Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2024 Jun 1.
Published in final edited form as: J Mech Behav Biomed Mater. 2023 Apr 6;142:105827. doi: 10.1016/j.jmbbm.2023.105827

CORRELATION ANALYSIS OF CARTILAGE WEAR WITH BIOCHEMICAL COMPOSITION, VISCOELASTIC PROPERTIES AND FRICTION

AMIN JOUKAR 1,2, AMY CREECY 3,4, SONALI KARNIK 2,4, HESSAM NOORI-DOKHT 1,2, STEPHEN B TRIPPEL 4, JOSEPH M WALLACE 3,4, DIANE R WAGNER 2,3,4,*
PMCID: PMC10175217  NIHMSID: NIHMS1893059  PMID: 37060715

Abstract

Healthy articular cartilage exhibits remarkable resistance to wear, sustaining mechanical loads and relative motion for decades. However, tissues that replace or repair cartilage defects are much less long lasting. Better information on the compositional and material characteristics that contribute to the wear resistance of healthy cartilage could help guide strategies to replace and repair degenerated tissue. The main objective of this study was to assess the relationship between wear of healthy articular cartilage, its biochemical composition, and its viscoelastic material properties. The correlation of these factors with the coefficient of friction during the wear test was also evaluated. Viscoelastic properties of healthy bovine cartilage were determined via stress relaxation indentation. The same specimens underwent an accelerated, in vitro wear test, and the amount of glycosaminoglycans (GAGs) and collagen released during the wear test were considered measures of wear. The frictional response during the wear test was also recorded. The GAG, collagen and water content and the concentration of the enzymatic collagen crosslink pyridinoline were quantified in tissue that was adjacent to each wear test specimen. Finally, correlation analysis was performed to identify potential relationships between wear characteristics of healthy articular cartilage with its composition, viscoelastic material properties and friction. The findings suggest that stiffer cartilage with higher GAG, collagen and water content has a higher wear resistance. Enzymatic collagen crosslinks also enhance the wear resistance of the collagen network. The parameters of wear, composition, and mechanical stiffness of cartilage were all correlated with one another, suggesting that they are interrelated. However, friction was largely independent of these in this study. The results identify characteristics of healthy articular cartilage that contribute to its remarkable wear resistance. These data may be useful for guiding techniques to restore, regenerate, and stabilize cartilage tissue.

Keywords: cartilage, wear, friction, glycosaminoglycans, collagen, crosslinks, pyridinoline, viscoelastic

1. INTRODUCTION

Articular cartilage is a remarkably resilient tissue. Cartilage covering the bones in synovial joints endures repeated and persistent relative motion between the joint surfaces with daily activities. Yet, healthy cartilage exhibits remarkable resistance to wear, sustaining mechanical loads and relative motion for several decades of lifespan in many individuals. However, when cartilage becomes damaged and degenerated, replacement tissues are much less long lasting. For example, osteochondral grafts have limited life span and durability. Osteochondral grafts for cartilage repair have graft failure rate of up to 36% over 10 years (Ulstein et al., 2014). Tissue engineering has emerged as an alternative procedure to replace damaged tissues, however, generating functional cartilage constructs remains challenging (Cipollaro et al., 2019), and the wear characteristics of engineered cartilage are not typically studied. A better understanding of the features and characteristics of healthy cartilage that contributes to its wear resistance could be beneficial in guiding techniques that aim to repair, regenerate or replace cartilage tissue.

The three main extracellular components of articular cartilage are the crosslinked collagen fibril network, negatively charged proteoglycans, and water. A number of studies have been conducted on the relationship between tissue stiffness and the composition of cartilage (Julkunen et al., 2010, 2007; P. Julkunen et al., 2008; Petro Julkunen et al., 2008a, 2008b; Kempson et al., 1970; Kiviranta et al., 2009; Korhonen et al., 2002; R Krishnan et al., 2004; Laasanen et al., 2002). Proteoglycans rich in negatively charged glycosaminoglycans (GAGs) are mainly associated with the compressive modulus of cartilage tissue, while the density of the collagen network and the degree to which it is crosslinked are related to the tensile modulus and strength of the tissue (Brown et al., 2019; Chen et al., 2002; Williamson et al., 2003; Yan et al., 2009). However, whether the composition and/or material properties of cartilage are predictive of its wear resistance is not known. Many recent advances have been made in understanding cartilage lubrication and its notably low coefficient of friction (Bonnevie and Bonassar, 2020; Eschweiler et al., 2021; Moore et al., 2017). Wear and friction are functional properties that are important when considering the selection of osteochondral autografts and allografts, and in developing techniques and constructs to repair cartilage defects.

The main objective of this study was to assess the relationship between wear of healthy articular cartilage and its biochemical composition and viscoelastic characteristics. The correlation of the coefficient of friction during the wear test with biochemical composition, viscoelastic properties, and mechanical wear was also assessed. The instantaneous modulus, equilibrium modulus, and relaxation time constant of bovine cartilage specimens were determined via stress relaxation indentation. Then the same specimens underwent an accelerated, in vitro wear test, and the amount of GAG and hydroxyproline (HYP) released during the wear test were quantified as measures of wear. HYP is an amino acid that is found almost exclusively in collagen, therefore the quantification of HYP released during the wear test indicated the collagen lost from the specimen due to wear. The frictional response was also recorded during the wear test, and the initial and equilibrium coefficient of friction (COF) were reported to capture the time-dependent frictional behavior. The composition of the cartilage tissue that was adjacent to each specimen was quantified. These measurements included GAG content of the tissue, HYP content as a determination of the collagen content in the tissue, the concentration of the enzymatic collagen crosslink pyridinoline (PYD), and the tissue water content. Finally, correlation analysis was performed to identify potential relationships between wear and frictional characteristics of healthy cartilage tissue and its composition and viscoelastic material properties.

