 $ caveolin 1; FSCN1 $ fascin homolog, actin-bundling protein 1; AKT-2 $ v-akt murine thymoma viral oncogene
homolog.
role in hematopoietic differentiation and function (44).
As an example, miR-155 is the prototype of a multifunctional miRNA that controls the development of
inflammatory Th1 and Th17 cells as well as FoxP3#
regulatory T cells, the IgG class switch in B cells, the
maturation of dendritic cells, and the susceptibility of
CD4# T cells to natural Treg cell–mediated suppression (45). The second critical cellular function that is
revealed when searching for the validated target genes
of the miRNAs that are abnormally expressed in RA is
apoptosis/proliferation. The miRNAs concerned (miR16, miR-124a, miR-133, miR-155, miR-203, and miR223) were observed to be dysregulated in synovial fibroblasts and in both circulating and infiltrating PBMCs
known to play a role in the pathogenesis of RA due to
an imbalance between proapoptotic and antiapoptotic
factors (46). Other studies have implicated miRNAs
in cancer; for example, miR-155 has been referred to as
an oncomiR, and miR-223 (the prototype of myeloidspecific miRNA) is well documented in myeloid leukemia (47).
The identification of miRNAs playing a key
causal role in RA pathogenesis would be expected to
lead to the development of a novel efficient therapeutic
strategy. Many more functional analyses would be ini-
tially required before planning to develop miRNA-based
treatments.
Novel therapeutic strategies? Since miRNAs are
a novel class of highly effective, highly specific, and
sometimes even tissue- and developmental stage–
specific regulators of gene expression, they represent
very attractive candidates for the design of innovative
therapeutic strategies in several important human disorders. Considering the dogma that miRNAs are important regulators not only for single genes but also for
whole gene networks, they theoretically present enormous advantages over current drug design strategies.
For clinical development, however, the use of miRNAbased therapeutics would first need to overcome several
major hurdles, mainly those concerning targeted delivery and safety issues (48). In addition, several requirements should be met, as follows: the protection of
pre/antago-miRNAs from serum degradation during
transport; high delivery and uptake by target tissues
and/or cells; and internalization of the vehicle with
release of antago-miRNAs into the cytosol for direct
access to the RNA interference pathway or with introduction of miRNA-expressing vectors into the nucleus
for long-term expression.
Thus, although the in vivo targeting of miRNAs
18
has been validated in many experimental models, mostly
cancer, the greatest challenge now is to achieve effective
inhibition or overexpression of specific miRNAs in
targeted cell types and compartments without inducing
side effects, using clinically applicable technologies. To
overcome these hurdles, different vehicles have been,
and are still being, developed by groups focusing on gene
therapy delivery systems. The growing interest in
miRNAs among rheumatology researchers, especially
concerning their potential as novel therapeutic candidates, should energize the field of gene therapy in RA.
In 2008, Santaris Pharma initiated the first clinical trial aimed at targeting miRNA in vivo. In a doubleblind, randomized, dose-escalation, phase I study, 48
healthy male volunteers were injected with miravirsen
(SPC3649; Santaris Pharma), a locked nucleic acid
(LNA)–based antisense molecule against miR-122 that
was developed for the treatment of hepatitis C virus
(HCV). This liver-specific miR-122 is essential for HCV
replication and, as expected, antagonizing it by injecting
LNA-modified oligonucleotides in chimpanzees with
chronic infection down-regulated IFN-regulated genes
and improved HCV-induced liver pathology (49), thus
providing proof-of-concept for the development of a
new type of antiviral intervention. In 2010, a phase IIa
clinical trial with miravirsen was launched to assess its
safety and tolerability in treatment-naive patients with
chronic HCV infection.
