Abstract
Purpose:
The purpose of this study was to characterize the influence of SDF-1 on cell migration/adhesion and temporal gene expression of human cartilage mesenchymal progenitor cells (C-PCs); and to utilize SDF-1 conditioned mesenchymal progenitors to stimulate reintegration of human meniscus fibrocartilage breaks.
Materials and Methods:
Characterization of SDF-1-induced cell migration was achieved using hydroxypropyl cellulose (HPC) scaffolds pretreated with SDF-1. Fluorescence microscopy and cell counting were used to visualize and quantify the extent of cell migration into scaffolds, respectively. Relative mRNA expression analysis was used to characterize the temporal effects of SDF-1 on C-PCs. Tissue reintegration experiments were conducted using cylindrical human meniscal tissue punches, which were then placed back together with an HPC scaffold embedded with C-PCs. Tensile testing was used to evaluate the extent of tissue reintegration stimulated by human mesenchymal progenitors.
Results:
C-PCs migrate into scaffolds in response to SDF-1 with the same efficiency as mesenchymal progenitors from human marrow (BM-MSCs). SDF-1 treatment of C-PCs did not significantly alter the expression of early and late stage chondrogenic differentiation genes. Scaffolds containing SDF-1 preconditioned C-PCs successfully adhered to fibrocartilage breaks and migrated from the scaffold into the tissue. Tensile testing demonstrated that SDF-1 preconditioned C-PCs stimulate reintegration of fibrocartilage tears.
Conclusion:
C-PCs migrate in response to SDF-1. Exposure to SDF-1 does not significantly alter the unique mRNA profile of C-PCs that make them desirable for cartilaginous tissue repair applications. SDF-1 pretreated mesenchymal progenitors successfully disperse into injured tissues to help facilitate tissue reintegration.
Keywords: knee meniscus, cartilage progenitor, SDF-1, fibrocartilage, HPC scaffold
INTRODUCTION
The inadequate healing capabilities of articular cartilage and meniscal fibrocartilage makes stimulating injury repair a clinical challenge1. Poor healing capacity has been attributed to the inherent hypocellularity and lack of blood supply that is characteristic of these tissues2,3. Additionally, meniscus contains dense collagen networks that need to be rebuilt/replenished for healing to be achieved. Cell-mediated tissue repair/engineering strategies offer a potential solution to this classic challenge. Recent pre-clinical studies have used both ex-vivo and in-vivo systems to demonstrate healing potential of various mesenchymal stem cell populations4,5. Concurrent studies are focused on identifying the best possible combination of cells, growth factors and/or scaffolds required to stimulate healing following a sustained traumatic injury6–8. Tissue engineering innovations in biomaterial scaffolds and treatment with cellular regenerative techniques have been recently gaining more attention in the field of musculoskeletal medicine9–11. It has been previously reported that cellulose-based scaffolds are biocompatible and possess attractive features including cell adhesion capabilities and increased tensile strength for tissue engineering of cartilage12,13.
Our laboratory has focused on the utilization of cartilage-derived mesenchymal progenitor cells (C-PCs) for meniscus and cartilage tissue repair because they are highly proliferative14 and exhibit low basal expression of type X collagen (COL10)6,15, which is a marker of cell hypertrophy that is reported to be upregulated in cartilage during osteoarthritis pathogenesis16–18. Although it has been hypothesized that C-PCs are ideal for cartilaginous tissue repair and regeneration applications15, their efficacy has not been previously tested in human tissues. Our laboratory recently reported the successful utilization of C-PCs to stimulate reintegration of meniscus fibrocartilage tears in a rat meniscus tissue explant model6. Moreover, we demonstrated that C-PCs, like marrow derived mesenchymal stem cells (BM-MSCs), will migrate to the tissue in response to Stromal Cell-Derived Factor 1 (SDF-1). Veritably, we also found that inhibiting SDF-1 cell signaling by blocking the SDF-1 receptor (CXCR4) noticeably hindered the ability of these cells to stimulate fibrocartilage tissue repair6.
In light of these findings, our long-term goal is to develop strategies for the effective utilization of C-PCs in human cartilaginous tissue repair and tissue engineering applications. We hypothesized that SDF-1 preconditioned hydroxypropyl cellulose (HPC) scaffolds and collagen-coated HPC scaffolds can harbor and distribute C-PCs to regions of human meniscus tissue injury, using an explant culture model. Our results demonstrate that SDF-1 does not significantly alter the expression profile of important early and late stage chondrogenesis genes in C-PCs. Lastly, we demonstrate that meniscus-scaffold constructs that contained C-PCs seeded into an HPC scaffold can improve the tensile integrity at the meniscus-scaffold interface.
MATERIALS AND METHODS
Cells and in-vitro culture conditions.
The previously established human cartilage-derived mesenchymal progenitor cell line (CPCLA1)6 was used to perform all C-PC experiments in this study. A stable human BM-MSC cell line was generated using the pRetro-E2 SV40 construct (Applied Biomaterials Inc., Richmond, BC, Canada), according to the manufacturer’s standard protocol, and used in order to complete all experiments involving BM-MSCs. Cells were grown and maintained in Dulbecco’s modified Eagle’s medium (DMEM) with 10% FBS, 1% Pen Strep,100 mM HEPES, 2 mM L-glutamine, 0.1 mM ascorbic acid,0.1 mM sodium pyruvate, and 2.7μM L-glucose (DMEM++). Chondrogenesis induction in culture was achieved using Stempro ® Chondrocyte Differentiation Media (Life technologies, Grand Island, NY, USA).
Acquisition and use of human tissues for explant culture models.