2. METHODS

2.1. Specimen Preparation

Bovine knee joints from skeletally mature animals were obtained from a local abattoir (Knightstown Locker, Knightstown, IN) and were stored at −20 °C until use. Thawed joints were dissected to isolate the femoral condyles, which were hydrated with phosphate buffered saline (PBS). Twenty-five osteochondral specimens with 9.5 mm diameter and approximate height of 4 mm were cored from the flattest regions of the condyles using a drill press with the coring axis perpendicular to the articular surface. Specimens were not collected from any tissue that was fibrillated, damaged, or discolored. The osteochondral plugs were stored at −20 °C until subsequent indentation and wear testing. Tissue that had been adjacent to the cored specimens was also stored at −20 °C for the composition analyses.

2.2. Composition Quantification

Full thickness cartilage tissue that was adjacent to each osteochondral specimen was removed from the subchondral bone with a scalpel. The wet weights of samples were measured and then specimens were lyophilized. The dried specimens were reweighed, and the water content proportion was calculated as (wet weight – dry weight)/wet weight. Following that, the specimens were digested with papain. Sulfated glycosaminoglycan (GAG) content was quantified using the dimethylmethylene blue dye-binding assay (DMMB; Sigma Aldrich, St. Louis, MO) with a chondroitin sulfate standard. To quantify the hydroxyproline (HYP) content and pyridinoline (PYD) crosslink density, aliquots from papain digested samples were hydrolyzed with 6N HCL. The acid was evaporated, and samples were resuspended in HPLC-grade water and filtered. One aliquot of hydrolysate was used to quantify HYP content using a chloramine-T assay. Another aliquot was suspended in a solution of pyridoxine in HPLC-grade water. These were run on a Spherisorb® ODS2 (4.6x150mm, PSS831913, Waters, Milford, MA) using mobile phases heptafluorobutyric acid and acetonitrile to detect PYD in the samples (Creecy et al., 2017). Concentration of PYD was determined by comparison to a standard curve of known concentrations. GAG and HYP content were normalized to wet weight, while PYD values were normalized to HYP content. Two replicates per sample were conducted in each test and were averaged to quantify HYP, GAG, and PYD in the cartilage tissue.

2.3. Viscoelastic Properties from Indentation

Thawed osteochondral cores were submerged in PBS containing protease inhibitors. Stress relaxation testing was performed via indentation using a flat punch indenter of 0.3 mm diameter to obtain viscoelastic properties of the cartilage. Biomomentum (Montreal, QC) mechanical tester model Mach-1 v500css was equipped with a load cell (AL312AL, Honeywell Industries Inc., Columbus, OH) with a 150 gf range and 0.0075 gf resolution. The stress-relaxation protocol, consisting of 20 sec ramp to 50 um peak displacement followed by 20 sec hold at this displacement, was previously developed to produce repeatable measurements (McGann et al., 2014). Displacement was measured from the cartilage surface, which was defined with a threshold load value of 0.0875 gf. Indentation was executed in quadruplicate at five locations on each specimen.

Best-fit values of the standard linear solid (SLS) model was found from experimental stress relaxation data (Noori-Dokht et al., 2022). The instantaneous (Einst) and equilibrium moduli (Eeq) and time constant (τ) were calculated from the best-fit values of the SLS model. Average values were calculated for each specimen.

2.4. Wear and Friction

After indentation testing, osteochondral specimens were glued to a cylindrical bone core to make specimens with overall length of 18 mm for the wear test. An accelerated pin-on-disc wear test was performed with a reciprocal motion against a T316 stainless steel plate with a #8 mirror finish using a UMT Tribolab (Bruker, San Jose, CA) equipped with a two-axis load cell and one-dimensional displacement sensor (Figure 1). The specimens were loaded to a constant 160 N, or an average contact pressure of approximately 2.2 MPa. This load was maintained for 14000 reciprocating cycles of 18 mm length in each direction at a speed of 4 mm/s with 1 sec pause at each end. The 43.5 hours wear test was conducted at room temperature in a solution containing protease inhibitors in PBS (Hossain et al., 2020; Noori-Dokht et al., 2022).

Figure 1.

Figure 1.

Open in a new tab

Wear test set-up. Osteochondral specimens underwent reciprocating motion while loaded against a polished T316 stainless-steel plate. The wear test was performed in PBS containing protease inhibitors. Friction and normal forces were collected from the two-axis load cells and the coefficient of friction was calculated. Adapted from Hossain et al., 2020.

Friction and normal loads were recorded by UMT Tribolab load sensor at a rate of 100 Hz during the wear test and initial and equilibrium COF were extracted. For calculating the COF, a Matlab (Mathworks, Natick, MA) code was developed to discard data from the acceleration and deceleration portion of each cycle so that only data at 4 mm/s speed was considered. The average COF was calculated for each reciprocating cycle. The average of the first reciprocating cycle was considered the initial COF and the average of the last 8000 cycles was considered the equilibrium COF (Hossain et al., 2020; Noori-Dokht et al., 2022).

Wear was measured by biochemically quantifying the amount of GAG and HYP that were released to the solution bath during the wear test. The solution in the bath along with any particles that had been released to the bath during wear was collected for each specimen, lyophilized, and was digested in a papain solution. GAG content was measured using the DMMB assay (Ateshian et al., 2005; Oungoulian et al., 2015; Radin et al., 1982), as described above. Separate aliquots were hydrolyzed in hydrochloric acid and were assessed for HYP content with a chloramine-T assay (Bonitsky et al., 2017; Lipshitz et al., 1975; McGann et al., 2015, 2012; Oungoulian et al., 2015), as in the composition analysis.

