Photoswitchable antibiotics
Spirofloxacin is a conjugate of the broad-spectrum antibiotic ciprofloxacin, and a spiropyran with photoswitching activity (see Figures 1 and 2) [1]. The resulting compound has a much higher antibiotic activity after irradiation with 365 nm light as a result of a conformational change in the spiropyran moiety [1]. Spirofloxacin quickly deactivates under natural light, and more slowly in the dark [1]. This prevents the activation of the antibiotic by sequential irradiation with all wavelengths. In principle, photoswitchable antibiotics can be tuned to respond to any wavelength using molecular engineering techniques [1].
Spirofloxacin was kindly provided by the research group of Ben L. Feringa. This antibiotic was used by us as a component in a photoresponsive biological lock system. Spores of the key strain (resistant to the antibiotic) are mixed with spores of non-resistant, decoy strains. The antibiotic (in its ‘off’ state) is given to the recipient together with the spores. The recipient must activate the antibiotic with the ‘secret’ wavelength before adding it to the culture medium. As a result, the resistant key strain will grow, while the susceptible decoy strains will not. When DNA is extracted from a correctly treated culture, it will be predominantly the DNA of the key strain. If the culture is not treated correctly, the decoy strains will outcompete the key strain, making it nearly impossible to find the correct DNA sequence.
Experiments
We carried out two pilot experiments to determine if Bacillus subtilis 168 is resistant enough to UV to survive the exposure needed to activate the antibiotic, and to demonstrate the switching activity of the antibiotic.
UV resistance test:
- Resuspend spores in 50mL LB medium
- Transfer to sterile, open beaker (glass and plastic absorb UV) and place under 365nm UV light
- Take a 1mL sample for plating and 1mL sample for OD600 measurement at t = 0, 10, 20, 30, 35, 40, 45, 50, 55, and 60 minutes.
- Plate 100 µL, 10 µL and 1 µL in duplicate on LB agar.
- Culture overnight at 37°C
- Compare the growth on the plates
| t(min) | OD600 |
|---|---|
| 0 | 0.055 |
| 10 | 0.055 |
| 20 | 0.054 |
| 30 | 0.053 |
| 35 | 0.053 |
| 40 | 0.053 |
| 45 | 0.051 |
| 50 | 0.052 |
| 55 | 0.049 |
| 60 | 0.050 |
There is a lot of growth on all the plates, even the 60 minute plate with 1 µL plated culture, as shown in Figures 3 and 4. This result, along with the OD600 measurements suggest that Bacillus subtilis 168 spores are resistant enough to UV (365nm) to survive UV exposure necessary to activate a spiropyran-ciprofloxacin conjugate.
Antibiotic activity spot test:
To demonstrate that spirofloxacin possesses antibiotic activity, a spot test was carried out. B. subtilis 168 was plated on LB agar, and a small amount of spirofloxacin powder was placed in the middle of the plate. There is an area around the antibiotic that has no growth (see Figure 5). This result suggests that even in its non-irradiated state, spirofloxacin has some antibiotic activity.
Photoswitching activity test
The activated (365nm irradiated merocyanin state, open ring) spirofloxacin has a higher absorption peak at 590 nm than the deactivated (thermally adapted spiropyran state, closed ring), as reported by Feringa and co-workers [1] (see Figure 6).
In order to demonstrate that spirofloxacin undergoes a change in molecular conformation upon irradiation with 365 nm light, we set up the following experiment.
- Stock solution of spirofloxacin was prepared by dissolving 6.94 mg of spirofloxacin in 10 mL DMSO. Working solutions were prepared by diluting the stock solution in MilliQ H2O to the appropriate concentration.
- Prepared 2 mL [100 µM] spirofloxacin in MilliQ H2O
- Irradiated 1 mL of the solution for 60 minutes under 365 nm light
- The other 1 mL was kept at ambient conditions
- Measured OD590, using MilliQ H2O as a blank (open ring form of spirofloxacin absorbs at that wavelength).
| UV Irradiation | OD590 |
|---|---|
| λ- | 0.014 |
| λ+ | 0.108 |
Conclusion: There is a clear difference in absorption after irradiation (λ+). This suggests that a change in molecular conformation has taken place as a result of irradiation with 365 nm light. This result indicates that the photoswitch is active, and undergoes photoisomerization under UV light.
MIC and MBC tests – Ciprofloxacin, Wild-Type B. subtilis 168
Minimal Inhibitory Concentration (MIC) is the lowest concentration of antibiotic which prevents growth of bacteria. Minimal Bactericidal Concentration (MBC) is the lowest concentration that kills bacteria. Ciprofloxacin MIC and MBC tests were carried out on wild-type Bacillus subtilis 168 to determine the concentration of the antibiotic that this strain could tolerate.