Surprisingly, despite the fact that both preclinical
and clinical studies based on the modulation of miRNA
activity hold great promise for future drug development,
only few publications report data supporting the therapeutic potential of miRNA-based treatment in RA. The
first data concerning the role of miR-15a in a preclinical
model of RA were published in 2009 (50). Using repetitive intraarticular injections of double-stranded miR15a atelocollagen complex into the knee joints of mice
with autoantibody-mediated arthritis, Nagata and colleagues showed that enforced local expression of miR15a was able to induce synovial membrane cell apoptosis
by negatively regulating the local expression of Bcl-2.
However, because the investigators presented no data
on the clinical effect of local miR-15a overexpression in
experimental arthritis, it remains unclear as to whether
this approach represents a valuable design strategy for
treating RA.
In contrast, 2 recently published reports (51,52)
do provide evidence for a miRNA-based therapeutic
strategy in RA. Again, although they are not specific for
RA, both miR-146a and miR-155 have been a particular
focus for investigators, because they are induced by
DUROUX-RICHARD ET AL
inflammatory conditions, are overexpressed in RA, and
play a broad regulatory role in the immune response and
inflammation. As was previously shown in the mouse
model of experimental allergic encephalitis (51), mice
deficient for miR-155 are protected from collageninduced arthritis (CIA) and consequently show an impaired generation of pathogenic autoreactive T cells and
B cells, as assessed by a marked reduction of an anticollagen antibody response and an impaired antigenspecific T cell response (52). However, using the K/BxN
mouse model of arthritis that depends primarily on
innate immune system cells and molecules, the lack of
miR-155 expression only reduced bone erosion without
impacting inflammation, thus suggesting a role of miR155 in osteoclast differentiation.
The same observation (i.e., prevention of joint
destruction but no effect on inflammation) was made
when miR-146a was systemically overexpressed in mice
with CIA (53). No information about the cell type(s)
mediating the effect of the injected miR-146a atelocollagen was provided in that report. Importantly, considering that miR-146a is overexpressed in RA tissues
and plays a key role in inflammatory responses, it would
have been more logical to investigate the clinical effect
of inhibiting miR-146a in mice with CIA rather than
enforcing its expression. Overall, in both studies, modulation of the expression levels of miR-146a or miR-155
in experimental arthritis models proved that both
miRNAs play important roles in disease pathogenesis;
however, neither study convincingly proved that targeting these specific miRNAs would fulfill expectations for
RA treatment.
More knowledge about the functional roles of
other miRNAs is needed before miRNA-based therapeutics for the treatment of RA can undergo further
development. One route that remains insufficiently studied and could provide an alternative strategy for modulating miRNA expression is the regulation of their
transcription. Given that RA is a disease that affects
predominantly female individuals, and that estrogens
are shown to regulate miRNAs in the immune system,
we should also consider sex hormone regulation of
miRNAs in inflammatory responses.
Conclusion
Over the past 3 years, although a dozen miRNAs
have been reported to be dysregulated in RA, most
studies are descriptive, and in-depth functional analyses
are still lacking. Only the very recent studies performed
in animal models of preclinical arthritis have (or have
MicroRNAs IN RHEUMATOID ARTHRITIS
not) provided evidence that some miRNAs actively
contribute to the molecular mechanisms underlying RA.
Thus, although the identification, characterization, and
modulation of miRNA expression in RA is a matter of
great interest, little evidence exists supporting miRNAs
as novel candidates for the development of immunomodulatory drugs. The potential of miRNAs is, however, promising, and rapid progress toward providing
strong proof of concept in preclinical models for the use
of miRNA-based therapeutic strategies in RA should
now be expected. Nevertheless, as with all gene therapy
strategies developed over the past few decades, the
principal hurdles are likely to be more related to vector
limitations than to the identification of a key miRNA in
a specific cellular subset.
Overall, considering specificity and safety issues,
the main challenge now remains the development of
optimized vehicles for in vivo miRNA-based therapies
that avoid off-target effects while providing efficient cell
targeting, and increasing our understanding of the cellular functions of dysregulated miRNAs.
AUTHOR CONTRIBUTIONS
19
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
All authors were involved in drafting the article or revising it
critically for important intellectual content, and all authors approved
the final version to be published.
