Fabrication and Characterization of Layer-By-Layer Janus Base Nano-Matrix to Promote Cartilage Regeneration

Summary

This protocol describes the assembly of a layer-by-layer Janus base nano-matrix (JBNm) scaffold by adding Janus base nanotubes (JBNts), matrilin-3, and Transforming Growth Factor Beta-1 (TGF-β1) sequentially. The JBNm was fabricated and characterized; additionally, it displayed excellent bioactivity, encouraging cell functions such as adhesion, proliferation, and differentiation.

Abstract

Various biomaterial scaffolds have been developed to guide cell adhesion and proliferation in hopes to promote specific functions for in vitro and in vivo uses. The addition of growth factors into these biomaterial scaffolds is generally done to provide an optimal cell culture environment, mediating cell differentiation and its subsequent functions. However, the growth factors in a conventional biomaterial scaffold are typically designed to be released upon implantation, which could result in unintended side effects on surrounding tissue or cells. Here, the DNA-inspired Janus base nano-matrix (JBNm) has successfully achieved a highly localized microenvironment with a layer-by-layer structure for self-sustainable cartilage tissue constructs. JBNms are self-assembled from Janus base nanotubes (JBNts), matrilin-3, and transforming growth factor beta-1 (TGF-β1) via bioaffinity. The JBNm was assembled at a TGF-β1:matrilin-3:JBNt ratio of 1:4:10, as this has been the determined ratio at which proper assembly into the layer-by-layer structure could occur. First, the TGF-β1 solution was added to the matrilin-3 solution. Then, this mixture was pipetted several times to ensure sufficient homogeneity before the addition of the JBNt solution. This formed the layer-by-layer JBNm, after pipetting several times again. A variety of experiments were performed to characterize the layer-by-layer JBNm structure, JBNts alone, matrilin-3 alone, and TGF-β1 alone. The formation of JBNm was studied with UV-Vis absorption spectra, and the structure of the JBNm was observed with transmission electron microscopy (TEM). As the innovative layer-by-layer JBNm scaffold is formed on a molecular scale, the fluorescent dye-labeled JBNm could be observed. The TGF-β1 is confined within the inner layer of the injectable JBNm, which can prevent the release of growth factors to surrounding areas, promote localized chondrogenesis, and promote an anti-hypertrophic microenvironment.

Introduction

Scaffolds in tissue engineering play a vital role in providing structural support for cell attachment and subsequent tissue development¹. Typically, conventional tissue constructs without any scaffolding rely on the cell culture environment and added growth factors to mediate cell differentiation. Furthermore, this addition of bioactive molecules into scaffolds is often the preferred approach in guiding cell differentiation and function^2,3. Some scaffolds can mimic the biochemical microenvironment of native tissues independently, while others can directly influence cell functions via growth factors. However, researchers often encounter challenges in selecting scaffolds that could positively affect cell adhesion, growth, and differentiation, while providing optimal structural support and stability over a long period^4,5. The bioactive molecules are often loosely bound to the scaffold leading to rapid release of these proteins upon implantation, resulting in their release in undesired locations. This culminates in side effects on tissues or cells that were not intentionally targeted^6,7.

Scaffolds are typically made of polymeric materials. The Janus base nano-matrix (JBNm) is a biomimetic scaffold platform created with a novel layer-by-layer method for self-sustainable cartilage tissue construct⁸. These novel DNA-inspired nanotubes have been named Janus base nanotubes (JBNts), as they properly mimic the structure and surface chemistry of collagen found in the extracellular matrix (ECM). With the addition of bioactive molecules, such as matrilin-3 and Transforming Growth Factor Beta-1 (TGF-β1), the JBNm can create an optimal microenvironment which can then stimulate desired cell and tissue functionality⁹.

JBNts are novel nanotubes derived from synthetic versions of the nucleobase adenine and thymine. The JBNts are formed through self-assembly¹⁰; six synthetic nucleobases bond to form a ring, and these rings undergo π-π stacking interactions to create a nanotube 200-300 µm in length¹¹. These nanotubes are structurally similar to collagen proteins; by mimicking an aspect of the native cartilage microenvironment, JBNts have been shown to provide a favorable attachment site for chondrocytes and human mesenchymal stem cells (hMSCs)^11,12,13,14. Because the nanotubes undergo self-assembly and do not require any sort of initiator (such as UV-light), they show exciting potential as an injectable scaffold for hard-to-reach defect areas¹⁵.

Matrilin-3 is a structural extracellular matrix protein found in cartilage. This protein plays a significant role in chondrogenesis and proper cartilage function^16,17. Recently, it has been included in biomaterial scaffolds, encouraging chondrogenesis without hypertrophy^9,18,19. By including this protein in the JBNm, cartilage cells are attracted to a scaffold that contains similar components to that of its native microenvironment. Additionally, it has been shown that matrilin-3 is needed for proper TGF-β1 signaling within chondrocytes²⁰. Growth factors function as signaling molecules, causing specific growth of a certain cell or tissue. Thus, to achieve optimal cartilage regeneration, matrilin-3 and TGF-β1 are essential components within the JBNm. The addition of TGF-β1 into the layer-by-layer scaffold can further promote cartilage regeneration in a tissue construct. TGF-β1 is a growth factor employed to encourage the healing process of osteochondral defects, encouraging chondrocyte and hMSC proliferation and differentiation^21,22. Thus, TGF-β1 plays a key role in the cartilage regeneration JBNm (J/T/M JBNm)²³, encouraging proper growth especially when it is localized within the JBNm layers.

