SEARCH

SEARCH BY CITATION

Keywords:

  • OSTEOARTHRITIS;
  • KNEE JOINT PAIN;
  • MICRORNAS;
  • MICROGLIAL CELLS

ABSTRACT

  1. Top of page
  2. ABSTRACT
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgments
  9. References

The objective of this study was to examine whether altered expression of microRNAs in central nervous system components is pathologically linked to chronic knee joint pain in osteoarthritis. A surgical animal model for knee joint OA was generated by medial meniscus transection in rats followed by behavioral pain tests. Relationships between pathological changes in knee joint and development of chronic joint pain were examined by histology and imaging analyses. Alterations in microRNAs associated with OA-evoked pain sensation were determined in bilateral lumbar dorsal root ganglia (DRG) and the spinal dorsal horn by microRNA array followed by individual microRNA analyses. Gain- and loss-of-function studies of selected microRNAs (miR-146a and miR-183 cluster) were conducted to identify target pain mediators regulated by these selective microRNAs in glial cells. The ipsilateral hind leg displayed significantly increased hyperalgesia after 4 weeks of surgery, and sensitivity was sustained for the remainder of the 8-week experimental period (F = 341, p < 0.001). The development of OA-induced chronic pain was correlated with pathological changes in the knee joints as assessed by histological and imaging analyses. MicroRNA analyses showed that miR-146a and the miR-183 cluster were markedly reduced in the sensory neurons in DRG (L4/L5) and spinal cord from animals experiencing knee joint OA pain. The downregulation of miR-146a and/or the miR-183 cluster in the central compartments (DRG and spinal cord) are closely associated with the upregulation of inflammatory pain mediators. The corroboration between decreases in these signature microRNAs and their specific target pain mediators were further confirmed by gain- and loss-of-function analyses in glia, the major cellular component of the central nervous system (CNS). MicroRNA therapy using miR-146a and the miR-183 cluster could be powerful therapeutic intervention for OA in alleviating joint pain and concomitantly regenerating peripheral knee joint cartilage. © 2013 American Society for Bone and Mineral Research.


Introduction

  1. Top of page
  2. ABSTRACT
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgments
  9. References

Pain is one of the most frequent causes of physical disability among adults and is the most prominent symptom of osteoarthritis (OA), which affects >20 million Americans.[1] Many studies focus on the molecular mechanisms related to cartilage loss in OA. However, pain may precede cartilage loss, and there is no effective way to relieve OA pain. To treat this common disease optimally, it is essential to gain a greater understanding of the causes and consequences of OA-related pain. Peripheral tissue injury activates nociceptive pathways, and this alteration may contribute to the development and maintenance of OA pain. Nociceptors are located throughout joint components, having been identified in the capsule, ligaments, menisci, periosteum, and subchondral bone.[2] Joint movement activates ion channels at the terminals of sensory nerves and results in membrane depolarization, reflected by the propagation of action potentials toward the central nervous system (CNS). If joint movement becomes painful, the rate of nerve firing increases dramatically and neural and non-neuronal (glial) elements in the CNS interpret these signals as pain. Nociceptive stimuli appear to be related to but fundamentally different from those producing cartilage loss,[3] and it is conceivable that the activity of spinal components may be involved in joint pain. Better appreciation for these processes will facilitate the development of new treatments for OA-mediated pain.

MicroRNAs (miRNAs) are key factors that regulate global gene expression in diverse cellular process. Typically, they bind to the 3′-untranslated region of their target mRNAs and repress protein expression by affecting mRNA translation and/or destabilization.[4] miRNAs have emerged as key regulators of innate and adaptive immune responses,[5-7] inflammation,[8-11] and chronic pain. For example, miR-146a, a critical regulator of inflammatory and immune responses, alters genes linked to toll-like receptor (TLR) signaling, NF-κB pathways, tumor necrosis factor α (TNFα), interleukin-1β (IL-1β), IL-17[5], and regulated on activation, normal T cell expressed and secreted (RANTES).[12] Overexpression of miR-146a in human primary chondrocytes significantly decreases matrix metalloproteinase 13 (MMP13) and a disintegrin and metalloproteinase with thrombospondin motifs 4 (ADAMTS4), two potent cartilage-degrading enzymes.[12]

Accumulating evidence suggests that, at the level of the spinal cord, glial cells mediate the development and maintenance of persistent chronic pain.[13, 14] Glia dynamically modulate sensory functions under pathological conditions, including rheumatoid arthritis.[15-17] Upon activation, glia release inflammatory cytokines (TNFα, IL-1, IL-17) and pain modulators leading to hypersensitivity.[18] Pharmacological attenuation of glial activity may represent a novel approach for controlling neuroinflammatory diseases and neuropathic pain. Importantly, pathologically altered levels of miR-183 family members in the CNS are associated with glial activity and neuropathy.[19]

We hypothesize that OA-induced knee joint hypersensitivity is associated with pathologically aberrant expression patterns of distinct miRNAs in spinal tissue that regulate nociceptive glial activity. In the current study, we demonstrate that dysregulation of miRNA expression in peripheral and central nervous systems is directly linked to knee joint OA-induced hyperalgesia.

Materials and Methods

  1. Top of page
  2. ABSTRACT
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgments
  9. References

Induction of osteoarthritis

Animal handling and procedures used in this study were in agreement with the guidelines of the Rush Institutional Animal Care and Use Committee (IACUC). OA was induced in male Sprague-Dawley rats (250 to 300 g) by medial meniscus transection (MMT).[20] Briefly, rats were anesthetized with 1.5% isoflurane (Abbott Laboratories, North Chicago, IL, USA) in oxygen, and the surgery was performed with aseptic technique. The surgery began with a 1-cm longitudinal skin incision over the patella in the left knee. The medial meniscus was freed from ligaments attached to the tibia, and with serrated forceps the meniscus was elevated about 1 mm. Using iris scissors, the meniscus was cut completely through at a location slightly under the patella. The skin was closed with 4-0 nylon sutures.

