Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Sep 5.
Published in final edited form as: Gene. 2019 Jun 6;712:143911. doi: 10.1016/j.gene.2019.06.001

Circulating MicroRNA-23b as a New Biomarker for Rheumatoid Arthritis

Xi Liu 1,4,†, Su Ni 2,†, Chenkai Li 2, Nanwei Xu 3, Wenyang Chen 2, Min Wu 1, Andre J van Wijnen 5,*, Yuji Wang 3,5,*
PMCID: PMC6724744  NIHMSID: NIHMS1534561  PMID: 31176730

Abstract

MicroRNA-23b (miR-23b) is associated with inflammation and autoimmune diseases. This study evaluated miR-23b expression and assessed its potential as a biomarker of disease activity for rheumatoid arthritis (RA). Differential expression of microRNAs was determined by miRNA microarray analysis in fibroblast-like synoviocytes (FLSs) from four trauma patients as healthy controls (HCs) and eight RA patients. The microarray results showed elevated expression of miR-23b in FLSs from RA patients and this finding was corroborated by real-time quantitative polymerase chain reaction (RT-qPCR) and in situ hybridization using synovial tissues (STs). Furthermore, we found miR-23b levels in plasma of RA patients were significantly higher than in HCs, and plasma miR-23b levels positively correlated with the erythrocyte sedimentation rate (ESR), hypersensitive C-reactive protein (hs-CRP), C-reactive protein (CRP), DAS28, and platelet (PLT) count (P < 0.05). MiR-23b levels in plasma inversely correlated with the levels of hemoglobin (Hb), total bilirubin (TBIL), direct bilirubin (DBIL), indirect bilirubin (IBIL), total cholesterol (TC) and low-density lipoprotein cholesterol (LDL-C) (P < 0.05), but not with rheumatoid factor (RF) or anti-cyclic citrullinated peptide antibodies (ACPA) (P > 0.05). Moreover, patients with anorexia showed higher levels of miR-23b in plasma than those without anorexia. Similar results were observed with fatigue. Appropriate treatment for RA not only ameliorated the disease condition but also reversed the elevated plasma miR-23b level remarkably. These results suggest that circulating miR-23b may be a promising biomarker for RA disease activity.

Keywords: Circulating microRNA-23b, rheumatoid arthritis, biomarker

INTRODUCTION

Rheumatoid arthritis (RA) is an autoimmune disease (Smolen, Aletaha, & McInnes) which affects about 1% of the global population (Mcinnes & O’Dell, 2010). It chronically impairs the heart, skin, and many other organs. It is characterized by erosive alterations of the joint surfaces leading to destruction and deformity of the joints and eventual disability (Hellmich, 2014; Torpy, 2011). In addition to the joint-related symptoms, extra-articular symptoms and systemic lesions, such as anorexia, fatigue, morning stiffness, lymph nodes, and interstitial lung disease (ILD) can also be observed in patients with active RA (Benaglio et al., 2015; Kim, Kim, Moon, Lee, & Kim, 2015; Zhang, Li, Wu, Xin, & Yi, 2017). Nevertheless, the pathology of RA is not fully elucidated and diverse factors, such as genes, environment, gender, infection, and even dietary components are reported to get involved in the initiation and progression of RA (Vojdani, 2014). Recently, a growing body of evidence has pointed out that a cohort of microRNAs (miRNAs) existing in the peripheral blood gets elevated in the plasma or serum of RA patients and are involved in the pathogenesis of RA (Churov, Oleinik, & Knip, 2015).

miRNAs are endogenous small (approximately 22 nucleotides long) noncoding RNAs (Murata et al., 2010). They destabilize the target mRNAs to inhibit protein synthesis and regulate crucial pathways and cellular processes, such as cell growth, differentiation, proliferation, and cell death (Hruskova et al., 2016). An increasing number of miRNAs are being found to be involved in inflammatory and autoimmune responses, such as miR-346, which indirectly regulates the release of the pro-inflammatory cytokine interleukin (IL)-18 (Alsaleh et al., 2009). Wang, et al. have identified the circulating miRNAs that are specifically altered in patients with systemic lupus erythematosus (SLE) as compared to RA and healthy controls (HC). Eight miRNAs, namely, miR-126, miR-16, miR-451, miR-223, miR-21, miR-125a-3p, miR-146a, and miR-155 were given importance in subsequent clinical studies since these miRNAs were identified as important regulators of the immune cells responsible for the pathogenesis of SLE (Reid, Kirschner, & van Zandwijk).

MicroRNA-23b (miR-23b), a member of the miR-23b/27b/3074/24-1 cluster, harbored in chromosome 9, has been found to play a vital role in tumorigenesis (Au Yeung, Tsang, Yau, & Kwok, 2017; Cao et al., 2017; Fukumoto et al., 2016). An increasing number of reports have suggested that miR-23b has a close relationship with inflammation and autoimmune diseases (Zheng et al., 2012). Ayyadurai, et al. demonstrated that miR-23b was secreted and transported between the cells to impose a gene-silencing effect on the recipient intestinal macrophages, and regulated Marcksl-1 in the macrophages during intestinal inflammation (Ayyadurai et al., 2014). miR-23b has also been found to regulate the expression of inflammatory factors in vascular endothelial cells during sepsis (M. Wu, Gu, Yi, Tang, & Tao, 2015). In bone marrow mesenchymal stem cells, blocking the activation of NF-κB pathway through miR-23b overexpression resulted in the repression of maturation and differentiation of dendritic cells (DCs), which is the initiating factor in immune response (J. Wu et al., 2017). Moreover, Qi, et al. have demonstrated that a decreased level of miR-23b was involved in the activation of the Notch signaling pathway which contributes to the increase in the number of Th1/Th17 cells leading to Behcet’s disease (Jian et al., 2014). Most importantly, the level of miR-23b was reported to be decreased in both synovial tissues of RA patients and kidney tissues of SLE patients after suppressing IL-17-associated autoimmune inflammation through TAB2, TAB3, and IKK-α(Zhu et al., 2012).