As mentioned previously, growth factors are typically assembled on the outside of scaffolds with no specific methods of incorporation. Here, with the precisely designed nano-architecture of the biomaterials, the JBNm was developed for specific targeting of intended cells and tissues. The JBNm is composed of TGF-β1 adhered on JBNt surfaces in the inner layer and matrilin-3 adhered on JBNt surfaces in the outer layer^24,25. The incorporation of TGF-β1 in the inner layer of the layer-by-layer structure allows for a highly localized microenvironment along the JBNm fibers, creating a homeostatic tissue construct with a much slower release of the protein¹². The injectability of the JBNm makes it an ideal cartilage tissue construct for various future biomaterial applications²⁶.

Protocol

1. Synthesis of JBNts

1. Prepare the JBNt monomer utilizing previously published methods, involving the synthesis of a variety of compounds¹².
2. Purify crude JBNt monomer after it has been synthesized with high-performance liquid chromatography (HPLC) using a reverse-phase column. Use solvent A: 100% water, solvent B: 100% acetonitrile, and solvent C: HCl water solution with pH = 1. Use a flow rate of 3 mL/min. Collect the largest peak obtained in the HPLC at 7.2 min.

2. Fabrication for JBNt/Matn1/TGF-β1 (Video 1)

NOTE: Video 1 shows that the JBNm is an injectable solid formed in a physiological environment (water solution, no UV light, no chemical additives, and no heating), which is also biologically inspired.

1. Add 8 µL of 1 mg/mL TGF-β1 suspended in H₂O to 32 µL of 1 mg/mL matrilin-3 suspended in H₂O. Pipette to ensure proper mixing of these proteins.
2. Add 80 µL of 1 mg/mL JBNts suspended in H₂O to the TGF-β1/matrilin-3 solution. Pipette repeatedly to ensure proper blending. The presence of white floccules can be observed immediately after the addition of JBNts, indicating the formation of the JBNm (Video 1).

3. Observation of specimens with ultraviolet-visible (UV-Vis) absorption

NOTE: The UV-Vis absorption spectra were studied to characterize the assembly of the JBNm. This measurement was analyzed for four categories: JBNts alone, matrilin-3 alone, TGF-β1 alone, and the full layer-by-layer JBNm, comprised of all three parts. All initial concentrations are suspended in H₂O.

1. For the JBNt group, add 5 µL of 1 mg/mL JBNts to 50 µL of H₂O to make a 90.9 µg/mL solution.
2. For the matrilin-3 group, add 40 µL of 10 µg/mL matrilin-3 to 15 µL of H₂O to make a 7.3 µg/mL solution.
3. For the TGF-β1 group, add 10 µL of 10 µg/mL TGF-β1 to 45 µL of H₂O to make a 1.8 µg/mL solution.
4. For the JBNm group, add 40 µL of 10 µg/mL matrilin-3 to 10 µL of 10 µg/mL ^TGF-β1. Agitate to ensure proper mixing, and then add 5 µL of 1 mg/mL JBNts to the solution, pipetting repeatedly to mix the sample.
5. Using a spectrophotometer, measure the absorption spectrum of each group.

4. Zeta-potential measurement of specimens

NOTE: The zeta-potential was analyzed to better predict how the JBNm would interact with in vivo tissue. Three groups were measured: matrilin-3 alone, matrilin-3 with TGF-β1, and the full layer-by-layer JBNm.

1. For the matrilin-3 group, add 160 µL of 10 mg/mL matrilin-3 to 640 µL of H₂O to make an 800 µL solution.
2. For the matrilin-3/TGF-β1 compound, add 160 µL of 10 µg/mL matrilin-3 to 40 µL of 10 µg/mL TGF-β1. Pipette repeatedly to ensure proper homogeneity. Then, add it to 600 µL of H₂O resulting in an 800 µL solution.
3. For the JBNm group, add 160 µL of 10 µg/mL matrilin-3 to 40 µL of 10 µg/mL TGF-β1. Pipette multiple times to mix the proteins. Then, add 20 µL of 1 mg/mL JBNts to the solution and pipette to ensure homogeneity. Finally, add the JBNm solution to 580 µL of H₂O to produce an 800 µL solution.
4. Measure the zeta-potential values for the three groups.

5. Preparation of JBNt/matrilin-3 nano-matrix for transmission electron microscopy (TEM)

NOTE: TEM characterization is performed to characterize the morphology of JBNts and JBNm.

1. Insert two grids into a plasma cleaner to properly clean the grids before the negative staining of the JBNts and JBNm according to the published protocols and manufacturer's protocols¹².
2. Create a 200 µg/mL JBNt solution by mixing 10 µL of 1 mg/mL JBNts with 40 µL of distilled water.
3. Mix 30 µL of 100 µg/mL matrilin-3 with 20 µL of 100 µg/mL TGF-β1 and pipette several times. Add 10 µL of 1 mg/mL JBNts to the matrilin-3/TGF-β1 mixture to prepare the JBNm sample. Again, pipette repeatedly.
4. Add 3 µL of the JBNt solution (200 µg/mL) and 3 µL of the JBNm solution on separate grids, and leave them for 2 min.
5. Rinse each grid with 100 µL of uranyl acetate solution (0.5%). Use filter papers to remove the excess solution and allow the grids to air dry.
6. Operate a transmission electron microscope to properly observe and characterize the samples, as published previously^11,12.

