Identification of differentially expressed miRNAs associated with chronic kidney disease–mineral bone disorder

Abstract

The purpose of this study is to characterize a meta-signature of differentially expressed mRNA in chronic kidney disease (CKD) to predict putative microRNA (miRNA) in CKD–mineral bone disorder (CKD–MBD) and confirm the changes in these genes and miRNA expression under uremic conditions by using a cell culture system. PubMed searches using MeSH terms and keywords related to CKD, uremia, and mRNA arrays were conducted. Through a computational analysis, a meta-signature that characterizes the significant intersection of differentially expressed mRNA and expected miRNAs associated with CKD–MBD was determined. Additionally, changes in gene and miRNA expressions under uremic conditions were confirmed with human Saos-2 osteoblast-like cells. A statistically significant mRNA meta-signature of upregulated and downregulated mRNA levels was identified. Furthermore, miRNA expression profiles were inferred, and computational analyses were performed with the imputed microRNA regulation based on weighted ranked expression and putative microRNA targets (IMRE) method to identify miRNAs associated with CKD occurrence. TLR4 and miR-146b levels were significantly associated with CKD–MBD. TLR4 levels were significantly downregulated, whereas pri-miR-146b and miR-146b were upregulated in the presence of uremic toxins in human Saos-2 osteoblast-like cells. Differentially expressed miRNAs associated with CKD-MBD were identified through a computational analysis, and changes in gene and miRNA expressions were confirmed with an in vitro cell culture system.

Keywords

chronic kidney disease / microRNA / mineral bone disorder / uremia

Introduction

Chronic kidney disease (CKD) is characterized by reduced kidney function, which can result in uremia [1]. Uremia is defined as the presence of symptoms secondary to renal uremic toxins or uremic retention solute [2] and has been associated with significant changes in energy metabolism, insulin resistance, cardiovascular disease, and mineral-bone disorders [3–5]. CKD–mineral bone disorder (CKD–MBD) is a systemic mineral metabolic disorder associated with renal dysfunction. The bone lesions that result from CKD–MBD are a consequence of abnormal mineral metabolism [6]. Abnormal mineral metabolism can promote arterial calcification and arterial stiffness, which may lead to left ventricular hypertrophy and potentiation of atherogenesis [7]. Therefore, the changes in mineral metabolism caused by CKD represent a risk factor that contributes to the high cardiovascular mortality of CKD patients [8].

MicroRNA (miRNA) are single-stranded, non-coding RNAs that regulate gene expression at the post-transcriptional level [9]. Changes in miRNA expression are associated with cell proliferation, differentiation, and apoptosis as well as function-regulating mechanisms that link inflammation, cancer, immune response, and development [10–12]. In addition, miRNAs are promising biomarkers of early disease detection and progression as well as targets of highly efficient treatments [13,14].

Abnormalities in mineral homeostasis due to kidney failure result in decreased renal phosphate excretion, elevated fibroblast growth factor 23 levels, and increased levels of the parathyroid hormone (PTH) [6]. However, in early CKD, serum PTH, vitamin D, calcium, and phosphorus concentrations are still within the normal range due to compensatory mechanisms [15]. Several studies have identified critical cellular and molecular mediators underlying CKD–MBD [16]; however, the changes in miRNA expression correlated with CKD–MBD remain unexplored. miRNA may be used as a sensitive biomarker for early detection of CKD–MBD or to relate the pathology of this disease.

Although previous studies have identified differentially expressed miRNAs in uremia [17,18], these studies present inconsistencies due to small sample sizes and varying results obtained by different groups; the latter can be explained by different laboratory protocols, microarray platforms, and analysis techniques employed [19]. Recent studies have shown that systematic integration of gene expression data from multiple sources (meta-analyses) can increase the statistical power of detecting differentially expressed genes while allowing for heterogeneity assessment. These types of analyses may lead to accurate, highly robust, reproducible predictions [20].

In this study, we performed a meta-analysis of array-based gene expression datasets from uremia studies to determine mRNA expression changes and predict miRNA–gene expression interactions associated with CKD. miRNA–gene expression interactions related to CKD–MBD were predicted using bioinformatics resources and tools. The changes in these genes and miRNA expressions under uremic conditions were confirmed with an in vitro cell culture system.