In the sham control group, only the skin was open and closed. For comparison, we also generated a neuropathic pain animal model by L5 spinal nerve ligation using procedures we described previously[21] (n = 8 for each group). Throughout the study, the animals were housed in groups of two (animals with surgical induction of OA and control sham group) and were not mixed with any other chronic pain animal group (ie, neuropathy).

Animal behavioral tests

The method of testing for mechanical allodynia (von Frey sensitivity) followed that of Chaplan and colleagues.[22] To obtain consistent results, we allowed animals to adapt to the grid environment for 15 minutes. A calibrated set of von Frey filaments (Stoelting, Wood Dale, IL, USA) was then applied from below to the plantar hind paw to determine the 50% force withdrawal threshold using an iterative method. The filament forces ranged from 0.04 to 15 g, beginning with 2.0 g. The filament was applied to the skin with enough pressure to buckle and was maintained for up to 6 seconds. A brisk lifting of the foot was recorded as a positive response. If no response was observed, the filament with the next highest force was applied, whereas the filament with the next lowest force was applied upon a positive response. All behavioral tests were performed by the same technician who was blinded to the study groups and identification of animals in order to avoid subjective differences in interpretation, which could occur with different observers.

General tissue preparation and histopathological assessment

At 4 and 8 weeks post-MMT surgery or sham control surgery, the animals were euthanized with halothane anesthesia. The entire knee joints were then dissected for histology, microscopic analyses, and micro-computed tomography (µCT) imaging evaluation. Bilateral lumbar dorsal root ganglia (DRG) and spinal cords were harvested under a light microscope. Each dissected knee was fixed in 4% paraformalin and decalcified in EDTA, which was changed every 5 days. The decalcified knee was cut in the mid-sagittal plane and paraffin embedded. Serial knee sections of exact 5 µm thickness from the middle part of the knee were prepared for Safranin-O/Fast Green staining. Cartilage degradation was quantified using the modified Mankin grading system.[23] An unblinded investigator grouped the slides and randomly numbered them; these groups were then graded by two different blinded investigators (HJI, XL). A relative grade was assigned from 0 to 4, where 0 signifies no staining (PG loss) and 4 signifies the most intense stain (normal cartilage) based on Safranin-O. Two independent examinations were performed, and the repeatability of grading on the two occasions was determined using Cohen kappa statistics.

µCT and X-ray imaging analyses

Structural alterations of articular cartilage surface and subchondral bone architecture were evaluated by X-ray and µCT scan using standard procedures described previously.[21] Briefly, freshly dissected knee joints were immediately fixed in 10% formalin followed by µCT imaging analyses in the Rush Imaging Core Facility, with a Scanco Model 40 Desktop µCT (Scanco Medical, Bruttisellen, Switzerland). Scanning was conducted within a 10-mm region of the intact rat knees at 10-µm resolution (20-mm tube, high resolution at 55 kVP, 145 µA, 300 ms integration time). The X-ray beam was oriented perpendicular to the long axis of the joints.

MicroRNA expression profiling

Expression profiling of miRNA arrays were carried out using the instructions provided by the manufacturer. Briefly, after having passed sample quality control using a Bioanalyzer 2100 (Agilent Technologies Inc., Santa Clara, CA, USA) and Nanodrop RNA quantification, samples were labeled using the miRCURY Hy3/Hy5 Power labeling kit and hybridized on the miRCURY LNA Array (v.11.0 hsa, mmu and rno). The quantified signals (background corrected) were normalized using the global LOWESS (locally weighted scatterplot smoothing) regression algorithm, which we have found produces the best within-slide normalization to minimize intensity-dependent differences between the dyes. The beneficial effects of this normalization are illustrated in the MA plots for each slide. MicroRNA sequences and target predictions were retrieved from the Sanger Institute mirBase registry and mirBase target database release 10.1 (http://microrna. sanger.ac.uk/index.shtml).

Real-time qPCR analysis of microRNA expression

For each DRG (L4/L5) and spinal dorsal horn, RNA fractions were extracted from tissue samples using Trizol reagent (Invitrogen, Carlsbad, CA, USA) following the instructions provided by the manufacturer. Expression of miR-146a (Assay ID 000468) and miR-183 (Assay ID 002269) was examined with the TaqMan MicroRNA Assay Kit. U6 (Assay ID 001973) was used as an internal control. RT was carried out with 1 µg total RNA using the ThermoScript RT-PCR system (Invitrogen) for first-strand complementary DNA (cDNA) synthesis. For real-time qRT-PCR, cDNA was amplified using the MyiQ Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). Cycle threshold (Ct) values were obtained from each amplification curve using iQ5 Optical System Software provided by the manufacturer (Bio-Rad). Relative messenger RNA (mRNA) expression was determined using the ΔΔCT method.[24] 18S RNA was used as the internal control. Primer sequences and details for their use will be provided upon request.

Immunoblotting

Total protein concentrations of cell lysates were determined by the bicinchoninic acid (BCA) protein assay (Pierce, Rockford, IL, USA). In each case, an equal amount of protein was resolved by 10% SDS-polyacrylamide gels and transferred to nitrocellulose membranes for immunoblot analyses as described previously.[25] Immunoreactivity of NF-κB (Cell Signaling, Billerica, MA, USA), TRAF6 (Thermo Scientific, Rockford, IL, USA), TLR2 (Thermo Scientific), and TLR4 (Novus Biological, Littleton, CO, USA) was visualized using the ECL system (Amersham Biosciences, Piscataway, NJ, USA) and the Signal Visual Enhancer system (Pierce) to magnify signal intensity.

