Abstract
Background
Osteoarthritis (OA) is a common joint degenerative disease that can affect multiple joints. Genetic events may play an important regulatory role in the early stages of the disease, but the specific mechanisms have not yet been fully elucidated. The main purpose of this study was to screen for disease-causing hub genes and effective small molecule drugs to reveal the pathogenesis of OA and to develop novel drugs for treatment.
Methods
Two gene expression profile datasets, GSE55235 and GSE55457, were integrated and further analyzed. The consistently differentially expressed genes (DEGs) were identified, and functional annotation and pathway analysis of these genes were performed with GO and KEGG. A protein–protein interaction network (PPI) of the DEGs was generated using STRING, and potential small molecule drug screening was performed on the connectivity map (CMap).
Results
A total of 158 consistently differentially expressed genes were identified from the two profile datasets. The functions of these DEGs are mainly related to the TNF signaling pathway, osteoclast differentiation, MAPK signaling pathway and so on. The PPI network contains 127 nodes and 1802 edges, and the ten hub genes were interleukin 6 (IL6), vascular endothelial growth factor A (VEGFA)and so on. 7 small molecule drugs were identified as potential interactors with these hubs.
Conclusions
This study explains the disorder of expression in the pathological process of OA at transcriptome, which will help to understand the pathogenesis of OA.
Keywords: Osteoarthritis (OA), Differentially expressed genes (DEGs), Small molecule drugs, Connectivity map (CMap)
Highlights
-
•
This study identified 158 DEGs in in OA synovial tissue compared with normal synovial tissue, including 71 upregulated DEGs and 87 downregulated DEGs.
-
•
The 158 DEGs were found enriched in multiple signaling pathways including osteoclast differentiation, NF-kappa B signaling and MAPK signaling.
-
•
Ten hub genes were identified from 127 DEGs constructing a PPI network, which might play an important function in the pathological process of OA.
-
•
Four drugs were identified as newly potential treatments for osteoarthritis based on CMap analysis of DEGs.
1. Background
Osteoarthritis (OA) is a very common orthopedic disease, with an incidence of 10–20% in people over 60 years of age [1]. The main treatments for OA are early pain control and advanced joint replacement surgery. However, these treatments often fail to effectively improve patient prognosis [2,3]. The pathogenesis of OA is very complex and involves many factors, including genetic, biological, and biomechanical factors [4]. However, the whole pathological process of OA is accompanied by the disorder of gene expression. It is regrettable that the disorder of genomic expression in the pathological process of OA has not yet been clarified. Therefore, this study aims to elucidate the pathogenesis at the genetic level by systematic bioinformatics methods. Specifically, we identify disease-causing hub genes and signaling pathways as early diagnostic and therapeutic targets and novel small molecule drugs to provide new solutions for their treatment.
In recent decades, high-throughput technologies to quickly detect all genes expressed in the same tissue sample have been widely used, and the resulting datasets can be analyzed to identify the expression of DEGs [5]. With the widespread use of genetic testing technology, massive amounts of gene chip data are generated and further stored in public databases such as the GEO database and the TCGA database. Microarray expression profile data of many diseases have been analyzed to obtain numerous DEGs for the diagnosis and treatment of diseases such as colorectal cancer, cervical cancer, and ovarian epithelial cancer [[6], [7], [8]]. However, there is currently no comprehensive bioinformatics analysis of OA. Therefore, no reliable biomarkers have been identified for early diagnosis and therapeutic markers of this disease.
To date, many drugs, such as chondroitin, glucosamine, doxycycline, bisphosphonates, and anakinra, have been shown to have their own advantages and disadvantages and to play a certain role in the treatment of OA, but none of them can hinder the pathological process of the disease [4]. Therefore, the development of new therapeutic drugs is urgently needed. Connectivity Map (CMap) is a public data platform for disease-based gene expression profiling data to obtain corresponding targeted therapeutics [9].
Therefore, in order to explore the key expression disorder genes in the pathological process of OA, the study first retrieved two data sets, GSE55235 and GSE55457, in the GEO database. Subsequently, we screened 158 DEGs from synovial tissue of osteoarthritis patients and normal synovial tissue. Subsequently, GO functional annotation analysis, KEGG cell signal pathway analysis, PPI network construction, and CMap analysis of small molecule drugs were performed. Finally, we obtained several hub genes as well as potential therapeutic small molecule drugs for OA.
2. Materials and methods
2.1. Data collection
The Gene Expression Omnibus (GEO) database is a public data repository containing a collection of gene expression data associated with many diseases that can be downloaded freely by researchers [10]. In this study, in order to obtain the microarray data of osteoarthritis (OA) from multiple data sets, it can make the analysis results more accurate and more scientific. we searched the entire GEO database about osteoarthritis (OA) datasets. The results showed that only GSE55235 and GSE55457 datasets had the same organization source, platform number, and no drug or other treatment. First of all, we log in to the official GEO website (http://www.ncbi.nlm.nih.gov/geo/) and input two keywords: osteoarthritis and synovial tissue to search. Second, after reading the search results, we found that only two data sets, GSE55235 and GSE55457, met our inclusion criteria. Finally, we download the GSE55235 (10 synovial tissue samples from patients with osteoarthritis and 10 normal synovial tissue samples) and GSE55457 (10 synovial tissue samples from patients with osteoarthritis and 10 normal synovial tissue samples) series matrix files and the corresponding GPL96 platform [HG-U133A] Affymetrix Human Genome U133A Array one by one. Finally, the detailed information of the sample has been included in Table S1.
2.2. Identification of DEGs
After downloading the series matrix file data, the probe IDs were converted to gene symbols based on the GPL96 platform using Perl scripts [11]. We used the limma package of Bioconductor in R to standardize the gene expression profile data and then took the log2 values. Empirical Bayesian methods were applied to obtain differentially expressed genes (DEGs) from the synovial tissues of osteoarthritis patients and normal synovial tissues according to the cutoff criteria of P value < 0.05 and logFoldChange >1 [12]. The heat map was produced in the package pheatmap in R to show the distribution of the DEGs [13]. The online tool Venny (http://bioinfogp.cnb.csic.es/tools/venny/) was applied to perform integrated analysis of the GSE55457 and GSE55235 data sets [14].
