Osteoarthritis Is A Chronic Musculoskeletal Disorder Biology Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

Osteoarthritis is a chronic musculoskeletal disorder characterized mainly by the progressive degradation of the articular cartilage and aging. Diagnosis of osteoarthritis is usually made at an advanced stage of cartilage destruction. Efforts are on-going to identify specific markers that might help in early diagnosis, monitoring disease progression and therapeutic intervention. To facilitate the identification of such markers, an in-depth proteomic analysis of osteoarthritis synovial fluid, the proximal joint fluid, is highly desirable. To explore the synovial fluid proteome of osteoarthritis, we employed multiple fractionation strategies followed by high resolution mass spectrometry analysis using LTQ-Orbitrap Velos. In addition, lectin affinity chromatography was also carried out to enrich the glycoproteins in osteoarthritis synovial fluid . In total, 677 proteins were identified and 550 of them were novel to this study. This is the largest repertoire of human osteoarthritis synovial fluid proteins reported till date.


Osteoarthritis (OA) is a degenerative joint disorder characterized by the damage to the articular cartilage, formation of osteophytes, increased subchondral plate thickness, development of subchondral bone cysts, inflammation of synovial membrane and neovascularization [1]. It is one of the leading causes for disability among the aging population [2]. Despite the high prevalence of OA, its mechanism of pathogenesis still remains unclear [3]. The diagnostic tools currently used have their own limitations and provide an insensitive assessment of disease progression [4, 5]. The drugs currently used for the treatment of OA are targeted mainly to reduce the pain and do not possess any disease modifying activity [6].

Advances in proteomic technologies have facilitated extensive proteomic characterization of body fluids [7-13]. Analyzing the proteome of body fluids will provide cues about the molecules that might serve as potential diagnostic, prognostic or therapeutic markers for various diseases. Though easily accessible, profiling the plasma and serum proteome suffer from limitations due to their high complexity and wide dynamic range of protein concentration [14]. Studying the synovial fluid (SF) proteome in OA is more beneficial than serum or plasma considering the fact that the concentration of the cartilage degradation products is higher in SF [15]. Additionally, proteins from other joint tissues including synovium, ligament, meniscus, joint capsule and bone might also be present in SF, thus reflecting the severity of the disease and its progression.

In one of the earliest proteomic studies of OA SF using 2-DE, Yamagiwa et al. have demonstrated a high level of inter-individual variations among SF samples[16]. A total of 135 SF proteins were identified in a study involving patients with early, advanced stages of OA and control individuals. This study showed that the protein composition of OA SF was not largely altered irrespective of the disease duration [3]. A method of endogenous profiling of peptides from OA SF that resulted in the identification of 40 proteins was described by Kamphorst et al. [17]. In a recent proteomic study, abnormally high levels of complement components were shown to be activated in OA SF indicating their significant role in OA pathogenesis [18]. The role of plasma proteins in OA SF was also investigated in a recent study in which 108 SF proteins were identified [19]. The above mentioned studies were done using low resolution mass spectrometers and did not involve extensive fractionation of the SF samples. A summary of the major proteomic studies published on OA SF is provided in Table. 1. In the present study, we have carried out an in-depth qualitative analysis of OA SF proteome by using multiple fractionation methods followed by high resolution mass spectrometry analysis. We identified 677 proteins of which 550 were not reported earlier in OA SF.

2. Materials and methods

2.1 Sample collection and processing

SF was collected from five OA patients from the affected knees after obtaining patient consent and institutional human ethics committee approval. The details of patients in the study are provided in Supplementary Table.1. Around 4-5 ml of SF was collected from each patient in heparin containing vacutainers (Becton, Dickinson and Company, New Jersey, USA). The SF was then centrifuged at 3,500 rpm for 15 minutes to remove the cells. The supernatant was then filtered using 0.45 micron filters (Millipore, USA) and then stored at -80ËšC until further processing.

