BC material has high porosity, high water retention, strong plasticity, and good biocompatibility compared to most other biochemical materials. Therefore, we chose BC as the tissue engineering scaffold for large areas and deep trauma, which makes it easier for cells to cling and proliferate, and finally the wound can heal more quickly and even restore to its original state¹. According to the previous studies, we intended to use Acetobacter xylinum to produce BC. Its production efficiency is high and the synthesis mechanism is thoroughly studied. It has been selected by many iGEM teams (BNDS_China 2020, TUST_China 2017, etc.) to produce BC in their projects.

Fig.1 Schematic representation of the production of BC membrane by Acetobacter xylinum

Based on BC, we planned to carry on the preliminary modification. The macromolecular proteins beneficial to wound healing, such as sericin and collagen, will be difficult to embed directly into the BC film. Although the resulting BC membrane can be self-assembled into protein molecules by adding it directly to the culture medium of Acetobacter xylinum, the BC membrane produced under this culture method is difficult to be sterilized completely and the protein activity can not be guaranteed after sterilization. Therefore, we only considered adding small molecular substances such as chitosan and hyaluronic acid to make the preliminary modification. Chitosan has excellent biocompatibility, biodegradability, antibacterial ability, and ability of promoting wound healing. It is an ideal raw material for biomedical materials, especially tissue regeneration materials. Hyaluronic acid is an important component of extracellular matrix, which plays an important role in cell signal transmission and wound healing².

Fig.2 Schematic representation of the production of BC membranes by Acetobacter xylinum

As the main body of the system, the addition of the above two substances will make the BC scaffold have better performance. On this basis, we also design more modules to improve the system.

Although BC membrane and modified small molecular substances can promote wound healing to a certain extent, we hoped to indirectly modify the cells in the skin wound in other ways to enhance the proliferation and antibacterial ability of skin cells.

The whole process of wound healing is mainly regulated by growth factors. In this process, the exogenous supplement of growth factors or increasing the activity of endogenous growth factors can promote wound healing. Epidermal growth factor (EGF) can promote epidermal proliferation and keratinization, epithelial cell division and enhance cell activity, and promote wound healing of skin and mucosa³. The antibacterial ability of the stent system is also very important. Given a lot of side effects of drug resistance caused by antibiotics, we chose to use antimicrobial peptides to ensure the antibacterial ability of the system. LL-37⁴ is one of the well-studied, most promising, and safe antimicrobial peptides in mammals. It has wide antimicrobial spectrum such as gram-positive and gram-negative bacteria, viruses, fungi, parasites.

To make the cells in the skin wound secrete EGF and LL-37 rapidly and efficiently, the construction of AAV is the most effective form. Except for the high efficiency of gene expression, its safety is also extremely excellent. The virus maintains an attachment to a large extent and has almost no tendency to integrate into the genome⁵. Therefore, we planned to use adeno-associated virus as the carrier of two functional peptides. By injecting the modified adeno-associated virus under the wound source, it can contact and invade the cells of the human wound, and then transfer the target gene into cells, so that fibroblasts and other cells in the wound can independently express functional peptides.

Fig.3 EGF pathway and the effect diagram

Fig.4 LL-37 pathway and the effect diagram

The initial design of the project was to construct only two kinds of recombinant AAV, epidermal growth factor (EGF) and antimicrobial peptide LL-37. After consulting professionals, we knew that EGF has a corresponding effect mainly on epidermal cells, while basic fibroblast growth factor (bFGF) affects full-layer skin cells. The bFGF⁶ can regulate the synthesis and deposition of extracellular matrix components, increase the migration of keratinocytes and promote the migration of fibroblasts, thus accelerating the formation and re-epithelialization of granulation tissue. Intaking exogenous bFGF can also increase the expression of cytokines such as VEGF and TGF- β in the wound, thus promoting the mitosis of vascular endothelial cells, increasing endothelial permeability, and further promoting capillary fusion and macroangiogenesis. Therefore, we decided to increase the construction of adeno-associated virus vectors containing effective fragments of bFGF based on EGF and LL-37. At this point, we hoped that the tissue engineering scaffold can help to assist in healing with anti-bacterial infection ability.

Fig.5 bFGF pathway and the effect diagram

The main material of the tissue engineering scaffold we designed is bacterial cellulose, which can not be degraded by the human body in skin healing, and may cause a scar or immune infection. Therefore, we hoped that when the wound has healed properly, BC can be biodegraded and its degradable substances can be absorbed and utilized by the human body. Based on this idea, we added engineered living cells to achieve this goal. According to the studies, the effect of allogeneic fibroblasts in tissue engineering scaffolds is much stronger than acellular scaffolds⁷. This cell can be transformed into other cellular tissues and produce growth factors, collagen, fibronectin and GAG, which play an important role in the wound healing process. Therefore, we selected and modified the allogeneic human fibroblast cell line and introduced the blue light-activated LightOn gene expression system to enable the fibroblast cell line to start the secretory expression of cellulase under blue light and degrade the subcutaneous nonfunctional cellulose scaffold into glucose.