Gain- and loss-of-function analyses of microRNA in glial cells

Astrocytes were prepared from postnatal (D7) mice and isolated from mixed glial cultures as described earlier[26, 27] in accordance with the Giulian and Baker[28] procedure. Briefly, on day 9, the mixed glial cultures were washed three times with Dulbecco's modified Eagle's medium/F-12 (DMEM/F-12) and subjected to shaking at 240 rpm for 2 hours at 37°C on a rotary shaker to remove microglia. After 2 days, shaking was repeated for 24 hours to remove oligodendroglia. The attached cells were trypsinized and seeded in 12-well plates for 24 hours before experimentation. BV2 microglial cells[23] were likewise seeded in 12-well plates before experimentation. Analyses for gain- or loss-of-function of microRNAs were conducted using the Lipofectamine 2000 system (Invitrogen) following the manufacturer's instructions. Briefly, commercially validated miRNA mimics or inhibitors (Dharmacon, Lafayette, CO, USA) were diluted in OPTI-MEM I medium (Invitrogen) to achieve a final concentration of 20 nM and supplemented with an equal volume of medium with Lipofectamine 2000. After incubation for 20 minutes, the oligomer-lipofectamine 2000 complexes were added to the cells. After transfection, cells were stimulated with 1 mg/mL of lipopolysaccharides (LPS, Sigma, St. Louis, MO, USA) or 10 ng/mL of IL-1β (PeproTech, Rocky Hill, NJ, USA). PBS was used as a negative control. Total RNA and protein were prepared to assess changes in the expression of inflammatory or pain-related mediators.

Statistical analysis

Statistical significance was determined by Student's t test or one-way repeated measures ANOVA followed by Sidak post hoc testing using the SPSS 17 program. The evaluation of real-time PCR data was done by one-way ANOVA with Tukey's post hoc test using 2-ΔΔct values of each sample.[24] In each case, p values less than 0.05 were considered significant.

Results

  1. Top of page
  2. ABSTRACT
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgments
  9. References

Chronic OA pain is correlated with histopathological and structural changes in knee joints

Using an established surgical model for OA involving transection of the medial meniscus in the rat knee joint, we examined the development of joint pathology in relation to the experience of pain. In the ipsilateral knee joint, significantly increased sensitivity was observed around 4 weeks after OA induction (early pain stage) as assessed by mechanical allodynia (von Frey filament test), and the hyperalgesia was sustained throughout the entire experimental period (>8 weeks), suggesting the development of chronic joint pain by an experimental OA induction (Fig. 1A; p < 0.001, F = 341, n = 8 per group).

image

Figure 1. (A) The OA knee joint pain model was generated by medial meniscus transection (MMT), followed by behavioral pain tests that assess mechanical allodynia (n = 8, *p < 0.05). (B) Rats were euthanized at 2, 4, and 8 weeks after surgery, and the knee joints were subjected to Safranin O/Fast Green staining followed by blinded grading (upper right panel). Thinning cartilage area is marked by the solid triangles. (C) Structural changes in the articular cartilage surface and joint space narrowing were examined by µCT and radiographic imaging analyses (X-ray), respectively (sham control [CTL] versus 8 weeks after surgery).

Download figure to PowerPoint

To evaluate the relationship between pain symptoms and joint pathology, alterations in structural components of the knee joint tissues were examined by histopathological assessments. Rats that received MMT or sham surgery were euthanized at three specified time points representing pre-onset of OA (2 weeks), early OA (4 weeks), and chronic OA (8 weeks) pain stages. Relative histopathological grading by two blinded independent investigators demonstrate that significant joint pathology was observed at 4 and 8 weeks after OA induction (p < 0.05 and p < 0.01, respectively). Representative results are shown in Fig. 1B (n = 8 for each group).

Histopathological changes in articular cartilage were corroborated by imaging analyses. Three-dimensional µCT scans and radiographic analyses (X-ray) of knee joints 8 weeks after OA induction revealed structural alterations with undulating (ie, heaved and sunken) articular cartilage surfaces (Fig. 1C, upper right panel) and narrowing of the knee joint space (Fig. 1C, lower right panel) compared with the sham control group that maintained integrity of the knee joint (upper and lower left panel, respectively). Each rat knee shown is representative for a sample of n = 8. Collectively, these results suggest that OA-induced joint pain is closely correlated with structural changes in knee joint components.

Expression of miR-146a and the miR-183 cluster is significantly decreased in the DRG and spinal cord of animals with severe OA pain

Comparative miRNA profiling using the miRCURY LNA Array identified nine distinct miRNAs that were significantly (>twofold changes, p < 0.03) altered in the spinal dorsal horn from animals with severe chronic OA pain (8 weeks after OA induction) compared with the sham control group. Two miRNA groups were upregulated (ie, miR-140, -155) and seven were downregulated (miR-146a, miR-183 cluster [miR-183, -96 and -182], as well as miR-125b, let-7e, and -27b) (Fig. 2A). Altered expression patterns of miRNAs were comparable in both animal models of OA, MMT, and intra-articular injection of MIA (Fig. 2A, lanes 3 versus 4).

image

Figure 2. (A) Rat spinal cords were harvested and microRNA profiles were determined using an array-based method. The color scale illustrates the relative expression level of a microRNA across all samples: red represents an expression value lower than the mean, and blue represents values higher than the mean. (B) Expression of miR-146a and miR-183 was determined in both spinal cord dorsal horn and dorsal root ganglion (DRG) 2, 4, and 8 weeks after surgery using real-time quantitative PCR.

Download figure to PowerPoint

Based on the miRNA array data and our previous report,[12] we selected two different classes of miRNAs for further analyses: miR-146a and the miR-183 family (ie, miR-96, -182, and -183). These miRNAs were chosen because of their specific roles in cartilage homeostasis, inflammatory pain pathways, and chronic pain.[29, 30] Quantitative RT-PCR analyses of individual miRNAs revealed that these two classes were indeed significantly downregulated in both DRG sensory neurons (Fig. 2B, left panel, p < 0.01) and the spinal dorsal horn (right panel, p < 0.01) from animals with joint pain (4 and 8 weeks after OA induction by MMT) compared with sham control. There was no significant change in miRNAs 2 weeks after OA induction, a prodromal OA stage.