2.3. Gene Ontology (GO) analysis of DEGs
GO is considered to be a comprehensive database of computable information resources of genes and their products, and the annotations are mainly divided into three functional components: molecular function, cellular component and biological process [15]. DAVID is an online tool widely used in bioinformatics analysis, with advantages such as multiple functions and convenient operation (https://david.ncifcrf.gov/) [16]. The DEGs were uploaded to the official DAVID website to obtain the GO functional annotations with the cutoff criterion of P value < 0.05.
2.4. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways of DEGs
KEGG is well regarded as an international database of gene functions, and the enrichment of cell signaling pathways is one of its important analytical functions [17]. KOBAS is a versatile online bioinformatics analysis tool that provides information on gene annotation, cell pathway analysis, and disease associations [18]. The list of gene symbols for DEGs was first converted to a list of Ensembl Gene IDs on the DAVID website and then uploaded to KOBAS 3.0 (http://kobas.cbi.pku.edu.cn/) for KEGG pathway analysis. With the cutoff criterion of P value < 0.05, the results of meaningful cell signaling pathways were screened and then visualized using Cytoscape.
2.5. Protein‐protein interaction (PPI) network construction and analysis of DEGs
Protein-protein interaction (PPI) describes the process in which two or more protein molecules form a protein complex through noncovalent bonds. STRING is a public database that searches for interactions between known proteins and predicted proteins [19]. We uploaded DEGs to the official website of the STRING database to obtain their relationship with a minimum required interaction score>0.4 and then used Cytoscape software to represent these relationships graphically. Based on the number of connections between DEGs, the hub genes were determined.
2.6. Identification of potential small molecule drugs for the treatment of OA
Connectivity map (CMap) is a technology and data resource for studying drug mechanisms and drug relocation based on RNA microarrays [20]. We divided the DEGs obtained from osteoarthritis into upregulated genes and downregulated genes and then converted them to probesets, which were loaded into the CMap official website to query potential small molecule drugs. In this way, we identified small molecule drugs that might be useful for the treatment of osteoarthritis, according to the threshold value of P < 0.01 and enrichment <0 as a likely negatively correlated drug.
3. Results
3.1. DEG identification
According to the threshold value of P < 0.05 and logFoldChange >1, 889 DEGs were identified in the screening of the gene expression profile data of GSE55235, including 472 upregulated DEGs and 417 downregulated DEGs, and 506 DEGs were identified in the screening of the gene expression profile data of GSE55457, including 209 upregulated DEGs and 297 downregulated DEGs. Using the pheatmap package in R, a heat map of DEG distribution was drawn in which 71 upregulated DEGs and 87 downregulated DEGs were identified between GSE55235 and GSE55457 (Fig. 1A and B). Through integrated analysis, 158 DEGs were consistently expressed in the 2 groups of gene expression profiles (Fig. 2A and B), including 71 upregulated DEGs and 87 downregulated DEGs in OA synovial tissue compared with normal synovial tissue (Table 1).
Fig. 1.
The distribution heat map of 158 consistent DEGs in OA (red: upregulated; green: downregulated); A, GSE55235 data set; B, GSE55457 data set. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
Fig. 2.
Integrated analysis shows 158 consistent DEGs in 2 groups of gene expression profiles; (A), representing 71 consistently upregulated DEGs; (B), representing 87 consistently downregulated DEGs.
Table 1.
Consistent expression of 158 DEGs, including 71 upregulated genes and 87 downregulated genes, was identified in the GSE55235 data set and GSE55457 data set according to the cutoff criteria of P value < 0.05 and logFoldChange >1.
| DEGs | Gene Name |
|---|---|
| up-regulated 71 | RRAS,GPR88,LINC01140,GUCA1A,NUDT1,PDZRN4,STMN2,CBR3,HLA-DQB1,GSTZ1,MYOM2,PTHLH,B3GALNT1,FDXR, PNMAL1,NDUFA4L2,ZIC1,TMEM106C,CLIC3,ST8SIA1,HSD11B2,PTN,TDO2,NAP1L3,THBS4,NUDT11,DPYS, TRIL,FANCF,SIL1,FGGY, CAPG,RTN1,PART1,WIF1,RTP4,RPE65,ERAP2,LIPC,RGS13,DPT,MSTN, NELL1,ZBED8,TAC1,C1QTNF3,SLC5A3,HTR2B,IGK,ANKH,KAL1,GPR1,LRRC17,SLC18A2,MTUS2,MGAT4C,TNFSF11,XIST, EPYC,IGKC,TLR7,LRRC15,IGHM,WNT5B,TREM2,SCRG1,OGN,CX3CR1,IGLV1-44,LTC4S,IGLC1. |
| down-regulated 87 | FOSB, MAFF,NFIL3,FKBP5,CXCL2,CXCL3,APOLD1,SIK1,GADD45B,PTGS2,ATF3,SLC2A3,RPS4Y1,KLF9,SLC19A2,IL6,DDX3Y,MYC,NAMPT, FOSL2,NR4A2,ZFP36,SLC7A5,CDKN1A,SLC16A7,VEGFA, FOSL1,CYR61,TNFAIP3,SNORA21,KDM5D,JUN,CRISPLD2,RND1,CCNL1,RHOB, FOXC2,HAS1,LOC100506282,SELE, MTHFD2,INHBB, TIPARP,NR4A1,LRCH1,AREG, NPAS2,AF007147,SOCS3,KLF4,ADAMTS1,DUSP5,USP9Y,STC1,JUNB, SBNO2,RP11-769O8.3,CLTC-IT1,PKD1P1,KLF6,NLGN4Y,NR4A3,DUSP1,EGR1,PHACTR1,RGS16,DUSP4,NAA15,SPRY1,LGALSL, PFKFB3,MCL1,AC007292.3,FASN,CCL25,NFKB2,BIN3-IT1,AF090939,ANKRD11,UGCG, ACACB,THBD, ARID5A,THBS1,RUFY2,SLC9A5,IL3RA. |
3.2. GO annotation analysis of DEGs
The 158 DEGs were analyzed by GO, with annotations divided into three main parts (molecular function, biological process and cellular component); the top ten annotations in each part are graphically displayed in Fig. 3. The top three most enriched biological processes were response to cAMP, positive regulation of transcription from the RNA polymerase II promoter and drug response. The top three most enriched cellular components were extracellular space, proteinaceous extracellular matrix and extracellular region. The top three most enriched molecular functions were transcriptional activator activity, RNA polymerase II core promoter proximal region sequence-specific binding, heparin binding and transcription factor activity, and RNA polymerase II core promoter proximal region sequence-specific binding. We summarize the top 10 meaningful enrichments for each part of the GO analysis in Table 2.