Focusing on the pathology of RA, we applied microarray analysis with microRNAs obtained from RA patients and HC. Our data showed a substantial elevation of miR-23b in fibroblast-like synoviocytes (FLS) obtained from RA patients. Till date, there is no report on the level of circulating miR-23b in RA. The present study was aimed to evaluate the miR-23b levels in synovial tissues (STs) and plasma and to further explore the clinical potential of miR-23b as a biomarker for RA disease activity.

MATERIAL AND METHODS

Samples obtained from RA patients and HC

All RA patients included in this study were diagnosed with RA according to the 2010 American College of Rheumatology/European League Against Rheumatism (ACR/EULAR) criteria (Aletaha et al., 2010). The samples of synovial tissues from 34 RA patients were obtained by total knee joint replacement. The control knee tissue samples were taken from 36 trauma patients (as re. These trauma patients had healthy joints with intact synovium. All tissue samples were stored at −80 °C immediately upon collection until use for miRNA extraction.

Peripheral blood was collected from 109 RA patients and 48 HC having no history of autoiimnune disease, severe cardiovascular disease, hepatic disease, renal disease, and any other chronic disease. Before venipuncture, the skin was disinfected with alcohol, and the blood samples were collected in EDTA-coated tubes. The demographic characteristics of the RA patients and the HC are shown in Table 1.

Table 1.

Characteristics and plasma miR-23b levels of the subjects investigated

Characteristics RA HC RA(STs) HC (STs)
Number 109 48 34 36
Age (years)a 61(52-65) 56(39-66.5) 57(53-65) 54(44-59)
SEX(M/F) 20/89 13/35 16/18 17/19
MiR-23b levelsa 55.56(16.27,229.6) 37.78(9.57,96.59) 25.93(9.40,89.85) 6.39(1.67,27.44)
Open in a new tab
a

Expressed as the median (25th to 75th percentitle). RA, rheumatoid arthritis; HC, healthy controls; MiR-23b, microRNA-23b;HC, control trauma tissues; STs, synovial tissues.

In the groups providing plasma, there was no significant difference in age (P=0.3157, Mann whitney test) and sex (P=0.1295, Chi-square) between RA patients and HC. And in the groups providing STs, there was no significant difference in age (P=0.0901, Mann whitney test) and sex (P=0.9891, Chi-square) between RA patients and HC.

The levels of platelet (PLT), C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), high sensitivity C-reactive protein (hs-CRP), low density lipoprotein cholesterol (LDL-C), aspartate transaminase (AST), total bilirubin (TBIL), indirect bilirubin (IBIL), direct bilirubin (DBIL), antinuclear antibody (ANA), rheumatoid factor (RF), anti-cyclic citrullinated peptide antibody (ACPA), and other laboratory parameters were estimated in a laboratory. The patients were evaluated by Disease Activity Score 28 (DAS28) using ESR, number of tender and swollen joints, and a 100 mm visual analog scale (VAS) for calculating the pain score. The disease characteristics of the RA patients are shown in Table 2 and 3.

Table 2.

Laboratory indicator and DAS28 of RA patients investigateda

Characteristics n r P
WBC(×109/l)b 6.39(5.35, 7.37) 107 0.026 0.792
Hb(g/l)c 113.90±20.62 107 −0.338 0.0004**
PLT(×109/l) b 237.00(195.75, 288.00) 104 0.364 0.0001**
CRP(mg/l) b 19.00(7.70, 43.00) 93 0.247 0.0169*
ESR(mm/h) c 44.48:25.44 98 0.207 0.0407*
Hs-CRP(mg/l) b 9.16(3.00 22.45) 75 0.273 0.018*
ALT(u/l) b 15.00(11.00, 22.00) 105 −0.173 0.078
AST(u/l) b 18.00(14.00, 23.00) 105 −0.170 0.083
γ-GT(u/l) b 23.00(15.00, 31.00) 105 −0.047 0.635
ALP(u/l) b 78.00(65.00, 93.00) 105 0.045 0.649
LDH(u/l) c 200.65± 51.93 105 0.049 0.622
ALB(g/l) c 35.90±5.00 105 −0.026 0.790
TBIL(umol/l) b 5.20(4.00, 6.50) 105 −0.2899 0.0027**
DBIL(umol/l) b 1.80(1.50, 2.20) 105 −0.2681 0.0057**
IBIL(umol/l) b 3.20(2.60, 4.40) 105 −0.2809 0.0037**
Cr(umol/l) b 58.00(52.00, 67.00) 104 −0.1364 0.167
BUN(umol/l) b 4.36(3.60, 5.73) 104 −0.182 0.064
TC(mmol/l) c 4.41±1.13 66 −0.2444 0.048*
TG(mmol/l) b 1.28(0.89, 1.68) 66 −0.173 0.164
HDL-C(mmol/l) b 1.14(0.96, 1.35) 66 −0.154 0.217
LDL-C(mmol/l) c 2.17±0.72 66 −0.2981 0.0151*
TSH(uIU/ml) b 1.90(1.31, 3.02) 54 0.112 0.421
FT3(pmol/l) c 5.49±1.58 54 0.268 0.05
FT4(pmol/l) b 16.30(15.21, 18.34) 54 −0.134 0.334
IgG(g/l)c 14.22±4.38 70 0.038 0.754
IgA(g/l) b 3.05(2.44, 3.64) 70 −0.045 0.709
IgM(g/l) b 1.45(1.18, 1.69) 70 −0.195 0.105
C3(g/l) c 1.09±0.20 70 0.060 0.622
C4(g/l) c 0.21±0.06 70 −0.112 0.356
ASO(lu/ml) b 26.80(25.00, 62.80) 65 −0.063 0.619
DAS28 c 4.35±1.63 73 0.2597 0.0265*
Open in a new tab
a

