2.1 Design

The goal of the first cycle was to identify suitable collagen-binding domains (CBDs) for the sequential release of VEGF and BMP-4. Different CBDs were fused with Enhanced Green Fluorescent Protein (EGFP) to screen for those that minimally impact the bioactivity of the fusion proteins and can control the binding to collagen. We chose pET28a plasmid as the vector and Escherichia coli BL21 as the host cell. Below is the collagen binding module and plasmid design we selected.

Fig 2.1 Plasmid design

SPARC/OD-EGFP:
SPARC (Secreted Protein Acidic and Rich in Cysteine) and OD (Osteonectin Domain): This domain is involved in regulating the interaction between cells and the extracellular matrix (ECM), particularly binding to collagen and influencing matrix remodeling. It plays a role in tissue repair and bone mineralization.

CBD MMPs-EGFP:
CBD MMPs (Collagen-Binding Domain from Matrix Metalloproteinases): MMPs are enzymes responsible for the breakdown of ECM components like collagen. This domain has a strong affinity for collagen, making it suitable for applications requiring high binding affinity.

DB-EGFP:
DB (Discoidin Domain): Found in various ECM proteins, this domain is known for its collagen-binding properties. It’s important in cell-matrix interactions and tissue organization.

DDR-EGFP:
DDR (Discoidin Domain Receptor): DDRs are receptor tyrosine kinases activated by collagen. They are involved in regulating ECM remodeling and cell adhesion, with strong collagen-binding capabilities.

AGR-EGFP:
AGR (Agrin): A multi-domain proteoglycan involved in the organization of the ECM and synapse formation. It binds to collagen and influences cellular interactions with the ECM.

COMP-EGFP:
COMP (Cartilage Oligomeric Matrix Protein): COMP is a key ECM protein that binds collagen and stabilizes matrix structures. It plays a role in tissue integrity, especially in cartilage and tendons.

VWF-A Domain-EGFP:
VWF-A (von Willebrand Factor A Domain): This domain is known for mediating collagen binding, particularly in blood clotting and platelet adhesion. It is essential for collagen interaction in the vascular system.

FTD-EGFP:
FTD (Fibrinogen-like Domain): FTD is commonly found in proteins involved in coagulation and ECM remodeling, where it plays a role in binding collagen and other matrix components.

2.2 Build

We constructed several CBD-EGFP fusion proteins expressing plasmids and expressed them in bacterial systems to analyze their activity.

We first commissioned a company to synthesize the gene sequences of each CBD and then used seamless cloning to connect the gene sequences of each CBD with the EGFP sequence already in the laboratory.

In this stage, the colony PCR results show successful amplification of the eight different CBD-EGFP fusion proteins, confirming the correct construction of the engineered strains. The bands visible in the gel correspond to the expected sizes of the fusion proteins:

• SPARC/OD, DB, and DDR are shorter sequences. After seamless cloning with EGFP, their expected PCR product size is approximately 800 bp.
• The remaining five CBDs—CBD MMPs, AGR, COMP, VWF-A, and FTD—are longer tags. After seamless cloning with EGFP, their expected PCR product size is around 1200 bp.

The PCR bands in the gel align well with these predicted sizes, demonstrating that each fusion construct was correctly assembled, which is crucial for proceeding with protein expression and purification.

Fig 2.2 Plasmid design

We then used these E. coli to express proteins and measured the fluorescence value of each protein host bacteria at the same bacterial concentration. Some of the fusion proteins were then purified by Ni-NTA for further analysis.

2.3 Test

The fluorescence activity of each fusion protein was tested to evaluate whether the CBDs affected EGFP activity. The relative fluorescence values of the eight CBD-EGFP fusion proteins were measured to assess the impact of the collagen-binding domains (CBDs) on EGFP activity. As shown in the graph:

• SPARC/OD-EGFP, CBD MMPs-EGFP, DB-EGFP, VWF-A Domain-EGFP, and FTD-EGFP exhibited fluorescence values close to or higher than that of the control EGFP, indicating that these CBDs did not significantly affect the bioactivity of EGFP.
• DDR-EGFP, AGR-EGFP, and COMP-EGFP showed slightly lower fluorescence values, suggesting some impact on EGFP activity, but still maintaining measurable fluorescence.