2.2 Depletion of SF

A total of 12 mg of protein was pooled from five OA patients and depleted using Human 6-Multiple Affinity Removal LC Column (MARS-6) (Agilent Technologies, Santa Clara, USA) as per the manufacturer's instructions. For each round of depletion, 1 mg protein was loaded onto the column and 12 such depletion runs were carried out. The elution of proteins was monitored at 278 nm. The depleted SF samples from each round were pooled and their protein concentration was estimated by Lowry's method. Around 1 mg of protein was divided into three aliquots to be subsequently fractionated by in-gel, strong cation exchange (SCX) and OFFGEL.

2.3 In-gel digestion

300 µg of pooled depleted SF protein was resolved on a 10% SDS-PAGE (16X18 cm). The gel was then stained using CCB. Twenty eight gel bands were excised and destained using 40 mM ammonium bicarbonate in 50% acetonitrile (ACN). In-gel digestion was carried out as described previously [20]. Reduction was carried out using 5 mM DTT (60ËšC for 45 minutes) followed by alkylation using 20 mM iodoacetamide (room temperature for 10 min in dark). Trypsin digestion was done using trypsin (Sequencing grade, Promega, Madison, WI, US) at 37°C for 12-16 hrs. Peptides were extracted from the gel pieces sequentially using 0.4% formic acid in 3% ACN twice, once using 0.4% formic acid in 50% ACN and once using 100% ACN. The extracted peptides were dried and stored at -80ËšC until LC-MS/MS analysis.

2.4 In-solution digestion

500µg of depleted SF protein was reconstituted in 40mM ammonium bicarbonate. It was then reduced (5 mM DTT), alkylated (20 mM iodoacetamide) and digested using trypsin as mentioned above.

2.5 Strong cation exchange chromatography (SCX)

The digested peptides were acidified using 1 M phosphoric acid and equilibrated with 10 mM potassium phosphate buffer containing 25% acetonitrile, pH 2.85 (solvent A). 200µg equivalent of the digested peptides were fractionated using SCX on a Polysulfoethyl A column (PolyLC, Columbia, MD) (300 Å, 5 µm, 100 - 2.1 mm) using an 1100 HPLC system (Agilent Technologies, Santa Clara, USA) containing a binary pump, UV detector and a fraction collector. The peptides were eluted using a linear salt gradient (0 to 100%) between solvent A and solvent B (10 mM potassium phosphate buffer containing 25% acetonitrile, 350 mM KCl, pH 2.85). The fractions were completely dried and reconstituted in 0.1% trifluoroacetic acid to be further desalted using stage-tips [21]. Desalted fractions were dried on speedvac and reconstituted in 10 µl of 0.1% TFA prior to reversed-phase (RP) liquid chromatography based tandem mass spectrometry (LC-MS/MS) analysis.

2.6 OFFGEL fractionation

300 µg of in-solution digested depleted SF protein was used for OFFGEL fractionation. 3100 OFFGEL fractionator (Agilent Technologies, Santa Clara, USA) was used for iso-electric point based separation of peptides. As per the protocol, peptides were separated using pH 3-10 IPG strip. The peptides were focused for 50kVh with maximum current of 50µA and maximum voltage set to 4000V. 12 fractions were collected after fractionation and then acidified using 1% TFA prior to sample cleaning using stage-tips [21]

2.7 Lectin affinity enrichment

10 mg equivalent of the total protein pooled from 5 OA samples was diluted in 10 mM phosphate buffer, pH 7.8. For glycoprotein enrichment, the samples were incubated with a mixture of three agarose conjugated lectins- concanavalin A (Con A), wheat germ agglutinin and jacalin (Vector labs, USA) for 12 h at 4°C. The beads were then washed three times using wash buffer (10 mM phosphate buffer, pH 7.8) and the bound proteins were eluted using a mixture of carbohydrates (100 mM each of N-acetylglucosamine, melibiose and galactose). Using wash buffer, the eluate was dialyzed to remove free sugars and then concentrated using 3 kDa cut-off filters. The protein concentration was estimated by Lowry's method. Around 1 mg of the protein was found to be enriched using lectin affinity chromatography. 250 μg each of the concentrated protein was then resolved by 10% SDS-PAGE (16X18 cm). Twenty-six gel bands were excised and subjected to in-gel trypsin digestion procedure as previously described [20]. 250 μg of the enriched glycoprotein was also subjected to SCX fractionation as mentioned above.