6. Absorption spectra measurement of fluorescent-labeled proteins

NOTE: The structure of JBNm is verified by observing the structures of the JBNm with absorption spectral analysis.

1. Label the TGF-β1 by using a protein labeling kit following the manufacturer's instructions. Add 20 µL of 20 µg/mL labeled TGF-β1 to 25 µL of H₂O to make an 8.9 µg/mL test solution.
2. Label the matrilin-3 by using a separate labeling kit. Add 20 µL of 80 µg/mL labeled matrilin-3 to 25 µL of H₂O to make a 36 µg/mL test solution.
   NOTE: The fluorescent dye labeling of the TGF-β1 resulted in a final concentration of 20 µg/mL, while the labeling of the matrilin-3 resulted in a final concentration of 80 µg/mL.
3. Mix 20 µL of 80 µg/mL labeled matrilin-3 with 20 µL of 20 µg/mL labeled TGF-β1 and 5 µL of H₂O, resulting in a labeled TGF-β1/matrilin-3 compound. Pipette the mixture several times to mix.
4. Add 5 µL of 1 mg/mL JBNts to 40 µL of H₂O, resulting in a 111 µg/mL solution.
5. Add 20 µL of 20 µg/mL labeled TGF-β1 to 20 µL of 80 µg/mL labeled matrilin-3 and pipette several times. Add 5 µL of 1 mg/mL JBNts to the labeled TGF-β1/ matrilin-3 solution, and pipette repeatedly to properly mix the compound.
   NOTE: The final concentrations of the JBNt, labeled TGF-β1, and labeled matrilin-3 samples were 111 µg/mL, 8.9 µg/mL, and 36 µg/mL, respectively.
6. Transfer each sample group (i.e., labeled TGF-β1 (step 6.1), labeled matrilin-3 (step 6.2), labeled TGF-β1/matrilin-3 (step 6.3), JBNts (step 6.4), JBNm (step 6.5)) to their own well of a black 384-well plate.
7. Load the black 384-well plate into a multi-mode microplate reader and take measurements at excitation wavelengths of 488 nm and 555 nm following the manufacturer's protocol.