Two tests were carried out with ciprofloxacin. The growth of B. subtilis 168 in a 96-well plate with a range of ciprofloxacin concentrations were monitored for 18 and 24 hours. The positive control was used to construct the growth curve (see Figure 7). For the first test a range of 25 nM to 2,500 nM was used. Results showed that the MIC was somewhere between 100 nM and 200 nM (see Figure 8). The second time a range of 100 nM to 200 nM was tested to determine the MIC more precisely. Results showed that the MIC of ciprofloxacin on B. subtilis 168 was 160 - 170 nM (see Figures 9 and 10).
Now that we obtained the MIC value, we determined the MBC. 10 µL of 160 nM, 170 nM and 180 nM cultures from the 96-well plate were plated on LB agar in duplicate. These were incubated overnight at 37°C. Only the 180 nM plate had no growth, so we take that to be the MBC.
Results:
Engineering ciprofloxacin resistance in Bacillus subtilis 168
For our system, we needed a strain of B. subtilis that is resistant to ciprofloxacin (and the spiropyran conjugate, spirofloxacin). Such a strain was not found in literature, so we engineered resistance via two parallel approaches.
Quinolone resistance gene qnrS1
We introduced the gene qnrS1 into our B. subtilis strain. The product of this gene alters the surface topology of the enzyme DNA gyrase, which is the target of ciprofloxacin. As a result, the antibiotic can no longer bind to gyrase, and thus does not inhibit DNA transcription. (Learn more about the design of our ciprofloxacin resistance BioBrick)
Directed evolution
A single colony of B. subtilis 168 was found to be growing on a 190 nM ciprofloxacin agar plate. This was re-plated onto LB agar medium with increasing antibiotic concentrations. After two months, the strain was growing on 25 µM ciprofloxacin plates (see Figure 11).
MIC tests – Ciprofloxacin, wild-type E. coli top10, E. coli top10 (carrying qnrS1) and B. subtilis 168 (carrying qnrS1)
MIC tests were carried out as done previously. The MIC values for the wild-type strains was found to be between 100-130 nM for Escherichia coli and 160 - 170 nM for B. subtilis (see Figures 12 and 10). MIC for E. coli top10 and B. subtilis 168 carrying the qnrS1 ciprofloxacin resistance gene was determined to be between 1000 - 2000 nM, and 400 - 500 nM (see Figures 13 and 14). This is a significant improvement in antibiotic tolerance (approximately 10x more resistance for E. coli and 2.3x more resistance for B. subtilis).
Results:
A MIC test was carried out on three colonies of B. subtilis 168 transformed with the qnrS1 resistance gene. Colony 1 was the most resistant, with a MIC between 400 and 500 nM (see Figure 14).
A MIC test was carried out on ciprofloxacin resistant B. subtilis 168 obtained via directed evolution (see Figure 11). The MIC value was in excess of 2,000 nM (see Figure 17), so a second MIC test was set up using a higher concentrations of ciprofloxacin.
Since the MIC of our ciprofloxacin-resistant B. subtilis 168 isolate could not be determined in our previous experiment, we set up another MIC test with a higher range of ciprofloxacin concentrations. We examined growth of this isolate in range of ciprofloxacin concentrations from 500 nM to 30,000 nM and compared it to wild-type B. subtilis 168 under the same conditions (see Figure 18). The MIC for this isolate was determined to be between 20,000 and 30,000 nM (see Figure 19). The wild-type strain grew very poorly even with the lowest concentration of ciprofloxacin (500 nM), never surpassing OD600 = 0.11. In contrast, the resistant B. subtilis isolate reached an OD600 = 0.3 in the presence of 500 nM after approximately 12 hours of growth. Cultures with a concentration of 17,500 nM reached an OD600 = 0.17 after 18 hours of growth. Even the cultures with 27,500 nM ciprofloxacin showed some growth after 18 hours, although very little, reaching an OD600 = 0.12 (which is still more than the wild-type cultures with 500 nM ciprofloxacin).
To confirm that this was really Bacillus subtilis, a PCR was set up, using primers that amplify the 16s rRNA region specific to B. subtilis. We obtained a clear bands of identical size for both the resistant isolate and wild-type B. subtilis 168 (see Figure 20). To prove conclusively that our isolate is in fact B. subtilis, the PCR products were sent for sequencing. This isolate would serve as a perfect starting point for the engineering of a highly resistant B. subtilis strain.
