The authors have declared that no competing interests exist.
Conceived and designed the experiments: SR RD RB. Performed the experiments: RD RB KC GC AC BB. Analyzed the data: RD KC GC. Contributed reagents/materials/analysis tools: BB JL AA. Wrote the paper: SR RD RB KC.
We investigated the effectiveness of integrating tissue engineered cartilage derived from human bone marrow derived stem cells (HBMSCs) to healthy as well as osteoarthritic cartilage mimics using hydroxyapatite (HA) nanoparticles immersed within a hydrogel substrate. Healthy and diseased engineered cartilage from human chondrocytes (cultured in agar gels) were integrated with human bone marrow stem cell (HBMSC)-derived cartilaginous engineered matrix with and without HA, and evaluated after 28 days of growth. HBMSCs were seeded within photopolymerizable poly (ethylene glycol) diacrylate (PEGDA) hydrogels. In addition, we also conducted a preliminary
Articular cartilage lesions in the knee frequently occur following an injury. Curl [
Tissue engineering approaches have shown great potential in treating cartilage defects [
Recently, injectable hydrogels incorporating hydroxyapatite (HA) nanoparticles have shown great potential in enhancing the integration of engineered cartilage to bone matrix [
Ethics Statement concerning use of Animals: Isoflurane was used to anesthetize the rabbits. Euthanization of the rabbits was conducted at the end of the in vivo portion of the study via barbituate overdose. In brief, barbituate overdose using 100 mg/ml pentabarbitol and 1 ml/10 lbs was administered by intracardiac injection. Note that, Ketamine (10 mg/kg IM) and medetomidine (0.5 mg/kg IM) was given to sedate the rabbits prior to euthanasia injection. Institutional Animal Care and Use Committee (IACUC): Mississippi State University (MSU), MSU-IACUC approval was granted on: April 9th 2014. MSU-IACUC Approval Number: # 14–029.
HBMSCs (Science cell, Carlsbad, CA) were seeded onto Poly D-lysine coated T-75 flasks (Fisher Scientific, Pittsburg, PA). The cells were cultured in manufacturer supplied proprietary medium (AdvanceSTEM Mesenchymal Stem Cell Expansion Kit (Thermo Scientific, Waltham, WA) until passage three (P3).
HCs (Cell Applications Inc., San Diego, CA, USA) were sourced from non-pathologic articular cartilage. They were cryopreserved at passage 2 (P2) when shipped by the manufacturer. When the cells were received, they were plated in T-75 flasks (Fisher Scientific) and allowed to proliferate. The cells were cultured in the Human Chondrocyte Media (Cell Applications, San Diego, CA) until P4 and were subsequently used for the experiments.
HCOA (Cell Applications) were isolated from the articular cartilage from patients with Osteoarthritis. They were received from the manufacturer at P1 and prior to usage in the experiments, were cultured and expanded in human chondrocyte media (Cell Applications) until P4.
Agar is a well-established gel scaffold for culturing cartilage matrix
Tissue engineered osteoarthritic cartilage was prepared in the same manner as tissue engineered healthy cartilage except for the use of HCOAs instead of HCs.
HBMSC derived engineered cartilage was prepared using our previously established protocols [
As stated, we first created healthy/osteoarthritic chondrocyte-derived cartilage constructs in Agar gel contained within cylindrical molds via temperature-based gelation i.e., either healthy or osteoarthritic, to mimic adjacent healthy and diseased native articular cartilage environments respectively. Next, into the same mold, HBMSCs suspended in monomer solution of PEGDA with and without HA was poured on top the Agar constructs. The PEGDA solution subsequently underwent photopolymerization to form a gel structure. Finally, a steel pin (Fisher Scientific, Catalog # 26002–10) was pierced across the two gel constructs to physically secure them for a period of 4 weeks. A physical gap would present itself at the interface of the Agar and PEGDA layers which presumably would fill with
Cellular constructs were similarly made without HA. In sum, four groups of 2 layers specimens were ultimately prepared as follows: 1) HCs Agar-HBMSCs PEGDA- No HA (HCs encapsulated in Agar-HBMSCs encapsulated in PEGDA without HA) 2) HCs Agar-HBMSCs PEGDA-HA (HCs encapsulated in Agar-HBMSCs encapsulated in PEGDA with HA), 3) HCOAs Agar-HBMSCs PEGDA- No HA (HCOAs encapsulated in Agar-HBMSCs encapsulated in PEGDA without HA), and 4) HCOAs Agar-HBMSCs PEGDA-HA (HCOAs encapsulated in Agar-HBMSCs encapsulated in PEGDA with HA).
To assess the viability of HCs and HCOAs encapsulated in Agar constructs, a Live-Dead assay was conducted using Calcein AM/Ethidium homodimer (Life Technologies, Carlsbad, CA) staining following the manufacturer's protocol.
