Team:Michigan/Results

MSBT 2017

Results

Experiment #1: How do different promoters in pThermolyze affect cell growth at 37C and 25C?

We performed this experiment three times, but it proved difficult to ensure all cultures began at 0.2-0.4 OD, so for our analysis, we discarded the data for all trial runs that did not start with a time 0 OD of 0.2-0.4. If we had not done this, we would be comparing cultures that were at very different stages of growth. The remaining data are shown in the table below. Unfortunately, this method of pruning the data for consistency meant that we did not have any valid trials for promoter J grown at 37C and only one trial for that promoter at 25C, so we have excluded promoter J from our analysis. As the weakest of the three promoters that we tested, it is unlikely that its exclusion impacts the results. We see no consistent drop in OD at any time point once the cultures with our killswitch and promoters were switched to incubating at 25C. Such a drop would have indicated our killswitch was lysing cells as we had hoped.

Trial Sample Time 0 0:20 0:40 1:00 1:20 1:40 2:00 2:20 2:40 3:00 3:20 3:40 4:00 4:20 4:40 5:00 5:20 5:40 6:00
1 P-25 0.354 0.379 0.497 0.44 0.471 0.497 0.573 0.621 0.601 0.694 0.803 0.912 0.951 1.132 1.077 1.246 1.505 1.362 1.484
2 P-25 0.33 0.365 0.447 0.473 0.507 0.638 0.657 0.777 0.913 0.945 1.045 1.162 1.321 1.301 1.525 1.666 1.782  
1 C-25 0.297 0.346 0.474 0.388 0.46 0.449 0.495 0.59 0.634 0.668 0.717 0.883 0.856 0.964 1.096 1.029 1.207 1.126 1.287
2 C-25 0.243 0.262 0.302 0.311 0.373 0.44 0.462 0.507 0.562 0.585 0.639 0.747 0.806 0.828 0.879 0.977 1.1  
3 C-25 0.268 0.32 0.287 0.306 0.309 0.415 0.409 0.472 0.502 0.545 0.684 0.67 0.763 0.754 0.853 0.929 0.884 0.998 1.06
1 T-25 0.258 0.233 0.287 0.277 0.268 0.335 0.304 0.442 0.33 0.436 0.479 0.463 0.649 0.549 0.555 0.681 0.599 0.794
3 T-25 0.293 0.319 0.307 0.312 0.359 0.498 0.449 0.502 0.535 0.604 0.689 0.697 0.756 0.911 0.935 0.96 0.992 1.062 1.109
2 J-25 0.368 0.505 0.541 0.603 0.712 0.862 0.863 0.962 1.064 1.169 1.301 1.421 1.611 1.734 1.846 2.059 2.236  
1 L-25 0.28 0.371 0.468 0.374 0.413 0.423 0.553 0.581 0.595 0.633 0.715 0.755 0.839 0.984 1.106 1.095 1.225 1.33 1.341
2 L-25 0.36 0.431 0.435 0.533 0.604 0.741 0.776 0.841 0.952 1.123 1.175 1.286 1.517 1.671 1.681 1.921 2.11  
1 P-37 0.354 0.554 0.726 0.929 1.157 1.41 1.642 2.026 2.276 2.43  
1 C-37 0.363 0.646 0.658 0.913 1.025 1.227 1.492 1.718 2.061 2.35 2.452  
2 C-37 0.362 0.513 0.724 0.992 1.126 1.37 1.598 1.896 2.1 2.249  
3 C-37 0.331 0.472 0.773 1.033 1.212 1.428 1.604 1.991 2.187 2.465  
1 T-37 0.235 0.342 0.392 0.512 0.59 0.657 0.893 1.117 1.43 1.595 1.857 1.999 2.385  
2 T-37 0.213 0.249 0.315 0.437 0.53 0.736 0.895 1.093 1.369 1.531 1.893 2.208 2.339  
3 T-37 0.291 0.523 1.003 0.998 1.161 1.41 1.636 1.948 2.283 2.439  
1 L-37 0.332 0.706 0.744 1.023 1.223 1.433 1.718 2.275 2.443                    


For the remaining trials, we calculated generation time from 20 minutes after half the cultures were switched to 25C until the 100 minute time point. We used the formula 80/(3.3*Log(Abs100/Abs20)), to calculate the generation time. We chose these two time points because after 20 minutes at 25C, the cultures should have cooled to their new temperature and stabilized, and 200 minutes was the last time point where no culture went over the measurable limit of 2.5 OD.