B. subtilis 168 appears to be naturally more resistant to ciprofloxacin than E. coli top10, but not as resistant as E. coli carrying the qnrS1 resistance gene (see Figure 21).
MIC tests – spirofloxacin
Test 1
Nine tests were carried out to determine the MIC of spirofloxacin on B. subtilis 168, E. coli BL4, E. coli DH5α and E. coli CS1562. The test with E. coli were done due to the inconsistent results we obtained with B. subtilis. Spirofloxacin stock solution was diluted in LB to the working concentrations. 3 mL cultures with different concentrations of spirofloxacin were inoculated and incubated overnight in the dark at 37°C in a shaking waterbath. OD600 was measured the next day.
- Strain: B. subtilis 168
- 10 µL OD=0.001 spores per 2 mL culture
- UV (365nm) irradiation durations: 0, 30, 45, 60 minutes
- + control: spores, 0 µM spirofloxacin
- - control: no spores, 5 µM spirofloxacin
Conclusion: Repeat with higher spirofloxacin concentrations
| [Spirofloxacin] nM | OD600= (λ-) | OD600= (λ+) |
|---|---|---|
| 100 | overgrown | overgrown |
| 200 | overgrown | overgrown |
| 300 | overgrown | overgrown |
| 400 | overgrown | overgrown |
| 500 | overgrown | overgrown |
| 700 | overgrown | overgrown |
| 1000 | overgrown | overgrown |
| 1250 | overgrown | overgrown |
| 1500 | overgrown | overgrown |
| 2000 | overgrown | overgrown |
Test 2
- Strain: B. subtilis 168
- 10 µL OD=0.001 spores per 2 mL culture
- UV (365nm) irradiation durations: 0, 60 minutes
- + control: spores, 0 µM spirofloxacin
- - control: no spores, 5 µM spirofloxacin
Conclusion:
- A lot of growth in all cultures except -controls.
- Measured OD of 2 µM and 10 µM cultures. B. subtilis 168 appears to grow slightly better with spirofloxacin (which is counterintuitive, given that it's an antibiotic).
- There is slightly less growth in the 10 µM cultures than in the 2 µM cultures.
- There is a slight difference in activity between λ+ and λ- cultures. λ+ cultures had less growth, as expected, but the difference was not as big as expected.
- Difference in + controls can be attributed to UV-induced damage of the cells.
| [Spirofloxacin] µM | OD600= (λ-) | OD600= (λ+) |
|---|---|---|
| - control | 0 | 0.002 |
| 2 | 1.512 | 1.364 |
| 2.5 | - | - |
| 3 | - | - |
| 3.5 | - | - |
| 4 | - | - |
| 4.5 | - | - |
| 5 | - | - |
| 5.5 | - | - |
| 6 | - | - |
| 10 | 1.388 | 1.128 |
| + control | 1.06 | 0.821 |
Test 3
- Strains: B. subtilis 168, E. coli DH5α
- 10 µL OD=0.001 B. subtilis spores per 2 mL culture
- UV (365nm) irradiation durations: 0, 60 minutes
- + control: spores, 0 µM spirofloxacin
- - control: no spores, 10 µM spirofloxacin
- DMSO control: 10% DMSO + spores
- E. coli control: 2 µL overnight E. coli culture per 2 mL culture i
Conclusion:
- Growth is somewhat inhibited at 100 µM, indicating antibiotic activity.
- There is slightly more growth after irradiation, which is unexpected and contradicts our previous result.
- DMSO appears to affect bacterial growth at this concentration, but not to the extent observed with 100 µM.
- E. coli appears to respond differently to the antibiotic. Growth in both λ+ and λ- cultures, but less in the λ+ culture, as expected.
| [Spirofloxacin] µM | OD600= (λ-) | OD600= (λ+) |
|---|---|---|
| 10 | 0.885 | 0.905 |
| 20 | 0.76 | 0.68 |
| 100 | 0.015 | 0.076 |
| - control | 0.002 | 0.01 |
| + control | 0.89 | 1.086 |
| DMSO control | 0.502 | 0.218 |
| E. coli control | 0.772 | 0.663 |
Test 4
LB medium absorbs heavily in the UV region of the spectrum. To overcome this issue we diluted the spirofloxacin in MilliQ H2O, and irradiated it before to adding to the cultures.