Human meniscus tissue was obtained from total knee replacement surgeries from patients (age range 61 – 85 years old) with approval from the Rhode Island Hospital Institutional Review Board (IRB). Menisci were brought to the laboratory within an hour of removal and washed thoroughly in 1×Hanks’ balanced saline solution (HBSS). For the explant repair experiments, sterile disposable 3–4mm skin biopsy punches (Acu Punch) were used to produce several meniscus plugs in the red-white zone of each sample. Plugs were cut in half with a surgical blade to produce cylindrical plugs of equal size with a flat surface. CelluSponge three-dimensional porous hydroxypropyl cellulose (HPC) and Cellusponge-Col (HPC coated in highly purified Type I collagen) scaffolds (Sigma–Aldrich) were punched into 3–4mm discs. In the experiments involving SDF-1, scaffolds were pre-treated with SDF-1 recombinant protein (0.1 μg/mL; PeproTech Inc.) for 24 hours. Scaffolds were inserted between two meniscal plugs and anchored together with a sterilized stainless-steel pin through the center. Constructs were treated with 1.0×105 C-PCs (delivered in 30μL of DMEM++) fluorescently labeled red with Vibrant Dil (Thermo Fisher, Waltham MA). After 14 days in culture, stainless-steel pins were carefully removed and samples were prepped for imaging and biomechanical testing. Images were taken using a Nikon Eclipse TS2 microscope.
Scaffold infiltration experiments.
C-PCs and BM-MSCs were seeded (5.0×104 cells/well) into 12-well plates and cultured in DMEM++ for 24 hours. CelluSponge three-dimensional porous hydroxypropyl cellulose (HPC) and Collagen Type 1 coated HPC scaffolds (Sigma–Aldrich) were punched into 3–4mm discs with sterile disposable skin biopsy punches (Acu Punch). In sample groups involving SDF-1, scaffolds were pre-treated with SDF-1 recombinant protein (0.1 μg/mL; PeproTech Inc.) for 24 hours. Scaffolds were placed in the center of the cell monolayer and 400μl of DMEM++ was added to submerge the scaffold without floating. Scaffolds were cultured for 3 days in a 37°C cell incubator with media changes twice daily. Scaffolds were removed from wells and placed in a clean empty dish for imaging. Scaffolds were trypsinized with Trypsin-LE and live cell numbers were quantified by Trypan Blue (Sigma-Aldrich) using a hemocytometer. Images were taken using a Nikon Eclipse TS2 microscope.
Flow cytometry for cell surface CXCR4 expression.
C-PCs and BM-MSCs were stained with the pre-conjugated antibody CXCR4(APC) and IgG(APC) isotype control antibody (Miltenyi Biotec Inc., San Diego, CA, USA). Cells (1.0×105) were washed in 5mL of 1xPBS, centrifuged at 300xg and resuspended in 100 μl of 1xPBS. The pre-conjugated antibody (10 μl) was added, mixed, and incubated with the cells in the dark at 4°C for 10 minutes. Excess antibody was washed off with 1.0 mL of 1xPBS. Stained cells were resuspended in 300 μl of buffer and analyzed using Accuri C6 Flow Cytometer (BD Biosciences, San Jose, CA, USA).
MTT assay.
Collagen Type 1 coated HPC scaffolds (Sigma–Aldrich) were pretreated with SDF-1 protein (Peprotech) (0.1 μg/mL) for 24 hours. Cells (5.0×104) were plated for 24 hours and then the pretreated scaffold with SDF-1 were placed on it and returned to the incubator. At day 6 the scaffold was placed in 96-well plate and the manufacture’s protocol for Vybrant MTT Assay kit (ThermoScientific) was followed. The absorbance was read at 540 nm using a plate reader. A negative control with just MTT dye and DMSO was included in the experiment.
ELISA for SDF-1 protein release into culture medium from scaffolds.
Scaffolds pretreated with 0.1μg/mL of SDF-1 recombinant protein (for 24 hours) were placed in cells containing 400 μl of DMEM++. Medium was collected and replaced once a day for 4 days. At the end of the experiment, the media was analyzed for SDF-1 release into the culture at each day by SDF-1α Human ELISA kit (EHCXCL12A, ThermoScientific) according to the manufacture’s protocol.
GAG assay.
Cells (1.0×105) were cultured in a 12-well plate containing 400μl of chondrogenesis medium in each well. Chondrogenesis was induced both in the presence and absence of recombinant SDF-1 protein (0.1 μg/mL). Media was collected at the end of 5 days after chondrogenesis induction. Glycosaminoclycan (GAG) concentrations in the culture medium was determined using a standard 1,9-dimethylmethylene blue (DMMB) assay. Accordingly, 20 μl of culture medium was combined with 200μl of DMMB dye solution (pH 3.0) and the absorbance (525nm) of the mixture was read using a plate reader. A standard curve of serially diluted chondroitin sulfate was used to extrapolate absorbance values to determine actual GAG concentration (ng/mL). DNA from the cell monolayer was collected and quantified at the end of the 5-day culture period. These values were used to normalize the amount of GAG released into the culture medium based on DNA content.
Gene expression analysis.
Real-time quantitative polymerase chain reaction (RT-qPCR) was used to quantify mRNA expression levels. Total mRNA was isolated from tissue and/or cells via RNAqueous Kit (Thermo Fisher, Waltham MA) according to manufacturer. Messenger RNA was reverse transcribed into cDNA using iScript cDNA Synthesis Kit (Bio-Rad, Hercules, CA) according to the manufacturer. Messenger RNA levels were calculated using the delta Ct (ΔΔCt) method and normalized to a house-keeping gene (ribosomal RNA beta-actin). X=2−ΔΔCt, in which ΔΔCt = (Ct Exp target gene−Ct Exp house-keeping gene) − (Ct Ctl target gene−Ct Ctl house-keeping gene) and X= relative transcript; Ct Exp= Ct of experimental group, Ct Ctl= Ct of control group. Primers include COL1, COL2, SOX9, COL10, RUNX2, MMP13, and CXCR4.