In brief, we used 4.1mm long healthy/osteoarthritic chondrocytes-seeded in Agar constructs that were sectioned into 1 mm-thick slices; the slices were subsequently stained to assess cell viability. Since the depth of the constructs in later investigations represented the depth of the constructs used in the viability tests, we interpret the latter to be reflective of cell viability under the actual experimental situation. More precisely, samples from the different groups that were cultured to different time points (Day1, Day 7, Day 14 and Day 28) were as mentioned, sectioned to 1mm thin slices using a blade and immediately stained with Calcein AM and Ethidium homodimer in PBS solution and incubated for 30 minutes. Next, to remove the background fluorescence they were washed 3 times with PBS solution and then visualized under a fluorescent microscope (Olympus IX81, Olympus America Inc., Miami, FL) at excitation and emission wavelengths of 495/515 nm and 495/635 nm for Calcein AM and Ethidium homodimer respectively. Note that the viability of HBMSCs in PEGDA environment with and without the incorporation of HA was previously reported by our research group [
The interfacial shear strength between the PEGDA-based tissue engineered cartilage and the agar based healthy (or alternatively, diseased) engineered cartilage substrates were determined by performing shear testing at the interface of the 2 layers. Specifically, Interfacial shear stress mechanical testing was performed using a BOSE Electroforce 3200 mechanical testing device (Bose, Eden Prairie, MN). Data was collected real-time manufacturer software (WinTest, Bose) and was then exported to a spreadsheet (MS excel, Microsoft Corp., Redmond, WA) for additional processing. Samples to be tested (n = 6/group) were placed horizontally on the platens. One side of the sample was adhered to the upper platen while the other end was secured to the lower platen, both using super glue. The upper platen was subsequently displaced at a rate of 0.05 mm/sec with data logging conducted at every 0.03 mm of displacement. The test was ended when the 2 layers were observed to separate from each other and also accompanied with a sudden rapid decrease in the loads recorded.
Calcium and GAG matrix distribution at the interface of two layer constructs for Group 1–4 was evaluated using the Von-Kossa and Alcian Blue Stain kit (IHC world, Woodstock, MD, USA). Briefly the samples from each group were first fixed in 10% formalin. They were then embedded in molds using OCT. Sections were cut using cryostat (25 micron thickness) and were stained using the manufacturer’s protocols. In short, sectioned samples were first washed with PBS three times to remove the OCT. Then they were treated with silver nitrate solution and exposed to UV light for 60 minutes. Next, they were stained with Alcian Blue Stain, washed with distilled water and observed under a microscope to visualize both the calcium and cartilage matrix deposition at the same time on the same tissue section. A transition zone was identified as the region of
To quantify the composition of elements found within the transition zone of two distinct engineered tissue matrices treated with HA, energy-dispersive x-ray spectroscopy (EDS) and EDS mapping was performed using the suggested manufacturer protocol (JEOL 6330F Field Emission Scanning Electron Microscope (FEG-SEM), JEOL Ltd., Akishima-Shi, Tokyo, Japan). The EDS was focused spatially on histological specimens containing the transition zone between engineered cartilage and engineered bone, which possessed a relatively large spatial region for assessment, thereby yielding unambiguous spectra. The purpose of the EDS analysis was to quantify the constituents of the transition zone. To obtain unambiguous spectra we assessed a relatively large spatial region sandwiched between engineered cartilage and engineered bone. To make this assessment unambiguous, an EDS map of this transition zone was also conducted.
At 4 weeks following PEGDA with or without HA treatment, rabbit knee explants (see “2.7 In Vivo Studies” sub-section) were evaluated visually, and then placed in a tube with 10% buffered formalin Decalcification of the joint was performed using Decalcifying Solution-Lite (Sigma-Aldrich) according to the manufacturer’s protocol. The specimens were embedded in optimal cutting temperature compound (OCT; Polysciences, Inc., Warrington, PA) and sliced using a cryostat (Lecia, Buffalo Grove, IL) through the center of each defect. Samples were subsequently stained with hematoxylin and eosin (H&E) according to manufacturer’s protocol (ScyTeck Laboratories, UT) and viewed under a microscope (Amscope, Irvine, CA) to visualize cell and repair tissue morphology within the cartilage defects. For quantification purposes, cell counts were performed on the images of the H&E sections (n = 6 defects for microfracture + HA; n = 3 defects for microfracture alone) to quantify the number of cells present in the repair tissue (ImageJ, 1.48v, National Institutes of Health, Bethesda, MD).