22.
REFERENCES
23.
1. Lagos-Quintana M, Rauhut R, Lendeckel W, Tuschl T. Identification of novel genes coding for small expressed RNAs. Science
2001;294:853–8.
2. Calin GA, Sevignani C, Dumitru CD, Hyslop T, Noch E, Yendamuri S, et al. Human microRNA genes are frequently located at
fragile sites and genomic regions involved in cancers. Proc Natl
Acad Sci U S A 2004;101:2999–3004.
3. Kosik KS. MicroRNAs tell an evo-devo story. Nat Rev Neurosci
2009;10:754–9.
4. Seitz H. Redefining microRNA targets. Curr Biol 2009;19:870–3.
5. Lee Y, Jeon K, Lee JT, Kim S, Kim VN. MicroRNA maturation:
stepwise processing and subcellular localization. EMBO J 2002;
21:4663–70.
6. Lee Y, Kim M, Han J, Yeom KH, Lee S, Baek SH, et al.
MicroRNA genes are transcribed by RNA polymerase II. EMBO
J 2004;23:4051–60.
7. Lee Y, Ahn C, Han J, Choi H, Kim J, Yim J, et al. The nuclear
RNase III Drosha initiates microRNA processing. Nature 2003;
425:415–9.
8. Yi R, Qin Y, Macara IG, Cullen BR. Exportin-5 mediates the
nuclear export of pre-microRNAs and short hairpin RNAs. Genes
Dev 2003;17:3011–6.
9. Du T, Zamore PD. microPrimer: the biogenesis and function of
microRNA. Development 2005;132:4645–52.
10. Filipowicz W, Bhattacharyya SN, Sonenberg N. Mechanisms of
post-transcriptional regulation by microRNAs: are the answers in
sight? Nat Rev Genet 2008;9:102–14.
11. Hammond SM, Bernstein E, Beach D, Hannon GJ. An RNA-
24.
25.
26.
27.
28.
29.
30.
31.
32.
directed nuclease mediates post-transcriptional gene silencing in
Drosophila cells. Nature 2000;404:293–6.
Ghildiyal M, Xu J, Seitz H, Weng Z, Zamore PD. Sorting of
Drosophila small silencing RNAs partitions microRNA* strands
into the RNA interference pathway. RNA 2010;16:43–56.
Okamura K, Phillips MD, Tyler DM, Duan H, Chou YT, Lai EC.
The regulatory activity of microRNA* species has substantial
influence on microRNA and 3" UTR evolution. Nat Struct Mol
Biol 2008;15:354–63.
Orom UA, Nielsen FC, Lund AH. MicroRNA-10a binds the
5"UTR of ribosomal protein mRNAs and enhances their translation. Mol Cell 2008;30:460–71.
Lytle JR, Yario TA, Steitz JA. Target mRNAs are repressed as
efficiently by microRNA-binding sites in the 5" UTR as in the 3"
UTR. Proc Natl Acad Sci U S A 2007;104:9667–72.
Zhou X, Duan X, Qian J, Li F. Abundant conserved microRNA
target sites in the 5"-untranslated region and coding sequence.
Genetica 2009;137:159–64.
Seitz H. Redefining microRNA targets. Curr Biol 2009;19:870–3.
Bhanji RA, Eystathioy T, Chan EK, Bloch DB, Fritzler MJ. Clinical and serological features of patients with autoantibodies to
GW/P bodies. Clin Immunol 2007;125:247–56.
Gibbings DJ, Ciaudo C, Erhardt M, Voinnet O. Multivesicular
bodies associate with components of miRNA effector complexes
and modulate miRNA activity. Nat Cell Biol 2009;11:1143–9.
Pauley KM, Satoh M, Chan AL, Bubb MR, Reeves WH, Chan EK.
Upregulated miR-146a expression in peripheral blood mononuclear cells from rheumatoid arthritis patients. Arthritis Res
Ther 2008;10:R101.