3D Design Engineering and Printing of Adapters.
Meniscus-Scaffold culture Discs:
Cylindrical discs were designed and developed using Google SketchUp 2017 CAD software, sliced and prepared in Ultimaker Cura 4.1.0 software, and were printed out of Polylactic acid (PLA) material at 100% infill density on an open-source FDM Anet A8 3D printer. A stainless-steel pin was passed through to compress/secure the meniscus and scaffold pieces together when assembling in the plug-scaffold-plug construct.
Meniscus-Scaffold Platform Adapters:
Tissue-specific platform adapters were designed and developed using Google SketchUp 2017 CAD software, sliced and prepared in Ultimaker Cura 4.1.0 software, and were printed out of Polylactic acid (PLA) material at 100% infill density on an open-source FDM Anet A8 3D printer. Each adapter was composed of two segments that were conjoined; one being a flat rectangular platform to support the meniscus, and the other segment being a housing for a #6–32 nut that specifically matches the thread pitch on the Bose ElectroForce 3200 Series 1000g load cell (Eden Prairie MN). The two symmetrical adapters were designed in order to vertically and accurately align the sample in the machine without experiencing opposing torsional force.
Biomechanical tensile force testing.
After removal of the stainless-steel pin, the outer surface of the meniscus-scaffold construct was superglued to the flat-surfaced rectangular platform of the adapter. Once the glue has dried (30 seconds), the first platform adapter was carefully attached to the Bose ElectroForce 3200 Series material testing machine (Eden Prairie MN) with the construct hanging upside down. An identical platform adapter was secured to the opposite junction 1000g load cell. A drop of superglue was placed on the identical platform and slowly raised upward until it met the bottom outer surface of the meniscus-scaffold construct. Once dried (30 seconds), this effectively secured the top and bottom of the construct to the platform adapters which can be pulled apart uniformly, transmitting the forces to the scaffold interfaces during tensile testing. Biomechanical data were analyzed with WinTest 7 software (Eden Prairie MN). The construct failure load was identified at the load at which there was a sudden decrease from the highest peak of load from the load-displacement curves.
Statistical analysis.
Statistical analysis using a Student’s t test was performed on experiments containing two groups, or experiments that compared each experimental group to a single control group. N≥3 for all experiments. A one-way ANOVA was used to analyze data with three or more groups. Error bars illustrate ±1 standard deviation (SD) of the mean. P-values less than or equal to 0.05 were considered statistically significant.
RESULTS
SDF-1 enhances infiltration of C-PCs into scaffolds.
We have previously demonstrated that C-PCs stimulate reintegration and bridging of small rat meniscus tissue tears in a manner that depends on SDF-1/CXCR4 pathway activity6. In this present study, we sought to develop a strategy to precisely mobilize C-PCs to injury sites in a human meniscus tissue injury model. To do this, we needed an effective way to concentrate C-PCs to an area of injury. Our idea was to take advantage of C-PC responsiveness to SDF-1, in order to help mobilize them to the desired locations. First, we tested whether SDF-1 pretreated HPC scaffolds can effectively be used as cell reservoirs, which ideally could then be placed in the injury site to help concentrate migrating C-PCs to this region. Our analysis of these HPC scaffolds demonstrated that they can slowly release SDF-1 recombinant protein over a 48-hour culture period (Supplemental Fig 1A).
To conduct our migration experiments, we utilized a clonal C-PC cell line (CPCLA1) that we generated and thoroughly characterized in our previous study6. An SDF-1 pretreated HPC scaffold was placed over 5.0×104 cells that were seeded in a 12-well plate. Cells were left to migrate/infiltrate into the scaffold for 72 hours, at which point the scaffolds were removed and all adherent cells were detached, and viable cells were stained with trypan blue and counted. Results indicated that SDF-1 pretreated HPC scaffolds held more than twice the number of cells held by untreated HPC scaffold controls (Fig 1A; 1B). The infiltration efficiency of CPCLA1 into SDF-1 pretreated scaffolds was comparable to that of BM-MSCs (Fig 1B), which are known to be responsive to SDF-1/CXCR4 axis signaling. We repeated the experiment using collagen I coated HPC scaffolds (Fig 1C; 1D). This was done because collagen I is a major component of meniscus tissue and we wanted to determine whether its presence would significantly alter cell infiltration into the HPC scaffold. However, our findings demonstrate that collagen coating does not significantly alter the number of infiltrating cells. In both experimental models, SDF-1 pretreatment stimulated the infiltration of approximately 12,000 BM-MSCs (or approximately 8,500 C-PCs) into the scaffold, whereas less than half that quantity of both respective cell types infiltrated into scaffolds that were not preconditioned with SDF-1 (Fig 1B; 1D). To further support these findings, we also performed an MTT assay to quantify viable CPCs that migrated into collagen I coated HPC scaffolds in the presence and absence of SDF-1 pretreatment of the scaffold (Supplemental Fig 1B). These findings were consistent.
Figure 1. SDF-1 preconditioning facilitates the infiltration of C-PC into HPC scaffolds.