To assess the gene expression of HBMSC-derived,
Each sample was subsequently cut into 2 parts. One section comprised of the transition region of 1 mm, which included the interface location while the remainder consisted solely of engineered tissue derived from HBMSCs. Three samples from each group were cut in a similar fashion and, crushed and pooled together for analysis. This was repeated for another 2 samples from each group, i.e., n = 3 samples for each of the interfacial and distal cartilage locations. Total mRNA was extracted from each group three times (SV total RNA isolation kit, Promega, Madison, WI). RNA isolation, reverse transcription and the qRT-PCR was subsequently performed as previously reported [
The primers (
Genes | Forward Primer | Reverse Primer |
---|---|---|
GAPDH | AATGAAGGGGTCATTGATGG | AAGGTGAAGGTCGGAGTCAA |
Aggrecan | GCGAGTTGTCATGGTCTGAA | TTCTTGGAGAAGGGAGTCCA |
SOX9 | GTAATCCGGGTGGTCCTTCT | GTACCCGCACTTGCACAAC |
Collagen II | AGACTTGCGTCTACCCCAATC | GCAGGCGTAGGAAGGTCATC |
MMP 13 | ACATCCCAAAACGCCAGACAA | GATGCAGCCGCCAGAAGAAT |
Runx2 | AATCCTCCCCAAGTTGCCA | TTCTGTCTGTCCTTCTGGGT |
Type X Collagen | TGGATCCAAAGGCGATGTG | GCCCAGTAGGTCCATTAAGGC |
Type I Collagen | TGAGAGACCAAGAACTG | CCATCCAAACCACTGAAACC |
Osteocalcin | CACTCCTCGCCCTATTGGC | CCCTCCTGCTTGGACACACAAAG |
To further evaluate the utility of HA in promoting engineered to native cartilage tissue integration an
Two weeks following this surgery, the animals underwent a second surgical procedure in which 4 mm chondral defects (3 per stifle) were created within the femoral trochleas. The defects were treated with either PEDGA alone or with PEDGA + HA. Of note, the PEGDA utilized for these
Statistical analysis was performed for the
The viability of the cells in the engineered scaffolds was observed over a period of 28 days at 4 time points (Day 1, Day 7, Day 14 and Day 28). We found that ~ 97% of the HCs and ~ 86% of the HCOAs were viable in the Agar gel at 28 days of growth (
(a) healthy and (b) osteoarthritic chondrocytes during 28 days of culture. We found that ~ 97% of the HCs and ~ 86% of the HCOAs were viable in the Agar gel at 28 days of growth.
We found a significantly higher shear strength (p<0.05) in HCs Agar-HBMSCs PEGDA- No HA compared to HCs Agar-HBMSCs PEGDA-HA after both 7 and 28 days of tissue culture (
(a) Tissue engineered cartilage integrated with healthy cartilage mimics with and without the presence of HA. (b) Tissue engineered cartilage integrated with osteoarthritic cartilage mimics with and without presence of HA. The “*” indicates that the difference between the groups was statistically significant (p < 0.05).
Histological sections revealed that HCs encapsulated in Agar-HBMSCs encapsulated in PEGDA with HA did not form a transition zone, but instead presented with a physical spacing or gap between the two engineered constructs (
(a) Tissue engineered cartilage derived from HBMSCs integrated with HC-secreted cartilage matrix without HA at Day 1 and at Day 28. (b) Tissue engineered cartilage derived from HBMSCs integrated with HC-secreted cartilage matrix with HA at Day 1 and at Day 28. (c) Tissue engineered cartilage derived from HBMSCs integrated with HCOA-secreted cartilage matrix without HA incorporation at Day 1 and at Day 28. (d) Tissue engineered cartilage derived from HBMSCs integrated with HCOA-secreted cartilage matrix with HA incorporation at Day 1 and at Day 28—Progressive filling of the transition zone with calcium phosphate deposits (indicated by dotted yellow lines) in the group with HA was found to occur. (e) A zoom-in of the transition zone (additional 2.5X that of Fig 3(d)). (f) Even further zoom-in of the transition zone (additional 10X that of Fig 3(d)).
After 28 days of tissue culture, in HCOAs encapsulated in Agar-HBMSCs encapsulated in PEGDA with HA, a thin transition zone was formed between HBMSC-derived engineered cartilage and HCOA-derived cartilage; on the other hand, when HA was not utilized, only a narrowing of the gap between the two layers was observed as was previously seen in the HC groups (
EDS analysis in the transition zone between
(a) Elementary composition of the transition zone using (EDS). A large transition region between engineered cartilage and engineered bone treated with HA was analyzed after 28 days of culture. Elemental Calcium in the order of ∼ 6.41% was found to be present in the transition zone. (b) EDS map of the transition zone with the revelation of both robust presence of phosphorous and calcium, the two primary constituents of HA. The map thereby confirmed HA’s presence within the transition zone.
