Synovial Extracellular Vesicles: Structure and Role in Synovial Fluid Tribological Performances

Abstract

The quality of the lubricant between cartilaginous joint surfaces impacts the joint's mechanistic properties. In this study, we define the biochemical, ultrastructural, and tribological signatures of synovial fluids (SF) from patients with degenerative (osteoarthritis-OA) or inflammatory (rheumatoid arthritis-RA) joint pathologies in comparison with SF from healthy subjects. Phospholipid (PL) concentration in SF increased in pathological contexts, but the proportion PL relative to the overall lipids decreased. Subtle changes in PL chain composition were attributed to the inflammatory state. Transmission electron microscopy showed the occurrence of large multilamellar synovial extracellular vesicles (EV) filled with glycoprotein gel in healthy subjects. Synovial extracellular vesicle structure was altered in SF from OA and RA patients. RA samples systematically showed lower viscosity than healthy samples under a hydrodynamic lubricating regimen whereas OA samples showed higher viscosity. In turn, under a boundary regimen, cartilage surfaces in both pathological situations showed high wear and friction coefficients. Thus, we found a difference in the biochemical, tribological, and ultrastructural properties of synovial fluid in healthy people and patients with osteoarthritis and arthritis of the joints, and that large, multilamellar vesicles are essential for good boundary lubrication by ensuring a ball-bearing effect and limiting the destruction of lipid layers at the cartilage surface.

Keywords: lipid structures; lubricant; nanoparticles; osteoarthritis; rheumatoid arthritis; synovial fluid.

Figure 1

Lipid composition of SF from the knee joints of dogs and human healthy volunteers. The molar percentage of phospholipid classes in the total PL extract (A), acyl chain distribution in the total PL extract (B), and molar percentage of saturated, monounsaturated, and polyunsaturated acyls present in the total PL extract (C). Lipids were quantified as described in Materials and Methods. Data are presented as the mean ± SD of 5 dogs and 7 humans. p-values (t-test) less than 0.05 were considered significant (*).

Figure 2

Lipid composition of SF from the knee joints of human healthy volunteers and OA or RA patients. The molar percentage of phospholipid classes in the total PL extract (A), acyl chain distribution in the total PL extract (B), and molar percentage of saturated, monounsaturated, and polyunsaturated acyls present in the total PL extract (C). Lipids were quantified as described in Materials and Methods. Data are presented as mean ± SD of 7 healthy, 3 OA, and 3 RA humans. p-values (t-test) less than 0.05 were considered significant (*).

Figure 3

Alterations in protein and total lipid concentration in SF from the knee joints of human healthy volunteers and OA or RA patients. Protein concentration (A), PL total amount (B), and PL percentage are calculated as the ratio of PL acyl chain molar concentration over total lipid acyl chain concentration (C). Data are presented as mean ± SD of 7 healthy, 3 OA, and 3 RA humans. p-values (t-test) less than 0.05 were considered significant (*).

Figure 4

Ultrastructure of representative samples of normal and pathological SF: Transmission electron microscopy of SF healthy donors (A) or SF from patients with OA (B) or RA (C). Size distribution of vesicles present in healthy SF samples (D), OA SF samples (E), or RA SF samples (F); Multilamellar structures characteristic of healthy SF samples in a 5 times enlargement of figure A (G). Size distribution analysis: (D) Min-Max: 100–556 nm; Med ± SD 157 ± 6 nm; (E) Min- Max: 30–171 nm; Med ± SD: 45 ± 5 nm (F) Min-Max: 30–171 nm; Med ± SD: 45 ± 5 nm.

Figure 5

Rheological analysis: Lubricant viscosity as a function of the shear rate measured in a cone/plane configuration under shear stress. (A) SF from normal and pathological sources. (B) Biomimetic vesicular structures.

Figure 6

Tribological analyses using Heathy, RA, and OA SF: Fiction coefficient upon rubbing of tribometer plates in the presence (color) or absence (color) of lipid bilayers spread at the surface (lipid interface) obtained with SF of healthy, RA, or OA donors as a lubricant (A); Visualization by fluorescence microscopy of the lipid bilayer spread at the plate after friction tests (60-min friction) in the presence of healthy SF (B), OA SF (C), or RA SF (D) between the rubbing plates. NBD-PE was used as a fluorescent probe embedded in the spread lipid bilayer. The destruction of the lipid bilayers is shown by the fragmentation and accumulation of fluorescent material at the edge of the contact zone between the two rigid surfaces. Error bars represent standard deviation (m ± SD; n = 3).