(A) Live fluorescent imaging of cell infiltration into HPC scaffolds. BM-MSCs and C-PCs were stained with fluorescent red Vibrant Dil dye and plated (1.0×105 cells per well). After the cells had fully adhered to the plate, SDF-1 preconditioned and non-preconditioned (No SDF-1) HPC scaffolds were placed in the well and the cells were left to migrate up into the scaffold. After 3 days in culture, scaffolds were removed and imaged in a clean well. Fluorescent red channel only (left), and merged red channel with bright field (right). (B) The number of viable cells that had infiltrated into each HPC scaffold was quantified and compared. (C) Same experiment as panel A but using HPC scaffolds coated with Collagen I. (D) The number of viable cells that had infiltrated into each collagen I coated HPC scaffold was quantified and compared. (E) Flow cytometry graphs representing BM-MSCs and CPCs that express CXCR4 on their cell surface. (F) Histogram of cell percentages that express CXCR4 on the cell surface in BM-MSCs and C-PCs. N ≥ 3, ** P ≤ 0.01. Error bars represent ±1 SD of the mean.
We next quantified the cell surface protein expression of the SDF-1 receptor CXCR4 in both cultured BM-MSCs and C-PCs. Interestingly, CXCR4 was only expressed on the cell surface of approximately 11% of cultured C-PCs and 17% of cultured BM-MSCs (Fig 1E; 1F). However, these data taken together demonstrate that both BM-MSCs and C-PCs successfully migrate in response to SDF-1 in culture. Overall, our results demonstrate that SDF-1 pretreatment enhances the integration of C-PCs and BM-MSCs into HPC scaffolds.
SDF-1 treatment does not diminish the qualities of C-PCs that make them desirable for cartilaginous tissue repair applications.
Since our long-term goal is to develop strategies to use C-PCs for effective cartilaginous tissue repair and tissue engineering, we wanted to examine the effect of SDF-1 on C-PCs to test whether it will alter fundamental characteristics of C-PCs that make them desirable for this intended purpose. To do this, we performed mRNA analysis on cells that had successfully migrated and integrated into collagen I coated HPC scaffolds that were preconditioned with SDF-1. Once again, we opted to use a collagen coated scaffold for this experiment because this condition mimics the microenvironment of the human meniscus, which predominantly consists of type I collagen. Gene expression analysis was used to detect relative fold changes in early chondrocyte differentiation genes (i.e. Collagen I, Collagen II, SOX9), late-stage chondrocyte differentiation and hypertrophy genes (i.e. Collagen X, RUNX2, MMP13), and the SDF-1 receptor CXCR4 (Fig 2). Results indicated that SDF-1 preconditioning only altered the expression of SOX9, in BM-MSCs (Fig 2A). All other tested markers did not show significant fold changes as a result of SDF-1 preconditioning, in both BM-MSCs and C-PCs (Fig 2A; 2B). MMP-13 gene expression in C-PCs was too low to be detected, suggesting that these cells exhibit naturally low expression of this cartilage degrading enzyme. Additionally, we tested whether SDF-1 stimulation of cells in culture during chondrogenesis induction would alter their expression. The presence of SDF-1 did not significantly alter the expression of these genes during chondrogenesis (Supplemental Fig 2). However, chondrogenic induction demonstrated that Collagen II expression was significantly elevated C-PCs, in comparison to BM-MSCs. MMP-13 remained suppressed in C-PCs, even after chondrogenesis induction (Supplemental Fig 2E). Furthermore, RUNX2 expression was also significantly lower in chondrogenically induced C-PCs, compared to BM-MSCs (Supplemental Fig 2F). Lastly, we quantified soluble GAG content in culture media and our findings demonstrated that the presence of SDF-1 did not impair GAG release by C-PCs during chondrogenesis induction. Taken together, these findings indicated that exposure of C-PCs to SDF-1 does not alter the expression of early and late chondrocyte differentiation genes.
Figure 2. Chondrocyte differentiation gene expression is minimally impacted by SDF-1/CXCR4 axis activation in human BM-MSCs and C-PCs.
(A) Gene expression analysis of early-stage chondrocyte differentiation genes (Collagen I, Collagen II, SOX9) and late stage chondrocyte differentiation genes (Collagen X, RUNX2, Matrix Metalloproteinase 13) and the SDF-1 receptor (CXCR4) in BM-MSCs and (B) C-PCs following SDF-1 treatment. Relative gene expression was quantified using Real-time quantitative Polymerase Chain Reaction 3 days after culture. N ≥ 3, * P ≤ 0.05. Error bars represent ±1 SD of the mean.
SDF-1 pretreated cell and scaffold constructs facilitate cell distribution throughout meniscus injury site.
We next tested whether SDF-1 pretreated HPC scaffolds containing C-PCs can be used to distribute these cells into fibrocartilage breaks. To do this, we used an explant model of human meniscus tissue injury. Human menisci were obtained from total knee replacement surgeries at Rhode Island Hospital. Patient age range was 61 – 85. Human meniscus tissue plugs were punched out (3 – 4 mm diameter) from the red-white zone and cut along the transverse plane into two pieces (Fig 3A: i). These pieces were then reassembled into a three-layered cylindrical construct containing an SDF-treated (0.1 μg/mL) HPC scaffold disc in between the two tissue plugs (Fig 3A: ii, iii). Each scaffold was seeded with 1.0×105 C-PCs that were labeled with fluorescent red Vibrant Dil dye. After two weeks in culture, the constructs were whole-mounted and imaged. C-PCs that were seeded into scaffolds pre-treated with SDF-1 infiltrated the scaffolds showed localization at the intersection where the scaffold meets the meniscus (Fig 3B: i, ii), whereas scaffolds that were not pretreated with this chemokine contained fewer cells (Fig 3B: iii, iv). C-PCs also infiltrated the meniscus plugs past the scaffold intersection. Taken together, these results indicated that SDF-1 pre-treated scaffolds facilitate the effective distribution of seeded C-PCs to meniscal fibrocartilage tissues at injury sites.