2.8 LC-MS/MS analysis

Tandem mass spectrometric analysis of 112 fractions obtained from depleted total proteome and lectin enrichment was carried out using LTQ-Orbitrap Velos mass spectrometer (Thermo Electron, Bremen, Germany) interfaced with Agilent 1100 (Agilent technologies, Santa Clara, CA, USA) nano liquid chromatography system. The RP system consisted of a desalting column (3 cm - 75 u, C18 material 5 u pore size, 100 Å bead size) and an analytical column (10 cm - 75 mm, C18 material C18 material 5 u pore size, 100 Å bead size) packed in-house. Electrospray ionization source is fitted with an emitter tip 8 µm (New Objective, Woburn, MA) and maintained at 2000 V ion spray voltage. Peptide samples were loaded onto a desalting column in 0.1% formic acid, 5% ACN for 15 min and peptide separation carried out using a linear gradient of 7-35% solvent B (90% ACN in 0.1% formic acid) for 60 minutes at a constant flow rate of 350 nl/min. Data was acquired using Xcalibur 2.1 (Thermo Electron, Bremen, Germany). The MS spectra were acquired in a data-dependent manner in the scan range of 350 to 1800 m/z and survey scans were acquired in Orbitrap mass analyzer at a mass resolution of 60,000 at 400 m/z. The MS/MS was acquired at a resolution of 15,000 by targeting up to 20 most abundant ions and fragmented using normalized 39% higher energy collision mode and detected in Orbitrap mass analyzer. The once selected ions were excluded from MS/MS for 60 sec. Automatic gate control target was kept as 0.5X106 ions for full MS and 1X105 for MS/MS and the filling times are allowed 100 ms and 200 ms respectively. Lock mass option was enabled for real time calibration using polycyclodimethylsiloxane ions.

2.9 Data analysis

Mass spectrometry data was analyzed using multiple search engines to maximize the peptide identification. Proteome Discoverer 1.3 (Thermo Fisher Scientific, Bremen, Germany) was used to carry out the peak list generation and database searches. Precursor mass range of 500 to 8,000 Da and signal to noise ratio of 1.5 were used as the criteria for generation of peak list files. NCBI RefSeq 49 human protein database which includes already known contaminants (32,967 entries) was used as a reference database. Sequest and Mascot algorithms were used to carry out database searches. The parameters used for database searches include trypsin as a protease with allowed one missed cleavage, carbamidomethyl cysteine as a fixed modification, oxidation of methionine as dynamic modification. MS error window of 20 ppm and MS/MS error window of 0.1 Da were allowed. The raw data obtained were searched against decoy database to calculate 1% false discovery rate (FDR) cut-off [22]. Spectra that matched to the contaminants and those that did not pass the 1% FDR threshold were not considered for analysis.

2.10 Bioinformatics analysis

Gene Ontology (GO) [23] analysis was done to identify the biological processes associated with the identified proteins. Sub-cellular localization, post-translational modifications (PTMs), domains and motif information of the identified proteins was obtained from Human protein reference database (HPRD) (http://www.hprd.org) which is a GO compliant database [24, 25]. Plasma proteome database (PPD) (http://www.plasmaproteomedatabase.org/) [26] was used to compare the list of SF proteins identified in the study with that of plasma and/or serum.

2.10 Data availability

The raw data obtained in this study are submitted to public data repositories such as Human Proteinpedia (https://www.humanproteinpedia.org) and Tranche (https://www.proteomecommons.org/tranche/). Processed data and the database search results can be downloaded from Human Proteinpedia using HuPA_00698 code [27]. The following hash can be used to download the raw data from Tranche repository:jQquXSNp5ly3M7vOj66hnmxADXDp2DPU7BSyWzal5KdJPGKIxe6YFp2vVMPVDOaYCOD1DShgS4XN5gb87B4c/r9sE+sAAAAAAAA2CA==

