The goal of this protocol is to show the assembly of a biomimetic nanomatrix (NM) with Janus base nanotubes (JBNTs) and fibronectin (FN). When co-cultured with human mesenchymal stem cells (hMSCs), the NMs exhibit excellent bioactivity in encouraging hMSCs adhesion.
A biomimetic NM was developed to serve as a tissue-engineering biological scaffold, which can enhance stem cell anchorage. The biomimetic NM is formed from JBNTs and FN through self-assembly in an aqueous solution. JBNTs measure 200-300 µm in length with inner hydrophobic hollow channels and outer hydrophilic surfaces. JBNTs are positively charged and FNs are negatively charged. Therefore, when injected into a neutral aqueous solution, they are bonded together via noncovalent bonding to form the NM bundles. The self-assembly process is completed within a few seconds without any chemical initiators, heat source, or UV light. When the pH of the NM solution is lower than the isoelectric point of FNs (pI 5.5-6.0), the NM bundles will self-release due to the presence of positively charged FN.
NM is known to mimic the extracellular matrix (ECM) morphologically and hence, can be used as an injectable scaffold, which provides an excellent platform to enhance hMSC adhesion. Cell density analysis and fluorescence imaging experiments indicated that the NMs significantly increased the anchorage of hMSCs compared to the negative control.
Human mesenchymal stem cells (hMSCs) have shown the potential for self-renewal and self-differentiation along different mesenchymal lineages, which helps in the regeneration and maintenance of tissues1. Based on the differentiation potential, hMSCs are considered as candidates for mesenchymal tissue injuries and hematopoietic disorder therapy2. hMSCs have shown the ability to promote wound healing by increasing tissue repair, angiogenesis, and reducing inflammation3. However, without biochemical or biomaterials assistance, the efficiency for the hMSCs to reach a target tissue and function at the desired location is low4. Although various engineered scaffolds have been utilized to attract hMSCs to adhere onto the lesions, some sites such as growth plate fracture, in the middle of a long bone, are not easily accessible by the conventional pre-fabricated scaffolds, which may not fit perfectly into an irregularly shaped injured site.
Here, we have developed a biomimetic nanomaterial that can self-assemble in situ and be injected to a hard to reach target area. The injectable bio-scaffold NM is composed of Janus base nanotubes (JBNTs) and fibronectin (FN). JBNTs, also known as the Rosette Nanotubes (RNTs), are derived from DNA base pairs, specifically thymine and adenine, here5,6,7. As seen in Figure 1, the nanotubes are formed when six molecules of the derived DNA base pairs self-assemble via hydrogen bonds to form a plane6. Six molecules are then stacked onto each other in a plane via a strong pi-stacking interaction7, which can be up to 200-300 µm in length. The JBNTs are designed to morphologically mimic collagen fibers so that FN will react with them.
FN is a high molecular weight adhesive glycoprotein, which can be found in the extracellular matrix (ECM)9. These can mediate the attachment of stem cells to other components of the ECM, particularly collagen10. We designed JBNTs to morphologically mimic collagen fibers so FN can react with them to form NM in a few seconds via noncovalent bonding. Therefore, NM is a promising bio-scaffold to be injected into a bone fracture site that could not be accessible by the conventionally fabricated scaffolds. Here, the injectable NM presents an excellent ability to enhance hMSC anchorage in vitro, exhibiting their potential to serve as a scaffold for tissue regeneration.
1. Synthesis of JBNTs
NOTE: JBNT monomer was prepared as published previously11.
2. Fabrication for JBNT/FN
3. Observation of lyophilized specimens
NOTE: This step is performed to show the scaffold structure of NM made from JBNTs and FN.
4. Absorption spectra measurement
NOTE: Use the change of spectra for FN and JBNTs to present the self-assembled JBNT/FN NM12.
5. Preparation of JBNT/FN NM for transmission electronic microscopy (TEM)
6. In vitro biological function assay
Our studies discovered that the formation of the NM of JBNTs and FN is fast, which happened in 10 seconds. As shown in Figure 2, white floccule was obtained when the JBNT solution was mixed with the FN solution and pipetted several times. The formation process of NM is completely biomimetic. No external stimuli are needed. The process of fabrication is much easier than that of some emerging biomaterials, which is based on ultraviolet light or chemical initiator for crosslinking13.
We captured and analyzed the camera images of FN, JBNTs, and the NM (Figure 3). For the FN group, except for short white spots, no long fibers were observed. The short clusters were the protein agglomerations consisting of FN, which indicates that the scaffold structure cannot be formed with FN alone without JBNTs (Figure 3A). As shown in Figure 3B, the long and thinner fibers indicate the existence of JBNTs. The width of the white strands of the NM is wider compared to JBNTs, indicating that the JBNTs and FN crosslinked with each other forming NM. The NM fiber can grow up to several centimeters in length (Figure 3C).