DAS28, disease activity score28; RA, rheumatoid arthritis. WBC, wliite blood cells; Fib, hemoglobin; PLT, platelet; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; hs-CRP, high sensitivity C reactive protein; ALT, Alanine aminotransferase; AST, Aspartate transaminase; γ-GT, glutamyl transpeptidase; ALP, alkaline phosphatase; LDH, lactate dehydrogenase; ALB, albumin; TBIL, total bilirubin; DBIL, direct bilirubin; IBIL, indirect bilirubin; TB, total cholesterol; TG, triacylglycerol; HDL-C, high-density lipoprotein cholesterol; LDL-C, low-density lipoprotein cholesterol; TSH, thyroid stimulating hormone; FT3, free thyroid3; FT4, free thyroid 4; IgG, immunoglobulin G; IgA immunoglobulin A; IgM immunoglobulin M; C3, complement 3; C4, complement 4; ASO, anti-Streptolysin “O”.

b

Expressed as the median (25th to 75th percentitle)

c

expressed as the average±standard deviation

*

expressed as P< 0.05

**

expressed as P< 0.01.

Table 3.

Antibodies patterns of RA patients investigateda

Antibodies Positive rate(%) n r P
RF 84.06% 68 −0.128 0.299
ACCP 87.8% 40 −0.118 0.469
ANA 48% 50 - 0.0029**
Open in a new tab
a

RA, rheumatoid arthritis; RF, rheumatoid factor; ACPA, anti-cyclic cirullinated peptide antibodies; ANA, antinuclear antibody.

**

expressed as P< 0.01.

Collection of STs and blood samples was performed following the medical ethics regulation of Nanjing Medical University and Suzhou University. This study was approved by Nanjing Medical University and Suzhou University Review Board, and written consent was requested and obtained from all the patients and HC.

miRNA microarray with FLSs

FLSs were isolated and cultured as previously described (Ni et al., 2015). After the cells were grown to 80–90% confluence on passage 3, they were harvested by trypinizing and total RNAs were purified with NucleoZOL (MACHEREY-NAGEL GmbH & Co. KG, Germany) according to the manufacturer’s protocol. For microarray analysis, the samples are labeled using miRCURY™ Hy3™/Hy5™ Power Labeling Kit and hybridized using miRCURY™ LNA Array (v.11.0, Exiqon, Demnark). Axon GenePix 4000B microarray scanner was used for scanning. GenePix pro v 6.0 was used to read the raw intensity of the image. The ratio of red signal to green signal was calculated after background subtraction and normalization using the global Lowess (Locally Weighted Scatter Plot Smoothing) regression algorithm (MIDAS, TIGR Microarray Data Analysis System). Between slides normalization was performed by scale normalization to reduce the between-slide variability. The replicated spots on the same slides were averaged by obtaining a median ratio of the replicated spots. The threshold value used to screen the differentially expressed miRNAs was ≥1.5-fold change or ≤ 0.67-fold change (p ≤ 0.05).

Real time-PCR

Total RNA was extracted from STs and plasma using mirVana™ PARIS™ Kit (Thermo Fisher, USA) according to the manufacturer’s instructions. The total RNA was eluted from a Filter Cartridge using 60 μl nuclease-free water. On-column digestion of the contaminated DNA was performed using RNase-Free DNase Set (Qiagen, Germany). Total RNA was reverse-transcribed using TaqMan® miRNA Assays and TaqMan® miRNA Reverse Transcription Kit in an Eppendorf AG. Following that TaqMan® miRNA Assays and TaqMan® Universal PCR Master Mix II no UNG were used to quantify miRNA expression by quantitative polymerase chain reaction (qPCR) in a ViiA 7 Real-Time PCR System (all the TaqMan® kits, assays, and probes were purchased from Applied Biosystems, Thermo Fisher Scientific, USA). The primer sequences used for amplifying matured miR-23b and U6 control were designed by Applied Biosystems. The PCR cycles were performed following the manufacturer’s instructions as follows: 50 °C for 2 min and 95 °C for 10 min followed by 40 cycles at 95 °C for 15 s and 60 °C for 1 min. The threshold cycles (Ct) for the target miRNAs and the internal reference gene were used to represent their relative expression.

miRNA in situ hybridization

The ST samples were subjected to standard overnight fixation in 10% neutral-buffered formalin followed by paraffin embedding and were cut into 10 μm-thick slices. The slides were then deparaffinized in xylene and ethanol, incubated with Proteinase-K for 10 min at 37 °C, dehydrated with gradient ethanol, and hybridized with 1 nM LNA™ U6 snRNA probe and 50 nM double-DOG LNA™ miRNA probe complementary to miR-23b and LNA™ scrambled miRNA probe, respectively (all the probes were purchased from Exiqon, Demnark). After incubating with and removing the blocking solution, the slides were sequentially incubated with anti-DIG reagent and AP substrate (all were procured from Roche, Mannheim, Germany) protecting from light. Finally, KTBT buffer was added to stop the reaction. Nuclear Fast Red (Solarbio, China) was used for counterstaining of the nucleus. The slides were finally observed under light microscopy.