OA-evoked joint pain is linked to increases in multiple targets of miR-146 and the miR-183 cluster in the spinal dorsal horn and DRG

Because miR-146a is markedly reduced in DRG sensory neurons and the spinal cord of animals with OA pain (Fig. 2), we further investigated whether the target molecules of miR-146a are reciprocally upregulated. For example, miR-146a is a major regulator of TLR- and NF-κB-dependent targets, including multiple pro-inflammatory cytokines in various tissues, such as human articular chondrocytes.[12] Multiple inflammatory cytokines and signaling components were strikingly upregulated compared with the sham control at 4 weeks after induction of OA, and these highly induced levels were sustained throughout the entire experimental period (8 weeks after OA induction) (Fig. 3).

image

Figure 3. Expression of inflammatory mediators was determined by real-time qPCR using both spinal cord and DRG at 2, 4, and 8 weeks after OA induction. Statistical significance is based on p < 0.05 compared with control (single asterisk) or p < 0.01 compared with control (double asterisk).

Download figure to PowerPoint

Unlike miR-146a, which is abundantly expressed in various tissues and cell types, miR-183 exhibits a distinct neural tissue specific expression pattern with no or negligent expression levels in articular cartilage (data not shown). It has been reported that the miR-183 cluster family (miR-183, -96, and -182) plays a key role in chronic neuropathy and inflammatory pain by regulating pain-related ion channel genes (ie, Nav1.3 or Nav1.7), which are present at abnormally high levels in injured sensory neurons.[19, 31] Because miR-183 was significantly downregulated in DRG sensory neurons and in the spinal cord of animals with OA pain, we examined potential alterations of miR-183 cluster-regulated genes such as ion channels and pain-related neurotrophic factors that modulate function of ion channels.[32] At 4 weeks after MMT surgery, expression of ion channel proteins (eg, Nav1.3, Nav1.7, and TRPV1), as well as neurotrophic factors and peptide neurotransmitters (eg, brain-derived neurotrophic factor [BDNF], nerve growth factor [NGF], and neuropeptide Y [NPY]) are significantly increased in DRG sensory neurons and spinal cords; furthermore, these increased levels are maintained throughout the entire experimental period (8 weeks) (Fig. 4). We noted that upregulation of TRPV1 in both DRG sensory neurons and spinal cords of animals experiencing OA joint pain is quite remarkable as evidenced by a >100-fold induction at the chronic stage of OA pain (Fig. 4; p < 0.001; 8 weeks after OA induction).

image

Figure 4. Expression of pain-related factors determined by real-time qPCR using both spinal dorsal horn and DRG harvested 2, 4, and 8 weeks after surgical induction of OA. Statistical significance is based on p < 0.05 (single asterisk), p < 0.01 (double asterisk), or p < 0.001 (triple asterisk) compared with sham control group.

Download figure to PowerPoint

Gain- or loss-of-function analyses reveal that miR-146a and the miR-183 cluster regulate factors associated with inflammation and hyperalgesia in glial cells

We observed close correlation between decreases in miR-146a and the miR-183 cluster and increases in their respective targets including TLR/NF-κB pathway-dependent pro-inflammatory cytokines (targets of miR-146a) and pain-related ion channel genes (targets of miR-183 cluster) in the central compartments. Thus, we further determined the roles of these key miRNAs in glia by transfection of miRNA mimics or inhibitors in the presence and absence of LPS or IL-1β widely used glial activators.[33] Based on our in vivo animal studies (Figs. 3 and 4), we selected the most prominent factors that were markedly induced in the spinal cord and/or DRG sensory neurons in animals experiencing OA pain. These targets include TNFα, IL-6, IL-1β, TRAF6, NF-κB, TLR-2, TLR-4, NGF, NPY, and the ion channels (TRPV1, Nav1.3, Nav1.7, and Nav1.8). miR-146a mimics significantly suppressed LPS or IL-1β-induced TNFα, IL-6, IL-1β (data not shown), TRAF6, NF-κB, TLR-2, TLR-4, and NGF in both mRNA (Fig. 5A; p < 0.05 or p < 0.01) and/or protein levels in microglial cells (Fig. 5B, C; p < 0.01).

image

Figure 5. BV2 cells were transfected with miR-146a mimic or inhibitor followed by stimulation of cells with LPS (1 µg/mL) or IL-β (10 ng/mL) for 24 hours. Altered gene expression and/or protein levels were analyzed by qPCR (A), immunoblotting (B), or ELISA (C). Statistical significance at different levels of probability is indicated by asterisks (*p < 0.05; **p < 0.01).

Download figure to PowerPoint

Although miR-146a gain-of-function significantly suppresses the TLR-/NF-κB-dependent pro-inflammatory cytokines expression, miR-146a had no effect on ion channel genes (Nav1.3, Nav1.7, and TRPV1) or the neurotransmitter, NPY, in microglia (data not shown). These ion channels and neurotransmitter genes are effectively regulated by exogenous supplementation of miR-183 cluster (ie, miR-183, -96, or -182) in which overexpression of the miR-183 cluster significantly mitigated LPS- or IL-1β stimulation of these targets (Fig. 6; p < 0.01). Inhibitors of either miR-146a or the miR-183 cluster showed no statistically significant changes in their targets.

image

Figure 6. BV2 cells were transfected with miR-183 family (miR-183, miR-182, and miR-96) mimic or inhibitor followed by stimulation of cells with LPS (1 µg/mL) or IL-β10 ng/mL for 24 hours. Altered gene expression were analyzed by qPCR analyses. Statistical significance at different levels of probability is indicated by asterisks (*p < 0.05; **p < 0.01).

