Migration of Articular Chondrocytes to Fibronectin and Type I Collagen Over a Semi-Porous Membrane

By Ace Indahole and Deuce Iswild

Abstract

This project attempted to find a method that would create an environment in vitro in which articular chondrocytes could migrate. A simple framework of common extra cellular matrix (ECM) proteins provided an environment that allowed articular chondrocytes to migrate over a semi-porous membrane. The matrix is a dynamic structure made of chondrocytes surrounded by fibronectin, collagen, and proteoglycans. Collagen and fibronectin were used as the extra-cellular matrix proteins because they are prevalent proteins in the ECM. Chondrocytes were found to migrate when they were in the presence of fetal bovine serum. The fibronectin and type I collagen created an environment where the chondrocytes could migrate. The protein coating creates a framework of integrins that allows the chondrocytes to migrate. In vivo, chondrocytes do not migrate. Because of this, it is interesting that chondrocytes retain motile characteristics and under certain conditions will migrate even though they are encased in a strict ECM. Fissures in articular cartilage do not heal because the chondrocytes do not migrate naturally to the place of injury. Our research concluded that mature articular chondrocytes can migrate under certain conditions.

Introduction

Articular cartilage is the elastic tissue at the end of long bones that provides low friction, load transferring surfaces. Once damaged, articular cartilage does not heal because chondrocytes in the cartilage do not naturally migrate to repair tissue. This project attempted to establish an in vitro protein framework across which chondrocytes could migrate. The framework used was a lattice of common extra-cellular matrix proteins, including fibronectin and type I collagen.

Hypothesis

The hypothesis for this project was that a simple framework of common extra cellular matrix proteins will provide an environment that would allow articular chondrocytes to migrate in vitro.

Background

Chondrocytes are uniquely found in cartilage. Chondrocytes are imbedded in an organized extracellular matrix (ECM) which is made up of 10-20% by mass type II collagen, 80% by mass water, and a fractional amount of fibronectin, collagen types IX through XI, and proteoglycans. As seen in Figure 1, cartilage is organized into three zones of the ECM. The upper zone is called the superficial tangent, which is the upper 10-20% of the depth of cartilage. The central 40-60% of cartilage depth is called the middle zone. The deep zone makes up the bottom 30% of the matrix (Mankin). The primary purpose of cartilage in the superficial tangent is to form a lubricative barrier between bones that are in contact at joints. This ability of cartilage to serve as a gliding surface between bones is due to its extremely low coefficient of friction (Stark).



In the superficial zone, proteoglycan content is the lowest, and the collagen fibrils are arranged parallel to the surface. In the middle zone of cartilage, chondrocytes are elongated and arranged with their long sides parallel to the surface. The deep zone is the region where proteoglycan concentration is greatest, and the water level is lowest. Here, the chondrocytes are elongated as in the middle zone but are arranged perpendicular to the cartilage surface (Mankin).

Chondrocytes seem to have a regional specificity within articular cartilage. It is unknown whether chondrocytes possess motile characteristics between these regions. Unlike most other cells in the body, chondrocytes do not migrate naturally in vivo in response to injury because they are encased in the restrictive ECM. In specific environments, research has established that chondrocytes can be stimulated to migrate in vitro by allowing chondrocyte receptors (integrins) to bind onto type I collagen and fibronectin. Integrins facilitate this migration. "Integrins are cell surface receptors involved in... adhesion, migration, and matrix assembly." Integrins are composed of pairs of alpha and beta subunits. The combination of alpha and beta pairs determines the specifications for peptide sequences with different matrix proteins. Beta 1 integrin subunits with several different alpha subunits mediate the interaction of fibronectin and type I collagen with chondrocytes that leads to adhesion (Enomoto-Iwamoto).



