In order to fight bacteria and make the wound heal better, we designed an auxiliary healing module. Based on a AAV-based vector, we successfully constructed three plasmids (i.e., pAAV-CMV-EGF, pAAV-CMV-LL-37, and pAAV-CMV-bFGF) (Fig.1). With the help of auxiliary plasmid pHelper that ensures the recombinant AAV to successfully infect cells, the target cells can successfully express epidermal growth factor (EGF), fibroblast growth factor (bFGF), and a human antibacterial peptide (LL-37). These three proteins ensure the antibacterial and healing functions of the system.

Fig.1 The AAV plasmids that were constructed for overexpressing growth factor or antibacterial peptide. a. pAAV-CMV-EGF b. pAAV-CMV-bFGF c. pAAV-CMV-LL-37

This plasmid was offered by another laboratory.

We obtained the linearized vector by double digestion of vector pAAV with endonucleases BamHI and NotI (Fig.2). We amplified the target gene LL-37 from another plasmid containing the target fragment by PCR (Fig.3). We then ligated the amplified fragment into the linearized vector by T4 DNA ligase when the ratio of vector to fragment was 1:4, 1:5, or 1:10. We carried out colony PCR (Fig.4) and sanger sequencing to verify whether the LL-37 was successfully ligated into the vector. From the result of colony PCR, it can be seen that the vector/fragment ratio of 1:10 is the most appropriate ratio for constructing this plasmid (Fig.4).

After obtaining the IL2-bFGF-6xHis fragment synthesized by the company, we successfully integrated it with the linearized vector by homologous recombination.(Fig.5)

Fig.2 The gel diagram of linearized pAAV after digestion

Fig.3 The gel diagram of the amplification LL-37 under different annealing temperatures

Fig.4 Colony PCR were used to confirmed whether LL-37 was successfully ligated with the vector

Fig.5 The gel diagram of pAAV-CMV-bFGF recombinant plasmid

After the pAAV-CMV-EGF and pAAV-CMV-LL-37 plasmids were successfully extracted, we cotransfected the pAAV-CMV-EGF or pAAV-CMV-LL-37 plasmids with pHelper to HEK293T cells for AAV6 packaging. After three days, we observed the cell status (Fig.6).

Fig.6 The status of HEK293T before (a) and after (b) transfection

After transfection, AAV6-EGF and AAV6-LL-37 were packaged and were still remained in cells. Therefore, we split the cells and ultracentrifuged the cell lysate to remove the empty shell virus and impurities and to separate and purify AAV. We then took the liquid at the scale line F (Fig.7) for ultrafiltration, concentrated the sample for 15 times and finally obtained the concentrated viruses.

Fig.7 The procedures of overspeed centrifugation. a. Comparison diagram before and after centrifugation (left: before centrifugation, right: after centrifugation). A, Cell lysate. B, 7.3ml 15% lodixanol density layer. C, 4.9ml 25% lodixanol density layer. D, 4ml 40% lodixanol density layer. e, 4ml 60% lodixanol density layer. f, Location of liquid containing purified AAV6. b. The rotor for ultracentrifugation

After obtaining the concentrated AAV solution, we conducted qPCR with SYBR Green I method to detect the titer of AAV. According to the standard curve and amplification curve (Fig.8). The titer of AAV6-LL-37 is 9.126×10⁴Vp/μL(Table 1). The titer of AAV6-EGF is 2.108 ×10⁵Vp/μL (Table 1). The results of qPCR are shown in Fig.9. We successfully obtained AAV6-LL-37 and AAV6-EGF.

Fig.8 qPCR results of AAV6-EGF and AAV6-LL-37 a. Standard Curve b. Amplification Curve

Table 1 The titer of the concentrated AAV solution

Fig.9 Calculation of the quantity of AAV based on qPCR. a. Histogram of the titer of AAV6-EGF. b. Broken line graph of the CT value of AAV6-EGF. c. Histogram of the titer of AAV6-LL-37. d. Broken line graph of the CT value of AAV6-LL-37.