- Strains: B. subtilis 168
- 10 µL OD=0.001 B. subtilis spores per 2 mL culture
- UV (365nm) irradiation durations: 0, 60 minutes
- + control: spores, 0 µM spirofloxacin
- - control: no spores, 10 µM spirofloxacin
| [Spirofloxacin] µM | OD600= (λ-) | OD600= (λ+) |
|---|---|---|
| 0.25 | 0.474 | 0.543 |
| 0.5 | 0.585 | 0.703 |
| 1 | 0.769 | 0.667 |
| 5 | 0.775 | 0.718 |
| 10 | 0.685 | 0.907 |
| - control | 0.155 | 0.217 |
| + control | 0.439 | 0.62 |
Test 5
There was contamination in our last - control. This time, additional steps were taken to minimize contamination and maximize UV exposure. Every surface was thoroughly cleaned with ethanol. Tips and LB were already autoclaved, but were autoclaved a second time right before the experiment. Spirofloxacin was irradiated on a sterile petri dish and filter sterilized with a 0.22 µm Whatman filter before being added to the cultures.
- Strains: B. subtilis 168, E. coli BL4
- 10 µL OD=0.001 B. subtilis spores per 2 mL culture
- 2 µL overnight E. coli culture per 2 mL culture
- UV (365nm) irradiation durations: 0, 60 minutes
- + control: spores, 0 µM spirofloxacin
- - control: no spores, 10 µM spirofloxacin
- E. coli control: 10 µM spirofloxacin
| [Spirofloxacin] µM | OD600= (λ-) | OD600= (λ+) |
|---|---|---|
| 10 | 2.094 | 2.03 |
| - control | 0.01 | 0.009 |
| + control | 2.746 | 2.8 |
| E. coli control | 0.021 | 0.012 |
Conclusion:
- No significant difference in growth between λ- and λ+ cultures, or the +/- controls
- Growth of E. coli is inhibited, but appears to be slightly higher with λ-, which is an expected result.
Test 6
Due to the inconsistent results we observed, we set up an experiment to test the activity of spirofloxacin on E. coli CS1562. This is the strain used by the authors of the paper where spirofloxacin was first reported [1].
- Strain: E. coli CS1562
- 7 µL overnight E. coli culture per 2 mL culture
- UV (365nm) irradiation durations: 0, 60 minutes
- + control: E. coli, 0 µM spirofloxacin
- - control: no cells, 1.25 µM spirofloxacin
- Contamination control: 0 µM spirofloxacin, no cells
Conclusion:
- No significant difference in growth between λ- and λ+ cultures, except with .25 µM (λ+ grows better) and 1.25 µM (opposite effect). Results are contradictory.
- No significant difference in growth between different spirofloxacin concentrations, except at 10 µM where growth was inhibited.
- +/- controls look as expected.
- There is contamination in the contamination control cultures, so it is uncertain if these results are reliable.
| [Spirofloxacin] µM | OD600= (λ-) | OD600= (λ+) |
|---|---|---|
| 0.25 | 0.177 | 1.036 |
| 0.625 | 1.265 | 1.289 |
| 1 | 1.244 | 1.277 |
| 1.25 | 1.231 | 0.515 |
| 2 | 1.238 | 1.309 |
| 10 | 0.016 | 0.015 |
| - control | 0.009 | 1.131 |
| + control | 0.864 | 0.535 |
| Contamination control | 0.522 | 0.57 |
Test 7
- Strain: E. coli CS1562
- 1 µL overnight E. coli culture per 2 mL culture
- UV (365nm) irradiation durations: 0, 60 minutes (filter sterilized after irradiation)
- + control: E. coli, 0 µM spirofloxacin
- - control: no cells, 5 µM spirofloxacin
- Contamination control: 0 µM spirofloxacin, no cells.
Conclusion:
- Most λ+ cultures have slightly less growth than their counterpart λ- cultures, which is unexpected.
- No significant difference in growth between different spirofloxacin concentrations
- This time there is growth in the 10 µM cultures
- λ-, 5 µM appears to be an anomaly. Perhaps we forgot to add spirofloxacin to that culture?
- All controls look as expected.