Figure 3. SDF-1 pre-treated HPC scaffolds can be used as cell reservoirs to facilitate re-adherence of meniscus tissue breaks.
(A) Assembly of meniscus (M) scaffold (S) constructs from human meniscal fibrocartilage plugs. Scaffolds were pretreated with SDF-1 and seeded with 1.0 × 105 cells. Polylactic Acid (PLA) disks were used to sandwich/compress the meniscus-scaffold construct and a stainless-steel pin was used as a rail to keep constructs held together. (B) Constructs were incubated in aseptic tissue-culture conditions for 2 weeks at which point fluorescent imaging was performed. Panels i, ii represent meniscus-scaffold-meniscus constructs that contained scaffolds that were pretreated with SDF-1 and panels iii, iv represents constructs containing scaffolds that were not pretreated with SDF-1. Cells are fluorescently red labeled in images. (i, iii – fluorescent channel only) (ii, iv – fluorescent channel and bright field merged). The label “M” signifies meniscus plugs and the label “S” signifies the scaffold. Experiments were repeated eight times.
C-PCs stimulate adherence of meniscus tissue to HPC scaffolds constructs.
Meniscus-scaffold three-piece constructs (plug-scaffold-plug) that were seeded with SDF-1 conditioned C-PCs were also subjected to biomechanical tensile testing. Plug-scaffold-plug constructs were pulled apart to failure using custom 3D printed PLA adaptors that were each meniscus tissue plug in place (Fig 4A). Constructs containing scaffolds with cells exhibited a measurable degree of adherence/integration, as determined by tensile force testing (Fig 4B). The control group that was used for comparison consisted of plug-scaffold-plug constructs containing SDF-1 in the scaffold, but without C-PCs. The majority of plug-scaffold-plug constructs in the control group did not support a tensile load (0 grams of tensile force required to separate the construct interface). These samples either failed during testing or when the constructs were being mounted to the testing platform due to weak interface adherence. On the other hand, the C-PC treated samples exhibited effective mechanical adherence ranging from 0.6 to 89.7 gram-force to failure, with the mean adherence being 24.6 gram-force to failure. Despite this large variation, the difference in gram-force to failure between C-PC-treated and untreated groups was statistically significant (p-value = 0.025). We surmise that the large variation in adhesion force can at least partly be attributed to the heterogeneity of patient tissues, considering that each tissue comes from a different patient that required complete knee replacement.
Figure 4. SDF-1 conditioned C-PCs increase the mechanical integrity of meniscus-scaffold constructs.
(A) Photo of biomechanical tensile testing setup taken during testing of meniscus-scaffold construct depicting the custom-made 3D engineered adaptors and the orientation in which the construct is attached/glued to the adaptor platforms. (B) The tensile strength of SDF-1 pretreated meniscus-scaffold constructs containing C-PCs in the scaffold was compared to constructs that did not contain C-PCs. Constructs containing C-PCs exhibited and average adherence of 24.6 gram-force to failure. Constructs without C-PCs exhibited and average adherence of 1.29 gram-force to failure. *, P ≤ 0.05. N ≥ 8.
DISCUSSION
Meniscus tears are common orthopaedic injuries that are challenging to treat due to the poor healing capacity of meniscal fibrocartilage19,20. It is surmised that a viable strategy to improve meniscus tissue repair is to increase trafficking of mesenchymal progenitor cells to injury sites, because they have been shown to help facilitate fibrocartilage healing6,7. Veritably, our group and others have previously reported that SDF-1/CXCR4 pathway signaling mediates stem/progenitor cell migration to sites of tissue injury and can improve healing6,21. The chemokine SDF-1 and its receptor CXCR4 are major regulators of stem cell trafficking and mobilization22–24. Elevated expression of CXCR4 in MSCs is reported to increase cell migration, increase cell engraftment in the site of injury, and improve tissue regeneration in cell-based myocardial repair25,26 and ischemic tissue repair27.
The goal of the present study was to investigate the efficacy of using SDF-1 as a chemoattractant to help localize mesenchymal progenitor cells to sites of meniscus fibrocartilage injury. Using an ex-vivo human meniscus tissue injury model we tested whether SDF-1-mediated cell homing can stimulate tissue reintegration and restore a degree of mechanical integrity. Specifically, our interest was to determine how progenitor cells that are naturally derived from the cartilage milieu can be utilized according to this strategy.
Work performed by our group and others have demonstrated that C-PCs are a population of chondrogenic cells that resist cellular hypertrophy, thereby potentially making them an ideal fit for cell-based cartilaginous tissue repair applications6,14,15. In this study, we first demonstrated that SDF-1 pretreatment of HPC scaffolds stimulate their migration into the scaffold with the same efficiency as that of BM-MSCs. Gene expression analysis revealed that SDF-1 conditioned BM-MSCs and C-PCs do not exhibit altered expression of important chondrogenesis genes, nor do they result in the increase of cell hypertrophy genes and the cartilage degradative enzyme MMP-13, unlike the case with mature chondrocytes28,29. This finding is important because it demonstrates that utilizing SDF-1 to concentrate progenitor cells to a focal region does not compromise cellular characteristics that make them suitable for cartilaginous tissue repair applications. We also demonstrated in this study that SDF-1 pretreated HPC scaffolds can be used to disperse cells throughout injured meniscus tissue. HPC scaffolds were chosen due to the biocompatible makeup of HPC itself. Additionally, HPC is side chain modified to allow crosslinking and ligand conjugation, which may be useful in the future for tightly regulating the scaffold microenvironment during cell-based tissue repair. This finding is relevant to applications of cell-based tissue repair because it indicates that these scaffold constructs facilitate the free movement of cells into the injured tissue once they are placed in close proximity to an injury site.