Materials and methods

Selection of datasets

Studies on uremia expression profiling were identified on February 11, 2015, by searching the Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo). The keywords “chronic kidney disease,” “uremia,” “mRNA,” “microarray,” “end-stage renal disease,” “renal failure,” “mineral bone disorder,” and “human [organism]” and their combinations were used. Only original experimental articles that compared the expression of mRNAs between the blood of uremic patients and that of normal controls were retained. All eligible publications met the following criteria: (1) the studies were associated with the diagnostic value of mRNA for the diagnosis of uremia and (2) the studies provided sufficient data to assess the diagnostic value of mRNA in uremia. The exclusion criteria were as follows: (1) duplicate publications, (2) studies without sufficient data, and (3) letters, reviews, editorials, meeting abstracts, and case reports.

Data extraction

Information was extracted from each identified study. This information included GEO accession number, sample type, platform, number of cases and controls, references, and expression data. Two independent reviewers extracted data from the original studies, and when no consensus was achieved, a third reviewer resolved the discrepancies between the two initial reviewers. The studies used the gene expression platforms HG-U133A, HG-U133Plus 2.0, and HG-U133A_2. Twelve studies were initially identified, among which three studies (GSE15072, GSE37171, and GSE43484) were extracted. The detailed information of these downloaded datasets is summarized in Table 1.

Gene expression analysis

The raw data of the cell file of the mRNA transcription array sets were processed with the Bioconductor Package [21], and the “fRMA” (frozen robust multiarray analysis) was implemented in R software (version 3.4, http://www.R-project.org/) [22]. To identify differentially expressed genes, a significance analysis of microarrays was performed using Student’s t-tests between uremic patients and healthy controls.

miRNA prediction

Apprehensive human miRNA target datasets (miRNOME) were generated by merging TargetScan, miRanda, miRDB, and miRTarBase. The downloaded versions are described in Supplementary Table S1. miRNOME contains 5813 distinct human miRNAs, 22 174 predicted putative miRNA gene targets, and 1 767 926 distinct miRNA–target relationships. Upregulated (or downregulated) miRNAs were set up by using the IMRE method (http://www.lussierlab.org/IMRE) [23]. The scores representing miRNA expression levels were calculated based on three downloaded mRNA expression sets. After calculating the scores representing the expression levels of each miRNA for each study subject, we listed potential miRNAs that distinguish the uremic condition of patients with CKD from that of healthy controls. For each miRNA, an empirical P value was obtained from 1000 permutations. The cutoff P value to select potential miRNAs was equal to or less than 0.1. We used this liberal value to reduce type II errors in the initial analysis. Analysis was performed with the R statistical package with the Bioconductor package Twilight [24].

Classification of differentially expressed genes

To select genes and pathways related to osteogenesis and bone metabolism, genes were annotated through gene set enrichment analysis (GSEA, http://www.broad.mit.edu/gsea), classification by gene ontology (GO) categories (Gene Ontology, http://www.geneontology.org/), and DAVID gene functional classification (https://david.ncifcrf.gov/home.jsp). A pathway enrichment analysis was performed based on the Kyoto Encyclopedia of Genes and Genomes Database (KEGG, http://www.genome.jp/kegg). Genes that showed nominal significance levels of P<0.05 were selected, and out of those selected, a set of downregulated genes with upregulated miRNA was further selected for in vitro experiments to identify miRNAs as diagnostic biomarkers.

Reagents

All chemicals, including PTH, were purchased from Sigma-Aldrich Co. (St. Louis, MO, USA) unless otherwise specified. Diallyl disulfide (DADS), diallyl trisulfide (DATS), NG, and NG-dimethyl-L-arginine (ADMA) were purchased from Cayman Chemical Co. (Ann Arbor, MI, USA). Indoxyl sulfate was purchased from Santa Cruz Biotechnology, Inc. (Dallas, TX, USA), p-cresylsulfate (pCS) and L-homocysteine were purchased from Tokyo Chemical Industry, Co., Ltd. (Tokyo, Japan), p-cresylglucurodine (pCG) was procured from Toronto Research Chemicals Inc. (Toronto, ON, Canada), and StemTAG™ alkaline phosphatase activity assay kits were obtained from Cell Biolabs, Inc. (San Diego, CA, USA).