Download figure to PowerPoint

Next, we tested the function of these key miRNAs in astrocytes, a major glial component of the CNS. Primary astrocytes were isolated and transfected with miRNA mimics or inhibitors. Similar results were observed using astrocytes as were seen in microglia in the presence of miR-146a mimics and/or inhibitors. For example, miR-146a overexpression significantly reduced expression level of IL-1β-induced TRAF6, NF-κB, TLR-2, and NGF (Fig. 7; p < 0.01). Interestingly, ion channel genes (ie, Nav1.3, TRPV1), which are regulated strictly by miR-183 cluster but not by miR-146a in microglial cells, were significantly suppressed by miR-146a mimics in astrocytes (p < 0.01), suggesting the possibility that miR-146a may exert broader regulatory functions in astrocytes compared with microglial cells.

image

Figure 7. Mouse primary astrocytes were transfected with miR-146a mimic or inhibitor followed by stimulation of cells with IL-β (10 ng/mL) for 24 hours. Altered gene expression and/or protein levels were analyzed by qPCR. Statistical significance at different levels of probability is indicated by asterisks (**p < 0.01).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. ABSTRACT
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgments
  9. References

Our findings suggest that OA-induced knee joint hypersensitivity may be regulated by pathologically altered expression patterns of miR-146a and -183 family members in the nociceptive pathway. We observed significant downregulation of miR-146a and members of the miR-183 cluster (-183, -96, and -182) in the DRG and spinal cord from the animals with OA-induced knee joint pain. The primary role of microRNAs is to inhibit expression of target genes by post-transcriptional suppression or destabilization of target mRNAs. In the current in vivo studies, we found that rats experiencing OA pain have highly increased levels of inflammatory cytokines, neuropeptides, and pain-related ion channel genes in the sensory neurons in the DRG and spinal dorsal horn. In vitro gain- or loss-of-functional analyses in glial cells demonstrate that miR-146a controls TLR/NF-κB-dependent target inflammatory cytokines and TLRs, whereas members of the miR-183 cluster potently modulates pain-related ion channel genes and peptide neurotransmitter in glial cells. The combined results from both in vitro and in vivo studies provide compelling evidence for the concept that downregulated key anti-inflammatory and anti-nociceptive miRNAs at the spinal level may be responsible for upregulation of pain-related genes, which may prime OA-related chronic pain and associated pathological condition.

Our current studies, which assessed the direct role of miRNAs in joint inflammation and disease, complement several other studies that address the contribution of miRNAs to differentiation and function of hematopoietic cells involved in inflammation or the regulation of dendrite cell functions in inflammation.[3] Our previous[12] and current data collectively suggest that inflammation (both peripheral joint tissue inflammation and spinal inflammation) is a critical event for the initiation and progression of OA-evoked chronic pain. Thus, miRNA-dependent gene regulation of the nociceptive pathways associated with peripheral and central inflammation could be a major target in the control of chronic OA pain. Our miRNA array studies using the spinal dorsal horns identified a significantly altered set of miRNAs in animals experiencing chronic OA pain. Interestingly, many of these dysregulated miRNAs have been reported to play a role in articular cartilage homeostasis and share common regulatory functions in controlling NF-κB-dependent cytokine production, inflammation, and innate immunity.[5, 34-39] For example, miR-146a has been the central focus of many studies because of its enhanced expression after TLR activation[5] and its function as a negative regulator in immune and inflammatory responses.[5, 6, 40, 41] A polymorphism in the 3′ UTR of the mRNA encoding the miR-146a target TRAF6 is associated with a susceptibility to rheumatoid arthritis (RA)[42] and psoriatic arthritis.[43] Previously, we reported that miR-146a gain-of-function significantly reduces NF-κB-dependent targets including TRAF6 and multiple pro-inflammatory pain mediators in peripheral human joint tissues (ie, primary articular chondrocyte and fibroblast-like synovial cells).[12] Our current studies revealed that miR-146a is highly expressed in sensory neurons in the DRG and spinal cord in normal physiological condition, and this expression level is significantly reduced in pathological chronic pain condition. One of our exciting findings is a dramatic induction (∼100-fold) of TRPV1 in the DRG and spinal cord in the chronic OA pain group. Previously, it was reported that TRPV1 receptor-knockout mice show less inflammatory or neuropathic pain-induced glial activity compared with wild-type mice.[44] Thus, TRPV1 receptor could be involved in activating spinal glia in different pain pathways and may operate through distinct mechanisms at different stages during the progression of the pain state.[44]

Mechanoeception and nociception could each influence observed changes in the DRG and spinal cord. For example, patients with OA and, in particular, those with more advanced OA, complain of severe pain even when resting, suggesting that the nociceptive input may be an overriding factor that leads to the changes in the DRG and the spinal cord. Unlike miR-146a, we were unable to detect expression of the miR-183 family, an evolutionarily conserved sensory organ-specific miRNA cluster that encodes miR-183, -96, and -182, in peripheral joint tissues regardless of the severity of the degenerative condition of the joint. The miR-183 cluster is abundantly expressed in the DRG and spinal cord, suggesting tissue-specific co-expression and potential unique functions of nociceptive and mechanosensitive primary afferent neurons. The members of the miR-183 cluster share functional roles in similar target molecules[19, 45] and are known to be associated with neuronal plasticity by influencing pain pathways in which miR-183 cluster is dramatically reduced in neuropathy- and/or inflammatory chronic pain.[19, 29] Predicted targets of the miR-183 family are known to be involved in neuropathy, suggesting that the miR-183 family regulates nociceptive sensitivity by regulating pain-related ion channel genes such as Nav1.3, Nav1.7; these mRNAS are present at abnormally high levels in injured sensory neurons.[46] Like miR-146a, we demonstrated that expression of the miR-183 cluster is markedly reduced in sensory neurons in the DRG and spinal cords from the animals with OA-induced chronic joint pain.

