Biotribological Testing and Analysis of Articular Cartilage Sliding against Metal for Implants

Summary

This protocol describes the preparation, biotribological testing, and analysis of osteochondral cylinders sliding against metal implant material. Outcome measures included in this protocol are metabolic activity, gene expression and histology.

Abstract

Osteochondral defects in middle-aged patients might be treated with focal metallic implants. First developed for defects in the knee joint, implants are now available for the shoulder, hip, ankle and the first metatarsalphalangeal joint. While providing pain reduction and clinical improvement, progressive degenerative changes of the opposing cartilage are observed in many patients. The mechanisms leading to this damage are not fully understood. This protocol describes a tribological experiment to simulate a metal-on-cartilage pairing and comprehensive analysis of the articular cartilage. Metal implant material is tested against bovine osteochondral cylinders as a model for human articular cartilage. By applying different loads and sliding speeds, physiological loading conditions can be imitated. To provide a comprehensive analysis of the effects on the articular cartilage, histology, metabolic activity and gene expression analysis are described in this protocol. The main advantage of tribological testing is that loading parameters can be adjusted freely to simulate in vivo conditions. Furthermore, different testing solutions might be used to investigate the influence of lubrication or pro-inflammatory agents. By using gene expression analysis for cartilage-specific genes and catabolic genes, early changes in the metabolism of articular chondrocytes in response to mechanical loading might be detected.

Introduction

The treatment of osteochondral defects is demanding and requires surgery in many cases. For focal osteochondral lesions in middle-aged patients, focal metallic implants are a viable option, especially after the failure of primary treatment, like bone marrow stimulation (BMS) or autologous chondrocyte implantation (ACI)¹. Partial surface replacements can be considered salvage procedures that can reduce pain and improve the range of motion². These implants are typically composed of a CoCrMo alloy and are available in different sizes and offset configurations to match the normal anatomy³. While initially developed for defects on the medial femoral condyle in the knee, such implants are now available and in use for the hip, ankle, shoulder, and elbow^4,5,6. For a satisfactory outcome, it is crucial to assess the mechanical joint alignment and condition of the opposing cartilage. Furthermore, correct implantation without protrusion of the implant has been shown to be fundamental⁷.

Clinical studies demonstrated excellent short-term results in terms of pain reduction and improvement of function in middle-aged patients for various locations^5,6,8. Compared with allograft implantation, focal metal implants allow early weight bearing. However, the opposing articular cartilage showed accelerated wear in a considerable number of patients^9,10. Hence, even with proper placement, in many cases degeneration of the native cartilage seems inevitable, while the underlying mechanisms remain unclear. Similar degenerative changes have been observed after bipolar hemiarthroplasty of the hip¹¹ and are increased with activity and loading¹².

Tribological experiments provide the possibility to study such pairings in vitro and simulate different loading situations occurring under physiological conditions¹³. The use of osteochondral pins offers a simple geometry model to investigate the tribology of articular cartilage sliding against native cartilage or any implant material¹⁴ and might further be used in whole joint simulation models¹⁵. Metal-on-cartilage pairings show accelerated cartilage wear, extracellular matrix disruption, and decreased cell viability in the superficial zone compared with a cartilage-on-cartilage pairing¹⁶. Damage to the cartilage occurred mainly in the form of delamination between the superficial and middle zones¹⁷. However, the mechanisms leading to cartilage degeneration are not fully understood. This protocol provides a comprehensive analysis of the biosynthetic activity of articular cartilage. By the determination of metabolic activity and gene expression levels of catabolic genes, early indications for cartilage breakdown might be identified. The advantage of in vitro tribological experiments is that loading parameters can be adjusted to imitate various loading conditions.

Hence, the following protocol is suitable to simulate a metal-on-cartilage pairing, representing an experimental hemiarthroplasty model.