After AAV6-EGF and AAV6-LL-37 were generated, we infected HEK293T with concentrated viruses. After incubation for 24h or 72h, the supernatant was harvested and the levels of EGF and LL-37 in the supernatant were measured by ELISA.

We established the standard curve to accurately quantify EGF in the cell culture supernatant (Fig.10a). We calculated the content of EGF in the supernatant of the experimental group and the control group at 24h and 72h. The level of EGF in the supernatant of the experimental group was 107.25% higher than that of the control group after 24 hours of incubation. The level of EGF in the supernatant of the experimental group was 53.10% higher than that of the control group after 72 hours of infection (Fig.10b, 10c). These data suggested that AAV6-EGF successfully elevated the level of EGF in the supernatant.

Fig.10 Determination of EGF content in cell culture supernatant by ELISA. a. The standard curve of EGF. b. Table of EGF content in culture supernatant of experimental group and control group at different time. c. Comparison of EGF content in culture supernatant between experimental group and control group at 24h or 72h.

To accurately measure LL-37 in the cell culture supernatant, we established the standard curve for LL-37 (Fig.11a). According to this standard curve, we calculated the expression levels of LL-37 in the supernatant of the experimental group and the control group after incubating HEK293T cells with AAV for 24h or 72h. As can be seen from the results, the level of LL-37 in the supernatant of the experimental group was 1746.71% higher than that of the control group after 24 hours of infection (Fig.11b, 11c). The level of LL-37 in the supernatant of the experimental group was 75.28% higher than that of the control group after 72 hours of infection (Fig.11b, 11c). Therefore, the infection of AAV6-LL-37 led to the significant increase of LL-37 in the supernatant.

Fig.11 Determination of LL-37 content in cell culture supernatant by ELISA. a. The standard curve of LL-37. b. Table of LL-37 content in culture supernatant of experimental group and control group at 24h or 72h. c. Comparison of LL-37 content in culture supernatant between experimental group and control group at different time.

The above results showed that the expression of EGF and LL-37 could be effectively increased by infecting cells with corresponding adeno-associated viruses. EGF can continuously accumulate in the cell culture supernatant as the expression of EGF in the experimental group was 94.37% higher after 72h than that after 24h. However, LL-37 may be expressed quickly but degraded quickly as the level of LL-37 in the experimental group was 65.28% lower than that in the 24h group. We will conduct repeated experiments to confirm the reliability of this result. If it is confirmed, we have to control the sequence and timing of virus infection. We will also determine the role of EGF in promoting cell growth by measuring the growth curve of cells and measure the antibacterial performance of LL-37 by antibacterial experiments.

When the wound was healed to an appropriate extent, it was assumed that the cellulase can be secreted out of cells to degrade the cellulose scaffold that was useless under the skin, aiming to prevent scar formation and possible immune reaction. As for the project, we planned to introduce LightOn system, a blue light activated gene expression system, into the engineering cells. After blue light irradiation, transcription activating factor (GAVPO) can be rapidly dimerized and bind to the 5xUAS_G sequence upstream of TATA box to induce the transcript accumulation of cellulase to degrade cellulose scaffolds. The experimental design constructed a cellulase eukaryotic expression system, and we chose human fibroblasts as our target host cells. However, due to time constraints, we transiently transferred the cellulase gene into HEK293T cells and achieved blue light induced expression. After that, we packaged lentivirus and transfected HEK293T cells to achieve stable expression. Subsequently, we used purinomycin to screen the stably transfected positive cell clones.

The bacteria DNA kit was used to extract the genomic DNA of cellulomonas fimi, and the cellulase genes CenA (endoglucanase), Cex (exo-β-1,4-glucanase) and Bglx (β-glucosidase) were amplified by PCR. The signal peptide sequence of interleukin-2 (IL-2) was fused at the N-terminal of genes by overlap PCR. The plasmids skeleton containing the LightOn system was digested with BamHI and PacI enzymes to obtain linearized plasmids (Fig.1~Fig.3), after which we connected cellulase genes to the plasmids through homologous recombination to obtain positive clone recombinants.