2.5. Correlation and Statistics

Pearson correlation was conducted between mechanical properties of cartilage obtained from indentation data, cartilage composition, wear and friction using GraphPad Prism Software with p < 0.05 considered as significant. Significant correlations were displayed with a red line passing through the data points.

3. RESULTS

3.1. Relationship Between Different Measures of Wear and Between Different Extracellular Constituents

Two measures of wear were evaluated for each specimen in the accelerated in vitro wear test: the amount of GAG released to the hydrating fluid during the wear test (GAG wear), and the amount of HYP released (HYP wear). To determine whether these measures of wear were independent from one another, the correlation between them was evaluated. A positive significant linear correlation was found between these two measures of cartilage wear (Figure 2A), indicating that these wear measurements were not independent from one another. Likewise, the concentration of individual constituents that make up the composition of cartilage were not independent parameters. A positive correlation was also detected between HYP and GAG content of cartilage (Figure 2B). Because of these dependencies, multilinear regressions were not conducted for subsequent analyses.

Figure 2.

Figure 2.

Open in a new tab

Positive correlations between A) the GAG and HYP released in the wear of articular cartilage; B) HYP and GAG content of cartilage, each normalized to tissue wet weight (WW).

3.2. Correlation of Cartilage Wear, Biochemical Composition and Viscoelastic Properties

The amount of HYP released during the wear test was negatively correlated with HYP and GAG content of cartilage, PYD concentration, and water content (Figure 3A, C, E and G). A negative correlation was also detected between the amount of GAG released during the wear test and HYP and GAG content and PYD concentration (Figure 3B, D, F and H). The strongest correlations of these were between HYP wear and GAG and HYP content (Figure 3A, C).

Figure 3.

Figure 3.

Open in a new tab

Negative correlations between matrix constituents and cartilage wear. HYP content normalized to tissue wet weight (WW) (A, B); GAG content normalized to tissue WW (C, D); PYD density (E, F); water content (G, H); HYP released due to wear (A, C, E, G); GAG released due to wear (B,D, F, H).

Correlations between cartilage wear and viscoelastic material properties were also observed. Instantaneous and equilibrium modulus of cartilage was negatively correlated with HYP and GAG removed from the cartilage tissue during the wear test (Figure 4A, B, D and E). A significant correlation was also found between the time constant with HYP wear of cartilage (Figure 4C). There was no correlation detected between the time constant and GAG wear (Figure 4F). The strongest correlations were between HYP wear and instantaneous and equilibrium modulus.

Figure 4.

Figure 4.

Open in a new tab

Relationships between viscoelastic material properties and cartilage wear. HYP released due to wear (A – C); GAG released due to wear (D-F); instantaneous modulus, Einst (A, D); equilibrium modulus, Eeq (B, E); time constant, τ (C, F).

A positive correlation was found between instantaneous modulus of cartilage and HYP and GAG content and PYD concentration (Figure 5A, D and G). The equilibrium modulus was also positively correlated with HYP, GAG and PYD content (Figure 5B, E and H). The time constant was correlated with HYP and GAG content of cartilage (Figure 5C and F), however, no correlation was detected between the time constant and PYD concentration in the collagen network (Figure 5I). Additionally, no correlation was detected between the water content and any of the viscoelastic material properties (data not shown). The strongest correlation was between GAG content and time constant.

Figure 5.

Figure 5.

Open in a new tab

Relationship between cartilage composition and viscoelastic material properties. HYP content normalized to tissue wet weight (WW) (A – C); GAG content normalized to tissue WW (D-F); PYD density (G-I); instantaneous modulus, Einst (A, D, G); equilibrium modulus, Eeq (B, E, H); time constant, τ (C, F, I).

3.3. Correlation of Cartilage Friction with Composition, Viscoelastic Properties and Wear

During the wear test, the COF was initially low and increased to an approximately equilibrium value (Figure 6). No correlation was detected between cartilage matrix constituents HYP, GAG and PYD, and the initial and equilibrium coefficient of friction (Figure 7A-F). There was a negative correlation between the water content and the equilibrium COF (Figure 7H), but not between the water content and the initial COF (Figure 7G). Additionally, no correlation was detected between any viscoelastic material properties of cartilage and friction characteristics (Figure 8A-F). Finally, no trend was observed in the relationship between cartilage wear and coefficient of friction (Figure 9A-D). The relationship between equilibrium COF and HYP released during the wear test had the lowest p-value at 0.15.

Figure 6.

Figure 6.

Open in a new tab

Representative mean coefficient of friction (COF) value calculated for each reciprocating cycle of the wear test.

Figure 7.

Figure 7.

Open in a new tab

Relationship between cartilage composition and coefficient of friction. HYP content normalized to wet weight (WW) (A, B); GAG content normalized to WW (C, D); PYD density (E, F); water content (G, H); initial coefficient of friction (A, C, E, F); equilibrium coefficient of friction (B, D, F, H).

Figure 8.

Figure 8.

Open in a new tab

Relationship between coefficient of friction and viscoelastic material properties in cartilage. Initial coefficient of friction (A – C); Equilibrium coefficient of friction (D-F); Instantaneous modulus, Einst (A, D); Equilibrium modulus, Eeq (B, E); Time constant, τ (C, F).

Figure 9.

Figure 9.

Open in a new tab

Relationship between cartilage wear and coefficient of friction. Initial coefficient of friction (A, C); Equilibrium coefficient of friction (B, D); HYP released due to wear (A, B); GAG released due to wear (C, D).

4. DISCUSSION

The main objective of this study was to evaluate the relationships between wear of healthy articular cartilage and its biochemical composition and viscoelastic material properties. The data reveal that lower wear rates are found in tissue with high GAG, HYP and water content, and high collagen crosslink density. As HYP is found almost exclusively in collagen, and crosslinked type II collagen and GAG-rich proteoglycans are the primary solid matrix components of cartilage, results overall suggest that a dense, highly interconnected and well-hydrated matrix is associated with low cartilage wear. Finding from the study also indicate that higher material stiffness of cartilage is associated with cartilage wear resistance, but that many of these parameters may be independent of COF in healthy cartilage. These results identify characteristics of healthy native cartilage that contribute to its remarkable wear resistance.