Protocol

1. Preparation of metal cylinders

1. Analyze cylindrical cobalt-chromium-molybdenum (CoCrMo) rods fulfilling the standard specifications for surgical implants for their chemical composition using scanning electron microscopy (SEM) with energy dispersive x-ray spectroscopy per manufacturer’s protocol to confirm provided values.
   NOTE: The elemental composition of the CoCrMo alloy used for this experiment is 65% Co, 28% Cr, 5% Mo and 2% others.
2. Wet grind the samples with silicon carbide grinding paper starting with a grain size of 500. Use grinding paper in increasing order up to a grain size of 4000.
3. Polish the cylinder with 3 µm and 1 µm paste to achieve a surface roughness that is within the tolerance level of surface finish requirements for metallic surgical implants (ISO 5832-12:2019) and total and partial joint replacement implants (ISO 21534:2007).
   NOTE: The average surface roughness is determined using a confocal microscope.
4. Cut CoCrMo rods (Ø of 6 mm) to cylinders with a length of 10 mm.

2. Harvesting of osteochondral cylinders

1. Use bovine stifle joints from skeletally mature animals (aged 18-24 months at the time of sacrifice) and keep them contained and cooled until dissection within 24 h after sacrifice.
   NOTE: Joints are purchased from the local butcher. The joint remains closed until dissection.
2. To harvest cylindrical osteochondral plugs under aseptic conditions, disinfect the knee and perform an arthrotomy and expose the medial femoral condyle.
   NOTE: The dissection has to be performed with caution not to damage the articular surface.
3. Inspect the articular surface for macroscopic damages.
   NOTE: Discard the sample if the cartilage lacks its whitish, smooth and glossy appearance or if there is blistering, fissures or larger defects.
4. Align the cutting tube perpendicular to the articular surface of the weight-bearing area and drive the device into the cartilage and subchondral bone by firm strokes with a hammer. At 15 mm penetration depth, twist the device clockwise with a sudden motion.
5. Remove the device, insert the white knob and screw it in until the bottom end of the osteochondral plug is visible.
6. Mark the anteroposterior orientation of the samples with a sterile marker in order to arrange the osteochondral cylinder accordingly during testing.
   NOTE: The three-dimensional collagen network and its complex architecture facilitate the unique mechanical properties of articular cartilage and should be considered in the orientation of the samples.
7. Rinse the sample with phosphate-buffered saline (PBS) to wash off blood and fat tissue.
8. Repeat the steps mentioned above to harvest the desired number of osteochondral plugs (8 mm diameter, 15 mm length).
   NOTE: Typically, 9 to 12 osteochondral cylinders can be harvested from the weight bearing area on the medial femoral condyle.
9. Place the samples in Dulbecco’s modified Eagle’s medium containing 10% fetal bovine serum, supplemented with antibiotics (penicillin 200 U/mL; streptomycin 0.2 mg/mL) and amphotericin B 2.5 µg/mL and store them at 4 °C until testing to maintain viability.
10. Analyze control osteochondral plugs immediately after harvesting to establish baseline values (see analysis section).

3. Tribological testing

1. Perform the experiments using a commercially available reciprocating tribometer with a cylinder-on-plate configuration. Requirements for the device are vertical loading and adjustable load and sliding speed. Furthermore, a liquid cell enables to perform the tests in a lubricating solution.
2. Determine the contact pressure in the CoCrMo-on-cartilage system using a pressure measurement film. Place the pressure measurement film at the interface and apply static load for 30 s to determine initial contact pressure, contact size and shape. Owing to the convexity of the metal cylinder and the articular cartilage, the initial contact area has an elliptical shape in this configuration.
   NOTE: The pressure measurement film reacts to the applied pressure showing red discoloration of zones where the threshold pressure is reached or exceeded. For 1 N of load, the contact pressure was determined around 2 MPa by visual comparison with defined contact pressures.
3. Fix the osteochondral cylinders on the bottom sample holder with the marking aligned with the sliding direction, and mount the CoCrMo cylinders onto the upper load cell.
4. Add the testing solution (PBS with 3 g/L hyaluronic acid) into the liquid cell that results in submerging the osteochondral cylinder and covering the metal-cartilage sliding interface.
5. Set the testing parameters (prescribed normal force, stroke and sliding speed), which are then applied and maintained throughout the test.
   NOTE: The stroke length of the reciprocating motion must be set according to the contact area to create a migrating contact area (MCA). For plugs that are 8 mm in diameter, a 2 mm stroke allows adequate rehydration of the cartilage.
6. Start reciprocal sliding of the CoCrMo cylinder against the articular cartilage immersed in the lubricating solution with the set loading parameters.
7. Monitor the coefficient of friction (COF) during the experiments.
   NOTE: The COF is assessed automatically but can be calculated using the equation μ=F/W (μ - coefficient of friction; F - frictional force; W - normal load applied by the system).
8. Terminate the experiment after the desired testing period.
9. Remove the osteochondral plug from the sample holder, rinse it with PBS and store it in medium until further biological analysis (see below).
10. Submerge control samples in the testing solution at room temperature for the duration of the test and analyze together with samples that have been exposed to mechanical loading.