Fig.1 Construction of lentivirus expression vectors for three cellulases. a.Blue light induced Cex expression plasmid pCex; b.Blue light induced CenA expression plasmid pCenA; c.Blue light induced Bglx expression plasmid pBglx.

Fig.2 Agarose gel electrophoresis analysis of sigIL-2_ Cex (1407 bp), sigIL-2_ CenA (1343 bp) and sigIL-2_Bglx (1749 bp). SigIL-2 is interleukin 2 signal peptide fused at the N-terminal of the gene. Red frames outlined in the figure are target genes.

Fig.3 Agarose gel electrophoresis analysis of lentivirus mammalian cell expression plasmids with double enzyme digestion. The plasmids were digested with BamHI and PacI, liner plasmid I and II were both digested products.

The constructed cellulase expression plasmids pCex, pCenA and pBglx were respectively mixed with the transfection reagent Lipo8000 and transiently transfected into HEK293T cells. One day after transfection, the culture flask was irradiated with blue light for 3 hours. The culture flask was illuminated by 1mW·cm^-2 blue light (460-479nm) for 3 hours. The supernatant of the culture medium and the cell lysate were analyzed by SDS-PAGE and identified by Coomassie brilliant blue staining and Western blot. The results of Coomassie brilliant blue (Fig.4) showed that both the supernatant and cell lysate samples had a large quantity of proteins in the target size of 45 kDa-60 kDa. It was hard to determine whether the target protein was in them, and further experimental detection was required. According to Western blot results (Fig.5), the cells transfected with empty plasmids and plasmids connected with cellulase could express GAVPO protein (an important element of LightOn system on the plasmid skeleton), which means that our plasmids had been successfully transferred into cells and expressed.

Fig.4 SDS-PAGE analysis of cellulase expression in supernatant of medium and cell sediment. Intra=intracellular, the samples are cell precipitation fragments; extra=extracellular, the samples are the cell precipitation fragments.

Fig.5 Western blot analysis of GAVPO. a. GAVPO is a constitutive expression protein (56 kDa) on the plasmid skeleton, and blue-light induces dimerization; b. GAPGH is a reference protein in cells with a molecular weight of 36 kDa.

Western blotting proved that cells could express the target cellulases respectively after blue-light irradiation (Fig.6). CenA is an endoglucanase secreted to the environment in the form of glycosylated protein. In previous studies, Western blot of CenA revealed a complex pattern comprising multiple bands, including precursor and many smaller bands¹. CenA is susceptible to non-enzymatic attack modes in the cytoplasm, thus leading to auto-catalytic or spontaneous cleavages (SC). Different intensities of the protein bands reflected that the peptide bonds in CenA were susceptible to unequal rates or frequencies of cleavage². Our Western blot results showed that the CenA did contain multiple bands, of which two bands between 45 kDa-60 kDa were the strongest, and there were faint bands of other lengths above and below them. Our experimental results were consistent with the previous studies, which revealed that CenA was successfully secreted to the outside of the cell. Afterwards, we collected the supernatant of the medium for Western blot, and the results illustrated that endoglucanase (CenA) was successfully secreted into the extracellular medium after expression (Fig.6c).

Fig.6 Western blot analysis of cellulase. Notes: CenA molecular weight is about 48 kDa; Cex molecular weight is about 51 kDa; Bglx molecular weight is about 66 kDa. a. and b. are the Western blot analysis diagram of cell lysates, and c. is the Western blot analysis diagram of culture medium supernatant.

Based on Western blot, all these plasmids we constructed were successfully expressed in cells after transfection, and the LightOn system was able to work as we designed. Moreover, the downstream cellulase would be expressed and secreted to perform functions.