Overall, the majority of CBDs retained sufficient EGFP activity, which confirms that they can be used in further experiments for controlled release without significantly impairing the functionality of the fusion proteins. These results support the selection of CBDs that balance collagen binding with maintaining protein bioactivity.

Fig 2.3 The fluorescence activity of each fusion protein

Binding affinity tests (via microscale thermophoresis) were also conducted to determine which CBDs had high binding affinity (for BMP-4) and low binding affinity (for VEGF). Based on the binding experiment shown in the graph, the results illustrate the relative binding affinities of different CBD-EGFP fusion proteins to collagen, with the binding affinity represented by the dissociation constant (Kd). The midpoint of each curve on the x-axis corresponds to the Kd value, indicating how tightly the CBDs bind to collagen.

• FTD-EGFP shows the strongest binding to collagen, with the lowest Kd value. From the graph, the FTD curve's midpoint is approximately 8.3*10^-7 M, indicating a high binding affinity.
• CBD MMPs-EGFP exhibits the weakest binding to collagen, with the highest Kd value. The midpoint of the CBD MMPs curve is approximately at 6.7*10^-6 M, indicating a lower binding affinity compared to the other CBDs.

These Kd values highlight that FTD is the most suitable for applications requiring strong binding, while CBD MMPs may be used in situations where weaker, more transient binding is needed.

2.4 Learn

From the binding results, FTD was chosen for BMP-4 due to its high affinity for collagen, ensuring a more sustained release. Meanwhile, CBD MMPs were selected for VEGF because of their low affinity, allowing for a faster and more transient release. These selections laid the foundation for the next cycle, where these domains were fused with the respective cell factors to achieve controlled and sequential release, mimicking natural bone healing processes.

Fig 2.4 What we learned in Cycle 1

3.1 Design

Based on the selections from Cycle 1, the second cycle aimed to design a dual cell factor release system. The goal was to ensure that VEGF, responsible for promoting angiogenesis, is released first with a low-affinity collagen-binding domain (CBD MMPs), followed by the controlled, delayed release of BMP-4 for osteogenesis, using the high-affinity binding domain (FTD). This sequential release would better mimic natural bone healing processes.

Fig 3.1 Fusion Cell factor expression plasmid

3.2 Build

In the Build stage of Cycle 2, the primary objective was to successfully express and purify two functional fusion proteins: CBD MMPs-VEGF and FTD-BMP-4, which are critical for the controlled release system. The key to achieving the desired gradient release lies in the intrinsic binding affinity of the collagen-binding domains (CBDs) to the collagen matrix. Since the release timing depends on the CBDs’ strength of interaction with collagen, we could focus solely on the expression of the cell factors without needing to alter the hydrogel’s crosslinking properties.

The success of this cycle hinged on producing ,soluble, bioactive proteins. For CBD MMPs-VEGF, expression in E. coli BL21 (DE3) yielded high concentrations of soluble protein under standard conditions, confirming that this fusion protein could be expressed easily and efficiently.

In contrast, FTD-BMP-4 proved more challenging. The presence of multiple disulfide bonds in BMP-4 led to the formation of inclusion bodies when expressed in BL21, which made the protein insoluble and inactive. To overcome this, we explored various optimized bacterial strains designed to facilitate the proper folding of disulfide bond-containing proteins.

Several strains were tested:

• E. coli Origami (DE3) improved disulfide bond formation but still resulted in considerable protein aggregation.
• E. coli Shuffle T7-B enhanced soluble expression, but its slow growth rate limited its practicality for large-scale production.
• Finally, ArcticExpress (DE3) pRARE was selected as the most suitable expression host, balancing both high protein yield and improved solubility of FTD-BMP-4.