Fig.6 Blue light initiates the degradation of the BC membrane

The LightOn system⁸ contains only a single photosensitive transcription factor GAVPO, which does not bind to the target transcription unit under dark conditions, and the target gene is not expressed; after blue light irradiation, GAVPO will be dimerized to bind to the UAS_G sequence of the target transcription unit and activate the expression of the target gene. The three bacterial cellulase genes^9,10,11 induced by LightOn system are all from Cellulomonas fimi. Among them, the Cex and CenA codes for an exo-1,4-D-glucanase and 1,4-D-glueanohydrolase, which degrade cellulose to cellobiose respectively. Both enzymes contain a cellulose binding domain that enhances the affinity to the cellulose substrate. The introduced third bacterial cellulase gene encoding for β-glucosidase could further hydrolyze cellobiose to glucose. Since all three genes mentioned above are prokaryotic expression genes, a fusion of eukaryotic signal peptide IL2-sig (signal peptide of Homo sapiens interleukin 2) at its C-terminal end was designed in order to enable their secretory expression in our engineered BJ.

Fig.7 Blue light initiates the cellulase expression pathway

The function of fibroblasts to degrade BC scaffolds needs to be healed to a certain extent, so the expression of cellulase must be stable and effective for a long time. At the same time, the molecular weight of the whole LightOn system and bacterial cellulase is large, so lentivirus is selected as the carrier to stably transfect LightOn system and bacterial cellulase into fibroblasts to degradate BC. Considering that the lentiviral vector system can only package plasmids with target genes of a certain size at one time, we split three kinds of cellulases and constructed three lentiviral plasmids containing transcription factor GAVPO and cellulase respectively. After the viruses are packaged, it can infect different fibroblasts, so that these fibroblasts can be induced by blue light to secrete and express their corresponding cellulases, thus degradating of the whole BC scaffold.

The plasmid used the constitutive promoter CMV to express the GAVPO protein, and the upstream initiation region of the cellulase gene contains a UAS_G sequence regulated by transcription factor activation, allowing for light-induced expression regulation. Among the plasmids containing LightOn system used in previous studies, the N-terminal of UAS_G containing polyA and pause site can eliminate background interference expression¹². However, lentiviruses are retroviruses, the insertion of polyA in the viral genome may prematurely end the transcription process leading to transcription failure or reduced titers¹³. We decided to delete the N-terminal polyA and pause site. We also attempted to adapt the target transcription units initiated by LightOn system upstream of the CMV promoter to prevent the strong promoter CMV from affecting the transcription of the downstream genes. Considering that screening of effective cells was required after subsequent stable transfection of cells, we added the PuroR resistance gene to the plasmid.

Fig.8 Blue-light lentiviral vector expressing CenA

Fig.9 Blue-light lentiviral vector expressing Cex

Fig.10 Blue-light lentiviral vector expressing Bglx

After fibroblasts secreted a series of cellulases to completely break down BC into glucose under the regulation of blue light, we added a red/far-red light suicide regulatory system to control the expression of downstream toxin protein MazF in fibroblasts to prevent them from continuously secreting cellulases and causing unnecessary immune responses. MazF¹⁴ is an Escherichia coli toxin, which is highly conserved in prokaryotes. When MazF is induced, it can produce a specific cleavage reaction between the first A and C at the ACA site of single stranded RNA, enabling protein synthesis be effectively inhibited¹⁵. MazF also has a programmed killing effect on normal and tumor cells of mammals. When irradiated with red light (660 nm), engineered fibroblasts will synthesize MazF toxin to interfere with the normal synthesis of their own proteins, resulting in a series of reactions and eventually leading to self apoptosis. The apoptotic mechanism of these cells will be terminated by far-red irradiation (730 nm). The implementation of this system also depends on lentivirus vector transfection of fibroblast cell line.

Fig.11 Red light initiates cell suicide

The red/far-red light suicide control system REDMAP¹⁶ is so sensitive to light that as low as 0.1 mW cm^-2 red light (far-red light) can be sensed. The specific mechanism of this system is that when exposed to red light (660 nm), the transactivator (FHY1–VP64) can specifically bind to the light sensor domain (ΔPhyA–Gal4) in the presence of the photosensitive pigment Phycocyanobilin (PCB), and the combined protein complex translocates into the nucleus where it can bind to its synthetic promoter (P_5 × UAS, 5 × UAS-P_hCMVmin) to initiate expression of MazF. Following exposure to far-red light (730 nm), the transactivator dissociates from the light sensor domain (ΔPhyA–Gal4), thereby terminating the expression of MazF. At the same time, in order to prevent the CMV strong promoter from directly activating the expression of MazF, we adjusted the MazF region to the upstream of the CMV promoter. Since the REDMAP and LightOn system require stable transfection to the same fibroblast cell line (BJ), we added the resistance gene Neo that is different from PuroR to easily select cell lines stably transfected with the REDMAP and LightOn systems.

Fig.12 Red light initiates the MazF expression pathway

Fig.13 Red light system lentiviral vector

The system we designed fully reduces the autoimmune risk of engineered allogeneic fibroblasts, and greatly enhances the security performance.