Previous cell transient experiment verified that the plasmids we constructed could achieve the goal of blue light induced expression of cellulase. In order to enable engineered cells fibroblasts to stably express cellulase under blue light, we decided to construct a stably expressing cell line by packaging lentivirus infecting cells, and then we obtained lentivirus packaging plasmids psPAX2 and pMD2.G through BIOSCI Co., Ltd. We extracted and sequenced the plasmids. Cellulase expression plasmids and lentivirus packaging plasmids were mixed and transferred into packaging cell HEK293T with Lipo8000, and virus supernatants transfected for 48h and 72h were collected.

However, we decided to choose HEK293T cells as the target host cells of lentivirus infection for the long growth cycle of human fibroblasts and the urgency of the experimental time affected by the pandemic situation. After infecting the cells with the collected virus supernatant, we irradiated the cells with blue light for 2h, after which we used puromycin to screen HEK293T cells 48 hours later. Before the formal drug screening, we used different concentrations of puromycin to treat the cells (Fig.7), and selected the lowest antibiotic concentration that would kill all cells within 3-4 days to conduct the drug screening test. Through the pre-experiment, it was determined that the proper drug screening concentration of puromycin was 1 μ g/ml. However, on the fourth day of drug screening, the cells in control groups and transfected groups became round and floating, and it was difficult to find cells with normal morphology under the microscope for expansion screening of monoclonal (Fig.8). The reason for the frustrating results may be that the lentivirus does not work. Meanwhile, we also suspected that virus does not add appropriate transfection promoting reagents when infecting cells or the cells before transfection have a large degree of fusion, thus leading to the low transfection efficiency of lentivirus. Therefore, it is difficult to obtain positive clone cells. Later, we will adjust the experimental scheme and continue to try to obtain stable cell lines.

Fig.7 Cell density and morphology after puromycin screening at different concentrations for 1 day. The picture shows the field of vision in 24 wells plate under 4× microscope. The number of cells in the figure a-h is basically the same after the initial passage, and is about half of the number of cells in Figure i-l. Puro concentration in Figure a and b is 0.5 μg/ml, puro concentration in Figure c and d is 0.8 μg/ml, puro concentration in Figure e and f is 1 μg/ml, and puro concentration in Figure g and h is 1.5 μg/ml. Puro concentration in Figure i is 0.5 μg/ml, puro concentration in Figure j is 0.8 μg/ml, the puro concentration in Figure k is 1 μg/ml, and the puro concentration in Figure l is 1.5 μg/ml.

Fig.8 Morphology of HEK293T cells on puromycin drug screening (2 days later). a. cells in control group; b. cells transfected pBglx plasmid; c. cells transfected pCex plasmid; d. cells transfected pCenA plasmid. The figure shows the field of vision of 10cm² dish under 4× microscope.

The initial idea of our module was that when the wound healed to a certain stage, the cellulose scaffold would be degraded by inducing the expression and secretion of cellulase in cells through external irradiation of blue light. Therefore, the following experiments were designed to verify whether the cellulase secreted from our engineering cells can degrade BC membrane.

Verify the degradation function of the protein that was secreted by cells in the transient system on BC membrane. After the pretreatment of BC membrane, we added cellulase (as positive control) and cell culture medium containing cellulase secreted by cells for co-culture to observe whether BC membrane can be degraded. According to the co-culture results (Fig.9), the BC membrane was basically degraded in the positive control group one day later. Three days later, it was observed that the surface of the BC membrane in well D was not flat and there were small cracks as well. Under the microscope, it was observed that there were obvious cracks and flocculent floaters on the surface of the membrane, while the degradation of well C was unclear and only a small number of flocculent substances appeared. According to the results, our cellulase did exist in the culture medium, which could degrade BC membrane, but the effect was not satisfactory.