3. Results and Discussion

3.1. Identification of proteins from OA SF

SF from 5 OA patients was pooled and depleted using MARS-6 column to remove the 6 most abundant proteins. The depleted protein was then subjected to multiple fractionation methods: SDS-PAGE, SCX and OFFGEL to reduce the complexity of the sample. In addition, lectin enrichment strategy was also employed to enrich the glycoprotein using a mixture of three different lectins: wheat germ agglutinin, Con A and jacalin. These lectins have different binding specificities and thereby ensure a broader coverage of glycoproteins. Subsequently the lectin enriched fractions were subjected to SDS-PAGE analysis and SCX fractionation. A work flow illustrating the steps involved in the proteomic analysis of OA SF is provided in Figure 1. All the fractions were analyzed on high resolution LTQ-Orbitrap Velos mass spectrometer.

In total, 677 proteins were identified from both depleted and lectin enriched fractions. Approximately 80% of the identified proteins have not been previously reported in OA SF. Some of the SF proteins reported in earlier studies were not identified in the present study due to the fact that different experimental methodologies and mass spectrometers were used as mentioned in Table. 1. The number of proteins identified from the depleted and lectin enriched fractions using the different fractionation methods have been summarized in Supplementary Figure 1A and 1B, respectively. Among the 301 lectin enriched proteins identified, 172 proteins were already known to be glycosylated from the data available in HPRD [24, 25]. 287 proteins were identified from single peptide hits and 30% of them were supported by multiple peptide spectral matches. The complete list of all the proteins and peptides identified in the study are provided in Supplementary Tables 2 and 3 (Supporting Information), respectively. MS/MS spectra of all the single peptide hits are provided in Supplementary Figure 2 (Supporting Information)

3.2. Classification based on Gene Ontology annotation

GO was used to categorize the proteins based on their sub-cellular localization, molecular function and biological processes. Signal peptide analysis of the identified proteins was done by using the domain information available in HPRD [24, 25] Out of 677 proteins, 482 proteins were found to have a signal peptide sequence. Classification-based on the sub-cellular localization (Figure 2A), indicated that 40% of proteins were extracellular. Several proteins were also localized to cytoplasm, plasma membrane and nucleus. Based on molecular function (Figure 2B), proteins were classified as constituents of the extracellular matrix and those involved in transporter activity, cell adhesion molecule activity, protease inhibitor activity and complement activity. Biological processes-based (Figure 2C) categorization showed that a majority of them played a role in cell communication and signaling, cell growth and/or maintenance, protein metabolism and immune response.

3.3. Proteins previously reported in OA SF

Several SF proteins reported earlier in OA SF were identified in our study confirming the validity of the experimental approach employed by us. Collagen proteins provide the required strength and stiffness to the cartilage [28]. Several type I, III, V, and VI (COL1A1, COL1A2, COL3A1, COL5A1, COL5A2 COL6A1 and COL6A3) collagens, aggrecan (ACAN), cartilage oligomeric protein (COMP), cartilage intermediate layer protein, matrix Gla protein, extracellular matrix protein 1, lumican and vitronectin identified in this study were already reported in OA SF [3, 17]. ACAN is the major proteoglycan and it confers the load bearing properties to the cartilage [29]. The levels of COMP and ACAN were found to be significantly elevated in the serum and SF of OA patients [30, 31] demonstrating its significance in OA pathogenesis. Xie et al. have shown an increased expression of fibronectin 1 (FN1) in the articular cartilage and SF of OA patients [32]. The presence of several serine protease inhibitors (SERPINs), SERPINA1, SERPINA3, SERPINA6, SERPINC1, SERPINF1, SERPING1 that regulated the proteases involved in the degradation of ECM were also confirmed in our study [3, 19]. The primary lubricating macromolecule in SF, proteoglycan 4 (PRG4) was found to be higher in the SF samples of patients in the advanced stage of OA [33]. Various complement components (C2, C3, C4A, C4B, C5, C7 and C9) that have been shown to contribute to the inflammation in OA joints were also identified in this study [18].