Cell culture and treatment with uremic toxins

Osteoblast-like osteosarcoma cell lines, namely, Saos-2 cells, were cultured in RPMI 1640 containing 10% fetal bovine serum (Gibco, Gran Island, NY, USA) and 10 U/ml of penicillin/streptomycin (Gibco, Gran Island, NY, USA) under 5% CO₂ at 37 °C. The cells (1×10⁶ cells/ml) were seeded on a dish and incubated at different times with uremic toxins [4] until the following day. The final concentrations were as follows: pCS, 0.48 mmol/L; pCG, 80 mmol/L; PTH, 106 pmol/L (1000 pg/ml); indoxyl sulfate, 80 mmol/L; ADMA, 1 mmol/L; homocysteine, 100 mmol/L; DADS, 200 mmol/L; and DATS, 200 mmol/L. Cells were collected and frozen immediately at −80 °C until further processing.

Quantitative real-time polymerase chain reaction (PCR)

Total RNA was isolated with the RNeasy Mini Kit (Qiagen, Valencia, CA, USA) following the manufacturer’s instructions. RNA concentration was measured with Nanodrop ND 1000 (Thermo Fisher Scientific, Waltham, MA, USA). cDNA was transcribed using the AMPIGENE cDNA synthesis kit (Enzo Life Sciences, Inc., Villeurbanne, France), and specific mRNAs were detected through real-time PCR with specific primers (Table 2) and the qPCR premix kit (AccuPower, Bioneer Co., Daejeon, Republic of Korea) in a 7500 real-time PCR system (Life Technologies, Gaithersburg, MD, USA). ACTB was used as a control gene, and miRNA was isolated using the miRNeasy Mini Kit (Qiagen) according to the manufacturer’s protocols. To analyze the miR-146b expression, TaqMan Advanced miRNA cDNA Synthesis Kit (Applied Biosystems, Foster City, CA, USA) and TaqMan Universal PCR Master Mix (Applied Biosystems) were used. RNU6B was utilized as a small nuclear RNA control, and relative mRNA or miRNA expression was determined with the comparative threshold (Ct) method (2^-ΔΔCt).

Alkaline phosphatase (ALP) activity assay

Osteoblast differentiation was evaluated by measuring ALP activity. An aliquot of cell lysate was added to the ALP substrate buffer containing 2 mg/ml of p-nitrophenyl phosphate, 0.1 mol/L of diethanolamine, 1 mmol/L of MgCl₂, and 0.1% Triton X-100 (pH 9.8). The mixture was incubated at 37 °C for 30 min. The enzyme reaction was stopped with the addition of 0.5 mol/L of NaOH, and the absorbance was read at 405 nm. A calibration line was constructed from different concentrations of p-nitrophenol.

Statistical analyses

Statistical analyses were performed with R software. A probability of P<0.05 was considered statistically significant. Different groups were compared through independent sample t-tests. Data are presented as mean±standard deviation (SD).

Results

Short overviews of the included studies

After the search and selection process, three studies met the inclusion criteria. GEO37171 involved 75 uremic patients (stage 5) and 20 normal patients. GEO15072A involved 8 normal and 17 uremic patients (stage 5) after excluding 9 CKD patients without dialysis. Three uremic patients (stages 4–5) and three healthy control samples were downloaded from GSE43484. The study samples included 31 healthy controls and 95 uremic patients.

Differentially expressed genes in uremia

With a false discovery rate (FDR) of 0.05 and by applying a minimal fold change of 1.2, a total of 840 genes were found to have altered expressions in the sample of uremic patients compared with the healthy controls. The volcano plot showed that 178 genes were upregulated and 662 genes were downregulated (Fig. 1). The genes in which expression levels were increased in uremic patients included PRKD2 (1.33, P = 1.13×10^-27), RUNX3 (1.32, P = 8.65×10^-20), TMEFFR2 (1.31, P = 1.09×10^-17), DLEU2 (0.70, P = 4.55×10^-10), THEMIS (0.7, P = 6.63×10^-8), and RAB8B (0.71, P = 8.46×10^-10).

Classification of differentially expressed genes

Classification by GO categories revealed 17 biological processes, 12 molecular functions, and 20 cellular compartments. A total of 13 categories were identified through DAVID gene functional classification, and 40 pathways were determined from KEGG. The toll-like receptor (TLR) signaling pathway was significantly correlated with the healthy controls compared with the uremic patients (enrichment score of 0.525, P-value of 0.024; Supplemental Fig. S1). A total of 23 genes were differentially expressed, and TLR4, TLR3, and MAP3K7 were among the top ranked in the core set enriched by GSEA.