Glial cells are the major cellular component of the CNS and markedly outnumber neurons by at least by a factor of 10.[47] It has been extensively documented that, at the level of spinal cord, glia play a key role in the development and maintenance of persistent chronic pain.[13, 14, 48] For example, glia have been shown to dynamically modulate the function of the sensory system under pathological conditions, including rheumatoid arthritis.[18] Specific features of activated glial responses involve the release of cytokines, chemokines, neurotrophins, and proteases that alter neuronal activity and contribute to nociceptive processing.[49] Upon activation, glial cells release inflammatory cytokines (ie, TNFα, IL-1) and pain molecules leading to hypersensitivity.[49, 50] For these reasons, the pharmacological attenuation of glial activity has been a novel approach for controlling neuropathic pain. We show that introduction of miR-146a mimics significantly attenuates LPS- or IL-1-mediated inductions of the microglial marker CD11b (also known as OX-42).

Both in vitro and in vivo studies demonstrate that decreases in the expression of miR-146a and the miR-183 cluster are reciprocally related to increases in multiple inflammatory factors, neurotransmitters, neuropeptides and pain-related ion channels, which are regulated by these key miRNA signatures. Our functional analyses using individual cellular components of the spinal cord (ie, astrocytes, microglia and neurons) reveal that miRNAs may play a role in a cell-type specific manner, and the results may depend on the cellular environment. For example, in microglial cells, miR-146a controls TRAF6, NF-κB, IL-1/TLR-associated downstream signaling pathway targets and NGF, but fails to regulate ion channel genes, whereas miR-183 cluster specifically regulates TRPV1, NPY and sodium channel genes. In astrocytes, however, miR-146a appears to exert broader cellular regulatory functions: miR-146a has the capacity to control not only its direct targets (TLR/NF-κB-dependent genes), but also ion channels (ie, TRPV1), providing mechanistic basis that targeting miR-146a in astrocytes would be more efficient to control the nociceptive pathway. In neuronal cells (PC-12 cells), miR-146a gain-of-function fails to reduce NF-κB-dependent genes, including IL-1β, TNFα, and IL-6 and shows no controlling capacity on pain-related ion channel genes (data not shown).

We have previously presented in vivo evidence that OA pain is caused by central sensitization through communication between peripheral OA nociceptors and the central sensory system, suggesting a mechanistic overlap between OA-induced pain and neuropathic pain.[21] In this study, we observed similar expression patterns of the spinal miRNAs in both knee joint OA-induced pain and neuropathy along with glial activation as determined by the specific microglial marker CD11b (data not shown). Because miRNAs regulate global gene expression, our new findings corroborate our previous results, suggesting that (i) knee joint OA-induced chronic pain is centralized and involves sensory neuronal plasticity, and (ii) OA-evoked nociceptive pain pathways overlap, in part, with neuropathic pain pathways.

One limitation of our study is that we have not yet resolved which spinal factor(s) are specifically associated with OA-induced nociceptive pathways and thus play a key role in the pain perception. For example, OA pain significantly stimulates calcium channel subunit α2δ1 in central compartments; however, pregabalin (otherwise known as gabapentin), a pharmacological inhibitor of α2δ1 effective for neuropathy, has no effect on reducing OA pain (data not shown). These results suggest that although there are overlaps between neuropathy and OA-induced nociceptive pathways, OA-evoked pain may be distinct from other pain pathways (ie, neuropathy). Thus, a better appreciation for OA-specific nociceptive processes and the identification of key proteins that are specifically associated with OA-evoked pain will greatly facilitate the development of effective treatments for OA pain.

In summary, our studies demonstrate that downregulation of miR-146a and miR-183 cluster is causally related to increased inflammatory cytokines, ion channels, and pain-related neurotransmitters at the spinal level by controlling these targets in glial cells. With their dual function for alleviation of joint pain and concomitant regeneration of peripheral knee joint cartilage, miR-146a and miR-183 cluster could serve as powerful therapeutic targets for OA intervention.

Acknowledgments

  1. Top of page
  2. ABSTRACT
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgments
  9. References

This work was supported by NIH NIAMS R01 grants AR053220 and AR062136 (to HJI) as well as R01 AR055915 (to DC) and R01AR049069 (to AJvW). We also thank the Arthritis National Research Foundation (to XL).

Authors' roles: Study design: XL, DC, and HJI. Study conduct: XL, JSK, RK, GTC, KP, SF, J-SK, GG, JS, SGK, and HJI. Data analysis: XL, JSK, DC, and HJI. Data interpretation: XL, JSK, RK, GG, JSK, KP, and HJI. Drafting manuscript: XL, JSK, RK, DC, AJW, and HJI. Revising manuscript: XL, RK, GG, SF, GTC, KP, AJW, and HJI. Approval of final version of the manuscript: XL, JSK, RK, GG, DC, GTC, KP, SF, J-SK, AJW, JS, SGK, and HJI.