3.4 Proteins not reported in OA SF

Out of 677 proteins identified, 550 were not reported earlier in OA SF. A partial list of novel proteins is provided in Table 2. A few of the novel molecules identified are described below based on their relevant functions. Representative MS/MS spectra of peptides identified from Nidogen 2, Kallikrein B, Osteomodulin and Osteoglycin are provided in Figure 3.

3.4.1. Extracellular matrix proteins

Degradation of the articular cartilage is a hallmark for OA. Damage to the cartilage caused irreversible changes in the extracellular matrix (ECM) that resulted in joint dysfunction [34]. We identified two ECM proteins, fibulin1 (FBLN1) and hemicentin 1 (HMCN1) that have not been reported in OA SF previously. Fibulin 1 has been shown to bound to the elastin and microfibril-associated proteins, fibrillins 1 and 2 present in the ECM [35]. The expression of fibulin-1 was increased in breast cancer [36] and gets accumulated in the arterial wall and plasma of patients with type 2 diabetes and was associated with the changes in the arterial extracellular matrix [37]. Hemicentin 1, is a recently added member of the fibulin family and its deficiency resulted in defective cell-cell and cell-matrix interactions [38]. Polymorphisms in HMCN1 have been shown to be associated with age-related macular degeneration and renal pathophysiology [39]. Nidogen 2 (NID2) is a basement membrane protein that interacted with collagen type I, IV, laminin-1 and perlecan present in the ECM [40]. In hepatocellular carcinoma (HCC) tissues and serum, the expression of NID2 was decreased [41] whereas in the serum of ovarian cancer patients, it is elevated [42]. Kreugel et al., have shown that NID2 expression was increased in late-stage OA cartilage when compared to healthy cartilage in humans and established its regenerative ability [43].

3.4.2 Proteolytic enzymes

Degradation of the ECM in OA synovial joint has been shown to be primarily catalyzed by the proteolytic enzymes [44]. Alterations in the activities and levels of these enzymes and their associated inhibitors have been shown to disturb the balance between anabolism and catabolism in the affected joints [44]. Kallikrein B (KLKB1) is a serine protease that cleaves kininogen to release the biologically active bradykinin and this reaction was shown to be inhibited by C1 inhibitor [45]. Polymorphisms in KLKB1 have been shown to be associated with hypertension [46], renal failure [47] and breast cancer [48]. ADAM-like, decysin 1 (ADAMDEC1) is a recently identified member of the disintegrin metalloproteinase family [49]. ADAMDEC1 overexpression was associated with atherosclerotic plaque instability [50] and pulmonary sarcoidosis [51]. Its expression was decreased during colorectal tumor progression [52]. Xu et al., have shown its expression was inhibited by tamoxifen in craniopharyngioma cells [53].

3.4.3 Cell adhesion molecules

Cell-cell and cell-matrix interactions are mediated by cell adhesion molecules. These interactions are critical for the regulation of a plethora of biological processes including synovial inflammation and tissue remodelling [54]. CD84, also known as SLAMF5 or Ly9b, is a member of the signaling lymphocyte activation molecule (SLAM) family [55]. Signaling pathways triggered by CD84 have been shown to be essential for platelet aggregation [56]. It negatively regulated Fc epsilon RI-mediated signaling in human mast cells [57, 58]. Osteomodulin is a keratan sulfate proteoglycan that promotes cell binding mediated by integrin alphaV beta3 in bone [59]. It is also known as osteoadherin and belongs to the small leucine rich protein family (SLRPs). OMD was detected in bovine mature osteoblasts and human odontoblasts suggesting its role in bone mineralization [60]. OMD expression in odontoblasts was also stimulated by TGF-beta 1. In vitro overexpression of OMD increased the differentiation and maturation of mature osteoblasts [61]. It may play a role as a osteoblast maturation marker [62]. It was identified in human trabecular meshwork [63] and as a mechanosensitive gene in osteoblasts [64]. Microarray analysis has revealed the association of OMD in osteoblast differentiation mediated by bone morphogenetic protein -2 [65]