Fig. 3.
Gene Ontology (GO) functional annotation analysis of DEGs in OA. GO analysis was divided into three parts: cell components (CC), molecular functions (MF), and biological processes (BP).
Table 2.
Gene Ontology (GO) analysis of DEGs is divided into three parts; the top ten most meaningful molecular functions (MF) and biological processes (BP) results, and only 6 meaningful enrichments in the cell components (CC) results, are presented.
| ID | Description | P.adjust | Count | Gene Names |
|---|---|---|---|---|
| GO:0001077 | transcriptional activator activity, RNA polymerase II core promoter proximal region sequence-specific binding | 4.64E-07 | 13 | EGR1, NR4A2, NR4A1, NR4A3, FOSB, NFKB2, ZIC1, JUNB, JUN, FOXC2, MYC, FOSL1, KLF4. |
| GO:0008201 | heparin binding | 6.74E-07 | 11 | OGN, CRISPLD2, VEGFA, MSTN, PTN, ADAMTS1, THBS1, EPYC, LIPC, THBS4, CYR61. |
| GO:0000982 | transcription factor activity, RNA polymerase II core promoter proximal region sequence-specific binding | 3.17E-05 | 5 | ATF3, FOSL2, JUN, FOSB, FOSL1. |
| GO:0008083 | growth factor activity | 3.41E-04 | 8 | INHBB, OGN, IL6, VEGFA, MSTN, PTN, AREG, THBS4. |
| GO:0000978 | RNA polymerase II core promoter proximal region sequence-specific DNA binding | 0.002343239 | 10 | ATF3, FOSL2, JUN, NFKB2, NR4A3, FOSB, ZIC1, FOSL1, MYC, JUNB. |
| GO:0005125 | cytokine activity | 0.003033723 | 7 | INHBB, NAMPT, IL6, TNFSF11, VEGFA, MSTN, AREG |
| GO:0043565 | sequence-specific DNA binding | 0.003157753 | 12 | EGR1, MAFF, ATF3, FOSL2, JUN, NR4A2, FOXC2, NR4A1, NR4A3, FOSB, NFIL3, MYC. |
| GO:0017017 | MAP kinase tyrosine/serine/threonine phosphatase activity | 0.004741599 | 3 | DUSP5, DUSP4, DUSP1. |
| GO:0070402 | NADPH binding | 0.00631602 | 3 | FASN, FDXR, CBR3. |
| GO:0034987 | immunoglobulin receptor binding | 0.018453293 | 3 | IGKC, IGHM, IGLC1. |
| GO:0005615 | extracellular space | 1.42E-09 | 34 | NAMPT, WNT5B, CXCL3, CXCL2, TAC1, LRRC17, IGHM, OGN, MTHFD2, CCL25, C1QTNF3, SCRG1, PTN, IGKC, THBS1, DPT, THBS4, IL6, NUDT1, MSTN, CBR3, PTHLH, INHBB, TNFSF11, THBD, NLGN4Y, VEGFA, SIL1, STC1, AREG, LIPC, EPYC, IGLC1, SELE. |
| GO:0005578 | proteinaceous extracellular matrix | 0.004526911 | 8 | OGN, WNT5B, CRISPLD2, VEGFA, ADAMTS1, EPYC, DPT, CYR61 |
| GO:0005576 | extracellular region | 0.005179072 | 23 | IL6, WNT5B, IGLV1-44, NELL1, CXCL3, CXCL2, APOLD1, TAC1, PTHLH, INHBB, CCL25, OGN, TNFSF11, CRISPLD2, VEGFA, WIF1, THBS1, IGKC, TREM2, LIPC, IGLC1, THBS4, CYR61. |
| GO:0005788 | endoplasmic reticulum lumen | 0.015941637 | 6 | WNT5B, PTGS2, SIL1, ERAP2, THBS1, LIPC. |
| GO:0005667 | transcription factor complex | 0.016268215 | 6 | NPAS2, JUN, NAA15, NR4A1, NR4A3, JUNB. |
| GO:0005654 | nucleoplasm | 0.017494652 | 32 | NAMPT, FOSL2, MCL1, PFKFB3, FKBP5, NFKB2, ZIC1, NPAS2, SPRY1, ANKRD11, FANCF, MYC, KDM5D, EGR1, MAFF, KLF9, ARID5A, NR4A2, NR4A1, NR4A3, ZBED8, CBR3, JUNB, PTHLH, DUSP5, DUSP4, CDKN1A, ATF3, JUN, CAPG, RPS4Y1, KLF4. |
| GO:0051591 | response to cAMP | 1.89E-06 | 7 | THBD, DUSP1, JUN, AREG, FOSB, FOSL1, JUNB. |
| GO:0045944 | positive regulation of transcription from RNA polymerase II promoter | 4.26E-06 | 24 | EGR1, NAMPT, MAFF, SBNO2, IL6, FOSL2, NR4A2, CCNL1, NR4A1, NFKB2, FOSB, NR4A3, ZIC1, JUNB, NPAS2, TNFSF11, ATF3, JUN, VEGFA, FOXC2, FOSL1, MYC, KLF4, CYR61. |
| GO:0042493 | response to drug | 7.38E-06 | 13 | IL6, PTGS2, ACACB, FOSB, JUNB, CDKN1A, JUN, HSD11B2, PTN, HTR2B, THBS1, MYC, FOSL1. |
| GO:0048661 | positive regulation of smooth muscle cell proliferation | 9.21E-06 | 7 | NAMPT, IL6, PTGS2, JUN, NR4A3, THBS1, MYC. |
| GO:0045444 | fat cell differentiation | 2.87E-05 | 7 | INHBB, WNT5B, C1QTNF3, NR4A2, NR4A1, NR4A3, KLF4. |
| GO:0032496 | response to lipopolysaccharide | 6.00E-05 | 9 | THBD, PTGS2, CXCL3, JUN, CXCL2, TAC1, NFKB2, SELE, JUNB. |
| GO:0007565 | female pregnancy | 8.81E-05 | 7 | PTHLH, NAMPT, THBD, HSD11B2, FOSB, EPYC, FOSL1. |
| GO:0009612 | response to mechanical stimulus | 1.18E-04 | 6 | INHBB, JUN, FOSB, THBS1, FOSL1, JUNB. |
| GO:0006954 | inflammatory response | 2.80E-04 | 12 | CCL25, IL6, PTGS2, CXCL3, CXCL2, TAC1, NFKB2, THBS1, TRIL, TNFAIP3, TLR7, SELE. |
| GO:0007623 | circadian rhythm | 3.67E-04 | 6 | NAMPT, NPAS2, KLF9, JUN, RPE65, NFIL3. |
Abbreviation:GO,Gene Ontology.