As seen in Figure 2, after the chondrocytes adhere via these receptors, the chondrocytes spread and form a spindle shape that appears as a long thin cell with two tails. In the process of elongation, the actin filaments in the cytoskeleton of chondrocytes are redistributed to form multiple lamellipodia. The elongated chondrocytes can be stimulated to migrate by introducing different chemotactic and haptotactic factors. When these factors are present, the chondrocytes extend their lamellipodia, which adhere via receptors on neighboring proteins. The chondrocytes move by the cortical tension of the lamellipodia. This process is repeated to allow chondrocytes to migrate (Cytoskeleton).

In osteoarthritis, cracks and fissures can form on the surface of articular cartilage. These cracks greatly reduce the effectiveness of cartilage for many reasons. The cracks increase the coefficient of friction of the cartilage, and, if the fissure is large enough, cartilage can flake off and the bones begin to rub against each other. Under physiologic conditions, cartilage does not seem to heal or regenerate. Cartilage lacks mesenchymal cells, which are instrumental in the standard healing process of other tissues. In other tissues, the process of scarring occurs when mesenchymal cells migrate and fill gaps that have opened due to injury. Unlike other tissues, which heals by scarring, cartilage has no mesenchymal cells that facilitate scarring. ÒIn addition to lacking blood vessels, cartilage lacks undifferentiated cells within the tissue that can migrate, proliferate, and participate in the repair responseÓ (Mankin). While sometimes, fibrocartilage synthesis can fill the defects, but the quality varies greatly from normal articular cartilage (Oegema). Developing a process to promote migration of chondrocytes would seem important in the regeneration of cartilage, because chondrocytes may be more likely to fill gaps of articular cartilage.

Methods

Chondrocyte Isolation: Bovine cartilage slices were placed in Dulbeccos Modified Eagle Medium (DMEM) and weighed. The DMEM was replaced with digestion medium consisting of 100 mL DMEM with 4.5 mg DNAse and 130 mg pronase and incubated for one hour. The first digestion medium was removed and replaced with the second digestion medium consisting of 100 mL DMEM with 4.5 mg of DNAse and 20 mg of collagenase. The cartilage slices were incubated for 20 hours at 37¡C in 5% carbon dioxide and 95% air while on a shaker. Both digestion mediums were sterilized using a 22 micrometer filter.

Culturing Chondrocytes: The cells were removed from the digestion medium using a 70 µL filter to remove all clumps. The cells were then and microfuged at 2500 rpm for 12 minutes to form a pellet of cells. The cells were resuspended using DMEM with 5% fetal bovine serum (FBS). The cells were counted using a hemocytometer and the number of cells calculated by multiplying the number of cells in the grid by 10,000. DMEM was added to dilute the cells to 200,000 cells/mL. The wells of a non-adhesive tissue culture plate were filled to 2 mL using media with FBS. The cells were grown in suspension on a shaker at 37ûC in a 5% carbon dioxide atmosphere. Half of the media was replace every two days.

Alginate Bead Preparation: Cells grown in suspension did not separate but clumped together in large groups that proved to be difficult to separate. Trypsin, collagenase, pronase, and sodium citrate were applied to the clumps of cells in hopes that they would cause the removal of the matrix that had built up around the cultured chondrocytes. These applications were unsuccessful, so a new method of culturing was used.

The cultured cells were suspended in digestion medium #2. The cells were microfuged at 2500 rpm for 12 minutes. The collagenase was removed and the cells were resuspended alginate solution (1.2% alginate in sterile saline solution) at a concentration of 2x106 cells/mL. A sterile six-well plate was filled with 10 mM calcium chloride. Using a 12 mL syringe with a 26G needle, the cells were drawn out of the culture well. A volume of 1.5 mL of the cells were dispersed dropwise in each of the six wells. After the alginate was placed in the solution, the beads with the cells were allowed to incubate for ten minutes, shaking every two minutes. The cells were washed three times with saline solution, two times with DMEM, and once with DMEM with 5% FBS.