Statistical Analyses:

All statistical analyses were performed using Prism 6.0 (GraphPad Prism, San Diego, CA, USA) and SPSS 22.0 (SPSS Inc., Chicago, IL, USA). Mann-Whitney U test was performed to compare the expression of miR-23b in STs and plasma between RA and HC. The relationships between the expression of miR-23b in plasma and other parameters were evaluated using the Spearman correlation. Wilcoxon matched paired test was used to compare the miRNA-23b levels of 30 patients before and after treatment. All tests were two-tailed, and a P value of < 0.05 was considered as statistically significant.

RESULTS

miR-23b levels were elevated in FLSs, STs, and plasma in RA patients

The expression profiles of miRNA in FLSs in 8 RA patients were determined by miRNA microarray analysis. Compared to FLSs obtained from 4 HC, the expression of 5 miRNAs (miR-28-3p, miR-378, miR-BART10, miR-432, and miR-23b) was significantly increased in FLS obtained from RA (Figure 1a).

Figure 1. Increased miR-23b expression in FLS and STs obtained from RA patients.

Figure 1.

Open in a new tab

(A) Compared to the trauma controls (H1-4), significantly altered expression of 20 miRNAs was found in FLS obtained from RA by miRNA microarray analysis (red: up regulation, blue: down regulation, grey: signals not detected). The threshold value of differentially expressed miRNAs was ≥1.5-fold change or ≤ 0.67-fold change (p ≤ 0.05). (B) miR-23b levels determined using RT-PCR in the STs obtained from RA patients (n = 34) were higher than that in HC (trauma control, n = 36). The P value derived by using Mann-Whitney U test is indicated. (C) In situ hybridization analysis showed that miR-23b expression in the STs obtained from RA patients was higher than that obtained from HC (scramble means negative control; u6 means positive control; FLS, fibroblast like synoviocytes; STs, synovial tissues; RA, rheumatoid arthritis).

To confirm our microarray data, miR-23b levels were estimated by RT-QPCR in the STs obtained from 34 RA patients. Compared to the STs obtained from 36 control trauma patients, the miR-23b levels were significantly up-regulated in that obtained from RA patients (Figure 1b). To obtain a deeper insight into the expression of miR-23b in STs, miRNA in situ hybridization was performed to compare the tissue levels of miR-23b in STs obtained from RA and HC. Consistent with the results of PCR, the miR-23b staining was at a very low level in the normal STs from HC, but was highly intense in the STs from RA patients. The miR-23b staining was clearly seen in the synoviocytes of the lining and sublining from the STs of RA patients (Figure 1c).

Increased miR-23b expression in plasma of RA patients

Escaping from RNase in the circulation, miRNA can exist in a stable form as an effective marker of a variety of diseases (Chakraborty & Das, 2016). Since miR-23b is a promising novel autoimmunity regulator molecule (Remakova, Svitalkova, Skoda, Vencovsky, & Novota, 2013), we evaluated the miR-23b expression in plasma. The relative expression of miR-23b was compared between 109 RA patients and 48 HC. As shown in Figure 2, the plasma miR-23b expression in RA was significantly higher than that in HC. We further analyzed the difference between the male and female patients, but there was no significance as P=0.254 (data not shown).

Figure 2. Increased miR-23b expression in the plasma of RA patients.

Figure 2.

Open in a new tab

The levels of miR-23b in plasma of 109 RA patients were higher than that in 48 healthy controls (HC). The P value derived using Mann-Whitney U test are indicated (RA, rheumatoid arthritis).

Plasma miR-23b level correlated with the clinical patterns of RA

Considering that miR-23b is an important inflammatory mediator (Zheng et al., 2012) and the activity of RA is closely related to the severity of inflammation (Smolen et al.), the plasma miR-23b level and the pattern of RA disease activity was thoroughly analyzed. The plasma miR-23b level showed a positive correlation with CRP, ESR, hs-CRP, and DAS28 (Table 2 and Figure 3a).

Figure 3. Plasma miR-23b levels correlate with the clinical patterns of RA.

Figure 3.

Open in a new tab

(A) Correlations of plasma miR-23b levels with CRP (n = 93), ESR (n = 98), hs-CRP (n = 75), DAS28 (n = 73), PLT (n = 104), Hb (n = 107), TBIL (n = 105), DBIL (n = 105), IBIL (n = 105), LDL-C (n=66) and TC (n = 66). Correlation coefficients (r) obtained from Spearman rank-order test are shown. (B) Comparison between the levels of plasma miR-23b in ANA-positive RA (n = 24) and ANA-negative RA (n = 26) are shown. (C) Comparison between the levels of plasma miR-23b in RA patients with (n = 31) or without anorexia (n=78), RA patients with (n = 31) or without fatigue (n=78), and RA patients with (n=43) or without ILD (n=48) are shown. Mann-Whitney U test was used (CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; hs-CRP, high sensitivity C reactive protein; DAS28, disease activity score28; PLT, platelet; Hb, hemoglobin; TBIL, total bilirubin; DBIL, direct bilirubin; IBIL, indirect bilirubin; ANA, antinuclear antibody; ILD, interstitial lung disease; TC, total cholesterol; RA, rheumatoid arthritis).

RA is a systemic inflammatory disease, and hence, we collected other clinical indices in addition to the patterns of disease activity, such as white blood cells (WBC), platelet (PLT), alanine aminotransferase (ALT), aspartate transaminase (AST), etc. The correlation between the levels of plasma miR-23b and the levels of these parameters was analyzed (Table 2). The levels of plasma miR-23b positively correlated with PLT count but negatively correlated with Hb, TBIL, DBIL, IBIL, TC and LDL-C levels (Figure 3a). The plasma miR-23b levels, however, showed no correlation with the other parameters (Table 2).

Many types of antibodies are present in the blood of RA patients because of the autoimmune nature of the disease. We further estimated the levels of 3 crucial antibodies, ACPA, RF, and ANA, and evaluated their correlation with miR-23b expression. The results showed that plasma miR-23b expression in the ANA-positive RA patients was higher as compared to that in the ANA-negative RA patients (Figure 3b). However, the plasma levels of miR-23b did not correlate with RF or ACPA (Table 3).