3.3. KEGG pathway analysis of DEGs
Based on the KEGG database, we obtained 24 meaningfully enriched cell signaling pathways. Cytoscape software was used to display 56 DEGs associated with 24 meaningfully cellular signaling pathways (Fig. 4). The five most significantly enriched signaling pathways were TNF signaling, HTLV-I infection, osteoclast differentiation, NF-kappa B signaling and MAPK signaling. We have summarized 24 meaningful enrichment pathways and related DEGs in Table 3.
Fig. 4.
KEGG Signal Pathway Analysis network of DEGs in OA. In this network. Red open octagons represent upregulated DEGs; cyan open octagons represent downregulated DEGs; yellow open diamonds indicate the KEGG ID. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
Table 3.
KEGG cell signaling pathway analysis of DEGs in OA.
| ID | Description | P.adjust | Count | Gene Names |
|---|---|---|---|---|
| hsa04668 | TNF signaling pathway | 3.51E-08 | 9 | CXCL2,TNFAIP3,SELE, SOCS3,CXCL3,PTGS2,IL6,JUNB,JUN. |
| hsa05166 | HTLV-I infection | 5.42E-08 | 12 | ATF3,WNT5B,EGR1,FOSL1,CDKN1A,MYC,RRAS,HLA-DQB1,IL6,ZFP36,NFKB2,JUN. |
| hsa04380 | Osteoclast differentiation | 1.31E-07 | 9 | TNFSF11,SOCS3,FOSL1,FOSB, NFKB2,JUNB, TREM2,FOSL2,JUN. |
| hsa04064 | NF-kappa B signaling pathway | 2.25E-05 | 6 | CXCL2,TNFAIP3,PTGS2,TNFSF11,GADD45B,NFKB2. |
| hsa04010 | MAPK signaling pathway | 2.31E-05 | 9 | DUSP5,DUSP1,MYC,NFKB2,GADD45B,RRAS, DUSP4,NR4A1,JUN. |
| hsa04151 | PI3K-Akt signaling pathway | 0.000196531 | 9 | MCL1,THBS4,VEGFA, CDKN1A,MYC,THBS1,IL3RA,IL6,NR4A1. |
| hsa04630 | Jak-STAT signaling pathway | 0.000389706 | 6 | MCL1,SOCS3,CDKN1A,MYC,IL3RA,IL6. |
| hsa04310 | Wnt signaling pathway | 0.001573118 | 5 | WNT5B,MYC,FOSL1,WIF1,JUN. |
| hsa04012 | ErbB signaling pathway | 0.002081472 | 4 | CDKN1A,MYC,AREG,JUN. |
| hsa01100 | Metabolic pathways | 0.00220972 | 16 | ST8SIA1,MTHFD2,GSTZ1,NDUFA4L2,FASN, ACACB,TDO2,MGAT4C,NAMPT,B3GALNT1,CBR3,PTGS2,LIPC,LTC4S,UGCG, DPYS. |
| hsa04066 | HIF-1 signaling pathway | 0.00355596 | 4 | CDKN1A,IL6,VEGFA, PFKFB3. |
| hsa00061 | Fatty acid biosynthesis | 0.003679887 | 2 | FASN, ACACB. |
| hsa04621 | NOD-like receptor signaling pathway | 0.005138163 | 3 | CXCL2,IL6,TNFAIP3. |
| hsa00590 | Arachidonic acid metabolism | 0.006671714 | 3 | LTC4S,CBR3,PTGS2. |
| hsa04115 | p53 signaling pathway | 0.008136417 | 3 | CDKN1A,THBS1,GADD45B. |
| hsa04210 | Apoptosis | 0.009881695 | 4 | MCL1,GADD45B,IL3RA,JUN. |
| hsa04550 | Signaling pathways regulating pluripotency of stem cells | 0.010350028 | 4 | WNT5B,INHBB,KLF4,MYC. |
| hsa04350 | TGF-beta signaling pathway | 0.013594772 | 3 | INHBB, THBS1,MYC |
| hsa04062 | Chemokine signaling pathway | 0.024879166 | 4 | CXCL2,CX3CR1,CXCL3,CCL25. |
| hsa04620 | Toll-like receptor signaling pathway | 0.025788047 | 3 | IL6,TLR7,JUN. |
| hsa04510 | Focal adhesion | 0.032499863 | 4 | THBS1,VEGFA, THBS4,JUN. |
| hsa04110 | Cell cycle | 0.036314952 | 3 | CDKN1A,MYC,GADD45B. |
| hsa04152 | AMPK signaling pathway | 0.03703588 | 3 | ACACB, FASN,PFKFB3. |
| hsa04068 | FoxO signaling pathway | 0.044645559 | 3 | CDKN1A,IL6,GADD45B |
Abbreviation:KEGG, Kyoto Encyclopedia of Genes and Genomes.
3.4. PPI network construction and identification of hub genes
A total of 127 DEGs from 158 DEGs in OA were used to construct a PPI network containing 127 nodes and 1802 edges. Cytoscape software was used to visualize the network, with red nodes representing upregulated DEGs and yellow nodes representing downregulated DEGs (Fig. 5A). The determination of hub genes was based on the degree of association between DEGs and are displayed graphically in Fig. 5B. The ten DEGs with the highest number of closely related genes are considered to be hub genes: interleukin 6 (IL6), vascular endothelial growth factor A (VEGFA), Jun proto-oncogene (JUN), MYC proto-oncogene (MYC), prostaglandin-endoperoxide synthase 2 (PTGS2), dual specificity phosphatase 1(DUSP1), early growth response 1 (EGR1), activating transcription factor 3 (ATF3), cyclin dependent kinase inhibitor 1A (CDKN1A) and nuclear receptor subfamily 4 group A member 1 (NR4A1). These genes may play an important regulatory function in the pathological process of OA.