Identification of candidate miRNAs associated with uremia

In GSE15072, 41 miRNAs were expected to be downregulated, and 2820 miRNAs were expected to be upregulated. Meanwhile, in GSE37171, 2366 miRNAs were expected to be downregulated, and 370 miRNAs were expected to be upregulated. No miRNA was expected in GSE43484. In total, 261 miRNAs were expected to be downregulated, and 713 miRNAs were expected to be upregulated (Supplemental Fig. S2).

Effect of uremic toxins on TLR4 mRNA levels in Saos-2 cells

To examine the effect of uremic toxins on TLR4 expression, TLR4 mRNA was analyzed through quantitative RT-PCR in Saos-2 cells in the presence of uremic toxins. The uremic toxins significantly reduced TLR4 expression until 24 h had passed (Fig. 2A).

Kinetics of pri-miR-146b and miR-146b induced by uremic toxins

We monitored the kinetics of pri-miR-146b and miR-146b treated with uremic toxins over a 24-h period. Pri-miR-146b was induced, peaked at approximately fourfold induction at the 4-h time point, and decreased subsequently (Fig. 2B). However, miR-146b increased more slowly, showing a 12-fold induction when miR-146b peaked at the 8-h period, and decreased (Fig. 2C).

Reduced osteogenic differentiation by uremic toxins

Bone differentiation markers were monitored in the presence of uremic toxins. The uremic toxins reduced SPP1 expression levels during the 24-h treatment with uremic toxins in comparison with the control medium (Fig. 3). In addition, ALP activity, a typical marker of osteogenic differentiation, decreased when Saos-2 cells were treated with uremic toxins over a 24-h period (Fig. 4). The ALP activity of the control and uremic groups decreased over a 24-h period, and ALP activity was lower in the uremic toxin-treated group compared with that in the control at each detected time point (P<0.05).

Discussion

We investigated the differential expression of genes and miRNA in the blood of healthy controls and uremic patients to understand the molecular mechanisms involved in the etiology of the disease and develop biomarkers or identify therapeutic targets by using publically available GEO datasets. Although the limitation of this study is that the clinical information of each patient was lacking, approximately 800 genes and 900 miRNA expression levels were differentially expressed in healthy controls compared with that in uremic patients.

A recent report has revealed that RUNX3 is associated with osteoporosis [25] and cancer-associated bone destruction [26]. DLEU2 is significantly correlated with total hip bone mineral density [27]. In addition, miR-454 is downregulated to suppress cell proliferation and invasion by directly targeting c-Met in osteosarcomas [28]. Analysis of differentially expressed genes has also shown that the TLR signaling pathway is associated with osteogenesis [29].

In our study, a set of downregulated genes with upregulated miRNA was selected for further in vitro experiments. Among those selected, a negative correlation was observed between TLR4 and miR-146b expressions. TLR4 expression constitutively decreases in CKD patients with reduced expression [30]. In addition, the peripheral blood mononuclear cell expression of TLR4 is downregulated in uremic patients compared with healthy controls [31]. The TLR4/MyD88/NF-кB pathway is related to osteogenic differentiation through crosstalk with BMP/Smad signaling in mesenchymal stem cells [29]. TLR4 regulates the proliferation of bone marrow mesenchymal stromal cells and osteogenic differentiation through Wnt3a and Wnt5a signaling [32], and activation of the TLR4 receptor plays a role during osteogenic differentiation of adipose-derived mesenchymal stromal cells [33].

Members of the miR-146 family, such as miR-146a and miR-146b, are negative regulators of inflammatory gene expression in numerous cell types, including monocytes [34], fibroblasts [35], endothelial [36], airway smooth muscle [37], and epithelial cells [38]. miR-146b directly suppresses TLR4 expression and signaling [8,39]. Therefore, we selected osteoblast-like osteosarcoma cell lines, Saos-2 cells, that express TLR4 [40] and measured the expression changes in TLR4, pri-miR-146b, and miR-146b in these cells after treatment with uremic toxins. Uremic toxins strongly induced pri-miR-146b and miR-146b, whereas uremic toxins reduced TLR4 expression. In colon cancer cells, the calcium-sensing receptor (CASR) mRNA is known to be the target of miR-146b [41], and CASR is involved in the pathogenesis of CKD–MBD [42]. Additionally, miR-146b has been studied for its role in regulating osteogenic differentiation of PDGFRa (+) cells [43]. Uremia is associated with a state of immune dysfunction characterized by immunodepression [44]. The immunosuppressive status due to uremia might affect downregulated TLR4 expression in uremic patients and in uremic toxin-treated Saos-2 cells. We confirmed that uremic toxins reduce SPP1 expression and ALP activity as osteogenic differentiation markers.