We further obtained data regarding the clinical features of RA patients involved in this study and compared the plasma miR-23b levels in patients having positive clinical features with those having negative clinical features. The patients having anorexia showed higher levels of miR-23b in plasma than those without anorexia. A similar result was found in patients with fatigue (left and middle panel in Figure 3c). There was no significant difference between patients having extra-articular symptoms, systemic lesions, and articular symptoms including swollen joints, tender joints, and joint deformity and the patients not having these symptoms (Table 4). However, the patients having ILD showed a remarkably lower level of miR-23b compare to those without ILD (right panel in Figure 3c).

Table 4.

Clinical manifestation of RA patientsa

Manifestation Positive number (total number) Percentage (%) p
Morning stiffness 20 (109) 18.35 0.133
Swollen joints 79 (109) 72.48 0.519
Tender joints 86 (109) 78.90 0.782
Joint deformity 22 (109) 20.18 0.319
Fatigue 31 (109) 28.44 0.0161*
Inappetence 31 (109) 28.44 0.0365*
ILD 43 (91) 46.15 0.0076**
Lymph nodes 27 (76) 35.52 0.737
Open in a new tab
a

RA, rheumatoid arthritis; ILD, interstitial lung disease.

*

expressed as P< 0.05

**

expressed as P< 0.01.

Effects of treatment on plasma miR-23b levels in RA patients

We performed follow-up studies with plasma miR-23b levels in 30 RA patients after treatment with infliximab and/or disease-modifying antirheumatic drags (DMARDs) for an average of 37 days. Among these 30 RA patients, 8 achieved ACR20, 9 achieved ACR50, 4 achieved ACR70, and 9 were non-responders. The treatment plan included infliximab combined with DMARDs and non-steroidal anti-inflammatory drags (NSAIDs) for 1 patient, DMARDs only or combined with traditional Chinese medicinal preparation for 5 patients, NSAIDS combined with DMARDs and/or traditional Chinese medicinal preparation for 19 patients, glucocorticoid combined with DMARDs and/or traditional Chinese medicinal preparation for 4 patients, and irregular treatment during two follow-up visits for 1 patient. As shown in Figure 4, the plasma miR-23b levels in 21 patients with active RA were markedly decreased following clinical improvement with treatment (Wilcoxon matched paired test, P = 0.029). Importantly, for the 9 patients who did not achieve an ACR20 response, the plasma miR-23b levels did not change significantly before and after treatment (Wilcoxon matched paired test, P = 0.25).

Figure 4. Plasma levels of miR-23b in RA patients before and after therapy.

Figure 4

Open in a new tab

Paired plasma samples were obtained from 30 patients before treatment and after an average of 37 days of treatment with DMARDs and/or NSAIDS. (a) 21 patients achieved at least an ACR20 response, (b) 9 non-responders did not achieve an ACR20 response [RA, rheumatoid arthritis; DMARDs, disease-modifying anti-rheumatic drugs; NSAIDS, nonsteroidal anti-inflammatory drugs; ACR20, improvement of ≥ 20% according to the American College of Rheumatology (ACR) response criteria].

DISCUSSION

We investigated the relationship between miR-23b and RA in the present study. Our results revealed that RA patients show higher levels of miR-23b in STs and plasma than HCs. The levels of plasma miR-23b correlated with the serological and clinical features of RA disease activity. Further, we have observed that plasma miR-23b levels in RA patients diminished following successful treatment and clinical improvement.

miR-23b lias been demonstrated to play complicated roles in several pathological and physiological process, such as diabetes (Grieco et al., 2017) tumor (Kou, Zhou, Han, Zhuang, & Qian, 2016), and inflammation and immune disorders (Zheng et al., 2012). Previous reports have demonstrated that miR-23b is dysregulated in inflammatory and autoimmune diseases. For instance, miR-23b was found to be up-regulated in the colonic mucosa of patients with Crohn’s disease (Lin et al., 2013) and in the kidneys of mice with SLE (He et al., 2016). The present study demonstrated that higher miR-23b levels were detected in plasma in RA patients as compared to that in HC. We found a significant increase in the miR-23b levels in STs in RA patients as compared to that in HC. This is contrary to the results of Zhu, et al. who showed that miR-23b levels deceased in STs in RA patients as compared to that in HC (Zhu et al., 2012). The inconsistency between these results might be attributed to the difference in sample size. The number of ST samples obtained from RA patients and HC were 17 and 3, respectively, in the study of Shu Zhu, et al. (Zhu et al., 2012), whereas, the number of ST samples obtained from RA patients and HC were 34 and 36 in our study. Besides, our result of RT-QPCR with STs is consistent with the result of in situ hybridization. The plasma miR-23b levels correlated well with the clinical features of RA and the level was down-regulated after treatment. However, an increasing miR-23b level has been reported to promote Tregs differentiation, while number of Tregs has been demonstrated to be decreased in RA patients (Zheng et al., 2012). We hypothesized that the up-regulation of plasma miR-23b might be due to feedback from the significantly decreased number of Tregs in RA(Al-Zifzaf et al., 2015).

We further assessed the relationship between miR-23b and clinical patterns of RA. Our results revealed that plasma miR-23b levels correlated well with several important biological and clinical indices which could assess the RA disease activity, including CRP, ESR, hs-CRP(Benhadjmohamed et al., 2017), and DAS28 score. The RA patients with anorexia showed higher levels of miR-23b in plasma than those with preserved appetite, and similar results were obtained with fatigue. Importantly, the level of plasma miR-23b was significantly decreased along with improvement of clinical symptoms. These results indicated that plasma miR-23b might be a potential biomarker for RA disease activity. The existing laboratory markers of RA often lack sensitivity and specificity without being 100% truly positive in RA or 100% truly negative in normal subjects (Pincus & Sokka, 2009). The plasma miR-23b level could be a useful marker in RA patients who are negative to the existing test results.