Fig. 5.
(A) The PPI network constructed from 127 DEGs; the numbers in red octagons represent the upregulated DEGs, and the numbers in yellow octagons represent the downregulated DEGs. (B) The number of connections between the DEGs is used to determine the hub genes. (For interpretation of the references to color in this figure legend, the reader is referred to the Web version of this article.)
3.5. Screening for potential small molecule compounds for OA treatment
The results of CMap analysis of the identified DEGs in OA yielded a total of 7 drugs that met the threshold values of P < 0.01 and enrichment <0 (Table 4). A literature search showed that some of these small molecule drugs, such as hecogenin, lidocaine and ambroxol, have been reported to have significant anti-inflammatory or analgesic effects. The remaining four drugs, metolazone, pipemidic acid, yohimbic acid and chlorphenesin, have not been studied in OA or as anti-inflammatory treatments. We have summarized the 3D structures of these drugs in Fig. 6.
Table 4.
The 7 small-molecule compounds identified by CMap analysis.
| Cmap name | Enrichment | P-value |
|---|---|---|
| hecogenin | −0.931 | 0.00002 |
| metolazone | −0.779 | 0.00096 |
| lidocaine | −0.766 | 0.0013 |
| ambroxol | −0.774 | 0.00527 |
| pipemidic acid | −0.846 | 0.00735 |
| yohimbic acid | −0.845 | 0.00747 |
| chlorphenesin | −0.74 | 0.00899 |
Abbreviation:CMap, Connectivity Map.
Fig. 6.
The 3D structures of 7 small-molecule drugs for OA treatment.(A) hecogenin, (B) metolazone, (C) lidocaine, (D) ambroxol, (E) pipemidic acid, (F) yohimbic acid and (G) chlorphenesin.
4. Discussion
Osteoarthritis (OA) is the most common degenerative joint disease in the world, which leads to chronic disability in the elderly [21]. So far, the pathogenesis of osteoarthritis has not been fully elucidated. It is generally believed that mechanical wear and biological factors lead to the imbalance of degradation and synthesis of soft tissue, extracellular matrix and subchondral bone [22,23]. As the molecular pathogenesis of osteoarthritis progression is still poorly understood, there is currently no effective treatment to repair damaged cartilage or slow the progression of osteoarthritis. Therefore, to clarify the underlying mechanism of joint degeneration, which can effectively delay joint degeneration in the early stage, is essential for the treatment of osteoarthritis.
In the past few decades, a massive amount of research has been conducted to reveal the causes and the underlying regulatory mechanisms of OA. To date, there have been many applications of bioinformatics to explain the pathophysiological mechanism of OA from a genetic perspective. However, the therapeutic effect has not been satisfactory, and no precise biomarkers were screened out in OA, probably because most studies have focused on individual genetic changes or expression data from a single sample, failing to correlate gene expression data with therapeutic drugs [24,25]. The innovation of this study is to use the gene expression profiling data of two osteoarthritis-related datasets to further integrate the analysis and obtain a total of 158 consistent DEGs, avoiding some of the disadvantages of using a single cohort, such as a smaller number of tissue samples and experimental errors in a single study. In particular, our approach uses differentially expressed genes to identify some small molecule drugs that may become targeted drugs for the future treatment of OA.
The results of our KEGG analysis have been verified by some previous reports. For example, the tumor necrosis factor (TNF) family is a group of cytokines that are produced by a variety of cells, and up-regulated matrix metalloproteinase (MMP) and thrombospondin motifs (ADAMTS), as a kind of pathogenic molecules, can increase the activity of TNF signaling pathway, and then destroy articular cartilage to promote the pathological progress of osteoarthritis [26,27]. Increased activity of Human T cell leukaemia virus type I (HTLV-I) infection cell signaling pathway may promote the release of inflammatory factors that may alter the pathological process of osteoarthritis, but HTLV-I infection alone may not be a causative factor of osteoarthritis [28]. Peripheral Blood Mononuclear Cells (PBMCs) are considered to be precursor cells that can be transformed into osteoclasts in vitro and play an important role in the pathological process of osteoarthritis [29]. In the rabbit osteoarthritis model, the early use of low-intensity pulsed ultrasound, an acoustic therapy, can significantly slow the degradation of cartilage, possibly by inhibiting the activation of the MAPK signaling pathway [30]. The PI3K-Akt signaling pathway is involved in the regulation of the development and progression of various diseases, and inhibition of the PI3K-Akt signaling pathway in the pathological process of osteoarthritis can significantly reduce the inflammatory response [31]; additional relevant literature reports are not repeated here. In short, the above discussion can infer that these signal pathways are involved in the pathological process of osteoarthritis.
The hub gene we screened out in the PPI network may be the core promoter gene in the pathological process of osteoarthritis. This speculation was supported by previous reports. For example, miR-139 can regulate the secretion of IL6 and thus become a driving factor to promote the destruction of cartilage in osteoarthritis. Therefore, reducing miR-139 expression may provide a benefit for the treatment of osteoarthritis [32]. VEGFA maintains the structural integrity of chondrocytes through a variety of pathways and thus can be an effective therapeutic target for osteoarthritis [33]. C-myc is an apoptosis-related protein that is closely associated with the destruction of articular cartilage and can be regarded as an independent factor in the pathogenesis of arthritis [34]. PTGS2 may be a diagnostic marker and a therapeutic target for osteoarthritis that participates in the inflammatory response pathway that promotes joint destruction [35]. DUSP1 has a negative regulatory effect on the MAPK signaling pathway and thus can be anti-inflammatory, so high expression of DUSP1 can delay the pathological progression of osteoarthritis [1]. Among the hub genes, JUN, ATF3, CDKN1A and NR4A1 are also known to be involved in the pathological process of osteoarthritis, but the role of EGR1 in this process has not been reported [[36], [37], [38], [39]]. These hub genes may play an important role in the pathological process of OA, but this study only uses biological information analysis and has no direct evidence to confirm its regulatory effect.