Nakasa T, Miyaki S, Okubo A, Hashimoto M, Nishida K, Ochi M,
et al. Expression of microRNA-146 in rheumatoid arthritis synovial tissue. Arthritis Rheum 2008;58:1284–92.
Stanczyk J, Pedrioli DM, Brentano F, Sanchez-Pernaute O,
Kolling C, Gay RE, et al. Altered expression of microRNA in
synovial fibroblasts and synovial tissue in rheumatoid arthritis.
Arthritis Rheum 2008;58:1001–9.
Duroux-Richard I, Presumey J, Courties G, Gay S, Gordeladze
JO, Jorgensen C, et al. microRNAs as new player in rheumatoid
arthritis. Joint Bone Spine 2011;78:17–22.
Li L, Chen XP, Li YJ. MicroRNA-146a and human disease. Scand
J Immunol 2010;71:227–31.
Lu LF, Liston A. MicroRNA in the immune system, microRNA as
an immune system. Immunology 2009;127:291–8.
Furer V, Greenberg JD, Attur M, Abramson SB, Pillinger MH.
The role of microRNA in rheumatoid arthritis and other autoimmune diseases. Clin Immunol 2010;136:1–15.
Tsitsiou E, Lindsay MA. microRNAs and the immune response.
Curr Opin Pharmacol 2009;9:514–20.
Li J, Wan Y, Guo Q, Zou L, Zhang J, Fang Y, et al. Altered
microRNA expression profile with miR-146a upregulation in
CD4# T cells from patients with rheumatoid arthritis. Arthritis
Res Ther 2010;12:R81.
Murata K, Yoshitomi H, Tanida S, Ishikawa M, Nishitani K, Ito H,
et al. Plasma and synovial fluid microRNAs as potential biomarkers of rheumatoid arthritis and osteoarthritis. Arthritis Res Ther
2010;12:R86.
Niimoto T, Nakasa T, Ishikawa M, Okuhara A, Izumi B, Deie M,
et al. MicroRNA-146a expresses in interleukin-17 producing T
cells in rheumatoid arthritis patients. BMC Musculoskelet Disord
2010;11:209.
Prevoo ML, van ’t Hof MA, Kuper HH, van Leeuwen MA, van de
Putte LB, van Riel PL. Modified disease activity scores that
include twenty-eight–joint counts: development and validation in a
prospective longitudinal study of patients with rheumatoid arthritis. Arthritis Rheum 1995;38:44–8.
Fulci V, Scappucci G, Sebastiani GD, Giannitti C, Franceschini D,
Meloni F, et al. miR-223 is overexpressed in T-lymphocytes of
20
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
DUROUX-RICHARD ET AL
patients affected by rheumatoid arthritis. Hum Immunol 2010;71:
206–11.
Nakamachi Y, Kawano S, Takenokuchi M, Nishimura K, Sakai Y,
Chin T, et al. MicroRNA-124a is a key regulator of proliferation
and monocyte chemoattractant protein 1 secretion in fibroblastlike synoviocytes from patients with rheumatoid arthritis. Arthritis
Rheum 2009;60:1294–304.
Stanczyk J, Ospelt C, Karouzakis E, Filer A, Raza K, Kolling C, et
al. Altered expression of microRNA-203 in rheumatoid arthritis
synovial fibroblasts and its role in fibroblast activation. Arthritis
Rheum 2011;63:373–81.
White NM, Fatoohi E, Metias M, Jung K, Stephan C, Yousef GM.
Metastamirs: a stepping stone towards improved cancer management. Nat Rev Clin Oncol 2011;8:75–84.
Gilad S, Meiri E, Yogev Y, Benjamin S, Lebanony D, Yerushalmi
N, et al. Serum microRNAs are promising novel biomarkers. PLoS
One 2008;3:e3148.
