Synovial Fluid-Derived Extracellular Vesicles of Patients with Arthritides Contribute to Hippocampal Synaptic Dysfunctions and Increase with Mood Disorders Severity in Humans

Abstract

Arthritides are a highly heterogeneous group of disorders that include two major clinical entities, localized joint disorders such as osteoarthritis (OA) and systemic autoimmune-driven diseases such as rheumatoid arthritis (RA). Arthritides are characterized by chronic debilitating musculoskeletal conditions and systemic chronic inflammation. Poor mental health is also one of the most common comorbidities of arthritides. Depressive symptoms which are most prevalent, negatively impact patient global assessment diminishing the probability of achieving the target of clinical remission. Here, we investigated new insights into mechanisms that link different joint disorders to poor mental health, and to this issue, we explored the action of the synovial fluid-derived extracellular vesicles (EVs) on neuronal function. Our data show that the exposure of neurons to different concentrations of EVs derived from both RA and OA synovial fluids (RA-EVs and OA-EVs) leads to increased excitatory synaptic transmission but acts on specific modifications on excitatory or inhibitory synapses, as evidenced by electrophysiological and confocal experiments carried out in hippocampal cultures. The treatment of neurons with EVs membrane is also responsible for generating similar effects to those found with intact EVs suggesting that changes in neuronal ability arise upon EVs membrane molecules' interactions with neurons. In humans with arthritides, we found that nearly half of patients (37.5%) showed clinically significant psychiatric symptoms (CGIs score ≥ 3), and at least mild anxiety (HAM-A ≥ 7) or depression (MADRS and HAM-D ≥ 7); interestingly, these individuals revealed an increased concentration of synovial EVs. In conclusion, our data showing opposite changes at the excitatory and inhibitory levels in neurons treated with OA- and RA-EVs, lay the scientific basis for personalized medicine in OA and RA patients, and identify EVs as new potential actionable biomarkers in patients with OA/RA with poor mental health.

Keywords: EVs; neurons; osteoarthritis; rheumatoid arthritis; synaptic transmission.

Figure 1

Characterization of synovial fluid-derived extracellular vesicles. (A) Representative NanoSight graph: dimension (nm) on x-axis, particles concentration (particles/mL) on y-axis; EVs derived from OA- and RA-synovial fluid are characterized by the same dimension, but a higher RA-particles concentration is obtained from the extraction. (B) No difference was observed in the dimension of EVs extracted both from OA- and RA-synovial fluid. (C) RA-EVs are characterized by a higher concentration in comparison to OA-EVs. Mann–Whitney test, ** p < 0.01. (D) Western Blot analysis revealed the presence of Alix in OA- and RA-EVs pellet. (E) FACs analysis revealed that OA- and RA-EVs display similar cellular origin; a significant enrichment of CD25-positive EVs was found in RA-derived vesicles, confirming the higher inflammatory cellular component present in synovial fluid collected from RA patients. Mann–Whitney test, * p < 0.05. (F) EVs originated from distinct cellular populations (CD14⁺, CD25⁺, CD62^+, and CD177⁺ cells) and expressed as % of total EVs collected from RA or OA synovial fluids. Two-Way ANOVA (Mixed Model) followed by Tukey’s multiple comparisons test, * p < 0.05.

Figure 2

EVs collected from RA and OA synovial fluids lead to increased glutamatergic function but through opposite synaptic changes: electrophysiological results. (A) Representative traces of mEPSCs and mIPSCs recorded in 14 DIV hippocampal neurons obtained from wt mouse embryos. (B) Analysis of mEPSCs frequency recorded after two hours exposure to OA- and RA-EVs at low (0.75 μg/mL) or medium/high (2.4 and 4 μg/mL) concentration from cultured neurons (low concentration: Kruskal–Wallis test followed by Dunn’s multiple comparison test, ns; number of recorded neurons: PBS = 21, OA = 18, RA = 23; medium/high concentration: Kruskal–Wallis test followed by Dunn’s multiple comparison test, ns; number of recorded neurons: PBS = 20, OA = 32, RA = 41). (C) Analysis of mEPSCs amplitude recorded after two hours exposure to OA- and RA-EVs at low (0.75 μg/mL) or medium/high (2.4 and 4 μg/mL) concentration from cultured neurons (low concentration: Ordinary one-way ANOVA followed by Tukey’s multiple comparison test, ns; number of recorded neurons: PBS = 21, OA = 18, RA = 23; medium/high concentration: Kruskal–Wallis test followed by Dunn’s multiple comparison test, ** p < 0.01; number of recorded neurons: PBS = 20, OA = 33, RA = 41). (D) Analysis of mIPSCs frequency recorded after two hours exposure to OA- and RA-EVs at low (0.75 μg/mL) or medium/high (2.4 and 4 μg/mL) concentration from cultured neurons (low concentration: Ordinary one-way ANOVA followed by Tukey’s multiple comparison test, ** p < 0.01; number of recorded neurons: PBS = 16, OA = 20, RA = 23; medium/high concentration: Kruskal–Wallis test followed by Dunn’s multiple comparison test, ns; number of recorded neurons: PBS = 17, OA = 39, RA = 40). (E) Analysis of mIPSCs amplitude recorded after two hours exposure to OA- and RA-EVs at low (0.75 μg/mL) or medium/high (2.4 and 4 μg/mL) concentration from cultured neurons (low concentration: Ordinary one-way ANOVA followed by Tukey’s multiple comparison test, ns; number of recorded neurons: PBS = 16, OA = 20, RA = 23; medium/high concentration: Kruskal–Wallis test followed by Dunn’s multiple comparison test, ns; number of recorded neurons: PBS = 17, OA = 39, RA = 40).