We presented putative miRNA–gene expression interactions related to CKD–MBD by performing a computational analysis and confirmed the changes in genes and miRNA expression by using an in vitro cell culture system. The findings of this study may lead to future identification of functional miRNAs relevant to understanding the mechanism of disease occurrence. Targeting the expression levels of these miRNAs will lead to new therapeutic approaches for renal bone disease. Our results provide insights into the search for non-invasive biomarkers that may be useful in identifying high-risk CKD–MBD patients. Therefore, further research is necessary to clarify the associated mechanisms between miR-146b and signal pathways related to osteogenesis. The regulation of miR-146b in CKD patients could be useful in the management of this complication of CKD, and this biomarker could be useful for risk evaluation and have a prognostic value in terms of treatment outcome.

References

[1] Meyer TW, Hostetter TH. Uremia. N Engl J Med 2007; 357(13): 1316–1325

[2] Duranton F, Cohen G, De Smet R, Rodriguez M, Jankowski J, Vanholder R, Argiles A; European Uremic Toxin Work Group. Normal and pathologic concentrations of uremic toxins. J Am Soc Nephrol 2012; 23(7): 1258–1270

[3] Cibulka R, Racek J. Metabolic disorders in patients with chronic kidney failure. Physiol Res 2007; 56(6): 697–705

[4] Lanza D, Perna AF, Oliva A, Vanholder R, Pletinck A, Guastafierro S, Di Nunzio A, Vigorito C, Capasso G, Jankowski V, Jankowski J, Ingrosso D. Impact of the uremic milieu on the osteogenic potential of mesenchymal stem cells. PLoS One 2015; 10(1): e0116468

[5] Meijers BK, Claes K, Bammens B, de Loor H, Viaene L, Verbeke K, Kuypers D, Vanrenterghem Y, Evenepoel P. p-Cresol and cardiovascular risk in mild-to-moderate kidney disease. Clin J Am Soc Nephrol 2010; 5(7): 1182–1189

[6] Moe S, Drüeke T, Cunningham J, Goodman W, Martin K, Olgaard K, Ott S, Sprague S, Lameire N, Eknoyan G; Kidney Disease: Improving Global Outcomes (KDIGO). Definition, evaluation, and classification of renal osteodystrophy: a position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int 2006; 69(11): 1945–1953

[7] Menon V, Gul A, Sarnak MJ. Cardiovascular risk factors in chronic kidney disease. Kidney Int 2005; 68(4): 1413–1418

[8] Hruska K, Mathew S, Lund R, Fang Y, Sugatani T. Cardiovascular risk factors in chronic kidney disease: does phosphate qualify? Kidney Int 2011; 79(S121): S9–S13 PMID: 26746860 

[9] Bartel DP. MicroRNAs: genomics, biogenesis, mechanism, and function. Cell 2004; 116(2): 281–297

[10] Alvarez-Garcia I, Miska EA. MicroRNA functions in animal development and human disease. Development 2005; 132(21): 4653–4662

[11] O’Connell RM, Rao DS, Chaudhuri AA, Baltimore D. Physiological and pathological roles for microRNAs in the immune system. Nat Rev Immunol 2010; 10(2): 111–122

[12] Tili E, Michaille JJ, Croce CM. MicroRNAs play a central role in molecular dysfunctions linking inflammation with cancer. Immunol Rev 2013; 253(1): 167–184

[13] Nana-Sinkam SP, Croce CM. MicroRNAs as therapeutic targets in cancer. Transl Res 2011; 157(4): 216–225

[14] Schöler N, Langer C, Döhner H, Buske C, Kuchenbauer F. Serum microRNAs as a novel class of biomarkers: a comprehensive review of the literature. Exp Hematol 2010; 38(12): 1126–1130