We obtained the UV-Vis spectra to indicate the assembly between JBNTs and FN. As shown in Figure 4, JBNTs exhibits two absorption peaks at 220 nm and 280 nm, respectively. Compared to JBNTs, the JBNT/FN NM has two absorption peaks at the same location with a significantly reduced value, which was from JBNTs but affected by the non-covalent self-assembled JBNTs and FN in between.
TEM was used to characterize the morphology of the JBNTs and NM. As shown in Figure 5A, the JBNTs are slender tubes with uniform diameters. Under neutral conditions, JBNTs are positively charged while FN are negatively charged. When mixed, they formed long fibroid NM via charge interactions (Figure 5B). When pH is lower than the isoelectric point of the FN (pI of 5.5-6.0), the NM bundles present self-release due to the positively charged FN. As shown in Figure 5C, when preserved in a solution with low pH (4.0), the NM was dissembled, and a lot of FN was released from the NM. The FN distributed alongside the nanotube is also a strong evidence that the NM was fabricated by JBNTs and FN10.
We explored the effect of the JBNT/FN NM on cell adhesion. As shown in Figure 6, the cell adhesion density of the NM group showed significant difference (p-value <0.05) value compared to the negative control. The difference may be because the NM are designed to morphologically mimic ECM, which provide a scaffold for cell adhesion14. The ECM was composed of collagens and cell adhesive glycoproteins, such as fibronectin15. Collagen has been presented with the ability to promote higher adhesion and proliferation of hMSCs16. Additionally, fibronectin has been shown to enhance the hMSCs adhesion in the injured site as well15. Fluorescence imaging also demonstrated that the hMSCs adherence of the NM group increased significantly compared to the negative control group (Figure 7)15,17.
Figure 1: Schematic illustration of the hierarchical self-assembly of JBNTs with a lysine side chain.
This figure is reprinted with permission from the previous publication12. Please click here to view a larger version of this figure.
Figure 2: Demonstration for the self-assembled process of the NM.
(A) FN solution. (B) JBNTs mixed with FN. This figure is reprinted with permission from the previous publication12. Please click here to view a larger version of this figure.
Figure 3: Brightfield photographs.
Brightfield photographs of (A) FN (B) JBNTs and (C) NM formed by JBNTs and FN. Each area has a diameter of 2 cm. This figure is reprinted with permission from the previous publication12. Please click here to view a larger version of this figure.
Figure 4: Absorption spectra of FN, JBNTs, and JBNT/FN NM.
This figure is reprinted from the previous publication12. Please click here to view a larger version of this figure.
Figure 5: TEM images.
TEM images of (A) JBNTs (B) JBNT/FN NM, and (C) released JBNT/FN NM. This figure is reprinted with permission from the previous publication12. Please click here to view a larger version of this figure.
Figure 6: Statistical analysis of cellular adhesion.
Cell adhesion density was recorded in this experiment. *p<0.01 compared to negative controls. **p<0.05 compared to JBNT alone. N=3. This figure is reprinted with permission from the previous publication12. Please click here to view a larger version of this figure.
Figure 7: Fluorescence images.
Fluorescence images of (A) the hMSCs and (B) the hMSC incubated with the JBNTs. (C) the hMSC incubated with the FN. (D) the hMSC incubated with the JBNT/FN NM. Scale bar: 100 μm. This figure is reprinted with permission from the previous publication12. Please click here to view a larger version of this figure.
Time | A% | B% | C% |
0.00 min | 98 | 0 | 2 |
15.00 min | 68 | 30 | 2 |
25.00 min | 8 | 90 | 2 |
26.00 min | 0 | 100 | 0 |
Table 1: Gradient elution time.
Supplemental File 1: Scheme for the synthesis of JBNT monomer. Please click here to download this figure.
Supplemental File 2: Video for self-assembly of the FN/JBNT NM. Please click here to download this video.
In this study, we developed a self-assembled biomimetic NM, which was formed with DNA-inspired JBNTs and FN. When preparing the JBNT solution, the JBNT lyophilized powder should be dissolved into the water instead of PBS because PBS will cause agglomeration of JBNTs, which inhibits their assembly. Moreover, the NM should also be assembled in water if we want to observe the nano-fibril structures of the NM, because the salt in PBS will bundle with NM fibers, which can greatly reduce the resolution of the images.