This study reveals the relationships between wear of cartilage with its compositional properties. First, when wear properties are investigated, the amount of GAG and HYP released during the wear test are positively correlated with each other (Figure 2A). Disruption of the collagen network due to wear may release proteoglycans which are trapped by the collagen network. Both the amount of HYP and GAG released during the wear test were negatively correlated with GAG and HYP content (Figure 3A - D). This may suggest that distributing the wear load across more matrix molecules reduces the load on each, and consequently reduces the failure and wear rate. As collagen and GAG content are positively correlated with each other (Figure 2B), it is difficult to separate their individual effect and contribution to wear resistance. Measures of wear were also negatively correlated with the PYD crosslink density of the matrix (Figure 3E - F). Factors that increase the wear resistance of cartilage may be similar to factors that strengthen cartilage tissue in a tensile test. Both collagen content and PYD density were correlated with increased tensile strength of cartilage (Williamson et al., 2003). Additionally, non-enzymatic crosslinks associated with advanced glycation end products (AGEs) also increase the tensile strength of cartilage (Chen et al., 2002). These data together with current results suggest that cartilage wear resistance may be related to other tissue failure properties such as the tensile strength.

Water content of the tissue also influenced cartilage wear and frictional properties. Higher water content of the cartilage tissue was correlated with decreased release of both HYP and GAGs during the wear test (Figure 3G-H), and with decreased equilibrium COF (Figure 7H). This may indicate that a higher tissue water content increases lubrication at the articular surface, decreasing the COF and consequently the mechanical wear of cartilage. Although a direct correlation was not detected between measures of cartilage wear and the COF, the correlation between the release of HYP due to wear and the equilibrium COF trended towards significance (p = 0.15).

The study also indicated that cartilage tissue with stiffer viscoelastic properties has lower wear (Figure 5). This may indicate that tissues with lower moduli experienced higher strain and elevated damage during the wear test, liberating more cartilage matrix from the test specimen. However, because cartilage wear, composition, and material properties all appear to be interrelated, it is difficult to tease out the specific effect of modulus and strain on wear. In engineering materials, a stiffer material is not necessarily a more wear resistance one. One example of this is found in ultra-high molecular weight polyethylene (UHMWPE), a polymer used in prosthetic joints. UHMWPE is typically crosslinked via irradiation to improve its wear resistance (McKellop et al., 1999b, 1999a; Muratoglu et al., 1999). However, irradiation can decrease the polymer crystallinity, producing a crosslinked material with a lower modulus (Bistolfi et al., 2009; Gomoll et al., 2002; Grobbelaar et al., 1978), even while increasing the wear resistance.

The relationship between the cartilage modulus and its collagen and GAG content has been extensively evaluated in previous studies (Julkunen et al., 2010, 2007; P. Julkunen et al., 2008; Petro Julkunen et al., 2008a, 2008b; Kempson et al., 1970; Kiviranta et al., 2009; Korhonen et al., 2002; R Krishnan et al., 2004; Laasanen et al., 2002). The current investigation corroborates these findings with a positive correlation between the instantaneous and equilibrium moduli with HYP and GAG content. The material stiffness of cartilage has also been associated with crosslink density, including native enzymatic and non-enzymatic crosslinks (Brown et al., 2019; McGann et al., 2014; Verzijl and DeGroot, 2002; Williamson et al., 2003) as well as exogenously induced crosslinks (Bonitsky et al., 2017; McGann et al., 2015; Noori-Dokht et al., 2022). Consistent with this, we found a correlation between PYD crosslink density of cartilage with the viscoelastic moduli (Figure 5C, F). The time constant of the stress relaxation, τ, is inversely proportional to the permeability, or the ability of fluid to flow through the tissue (Nia et al., 2011). Accordingly, a correlation was observed between collagen and GAG content, factors that would contribute to a denser matrix with lower permeability, and τ (Figure 5G, H). Conversely, no relationship was detected between τ and crosslink density (Figure 5I).

Interestingly, composition of the solid cartilage matrix and the material properties of the tissue had no detectable influence on the COF during the wear test (Figure 7A - F, Figure 8). This indicates that friction may be independent of bulk tissue properties in healthy cartilage under the conditions of our in vitro test. One exception was the correlation between equilibrium COF and the water content of the tissue (Figure 7H). Factors that were not measured may also regulate the friction, such as the amount of lubricin and hyaluronic acid at the articular surface (Bonnevie et al., 2015). Moreover, in this study, the composition of full thickness cartilage was evaluated, while friction may depend more heavily on the specific cartilage composition found at or near the articular surface. Although cartilage wear appeared to be independent of the COF in this study, in previous studies increased cartilage friction was associated with accelerated cartilage damage (Elsaid et al., 2005; Jay et al., 2007; Oungoulian et al., 2015). In these previous studies, cartilage damage was observed with an increase in the coefficient of friction of about 3 to 4-fold. In the current study, the initial COF of the specimens had a relatively wide range of ~0.001 to 0.008. However, the initial COF is unlikely to contribute to the wear, both because of its low magnitude and because the equilibrium COF is maintained for a longer time duration (Figure 6). The equilibrium COF for the healthy tissue in the current study had a relatively narrow range of ~0.15 – 0.3, or only about a 2-fold increase from the lowest to highest values. Because of these relatively consistent values, a relationship between wear and friction may be difficult to detect in healthy cartilage. As no significant correlations between cartilage wear and COF were detected, differences in wear in the current study may be primarily attributed to variations in tissue composition and structure.