[15] Isakova T, Gutiérrez OM, Patel NM, Andress DL, Wolf M, Levin A. Vitamin D deficiency, inflammation, and albuminuria in chronic kidney disease: complex interactions. J Ren Nutr 2011; 21(4): 295–302

[16] Fang Y, Ginsberg C, Seifert M, Agapova O, Sugatani T, Register TC, Freedman BI, Monier-Faugere MC, Malluche H, Hruska KA. CKD-induced wingless/integration1 inhibitors and phosphorus cause the CKD-mineral and bone disorder. J Am Soc Nephrol 2014; 25(8): 1760–1773

[17] Neal CS, Michael MZ, Pimlott LK, Yong TY, Li JY, Gleadle JM. Circulating microRNA expression is reduced in chronic kidney disease. Nephrol Dial Transplant 2011; 26(11): 3794–3802

[18] Beltrami C, Clayton A, Phillips AO, Fraser DJ, Bowen T. Analysis of urinary microRNAs in chronic kidney disease. Biochem Soc Trans 2012; 40(4): 875–879

[19] Feichtinger J, McFarlane RJ, Larcombe LD. CancerMA: a web-based tool for automatic meta-analysis of public cancer microarray data. Database (Oxford) 2012; 2012: bas055

[20] Ramasamy A, Mondry A, Holmes CC, Altman DG. Key issues in conducting a meta-analysis of gene expression microarray datasets. PLoS Med 2008; 5(9): e184

[21] Gentleman RC, Carey VJ, Bates DM, Bolstad B, Dettling M, Dudoit S, Ellis B, Gautier L, Ge Y, Gentry J, Hornik K, Hothorn T, Huber W, Iacus S, Irizarry R, Leisch F, Li C, Maechler M, Rossini AJ, Sawitzki G, Smith C, Smyth G, Tierney L, Yang JY, Zhang J. Bioconductor: open software development for computational biology and bioinformatics. Genome Biol 2004; 5(10): R80

[22] McCall MN, Bolstad BM, Irizarry RA. Frozen robust multiarray analysis (fRMA). Biostatistics 2010; 11(2): 242–253

[23] Lee Y, Yang X, Huang Y, Fan H, Zhang Q, Wu Y, Li J, Hasina R, Cheng C, Lingen MW, Gerstein MB, Weichselbaum RR, Xing HR, Lussier YA. Network modeling identifies molecular functions targeted by miR-204 to suppress head and neck tumor metastasis. PLOS Comput Biol 2010; 6(4): e1000730

[24] Scheid S, Spang R. twilight; a Bioconductor package for estimating the local false discovery rate. Bioinformatics 2005; 21(12): 2921–2922

[25] Bauer O, Sharir A, Kimura A, Hantisteanu S, Takeda S, Groner Y. Loss of osteoblast Runx3 produces severe congenital osteopenia. Mol Cell Biol 2015; 35(7): 1097–1109

[26] Kim HJ, Park J, Lee SK, Kim KR, Park KK, Chung WY. Loss of RUNX3 expression promotes cancer-associated bone destruction by regulating CCL5, CCL19 and CXCL11 in non-small cell lung cancer. J Pathol 2015; 237(4): 520–531

[27] Reppe S, Refvem H, Gautvik VT, Olstad OK, Høvring PI, Reinholt FP, Holden M, Frigessi A, Jemtland R, Gautvik KM. Eight genes are highly associated with BMD variation in postmenopausal Caucasian women. Bone 2010; 46(3): 604–612

[28] Niu G, Li B, Sun J, Sun L. miR-454 is down-regulated in osteosarcomas and suppresses cell proliferation and invasion by directly targeting c-Met. Cell Prolif 2015; 48(3): 348–355

[29] Huang RL, Yuan Y, Zou GM, Liu G, Tu J, Li Q. LPS-stimulated inflammatory environment inhibits BMP-2-induced osteoblastic differentiation through crosstalk between TLR4/MyD88/NF-kB and BMP/Smad signaling. Stem Cells Dev 2014; 23(3): 277–289160;

[30] Ando M, Shibuya A, Tsuchiya K, Akiba T, Nitta K. Reduced capacity of mononuclear cells to synthesize cytokines against an inflammatory stimulus in uremic patients. Nephron Clin Pract 2006; 104(3): c113–c119