References

  1. Top of page
  2. ABSTRACT
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Disclosures
  8. Acknowledgments
  9. References
  • 1
    Sprangers MA, de Regt EB, Andries F, van Agt HM, Bijl RV, de Boer JB, Foets M, Hoeymans N, Jacobs AE, Kempen GI, Miedema HS, Tijhuis MA, de Haes HC. Which chronic conditions are associated with better or poorer quality of life?. J Clin Epidemiol. 2000; 53(9):895907.
  • 2
    Ferreira-Gomes J, Adaes S, Castro-Lopes JM. Assessment of movement-evoked pain in osteoarthritis by the knee-bend and CatWalk tests: a clinically relevant study. J Pain. 2008; 9(10):94554.
  • 3
    Busch M, Zernecke A. microRNAs in the regulation of dendritic cell functions in inflammation and atherosclerosis. J Mol Med (Berl). 2012; 90(8):87785.
  • 4
    Bartel DP. MicroRNAs: target recognition and regulatory functions. Cell. 2009; 136(2):21533.
  • 5
    Taganov KD, Boldin MP, Chang KJ, Baltimore D. NF-kappaB-dependent induction of microRNA miR-146, an inhibitor targeted to signaling proteins of innate immune responses. Proc Natl Acad Sci USA. 2006; 103(33):124816.
  • 6
    Taganov KD, Boldin MP, Baltimore D. MicroRNAs and immunity: tiny players in a big field. Immunity. 2007; 26(2):1337.
  • 7
    Tsai CY, Allie SR, Zhang W, Usherwood EJ. MicroRNA miR-155 affects antiviral effector and effector Memory CD8 T cell differentiation. J Virol. 2013; 87(4):234851.
  • 8
    Tili E, Michaille JJ, Cimino A, Costinean S, Dumitru CD, Adair B, Fabbri M, Alder H, Liu CG, Calin GA, Croce CM. Modulation of miR-155 and miR-125b levels following lipopolysaccharide/TNF-alpha stimulation and their possible roles in regulating the response to endotoxin shock. J Immunol. 2007; 179(8):50829.
  • 9
    Murphy AJ, Guyre PM, Pioli PA. Estradiol suppresses NF-kappa B activation through coordinated regulation of let-7a and miR-125b in primary human macrophages. J Immunol. 184(9):502937.
  • 10
    Williams AE, Perry MM, Moschos SA, Larner-Svensson HM, Lindsay MA. Role of miRNA-146a in the regulation of the innate immune response and cancer. Biochem Soc Trans. 2008; 36(Pt 6):12115.
  • 11
    Chiyomaru T, Enokida H, Tatarano S, Kawahara K, Uchida Y, Nishiyama K, Fujimura L, Kikkawa N, Seki N, Nakagawa M. miR-145 and miR-133a function as tumour suppressors and directly regulate FSCN1 expression in bladder cancer. Br J Cancer. 2010; 102(5):88391.
  • 12
    Li X, Gibson G, Kim JS, Kroin J, Xu S, van Wijnen AJ, Im HJ. MicroRNA-146a is linked to pain-related pathophysiology of osteoarthritis. Gene. 2011; 480(1–2):3441.
  • 13
    Gwak YS, Hulsebosch CE. Remote astrocytic and microglial activation modulates neuronal hyperexcitability and below-level neuropathic pain after spinal injury in rat. Neuroscience. 2009; 161(3):89503.
  • 14
    Griffin RS, Costigan M, Brenner GJ, Ma CH, Scholz J, Moss A, Allchorne AJ, Stahl GL, Woolf CJ. Complement induction in spinal cord microglia results in anaphylatoxin C5a-mediated pain hypersensitivity. J Neurosci. 2007; 27(32):8699708.
  • 15
    Mika J, Osikowicz M, Rojewska E, Korostynski M, Wawrzczak-Bargiela A, Przewlocki R, Przewlocka B. Differential activation of spinal microglial and astroglial cells in a mouse model of peripheral neuropathic pain. Eur J Pharmacol. 2009; 623(1–3):6572.
  • 16
    Ji RR, Suter MR. p38 MAPK, microglial signaling, and neuropathic pain. Mol Pain. 2007; 3:33.
  • 17
    Temburni MK, Jacob MH. New functions for glia in the brain. Proc Natl Acad Sci USA. 2001; 98(7):363132.
  • 18
    Constandil L, Hernandez A, Pelissier T, Arriagada O, Espinoza K, Burgos H, Laurido C. Effect of interleukin-1beta on spinal cord nociceptive transmission of normal and monoarthritic rats after disruption of glial function. Arthritis Res Ther. 2009; 11(4):R105.
  • 19
    Aldrich BT, Frakes EP, Kasuya J, Hammond DL, Kitamoto T. Changes in expression of sensory organ-specific microRNAs in rat dorsal root ganglia in association with mechanical hypersensitivity induced by spinal nerve ligation. Neuroscience. 2009; 164(2):71123.
  • 20
    Glasson SS, Blanchet TJ, Morris EA. The surgical destabilization of the medial meniscus (DMM) model of osteoarthritis in the 129/SvEv mouse. Osteoarthritis Cartilage. 2007; 15(9):10619.
  • 21
    Im HJ, Kim JS, Li X, Kotwal N, Sumner DR, van Wijnen AJ, Davis FJ, Yan D, Levine B, Henry JL, Desevre J, Kroin JS. Alteration of sensory neurons and spinal response to an experimental osteoarthritis pain model. Arthritis Rheum. 2010; 62(10):29953005.
  • 22
    Chaplan SR, Bach FW, Pogrel JW, Chung JM, Yaksh TL. Quantitative assessment of tactile allodynia in the rat paw. J Neurosci Methods. 1994; 53(1):5563.
  • 23
    van der Sluijs JA, Geesink RG, van der Linden AJ, Bulstra SK, Kuyer R, Drukker J. The reliability of the Mankin score for osteoarthritis. J Orthop Res. 1992; 10(1):5861.
  • 24
    Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods. 2001; 25(4):4028.
  • 25
    Im HJ, Muddasani P, Natarajan V, Schmid TM, Block JA, Davis F, van Wijnen AJ, Loeser RF. Basic fibroblast growth factor stimulates matrix metalloproteinase-13 via the molecular cross-talk between the mitogen-activated protein kinases and protein kinase Cdelta pathways in human adult articular chondrocytes. J Biol Chem. 2007; 282(15):1111021.