The culmination of this stage was obtaining both fusion proteins in their active, soluble forms, laying the foundation for further testing of their controlled release and biological activity in subsequent phases. In Fig3.2, lanes 1 to 4, 5 to 8, and 9 to 12 are whole bacteria, supernatant after disruption, and precipitate after disruption, respectively. Each group consists of four lanes, and the order within each group is BL21, ArcticExpress (DE3) pRARE, Shuffle T7-B, Origami (DE3)

Fig 3.2 Results of expression system optimization a) SDS-PAGE; b) semi-quantitative analysis (Lanes 1 to 4, 5 to 8, and 9 to 12 are whole bacteria, supernatant after disruption, and precipitate after disruption, respectively. Each group consists of four lanes, and the order within each group is BL21, ArcticExpress (DE3) pRARE, Shuffle T7-B, Origami (DE3))

According to Fig3.2, ArcticExpress (DE3) pRARE stands out for its high proportion of soluble protein expression.The reasons for ArcticExpress's success in producing a high level of soluble protein are as follows:

In the gel (left) and the bar chart (right), ArcticExpress shows significantly improved soluble protein expression compared to other strains like BL21, which primarily produces inclusion bodies. This makes ArcticExpress the optimal choice for expressing soluble and active FTD-BMP-4, ensuring that a higher percentage of the total protein is in the desired soluble form.

3.3 Test

The system was tested through a series of in vitro cell assays to evaluate the controlled release of the two cell factors:
VEGF release was monitored to ensure rapid availability for angiogenesis.
BMP-4 release was tested to confirm delayed but sustained release for bone formation.
ALP (Alkaline Phosphatase) staining and other bioassays were conducted to assess the biological activity of the cell factors in promoting angiogenesis and osteogenesis.

3.3.1 Sequential release

This experiment monitored the release profiles of CBD MMPs-VEGF and FTD-BMP-4 from collagen hydrogels over time. The two fusion proteins were incorporated into collagen hydrogels, and their release was quantified using ELISA assays. The hydrogels were incubated in PBS at 37°C, with supernatant samples collected at regular intervals to measure the amount of protein released. After each sampling, fresh PBS was added to the hydrogels to maintain consistent conditions throughout the experiment.

Fig 3.3 Time-Dependent Release Profiles of CBD MMPs-VEGF and FTD-BMP-4 from Collagen Hydrogel

The results reveal a significant difference in the release kinetics between the two cell factors. CBD MMPs-VEGF showed a rapid and early release pattern, starting with an accelerated increase in the release rate during the first 10 days. By day 10, more than 50% of the VEGF had already been released from the hydrogel. This sharp increase continued, reaching approximately 75% release by day 20. Following this, the release rate gradually slowed, with the curve plateauing as nearly 100% of the VEGF was released by day 30. The release curve of VEGF shows an initial rapid release phase, followed by a stabilization period where the release rate gradually levels off. This fast-release behavior can be attributed to the lower binding affinity of the CBD MMPs domain, allowing VEGF to dissociate from the collagen matrix more quickly.

In contrast, FTD-BMP-4 exhibited a much slower and more controlled release. During the first 10 days, less than 20% of BMP-4 was released, indicating a more gradual release process. The release steadily increased, with the curve following a linear trend. By day 20, around 30% of BMP-4 had been released, and by day 30, approximately 50% was released. The release continued steadily, reaching close to 100% around day 40. This extended release profile reflects the higher binding affinity of the FTD domain to collagen, which ensures a slower and sustained release of BMP-4 throughout the experiment.

These contrasting release profiles demonstrate the effectiveness of using different collagen-binding domains to control the timing and rate of cell factor release, with CBD MMPs facilitating a faster release for VEGF and FTD ensuring a prolonged release for BMP-4.These results confirm that the controlled, sequential release of the two cell factors can be achieved by leveraging the different binding affinities of the CBDs, with CBD MMPs-VEGF releasing faster to initiate angiogenesis, followed by the slower release of FTD-BMP-4 to promote osteogenesis.

3.3.2 In vitro biological activity(VEGF)

The tube formation is a widely used in vitro method for assessing the ability of endothelial cells, such as HUVECs (Human Umbilical Vein Endothelial Cells), to form capillary-like structures, mimicking the process of angiogenesis. This assay is particularly useful for evaluating the angiogenic potential of cell factors, small molecules, or other treatments.

In the assay, endothelial cells are seeded on a Matrigel matrix, which provides an extracellular matrix environment that encourages the cells to migrate, align, and connect with each other, forming tubular networks. When angiogenic cell factors, such as VEGF, are present, these processes are enhanced, leading to more extensive and faster formation of tube-like structures.