In addition, our results showed that plasma miR-23b level positively correlated with PLT count and negatively correlated with Hb, TBIL, DBIL, IBIL, LDL-C and TC levels. Previous reports have shown that the PLT count in the peripheral blood and synovium of RA patients is higher than that in HC and this count decreases in the peripheral blood following treatment with TNF-α inhibitor (Łukasik, Makowski, & Makowska, 2018). Low Hb in patients with RA upon TNF-α inhibitor treatment indicates an ongoing radiographic progression of the disease(Möller et al., 2017). Pan, et al. indicated that RA patients have a lower serum bilirubin level than HCs and that serum bilirubin level negatively correlates with the DAS28 score (Peng, Wang, & Pan, 2017). Serum bilirubin level is an indicator of hepatic function, and hence, miR-23b may take part in the mechanism of hepatic injury in RA. Meanwhile, reports have shown that the TC and LDL-C level in RA patients is lower than that in HCs(Plutzky & Liao 2018) and LDL-C negatively correlates with CRP, ESR, and DAS28(Charles-Schoeman et al., 2016). The increase in TC and LDL-C level in RA patients after treatment with DMARDs correlates with the alleviation of inflammation(Plutzky & Liao 2018). Taking together, we can conclude that plasma miR-23b could serve as a new biomarker for RA disease activity.

Reports before have found that miR-23b can alleviate fibrosis in diabetic nephropathy(Zhao et al., 2016), and the miR-23b/27b cluster suppresses activation of hepatic stellate cells which plays a pivotal role in the fibrogenesis process during chronic liver injury(Zeng et al., 2016). And we found that miR-23b levels in RA patients without ILD were higher than those with ILD. But The role of miR-23b playing in the pathological process of RA complicating with ILD needs further research to explore.

Interestingly, the plasma miR-23b level in ANA-positive RA was higher than in ANA-negative RA, which may be because of the fact that miR-23b can silence B cell Blimp-1 expression, thereby dampening the class-switch and hypermutated autoantibody response including ANA and immunopathology(White et al., 2014).

Elucidation of the specific mechanism regulating miR-23b expression can provide additional support for the involvement of miRNAs in the pathogenesis of RA. This needs to be investigated in further studies. Further research is also required to explain the correlation between plasma miR-23b level and ANA and ILD in RA patients and to examine the specificity of our proposed biomarkers by analyzing plasma from patients with infection, injury, and other inflammatory diseases.

CONCLUSION

The miR-23b level was found to be increased in the synovial tissues and plasma of RA patients as compared to that in HCs. Furthermore, the plasma miR-23b level in RA patients positively correlated with CRP, ESR, hs-CRP, DSA28, and PLT, and negatively correlated with TBIL, DBIL, IBIL, Hb, TC and LDL-C. In addition, the plasma miR-23b level declined significantly following treatment. Since miRNA has high stability in body fluids(Mitchell et al., 2008) and plasma miR-23b can be easily accessed through various methods, miR-23b possesses every property for being considered as a promising biomarker reflecting the degree of inflammatory disease activities and therapeutic effects in patients with RA.

Acknowledgements

We would like to thank Dr. Dawei Li for helping in the analysis of miRNA microarray data. This work was supported by grants obtained from the National Natural Science Foundation of China (81171680 to Y.W.), the Social Development Project of Jiangsu Province (BE2015632 to Y.W.), the Natural Science Foundation of the Jiangsu for Youth (BK20180182 to S.N.), and the Changzhou Science and Technology Program (CJ20180057 to S.N.). Additional support was provided by the National Institutes of Health (R01 AR049069 to A.J. v. W.).

Abbreviations:

ACPA:

Anti-Cyclic Citrullinated Peptide Antibody

ANA:

Anti-Nuclear Antibody

AST:

Aspartate Transaminase

CRP:

C-Reactive Protein

DBIL:

Direct Bilirubin

ESR:

Erythrocyte Sedimentation Rate

FLS:

Fibroblast-Like Synoviocytes

HC:

Healthy Controls

Hs-CRP:

High Sensitivity C-Reactive Protein

IBIL:

Indirect Bilirubin

ILD:

Interstitial Lung Disease

LDL-C:

Low Density Lipoprotein Cholesterol

miRNAs:

microRNAs

PLT:

Platelet Count

RA:

Rheumatoid Arthritis

RF:

Rheumatoid Factor

SLE:

Systemic Lupus Erythematosus

STs:

Synovial Tissues

TBIL:

Total Bilirubin

Footnotes

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

Conflict of interest

The authors declare that they have no conflict of interest.