7. In vitro biological function assay

1. Cell adhesion test on pre-coated coverglass chambers
   1. Prepare two chambered coverglasses with five groups of samples, as one coverglass is used for each cell type.
   2. Add 1.25 µL of 1 mg/mL JBNts to 198.75 µL of distilled water, resulting in a 6.25 µg/mL solution.
   3. Add 10 µL of 10 µg/mL matrilin-3 to 190 µL of distilled water, resulting in a 0.5 µg/mLsolution.
   4. Add 2.5 µL of 10 µg/mL TGF-β1 to 197.5 µg/mLof distilled water, resulting in a 0.125 µg/mLsolution.
   5. For the JBNm group, add 10 µL of 10 µg/mL matrilin-3 to 2.5 µL of 10 µg/mLTGF-β1 and pipette repeatedly. Add 1.25 µL of JBNts with a concentration of 1 mg/mLto the mixture solution and pipette. Finally, add 186.25 µL of distilled water to result in a 200 µL JBNm solution.
      1. For the control group, use 200 µL of distilled water.
   6. Add each sample group into their own well of a No. 1.5 chambered coverglass. Place the coverglass into a -80 °C freezer for 1 h, and then freeze-dry it with a lyophilizer instrument.
   7. Seed human mesenchymal stem cells (hMSCs) and human chondrocyte cells (10,000 cells per well) into each well of the prepared chambered coverglasses.
   8. Incubate the two coverglasses for 4 h in a 37 °C incubator. Then, aspirate the cell culture medium and rinse twice with PBS.
   9. Fix cells with 4% paraformaldehyde for 5 min, and then remove and rinse samples with PBS. Rinse the sample a second time to completely remove the 4% paraformaldehyde.
   10. Incubate cells with 100 µL of 0.1% Triton-X for 10 min and wash with PBS. Add 100 µL of 0.165 µM rhodamine-phalloidin to each well. Wait for 30 min and then wash with PBS again.
   11. Add 0.1 µg/mL DAPI to each well to stain the nuclei. Wait for 5 min and rinse the wells with PBS twice.
   12. Use a spectral confocal microscope to observe cell morphology and capture fluorescent images.
   13. Further, analyze the cell number and morphology using image processing software. Open the software and load the images. Add a scale bar and calibrate the software. Count the number of cells in a given area for each sample. Then, use the measuring tool to collect data on cell morphology for each sample.
2. Cell proliferation
   1. Prepare three untreated 96-well plates, adding five groups of samples to each.
      1. For the JBNt group, add 3.75 µL of 1 mg/mL JBNts to 596.25 µL of distilled water, resulting in a 600 µL JBNt solution with a concentration of 6.25 µg/mL.
      2. For the TGF-β1 group, add 7.5 µL of 10 µg/mL TGF-β1 to 592.5 µL of distilled water, resulting in a 0.125 µg/mL test solution.
      3. For the matrilin-3 group, add 30 µL of 10 µg/mL matrilin-3 to 570 µL of distilled water, resulting in a 0.5 µg/mL test solution.
      4. For the JBNm group, mix 7.5 µL of 10 µg/mL TGF-β1 with 30 µL of 10 µg/mL matrilin-3 and pipette several times, followed by adding 3.75 µL of 1 mg/mL JBNts to the solution.
      5. Dilute JBNm solution with 558.75 µL of distilled water to obtain the final concentrations of 6.25 µg/mL, 0.125 µg/mL, and 0.5 µg/mL, respectively, for JBNt, TGF-β1, and matrilin-3 samples.
      6. For the control group, use 600 µL of distilled water.
   2. Divide each sample group into six wells (100 µL sample per well).
   3. Place the three plates containing all samples into a -80 °C freezer for 1 h and then freeze-dry with a lyophilizing instrument.
   4. Seed hMSCs onto these plates, each receiving a 100 µL cell suspension (per well) containing 5,000 cells. Incubate the three plates at 37 °C for 1 day, 3 days, or 5 days at 5% CO₂.
   5. Add 10 µL of CCK-8 solution to each well with the cells and incubate at 37 °C for another 2 h.
   6. Measure the absorption values of each well with multi-mode microplate readers at 450 nm.
   7. Seed a known gradient of hMSCs onto a 96-well plate and incubate for 4 h. Use the CCK-8 assay to measure the absorption values of the known number of cells and generate a standard curve.
   8. Calculate cell proliferation by using the absorption standard curve.
3. Stability test
   NOTE: A human TGF-β1 targeted ELISA kit was used to determine the percentage release of TGF-β1 from the JBNm in an agarose hydrogel.
   1. Prepare 2% agarose by adding 400 mg of agarose powder to 20 mL of PBS and heating it up to 100 °C to completely dissolve the agarose powder.
   2. Prepare the JBNm inside the cooled 2% agarose hydrogel.
      1. Combine 10 µL of 10 µg/mL TGF-β1, 40 µL of 10 µg/mL matrilin-3, 5 µL of 1 mg/mL JBNts, and 195 µL of PBS. Mix this solution with 250 µL of 2% agarose, creating the JBNm hydrogel.
   3. Add 500 µL of PBS, using it as a release solution, and ensure to change it every 3 days. Keep every released solution, and let the sample sit for 15 days.
   4. Utilize an ELISA kit to test each of the captured release solutions, by following the manufacturer's instructions.
      NOTE: By subtracting the amount of TGF-β1 released in the PBS from the theoretical loading value of 100%, the remaining amount of TGF-β1 can be determined.
4. Cell differentiation analysis with PCR²⁶.
   1. Prepare seven groups of samples with each sample containing 20 µL of cell suspension (4 x 10⁴ cells), 30 µL of the specific solution (or PBS, depending on the sample group), and 50 µL of 2 wt % agarose.
      1. For the TGF-β1 group, add 5 µL of 10 µg/mL TGF-β1 to 25 µL of PBS, resulting in a 1.6 µg/mL solution.
      2. For the matrilin-3 group, add 20 µL of 10 µg/mL matrilin-3 to 10 µL of PBS, resulting in a 6.7 µg/mL solution.
      3. For the JBNt group, add 2.5 µL of 1 mg/mL JBNts to 27.5 µL of PBS, resulting in an 83.3 µg/mL solution.
      4. For the J/T/M JBNm group, add 10 µL of 10 µg/mL matrilin-3 to 5 µL of 10 µg/mL TGF-β1 and pipette up and down to mix the solution. Then, add 2.5 µL of 1 mg/mL JBNts and pipette again. Finally, dilute with 2.5 µL of PBS.
      5. For each control group, use 30 µL of PBS as the sample.
   2. Add 0.5 mL of cell culture medium to each well, making sure to replace it every 3 days.
      1. For the positive control group, use a commercially available hMSC chondrogenic medium containing added TGF-β1. This additional TGF-β1 is equal to the amount in both the TGF-β1 and J/T/M JBNm groups.
      2. For one of the negative control groups, use a commercial hMSC chondrogenic medium that does not contain TGF-β1. For the other negative control group, use a DMEM cell culture medium containing 10% fetal bovine serum (FBS).
      3. For every other group, use a commercial hMSC chondrogenic cell culture medium without TGF-β1.
   3. Culture the samples for 15 days.
   4. Extract the samples' RNA and perform PCR to study the differentiation of the samples as described previously²⁶.
5. Immunostaining for type X collagen expression assay
   1. Prepare seven agarose-based samples in the same way as the cell differentiation with the PCR method section (step 7.4).
   2. Culture the samples for 15 days.
   3. Fix the samples with 4% formaldehyde for 1 day, soak in 30% sucrose solution overnight, and then freeze overnight at -80 °C with liquid nitrogen. Fix the samples using the optimal cutting temperature compound reagent (OCT), a blend of water-soluble glycols and resins, at -10 °C.
   4. Operate a cryostat microtome to obtain 20 µm thick frozen sections of each sample.
   5. Stain each section with an Anti-Collagen X antibody with 1:800 dilution and a fluorescent labeling secondary antibody diluted with PBS, according to the manufacturer's protocols.
   6. Observe each section with a confocal microscope, using an excitation of 488 nm and a magnification of 40x on the eyepiece.