The NM has shown great potential to serve as a novel tissue engineering scaffold although a lot of efforts have been made to fabricate tissue regenerative scaffolds, which were used to facilitate hMSCs adhesion and differentiation. However, most of these materials are pre-made with a specific shape, such as a 3D printed scaffold18,19. As such, they are not easy to be implanted into injured sites that are deep in the joint. However, the NM solution here can be injected as a liquid and then serve as a scaffold for cell adhesion.
One limitation is that the NM is not mechanically strong. Unlike polymers, they are fabricated via self-assembly of noncovalent interactions, so they cannot bear a lot of mechanical loading or tension. On the other hand, the noncovalent structure also presents very good biodegradability and biocompatibility suitable for biological functions8,20,21.
The injectable NM has shown great potential to provide a target location and a scaffold for MSC homing to enhance bone fracture healing in the body. However, when injected into the injured site, the noncovalent NM may not stay in place for a long time, which may reduce the therapeutic effect. In the future, we will explore to incorporate other materials (such as hydrogels) or to alternate the NM structures to further optimize the properties of the NM scaffold.
The authors have nothing to disclose.
This work is financially supported by NIH (Grants 1R01AR072027-01, 1R03AR069383-01), NSF Career Award (1653702) and University of Connecticut.
1,2-dichloroethane | Alfa Aesar | 39121 | |
2-cyanoacetic acid | Sigma-Aldrich | C88505 | |
4-Dimethylaminopyridine | TCI America | D1450 | |
8 wells Chambered Coverglass | Thermo Fisher | 155409 | |
96-well plate | Corning | 353072 | |
absolute ethanol | Thermo Fisher | BP2818500 | |
acetone | Sigma-Aldrich | 179124 | |
acetonitrile | Sigma-Aldrich | 34851 | |
allylamine | Sigma-Aldrich | 145831 | |
Basic Plasma Cleaner | Harrick Plasma | PDC32G | |
citric acid | Sigma-Aldrich | 251275 | |
concentrated hydrochloric acid | Sigma-Aldrich | H1758 | |
Deionized water | Thermo Fisher | 15230147 | |
dichloromethane | Sigma-Aldrich | 270997 | |
diethyl ether | Sigma-Aldrich | 296082 | |
Di-tert-butyl dicarbonate | Sigma-Aldrich | 361941 | |
ethyl acetate | Sigma-Aldrich | 319902 | |
ethylcarbamate | Sigma-Aldrich | U2500 | |
Fibronectin | Thermo Fisher | PHE0023 | |
Fixative Solution (4 % formaldehyde prepared in PBS) | Thermo Fisher | R37814 | |
guanidinium hydrochloride | Alfa Aesar | A13543 | |
hexanes | Sigma-Aldrich | 227064 | |
Human mesenchymal stem cells | Lonza | PT-2501 | |
methanol | Sigma-Aldrich | 34860 | |
methyl iodide | Sigma-Aldrich | 289566 | |
N,N-Diisopropylethylamine | Alfa Aesar | A17114 | |
N,N-dimethylformamide | Sigma-Aldrich | 227056 | |
N-Methylmorpholine N-oxide | Alfa Aesar | A19802 | |
Osmium tetraoxide | Alfa Aesar | 45385 | |
Penicillin-Streptomycin | Thermo Fisher | 15140163 | |
Phosphate Buffer Solution | Thermo Fisher | 20012050 | |
phosphoryl chloride | Sigma-Aldrich | 201170 | |
potassium carbonate | Sigma-Aldrich | 347825 | |
reverse phase column | Thermo Fisher | 25305-154630 | |
Rhodamine Phalloidin | Thermo Fisher | R415 | |
silica gel | TCI America | S0821 | |
sodium bicarbonate | Sigma-Aldrich | S6014 | |
sodium ethoxide | Alfa Aesar | L13083 | |
sodium periodide | Sigma-Aldrich | 71859 | |
sodium sulfate | Sigma-Aldrich | 239313 | |
sodium sulfite | Sigma-Aldrich | S0505 | |
sodium triacetoxyborohydride | Alfa Aesar | B22060 | |
spectrophotometer(NanoDrop One/Oneᶜ UV-Vis) | Thermo Fisher | ND-ONE-W | |
Stem Cell Growth Medium BulletKit | Lonza | PT-3001 | |
tetrahydrofuran | Sigma-Aldrich | 401757 | |
thioanisole | Sigma-Aldrich | T28002 | |
toluene | Sigma-Aldrich | 179418 | |
triethylamine | Alfa Aesar | A12646 | |
trifluoroacetic acid | Alfa Aesar | A12198 | |
Triton X-100 | Thermo Fisher | HFH10 | |
Trypsin-EDTA solution | Thermo Fisher | 25200056 |