Finally, CMap is an online tool used for drug discovery. The working principle of the CMap analysis is a professional drug research platform for obtaining corresponding corrective drugs based on gene expression differences of target diseases. Using this tool, we obtained a total of seven small molecule drugs as novel drugs for the treatment of osteoarthritis. Due to the pathological progress of osteoarthritis is inseparable from aseptic inflammation, anti-inflammatory drugs have certain therapeutic effects on delaying the progress of osteoarthritis. Previous research reports suggested that the small molecular drugs have anti-inflammatory effects obtained by CMap analysis. For example, hecogenin, a plant steroid, not only inhibits the proliferation of fibroblast-like synoviocytes but also promotes apoptosis and can therefore be an effective treatment for the rheumatoid arthritis [40]. Hecogenin has many pharmacological effects, such as pain-relieving, antibacterial, antihypertensive and anti-inflammatory effects; it can significantly inhibit the activity of COX mRNA enzymes and the release of inflammatory factors including TNF-a, IL-6, and IL-12, and thus has an anti-inflammatory effect in arthritis [41]. Pain is one of the main symptoms of osteoarthritis, and lidocaine, a local anesthetic, can block the sodium channels of the sensory nerve endings and inhibit nerve conduction for analgesia [42]. The release of inflammatory mediators is an important part of the pathological progression of osteoarthritis, and the anti-inflammatory properties of lidocaine are achieved by inhibiting synovial EP1 receptors. Lidocaine can also stabilize cell membranes and improve cell function [43]. Ambroxol, as a expectorant, has long been used to treat lung diseases, but many studies in recent years have confirmed that it has significant anti-inflammatory and antioxidative effects in the treatment of other diseases. Ambroxol inhibits the release of various inflammatory factors by mast cells and leukocytes and various immune-related factors by basophils [44]. The analysis method of gene biological information in this study obtained the drugs discussed above, so there is not enough evidence to show that the above drugs can be used for the clinical treatment of OA only provides an index. However, more experiments need to be performed to further verify their utility in OA.
Although our study comprehensively analyzed the differential gene expression and its functional mechanism in the pathological progress of osteoarthritis, and screened out several potential small molecule compounds that antagonize osteoarthritis. But some limitations were hard to avoid. The hub genes we obtained through PPI analysis are in osteoarthritis, and traditional experimental methods should be used to further verify the mechanism of action. However, due to the disorder of many disease-treatment genes, a single article cannot contain the specific mechanism of action of all genes, so we will verify them one by one in the following research. Second, although the joint analysis of multiple data sets has improved the accuracy and scientificity of the analysis results, we only found two data sets in the GEO database that contained 20 synovial tissues of osteoarthritis and 20 cases of normal synovial tissue. However, hundreds or even thousands of sample tissues and corresponding detailed patient clinical characteristics have not been obtained. This is due to the limitations of using public database analysis, it is difficult for us to expand the number of tissue samples in a short time. Finally, the drugs have obtained from the Cmap database, which have potential for the treatment of osteoarthritis were scientific conclusions, but we believe that we provide was only an index, so that more researchers pay attention to the potential therapeutic effects of these drugs on arthritis, so as to promote the development of drugs for osteoarthritis.
5. Conclusion
In this study, gene expression data from two osteoarthritis-related datasets and the CMap database were effectively combined to identify not only 10 core genes and their involvement in regulating the pathogenic pathway of OA but also several small molecule drugs. Therefore, it is believed that this study explains the disorder of expression in the pathological process of OA at transcriptome, which will help to understand the pathogenesis of OA.
Author contributions
Xingbo Cheng：Writing - Original Draft, Visualization, Supervision. Zhendong Liu：Software, Formal analysis, Writing - Review & Editing. Hongbo Wang：Data Curation. Jiangfen Zhang：Data Curation & Software. Yanzheng Gao: Conceptualization, Funding acquisition.
Consent for publication
The publication of this worked has received the permission of all authors.
Funding
This work was supported by The National Natural Science Foundations of China (82172438), The Medical Science and technology research plan in Henan Province (Grant ID: LHGJ20210013，LHGJ20220031), National Clinical Research Center for Orthopedics，Sports Medicine ＆ Rehabilitation of China (2021-NCRC-CXJJ-ZH-05).
Declaration of competing interest
All the authors agreed to the final manuscript, and then everyone declared that there was no conflict of interest in it.
Acknowledgments
This work was supported by The National Natural Science Foundations of China (82172438), The study benefits from this funding and GEO databasis. Therefore, we thank all the staff for this funding and the databases for providing such a good platform for researchers.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.bbrep.2023.101450.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
Data availability
Data will be made available on request.