The process of cell connection and tube formation involves several key cellular mechanisms:

1. Migration: Endothelial cells move toward each other, driven by chemotactic signals, such as VEGF, which promote cell motility. VEGF binds to its receptors on endothelial cells, activating intracellular signaling pathways that guide this migration.
2. Adhesion: As cells come into contact, they adhere to each other via cell-cell adhesion molecules such as VE-cadherin. These molecules facilitate stable connections between neighboring cells, which are critical for forming continuous structures.
3. Cell Structure Rearrangement: The internal framework of endothelial cells undergoes dynamic changes, allowing the cells to elongate and align with each other. This is crucial for shaping the cells into tube-like formations.
4. Lumen Formation: Once the cells are aligned and connected, they start forming hollow tubes that resemble capillaries. This involves the coordinated action of intracellular vacuoles and the fusion of these vacuoles between neighboring cells.

Overall, the tube formation assay provides insights into the angiogenic potential of treatments like CBD MMPs-VEGF, as seen in our experiment. The results suggest that CBD MMPs-VEGF promotes cell migration, adhesion, and network formation, confirming its ability to stimulate angiogenesis through effective cell factor release.

In the tube formation assay, human umbilical vein endothelial cells (HUVECs) were seeded onto a Matrigel-coated plate to facilitate tube formation. The experimental group was treated with CBD MMPs-VEGF at a final concentration of 100 ng/ml, while the control group received no VEGF. The cells were incubated at 37°C with 5% CO₂ for 4-6hours, during which tube formation was observed. The formation of capillary-like structures was monitored under a microscope to assess angiogenesis. The extent of tube formation between the experimental and control groups was compared to determine the impact of CBD MMPs-VEGF on promoting angiogenesis.

Fig 3.4 Comparison of Tube Formation in HUVECs a) Control; b)CBD MMPs-VEGF Treatment

Image a (Without CBD MMPs-VEGF): In the control group, where no CBD MMPs-VEGF was added, the HUVECs exhibited weak angiogenic activity. Only a few scattered cell connections were observed, with minimal evidence of network formation. Quantitatively, the tube formation rate (measured by parameters such as total tube length, number of branch points, and closed loops or tube closure rate) was significantly reduced. The tube closure rate, which measures the percentage of formed capillary-like loops, was close to zero, reflecting the lack of organized tube structures. This result confirms that without the presence of VEGF, the cells are unable to initiate or sustain effective angiogenesis. Image b (CBD MMPs-VEGF at 100 ng/ml): In the experimental group treated with 100 ng/ml CBD MMPs-VEGF, the HUVECs showed robust tube formation, with clearly defined and well-organized capillary-like structures. Quantitative analysis revealed a dramatic increase in several angiogenesis indicators. The tube closure rate increased to approximately 70-80%, indicating a high number of complete tube loops. Additionally, the total tube length and number of branch points were significantly higher compared to the control, demonstrating enhanced network complexity and cell connectivity. These findings suggest that the addition of VEGF significantly promoted angiogenesis, with the cells forming a dense and interconnected tubular network, characteristic of active capillary formation. This comparison highlights the essential role of CBD MMPs-VEGF in promoting angiogenesis, as quantified by standard tube formation metrics. The increased tube closure rate, total tube length, and branch points in the experimental group strongly indicate that the VEGF produced in the system retains its biological function and effectively stimulates endothelial cell organization and network formation. The comparison between these two images confirms that the CBD MMPs-VEGF produced in this study retains biological activity and effectively stimulates tube formation, validating its functional ability to promote angiogenesis.

3.3.3 In vitro biological activity(BMP-4)

In our osteogenic differentiation experiments, MC3T3-E1 cells, a well-established pre-osteoblast cell line derived from mouse calvaria, were used. These cells are commonly employed in bone research because they have the inherent capacity to differentiate into mature osteoblasts when cultured under osteogenic conditions. During the early stages of osteogenic differentiation, MC3T3-E1 cells express alkaline phosphatase (ALP), which is critical for mineralization. ALP staining can be used to measure the early osteogenic activity in these cells. As differentiation progresses, MC3T3-E1 cells begin to deposit calcium in the extracellular matrix, which is a key indicator of bone formation. This can be detected using Alizarin Red staining, which binds specifically to the calcium deposits and allows for the visualization of mineralization. Thus, by using both ALP staining and Alizarin Red staining, the osteogenic differentiation process of MC3T3-E1 cells can be effectively tracked from early osteoblast activity (ALP expression) to late-stage bone formation (calcium deposition).
So, ALP (Alkaline Phosphatase) Staining and Alizarin Red Staining are two key methods used to assess different stages of osteogenic differentiation in cells, providing insights into the bone formation process.

