Identification of Lipid Biomarkers for Chronic Joint Pain Associated with Different Joint Diseases

Abstract

Lipids, especially lysophosphatidylcholine LPC16:0, have been shown to be involved in chronic joint pain through the activation of acid-sensing ion channels (ASIC3). The aim of the present study was to investigate the lipid contents of the synovial fluids from controls and patients suffering from chronic joint pain in order to identify characteristic lipid signatures associated with specific joint diseases. For this purpose, lipids were extracted from the synovial fluids and analyzed by mass spectrometry. Lipidomic analyses identified certain choline-containing lipid classes and molecular species as biomarkers of chronic joint pain, regardless of the pathology, with significantly higher levels detected in the patient samples. Moreover, correlations were observed between certain lipid levels and the type of joint pathologies. Interestingly, LPC16:0 levels appeared to correlate with the metabolic status of patients while other choline-containing lipids were more specifically associated with the inflammatory state. Overall, these data point at selective lipid species in synovial fluid as being strong predictors of specific joint pathologies which could help in the selection of the most adapted treatment.

Trial registration: ClinicalTrials.gov NCT01867840.

Keywords: chronic joint pain; inflammation; lipidomics; metabolic status; synovial fluid.

Figure 1

Relative abundances of analyzed lipids in human synovial fluids (HSF) of a cohort of patients suffering from joint pain associated with different rheumatic diseases. Analysis of lipid species in HSF samples was carried out by a direct infusion into mass spectrometry (MS) using an electrospray ionization source (ESI) in positive ion mode. (a) Relative abundances of lipid classes (PhosphatidylCholine PC, LysoPhosphatidylCholine LPC and plasmalogen PC PCp). (b) Relative abundances of PC species. (c) Relative abundances of LPC species showing a high level of LPC 16:0. (d) Relative abundances of the different PCp species.

Figure 2

Comparison between lipid levels in HSF samples of the cohort of patients suffering from joint pain associated with different rheumatic diseases (Patients) and post-mortem controls (Controls). Lipids were extracted from HSF samples and the levels of PC, LPC and PCp species were measured based on ESI-MS analysis after the addition of lipid internal standards. The comparison between the total amount of each lipid class between the patients and the controls was based on an unpaired t-test, while the two-way ANOVA test was performed to compare multiple lipid species within each lipid class. (a,d) Histogram representing the comparison of the total PC concentration and individual PC species concentrations (µM) in HSF between patients and controls. (b,e) Histogram representing the comparison of total LPC concentration and individual LPC species concentrations between patients and controls. (c,f) Histogram representing the comparison of total PCp concentration and of individual PCp species concentrations between patients and controls. ns: not significant p > 0.05, *: 0.01 < p < 0.05, **: 0.001 < p < 0.01 and ***: p < 0.001.

Figure 3

Correlation of lipid levels in HSF with the body mass index (BMI) and age of patients. (a), Table of Spearman correlation between lipid levels in HSF and BMI. (b,c) Correlation plots of BMI with LPC16:0 and LPC18:0, respectively. (d) Table of Spearman correlation between lipid levels in HSF and the age of patients regardless of joint pathologies. (e,f) Correlation plots of age with LPC16:0 and LPC18:0, respectively. ns: not significant p > 0.05, *: 0.01 < p < 0.05, and **: 0.001 < p < 0.01.

Figure 4

Evaluation of lipid concentrations in the synovial fluid of patients according to the rheumatic diseases associated with chronic joint pain. (a–d) Concentrations (in µM) of PC, PCp, LPC16:0, and LPC18:0, respectively, in the synovial fluid of patients depending on joint pathologies: (RA, n = 6), (Gout, n = 5), (CCA, n = 12), (PA, n = 4), (SPA, n = 5), (OA, n = 18) and controls n = 10. The comparison between lipid concentrations in the individual joint pathologies was based on the one-way analysis of variance completed with Bonferroni’s Multiple Comparisons Test. ns: not significant p > 0.05, *: 0.01 < p < 0.05, **: 0.001 < p < 0.01 and ***: p < 0.001.

Figure 5

Lipid concentrations in the synovial fluid of patients according to the inflammatory state of the patients. Circulating C-Reactive Protein (CRP mg/l) was considered as an indicator of systemic inflammation. (a) Table of Spearman correlation results of circulating CRP with LPC16:0, LPC18:0, PC and PCp levels. (b–d), Correlation plots of CRP level with LPC16:0, LPC18:0, and PC levels, respectively. (e) Concentration (in mg/L) of circulating CRP according to the different joint diseases showing higher CRP levels in microcrystalline arthropathies patients. One-way analysis of variance completed with the Bonferroni’s Multiple Comparisons Test were performed to compare CRP concentrations in the individual joint pathologies; ns: not significant p > 0.05, *: 0.01 < p < 0.05, **: 0.001 < p < 0.01.

Figure 6

Evaluation of phospholipase A2 (PLA2) role in the variation of LPC16:0 levels in the synovial fluid of patients. (a) Represents the structure of LPC16:0. (b,c) Represents the concentration (in µM, left panel) and the structure (right panel) of PC(16:0/20:4) and FA20:4, respectively, in the different rheumatic diseases compared to controls. A one-way analysis of variance completed with the Bonferroni’s Multiple Comparisons Test were performed to compare lipid concentrations in the individual joint pathologies; ns: not significant p > 0.05, **: 0.001 < p < 0.01.

Figure 7

Assessment of potential associations between the number of cellular elements in human synovial fluids and the variation of lipid levels. (a) Table of Spearman correlation between the number of cellular elements per mm³ of synovial fluids and lipid concentrations (in µM) in the synovial fluids. (b–g) Correlation plots of the number of cellular elements with LPC16:0, LPC18:0, FA20:4, PC, PCp and C-Reactive Protein levels, respectively. ns: not significant p > 0.05, **: 0.001 < p < 0.01 and ***: p < 0.001.