Alginate Bead Dissociation: Sodium citrate (16.18 g sodium citrate, 1.86 g disodium EDTA, 8.766 g sodium chloride, and bovine serum albumin (BSA) 5 mg/mL) was used for the chelating the calcium and dissolving the beads to release the cells. The beads were washed with 6 mL of DMEM to remove the FBS. The media was then removed, 2 mL of chelating solution were added to the well. The plate was shaken every two minutes while it incubated for 15 minutes. The cells were then removed and loaded into four microfuge tubes and microfuged for twelve minutes at 2500 rpm. The supernatant was removed, then 650 µL of trypsin EDTA (0.25% by mass trypsin in 10 mM EDTA, pH 7.0) were added. The solution was allowed to sit for five minutes, and 650 µl of soybean trypsin inhibitor were added. The tubes were microfuged at the same settings, and the supernatant was removed. The cells were resuspended in basal media. From the suspension, 40 µL of cells were added to 12 microliter of trypan blue. Ten microliter of this solution were placed on hemocytometer to obtain a cell count. The cells were diluted to 1.2x105 cells/mL. The cells were placed in a sterile six-well plate with 4 mL of DMEM with FBS. Migration of cells isolated immediately after removal from the beads was tested and it was found that cells did not migrate. Since they could be changing their integrins on the cell surface, the cells were isolated and tested on different cultures. The cells were allowed to adhere in monolayer for seven days before being used in the migration assay. This was done because the cells had to establish a basis for the motile abilities needed for migration. The monolayer adhesion was dispersed using 0.25% trypsin EDTA for five to ten minutes. The wells were pooled and dispersed with a 2 mL pipette for 15 to 20 minutes. The trypsin was neutralized by adding 2 mL of STI. This solution was microfuged at 2500 rpms for five minutes. The cells were counted and resuspended at a concentration of 60,000 cells/mL.

Migration of Chondrocytes: To monitor the migration of the cells, a Boyden Chamber was used with porous membrane filters that separated the cells from a test solution. The filters were placed in petri dishes of fibronectin, collagen, RGD peptide, the cell attatchment peptide sequence from fibronectin and FBS, as a positive control, in DMEM and left for 20 hours at 37¡C in solution so that the proteins were bound to the filters and coated them. Different concentrations (80 microgram/mL, 40 microgram/mL, 20 microgram/mL, 10 microgram/mL, 2 microgram/mL) of each protein were used in a separate petri dishes. Then the filters were washed with phosphate buffer solution (PBS) and water. BSA was added to all of the solutions to prevent the chondrocytes from nonspecifically adhering to the plastic. In the lower well of the chamber, 30 microliter of DMEM with 5 mg/mL BSA were placed. In the upper chambers, 50 µL of cells at 60,000 cells/mL diluted in DMEM in 5 mg/mL BSA were added. The experiment was repeated twice. Staining the Filters: The filters were removed from the chambers after 24 hours. The filters were placed on numbered paper clips. The filters were dipped in fixative, and stained with hemotoxlin and eosin, and water for one minute to stain the cells. The filters were placed coated side down on the slides. The tops of the slides were wiped with a cotton swab dipped in PBS to remove cells that did not migrate. The slides were left to dry and then covered using coverslips and immersion oil. The cells were counted with a Geico microscope.

Results

Chondrocytes were found to migrate across a semi-porous matrix in a Boyden Chamber. In order for the chondrocytes to migrate and adhere, common extra cellular matrix proteins fibronectin and type I collagen were used to form a framework. The positive control was 5% fetal calf serum. Cells that were released from the alginate and only treated with trypsin did not migrate through the filters toward the BSA gradient. However, when the cells were allowed to attach to plastic plates, spread and multiply, they migrated in response to both the collagen and fibronectin in a time-dependent manner. It was found that a 5% serum (FBS) produced the best results. Collagen worked better than the fibronectin in the assays. The migration was time dependent and both of the proteins had better results when left in the incubator for 18 hours as opposed to five hours. Since different integrin pairs recognize different peptide sequences within the matrix protein, a control RGD peptide from fibronectin was added to the cell solution to inhibit migration.