We also observed that C-PCs seeded into an HPC scaffold improved the physical integrity of these constructs, as determined through mechanical tensile testing analysis. C-PC seeded SDF-1 pretreated HPC scaffolds stimulated an average tensile strength of 24.6 gram-force, whereas scaffolds without cells mechanically failed when subjected to an average of 1.29 gram-force.
The present study shows for the first time that, similar to mesenchymal progenitor/stem cells from bone marrow – BM-MSCs, mesenchymal progenitors from cartilage – C-PCs can also effectively mobilize in response to SDF-1. It is intriguing that our data shows the SDF-1 receptor CXCR4 is expressed by relatively few C-PCs in culture. However, this finding is consistent with others who have shown that MSCs exhibit lower expression of CXCR4 in normoxic culture conditions30. Overall this suggests that if the C-PCs were introduced into a hypoxic environment, such as the knee joint, CXCR4 surface expression may increase31,32 thereby potentially further elevating the chemotaxis of these cells towards SDF-1. This study demonstrates that SDF-1 pretreatment of scaffolds and tissues can be used as a viable strategy to mediate targeted C-PC trafficking to areas of interest. Moreover, exposure to SDF-1 does not alter the fundamental molecular characteristics in C-PCs that make them desirable for cell-based therapy. Most importantly (and surprisingly), our findings show that SDF-1 does not significantly elevate the expression of MMP-13 in either C-PCs or BM-MSCs. This is important particularly due to SDF-1 being reported to stimulate expression of hypertrophic markers type X collagen, RUNX2, and cartilage catabolism marker MMP-13 in mature chondrocytes. We also report for the first time that mobilization of C-PCs to HPC scaffolds facilitate improved mechanical integrity at the meniscus-scaffold interface. However, the broad range in gram-force required to dismember the meniscus-scaffold constructs during tensile testing can be attributed to a notable limitation of this study. Meniscus tissue heterogeneity was a variable beyond our control, since we can only collect human meniscus tissues from total knee replacement surgeries. However, regardless of this, C-PC-treated meniscus-scaffold constructs on average exhibited significantly greater effective mechanical integrity suggesting that C-PCs can be effective in applications of human meniscus injury repair.
Supplementary Material
Supplemental Figure 1. (A) ELISA results used to quantify the release of recombinant SDF-1 protein from HPC scaffold into culture medium over time. Scaffolds were pretreated with 0.1 μg/mL of recombinant SDF-1 protein, prior to culture experiments. N=4. (B) MTT cell viability assay performed on C-PCs that have migrated into Collagen I coated HPC scaffolds that were pretreated or left untreated with SDF-1 recombinant protein. N = 3. ***, P ≤ 0.005. Error bars represent ±1 SD of the mean in both experiments.
Supplemental Figure 2. (A) Gene expression analysis of early-stage chondrocyte differentiation genes Collagen I, (B) Collagen II, (C) SOX9 and late stage chondrocyte differentiation genes (D) Collagen X, (E) Matrix Metalloproteinase 13 (F) RUNX2 and (G) the SDF-1 receptor in BM-MSCs and C-PCs following SDF-1 treatment during chondrogenesis induction over a 5 day period. Relative gene expression was quantified using Real-time quantitative Polymerase Chain Reaction. N = 4, * P ≤ 0.05, ** P ≤ 0.01, *** P ≤ 0.005, **** P ≤ 0.001. Error bars represent ±1 SD of the mean.
Supplemental Figure 3. (A) Quantification of GAG concentration in culture medium following 5-day chondrogenesis induction of C-PCs (1.0 x 105 cells) in the presence and absence of SDF-1 recombinant protein. (B) GAG concentration normalized to DNA content in the cell culture monolayer. N = 3, ** P ≤ 0.01, ****, P ≤ 0.001 compared to the growth medium cultured control group. ## ≤ 0.01. Error bars represent ±1 SD of the mean.
Table 1.
List of forward and reverse primers, in 5′ to 3′ orientation, used for RT-qPCR. The table also lists the specific accession numbers of all target gene transcripts
| Gene | Forward Seq. | Reverse Seq. | Accession |
|---|---|---|---|
| Beta-Actin | GGACCTGACTGACTACCTCAT | CGTAGCACAGCTTCTCCTTAAT | NM_001101.4 |
| COL1 | CAGGAGGCACGCGGAGTGTG | GGCAGGGCTCGGGTTTCCAC | NM_000088.3 |
| COL2 | CTCCCAGAACATCACCTACCACT | CGTGAACCTGCTATTGCCCT | NM_001844.4 |
| COL10 | GCCCACAGGCATAAAAGGCCC | GAAGGACCTGGGTGCCCTCGA | NM_000493.3 |
| CXCR4 | GAAACCCTCAGCGTCTCAGT | AGTAGTGGGCTAAGGGCACA | NM_001008540.2 |
| MMP13 | ATGCGGGGTTCCTGATGTGG | GGCCCAGGAGGAAAAGCATG | NM_002427.4 |
| RUNX2 | CTCTGACTTCTGCCTCTGGC | GGTGTGGTAGTGAGTGGTGG | NM_001024630.4 |
| SOX9 | GGACCAGTACCCGCACTTGCA | GTTCTTCACCGACTTCCTCCGCCG | NM_000346.4 |
Acknowledgments
We would like to acknowledge Ms. Jenasis Ortega for her assistance with bench experiments.