ALP Staining Mechanism:
ALP is an early marker of osteogenic differentiation, highly expressed by pre-osteoblasts and osteoblasts during the early stages of bone formation. ALP plays a crucial role in the mineralization process by hydrolyzing phosphate-containing compounds, and releasing inorganic phosphate, which contributes to the formation of hydroxyapatite crystals. When cells are induced toward osteogenic differentiation, an increase in ALP activity is observed. During ALP staining, a substrate (e.g., BCIP/NBT) reacts with ALP to produce a blue-purple color, indicating active ALP expression. By measuring this staining, you can assess how efficiently the cells are entering the osteogenic pathway and preparing for mineral deposition.

Alizarin Red Staining Mechanism:
Alizarin Red S staining detects calcium deposits, a hallmark of late-stage osteogenic differentiation. As osteoblasts mature, they begin to produce and deposit extracellular matrix, which undergoes mineralization as calcium and phosphate ions crystallize into hydroxyapatite, the main mineral component of bone. Alizarin Red S binds specifically to calcium ions, forming an orange-red complex that can be observed microscopically. By assessing the intensity and extent of red staining, you can determine the extent of mineralization, indicating that the cells have progressed to the later stages of osteoblast differentiation and are actively contributing to bone matrix formation.

Together, ALP staining and Alizarin Red staining offer a comprehensive view of osteogenic differentiation. ALP staining identifies early differentiation events, while Alizarin Red staining confirms matrix mineralization, showing that the cells are functionally mature osteoblasts capable of bone formation.

For the ALP staining, cells were cultured for 7 and 14 days, then washed with PBS and fixed with 4% paraformaldehyde for 10-15 minutes at room temperature. After washing with PBS, the cells were incubated with ALP staining solution (e.g., BCIP/NBT) for 30-60 minutes at 37°C. ALP activity, indicating early osteogenic differentiation, was visualized as blue-purple staining under a microscope. For the Alizarin Red staining, cells were cultured for 21 and 28 days, washed with PBS, and fixed with 4% paraformaldehyde for 10-15 minutes. After washing with distilled water, Alizarin Red S solution (pH 4.1-4.3) was applied for 20-30 minutes to detect calcium deposits. The cells were then rinsed with distilled water, and red staining was observed to indicate matrix mineralization, a marker of mature osteoblast activity.

Fig 3.5 ALP and Alizarin Red Staining of MC3T3-E1 Cells at Different Time Points During Osteogenic Differentiation

ALP Staining Results (Day 7 and Day 14): From the images provided, on Day 7 (top-left images), the ALP staining shows minimal blue-purple staining, suggesting that osteogenic differentiation had only just begun. By Day 14 (bottom-left images), there is a noticeable increase in blue-purple staining, indicating elevated ALP activity, which correlates with increased osteoblast differentiation. The stronger ALP activity seen on Day 14 demonstrates that the cells are undergoing early osteogenic differentiation, which is being effectively stimulated by the BMP-4 cell factor, confirming its biological activity in promoting bone formation.

Alizarin Red Staining Results (Day 21 and Day 28): In the Day 21 images (top-right), there are initial signs of mineralization with scattered red deposits, indicating the onset of calcium deposition. By Day 28 (bottom-right), the staining intensity has significantly increased, with larger and more densely stained red areas, showing substantial mineralization of the extracellular matrix. This confirms the later-stage osteogenic activity, as BMP-4 continues to drive the maturation of osteoblasts and the deposition of calcium, which is a hallmark of bone tissue formation.

These results confirm that the BMP-4 produced in our system retains its biological activity, as it successfully stimulates both early (ALP activity) and late (mineralization) osteogenic differentiation.

3.4 Learn

From the results, we confirmed that the gradient release of the two cell factors was achieved due to the differences in the collagen-binding domains (CBDs) fused to each factor, rather than any adjustments to the crosslinking density or hydrogel composition. VEGF was released first due to the lower affinity of CBD MMPs to collagen, while BMP-4 exhibited a more sustained release profile thanks to the high-affinity binding of FTD. These findings successfully demonstrated the functionality of the dual cell factor release system, which not only provided controlled, gradient release but also maintained excellent biological activity of both factors. This system proves effective for further applications and sets the stage for future optimization.