3.4.4. Growth factors and Cytokines

Growth factors and cytokines are regulatory molecules that play a significant role in joint destruction and disease pathogenesis. Their levels are altered in case of joint injury or disease [44]. Osteoglycin also known as mimecan or osteoinductive factor belongs to the family of small leucine rich proteoglycans (SLRPs). They are associated with the commitment to osteogenic phenotype that leads to osteoblast differentiation [44]. Mice deficient in OGN showed an increase in bone density [66] and thicker corneal collagen fibrils, thus implying its role in collagen fibrillogenesis [67]. It was differentially expressed in the trabecular meshwork of eyes with open-angle glaucoma [68]. The expression of osteoglycin was increased in the irradiated synovial membrane of rheumatoid arthritis patients [69]. FAM3C is a cytokine widely expressed in human tissues [70]. It is essential for epithelial to mesenchymal transition (EMT), tumor formation and metastasis in epithelial cells of multiple human tumors [71, 72] as well as in retinal laminar formation [73]. Its expression was altered in pancreatic cancer secretome [74] idiopathic temporal lobe epilepsy [75] and in Fuchs endothelial corneal dystrophy [76]. Single nucleotide polymorphisms in FAM3C have been shown to be associated with bone mineral density [77].

3.5 Glycoproteins in OA SF

Glycosylation of proteins is a biologically significant and complex post-translational modification associated with membrane and secreted proteins [78]. Body fluids are rich in glycoproteins [78]. Characterizing the glycoproteome increases the dynamic range of protein profiling in an otherwise complex sample [78]. We identified several glycoproteins in OA SF by lectin affinity enrichment. The list of all the proteins identified by lectin enrichment has been provided in Supplementary Table. 4 (Supporting Information). A few of the glycoproteins identified are described below. Lectin, galactoside-binding, soluble, 3 binding protein (LGALS3BP) is a glycoprotein highly associated with cancer and infection by human immunodeficiency virus (HIV) [79]. It is upregulated in the secretome of esophageal squamous cell carcinoma (ESCC) [80] colorectal cancer [81] and in hepatitis C related live cirrhosis [82]. CELSR1 is a neural specific gene that encodes for seven pass transmembrane protein [83]. It was found to be one of the frequently amplified gene in ductal carcinoma in situ of breast [84] and mutations in this gene were associated with craniorachischisis (CRN), a severe neural tube defect [85]. Matrix associated remodelling protein (MXRA5) was identified only from lectin enriched fraction. It was identified in colon cancer [86] and was shown to be frequently mutated in non-small cell lung carcinoma [87].

3.6 Plasma proteins in OA SF

SF is a partial plasma dialysate and plasma proteins normally entered the SF by passive diffusion [88]. They might also enter the synovial compartment at the sites of tissue damage by compromising the endothelial barrier [89]. The level of plasma proteins in OA SF were shown to be higher than the healthy SF [19]. Studies have also shown that the plasma proteins in OA SF activated macrophages through the toll-like receptor 4 to produce inflammatory cytokines that contributed to inflammation in OA synovial joints [19]. In order to survey the plasma proteins in the OA SF, the proteins identified in the current study were compared with the PPD [26]. We found that 606 proteins in OA SF were already known to be present in plasma and/or serum. A list of all the OA SF proteins that are present in PPD is provided in Supplementary Table. 5.

4. Concluding remarks

Using high resolution mass spectrometry we identified 677 proteins in this study, the largest catalog of proteins reported till date in OA SF. We hope that this study will further enhance our knowledge on OA pathophysiology and open new research avenues to identify potential diagnostic or therapeutic markers for OA.


We thank the Department of Biotechnology, Government of India for research support to the Institute of Bioinformatics, Bangalore. We thank Agilent Technologies for the instrument support. Rajesh Raju is a recipient of Senior Research fellowship award from Council of Scientific and Industrial Research (CSIR), Government of India. Santosh Renuse and Dhanashree S. Kelkar are recipients of Senior Research fellowship award from the University Grants Commission (UGC), Government of India. Harsha Gowda is a Wellcome Trust/DBT India Alliance Early Career Fellow. Thottethodi Subrahmanya Keshava Prasad is a recipient of Young Investigator award from the Department of Biotechnology, India.

Conflict of interest

The authors have declared no conflicts of interest.