Team:TU Eindhoven/Demonstrate

Demonstrate

Demonstrate

In February this year, we started our project and set out on the goal to create a fast, specific and modular detection system for bacterial pathogens, with the greater goal of diminishing antimicrobial resistance in mind. Since then we have come a long way and have eventually been able to create such a system and coined it dCastect.

Fast and specific

The system starts with bacteriophage infection and ends with the detection of phage DNA which indicates the presence of the bacterial pathogen that is screened for. The bacteriophage binds to a specific bacterium, inserts its DNA and replicates to form progeny phages. Within a small timeframe, the cell lyses and the progeny phages are released (Figure 1, part 1 and 2). To demonstrate the functioning of our detection system, the clinically relevant T7 phage and E.coli bacteria were chosen. A dilution series of a T7 phage stock with known titer obtained from the QAMH was prepared. Subsequently, the phages were heated at 70 °C to extract their dsDNA (Figure 1, part 3) [1, 2]. Then, it was tested whether this specific T7 phage DNA resulting from these different titers of phages could be measured by using the paired dCas9-Split-NanoLuc system (Figure 1, part 4). This was the case for titers down to 1x109 PFU/mL (Figure 2). This titer corresponds with DNA concentrations in the picomolar range, assuming an approximately 1:1 relation between the amount of PFU and DNA, making these results in agreement with the earlier reported apparent limit of detection of our paired dCas9-Split-NanoLuc system.


Figure 1: dCastect system with paired dCas9-Split-NanoLuc.

Now for making the next step towards actually detecting bacteria, we performed phage/host dynamics monitoring experiments at QAMH to obtain parameters for our model, which could be used to determine a useful start titer of phages for detecting a clinically relevant concentration of bacteria. Based on the results described above (Figure 2), we know what minimal final phage titer to aim for so that the measured bioluminescence intensity of the end titer be significantly bigger than the intensity measured for the control sample, containing the initial phage titer. For a final titer of 1-10x109 PFU/mL, this intensity difference is significant if a start titer of ≤ 1x106 PFU/mL is used for instance, which means that when there is a titer increase from 1x106 to at least 1x109, we can say that the specific bacteria that was screened for, were indeed present. Unfortunately, we obtained too little experimental data to calculate the required modeling parameters to use the dynamic model for our specific phage/host pair and relevant conditions. However, judging by the phage growth data we have (Figure 3), this titer increase described above corresponds to an infection time of approximately 170 minutes in the case of a starting bacterial concentration of 6x107 CFU/mL. Moreover, we did manage to replicate the phage growth model for the case of Salmonella and its lytic phage as described by Santos (2014) [3], and based on this we can conclude a rise in phage titer from 1x107 PFU/mL to 1x1010 PFU/mL takes about 3 hours at an MOI of 1.8x10-2 (see the Dynamic system model). As our detection method is modular, it should also work for Salmonella phage DNA detection.

This is only a first proof-of-concept, and albeit suboptimal, it shows that with dCastect, specific phage DNA can be detected, indicating which specific bacterial strain was present in the sample due to the high specificity of the phage-bacteria couple. Moreover, this result can be generated relatively rapidly since the measurable titer can be reached within 3 hours, validating the second objective of dCastect as well.


Figure 2: Bioluminescence intensity for varying phage titers.


Figure 3: Phage concentrations determined by experiments at QAMH. Starting bacterial concentration is 6x107 CFU/mL.

Specific and modular

Specificity of dCastect is not only reached through use of the natural specificity of bacteriophages for bacteria but also by use of the specificity of the dCas9 protein. Addition of non-target DNA does not result in an increase in bioluminescence intensity, making the detection of the DNA itself specific as well, reducing the chance of false positives (Figure 2). The third objective of the detection system is modularity, so any bacteria can be detected and any bacterial infection can be diagnosed without having to make too many alterations to the detection system itself. This modularity is reached through the internal properties of dCas9, that can bind to any dsDNA when the right complementary guide RNA is added. The right guide RNA can easily be found by using our own guide RNA tool. Moreover, our whole system is not limited in signal in interspace distances from 15 until 110 base pairs, making the whole system even more modular due to minimal internal restrictions.

Conclusions

To conclude, in the past months we have developed a novel detection system that is fast, specific and modular and is able to determine the source bacteria, enabling faster diagnosis of bacterial infections. This can be used for more specific treatment of these infections and with that, reduce antimicrobial resistance.

References

  1. Fekete, A., Rontó, G., Feigin, L. A., Tikhonychev, V. V., & Modos, K. (1982). Temperature dependent structural changes of intraphage T7 DNA. Biophysics of structure and mechanism, 9(1), 1-9.
  2. Vörös, Z., Csík, G., Herényi, L., & Kellermayer, M. (2018). Temperature-dependent nanomechanics and topography of bacteriophage T7. Journal of virology, 92(20), e01236-18.
  3. Santos, S. B., Carvalho, C., Azeredo, J., & Ferreira, E. C. (2014). Population dynamics of a Salmonella lytic phage and its host: implications of the host bacterial growth rate in modelling. PloS one, 9(7), e102507.