4. Analysis

NOTE: Osteochondral cylinder are analyzed for metabolic activity and gene expression to investigate biological activity; histology is performed to study cartilage surface integrity and the underlying matrix.

1. Histology
   1. For histological analysis, immerse the osteochondral plugs in 4% buffered formaldehyde solution at room temperature until further processing.
   2. Rinse the samples with PBS and place them into a plastic vessel.
   3. Add an excess of the ready-to-use decalcifier-solution so that all samples are covered.
   4. Apply constant agitation for 4 weeks for complete decalcification.
   5. After decalcification, embed the samples in water soluble glycols and resins and store them at −80 °C.
   6. Obtain 6 µm sections by cryosectioning transversal to the contact area.
   7. Subsequently, prepare the samples for Safranin O staining and Fastgreen counterstaining using a manufacturer’s protocol.
   8. Capture histological images using a microscope and process using imaging processing software.
2. Metabolic activity
   NOTE: The metabolic activity of chondrocytes in the articular cartilage are investigated with an XTT-based ex vivo toxicology assay.
   1. Rinse the osteochondral plug using PBS and place the sample in a Petri dish.
   2. Place a 24-well plate on a scale and zero the scale.
   3. Cut off the cartilage from the osteochondral graft with a scalpel in one piece.
   4. Bisect the cartilage in two equal pieces so that the contact area is equally distributed onto both cartilage pieces and mince one half to 1 mm³ pieces. The second half is used for gene expression analysis.
   5. Transfer the minced cartilage into one well of the prepared 24-well plate and determine the tissue weight.
   6. Repeat the steps mentioned above for each sample and add 1 mL of growth medium to each well of the plate.
   7. Add the XTT solution (490 µL of XTT labelling reagent and 10 µL of activation reagent) according to the manufacturer’s instruction and mix.
   8. Incubate the plate at 37 °C and 5% CO₂ for 4 h.
   9. After incubation, remove the supernatant and transfer it to a 5 mL tube.
   10. Extract the tetrazolium product by adding 0.5 mL of dimethyl sulfoxide (DMSO) to the cartilage tissue in the 24-well plate and apply continuous agitation for 1 h at room temperature.
   11. Remove the DMSO solution and pool it with the previously collected XTT solution.
   12. Transfer 100 µL of the sample in triplicates in a 96-well plate on a plate reader and measure the absorbance at a wavelength of 492 nm and a reference wavelength at 690 nm.
   13. Normalize the resulting absorbance values to the wet weight of each sample and perform analysis using software.
3. Gene expression analysis
   1. RNA isolation
      NOTE: RNA isolation is carried out using a commercial kit (Table of Materials) according to the instructions provided by the manufacturer with small amendments.
      1. Mince the second half of the cartilage tissue obtained from the osteochondral plug into small pieces.
      2. Transfer them to a tube containing ceramic beads and 300 µL of Lysis Buffer (containing 1% β-mercaptoethanol).
         NOTE: The samples can be frozen in liquid nitrogen until further processing.
      3. Thaw the samples for 2 min and use the commercial lyser for homogenization of the tissue. Apply 6500 rpm for 20 s (homogenization step) four times with a 2 min cooling phase after each run (at 4 °C using the commercial lyser cooling device) to fully disrupt the tissue.