References

  1. Al-Zifzaf DS, Bakry SAE, Mamdouh R, Shawarby LA, Ghaffar AYA, Amer HA, … Rahman RA (2015). FoxP3+T regulatory cells in Rheumatoid arthritis and the imbalance of the Treg/TH17 cytokine axis. Egyptian Rheumatologist, 37(1), 7–15. [Google Scholar]
  2. Aletaha D, Neogi T, Silman AJ, Funovits J, Felson DT, Birnbaum NS, … Cohen MD (2010). 2010 Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis & Rheumatism, 62(9), 2569. [DOI] [PubMed] [Google Scholar]
  3. Alsaleh G, Suffert G, Semaan N, Juncker T, Frenzel L, Gottenberg JE, … Wachsmann D (2009). Bruton’s tyrosine kinase is involved in miR-346-related regulation of IL-18 release by lipopolysaccharide-activated rheumatoid fibroblast-like synoviocytes. Journal of Immunology, 182(8), 5088. [DOI] [PubMed] [Google Scholar]
  4. Au Yeung CL, Tsang TY, Yau PL, & Kwok TT (2017). Human papillomavirus type 16 E6 suppresses microRNA-23b expression in human cervical cancer cells through DNA methylation of the host gene C9orf3. Oncotarget. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Ayyadurai S, Charania MA, Xiao B, Viennois E, Zhang Y, & Merlin D (2014). Colonic miRNA Expression/Secretion, Regulated by Intestinal Epithelial PepT1, Plays an Important Role in Cell-to-Cell Communication during Colitis. PLoS ONE, 9(2), e87614. doi: 10.1371/journal.pone.0087614 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Benaglio F, Vitolo B, Scarabelli M, Binda E, Bugatti S, Caporali R, … Manzo A (2015). The draining lymph node in rheumatoid arthritis: current concepts and research perspectives. Biomed Research International, 2015(9), 420251. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Benhadjmohamed M, Khelil S, Dbibis MB, Khlifi L, Chahed H, Ferchichi S, … Miled A (2017). Hepatic Proteins and Inflammatory Markers in Rheumatoid Arthritis Patients. Iranian Journal of Public Health, 46(8), 1071–1078. [PMC free article] [PubMed] [Google Scholar]
  8. Cao J, Liu J, Long J, Fu J, Huang L, Li J, … Yan Y (2017). microRNA-23b suppresses epithelial-mesenchymal transition (EMT) and metastasis in hepatocellular carcinoma via targeting Pyk2. Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie, 89, 642–650. doi: 10.1016/j.biopha.2017.02.030 [DOI] [PubMed] [Google Scholar]
  9. Chakraborty C, & Das S (2016). Profiling cell-free and circulating miRNA: a clinical diagnostic tool for different cancers. Tumor Biology, 37(5), 5705–5714. doi: 10.1007/s13277-016-4907-3 [DOI] [PubMed] [Google Scholar]
  10. Charles-Schoeman C, Wang X, Lee YY, Shahbazian A, Navarro-Millan I, Yang S, … Curtis JR (2016). Association of Triple Therapy With Improvement in Cholesterol Profiles Over Two-Year Followup in the Treatment of Early Aggressive Rheumatoid Arthritis Trial. Arthritis Rheumatol, 68(3), 577–586. doi: 10.1002/art.39502 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Churov AV, Oleinik EK, & Knip M (2015). MicroRNAs in rheumatoid arthritis: Altered expression and diagnostic potential. Autoimmunity Reviews, 14(11), 1029. [DOI] [PubMed] [Google Scholar]
  12. Fukumoto I, Koshizuka K, Hanazawa T, Kikkawa N, Matsushita R, Kurozumi A, … Seki N (2016). The tumor-suppressive microRNA-23b/27b cluster regulates the MET oncogene in oral squamous cell carcinoma. International Journal of Oncology, 49(3), 1119. [DOI] [PubMed] [Google Scholar]
  13. Grieco FA, Sebastiani G, J J-M, Villate O, Marroqui L, Ladrière L, … Marchetti P (2017). MicroRNAs miR-23a-3p, miR-23b-3p, and miR-149-5p Regulate the Expression of Proapoptotic BH3-Only Proteins DP5 and PUMA in Human Pancreatic β-Cells. Diabetes, 66(1), 100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. He X, Zhang Y, Ai Z, Kang Z, Zhang X, Li G, … Qu S (2016). Suppression of interleukin 17 contributes to the immunomodulatory effects of adipose-derived stem cells in a murine model of systemic lupus erythematosus. Immunologic Research, 1–11. [DOI] [PubMed] [Google Scholar]
  15. Hellmich B (2014). Rheumatoide Arthritis. CME, 11(1), 53–64. [Google Scholar]
  16. Hruskova V, Jandova R, Vernerova L, Mann H, Pecha O, Prajzlerova K, … Senolt L (2016). MicroRNA-125b: association with disease activity and the treatment response of patients with early rheumatoid arthritis. Arthritis Research & Therapy, 18(1), 124. doi: 10.1186/s13075-016-1023-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Jian Q, Yan Y, Shengping H, Yanbin Q, Qian W, Hongsong Y, … Peizeng Y (2014). Increased Notch pathway activation in Beh?et’s disease. Rheumatology, 53(5), 810–820. [DOI] [PubMed] [Google Scholar]
  18. Kim KW, Kim BM, Moon HW, Lee SH, & Kim HR (2015). Role of C-reactive protein in osteoclastogenesis in rheumatoid arthritis. Arthritis Research & Therapy, 17(1), 41. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kou C, Zhou T, Han X, Zhuang H, & Qian H (2016). Downregulation of mir-23b in plasma is associated with poor prognosis in patients with colorectal cancer. Oncology Letters, 12(6), 4838–4844. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Lin J, Cao Q, Zhang J, Li Y, Shen B, Zhao Z, … Bronner MP (2013). MicroRNA expression patterns in indeterminate inflammatory bowel disease. Modern Pathology An Official Journal of the United States & Canadian Academy of Pathology Inc, 26(1), 148. [DOI] [PubMed] [Google Scholar]