To explore the rate of cellulose degradation by cellulases over time, we took photos of the cellulose membrane at regular intervals from the beginning to the end of the degradation for positive control. And we fit the degradation curve of the BC membrane through a mathematical way (Fig.10).It was speculated that the existence of Bglx may inhibit the enzymatic reaction of the other two cellulases. Through debugging parameters in modeling, it was found that the third enzyme showed a non-competitive inhibition on the activity of other two enzymes. The enzyme used in our positive control was the cellulase complex produced by Trichoderma viride, and its degradation rate was fast. The low concentration of cellulase in the culture medium in our test group may be the reason that the degradation rate is slow.

We will further improve the experimental scheme to better conform to the application approach of the project.

Fig.9 Degradation of bacterial cellulose (BC) membrane by cellulase. The picture shows the degradation of BC membrane of 6-well plate after being treated for 3 days under the 4X microscope field of vision. a. PBS was added into well A as negative control; b. 1 ml 25 mg/ml of commercial cellulase solution was added into well B; c. After ultrafiltration concentration at 1:1:1 was added in well C, 0.5 ml of pCex, pCenA and pBglx transiently expressed cell culture medium. DMEM culture medium was added up to 2 ml; d. 750 μ L of pCex and pCenA transient expression media after ultrafiltration concentration were added into well D at 1:1, respectively. DMEM medium was added to 2 ml.

Fig.10 Degradation curve of membrane superficial area with time. Notes: From the beginning to the end of the degradation, we took photos of the cellulose membrane at regular intervals. We use ImageJ to measure the superficial area of cellulose membrane aiming to determine the degradation area ratio. And we fit the degradation curve of the BC membrane through a mathematical way.

1 Wong, W. R. et al. Characterization and structure of an endoglucanase gene cenA of Cellulomonas fimi. Gene 44, 315-324 (1986).

2 Lai, C. Y., Ng, K. L., Wang, H., Lam, C. C. & Wong, W. K. R. Spontaneous Cleavages of a Heterologous Protein, the CenA Endoglucanase of Cellulomonas fimi, in Escherichia coli. Microbiology Insights 14, 11786361211024637 (2021).

In order to realize the function of red light regulating cell apoptosis, we selected REDMAP red light control system, and obtained all the sequences required by the red light control system including Gal4, PhyA, FHY1, VP64, etc. from the original literature¹. We selected PLVX as the eukaryotic expression vector, and gave the sequence required to VectorBuilder to construct the recombinant plasmid REDMAP. The plasmid vector is shown in Fig.1. After sequencing and verifying that the gene sequence is correct, we resuscitated and expanded the E. coli containing the plasmids and extracted the plasmids. The results are shown in Table 1.

Fig.1 REDMAP plasmid

Table 1. Concentration of plasmid

In view of the slow growth rate of BJ and the inappropriate passage ratio at the early stage of the experiment, which affected the cell activity, in order to successfully test whether our recombinant plasmid can be expressed in eukaryotic cells, we selected HEK293T cells for subsequent functional verification.

We used lipo8000 ™ transfection system to transiently transfer PLVX vector to HEK293T cells by liposome transfection and continued to culture cells for 43 h to ensure the normal state of cells. Then we added PCB with final concentration of 5 μM to the system, and irradiated the system with red light (660 nm) for 3 h. After illumination, we observed the morphology of the cells and analyzed the cell lysate by SDS-PAGE and Western blot.

We found that nearly 15% of the cells in the bottle flask after light had become smaller and rounder, showing a dead state (Fig.2). Other cells, though adhering to the wall and spreading, also showed abnormal state. This showed that cells had started suicide switch after red light irradiation, and apoptosis of cells had been successfully started.

Fig.2 Cell morphology of experimental group and control group after red light verification. The cells in the experimental group were transient cells, while the cells in the control group were non transient cells.

The size of VP64 is 31 kDa and the size of MazF is 14 kDa. In the first SDS-PAGE results, we found that the cell lysate of the experimental group and the control group had bands similar in size to VP64 and MazF (Fig.3a). Theoretically, VP64 bands and MazF bands should not appear in the cell lysate of the control group. Therefore, we believed that the suspected bands in the control group were other proteins of similar size. At the same time, because our protein was not purified and there were many heterobands, we subsequently carried out the Western blot experiment.

