Table of Contents
I. Growth Curve of E. coli (OD & CFU) at 37°C
Description
To model bacterial growth dynamics in wounds, we used E. coli strain DH5α as a representative organism to establish a preliminary framework for the growth function. In future applications, the data collected by users will be fed back into the system, enabling simulations based on this function and allowing the model to continuously adapt to real-world conditions through iterative learning.
Data
Analysis
Initially, we attempted to establish the growth curve of E. coli by measuring OD600, and observed that the bacteria entered the stationary phase after approximately 10 hours. However, validation using CFU counts revealed that this was not the case, as the stationary phase occurred much later. We hypothesize that as the OD approached 2.0, the instrument’s detection limit may have been reached, leading to a significant drop in accuracy.(Figure A-I)
To derive a more accurate growth curve, we extended the growth period and tracked CFU counts. It was found that E. coli(DH5α) entered the exponential phase at around 2 hours, reached the stationary phase after approximately 30 hours, and entered the death phase after about 50 hours.(Figure A-II)
II. Growth Curve of L. lactis (OD & CFU) at 26°C
Description
To assess the growth pattern of Lactococcus lactis MG1363 and predict its behavior within the hydrogel, as well as estimate the shelf life of the hydrogel-based bandage product, we generated a growth curve of MG1363 using CFU measurements.
Data
Analysis
The OD600 peak for MG1363 was approximately 1.2, which did not reach the detection limit of the machine, resulting in a growth curve that aligned well with the CFU data. The growth of MG1363 was not as robust as that of E. coli (DH5α), with its exponential phase occurring between 30 and 40 hours.
III. Cloning （BL21）
Cloning of pUC57 、pUCIDT-GM-CSF-PelB and pUCIDT-GM-CSF-Usp45 with pKJE7
Description
We conducted experiments to test whether the rhGM-CSF sequence and the chaperone protein pKJE7 could effectively function in E. coli for successful protein production. We utilized the BL21 strain for protein secretion. To prevent conflicts between antibiotic resistance and pKJE7, a low-temperature induction strain was employed.
To verify the efficacy and necessity of the chaperone protein and signal peptide, we transformed the following sequences:
| No. | 1 | 2 | 3 | 4 | 5 | 6 |
|---|---|---|---|---|---|---|
| Gene name | pUC57 | pUCIDT-GM-CSF-PelB | pUCIDT-GM-CSF-Usp45 | pKJE7 + pUC57 | pKJE7 + pUCIDT-GM-CSF-PelB | pKJE7 + pUCIDT-GM-CSF-Usp45 |
| Part | BBa_J70004 | BBa_K5361005 | BBa_K5361006 | BBa_K5361007 + BBa_J70004 | BBa_K5361007 + BBa_K5361005 | BBa_K5361007 + BBa_K5361006 |
| Resistance | Ampicillin | Ampicillin | Ampicillin | Chloramphenicol + Ampicillin | Chloramphenicol + Ampicillin | Chloramphenicol + Ampicillin |
| Screted peptide | X | O | O | X | O | O |
| Chaperone | X | X | X | O | O | O |
Note: Sequences marked with pKJE7 indicate that this plasmid was pre-transformed into the BL21 strain.
Data
Lane 1: Ladder, Lane 2: pUC57, Lane 3: pUCIDT-GM-CSF-PelB, Lane 4: pUCIDT-GM-CSF-Usp45, Lane 5: pKJE7 + pUC57, Lane 6: pKJE7 + pUCIDT-GM-CSF-PelB, Lane 7: pKJE7 + pUCIDT-GM-CSF-Usp45, Lane 8: pUC57 control, Lane 9: pUCIDT-GM-CSF-PelB control, Lane 10: pUCIDT-GM-CSF-Usp45 control, Lane 11: pKJE7 control.
Analysis
In lanes 2 through 4, pUC57, pUCIDT-GM-CSF-PelB, and pUCIDT-GM-CSF-Usp45 were successfully transformed into BL21. The middle band represents BL21’s endogenous plasmid dedicated to protein folding, while the top band appears to be an unidentified artifact.
In lanes 5 through 7, the top band corresponds to the pKJE7 plasmid, and the middle band represents BL21’s endogenous plasmid for protein folding. However, pUC57, pUCIDT-GM-CSF-PelB, and pUCIDT-GM-CSF-Usp45 were not successfully transformed into BL21.
Simultaneous Co-Transformation of pUCIDT-GM-CSF-PelB, pUCIDT-GM-CSF-Usp45 and pKJE7 in BL21
Description
Following the failure of our previous experiment, we explored an alternative approach by attempting a simultaneous transformation of two plasmids using heat shock. We hypothesized that since BL21 is not a strain specifically optimized for transformations and already contains plasmids assisting with protein folding, it may be less conducive to multiple transformations. Thus, we aimed to increase the success rate by transforming two plasmids at once.
Data
Lane 1: pUCIDT-GM-CSF-PelB & pKJE7 (selected with both chloramphenicol and ampicillin), Lane 2: pUCIDT-GM-CSF-PelB & pKJE7 (no antibiotic selection), Lane 3: pKJE7 control, Lane 4: pUCIDT-GM-CSF-PelB control.
Analysis
We successfully co-transformed pUCIDT-GM-CSF-PelB and pKJE7 into BL21, demonstrating that two different plasmids can be simultaneously introduced into the strain. However, the co-transformation of pUCIDT-GM-CSF-Usp45 and pKJE7 failed during antibiotic selection, as the cells were unable to survive in LB broth containing both chloramphenicol and ampicillin, preventing further electrophoresis analysis.