      4. Add 20 µL of proteinase K and 580 µL of RNase-free water to each tube and incubate them at 55 °C for 30 min.
      5. Centrifuge the samples for 3 min at 10,000 x g and transfer the supernatant to a 1.5 mL tube.
      6. Add 0.5 volumes of 90% ethanol to each tube and mix.
      7. Transfer 700 µL of the sample to an RNA binding column placed in a 2 mL collection tube and centrifuge at 8,000 x g for 15 s.
      8. Discard the flow-through and repeat the centrifugation step for the complete lysate.
      9. Add 350 µL of Buffer RW1 to the column, centrifuge at 8,000 x g for 15 s, and discard the flow-through.
      10. Mix 10 µL of DNase stock solution and 70 µL of Buffer RDD. Add the solution to the RNA purification membrane and incubate it at room temperature for 15 min.
      11. Add 350 µL of Buffer RW1 to the column and centrifuge at 8,000 x g for 15 s. Discard the flow-through.
      12. Add 500 µL of Buffer RPE and centrifuge at 8,000 x g for 15 s. Discard the flow-through.
      13. Add 500 µL of Buffer RPE to the RNA purification column and centrifuge at 8,000 x g for 2 min.
      14. Place the column in a 1.5 mL collection tube and add 30 µL of RNase-free water. Centrifuge at 8,000 x g for 1 min.
      15. Store the isolated RNA at -80 °C until cDNA synthesis.
   2. cDNA synthesis
      NOTE: To synthesize complementary DNA (cDNA) from messenger RNA (mRNA) a commercial kit (Table of Materials) was used. RNA from bacteriophage MS2 was added to stabilize isolated RNA during cDNA synthesis.
      1. Thaw and mix the reagents. The composition for a single reaction is shown in Table 1.
      2. Add 16 µL of RNA sample to the volume for a single reaction (14 µL).
      3. Perform cDNA synthesis in a thermal cycler using the following parameters: 10 min at 25 °C (primer annealing), 60 min at 50 °C (DNA synthesis), 5 min at 85 °C (denaturation) and 5 min at 20 °C (cooling phase).
      4. Store cDNA at -20 °C until real-time quantitative polymerase chain reaction (RT-qPCR).
   3. RT-qPCR
      NOTE: For RT-qPCR of bovine samples, primers and probes were designed by using commercial Real-Time qPCR software (e.g., IDT) for the genes GAPDH (Glyceraldehyde 3-phosphate dehydrogenase), COL2A1 (Collagen type 2), ACAN (Aggrecan), COL1A1 (Collagen type 1), MMP-1 (Matrix Metalloproteinase-1), and MMP-13 (Matrix Metalloproteinase-13). Bovine primers and double quenched probes were provided by IDT. The reagents used for a single reaction to evaluate the efficiency and gene expression are displayed in Table 2.
      1. Dispense the master mix of a single reaction (9 µL) to each well of a 96-well PCR plate and add 1 µL of cDNA to each reaction. Perform tests for each sample in triplicates.
      2. Close the PCR plate using sealing oil and centrifuge at 877 x g for 10 min at 4 °C.
      3. Perform RT-qPCR using a precision thermal cycler with the following protocol: 95 °C for 10 min, 45 cycles of amplification (95 °C for 10 s, annealing for 30 s, cDNA synthesis), and 37 °C for 30 s.
         NOTE: Specific annealing temperatures are required for each primer.
      4. Use GAPDH along with the target genes to confirm efficiency.
      5. Use the provided software to calculate the efficiency of each gene.
      6. Normalize the cycle threshold (CT) values to the expression of the reference gene GAPDH and use the ΔΔCT method for quantification.