We then averaged these generation times across each type of culture (ie: P at 25C, P at 37C, etc.).


There are two main takeaways from these generation times. First, we see negligible differences in generation time between the five types of culture at 37C. This is promising because it tells us that our construct has no negative impact on cell growth at working conditions. In theory, producing holin, endolysin, antiholin, and TlpA36 could have put undue metabolic strain on the cell, but we see this is not the case. Also promising is the fact that there is a noticeable increase in generation time between our two control cultures (P and C), and the two cultures with our kill switch and promoters. This suggests that even though our kill switch did not outright lyse the cells, they do seem to be impacting growth rate, supporting the possibility that the selection of a stronger promoter may yield a viable killswitch.

To further establish that this increase in generation time is due to our killswitch and not just due to the shift in temperature, we can examine the ratio of the 25C cultures’ generation time to their matching 37C cultures’ generation time.


This confirms that promoter T is causing some increase of the generation time at 25C independent of the increase in generation time inherent in growth at a lower temperature. It appears promoter L may possible have an effect. Again, our killswitch did not lyse all of the cells as we intended it to, but if it is reducing growth rate, it may be lysing some of them. Those results suggest that a strong enough promoter could make our killswitch functional.

Experiment #2: Can pThermolyze transformed cells be stored as glycerol stocks and successfully recovered?

A bacterial lawn grew on all five plates in this experiment.

Experiment #3: Experiment #3: Can pThermolyze transformed cells be recovered from liquid cultures after storing at room temperature for 4 hours?


Since the cultures had slightly different concentrations of cells to begin with, the raw data above is better analyzed if we normalize each experimental type to its initial plate by looking at the ratio of the number of colonies at each time point to the number colonies on the t=0 plate.


Unfortunately, we see that the results do not identify a clear trend with respect to cell viability after plating for the strains tested. This confirms that our killswitch does not operate as we intended. Ideally, it would have shown a drop in cells that could be recovered the longer a culture was kept at 25C, with no cells recovered after 4 hours.

Can pThermolyze transformed cells be recovered from plates after storing at 4C for 16 hours?


Turbid liquid cultures grew in all 15 cultures, further confirming that our killswitch is ineffective with these promoters.

Conclusion

All four experiments we performed showed that pThermolyze could not effectively kill bacteria at 25C within a reasonable timeframe. Experiment #1 suggests that the version of pThermolyze that used promoter T to express holin and endolysin had some impact on cell growth at 25C vs 37C, suggesting that our killswitch plasmid’s design may have some potential. Perhaps if a strong enough constitutive promoter were found, our temperature controlled killswitch design would be functional.

Future Directions

We had planned to test more than three promoters in our killswitch design, but due to issues with our cloning and delays in the construction of pThermolyze, we only had time to test these three. The UC Berkeley iGEM team from 2008 (LINK) used holin and endolysin as part of an inducible kill switch design and reported a drop in OD from 2.5 to 0.5 in 3.5 hours when they expressed T4 holin and T4 endolysin using the pBAD promoter and 10mM arabinose. T4 antiholin was expressed in their design using an unnamed constitutive promoter. To continue developing pThermolyze we plan to find a constitutive promoter that is at least as powerful as the pBAD promoter, either in the literature or by performing transcription assays using a known reporter gene. We could use this promoter as a starting point and continue to tune the balance between holin and the antiholin expressed by the TlpA promoter until we find a combination that meets our design goals. After discussing our project with Dr. Andrew Lowell, Dr. Gyorgyi Csankovszki, and Dr. Randy Stockbridge as a part of our integrated human practices, we considered future applications of our design. In the future we would incorporate our construct into both pathogenic strains such as MRSA and non-pathogenic strains used in research. An ideal application would be teaching labs in schools so that research based classes could be more involved. These are directions we would like to examine for our project in the future.

Another important task would be to ensure that the construct was episomally integrated. Our construct is currently maintained in a plasmid. That plasmid is maintained because the cells are cultured in LB-Ampicillin and the plasmid confers Ampicillin resistance. If the cells were to escape into the environment, the plasmid would be quickly lost. While episomal integrations are still vulnerable to inactivating mutations, our construct would be maintained in the genome. These genomic integrations could be validated using PCR.