Cloning of pMG36E
Description
We acquired the pMG36E plasmid from Addgene, which was stored in E. coli by the supplier. To proceed, we first amplified the plasmid-containing E. coli and isolated the plasmid using erythromycin for selection (Figure II).
Data
Lane 1: Ladder, Lane 2: pMG36E (at half the concentration of the control), Lane 3: BL21 transformed with pMG36E, Lane 4: pMG36E control.
Analysis
We successfully extracted the pMG36E plasmid from E. coli and attempted to transform it into BL21. However, post-electrophoresis, no visible bands were observed. We hypothesize that the low plasmid concentration from the extraction may have reduced transformation efficiency, and the erythromycin selection was not sufficiently stringent, allowing some BL21 cells without the pMG36E plasmid to survive.
According to the data presented, Erythromycin exhibits significantly lower bactericidal activity compared to Ampicillin. This difference is particularly evident under the resistance selection conditions for the pMG36E plasmid, which involves an Erythromycin concentration of 200 µg/mL.
IV. Antimicrobial Effect of AMP
The bactericidal and inhibitory effects of AMP on Escherichia coli
Description
To model the bactericidal effects of antimicrobial agents on Gram-negative pathogens (simulated using Escherichia coli), we aim to establish a functional relationship between bacterial killing and growth, integrating data from growth curves. In our experiments, we utilize hBD-3 as the antimicrobial peptide (AMP). According to the literature, the minimum inhibitory concentration (MIC) of hBD-3 is approximately 10 μg/mL. However, to enhance the flexibility of our model, we have escalated the experimental dosages to 25 μg/mL and 50 μg/mL.
We hypothesize that higher concentrations will yield stronger antibacterial effects, but may also compromise the viability of Lactococcus lactis MG1363, necessitating careful dose optimization. To address this, we are conducting neural network simulations using the bactericidal data for MG1363 under various hBD-3 concentrations, enabling us to recommend optimal dosages between 10 μg/mL and 50 μg/mL based on the specific needs of users.
Data
Analysis
At 25 μg/mL, E. coli (DH5α) growth is effectively suppressed, while 50 μg/mL further accelerates bacterial killing, reducing cell numbers even during rapid replication phases. Both concentrations demonstrate significant antibacterial effects lasting approximately 7 to 8 hours, with higher concentrations providing prolonged inhibition of bacterial regrowth.
We also observed bacterial debris under the microscope, indicating that hBD-3 has both bactericidal and inhibitory effects on E. coli.
Note
We have set the initial bacterial load threshold at 10^6 CFU (colony-forming units). Although our simulations were conducted using an approximate starting population of 500,000 CFU, the model, derived from the generated function, demonstrates that antimicrobial administration at doses between 10 μg/mL and 50 μg/mL, as well as bacterial loads between 500,000 and 3 million CFU, can effectively control bacterial population growth. This prevents the harmful bacterial community (simulated using E. coli) from proliferating excessively.
AMP Resistance Issue
Description
To investigate whether AMP induces resistance similar to antibiotics, we designed an experiment where a second dose of AMP(25 μg/mL) was administered after E. coli had reached its stationary growth phase. The goal was to assess overall survival rates and determine if there was a decreasing trend.
Data
Analysis
Figure D-IV shows the CFU results 30 minutes after the first AMP treatment, measuring the number of viable bacteria. Statistical analysis revealed that both the 25 μg/mL and 50 μg/mL concentrations showed statistically significant (p < .001) effects at the 30-minute mark. However, after a second AMP treatment (Figure D-V), no statistically significant differences were observed between Group 2 (25 μg/mL) and Group 3 (50 μg/mL), indicating a trend of reduced antimicrobial effectiveness. This suggests the potential development of resistance, which poses a potential challenge and limitation for the application of our product.
Effects of AMP on Lactococcus lactis (MG1363)
Description
Wounds are often colonized by both Gram-negative bacteria and Gram-positive species such as Staphylococcus aureus. To enhance the flexibility and versatility of our product, we selected hBD-3, which has the capacity to kill both Gram-negative and Gram-positive bacteria. In future product designs, incorporating aptamers that detect specific biomarkers could enable targeted drug delivery to inhibit different bacterial species. However, a critical consideration is that hBD-3 may also harm Lactococcus lactis (MG1363). To ensure the preservation of MG1363 during treatment and determine the optimal timing for dressing changes, we conducted an analysis of AMP’s effects on MG1363.
Analysis
At hBD-3 concentrations of 25 μg/mL and 50 μg/mL, L. lactis was not completely eradicated and continued to grow steadily. Even when some bacterial killing occurred, a rebound in growth was observed within approximately 1 to 3 hours, with sustained upward growth thereafter.
V. Diffusion of GM-CSF versus Time
Description
To investigate the diffusion of GM-CSF within the chitosan hydrogel, we conducted this experiment to examine the changes in GM-CSF concentration over time (0, 1, 3, 6, 12, and 24 hours) by hGM-CSF ELISA kit.
Data
Analysis
As shown in Figure D-I, the diffusion concentration reached approximately 50 pg/mL after 3 hours (in 200 μL of PBS), about 100 pg/mL after 6 hours, and nearly 150 pg/mL after 24 hours.