We detected VP64 specific bands in the cell lysate of the experimental group (Fig.3b), and no specific bands were detected in the control group, which proved that VP64 was indeed expressed in our cells, and it means that our vector has been successfully transferred into cells. It also proved that the bands near 30 kDa in the control group in SDS-PAGE were other proteins.

We detected the specific band of MazF in the cell lysate of the experimental group (Fig.3c), and no specific band was detected in the control group, which proved that MazF was indeed expressed in our cells, and it mean that our vector had been successfully transferred into cells. It also proved that the bands near 13 kDa in the control group in SDS-PAGE were other proteins.

Fig.3 a. REDMAP SDS-PAGE. VP64 stripes in the red box and MazF stripes in the green box. CG-C was the cell lysate of the control group and EG-C was the cell lysate of the experimental group. b. Western blot of VP64. CG-C is the cell lysate of the control group, EG-C is the cell lysate of the experimental group, and the theoretical size of VP64 is 30 kDa. c. Western blot of MazF. CG-C is the cell lysate of the control group, EG-C is the cell lysate of the experimental group, and MazF theoretical size is 13 kDa.

In this transient transfection, we hoped to detect MazF expressed in the system to ensure the correctness of plasmid construction. Therefore, we immediately observed the cells and carried out protein verification after red light. However, just because the time left for MazF expression was not enough, there were still a large number of cells alive when we observed them under the microscope, so we could not determine whether MazF had a good killing effect. Therefore, we optimized the experimental steps to verify this function.

We used lipo8000 ™ transfection system to transiently transfer REDMAP plasmids to HEK293T cells by liposome transfection and continued to culture for 8 h to ensure the normal state of cells. Then we added PCB with the final concentration of 5 μM to the system. After PCB entered the cells and combined with PhyA, we irradiated the system with 660 nm red light for 8 hours. Because it takes a certain time for our MazF to express and finally cause apoptosis, we continued to culture cells for 18 hours after light exposure, and then observed the cells at this time and performed subsequent Western blot analysis.

We observed the cells of the experimental group and the control group 46 hours after adding the transient transfection system, and it was obvious that most of the cells in the experimental group were in a dead state, their adhesiveness was greatly weakened, they became smaller, rounded and brighter, and began to float in large areas (Fig.4). From this result, we can conclude that the red light suicide system we built can operate effectively when it has enough expression time, which prompted us to carry out the subsequent experiments on packaging and stable transfection of REDMAP lentivirus.

Fig.4 Cell morphology of control group and experimental group after red light verification. The cells in the experimental group were transient cells, while the cells in the control group were non transient cells.

To verify the successful expression of our target protein, we collected the cells that had activated the red light suicide switch and their supernatant, split the cells and concentrated the supernatant, respectively, to obtain the cell lysate and supernatant concentrate. Then we carried out SDS-PAGE and Western blot experiments.

After SDS-PAGE, we performed Coomassie brilliant blue staining on the electrophoresis gel, and the results are shown in Fig.5a. The protein size of REDMAP should be around 30 kDa. Theoretically, the protein amount of cell lysate in the experimental group at 30 kDa should be higher than that in the control group. However, due to the apoptosis of cells and the degradation of some proteins when we collected cells and supernatants, the protein bands in the cell lysate of the experimental group were shallower than those in the control group.

We also carried out Western bolt verification on the results of SDS-PAGE, the results are shown in Fig.5 b - d. In general, the results of GAPDH (internal reference) were correct, but the cells had died before the experiment and some proteins, including MazF, had been degraded, thus no specific bands had been produced. However, the result of REDMAP showed specific band of correct size.