This research was funded in part by faculty startup funds provided to Dr. Jayasuriya by the Rhode Island Hospital Department of Orthopaedics; by a clinical translational research pilot funding awarded to Dr. Jayasuriya and Dr. Owens through Institutional Development Award Number NIH U54GM115677 from the National Institute of General Medical Sciences (NIGMS); by the Orthopaedic Research and Education Foundation (OREF); and by the Bioengineering Core of the COBRE for Skeletal Health and Repair 5P3 GM122732. We would further like to thank the Department of Orthopaedics for working with the lab to acquire the meniscus samples used in this study.
Footnotes
Disclosures
The authors of this article report no conflict of interest.
References
- 1.Arnoczky SP, Warren RF. Microvasculature of the human meniscus. Am J Sports Med 1982. Mar-Apr;10(2):90–5. [DOI] [PubMed] [Google Scholar]
- 2.Day B, Mackenzie WG, Shim SS, Leung G. The vascular and nerve supply of the human meniscus. Arthroscopy 1985;1(1):58–62. [DOI] [PubMed] [Google Scholar]
- 3.Bochynska AI, Hannink G, Grijpma DW, Buma P. Tissue adhesives for meniscus tear repair: An overview of current advances and prospects for future clinical solutions. J Mater Sci Mater Med 2016. May;27(5):85,016-5694-5. Epub 2016 Mar 12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Korpershoek JV, de Windt TS, Hagmeijer MH, Vonk LA, Saris DB. Cell-based meniscus repair and regeneration: At the brink of clinical translation?: A systematic review of preclinical studies. Orthop J Sports Med 2017. February 21;5(2):2325967117690131. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Bilgen B, Jayasuriya CT, Owens BD. Current concepts in meniscus tissue engineering and repair. Adv Healthc Mater 2018. March 15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Jayasuriya CT, Twomey-Kozak J, Newberry J, Desai S, Feltman P, Franco JR, Li N, Terek R, Ehrlich MG, Owens BD. Human cartilage-derived progenitors resist terminal differentiation and require CXCR4 activation to successfully bridge meniscus tissue tears. Stem Cells 2018. October 24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Tarafder S, Gulko J, Sim KH, Yang J, Cook JL, Lee CH. Engineered healing of avascular meniscus tears by stem cell recruitment. Sci Rep 2018. May 25;8(1):8150,018-26545-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Sasaki H, Rothrauff BB, Alexander PG, Lin H, Gottardi R, Fu FH, Tuan RS. In vitro repair of meniscal radial tear with hydrogels seeded with adipose stem cells and TGF-beta3. Am J Sports Med 2018. August;46(10):2402–13. [DOI] [PubMed] [Google Scholar]
- 9.Makris EA, Hadidi P, Athanasiou KA. The knee meniscus: Structure-function, pathophysiology, current repair techniques, and prospects for regeneration. Biomaterials 2011. October;32(30):7411–31. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Fox AJ, Bedi A, Rodeo SA. The basic science of human knee menisci: Structure, composition, and function. Sports Health 2012. July;4(4):340–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Pereira H, Frias AM, Oliveira JM, Espregueira-Mendes J, Reis RL. Tissue engineering and regenerative medicine strategies in meniscus lesions. Arthroscopy 2011. December;27(12):1706–19. [DOI] [PubMed] [Google Scholar]
- 12.Muller FA, Muller L, Hofmann I, Greil P, Wenzel MM, Staudenmaier R. Cellulose-based scaffold materials for cartilage tissue engineering. Biomaterials 2006. July;27(21):3955–63. [DOI] [PubMed] [Google Scholar]
- 13.Svensson A, Nicklasson E, Harrah T, Panilaitis B, Kaplan DL, Brittberg M, Gatenholm P. Bacterial cellulose as a potential scaffold for tissue engineering of cartilage. Biomaterials 2005. February;26(4):419–31. [DOI] [PubMed] [Google Scholar]
- 14.Williams R, Khan IM, Richardson K, Nelson L, McCarthy HE, Analbelsi T, Singhrao SK, Dowthwaite GP, Jones RE, Baird DM, et al. Identification and clonal characterisation of a progenitor cell sub-population in normal human articular cartilage. PLoS One 2010. October 14;5(10):e13246. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.McCarthy HE, Bara JJ, Brakspear K, Singhrao SK, Archer CW. The comparison of equine articular cartilage progenitor cells and bone marrow-derived stromal cells as potential cell sources for cartilage repair in the horse. Vet J 2012. June;192(3):345–51. [DOI] [PubMed] [Google Scholar]
- 16.Walker GD, Fischer M, Gannon J, Thompson RC Jr, Oegema TR Jr. Expression of type-X collagen in osteoarthritis. J Orthop Res 1995. January;13(1):4–12. [DOI] [PubMed] [Google Scholar]
- 17.Higashikawa A, Saito T, Ikeda T, Kamekura S, Kawamura N, Kan A, Oshima Y, Ohba S, Ogata N, Takeshita K, et al. Identification of the core element responsive to runt-related transcription factor 2 in the promoter of human type X collagen gene. Arthritis Rheum 2009. January;60(1):166–78. [DOI] [PubMed] [Google Scholar]
- 18.Bertrand J, Cromme C, Umlauf D, Frank S, Pap T. Molecular mechanisms of cartilage remodelling in osteoarthritis. Int J Biochem Cell Biol 2010. October;42(10):1594–601. [DOI] [PubMed] [Google Scholar]