  • 26
    Brahmachari S, Pahan K. Sodium benzoate, a food additive and a metabolite of cinnamon, modifies T cells at multiple steps and inhibits adoptive transfer of experimental allergic encephalomyelitis. J Immunol. 2007; 179(1):27583.
  • 27
    Corbett GT, Roy A, Pahan K. Sodium phenylbutyrate enhances astrocytic neurotrophin synthesis via protein kinase C (PKC)-mediated activation of cAMP-response element-binding protein (CREB): implications for Alzheimer disease therapy. J Biol Chem. 2013; 288(12):8299312.
  • 28
    Giulian D, Baker TJ. Characterization of ameboid microglia isolated from developing mammalian brain. J Neurosci. 1986; 6(8):216378.
  • 29
    Bai G, Ambalavanar R, Wei D, Dessem D. Downregulation of selective microRNAs in trigeminal ganglion neurons following inflammatory muscle pain. Mol Pain. 2007; 3:15.
  • 30
    Yamasaki K, Nakasa T, Miyaki S, Ishikawa M, Deie M, Adachi N, Yasunaga Y, Asahara H, Ochi M. Expression of MicroRNA-146a in osteoarthritis cartilage. Arthritis Rheum. 2009; 60(4):103541.
  • 31
    Obata K, Noguchi K. BDNF in sensory neurons and chronic pain. Neurosci Res. 2006; 55(1):110.
  • 32
    Tucker K, Fadool DA. Neurotrophin modulation of voltage-gated potassium channels in rat through TrkB receptors is time and sensory experience dependent. J Physiol. 2002; 542(Pt 2):41329.
  • 33
    Chen Z, Jalabi W, Shpargel KB, Farabaugh KT, Dutta R, Yin X, Kidd GJ, Bergmann CC, Stohlman SA, Trapp BD. Lipopolysaccharide-induced microglial activation and neuroprotection against experimental brain injury is independent of hematogenous TLR4. J Neurosci. 2012; 32(34):1170615.
  • 34
    Thulasingam S, Massilamany C, Gangaplara A, Dai H, Yarbaeva S, Subramaniam S, Riethoven JJ, Eudy J, Lou M, Reddy J. miR-27b*, an oxidative stress-responsive microRNA modulates nuclear factor-kB pathway in RAW 264.7 cells. Mol Cell Biochem. 2011; 352(1–2):1818.
  • 35
    Miyaki S, Sato T, Inoue A, Otsuki S, Ito Y, Yokoyama S, Kato Y, Takemoto F, Nakasa T, Yamashita S, Takada S, Lotz MK, Ueno-Kudo H, Asahara H. MicroRNA-140 plays dual roles in both cartilage development and homeostasis. Genes Dev. 2012; 24(11):117385.
  • 36
    Bocchini V, Mazzolla R, Barluzzi R, Blasi E, Sick P, Kettenmann H. An immortalized cell line expresses properties of activated microglial cells. J Neurosci Res. 1992; 31(4):61621.
  • 37
    Bhaumik D, Scott GK, Schokrpur S, Patil CK, Campisi J, Benz CC. Expression of microRNA-146 suppresses NF-kappaB activity with reduction of metastatic potential in breast cancer cells. Oncogene. 2008; 27(42):56437.
  • 38
    Kappelmann M, Kuphal S, Meister G, Vardimon L, Bosserhoff AK. MicroRNA miR-125b controls melanoma progression by direct regulation of c-Jun protein expression. Oncogene. 2012 Jul 16 [Epub ahead of print].
  • 39
    Wang DJ, Legesse-Miller A, Johnson EL, Coller HA. Regulation of the let-7a-3 promoter by NF-kappaB. PLoS One. 2012; 7(2):e31240.
  • 40
    Sheedy FJ, O'Neill LA. Adding fuel to fire: microRNAs as a new class of mediators of inflammation. Ann Rheum Dis. 2008; 67(Suppl 3):iii505.
  • 41
    Jin ZB, Hirokawa G, Gui L, Takahashi R, Osakada F, Hiura Y, Takahashi M, Yasuhara O, Iwai N. Targeted deletion of miR-182, an abundant retinal microRNA. Mol Vis. 2009; 15:52333.
  • 42
    Chatzikyriakidou A, Voulgari PV, Georgiou I, Drosos AA. A polymorphism in the 3'-UTR of interleukin-1 receptor-associated kinase (IRAK1), a target gene of miR-146a, is associated with rheumatoid arthritis susceptibility. Joint Bone Spine. 2010; 77(5):4113.
  • 43
    Chatzikyriakidou A, Voulgari PV, Georgiou I, Drosos AA. The role of microRNA-146a (miR-146a) and its target IL-1R-associated kinase (IRAK1) in psoriatic arthritis susceptibility. Scand J Immunol. 2010; 71(5):3825.
  • 44
    Chen Y, Willcockson HH, Valtschanoff JG. Influence of the vanilloid receptor TRPV1 on the activation of spinal cord glia in mouse models of pain. Exp Neurol. 2009; 220(2):38390.
  • 45
    Xu S, Witmer PD, Lumayag S, Kovacs B, Valle D. MicroRNA (miRNA) transcriptome of mouse retina and identification of a sensory organ-specific miRNA cluster. J Biol Chem. 2007; 282(34):2505366.
  • 46
    Alloui A, Zimmermann K, Mamet J, Duprat F, Noel J, Chemin J, Guy N, Blondeau N, Voilley N, Rubat-Coudert C, Borsotto M, Romey G, Heurteaux C, Reeh P, Eschalier A, Lazdunski M. TREK-1, K+ channel involved in polymodal pain perception. EMBO.J. 2006; 25(11):236876.
  • 47
    Azevedo FA, Carvalho LR, Grinberg LT, Farfel JM, Ferretti RE, Leite RE, Jacob Filho W, Lent R, Herculano-Houzel S. Equal numbers of neuronal and nonneuronal cells make the human brain an isometrically scaled-up primate brain. J Comp Neurol. 2009; 513(5):53241.
  • 48
    Levin ME, Jin JG, Ji RR, Tong J, Pomonis JD, Lavery DJ, Miller SW, Chiang LW. Complement activation in the peripheral nervous system following the spinal nerve ligation model of neuropathic pain. Pain. 2008; 137(1):182201.
  • 49
    McMahon SB, Malcangio M. Current challenges in glia-pain biology. Neuron. 2009; 64(1):4654.
  • 50
    Gomez-Nicola D, Valle-Argos B, Suardiaz M, Taylor JS, Nieto-Sampedro M. Role of IL-15 in spinal cord and sciatic nerve after chronic constriction injury: regulation of macrophage and T-cell infiltration. J Neurochem. 2008; 107(6):174152.