Representative Results

The contact area and contact pressure must be confirmed using a pressure measurement film (Figure 1). Physiological loading condition can be confirmed by comparing with reference imprints for defined contact pressures. During testing, the coefficient of friction is monitored constantly. With a migrating contact area, a low friction coefficient can be maintained for at least 1 h (Figure 2). Using Safranin O staining the extracellular matrix composition and structure can be determined (Figure 3). The intensity of Safranin O staining is proportional to the proteoglycan content. Fast Green counterstains the non-collagen sites and provides a clear contrast to the Safranin O staining. The proteoglycan content varies over the articular surface but should be uniform throughout the tissue section in baseline samples (Figure 3A). Control samples submerged in the testing solution show extraction of GAGs, which can be counteracted by mechanical loading (Figure 3B, 3C). Metabolic activity of the bovine articular chondrocytes is independent of the harvesting site, but shows an increase with mechanical loading compared with unloaded controls (Figure 4). The gene expression levels of cartilage-specific genes (COL2A1, ACAN) increase with physiological loading conditions, whereas catabolic genes (COL1A1 and MMP13) are upregulated with stationary contact area (Figure 5).

	Volume (µl)
Transcriptor RT Reactions Buffer 5x conc.	6
Protector RNase Inhibitor 40U/µl	0.75
Deoxynucleotide Mix 10 mM each	3
Random Hexamer Primer 600 µM	3
Transcriptor Reverse Transcriptase 20 U/µl	0.75
MS2 RNA (0,8 µg/µl)	0.375
Nuclease free distilled water	0.125
Total volume	14

Table 1: Reagents for a single reaction for cDNA synthesis.

	Volume (µl)
FastStart Probe Master 2X	5
Hydrolysis Probe 2,5 µM	1
Left PrimerGAPDH 5 µM
Right Primer GAPDH 5 µM
Nuclease free distilled water	3
Total Master Mix	9

Table 2: Reagents for the Master Mix for a single PCR.

Figure 1: Pressure measurement of the initial contact area at the metal-cartilage interface before testing. Due to the convexity of the metal cylinder and the articular surface and its elastic properties, the initial contact area is elliptical. During sliding, this initial contact area moves with a stroke of 2 mm, resulting in a larger area that is exposed to mechanical loading; scale bar = 2 mm. Please click here to view a larger version of this figure.

Figure 2: Time-depended coefficient of friction (duration 1 hour) for seven samples tested at 8 mm/s sliding speed and 1 N load (2 MPa contact pressure). Each colored line represents the COF of one osteochondral cylinder. The observed variability is within the limits for biological samples. Please click here to view a larger version of this figure.

Figure 3: Histological cross-sections of bovine osteochondral samples stained with Safranin-O and Fast Green. (A) Baseline samples show high GAG content throughout the articular cartilage. (B) Control sample submerged in testing solution without mechanical loading show less Safranin-O staining in the middle zone, indication an extraction of proteoglycans. (C) Tested samples show higher GAG content compared with controls, indicating mechanical stimulation; scale bar = 250 µm Please click here to view a larger version of this figure.

Figure 4: Metabolic activity of bovine articular chondrocytes after tribological testing with different loading variations and controls. The horizontal dotted line represents baseline levels. The nonparametric Kruskal–Wallis test was performed for comparison between testing groups followed by Dunn’s post hoc test in case of significance. *p < 0.05. This figure has been modified from Stotter et al.¹⁸. Please click here to view a larger version of this figure.

Figure 5: Gene expression of cartilage-specific genes after tribological testing with different loading conditions and controls. COL2A1=collagen type 2; ACAN=aggrecan; COL1A1= collagen type 1; MMP13= matrix metalloproteinase 13. The expression levels were normalized to the housekeeping gene GAPDH (glyceraldehyde 3-phosphate dehydrogenase). The horizontal dotted lines represent baseline expression levels. The nonparametric Kruskal–Wallis test was performed for comparison between testing groups followed by Dunn’s post hoc test in case of significance. *p < 0.05. This figure has been modified from Stotter et al.¹⁸. Please click here to view a larger version of this figure.

Discussion

Focal metallic implants represent a salvage procedure for osteochondral defects, especially in middle-aged patients and after failed primary treatment. Although clinical studies demonstrated promising short-term results, one observed complication is damage to the opposing, native cartilage¹⁰. Cadaver and biomechanical studies show clear evidence that proper implantation with flat or slightly recessed positioning maintains natural contact pressures¹⁹. Tribological experiments provide a possibility to test various cartilage pairings in vitro. In such, loading conditions, lubrication, material pairings and duration might be adjusted as desired.

Bovine cartilage is available in high quantity at the local abattoir. The cellularity and the zonal structure are very similar to the human femoral condyles²⁰. However, proteoglycan content is site-specific, whereas gene expression levels have been shown to be uniform over the articular surface. In this protocol, osteochondral plugs were harvested from the weight bearing area. Cartilage thickness, collagen architecture and resulting tribological properties show regional differences over the articular surface¹⁶. The limitation of using osteochondral plugs in an unconfined loading setup with a disrupted collagen network and changed fluid pressurization compared with whole joint models need to be considered.