It can be seen from Fig.5b that there is a shallow band near 30 kDa in the cell lysate of the experimental group, and the size of the REDMAP element is just around 30 kDa, indicating that we may have obtained the correct result. However, due to the fact that the cells had died and some proteins had been degraded during the experiment, REDMAP had also been partially degraded, resulting in a shallow band.

The theoretical size of GAPDH is 36 kDa. The size of the bands in Fig.5c is around 36 kDa, which shows that the results of the GAPDH are correct, and our experimental operation is accurate.

The theoretical size of MazF is 13 kDa. It can be seen from Fig.5d that there is no specific band in the experimental group. The reason for this result may be that the cells used in this experiment have died before the experiment, leading to partial protein degradation.

According to the results of the second round of cell experiment, we can see that the red light suicide system we built has successfully played a role, and the use of red light after the expression of related proteins can cause cell death. At the same time, the results of Western blot also proved that our REDMAP can be successfully expressed in cells. So far, we have preliminarily proved that our red light suicide system can kill cells under a certain amount of red light. However, since the cells had died before this experiment, some proteins have been degraded, we didn’t detect MazF. Thus we have preliminarily proved the expression of REDMAP and MazF, we will continue to conduct more perfect experiments to verify the function of red light activated suicide module.

Fig.5 a. Results of REDMAP Coomassie brilliant blue staining. b. Western blot result of REDMAP. The theoretical size of REDMAP is 30 kDa. c. Western blot result of GAPDH. The theoretical size of GAPDH is 36 kDa. d. Western blot result of MazF. The theoretical size of MazF is 13 kDa. CG-C was the cell lysate of the control group and EG-C was the cell lysate of the experimental group.

We used lipo8000 ™ in the transfection system, two lentivirus packaging plasmids PMD2. G and PSPAX2 were co transfected with the target plasmid of REDMAP, which was correctly sequenced. When the cell density reaches 70-80%, we added the transfection system which includes serum free culture medium, three plasmids and lipo8000 ™. We collected virus supernatants of transfected cells for 48 h and 72 h respectively and filtered them into a new centrifuge tube through a 0.45 μm filter, and stored them at - 80 ℃ for subsequent stable transfection.

After obtaining the supernatant containing REDMAP lentivirus, we hoped to determine the effectiveness of virus packaging by measuring its titer. Before qPCR titer determination, we concentrated the supernatant obtained by ultrafiltration, and ultrafiltration 10 mL supernatant twice to 500 μL. According to the standard curve and the CT value of REDMAP lentivirus determined by qPCR, the average quantity of REDMAP lentivirus can be calculated as 1.596 × 10⁵ Vp/μL. According to Fig.6a and b, we successfully packaged lentivirus containing the target gene.

Fig.6 a. CT value of REDMAP lentivirus (after ultrafiltration) and blank. b. Quantity of REDMAP lentivirus (after ultrafiltration) and blank.

The REDMAP lentivirus we packaged carries a gene that is resistant to G418. HEK293T cells, however, were resistant to the analog G418 because they carried Neomycin gene when being introduced into SV40T-antigen, so HEK293T cells were not suitable for drug screening after stable transfection. We selected BGC823 cells which are also widely used in experiments when stably transfecting and expressing the red-light system. We carried out two rounds of lentivirus infection for 48 hours to cells that had just been subcultured for 24 hours. After infection, we added G418 with final concentration of 600 ug/mL. On the fifth day of drug screening, we found that more than half of the cells in the whole dish were dead, so we reduced the drug screening concentration to 400 ug/mL and continued to culture. By the seventh day of drug screening, we observed that almost 90% of the cells in the whole culture dish had died and floated down. Therefore, we changed the drug screening culture medium into a fresh complete culture medium to expand the remaining cells, and verified the red light function of stable cells when the cells were expanded to the appropriate density.

Fig.7 Status of BGC823 cells on the seventh day of drug screening

1 Zhou, Y. et al. A small and highly sensitive red/far-red optogenetic switch for applications in mammals. Nature biotechnology (2022).