Mitchell PS, Parkin RK, Kroh EM, Fritz BR, Wyman SK,
Pogosova-Agadjanyan EL, et al. Circulating microRNAs as stable
blood-based markers for cancer detection. Proc Natl Acad Sci
U S A 2008;105:10513–8.
Hunter MP, Ismail N, Zhang X, Aguda BD, Lee EJ, Yu L, et al.
Detection of microRNA expression in human peripheral blood
microvesicles. PLoS One 2008;3:e3694.
Reid G, Kirschner MB, van Zandwijk N. Circulating microRNAs:
association with disease and potential use as biomarkers. Crit Rev
Oncol Hematol 2010. E-pub ahead of print.
Bennasser Y, Chable-Bessia C, Triboulet R, Gibbings D, Gwizdek
C, Dargemont C, et al. Competition for XPO5 binding between
Dicer mRNA, pre-miRNA and viral RNA regulates human Dicer
levels. Nat Struct Mol Biol 2011;18:323–7.
Sarnow P, Jopling CL, Norman KL, Schutz S, Wehner KA.
MicroRNAs: expression, avoidance and subversion by vertebrate
viruses. Nat Rev Microbiol 2006;4:651–9.
Triboulet R, Mari B, Lin YL, Chable-Bessia C, Bennasser Y,
Lebrigand K, et al. Suppression of microRNA-silencing pathway
by HIV-1 during virus replication. Science 2007;315:1579–82.
Sheedy FJ, O’Neill LA. Adding fuel to fire: microRNAs as a new
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
class of mediators of inflammation. Ann Rheum Dis 2008;67 Suppl
III:iii50–5.
Bi Y, Liu G, Yang R. MicroRNAs: novel regulators during the
immune response. J Cell Physiol 2009;218:467–72.
Stahl HF, Fauti T, Ullrich N, Bopp T, Kubach J, Rust W, et al.
MiR-155 inhibition sensitizes CD4# Th cells for TREG mediated
suppression. PLoS One 2009;4:e7158.
Hutcheson J, Perlman H. BH3-only proteins in rheumatoid arthritis: potential targets for therapeutic intervention. Oncogene
2008;27 Suppl 1:S168–75.
Pulikkan JA, Dengler V, Peramangalam PS, Peer Zada AA,
Muller-Tidow C, Bohlander SK, et al. Cell-cycle regulator E2F1
and microRNA-223 comprise an autoregulatory negative feedback
loop in acute myeloid leukemia. Blood 2010;115:1768–78.
Courties G, Presumey J, Duroux-Richard I, Jorgensen C, Apparailly F. RNA interference-based gene therapy for successful
treatment of rheumatoid arthritis. Expert Opin Biol Ther 2009;9:
535–8.
Lanford RE, Hildebrandt-Eriksen ES, Petri A, Persson R, Lindow
M, Munk ME, et al. Therapeutic silencing of microRNA-122 in
primates with chronic hepatitis C virus infection. Science 2010;327:
198–201.
Nagata Y, Nakasa T, Mochizuki Y, Ishikawa M, Miyaki S, Shibuya
H, et al. Induction of apoptosis in the synovium of mice with
autoantibody-mediated arthritis by the intraarticular injection of
double-stranded microRNA-15a. Arthritis Rheum 2009;60:
2677–83.
O’Connell RM, Kahn D, Gibson WS, Round JL, Scholz RL,
Chaudhuri AA, et al. MicroRNA-155 promotes autoimmune
inflammation by enhancing inflammatory T cell development.
Immunity 2010;33:607–19.
Bluml S, Bonelli M, Niederreiter B, Puchner A, Mayr G, Hayer S,
et al. Essential role of microRNA-155 in the pathogenesis of
autoimmune arthritis in mice. Arthritis Rheum 2011;63:1281–8.
Nakasa T, Shibuya H, Nagata Y, Niimoto T, Ochi M. The
inhibitory effect of microRNA-146a expression on bone destruction in collagen-induced arthritis. Arthritis Rheum 2011;63:
1582–90.