References
- 1.Peng H., Yun Z., Wang W., Ma B. Dual specificity phosphatase 1 has a protective role in osteoarthritis fibroblast-like synoviocytes via inhibition of the MAPK signaling pathway. Mol. Med. Rep. 2017;16:8441–8447. doi: 10.3892/mmr.2017.7617. [DOI] [PubMed] [Google Scholar]
- 2.Pivec R., Johnson A.J., Mears S.C., Mont M.A. Hip arthroplasty. Lancet. 2012;380:1768–1777. doi: 10.1016/S0140-6736(12)60607-2. [DOI] [PubMed] [Google Scholar]
- 3.Bijlsma J.W., Berenbaum F., Lafeber F.P. Osteoarthritis: an update with relevance for clinical practice. Lancet. 2011;377:2115–2126. doi: 10.1016/S0140-6736(11)60243-2. [DOI] [PubMed] [Google Scholar]
- 4.Glyn-Jones S., Palmer A.J., Agricola R., Price A.J., Vincent T.L., Weinans H., Carr A.J. Osteoarthritis. Lancet. 2015;386:376–387. doi: 10.1016/S0140-6736(14)60802-3. [DOI] [PubMed] [Google Scholar]
- 5.Vogelstein B., Papadopoulos N., Velculescu V.E., Zhou S., Diaz L.A., JR., Kinzler K.W. Cancer genome landscapes. Science. 2013;339:1546–1558. doi: 10.1126/science.1235122. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Guo Y., Bao Y., Ma M., Yang W. Identification of key candidate genes and pathways in colorectal cancer by integrated bioinformatical analysis. Int. J. Mol. Sci. 2017;18 doi: 10.3390/ijms18040722. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Tang X., Xu Y., Lu L., Jiao Y., Liu J., Wang L., Zhao H. Identification of key candidate genes and small molecule drugs in cervical cancer by bioinformatics strategy. Cancer Manag. Res. 2018;10:3533–3549. doi: 10.2147/CMAR.S171661. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Zhou Y., Layton O., Hong L. Identification of genes and pathways involved in ovarian epithelial cancer by bioinformatics analysis. J. Cancer. 2018;9:3016–3022. doi: 10.7150/jca.26133. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Lamb J., Crawford E.D., Peck D., Modell J.W., Blat I.C., Wrobel M.J., others The Connectivity Map: using gene-expression signatures to connect small molecules, genes, and disease. Science. 2006;313:1929–1935. doi: 10.1126/science.1132939. [DOI] [PubMed] [Google Scholar]
- 10.Clough E., Barrett T. The gene expression Omnibus database. Methods Mol. Biol. 2016;1418:93–110. doi: 10.1007/978-1-4939-3578-9_5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Macdonald J., Boutros P. Log::ProgramInfo: a Perl module to collect and log data for bioinformatics pipelines. Source Code Biol. Med. 2016;11:9. doi: 10.1186/s13029-016-0055-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Ritchie M., Phipson B., Wu D., Hu Y., Law C., Shi W., Smyth G. Limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43:e47. doi: 10.1093/nar/gkv007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Yang Q., Guo B., Sun H., Zhang J., Liu S., Hexige S., Yu X., Wang X. Identification of the key genes implicated in the transformation of OLP to OSCC using RNA-sequencing. Oncol. Rep. 2017;37:2355–2365. doi: 10.3892/or.2017.5487. [DOI] [PubMed] [Google Scholar]
- 14.Hu H., Jie H., Zheng X. vol. 40. 2016. Three circulating LncRNA predict early progress of esophageal squamous cell carcinoma; pp. 117–125. (Cellular Physiology and Biochemistry : International Journal of Experimental Cellular Physiology, Biochemistry, and Pharmacology). [DOI] [PubMed] [Google Scholar]
- 15.Expansion of the gene Ontology knowledgebase and resources. Nucleic Acids Res. 2017;45:D331–D338. doi: 10.1093/nar/gkw1108. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Huang D., Sherman B., Tan Q., Kir J., Liu D., Bryant D., Guo Y., Stephens R., Baseler M., Lane H., Lempicki R. DAVID Bioinformatics Resources: expanded annotation database and novel algorithms to better extract biology from large gene lists. Nucleic Acids Res. 2007;35:W169–W175. doi: 10.1093/nar/gkm415. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Kanehisa M., Goto S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 2000;28:27–30. doi: 10.1093/nar/28.1.27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Xie C., Mao X., Huang J., Ding Y., Wu J., Dong S., Kong L., Gao G., Li C., Wei L. KOBAS 2.0: a web server for annotation and identification of enriched pathways and diseases. Nucleic Acids Res. 2011;39:W316–W322. doi: 10.1093/nar/gkr483. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Franceschini A., Szklarczyk D., Frankild S., Kuhn M., Simonovic M., Roth A., Lin J., Minguez P., Bork P., VON Mering C., Jensen L. STRING v9.1: protein-protein interaction networks, with increased coverage and integration. Nucleic Acids Res. 2013;41:D808–D815. doi: 10.1093/nar/gks1094. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Abstracts of presentations at the association of clinical scientists 143 meeting louisville, KY May 11-14,2022. Ann. Clin. Lab. Sci. 2022;52:511–525. [PubMed] [Google Scholar]
- 21.Gleason B., Chisari E., Parvizi J. Osteoarthritis can also start in the gut: the gut-joint Axis. Indian J. Orthop. 2022;56:1150–1155. doi: 10.1007/s43465-021-00473-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kang D., Shin J., Cho Y., Kim H., Gu Y., Kim H., others Stress-activated miR-204 governs senescent phenotypes of chondrocytes to promote osteoarthritis development. Sci. Transl. Med. 2019;11 doi: 10.1126/scitranslmed.aar6659. [DOI] [PubMed] [Google Scholar]
- 23.Goldring S.R., Goldring M.B. Changes in the osteochondral unit during osteoarthritis: structure, function and cartilage-bone crosstalk. Nat. Rev. Rheumatol. 2016;12:632–644. doi: 10.1038/nrrheum.2016.148. [DOI] [PubMed] [Google Scholar]
- 24.Cai P., Jiang T., Li B., Qin X., Lu Z., LE Y., Shen C., Yang Y., Zheng L., Zhao J. Comparison of rheumatoid arthritis (RA) and osteoarthritis (OA) based on microarray profiles of human joint fibroblast-like synoviocytes. Cell Biochem. Funct. 2019;37:31–41. doi: 10.1002/cbf.3370. [DOI] [PubMed] [Google Scholar]