  21. Łukasik ZM, Makowski M, & Makowska JS (2018). From blood coagulation to innate and adaptive immunity: the role of platelets in the physiology and pathology of autoimmune disorders. Rheumatology International. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Mcinnes IB, & O’Dell JR (2010). State-of-the-art: rheumatoid arthritis. Annals of the Rheumatic Diseases, 69(11), 1898–1906. [DOI] [PubMed] [Google Scholar]
  23. Mitchell PS, Parkin RK, Kroh EM, Fritz BR, Wyman SK, Pogosova-Agadjanyan EL, … Allen A (2008). Circulating microRNAs as stable blood-based markers for cancer detection. Proceedings of the National Academy of Sciences of the United States of America, 105(30), 10513. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Möller B, Everts-Graber J, Florentinus S, Li Y, Kupper H, & Finckh A (2017). Low Hemoglobin Predicts Radiographic Damage Progression in Early Rheumatoid Arthritis - Secondary Analysis from a Phase III trial. Arthritis Care & Research. [DOI] [PubMed] [Google Scholar]
  25. Murata K, Yoshitomi H, Tanida S, Ishikawa M, Nishitani K, Ito H, & Nakamura T (2010). Plasma and synovial fluid microRNAs as potential biomarkers of rheumatoid arthritis and osteoarthritis. Arthritis Research & Therapy, 12(3), R86–R86. doi: 10.1186/ar3013 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Ni S, Miao K, Zhou X, Xu N, Li C, Zhu R, … Wang Y (2015). The involvement of follistatin-like protein 1 in osteoarthritis by elevating NFkappaB-mediated inflammatory cytokines and enhancing fibroblast like synoviocyte proliferation. Arthritis Res Ther, 17, 91. doi: 10.1186/s13075-015-0605-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Peng YF, Wang JL, & Pan GG (2017). The correlation of serum bilirubin levels with disease activity in patients with rheumatoid arthritis. Clinica Chimica Acta, 469, 187–190. [DOI] [PubMed] [Google Scholar]
  28. Pincus T, & Sokka T (2009). Laboratory tests to assess patients with rheumatoid arthritis: advantages and limitations. Rheum Dis Clin North Am, 35(4), 731–734. [DOI] [PubMed] [Google Scholar]
  29. Plutzky J, & Liao KP (2018). Lipids in RA: Is Less Not Necessarily More? Curr Rheumatol Rep, 20(2), 8. doi: 10.1007/s11926-018-0715-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Reid G, Kirschner MB, & van Zandwijk N Circulating microRNAs: Association with disease and potential use as biomarkers. Critical Reviews in Oncology / Hematology, 80(2), 193–208. doi: 10.1016/j.critrevonc.2010.11.004 [DOI] [PubMed] [Google Scholar]
  31. Remakova M, Svitalkova T, Skoda M, Vencovsky J, & Novota P (2013). The expression profile of miR-23b is not altered in peripheral blood mononuclear cells of patients with idiopathic inflammatory myopathies [version 1; referees: 2 approved] (Vol. 2). [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Smolen JS, Aletaha D, & McInnes IB Rheumatoid arthritis. The Lancet, 388(10055), 2023–2038. doi: 10.1016/S0140-6736(16)30173-8 [DOI] [PubMed] [Google Scholar]
  33. Torpy JM (2011). Rheumatoid Arthritis. Journal of the American Medical Association, 305(17), 1824.21540429 [Google Scholar]
  34. Vojdani A (2014). A Potential Link between Environmental Triggers and Autoimmunity. Autoimmune Diseases, 2014(437231), 437231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. White CA, Pone EJ, Lam T, Tat C, Hayama KL, Li G, … Casali P (2014). HDAC Inhibitors Upregulate B Cell microRNAs that Silence AID and Blimp-1 Expression for Epigenetic Modulation of Antibody and Autoantibody Responses. Journal of Immunology, 193(12), 5933–5950. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Wu J, Ji C, Jia Q, Cao F, Yong L, Zhang X, & Wang L (2017). Bone marrow mesenchymal stem cells inhibit dendritic cells differentiation and maturation by microRNA-23b. BSR20160436. [DOI] [PMC free article] [PubMed]
  37. Wu M, Gu J-T, Yi B, Tang Z-Z, & Tao G-C (2015). microRNA-23b regulates the expression of inflammatory factors in vascular endothelial cells during sepsis. Experimental and therapeutic medicine, 9(4), 1125–1132. doi: 10.3892/etm.2015.2224 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Zeng XY, Zhang YQ, He XM, Wan LY, Wang H, Ni YR, … Liu CB (2016). Suppression of hepatic stellate cell activation through downregulation of gremlin1 expression by the miR-23b/27b cluster. Oncotarget, 7(52), 86198–86210. doi: 10.18632/oncotarget.13365 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Zhang Y, Li H, Wu N, Xin D, & Yi Z (2017). Retrospective study of the clinical characteristics and risk factors of rheumatoid arthritis-associated interstitial lung disease. Clinical Rheumatology, 36(4), 1–7. [DOI] [PubMed] [Google Scholar]
  40. Zhao B, Li H, Liu J, Han P, Zhang C, Bai H, … Chu Y (2016). MicroRNA-23b Targets Ras GTPase-Activating Protein SH3 Domain-Binding Protein 2 to Alleviate Fibrosis and Albuminuria in Diabetic Nephropathy. J Am Soc Nephrol, 27(9), 2597–2608. doi: 10.1681/ASN.2015030300 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zheng J, Jiang HY, Li J, Tang HC, Zhang XM, Wang XR, … Xu G (2012). MicroRNA-23b promotes tolerogenic properties of dendritic cells in vitro through inhibiting Notch1/NF-κB signalling pathways. Allergy, 67(3), 362. [DOI] [PubMed] [Google Scholar]
  42. Zhu S, Pan W, Song X, Liu Y, Shao X, Tang Y, … Liu W (2012). The microRNA miR-23b suppresses IL-17-associated autoimmune inflammation by targeting TAB2, TAB3 and IKK-α. Nature Medicine, 18(18), 1077–1086. [DOI] [PubMed] [Google Scholar]

RESOURCES