In the majority of tribological studies, PBS alone is used as testing solution to generate more robust data. PBS is a buffer solution with isotonic osmolarity and helps to maintain a constant pH during biological experiments. Using PBS with hyaluronic acid provides boundary lubrication and reduced friction²¹. Accordingly, synovial fluid reduces the friction coefficient and improves fluid pressurization compared with saline²². The friction coefficient depends on various system properties, demonstrated by the classic Stribeck Curve. The Stribeck Curve relates the friction coefficient and viscosity, speed and load and presents the basic lubrication regimes: boundary, mixed, and hydrodynamic lubrication. Boundary lubrication can be obtained with PBS alone as lubricating fluid, but loading parameters would need to be adjusted accordingly. The COF delivered from the tests are average values over the stroke. Thus, it can be assumed that different lubrication conditions occur during the cycle. During standstill at reversal position, boundary conditions might be prevailing, while mixed lubrication might be predominant during sliding. Based on absolute duration during the sliding cycle, the latter would have had more influence on the mean COF value.

To investigate physiological conditions occurring in joints during daily activities, loading conditions can be adjusted accordingly in the tribometer software. Pressure sensitive measurements should be used to confirm the desired contact pressures. Reported femorotibial contact pressures range between 1 MPa during standing and up to 10 MPa during downhill running²³. With a focal resurfacing, implant pressures are just slightly elevated compared with healthy joints²⁴. Reported relative sliding velocities during the gait cycle are reported up to 100 mm/s with high variations during the different phases. This means that relative joint movements exceed the velocities that can be applied in this tribological setup. To mimic natural kinematic conditions and contact pressures in healthy knee joints, loading conditions range from 1 to 10 MPa contact pressure and 5 to 100 mm/s sliding speed. However, while high loads can be applied in this experimental setup, the range of sliding velocities is limited. Pathological loading conditions, both overload and inadequate loads, might also be simulated. Low sliding velocities or static loading equate immobilization, while higher loads represent nonphysiological mechanical stimulation.

As enzymatic digestion can affect the expression of cartilage-specific genes, a nonenzymatic tissue homogenization is described in this protocol. During cDNA synthesis, in addition to the instructions, RNA from bacteriophage MS2 is added for stabilization purposes. Gene expression levels, but not proteins, were analyzed to detect early changes in the biosynthetic activity of articular chondrocytes. In addition to histological sections and metabolic activity, these assays provide comprehensive information on the effects of mechanical loading on articular cartilage.

Materials

Name	Company	Catalog Number	Comments
Amphotericin B	Sigma?Aldrich Chemie GmbH	A-2942-100ML
buffered formaldehyde solution 4%	VWR	97131000
Cell Proliferation Kit II (XTT)	Roche Diagnostics	11465015001	XTT-based ex vivo toxicology assay
CoCrMo raw material	Acnis International		CoCrMo rods 6mm in diameter
CryoStar NX70 Cryostat	Thermo Fischer Scientific		cryosectioning device
dimethyl sulfoxide (DMSO)	Sidma-Aldrich Chemie	D 2438-10ML
Dulbecco’s modified Eagle’s medium	Sigma?Aldrich Chemie GmbH		medium
fetal bovine serum	Gibco
Hyaluronic acid	Anika Therapeutics Inc.		component of lubricating solution
iCycler	BioRad		thermal cycler
Leica microscope DM?1000	Leica		microscope for histology
LightCycler 480 Sealing Foil	Roche Diagnostics
LightCycler 96	Roche Diagnostics		thermal cycler for PCR
MagNA Lyser Green Beads	Roche Diagnostics	3358941001
Osteochondral Autograft Transfer System (OATS)	Arthrex Inc.		cutting tube for harvesting osteochondral cylinders
osteosoft	Merck	1017279010	decalcifier-solution
Penicillin /Streptomycin	Sigma?Aldrich Chemie GmbH	P4333-100ML
phosphate?buffered saline	Sigma?Aldrich Chemie GmbH		PBS
Prescale Low Pressure	Fujifilm		pressure indicating film
RNeasy Fibrous Tissue Kit	QIAGEN	74404
Synergy 2	BioTek Instruments		plate reader
Tetra?Falex MUST	Falex Tribology		Tribometer
Tissue? Tek O.C.T.	SAKURA	4583	embedding formulation
Transcriptor First Strand cDNA Synthesis Kit	Roche Diagnostics	40897030001
β-mercaptoethanol	Sidma-Aldrich Chemie	M3148