The Role of Hyaluronic Acid in Cartilage Boundary Lubrication

Abstract

Hydration lubrication has emerged as a new paradigm for lubrication in aqueous and biological media, accounting especially for the extremely low friction (friction coefficients down to 0.001) of articular cartilage lubrication in joints. Among the ensemble of molecules acting in the joint, phosphatidylcholine (PC) lipids have been proposed as the key molecules forming, in a complex with other molecules including hyaluronic acid (HA), a robust layer on the outer surface of the cartilage. HA, ubiquitous in synovial joints, is not in itself a good boundary lubricant, but binds the PC lipids at the cartilage surface; these, in turn, massively reduce the friction via hydration lubrication at their exposed, highly hydrated phosphocholine headgroups. An important unresolved issue in this scenario is why the free HA molecules in the synovial fluid do not suppress the lubricity by adsorbing simultaneously to the opposing lipid layers, i.e., forming an adhesive, dissipative bridge between them, as they slide past each other during joint articulation. To address this question, we directly examined the friction between two hydrogenated soy PC (HSPC) lipid layers (in the form of liposomes) immersed in HA solution or two palmitoyl-oleoyl PC (POPC) lipid layers across HA-POPC solution using a surface force balance (SFB). The results show, clearly and surprisingly, that HA addition does not affect the outstanding lubrication provided by the PC lipid layers. A possible mechanism indicated by our data that may account for this is that multiple lipid layers form on each cartilage surface, so that the slip plane may move from the midplane between the opposing surfaces, which is bridged by the HA, to an HA-free interface within a multilayer, where hydration lubrication is freely active. Another possibility suggested by our model experiments is that lipids in synovial fluid may complex with HA, thereby inhibiting the HA molecules from adhering to the lipids on the cartilage surfaces.

Keywords: cartilage lubrication; hyaluronic acid; hydration lubrication; phosphatidylcholine lipids; polymer bridging.

Conflict of interest statement

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

Figures

Figure 1

Illustrating the proposed boundary layer on articular cartilage [33], where lubricin molecules anchor hyaluronic acid (HA) chains at the articular cartilage surface and the HA is complexed with phosphatidylcholine (PC) layers exposing their highly hydrated phosphocholine groups at the very outer surface (the slip plane). Reproduced with permission from [33]. Copyright 2016, Annual Review.

Figure 2

Size distribution of HSPC and POPC SUVs in water determined by dynamic light scattering (DLS). The hydrodynamic size and polydispersity index (PDI) for HSPC and POPC SUVs in water are 73 ± 6 nm (PDI = 0.071 ± 0.015) and 68 ± 5 nm (PDI = 0.056 ± 0.021), respectively.

Figure 3

Atomic force microscope (AFM) images of HSPC SUVs adsorbed on mica under water (a) and HA solution (b). The height distance cross section of the surface along the red line is presented below each image.

Figure 4

Schematic illustration of the surface force balance (SFB) setup, where normal and shear forces between two surfaces are directly measured via the bending of two orthogonal springs. White light undergoes multiple beam interference in passing through the half-silvered mica sheets, resulting in Fringes of Equal Chromatic Order (FECO, top right inset). These yield absolute separation between the surfaces to within ±0.2 nm. A sectored piezoelectric tube can move the surfaces normally and laterally relative to each other. The bending of normal and shear springs in response to applied motion, monitored via the change in an air gap capacitor or through changes in the interface fringes, directly reveals the shear and normal forces, respectively.

Figure 5

(a) Normal force profiles Fn(D)/R versus separation D between two mica surfaces across pure water (black symbols) following overnight incubation in HSPC small unilamellar vesicle (SUV) dispersion (0.3 mM) followed by washing, and thereafter across HA solution (red symbols), normalized in the Derjaguin approximation by the mean radius of curvature of the mica sheets (R). (b) Normalized force profiles Fn(D)/R versus separation D between two mica surfaces across POPC dispersion (0.3 mM, black symbols) following overnight incubation, and then across the HA–POPC mixture (0.5 mg/mL:0.3 mM, red symbols). The zero of separation (D = 0) is with respect to air contact between bare mica surfaces. The insets show the ‘hard wall’ behavior on a larger, linear–linear scale with the same data points as in the main figure; in (a) the hard wall in HA-free water (black symbols) is generally at ca. 30 nm (and occasionally at ca. 20 nm), and in HA solution it is consistently at ca. 20 nm; in (b) the hard wall is at ca. 10 nm in both POPC dispersion and in the HA–POPC mixture. Different shaped symbols correspond to different contact positions based on 5 independent experiments with first approaches (filled symbols), second approaches (empty symbols), and receding profiles (crossed symbols).

Figure 6

(a) Frictional force Fs versus Fn between two mica surfaces across pure water following overnight incubation with HSPC SUVs dispersion (black symbols), where μ ≈ 0.00028 ± 0.00014, and across HA solution (red symbols), where μ ≈ 0.00041 ± 0.00016. (b) Frictional force Fs versus Fn between two mica surfaces across POPC SUVs dispersion following overnight incubation with POPC SUVs dispersion (black symbols) and across POPC-HA mixture (red symbols), where both of them showed μ < 0.0001. The inset shows directly measured shear traces Fs(t) for POPC layer across HA–POPC (2nd trace) and HSPC layer across HA (3rd trace) with similar normal force as a function of applied sliding motion (top trace). The low values of Fs are determined by fast Fourier transform of the shear traces [12].

Figure 6

(a) Frictional force Fs versus Fn between two mica surfaces across pure water following overnight incubation with HSPC SUVs dispersion (black symbols), where μ ≈ 0.00028 ± 0.00014, and across HA solution (red symbols), where μ ≈ 0.00041 ± 0.00016. (b) Frictional force Fs versus Fn between two mica surfaces across POPC SUVs dispersion following overnight incubation with POPC SUVs dispersion (black symbols) and across POPC-HA mixture (red symbols), where both of them showed μ < 0.0001. The inset shows directly measured shear traces Fs(t) for POPC layer across HA–POPC (2nd trace) and HSPC layer across HA (3rd trace) with similar normal force as a function of applied sliding motion (top trace). The low values of Fs are determined by fast Fourier transform of the shear traces [12].

Figure 7

Illustrating the two mechanisms which are attributed to enable low friction hydration lubrication between PC-coated layers, even in the presence of HA. (a) In the case of the gel-phase lipid layers (HSPC) in HA-free water, the slip plane is the mid-plane between the vesicles. (b) When HA is added, it may adsorb on the vesicles’ surfaces to bridge the gap between them, leading to high friction at the mid-plane interface. As a result, the slip plane shifts, as shown, to the inner interfaces, where hydration lubrication is unhindered by the HA bridging. (c) In the case of the liquid-phase POPC, where the surfaces are immersed in a POPC dispersion (see text), the addition of HA leads to interactions between the lipids in solution and the polysaccharide, as indicated; this suppresses their adsorption on the surface-attached POPC bilayers and thus enables hydration lubrication between them, unhindered by any HA bridging.