Representative Results

Following protocol, JBNts were successfully synthesized and characterized with UV-Vis absorption and TEM. The JBNm is an injectable solid scaffold that undergoes a rapid biomimetic process. After JBNts were added to a mixture of TGF-β1/matrilin-3 solution in a physiological environment, a solid white-mesh scaffold was formed indicating the successful assembly of JBNm, as seen in Figure 1. This was demonstrated in the characterization methods.

Under physiological conditions, matrilin-3 is negatively charged due to its isoelectric point²⁷ (Figure 2A). After the addition of the TGF-β1 solution to the matrilin-3 solution, the zeta-potential of the TGF-β1/matrilin-3 compound increased to a near-neutral value, indicating that the two proteins were bound via charge interactions. As seen in Figure 2A, the zeta-potential of JBNm is the highest among the three groups owing to the JBNts, as the isoelectric point of the lysine side chain is around 9.74 in a physiological environment. The increase in zeta-potential values of the JBNm indicates the successful assembly of its layer-by-layer structure.

UV-Vis absorption spectra (Figure 2B) clarify the formation of the hierarchical layer-by-layer interior structure of the JBNm. The aromatic rings of the lysine side chains and the JBNts contributed to the two absorption peaks at 220 nm and 280 nm, respectively. The decrease in absorption value was observed after the addition of TGF-β1, indicating that binding occurs between TGF-β1 and JBNts. After the addition of JBNts to matrilin-3, a more obvious decrease in the absorption intensity of the peaks was observed, once again indicating successful binding between matrilin-3 and JBNts. Similarly, after the addition of JBNts to a mixture of TGF-β1/matrilin-3, the JBNm was formed, and the absorption peaks decreased in intensity. The absorption peak of JBNm is closer to the matrilin-3/JBNts line than the TGF-β1/JBNts line, indicating that the JBNts prefer to bind to matrilin-3, forming a layer-by-layer outer structure with matrilin-3 while TGF-β1 resides in the inner layer. TEM is used to characterize the morphology of the JBNts and JBNm (Figure 2C). After combining with proteins, thick bundles of JBNm were observed forming a scaffold structure.

Fluorescence microscopy has confirmed the presence of the layer-by-layer structure (Figure 3A) and demonstrated the cross-section of the JBNm. After labeling TGF-β1 and matrilin-3, it was observed that the red fluorescent matrilin-3 envelops the JBNt bundles, forming the outer layer of the JBNm. In Figure 3B, the green-fluorescent TGF-β1 formed an inner layer, contributing to the capability of storing growth factors and allowing the localization of TGF-β1. Figure 3C displays the fluorescence resonance energy transfer (FRET) process between the fluorescent dye-labeled proteins as characterized by fluorescence spectra, with emission peaks at 520 nm and 570 nm for labeled TGF-β1 and matrilin-3 groups, respectively²⁸. After the addition of JBNts according to the protocol, the layer-by-layer structure is formed; peaks were observed at both 520 nm and 570 nm for the JBNm group, indicating a successful binding and assembly of the proteins and JBNts.

Additionally, the effect of the JBNm on hMSCs adhesion and cell proliferation was explored. As shown in Figure 4, the cell adhesion density of the JBNm coated on the surface of the chambered coverglasses was determined. The JBNm showed hMSCs clustered along itself (Figure 4A), whereas fewer cells adhered to JBNts. However, for the other groups, the hMSCs were evenly distributed without alignment as compared to the JBNm group. The alignment of cells, as well as the size of the cells on the JBNm, demonstrated that JBNts played a role in cell adhesion, while the proteins in the JBNm increased the affinity of cell adhesion (Figure 4B,C).

After day 1 of cell culture with the JBNm, JBNts, matrilin-3 alone, TGF-β1 alone, and negative control, the JBNm and TGF-β1 groups showed significant cell proliferation when compared to the other groups. When the cell culture duration was increased to 3 and 5 days, the JBNm and TGF-β1 groups demonstrated even more increased cell proliferation, as seen in Figure 5. A long-term function study was performed to determine the differentiation of hMSCs with JBNm and without JBNm (negative control). After 15 days, an enhanced Alcian blue stain was observed, indicating chondrogenesis of hMSCs growing alongside JBNm²⁶. Thus, the cells preferred the JBNts of the JBNm. This was likely due to its DNA-mimicking structure and ability to be disassembled into small-molecule units, the latter of which can be triggered by a low pH or sufficient enzymatic activity (such as uptake by cells)^14,29.

Video 1: Video recording of JBNm assembly. This recording depicts the formation of the JBNts, matrilin-3, and TGF-β1 into the JBNm. This figure has been modified from Zhou et al. (2021)²⁶. Please click here to download this Video.

Figure 1: Schematic illustration of the chemical structure of JBNts, the J/T/M JBNm, and the J/T/M JBNm's chondrogenic ability. (A) The chemical structure of the JBNts, highlighting their formation from monomer to rosette ring to JBNt. (B) The components that make up the J/T/M JBNm, as well as how they assemble into the JBNm. (C) Schematic of culturing human mesenchymal stem cells on the J/T/M JBNm. This figure has been modified from Zhou et al. (2021)²⁶. Please click here to view a larger version of this figure.