| [Spirofloxacin] µM | OD600= (λ-) | OD600= (λ+) |
|---|---|---|
| 0.5 | 1.104 | 0.766 |
| 0.625 | 0.28 | 0.78 |
| 1.25 | 1.216 | 1.101 |
| 3 | 1.207 | 1.094 |
| 5 | 0.002 | 1.083 |
| 7 | 1.24 | 1.135 |
| 8 | 1.134 | 1.102 |
| 9 | 1.102 | 1.081 |
| 10 | 0.929 | 0.89 |
| - control | 0.002 | -0.002 |
| + control | 0.711 | 0.854 |
| Contamination control | 0.015 | 0.007 |
Test 8
- Strain: E. coli CS1562
- 1 µL overnight E. coli culture / mL
- UV (365nm) irradiation durations: 0, 60 minutes (filter sterilized after irradiation)
- + control: E. coli, 0 µM spirofloxacin
- - control: no cells, 5 µM spirofloxacin
- Contamination control: 0 µM spirofloxacin, no cells
| [Spirofloxacin] µM | OD600= (λ-) | OD600= (λ+) |
|---|---|---|
| 5 | 0.003 | 0.003 |
| 10 | 0.005 | 0.741 |
| - control | -0.002 | 0 |
| + control | 0.999 | 1.036 |
| Contamination control | 0 | -0.004 |
Conclusion:
- No growth in both 5 µM cultures, but there is growth in 10 µM, λ+ culture, which is unexpected.
- There is growth in the 10 µM, λ+ culture, but not its counterpart λ- culture. This is the opposite of the expected result.
- All controls look as expected.
Test 9
- Strain: E. coli CS1562
- 1 µL overnight E. coli culture / mL
- UV (365nm) irradiation durations: 0, 30 minutes (filter sterilized after irradiation)
- + control: E. coli, 0 µM spirofloxacin
- - control: no cells, 5 µM spirofloxacin
- Contamination control: 0 µM spirofloxacin, no cells
Conclusion:
- No growth in any of the cultures, except with 0.625 µM spirofloxacin.
- No significant difference in growth between .625 µM, λ- and .625 µM λ+ cultures
- Controls do not look good.
| [Spirofloxacin] µM | OD600= (λ-) | OD600= (λ+) |
|---|---|---|
| 0.625 | 0.294 | 0.3 |
| 1.25 | 0.001 | 0.004 |
| 2 | 0.001 | 0 |
| 4 | 0.001 | 0.002 |
| 5 | 0.001 | 0.001 |
| 7 | 0 | 0.001 |
| 9 | 0 | 0.006 |
| 10 | 0.001 | 0.003 |
| - control | 0.366 | 0.322 |
| + control | 0.727 | 0.541 |
| Contamination control | 0.163 | 0.292 |
Overall conclusion – spirofloxacin:
Our results are inconsistent and sometimes contradictory. Some of the difference between λ- and λ+ cultures could be attributed to the difference in absorption of the two spirofloxacin photoisomers, or due to the damaging effect of UV light on DNA. We were unable to achieve a large enough difference in growth between cultures with irradiated and non-irradiated spirofloxacin to be able to incorporate it in our project. However, under the right conditions (which would take longer to determine), or with further engineering of the compound, we believe spirofloxacin could be used as a component in a photoresponsive biological lock system.
Conclusion and Discussion
Photoswitchable antibiotics present an interesting option for a biological lock system. Although we did not succeed in incorporating spirofloxacin in our project due to the insufficient difference in bioactivity between the two photoisomers, we believe that it could be done with further optimization of our experimental setup, or with further engineering of the compound. In order to endow our B. subtilis 168 strain with resistance to ciprofloxacin (and therefore, to spirofloxacin), we introduced the qnrS1 fluoroquinolone resistance gene. This gene was part of a resistance cassette consisting of the pATPI promoter, a ribosomal binding site, qnrS1, and a double terminator. The cassette was assembled in E. coli top10, making it approximately 10 times more resistant than the wild-type. When the cassette was introduced into B. subtilis, resistance increased by a factor of 2.35. The difference in observed resistance increase after transformation with qnrS1 could be attributed to the fact that qnrS1 codon usage was optimized for E. coli, and therefore did not function as well in B. subtilis. Finally, a B. subtilis 168 isolate resistant to ciprofloxacin was obtained in a parallel approach via directed evolution. This isolate was found to tolerate ciprofloxacin concentrations well over 20,000 nM, and is an ideal starting point for further engineering of ciprofloxacin resistance. In the future, this isolate could be transformed with the qnrS1 and aac(6')-Ib-cr resistance genes to obtain a strain of B. subtilis that is highly resistant to ciprofloxacin, and possible other fluoroquinolone antibiotics.
References
- [1] Ciprofloxacin−Photoswitch Conjugates: A Facile Strategy for Photopharmacology Velema, Hansen, Lerch, Driessen, Szymanski, Feringa Bioconjugate Chem. 2015, 26, 2592−2597 DOI: 10.1021/acs.bioconjchem.5b00591
- [2] Photoswitching azo compounds in vivo with red light. J Am Chem Soc. 2013 Jul 3;135(26):9777-84. doi: 10.1021/ja402220t. Epub 2013 Jun 21.