Cultures treated with the peptide showed little migration, and both peptide concentrations, 100 and 500 mg/mL, were equally effective in blocking migration. Table 1 shows the cell counts that were taken from different migration assays. The two RGD columns show little migration and demonstrate that RGD peptide inhibits chondrocyte migration. The average number of cells that migrated were 1 and 1.05 respectively for the two concentrations. The cell counts were obtained by counting the number of stained cells in different fields. Serum averaged 46.05 cells migrating per field, while PBS, the control, averaged 17.2 cells per field. The average number of migrated cells for fibronectin was 13.5, while type I collagen averaged 20.1 cells per field.

The pictures in Figures 3-7 show stained (purple) chondrocytes that migrated through the 8 micrometer holes which are visible as white circles. In each figure the chondrocytes have a spindle shape with elongated edges. In Figure 3, a large number of chondrocytes migrated through the filter and were stained purple when type I collagen was used at a concentration of 40 microgram/mL in an assay that lasted for 24 hours. Figure 4 shows chondrocyte migration when type I collagen at a concentration 20 microgram/mL was assayed as the common extracelluar protein for 24 hours. There was a significant decrease in number of migrated chondrocytes between Figure 3 and Figure 4. Figure 5 shows only a small number of chondrocytes migrated when fibronectin 40 microgram/mL was assayed was for 2 hours. Figure 6 shows chondrocyte migration when fibronectin 40 microgram/mL was assayed for 5 hours. After 5 hours, a significant amount of cells migrated through the holes and were stained purple. In Figure 7, fibronectin at a concentration 40 microgram/mL was assayed for 24 hours. This assay shows another increase in the number of migrated chondrocyte from the other assays with identical concentrations but shorter time periods.











Discussion

The migration of chondrocytes that was observed was time dependent, protein dependent, and concentration dependent. The longer that the migration was assayed, the more chondrocytes migrated through the wells. This was not the case for the protein concentration. It was thought that the higher concentration would produce a larger number of cells migrating. Type I collagen was observed to produce a larger number of migrate chondrocytes than fibronectin. When cultured in alginate beads, the chondrocytes maintained a round shape and established a new pericellular matrix. When the matrix was removed with trypsin, the chondrocytes were still unable to migrate. However, when the cells were allowed to spread and form a monolayer, they rapidly migrated in response to serum (the positive control) and to common matrix proteins. This suggests that in articular cartilage, the cells will have to release from the inhibiting matrix before effective healing of cartilage by indigenous chondrocytes can be promoted. Several reports in the literature demonstrate this inhibition using young chondrocytes. These reports have suggested there may also be inhibition of migration by mature chondrocytes, but this is the first report where older mature chondrocytes have shown the same inhibition. ( Enomoto)

Conclusion

The goal of the project was to establish assay conditions necessary for migration through a semi-porous membrane in vitro. A framework of ECM proteins that was established successfully allowed chondrocytes to migrate in vitro. The natural integrins that exist in common ECM proteins provide binding receptors to which the chondrocytes adhere. It is thought that chondrocytes would have to adhere to these integrins for migration to occur. It is particularly interesting that cells that naturally occur in a restrictive matrix should retain the ability to migrate. There are many potential clinical applications for this research on chondrocyte migration. Unlike osteoclasts in bone, chondrocytes are not known to migrate through the ECM. It is believed that they do not migrate because fissures in articular cartilage do not heal but remain gaping cracks in an otherwise smooth surface. The information that chondrocytes have the ability to migrate, even though they do not exercise this ability in vivo, demonstrates that it may be possible to stimulate the process in vivo to heal and regenerate cartilage. If chondrocytes can be stimulated to migrate in vitro, it may be possible for cartilage to be reformed and fissures to be filled by chondrocytes migrating in vivo. This may not be feasible at this point of time, but the future of cartilage healing and research is wide open and important to orthopedics.