Figure 2: Material characterization of the J/T/M JBNm. (A) Zeta-potential spectra of matrilin-3 alone, the TGF-β1/matrilin-3 compound, and the J/T/M JBNm. (B) UV-Vis absorption spectra of JBNts alone, matrilin-3 alone, the TGF-β1 alone, matrilin-3/JBNts compound, the TGF-β1/JBNts compound, and the J/T/M JBNm. (C) Images of JBNts alone and the J/T/M JBNm obtained from transmission electron microscopy (TEM). This figure has been modified from Zhou et al. (2021)²⁶. Please click here to view a larger version of this figure.

Figure 3: Fluorescent confocal imaging and fluorescence spectra of J/T/M JBNm. (A) 3D confocal image of the J/T/M JBNm, with red-fluorescent-labeled matrilin-3 and green-fluorescent-labeled TGF-β1. (B) 2D confocal images of J/T/M JBNm, with red-fluorescent-labeled matrilin-3, green-fluorescent-labeled TGF-β1, and a merged version. (C) Fluorescence spectra of various compounds to characterize the FRET process between the labeled proteins. This figure has been modified from Zhou et al. (2021)²⁶. Please click here to view a larger version of this figure.

Figure 4: Analysis of human mesenchymal stem cell (hMSC) culture. (A) Optical microscopy images of hMSCs with TGF-β1, matrilin-3, and JBNts, cultured on an agarose gel. (B) Confocal microscopy images of hMSCs cultured on the chambered cover glass coated with a variety of JBNm components. (C) Graph of the number of cells adhered per square millimeter for each group. Error bars denote standard deviations. (D) Graph of cell major axis length in µm per group. Error bars denote standard deviations. Note: N ≥ 3, *P< 0.05, **P< 0.01, ***P< 0.001, ****P< 0.0001 (compared against negative controls, NC). This figure has been modified from Zhou et al. (2021)²⁶. Please click here to view a larger version of this figure.

Figure 5: Comparison of cell numbers for various groups between days 1, 3, and 5. Cell number statistics of hMSCs after being cultured with different materials on days 1, 3, and 5. Error bars denote standard deviations. Note: N = 6, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. This figure has been modified from Zhou et al. (2021)²⁶. Please click here to view a larger version of this figure.

Discussion

The goal of this study is to develop a biomimetic scaffold platform, the JBNm, to overcome the limitations of conventional tissue constructs that rely on cell culture environments to mediate cell differentiation. The JBNm is a layer-by-layer structure scaffold for a self-sustainable cartilage tissue construct. The innovative design is based on novel DNA-inspired nanomaterials, the JBNts. The JBNm, composed of JBNts³⁰, TGF-β1, and matrilin-3, is assembled through a novel layer-by-layer technique where the self-assembly of the scaffold was controlled at the molecular level. The assembly of JBNm was observed and characterized with zeta-potential measurement, UV-Vis absorption, TEM, and fluorescence spectral analysis.

The UV-Vis spectra in Figure 2B (step 3) have demonstrated the formation of the hierarchical layer-by-layer interior of the JBNm. A difference in absorbance peaks can be observed due to a difference in binding affinity. More major interactions occur between JBNts and matrilin-3 than between JBNts and TGF-β1 indicating that JBNts mimic collagen proteins in terms of their fibrous morphology and lysine surface chemistry. This technique is necessary to observe the binding of JBNts to matrilin-3 rather than TGF-β1, allowing for the encapsulation of TGF-β1 and therefore providing a slow release of TGF-β1 into the surrounding tissue. The most critical step in the development of the JBNm is the combination of proteins with the JBNts. TGF-β1 and matrilin-3 must be added prior to the addition of JBNts to observe the full effect of the FRET phenomenon between the labeled matrilin-3 and TGF-β1. The limitation of this technique stems from the incorrect addition sequence of proteins and nanotubes, resulting in varied measurements. This is an important step for the formation of JBNm for future studies, and, therefore, will be applied to future JBNm fabrications and studies.

Matrilin-3 is a highly negative protein, whereas TGF-β1 is a slightly neutral protein, as seen in Figure 2A. The charge difference was observed to determine the binding of these proteins. From the negatively charged state of matrilin-3, zeta-potential increased to a near-neutral value after the addition of TGF-β1 (Figure 2A). This step is important to determine the first inner layer of the biomimetic scaffold, and to indicate that the combination of both proteins is through charge interaction. After the addition of JBNts, which is the second step of the JBNm fabrication, the zeta-potential of JBNm was measured at ~10 ± 5 mV (Figure 2A). Because JBNts are highly positive, the charge of JBNm was observed to increase as expected. Determination of charges with zeta-potential is thus important to observe the formation of JBNm step-by-step via charge interactions. Similar to UV-Vis spectroscopy, the sequence of addition is important in this step, and incorrect addition order may result in varied measurements. Similarly, it is important to pipette the mixture up and down prior to measurement.