4.1 Design

In the third cycle, the goal was to create a collagen hydrogel system capable of delivering two cell factors—VEGF and BMP-4—in a controlled and sequential manner. Collagen was selected as the base material for the hydrogel due to its numerous advantages in bone repair. As a naturally occurring protein, collagen is biocompatible, biodegradable, and has excellent bioactivity, making it an ideal scaffold for supporting cell adhesion, proliferation, and differentiation. Furthermore, its structural similarity to the extracellular matrix provides an optimal environment for promoting tissue regeneration. Collagen’s ability to be easily crosslinked and modified allows for the incorporation of cell factors and controlled release mechanisms, making it a versatile material for bone defect filling while simultaneously delivering bioactive molecules. Based on these advantages, we designed the collagen hydrogel to act as both a physical scaffold for bone tissue regeneration and a carrier for sequentially releasing VEGF and BMP-4. Therefore, we chose to express Type I collagen using E. coli, as Type I collagen is more robust and widely distributed in bone tissue.
For our project, we are adopting a similar approach in the selection and design of a recombinant Type I collagen protein for bone repair applications
Sequence Selection: The foundation of our design is based on the human Type I collagen sequence, chosen for its natural role in supporting bone structure and promoting cell adhesion. The selected sequence retains the key functional regions of native collagen to ensure that our product effectively mimics the biological properties of natural bone tissue.
Codon Optimization: To ensure efficient production in E. coli, we will optimize the codon usage of the collagen gene. This adjustment improves the yield of the recombinant protein while ensuring proper folding and stability during expression in bacterial systems, allowing for scalable production of the collagen hydrogel.
Enhancement of Bioactivity: To further enhance the cell adhesion properties, we will incorporate specific peptide sequences at the N-terminal of the collagen. These sequences are designed to improve interaction with cells, thereby increasing the efficacy of collagen in promoting osteogenic activity and bone regeneration.
Structural Modifications: Similar to the patented approach, we will introduce targeted modifications to the collagen sequence. These changes will ensure the protein maintains high structural stability and over 90% sequence homology to native collagen while enhancing its performance in tissue repair applications. The modifications may involve selective addition, deletion, or substitution of amino acids to optimize both biological function and structural integrity.

This design methodology will allow us to create a recombinant Type I collagen that is highly effective in bone repair, ensuring both the strength and bioactivity required for successful tissue regeneration.

We have designed the pET32a-Collagen I plasmid to enable the recombinant expression of Type I collagen in E. coli. Utilizing the pET32a backbone, the plasmid incorporates a strong T7 promoter to drive high-level transcription of the collagen gene.

Fig 4.1 Schematic of pET32a-Collagen I Expression Plasmid for Recombinant Collagen Production

4.2 Build

Using E. coli ClearColi, we expressed low-endotoxin recombinant collagen, which was then incorporated into the hydrogel system. The collagen hydrogel was prepared by embedding the previously engineered fusion proteins—CBD MMPs-VEGF and FTD-BMP-4—into the matrix. The hydrogel was designed to ensure VEGF would be released quickly to stimulate angiogenesis, followed by a slower, sustained release of BMP-4 to support osteogenesis.

4.3 Test

The release profiles of the two cell factors were monitored over time using ELISA. The results showed that VEGF was rapidly released during the first 20 days due to the lower binding affinity of CBD MMPs to collagen, while BMP-4 was released more gradually over 40 days, reflecting the stronger collagen-binding capacity of FTD. Both cell factors retained their bioactivity, as confirmed by tube formation assays for VEGF and ALP and Alizarin Red staining for BMP-4, demonstrating successful angiogenesis and osteogenesis, respectively.

4.4 Learn

The third cycle confirmed that the dual cell factor release system achieved both gradient release and maintained strong biological activity of the embedded cell factors. VEGF was released first to promote angiogenesis, followed by BMP-4 for bone formation. This hydrogel system showed excellent potential for tissue engineering applications without the need for further modification to the matrix. These insights validated the design and performance of the release system and set the foundation for future applications in regenerative medicine.