Previous studies presented contradictory evidence on the role of GAG and collagen content on the frictional response of cartilage. Several previous studies reported that GAG and collagen depletion had no effect on the frictional response of cartilage (Katta et al., 2009; Larson et al., 2017; Mochizuki et al., 2013; Pickard et al., 1998), consistent with the current study. However, other studies demonstrated an increase in the equilibrium COF and either no difference or an increase in the initial COF after GAG and/or collagen depletion (Basalo et al., 2006, 2005; Bauer et al., 2021; Katta et al., 2008; Naka et al., 2005). In other studies, the coefficient of friction was reduced after GAG loss and collagen degradation (Accardi, 2013; DuRaine et al., 2009). The conflicting outcomes can be partially explained by differing amounts of degradation, and types of lubricant (Bonnevie et al., 2018).

Although a number of significant correlations were identified between cartilage wear, composition, and viscoelastic properties, R2 values were all below 0.5, indicating that no one factor accounted for the majority of the wear resistance. However, the R2 values from the correlation analyses between composition and other properties were likely underestimations. Composition was measured in separate samples of tissue that had been adjacent to the specimens that were evaluated for viscoelastic properties, friction and wear. Although the two samples were adjacent to one another, differences in composition and mechanical behavior can be expected at different anatomic locations. The R2 values from the correlation analyses involving wear may have been further underestimated. We previously demonstrated that the wear rate depends in part on whether the test specimen is oriented such that the reciprocating motion is parallel to the dominant collagen fiber direction at the articular surface or perpendicular to this direction (Hossain et al., 2020). In the current study, the specimens were oriented randomly with respect to the primary fiber direction, which may account for a portion of the variability in wear. Other factors that were not evaluated also likely contributed to variation in wear. For example, additional crosslinks in the tissue, such as deoxypyridinoline and advanced glycation end products are expected to be present in lower concentrations than PYD but may also contribute to the wear resistance of cartilage. These sources of variability also imply that in some cases, correlations between parameters may have been obscured, making it more difficult to detect weaker relationships.

The main limitation of this study is that the in vitro wear test produced accelerated matrix loss compared to physiologic wear. Although the contact pressure in the in vitro wear test of approximately 2.2 MPa is in the range of cartilage contact pressure in the hip and knee during moderate activities (Brand, 2005) the load was applied for the entire duration of the wear test, and does not represent a physiologic load history. The prolonged load induces fluid exudation from the tissue, increasing the COF and transferring the load to the solid matrix (Krishnan et al., 2004) to accelerate the wear. The opposing stainless steel wear surface is also non-physiologic, except in the case of a metal hemiarthroplasty. Despite the accelerated nature of the wear test, we’ve previously shown that it generates some of the hallmarks of osteoarthritis, including tissue loss and collagen damage (Hossain et al., 2020). This accelerated test enables the study of cartilage wear under laboratory conditions.

Results of this study could be used to direct the selection of cartilage tissue for autograft or allograft for individuals who suffer from OA. The transplanted tissue should ideally not be collected from a site where the collagen, GAG and water content, collagen crosslink density are low to avoid possible higher wear in the cartilage. The cartilage composition and wear are related to stiffness, therefore, an indentation test, which is non-destructive, could theoretically be performed to select the tissue. Also, crosslinking agents could be used to improve the mechanical properties and resistance of cartilage grafts to enzymatic degradation and wear, including grafts that are decellularized (Elder et al., 2017). In addition, collagen and GAG content, collagen crosslink density, hydration, and mechanical properties could be an appropriate first screen for the durability of tissue engineered cartilage constructs. Those composed of high collagen and GAG density or have a highly crosslinked collagen network and elevated water content, with correspondingly high material properties, potentially have high wear resistance. This could be an initial indicator for wear properties of tissue engineered cartilage without a time consuming and destructive wear test. Finally, in vitro studies demonstrate that stabilizing the native cartilage matrix via exogenous crosslinking may be a strategy to preserve cartilage tissue. Some of these exogenous crosslinkers have been applied directly to native cartilage, including genipin and CASPC, which increased both the modulus and wear resistance of the cartilage tissue (Bonitsky et al., 2017; McGann et al., 2015; Noori-Dokht et al., 2022). Other studies demonstrated that crosslinking additional macromolecules such as hyaluronic acid to the cartilage tissue improves the damaged cartilage biomechanics by enhancing the compressive modulus and reducing tissue permeability (Grenier et al., 2015; Kowalski et al., 2022; Patel et al., 2021)

In conclusion, this study demonstrates the relationships between wear and friction behavior of healthy articular cartilage in an accelerated in vitro wear test with its viscoelastic material properties and composition. The findings suggest that well-hydrated and stiffer cartilage with higher GAG and collagen content has a higher wear resistance. Enzymatic collagen crosslinks also enhance the wear resistance of the collagen network. The parameters of wear, composition, and mechanical stiffness of cartilage were all correlated with one another, suggesting that they are interrelated. However, friction was independent of these in this study. This data may be useful for guiding cartilage treatments such as cartilage transplantation and techniques to restore, regenerate, and stabilize cartilage tissue.

Acknowledgements

This research was supported by grants from the National Institutes of Health, including NIAMS AR069657 and AR065970. The authors thank M. Jayed Hossain for his early work on this project.

Diane Wagner reports financial support was provided by National Institutes of Health.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationship: that could have appeared to influence the work reported in this paper.