Figure 3

EVs collected from RA and OA synovial fluids lead to increased excitatory transmission but through opposite synaptic changes: confocal analysis. (A) Immunocytochemical experiments were performed in wt 14 DIV cultured hippocampal neurons against vGLUT1 (blue) and BIII tubulins (red) after two hours of exposure to OA- and RA-EVs. (B,C) Analysis of vGLUT1 density shows no differences among PBS and OA- / RA-EVs treated neurons, neither at low nor at medium/high concentration (low concentration: Kruskal–Wallis test followed by Dunn’s multiple comparison test, ns; field analysed for each condition: PBS = 37; OA = 35; RA = 40; medium/high concentration: Ordinary one-way ANOVA followed by Tukey’s multiple comparison test, ns; field analysed for each condition: PBS = 37; OA = 69; RA = 71). (D) Immunocytochemical experiments performed in wt 14 DIV cultured hippocampal neurons against vGAT (green) and BIII tubulins (red) after two hours of exposure to OA- and RA-EVs. (E,F) vGAT density was found increased after exposure to RA-EVs at low concentration; whereas no differences were revealed at medium/high concentration (low concentration: Ordinary one-way ANOVA followed by Tukey’s multiple comparison test, * p < 0.05; field analysed for each condition: PBS = 39; OA = 35; RA = 40; medium/high concentration: Kruskal–Wallis test followed by Dunn’s multiple comparison test, ns; field analyzed for each condition: PBS = 39; OA = 71; RA = 74).

Figure 4

EVs membranes contribute to synaptic alterations. (A) Representative traces of mEPSCs recorded in 14 DIV hippocampal neurons obtain after two hours of exposure to OA-EVs content and membrane at high concentration. (B) Representative traces of mIPSCs recorded in 14 DIV hippocampal neurons obtain after two hours of exposure to RA-EVs content and membrane at low concentration. (C) Analysis of mEPSCs frequency and amplitude recorded after two hours of exposure to OA-EVs content and membrane at high concentration from cultured neurons. No differences in terms of frequency were detected, whereas OA-EVs membranes increase mEPSCs amplitude (frequency: Kruskal–Wallis test followed by Dunn’s multiple comparison test, ns; number of recorded neurons: PBS = 29, OA content = 32, OA membrane = 19; amplitude: Kruskal–Wallis test followed by Dunn’s multiple comparison test, * p < 0.05; number of recorded neurons: PBS = 29, OA content = 32, OA membrane = 18). (D) Analysis of mIPSCs frequency and amplitude recorded after two hours of exposure to RA-EVs content and membrane at low concentration from cultured neurons. No differences in terms of amplitude were detected, whereas RA-EVs membranes decrease mIPSCs frequency (frequency: Kruskal–Wallis test followed by Dunn’s multiple comparison test, * p < 0.05; number of recorded neurons: PBS = 19, RA content = 20, RA membrane = 14; amplitude: Ordinary one-way ANOVA followed by Holm–Sidak’s multiple comparison test, ns; number of recorded neurons: PBS = 21, RA content = 28, RA membrane = 22). (E) Representative traces of mEPSCs recorded in 14 DIV hippocampal neurons obtain after two hours of exposure to OA-EVs membrane (high concentration) and OA-EVs membrane (high concentration) + ETN. (F) Representative traces of mIPSCs recorded in 14 DIV hippocampal neurons obtain after two hours of exposure to RA-EVs membrane (low concentration) and RA-EVs membrane (low concentration) + ETN. (G) Analysis of mEPSCS frequency and amplitude recorded after two hours of exposure to OA-EVs membrane (high concentration) + ETN from cultured neurons. No differences in terms of frequency were detected, whereas ETN treatment reduces mEPSCs amplitude (frequency: Mann–Whitney test, ns; number of recorded neurons: OA membrane = 20, OA membrane + ETN = 26; amplitude: Mann-Whitney test, * p < 0.05; number of recorded neurons: OA membrane = 16, OA membrane + ETN = 21). (H) Analysis of mIPSCs frequency and amplitude recorded after two hours of exposure to RA-EVs membrane (low concentration) + ETN from cultured neurons. No differences in terms of both frequency and amplitude were detected upon ETN treatment (frequency: Mann-Whitney test, ns; number of recorded neurons: RA membrane = 20, RA membrane + ETN = 20; amplitude: Mann-Whitney test, ns; number of recorded neurons: RA membrane = 20, RA membrane + ETN = 20).

Figure 5

Mood disorder severity and EVs levels in human synovial fluids. (A) Table summarizing age, gender, and psychiatric scales (MADRS, HAM-D, HAM-A, CGI, BFI-Neuroticism, and AQ-10) for each RA and OA patient. 37.5% of the total patients report at least mild psychiatric symptoms, mild anxiety, and depressive symptoms. AQ-10: 10-item Autism Spectrum Quotient; BFI: 10-item Big Five Inventory, Neuroticism is calculated as the sum of item 4 and 9; CGIs: Clinical Global Impression-Severity of Illness; HAM-A: Hamilton Anxiety Rating Scale; HAM-D: Hamilton Depression Rating Scale; MADRS: Montgomery Asberg Depression Rating Scale; OA: osteoarthritis; RA: rheumatoid arthritis; SpA: spondyloarthritis. (B) EVs concentration is higher in patients presenting mild and moderate depressive symptoms (MADRS > 6; Unpaired t-test, * p < 0.05).