[31] Wang ZS, Xu DM, Guan GJ, Cui MY, Wei Y, Tang LJ, Jia XY, Li WB. Clinical significance of toll-like receptor 4 expression on the surface of peripheral blood mononuclear cells in uremic patients. Natl Med J China (Zhonghua Yi Xue Za Zhi) 2010; 90(34): 2389–2391 (in Chinese)

[32] He X, Wang H, Jin T, Xu Y, Mei L, Yang J. TLR4 activation promotes bone marrow MSC proliferation and osteogenic differentiation via Wnt3a and Wnt5a signaling. PLoS One 2016; 11(3): e0149876

[33] Herzmann N, Salamon A, Fiedler T, Peters K. Lipopolysaccharide induces proliferation and osteogenic differentiation of adipose-derived mesenchymal stromal cells in vitro via TLR4 activation. Exp Cell Res 2017; 350(1): 115–122

[34] Taganov KD, Boldin MP, Chang KJ, Baltimore D. NF-κB-dependent induction of microRNA miR-146, an inhibitor targeted to signaling proteins of innate immune responses. Proc Natl Acad Sci USA 2006; 103(33): 12481–12486

[35] Sato T, Liu X, Nelson A, Nakanishi M, Kanaji N, Wang X, Kim M, Li Y, Sun J, Michalski J, Patil A, Basma H, Holz O, Magnussen H, Rennard SI. Reduced miR-146a increases prostaglandin E₂ in chronic obstructive pulmonary disease fibroblasts. Am J Respir Crit Care Med 2010; 182(8): 1020–1029

[36] Cheng HS, Sivachandran N, Lau A, Boudreau E, Zhao JL, Baltimore D, Delgado-Olguin P, Cybulsky MI, Fish JE. MicroRNA-146 represses endothelial activation by inhibiting pro-inflammatory pathways. EMBO Mol Med 2013; 5(7): 1017–1034

[37] Larner-Svensson HM, Williams AE, Tsitsiou E, Perry MM, Jiang X, Chung KF, Lindsay MA. Pharmacological studies of the mechanism and function of interleukin-1β-induced miRNA-146a expression in primary human airway smooth muscle. Respir Res 2010; 11(1): 68

[38] Perry MM, Moschos SA, Williams AE, Shepherd NJ, Larner-Svensson HM, Lindsay MA. Rapid changes in microRNA-146a expression negatively regulate the IL-1β-induced inflammatory response in human lung alveolar epithelial cells. J Immunol 2008; 180(8): 5689–5698

[39] Curtale G, Mirolo M, Renzi TA, Rossato M, Bazzoni F, Locati M. Negative regulation of Toll-like receptor 4 signaling by IL-10-dependent microRNA-146b. Proc Natl Acad Sci USA 2013; 110(28): 11499–11504

[40] Asai Y, Hirokawa Y, Niwa K, Ogawa T. Osteoclast differentiation by human osteoblastic cell line SaOS-2 primed with bacterial lipid A. FEMS Immunol Med Microbiol 2003; 38(1): 71–79

[41] Fetahu IS, Tennakoon S, Lines KE, Gröschel C, Aggarwal A, Mesteri I, Baumgartner-Parzer S, Mader RM, Thakker RV, Kállay E. miR-135b- and miR-146b-dependent silencing of calcium-sensing receptor expression in colorectal tumors. Int J Cancer 2016; 138(1): 137–145

[42] Bover J, Aguilar A, Baas J, Reyes J, Lloret MJ, Farré N, Olaya M, Canal C, Marco H, Andrés E, Trinidad P, Ballarin J. Calcimimetics in the chronic kidney disease-mineral and bone disorder. Int J Artif Organs 2009; 32(2): 108–121

[43] Oishi T, Uezumi A, Kanaji A, Yamamoto N, Yamaguchi A, Yamada H, Tsuchida K. Osteogenic differentiation capacity of human skeletal muscle-derived progenitor cells. PLoS One 2013; 8(2): e56641

[44] Kato S, Chmielewski M, Honda H, Pecoits-Filho R, Matsuo S, Yuzawa Y, Tranaeus A, Stenvinkel P, Lindholm B. Aspects of immune dysfunction in end-stage renal disease. Clin J Am Soc Nephrol 2008; 3(5): 1526–1533

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag Berlin Heidelberg