References

  1. Accardi M, 2013. Numerical and experimental characterisation of articular cartilage: a study on biomechanics and biotribology, osteoarthritis and tissue engineering solutions. [Google Scholar]
  2. Ateshian G. a, Mow VC, Huiskes R, 2005. Friction, lubrication, and wear of articular cartilage and diarthrodial joints. Basic Orthop. Biomech. Mechano-Biology 10.1186/1475-925X-4-28 [DOI] [Google Scholar]
  3. Basalo IM, Chen FH, Hung CT, Ateshian GA, 2006. Frictional response of bovine articular cartilage under creep loading following proteoglycan digestion with chondroitinase ABC. J. Biomech. Eng 128, 131–134. 10.1115/1.2133764 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Basalo IM, Raj D, Krishnan R, Chen FH, Hung CT, Ateshian GA, 2005. Effects of enzymatic degradation on the frictional response of articular cartilage in stress relaxation. J. Biomech 38, 1343–1349. 10.1016/j.jbiomech.2004.05.045 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bauer C, Göçerler H, Niculescu-Morzsa E, Jeyakumar V, Stotter C, Klestil T, Franek F, Nehrer S, 2021. Biotribological Tests of Osteochondral Grafts after Treatment with Pro-Inflammatory Cytokines. Cartilage, 10.1177/1947603521994900 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bistolfi A, Turell MB, Lee YL, Bellare A, 2009. Tensile and tribological properties of high-crystallinity radiation crosslinked UHMWPE. J. Biomed. Mater. Res. - Part B Appl. Biomater 90B, 137–144. 10.1002/jbm.b.31265 [DOI] [PubMed] [Google Scholar]
  7. Bonitsky CM, McGann ME, Selep MJ, Ovaert TC, Trippel SB, Wagner DR, 2017. Genipin crosslinking decreases the mechanical wear and biochemical degradation of impacted cartilage in vitro. J. Orthop. Res 10.1002/jor.23411 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bonnevie ED, Bonassar LJ, 2020. A Century of Cartilage Tribology Research Is Informing Lubrication Therapies. J. Biomech. Eng 142, 031004–031004. 10.1115/1.4046045 [DOI] [PubMed] [Google Scholar]
  9. Bonnevie ED, Galesso D, Secchieri C, Bonassar LJ, 2018. Degradation alters the lubrication of articular cartilage by high viscosity, hyaluronic acid-based lubricants. J. Orthop. Res 36, 1456–1464. 10.1002/jor.23782 [DOI] [PubMed] [Google Scholar]
  10. Bonnevie ED, Galesso D, Secchieri C, Cohen I, Bonassar LJ, 2015. Elastoviscous Transitions of Articular Cartilage Reveal a Mechanism of Synergy between Lubricin and Hyaluronic Acid, 10.1371/journal.pone.0143415 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Brand RA, 2005. Joint contact stress: a reasonable surrogate for biological processes? Iowa Orthop. J [PMC free article] [PubMed] [Google Scholar]
  12. Brown WE, DuRaine GD, Hu JC, Athanasiou KA, 2019. Structure-function relationships of fetal ovine articular cartilage. Acta Biomater. 87,235–244. 10.1016/j.actbio.2019.01.073 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Chen AC, Temple MM, Ng DM, Verzijl N, DeGroot J, TeKoppele JM, Sah RL, 2002. Induction of advanced glycation end products and alterations of the tensile properties of articular cartilage. Arthritis Rheum 46, 3212–3217. [DOI] [PubMed] [Google Scholar]
  14. Cipollaro L, Ciardulli MC, Porta G. Della, Peretti GM, Maffulli N, 2019. Biomechanical issues of tissue-engineered constructs for articular cartilage regeneration: In vitro and in vivo approaches. Br. Med. Bull 132, 53–80. 10.1093/bmb/ldz034 [DOI] [PubMed] [Google Scholar]
  15. Creecy A, Uppuganti S, Merkel AR, Neal DO, Makowski AJ, Granke M, Voziyan P, Nyman JS, 2017. Progression of Type 2 Diabetes in the ZDSD Rat 99, 289–301. 10.1007/s00223-016-0149-z.Changes [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. DuRaine G, Neu CP, Chan SMT, Komvopoulos K, June RK, Reddi AH, 2009. Regulation of the friction coefficient of articular cartilage by TGF-nβ1 and IL-1β. J. Orthop. Res 27, 249–256. 10.1002/jor.20713 [DOI] [PubMed] [Google Scholar]
  17. Elder S, Pinheiro A, Young C, Smith P, Wright E, 2017. Evaluation of genipin for stabilization of decellularized porcine cartilage. J. Orthop. Res 35, 1949–1957. 10.1002/JOR.23483 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Elsaid KA, Jay GD, Warman ML, Rhee DK, Chichester CO, 2005. Association of articular cartilage degradation and loss of boundary-lubricating ability of synovial fluid following injury and inflammatory arthritis. Arthritis Rheum. 52, 1746–1755. 10.1002/ART.21038 [DOI] [PubMed] [Google Scholar]
  19. Eschweiler J, Horn N, Rath B, Betsch M, Baroncini A, Tingart M, Migliorini F, 2021. The Biomechanics of Cartilage — An Overview. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gomoll A, Wanich T, Bellare A, 2002. J-integral fracture toughness and tearing modulus measurement of radiation cross-linked UHMWPE. J. Orthop. Res 20, 1152–1156. 10.1016/S0736-0266(02)00073-6 [DOI] [PubMed] [Google Scholar]