- 19.Majewski M, Susanne H, Klaus S. Epidemiology of athletic knee injuries: A 10-year study. Knee 2006. June;13(3):184–8. [DOI] [PubMed] [Google Scholar]
- 20.Clayton RA, Court-Brown CM. The epidemiology of musculoskeletal tendinous and ligamentous injuries. Injury 2008. December;39(12):1338–44. [DOI] [PubMed] [Google Scholar]
- 21.Shen W, Chen J, Zhu T, Chen L, Zhang W, Fang Z, Heng BC, Yin Z, Chen X, Ji J, et al. Intra-articular injection of human meniscus stem/progenitor cells promotes meniscus regeneration and ameliorates osteoarthritis through stromal cell-derived factor-1/CXCR4-mediated homing. Stem Cells Transl Med 2014. March;3(3):387–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Liu H, Xue W, Ge G, Luo X, Li Y, Xiang H, Ding X, Tian P, Tian X. Hypoxic preconditioning advances CXCR4 and CXCR7 expression by activating HIF-1alpha in MSCs. Biochem Biophys Res Commun 2010. October 29;401(4):509–15. [DOI] [PubMed] [Google Scholar]
- 23.Kitaori T, Ito H, Schwarz EM, Tsutsumi R, Yoshitomi H, Oishi S, Nakano M, Fujii N, Nagasawa T, Nakamura T. Stromal cell-derived factor 1/CXCR4 signaling is critical for the recruitment of mesenchymal stem cells to the fracture site during skeletal repair in a mouse model. Arthritis Rheum 2009. March;60(3):813–23. [DOI] [PubMed] [Google Scholar]
- 24.Lataillade JJ, Clay D, Dupuy C, Rigal S, Jasmin C, Bourin P, Le Bousse-Kerdiles MC. Chemokine SDF-1 enhances circulating CD34(+) cell proliferation in synergy with cytokines: Possible role in progenitor survival. Blood 2000. February 1;95(3):756–68. [PubMed] [Google Scholar]
- 25.Cheng Z, Ou L, Zhou X, Li F, Jia X, Zhang Y, Liu X, Li Y, Ward CA, Melo LG, et al. Targeted migration of mesenchymal stem cells modified with CXCR4 gene to infarcted myocardium improves cardiac performance. Mol Ther 2008. March;16(3):571–9. [DOI] [PubMed] [Google Scholar]
- 26.Zhang D, Fan GC, Zhou X, Zhao T, Pasha Z, Xu M, Zhu Y, Ashraf M, Wang Y. Over-expression of CXCR4 on mesenchymal stem cells augments myoangiogenesis in the infarcted myocardium. J Mol Cell Cardiol 2008. February;44(2):281–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Kim M, Kim DI, Kim EK, Kim CW. CXCR4 overexpression in human adipose tissue-derived stem cells improves homing and engraftment in an animal limb ischemia model. Cell Transplant 2017. February 16;26(2):191–204. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Wei L, Kanbe K, Lee M, Wei X, Pei M, Sun X, Terek R, Chen Q. Stimulation of chondrocyte hypertrophy by chemokine stromal cell-derived factor 1 in the chondro-osseous junction during endochondral bone formation. Dev Biol 2010. May 1;341(1):236–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Chiu YC, Yang RS, Hsieh KH, Fong YC, Way TD, Lee TS, Wu HC, Fu WM, Tang CH. Stromal cell-derived factor-1 induces matrix metalloprotease-13 expression in human chondrocytes. Mol Pharmacol 2007. September;72(3):695–703. [DOI] [PubMed] [Google Scholar]
- 30.Xu W, Xu R, Li Z, Wang Y, Hu R. Hypoxia changes chemotaxis behaviour of mesenchymal stem cells via HIF-1alpha signalling. J Cell Mol Med 2019. March;23(3):1899–907. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Schioppa T, Uranchimeg B, Saccani A, Biswas SK, Doni A, Rapisarda A, Bernasconi S, Saccani S, Nebuloni M, Vago L, et al. Regulation of the chemokine receptor CXCR4 by hypoxia. J Exp Med 2003. November 3;198(9):1391–402. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Liu H, Liu S, Li Y, Wang X, Xue W, Ge G, Luo X. The role of SDF-1-CXCR4/CXCR7 axis in the therapeutic effects of hypoxia-preconditioned mesenchymal stem cells for renal ischemia/reperfusion injury. PLoS One 2012;7(4):e34608. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Figure 1. (A) ELISA results used to quantify the release of recombinant SDF-1 protein from HPC scaffold into culture medium over time. Scaffolds were pretreated with 0.1 μg/mL of recombinant SDF-1 protein, prior to culture experiments. N=4. (B) MTT cell viability assay performed on C-PCs that have migrated into Collagen I coated HPC scaffolds that were pretreated or left untreated with SDF-1 recombinant protein. N = 3. ***, P ≤ 0.005. Error bars represent ±1 SD of the mean in both experiments.
Supplemental Figure 2. (A) Gene expression analysis of early-stage chondrocyte differentiation genes Collagen I, (B) Collagen II, (C) SOX9 and late stage chondrocyte differentiation genes (D) Collagen X, (E) Matrix Metalloproteinase 13 (F) RUNX2 and (G) the SDF-1 receptor in BM-MSCs and C-PCs following SDF-1 treatment during chondrogenesis induction over a 5 day period. Relative gene expression was quantified using Real-time quantitative Polymerase Chain Reaction. N = 4, * P ≤ 0.05, ** P ≤ 0.01, *** P ≤ 0.005, **** P ≤ 0.001. Error bars represent ±1 SD of the mean.
Supplemental Figure 3. (A) Quantification of GAG concentration in culture medium following 5-day chondrogenesis induction of C-PCs (1.0 x 105 cells) in the presence and absence of SDF-1 recombinant protein. (B) GAG concentration normalized to DNA content in the cell culture monolayer. N = 3, ** P ≤ 0.01, ****, P ≤ 0.001 compared to the growth medium cultured control group. ## ≤ 0.01. Error bars represent ±1 SD of the mean.