Next, TGF-β1 and matrilin-3 are labeled so that the development of JBNm can be observed with confocal microscopy and fluorescence spectral analysis with multimode microplate readers. The samples were excited with a 488 nm laser and showed emission peaks at 520 nm and 570 nm (Figure 3). No emission peak was observed for JBNts alone because they are not labeled. FRET occurred from the TGF-β1 donor to the matrilin-3 acceptor, thereby reducing the emission peak at 520 nm and increasing the emission peak at 570 nm; the two components are 10 nm apart in distance, allowing the FRET phenomenon to occur. Therefore, the assembly of JBNm is based on spatial (i.e., physical distance) processes together with charge interactions. The most critical step in the development of JBNm is the addition sequence; TGF-β1 and matrilin-3 must be added prior to the addition of JBNts to observe the full effect of the FRET phenomenon. Magnification of at least 40x on the eyepiece is required to observe the fluorescent JBNm. The limitation of this technique is that the JBNts are not labeled and, therefore, could not be observed with confocal microscopy. However, the protein labeling technique can be applied to label JBNt with some modifications for future applications.

JBNm can aid cell functions, such as cell adhesion and cell proliferation. In this study, hMSCs were cultured (starting with 5,000 cells) on different materials (JBNm, JBNts, TGF-β1, matrilin-3, and negative controls without additives) to compare the cell numbers after incubation for 1, 3, and 5 days using CCK-8 solution, following the exact manufacturer's protocols to determine cell proliferation. JBNm, especially in the presence of TGF-β1, and TGF-β1 alone showed significant cell proliferation compared to the other three groups. JBNts mimic collagen in the ECM morphologically while matrilin-3 is a cartilage-specific protein, resulting in increased cell proliferation activity after incubation at 37 °C for 3 and 5 days (Figure 5). The JBNm has shown great potential to serve as an injectable and biomimetic tissue engineering scaffold to overcome the limitations of traditional cell constructs for various future applications as biomaterials^29,31. JBNm mimics the cartilage ECM morphologically, providing adhesion sites and allowing the release of pro-chondrogenic differentiative factors into the microenvironment, such as TGF-β1^32,33. The ability of the JBNm to incorporate TGF-β1 in the inner layer of the scaffold prevents it from leaking to undesired areas, thus improving the effectiveness of the scaffold (Figure 4 and Figure 5). Many hMSCs are seen to cluster along the JBNm scaffold, while the cells are sparsely distributed in matrilin-3, TGF-β1, and negative control groups (Figure 4). The morphology of the cells was also observed, indicating excellent affinity with JBNm surfaces. These precisely designed biomaterials prevent the rapid release of incorporated bioactive molecules into the cell culture environment and improve the self-sustainability of the tissue constructs within the matrix³⁴. A limitation of this technique is that the starting cell numbers are critical for determining the proliferation of cells. A starting cell number of 5,000 was found most optimal for this study. A standard curve is also needed to determine the cell number by plotting a known number of cells and measuring them in the multimode microplate reader. This technique can be applied in the future as a standardized method to determine cell proliferation across all studies relating to JBNm scaffolds.

hMSCs were cultured in 3D agarose pellets with and without JBNm to determine the long-term function study for 15 days. After 15 days, total RNA was extracted from the hMSCs in the agarose pellets, containing positive control (pellets were supplied with fresh TGF-β1 every time medium was changed), JBNm, JBNts, matrilin-3, TGF-β1, and no additives. Real-time qPCR was carried out for gene analysis, testing out for chondrogenic differentiation markers, Aggrecan (ACAN), and hypertrophy marker type X collagen (COL X). ACAN expression in the JBNm group increased significantly compared to other groups, demonstrating that JBNm promotes stem cell chondrogenic differentiation significantly while inhibiting COL X. Meanwhile, the positive control has enhanced chondrogenesis, as noted in the increase in ACAN, and also hypertrophy of the differentiated cell²⁶. A further limitation of this study is that the RNA extraction is from the cells embedded in a hydrogel. The total RNA obtained in this protocol, therefore, was low in concentration and purity. To overcome this limitation, more samples were cultured and extracted to obtain the optimal concentration and purity for the next step, real-time qPCR. While qPCR is important for the study, minor modification of this technique is necessary for future studies to ensure efficient sample extraction without sacrificing a large number of samples.

A stability test was performed with a human TGF-β1 ELISA kit to test the release of protein from the JBNm hydrogel. In this study, TGF-β1 is expected to be encapsulated in the J/T/M JBNm, and therefore no TGF-β1 release should be observed (i.e., the theoretical loading value is 100%). A limitation of this step is that all TGF-β1 is assumed to be encapsulated into the JBNm layer-by-layer scaffold. After 15 days, solutions released from the hydrogel are collected and tested with the kit, following the manufacturer's protocols. Then, the amount of TGF-β1 was calculated by subtracting the amount of the released TGF-β1 in PBS from the theoretical loading value. Because TGF-β1 was encapsulated in the inner layer of the JBNm, the slow release of the protein was anticipated. The study demonstrated that the TGF-β1 is localized within the JBNm scaffold and does not release rapidly into the surrounding areas²⁶.

This innovative layer-by-layer JBNm cartilage tissue construct was realized by highly organized and controlled self-assembly on the molecular level. The TGF-β1 is confined in the inner layer of the matrix fibers, preventing its leakage to undesired locations, and promoting localized chondrogenesis simultaneously. In addition, matrilin-3 is localized in the outer layer of the matrix fibers, creating an anti-hypertrophic microenvironment³⁵. JBNts have been shown to not only serve as a structural scaffold backbone but also enhance stem cell anchorage and adhesion to localize cells along the matrix fibers^30,36. As for future work, the layer-by-layer design of the JBNt-based scaffold will be customized for applications in various tissues^37,38,39.