  21. Grenier S, Donnelly PE, Gittens J, Torzilli PA, 2015. Resurfacing damaged articular cartilage to restore compressive properties. J. Biomech 48,122–129. 10.1016/j.jbiomech.2014.10.023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Grobbelaar CJ, Du Plessis TA, Marais F, 1978. The radiation improvement of polyethylene prostheses. A preliminary study. J. Bone Jt. Surg. - Ser. B 60 B, 370–374. 10.1302/0301-620x.60b3.681412 [DOI] [PubMed] [Google Scholar]
  23. Hossain MJ, Noori-Dokht H, Karnik S, Alyafei N, Joukar A, Trippel SB, Wagner DR, 2020. Anisotropic properties of articular cartilage in an accelerated in vitro wear test. J. Mech. Behav. Biomed. Mater 109, 103834. 10.1016/j.jmbbm.2020.103834 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Jay GD, Torres JR, Rhee DK, Helminen HJ, Hytinnen MM, Cha CJ, Elsaid K, Kim KS, Cui Y, Warman ML, 2007. Association between friction and wear in diarthrodial joints lacking lubricin. Arthritis Rheum. 56, 3662–3669. 10.1002/art.22974 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Julkunen P, Jurvelin JS, Isaksson H, 2010. Contribution of tissue composition and structure to mechanical response of articular cartilage under different loading geometries and strain rates. Biomech. Model. Mechanobiol 9, 237–245. 10.1007/s10237-009-0169-y [DOI] [PubMed] [Google Scholar]
  26. Julkunen P, Kiviranta P, Wilson W, Jurvelin JS, Korhonen RK, 2007. Characterization of articular cartilage by combining microscopic analysis with a fibril-reinforced finite-element model. J. Biomech 40, 1862–1870. 10.1016/j.jbiomech.2006.07.026 [DOI] [PubMed] [Google Scholar]
  27. Julkunen Petro, Korhonen RK, Herzog W, Jurvelin JS, 2008a. Uncertainties in indentation testing of articular cartilage: a fibril-reinforced poroviscoelastic study. Med. Eng. Phys 30, 506–515. [DOI] [PubMed] [Google Scholar]
  28. Julkunen P, Korhonen RK, Nissi MJ, Jurvelin JS, 2008. Mechanical characterization of articular cartilage by combining magnetic resonance imaging and finite-element analysis - A potential functional imaging technique. Phys. Med. Biol 53, 2425–2438. 10.1088/0031-9155/53/9/014 [DOI] [PubMed] [Google Scholar]
  29. Julkunen Petro, Wilson W, Jurvelin JS, Rieppo J, Qu CJ, Lammi MJ, Korhonen RK, 2008b. Stress-relaxation of human patellar articular cartilage in unconfined compression: Prediction of mechanical response by tissue composition and structure. J. Biomech 41, 1978–1986. 10.1016/j.jbiomech.2008.03.026 [DOI] [PubMed] [Google Scholar]
  30. Katta J, Jin Z, Ingham E, Fisher J, 2009. Effect of nominal stress on the long term friction , deformation and wear of native and glycosaminoglycan deficient articular cartilage. Osteoarthr. Cartil 17, 662–668. 10.1016/j.joca.2008.10.008 [DOI] [PubMed] [Google Scholar]
  31. Katta J, Stapleton T, Ingham E, Jin ZM, Fisher J, 2008. The effect of glycosaminoglycan depletion on the friction and deformation of articular cartilage. Proc. Inst. Mech. Eng. Part H J. Eng. Med 222, 1–11. 10.1243/09544119JEIM325 [DOI] [PubMed] [Google Scholar]
  32. Kempson GE, Muir H, Swanson SAV, Freeman MAR, 1970. Correlations between stiffness and the chemical constituents of cartilage on the human femoral head. BBA - Gen. Subj 215, 70–77. 10.1016/0304-4165(70)90388-0 [DOI] [PubMed] [Google Scholar]
  33. Kiviranta P, Lammentausta E, Toyras J, Nieminen HJ, Julkunen P, Kiviranta I, Jurvelin JS, 2009. Differences in Acoustic Properties of Intact and Degenerated Human Patellar Cartilage During Compression. Ultrasound Med. Biol 35, 1367–1375. 10.1016/j.ultrasmedbio.2009.03.003 [DOI] [PubMed] [Google Scholar]
  34. Korhonen RK, Wong M, Arokoski J, Lindgren R, Helminen HJ, Hunziker EB, Jurvelin JS, 2002. Importance of the superficial tissue layer for the indentation stiffness of articular cartilage. Med. Eng. Phys 24, 99–108. 10.1016/S1350-4533(01)00123-0 [DOI] [PubMed] [Google Scholar]
  35. Kowalski MA, Fernandes LM, Hammond KE, Labib S, Drissi H, Patel JM, 2022. Cartilage-penetrating hyaluronic acid hydrogel preserves tissue content and reduces chondrocyte catabolism. J. Tissue Eng. Regen. Med 1–24. 10.1002/term.3352 [DOI] [PubMed] [Google Scholar]
  36. Krishnan R, Kopacz M, Ateshian GA, 2004. Experimental verification of the role of interstitial fluid pressurization in cartilage lubrication. J Orthop Res 22, 565–570. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Krishnan Ramaswamy, Kopacz M, Ateshian GA, 2004. Experimental verification of the role of interstitial fluid pressurization in cartilage lubrication. J. Orthop. Res 10.1016/j.orthres.2003.07.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Laasanen MS, Toyras J, Korhonen RK, Rieppo J, Saarakkala S, Nieminen MT, Hirvonen J, Jurvelin JS, 2002. Biomechanical properties of knee articular cartilage. Biorheology 40, 133–140. [PubMed] [Google Scholar]
  39. Larson KM, Zhang L, Elsaid KA, Schmidt TA, Fleming BC, Badger GJ, Jay GD, 2017. Reduction of friction by recombinant human proteoglycan 4 in IL-1α stimulated bovine cartilage explants. J. Orthop. Res 35, 580–589. 10.1002/jor.23367 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Lipshitz H, Etheredge R, Glimcher M, 1975. In vitro wear of articular cartilage. J. Bone Jt. Surg 57, 527–534. 10.2106/00004623-197557040-00015 [DOI] [PubMed] [Google Scholar]