Gross morphology as well as histology suggested robust tissue filling of defects treated with PEGDA-alone as well as those treated with PEGDA+HA (
(a) Macroscopic view of bone joint morphology. Three chondral defects per rabbit knee were created in the trochlear groove, shown by the yellow arrow with microfracture of the subchondral bone. (b) H&E staining (40X magnification) of chondral defect (c) H&E staining (250X magnification) of: (Left) without the incorporation of HA and (Right) with HA incorporation. The inclusion of HA nanoparticles promoted improved cellular organization within the newly formed tissue in the defect.
After 28 days of culture in chondrogenic media, it was observed that HBMSCs derived tissue engineered cartilage in HCs Agar-HBMSCs PEGDA- No HA, exhibited relatively high expression of Aggrecan, SOX9 and Collagen Type II genes in both the proximal and distal regions (relative to the interface with HCs-derived engineered cartilage;
(a) q(RT-PCR) of HBMSCs derived engineered cartilage integrated to HC-secreted cartilage matrix at a region of 1 cm (proximal) and 4cm (distal) from the interface within the HBMSC construct. In samples both with and without HA, a high expression of Aggrecan, SOX9 and Collagen Type II at both the proximal and distal positions was found, indicative of a healthy articular cartilage phenotype. In addition, in the samples with HA, high gene expression of MMP13, Runx2 Collagen Type X and significant expression of Osteocalcin (p < 0.05) at the proximal location was found. The absence of these bone ECM genes thus preserved the cartilage phenotype when HA was not incorporated in healthy engineered cartilage matrix. (b) q(RT-PCR) of HBMSCs derived engineered cartilage integrated with HCOA-secreted cartilage matrix at a region of 1 cm (proximal) and 4cm (distal) from the interface within the HBMSC portion. In samples with HA, a significantly lower expression of Collagen Type I (p < 0.05) at the proximal location was found, i.e., preservation of the articular cartilage phenotype (while still maintaining robust expression of Aggrecan, SOX9 and Collagen Type II). In the specimens with HA, distal locations deep within the HBMSC-derived engineered tissue did not or negligibly expressed MMP13, Runx2 and Collagen X (p < 0.05). We speculate that the presence of HA at the interface and the subsequent creation of a calcium phosphate-rich transition zone served to potentially further reduce the spread of osteoarthritic conditions to the
In the HBMSCs derived cartilage from HCOAs Agar-HBMSCs PEGDA-HA group, it was found that there was significantly lower gene expression (p < 0.05) of Collagen Type I in both the proximal and distal regions from the interface (
Over the past decade, studies have focused on augmenting the integration of engineered cartilage to bone [
We found a significantly higher shear strength (p<0.05;
A narrowing of the gap in both HCs Agar-HBMSCs PEGDA- No HA and HCs Agar-HBMSCs PEGDA-HA samples was observed after 28 days of tissue culture (
On the other hand, after 28 days of culture in HCOAs Agar-HBMSCs PEGDA-HA samples, we observed the formation of a thin transition zone (
To further evaluate the effects of a more stable interface via HA incorporation in diseased environments, we conducted a preliminary
We emphasize however that the limitations of the explant studies are that it is preliminary and we can for the moment, only conjecture on enhanced stability indirectly based on the higher concentrations of cells within the defect, i.e., more cells will create more ECM, enabling greater tissue filling, hence greater repair and hence, stability. On the other hand the
We observed
Finally in osteoarthritic environments, in the specimens with HA, distal locations deep within the HBMSC-derived engineered tissue did not or negligibly expressed MMP13, Runx2 and Collagen X (p < 0.05;
Even though our results appear promising and our HA-based protocol could potentially be transferred directly to clinical photopolymerizable tissue engineered cartilage strategies currently being investigated [
In summary, we were able to demonstrate the utility of HA particles for chondral-chondral integration when the chondrocyte-derived cartilage matrix is osteoarthritic. A calcified cartilage matrix manifests itself between HBMSC and HCOA derived engineered cartilage tissues in the form of a spatial transition zone, thereby forming a stronger and a more stable interface. From an
(XLSX)
(XLSX)
(XLSX)
(DOCX)
(XLSX)
(XLSX)
Financial support for this study was provided by the College of Engineering and Computing at Florida International University (FIU). Additionally, a Dissertation Year Fellowship to support Rupak Dua, provided by the university graduate school at FIU is gratefully acknowledged. The authors sincerely acknowledge the equipment and services provided by the Advanced Materials Engineering Research Institute (AMERI) at FIU, related to the FEG-SEM with EDS capabilities. Finally, the authors wish to thank Dr. Nicolas Tsoukias for permitting access and use of facilities in his laboratory.