Materials

Name	Company	Catalog Number	Comments
10 % Normal Goat Serum	Thermo Fisher	50062Z	Agent used to block nonspecific antibody binding actions during staining.
24-well plate	Corning	07-200-740	24-well plate used for comparative cell culture.
384-Well Black Untreated Plate	Thermo Fisher	262260	384-well plate used for absorption measurements.
8-well chambered coverglass	Thermo Fisher	155409PK	8-well coverglass used for comparative cell culture.
96-well flat bottom	Corning	07-200-91	96-well plate used for comparative cell culture.
96-Well Plate non- treated	Thermo Fisher	260895	96-well plate used for comparative cell culture and analysis.
Agarose Gel	Sigma-Aldrich	A9539	Hydrogel used for cell culture.
Agarose Gel	Sigma Aldrich	A9539	Hydrogel used as an environment for cell culture.
Alexa Fluor Microscale Protein Labeling Kit	Thermo Fisher	A30006 (488) and A30007 (555)	Fluorescent dye used to label proteins.
Anti-Collagen X Antibody	Thermo Fisher	41-9771-82	Antibody used to stain collagen-X.
Bio-Rad PCR Machine	Bio-Rad		Equipment used to perform PCR on samples.
C28/I2 Chondrocyte Cell Line			Cells used to analyze proliferative abilities of various samples.
Cell Counting Kit 8	Milipore Sigma	96992	Cell proliferation assay.
Cell Profiler	Broad Institute		Software used to analyze cell images.
Cryostat Microtome			Equipment used to produce thin segments of samples for use in staining and microscopy.
DAPI	Invitrogen	D1306	Blue fluorescent stain that binds to adenine-thymine DNA regions.
Disposable cuvettes	FISHER Scientific	14-955-128	Container used for spectrophotometry.
DMEM Cell Culture Medium	Thermo Fisher	10566032	Media used to support cellular growth.
Fetal Bovine Serum	GIBCO	A4766801	Serum used in cell culture medium to support cell growth.
Fluoromount-G Mounting Medium	Thermo Fisher	00-4958-02	Solution used to mount slides for immunostaining.
Formaldehyde			Compound used to fix samples prior to microtoming.
Goat anti-Rabbit IgG (H+L) Highly Cross-Adsorbed Secondary Antibody	Thermo Fisher	A16110	Antibody used for protein staining.
Human Mesenchymal Stem Cells	LONZA	PT-2501	Cells used to analyze differentiative abilities of various samples.
Human Mesenchymal Stem Chondrogenic Medium	LONZA	PT-3003	Cell medium used to promote chondrogenic differentiation.
ImageJ	National Institutes of Health		Image analysis software used in conjunction with microscopy.
itaq Universal SYBR Green One-Step Kit	BioRad	1725150	Kit used for PCR.
Janus-base nanotubes (JBNts)			Nanotube made from synthetic nucleobases to act as cell scaffolding tool.
LaB6 20-120 kV Transmission Electronic Microscope	Tecnai		Equipment used to perform transmission electron microscopy on a sample.
MATLAB	MathWorks		Statistical software used for modeling and data analysis.
Matrilin-3	Fisher Scientific	3017MN050	Structural protein used as adhesion sites for chondrocytes.
NanoDrop Spectrophotometer	Thermo Fisher		Equipment used to measure absorption values of a sample.
Nikon A1R Spectral Confocal Microscope	Nikon	A1R HD25	Confocal microscope used to analyze samples.
Number 1.5 Chamber Coverglass	Thermo Fisher	152250	Environment for sterile cell culture and imaging.
Optimal Cutting Temperature Compound Reagent			Compound used to embed cells prior to microtoming.
Paraformaldehyde	Thermo Scientific	AAJ19943K2	Compound used to fix cells.
PDC-32G Plasma Cleaner	Harrick Plasma		Cleaner used to prepare grids prior to transmission electron microscopy.
penicillin-streptomycin	GIBCO	15-140-148	Antibiotic agent used to discourage bacterial growth during cell culture.
Phosphate Buffered Saline	Thermo Fisher	10010023	Solution used to wash cell medium and act as a buffer during experimentation.
Rhodamine-phalloidin	Invitrogen	R415	F-Actin red fluorescent dye.
Rneasy Plant Mini Kit	QIAGEN	74904	Kit used to filter and homogenize samples during RNA extraction.
Sucrose Solution			Solution used to process samples prior to microtoming.
TGF beta-1 Human ELISA Kit	Invitrogen	BMS249-4	Assay kit used to determine the presence of TGF-β1 in a sample.
TGF-β1	PEPROTECH	100-21C	Growth factor used for the stimulation of chondrogenic differentiation and proliferation.
Triton-X	Invitrogen	HFH10	Compound used to lyse cells not fixed during staining process.
TRIzol Reagent	Thermo Fisher	15596026	Reagent used to isolate RNA.
Zetasizer Nano ZS	Malvern Panalytical		Equipment used to measure zeta-potential values of a sample.