  41. McGann ME, Bonitsky CM, Jackson ML, Ovaert TC, Trippel SB, Wagner DR, 2015. Genipin crosslinking of cartilage enhances resistance to biochemical degradation and mechanical wear. J. Orthop. Res 33, 1571–1579. 10.1002/jor.22939 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. McGann ME, Bonitsky CM, Ovaert TC, Wagner DR, 2014. The effect of collagen crosslinking on the biphasic poroviscoelastic cartilage properties determined from a semi-automated microindentation protocol for stress relaxation. J. Mech. Behav. Biomed. Mater 34, 264–272. 10.1016/j.jmbbm.2014.02.013 [DOI] [PubMed] [Google Scholar]
  43. McGann ME, Vahdati A, Wagner DR, 2012. Methods to assess in vitro wear of articular cartilage. Proc. Inst. Mech. Eng. Part H J. Eng. Med 226, 612–622. [DOI] [PubMed] [Google Scholar]
  44. McKellop H, Shen FW, DiMaio W, Lancaster JG, 1999a. Wear of gamma-crosslinked polyethylene acetabular cups against roughened femoral balls, in: Clinical Orthopaedics and Related Research. Lippincott Williams and Wilkins, pp. 73–82. 10.1097/00003086-199912000-00008 [DOI] [PubMed] [Google Scholar]
  45. McKellop H, Shen FW, Lu B, Campbell P, Salovey R, 1999b. Development of an extremely wear-resistant ultra high molecular weight polyethylene for total hip replacements. J. Orthop. Res 17, 157–167. 10.1002/jor.1100170203 [DOI] [PubMed] [Google Scholar]
  46. Mochizuki S, Yanagida S, Fujie H, 2013. Effect of enzymatic degeneration on the frictional property of articular cartilage. ASME 2013 Summer Bioeng. Conf. SBC 2013 1 B, 1–2. 10.1115/SBC2013-14461 [DOI] [Google Scholar]
  47. Moore AC, Schrader JL, Ulvila JJ, Burris DL, 2017. A review of methods to study hydration effects on cartilage friction. Tribol. - Mater. Surfaces Interfaces 11, 202–214. 10.1080/17515831.2017.1397329 [DOI] [Google Scholar]
  48. Muratoglu OK, Bragdon CR, O’Connor DO, Jasty M, Harris WH, Rizwan G, McGarry F, 1999. Unified wear model for highly crosslinked ultra-high molecular weight polyethylenes (UHMWPE). Biomaterials 20, 1463–1470. 10.1016/S0142-9612(99)00039-3 [DOI] [PubMed] [Google Scholar]
  49. Naka MH, Hattori K, Ohashi T, Ikeuchi K, 2005. Evaluation of the effect of collagen network degradation on the frictional characteristics of articular cartilage using a simultaneous analysis of the contact condition. Clin. Biomech 20,1111–1118. 10.1016/j.clinbiomech.2005.06.009 [DOI] [PubMed] [Google Scholar]
  50. Nia HT, Han L, Li Y, Ortiz C, Grodzinsky A, 2011. Poroelasticity of cartilage at the nanoscale. Biophys. J 101, 2304–2313. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Noori-Dokht H, Joukar A, Karnik S, Williams T, Trippel SB, Wagner DR, 2022. A Photochemical Crosslinking Approach to Enhance Resistance to Mechanical Wear and Biochemical Degradation of Articular Cartilage. Cartilage. 10.1177/19476035221093064 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Oungoulian SR, Dumey KM, Jones BK, Ahmad CS, Hung CT, Ateshian GA, 2015. Wear and damage of articular cartilage with friction against orthopedic implant materials. J. Biomech 48, 1957–1964. 10.1016/j.jbiomech.2015.04.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Patel JM, Loebel C, Saleh KS, Wise BC, Bonnevie ED, Miller LM, Carey JL, Burdick JA, Mauck RL, 2021. Stabilization of Damaged Articular Cartilage with Hydrogel-Mediated Reinforcement and Sealing. Adv. Healthc. Mater 10. 10.1002/adhm.202100315 [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Pickard J, Ingham E, Egan J, Fisher J, 1998. Investigation into the effect of proteoglycan molecules on the tribological properties of cartilage joint tissues. Proc. Inst. Mech. Eng. Part H J. Eng. Med 212, 177–182. 10.1243/0954411981533953 [DOI] [PubMed] [Google Scholar]
  55. Radin EL, Swann DA, Paul IL, Mcgrath PJ, 1982. Factors influencing articular cartilage wear in vitro. Arthritis Rheum. 25, 974–980. [DOI] [PubMed] [Google Scholar]
  56. Ulstein S, Årøen A, Røtterud JH, Løken S, Engebretsen L, Heir S, 2014. Microfracture technique versus osteochondral autologous transplantation mosaicplasty in patients with articular chondral lesions of the knee: A prospective randomized trial with long-term follow-up. Knee Surgery, Sport. Traumatol. Arthrosc 10.1007/s00167-014-2843-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Verzijl N, DeGroot J, 2002. Crosslinking by Advanced Glycation End Products Increases the Stiffness of the Collagen Network in Human Articular Cartilage. Arthritis … 46, 114–123. 10.1002/art.10025 [DOI] [PubMed] [Google Scholar]
  58. Williamson AK, Chen AC, Masuda K, Thonar EJ, Sah RL, 2003. Tensile mechanical properties of bovine articular cartilage: variations with growth and relationships to collagen network components. J Orthop Res 21, 872–880. [DOI] [PubMed] [Google Scholar]
  59. Yan D, Zhou G, Zhou X, Liu W, Zhang WJ, Luo X, Zhang L, Jiang T, Cui L, Cao Y, 2009. The impact of low levels of collagen IX and pyridinoline on the mechanical properties of in vitro engineered cartilage. Biomaterials 30,814–821. 10.1016/j.biomaterials.2008.10.042 [DOI] [PubMed] [Google Scholar]

RESOURCES