- 25.Lin J., Wu G., Zhao Z., Huang Y., Chen J., Fu C., Ye J., Liu X. Bioinformatics analysis to identify key genes and pathways influencing synovial inflammation in osteoarthritis. Mol. Med. Rep. 2018;18:5594–5602. doi: 10.3892/mmr.2018.9575. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Hosseinzadeh A., Kamrava S., Joghataei M., Darabi R., Shakeri-Zadeh A., Shahriari M., Reiter R., Ghaznavi H., Mehrzadi S. Apoptosis signaling pathways in osteoarthritis and possible protective role of melatonin. J. Pineal Res. 2016;61:411–425. doi: 10.1111/jpi.12362. [DOI] [PubMed] [Google Scholar]
- 27.Zhao Y., Liu B., Tian Q., Wei J., Richbourgh B., Liu C. Progranulin protects against osteoarthritis through interacting with TNF-α and β-Catenin signalling. Ann. Rheum. Dis. 2015;74:2244–2253. doi: 10.1136/annrheumdis-2014-205779. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Yoshihara Y., Tsukazaki T., Osaki M., Nakashima M., Hasui K., Shindo H. Altered expression of inflammatory cytokines in primary osteoarthritis by human T lymphotropic virus type I retrovirus infection: a cross-sectional study. Arthritis Res. Ther. 2004;6:R347–R354. doi: 10.1186/ar1193. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Durand M., Komarova S., Bhargava A., Trebec-Reynolds D., Li K., Fiorino C., others Monocytes from patients with osteoarthritis display increased osteoclastogenesis and bone resorption: the in Vitro Osteoclast Differentiation in Arthritis study. Arthritis Rheum. 2013;65:148–158. doi: 10.1002/art.37722. [DOI] [PubMed] [Google Scholar]
- 30.Li X., Li J., Cheng K., Lin Q., Wang D., Zhang H., An H., Gao M., Chen A. Effect of low-intensity pulsed ultrasound on MMP-13 and MAPKs signaling pathway in rabbit knee osteoarthritis. Cell Biochem. Biophys. 2011;61:427–434. doi: 10.1007/s12013-011-9206-4. [DOI] [PubMed] [Google Scholar]
- 31.Xue J., Shi Z., Zou J., Li X. Inhibition of PI3K/AKT/mTOR signaling pathway promotes autophagy of articular chondrocytes and attenuates inflammatory response in rats with osteoarthritis. Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie. 2017;89:1252–1261. doi: 10.1016/j.biopha.2017.01.130. [DOI] [PubMed] [Google Scholar]
- 32.Makki M., Haqqi T. miR-139 modulates MCPIP1/IL-6 expression and induces apoptosis in human OA chondrocytes. Exp. Mol. Med. 2015;47:e189. doi: 10.1038/emm.2015.66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Chen H., Tian Y. MiR-15a-5p regulates viability and matrix degradation of human osteoarthritis chondrocytes via targeting VEGFA. Bioscience trends. 2017;10:482–488. doi: 10.5582/bst.2016.01187. [DOI] [PubMed] [Google Scholar]
- 34.Yatsugi N., Tsukazaki T., Osaki M., Koji T., Yamashita S., Shindo H. Apoptosis of articular chondrocytes in rheumatoid arthritis and osteoarthritis: correlation of apoptosis with degree of cartilage destruction and expression of apoptosis-related proteins of p53 and c-myc. J. Orthop. Sci. : Off J Jpn Orthop Assoc. 2000;5:150–156. doi: 10.1007/s007760050142. [DOI] [PubMed] [Google Scholar]
- 35.Schwarzbach S., Melo C., Xavier P., Roballo K., Cordeiro Y., AmbróSIO C., Fukumasu H., Carregaro A. Morphine, but not methadone, inhibits microsomal prostaglandin E synthase-1 and prostaglandin-endoperoxide synthase 2 in lipopolysaccharide-stimulated horse synoviocytes. Biochimie. 2019;160:28–33. doi: 10.1016/j.biochi.2019.02.004. [DOI] [PubMed] [Google Scholar]
- 36.Ye Z., Chen Y., Zhang R., Dai H., Zeng C., Zeng H., Feng H., DU G., Fang H., Cai D. c-Jun N-terminal kinase - c-Jun pathway transactivates Bim to promote osteoarthritis. Can. J. Physiol. Pharmacol. 2014;92:132–139. doi: 10.1139/cjpp-2013-0228. [DOI] [PubMed] [Google Scholar]
- 37.Iezaki T., Ozaki K., Fukasawa K., Inoue M., Kitajima S., Muneta T., others ATF3 deficiency in chondrocytes alleviates osteoarthritis development. J. Pathol. 2016;239:426–437. doi: 10.1002/path.4739. [DOI] [PubMed] [Google Scholar]
- 38.Kihara S., Hayashi S., Hashimoto S., Kanzaki N., Takayama K., Matsumoto T., Chinzei N., Iwasa K., Haneda M., Takeuchi K., Nishida K., Kuroda R. vol. 32. 2017. Cyclin-dependent kinase inhibitor-1-deficient mice are susceptible to osteoarthritis associated with enhanced inflammation; pp. 991–1001. (Journal of Bone and Mineral Research : the Official Journal of the American Society for Bone and Mineral Research). [DOI] [PubMed] [Google Scholar]
- 39.Shi X., Ye H., Yao X., Gao Y. The involvement and possible mechanism of NR4A1 in chondrocyte apoptosis during osteoarthritis. Am. J. Tourism Res. 2017;9:746–754. [PMC free article] [PubMed] [Google Scholar]
- 40.Liagre B., Vergne-Salle P., Leger D., Beneytout J. Inhibition of human rheumatoid arthritis synovial cell survival by hecogenin and tigogenin is associated with increased apoptosis, p38 mitogen-activated protein kinase activity and upregulation of cyclooxygenase-2. Int. J. Mol. Med. 2007;20:451–460. [PubMed] [Google Scholar]
- 41.Ingawale D., Patel S. Hecogenin exhibits anti-arthritic activity in rats through suppression of pro-inflammatory cytokines in Complete Freund's adjuvant-induced arthritis. Immunopharmacol. Immunotoxicol. 2018;40:59–71. doi: 10.1080/08923973.2017.1405439. [DOI] [PubMed] [Google Scholar]
- 42.Ferreira-Gomes J., AdãES S., MendonçA M., Castro-Lopes J. Analgesic effects of lidocaine, morphine and diclofenac on movement-induced nociception, as assessed by the Knee-Bend and CatWalk tests in a rat model of osteoarthritis. Pharmacol. Biochem. Behav. 2012;101:617–624. doi: 10.1016/j.pbb.2012.03.003. [DOI] [PubMed] [Google Scholar]
- 43.Eker H., Cok O., Aribogan A., Arslan G. The efficacy of intra-articular lidocaine administration in chronic knee pain due to osteoarthritis: a randomized, double-blind, controlled study. Anaesthesia, critical care & pain medicine. 2017;36:109–114. doi: 10.1016/j.accpm.2016.05.003. [DOI] [PubMed] [Google Scholar]
- 44.Gibbs B., Schmutzler W., Vollrath I., Brosthardt P., Braam U., Wolff H., Zwadlo-Klarwasser G. vol. 48. 1999. Ambroxol inhibits the release of histamine, leukotrienes and cytokines from human leukocytes and mast cells; pp. 86–93. (Inflammation research : official journal of the European Histamine Research Society ... [et al.]). [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data will be made available on request.