Team:William and Mary/Safety
Safety
While TheraPUFA is completely focused on design and modeling, safety has nevertheless been a key consideration throughout our entire project, given that we want to design, model, and critically evaluate a probiotic that could realistically be used as a broad-spectrum antiviral therapeutic. Therefore ensuring the safety of our design is paramount. Moreover, this need for safety has been emphasized through our integrated human practices, where we found that experts considered the safety of the probiotic both within, and outside the human body to be a top priority. The experts from our human practices have provided us with extensive suggestions on how to incorporate safety features in our design. We wished to ensure that the probiotic was safe for the person being treated with the probiotic; in addition, we needed to ensure that we incorporated safety features that addressed the probiotic in the environment when shed from patients. Our safety features are highlighted by several key aspects of TheraPUFA: our choice and design of our AA/DHA producing circuits; our use of a safe and commensal bacterium as our probiotic chassis; and through our design of two separate kill switches.
.
Choice of PUFAs and Circuit Design
After reviewing the literature for biological compounds that exert antiviral effects, we found that polyunsaturated acids, PUFAs, were the most broad-spectrum, and generally safest option. Our circuit design of switching from AA to DHA production upon detection of high cytokine levels is another safety feature that is unique to TheraPUFA. AA has both pro-inflammatory and anti-inflammatory metabolites, while DHA has mostly anti-inflammatory metabolites. As we don’t want the immune system to be suppressed during early stages of infection, our circuit only produces AA so that the immune system can mount the necessary response against the virus. However, if our circuit detects high levels of cytokines (such as during a cytokine storm), AA production will be halted and DHA will be produced. The produced DHA will then help to reduce inflammation. This switching behavior between AA and DHA ensures that the direct antiviral effects of these lipids will always be present, while also maintaining a safe level of inflammatory response.
.
Choice of Chassis
After consulting with stakeholders and experts, we decided on three criteria that would need to be fulfilled by the organism that we would decide to use as our probiotic chassis. First, the probiotic organism would need to be a commensal part of the healthy nasopharyngeal microbiome. Second, the organism must not cause disease, and must not be implicated with increased disease risk in healthy individuals. Finally, we would need to use a genus of bacteria that has already been demonstrated as a nasopharyngeal probiotic.
Consulting the literature, we found that Neisseria cinerea was able to fulfill all three of these requirements. First, N. cinerea is found in the nasal and oral microbiota of healthy adults and children, and is considered to be a commensal member (Custodio et al., 2020). Second, N. cinerea is a BSL1 organism, and is neither disease-causing, nor is known to be associated with disease (Custodio et al., 2020). Moreover, it has been found that N. cinerea may actually be beneficial, as it reduces the ability of N. meningitidis to associate with epithelial cells, potentially interfering with the ability of N. meningitidis to cause disease (Custodio et al., 2020). Finally, Neisseria species have been demonstrated as safe to use as nasopharyngeal probiotics. Neisseria lactamica Y92-1009 has been used in human clinical trials to help prevent colonization by N. meningitidis (Deasy et al., 2015). The authors found that the probiotic was well-tolerated by the volunteers, with no serious side effects reported, thereby demonstrating that commensal Neisseria species can be safely used as probiotics in the nasopharyngeal microbiome (Deasy et al., 2015).
Additionally, from the interviews we conducted with experts, we received concern that a nasopharyngeal probiotic will quickly end up migrating down into the lower respiratory tract and lungs. Therefore, we need to ensure that our probiotic is safe in these environments. Given the administration of N. lactamica by authors Deasy et al. was mediated by nasal droplets, it is likely that this probiotic experienced this migratory effect into the lower respiratory tract (Deasy et al., 2015). No serious side effects were reported by any of the roughly 300 volunteers who were administered the probiotic (Deasy et al., 2015). Therefore, lower respiratory complications are not likely to be caused by administration of a commensal, BSL1 Neisseria species, such as N. cinerea.
As Neisseria cinerea meets all of the requirements we set out to identify a safe-to-use nasal probiotic species, we decided to utilize this bacterium as the chassis for our nasopharyngeal probiotic.
.
Kill Switch Designs
Environmental Kill Switch
To ensure that our bacteria are unable to replicate outside of the body, we created a novel, temperature-sensitive kill switch for Neisseria cinerea. With this designed kill switch, our probiotic should be unable to replicate below a temperature of about 37 degrees Celsius (such as when outside of the body). Such circuits have been designed for E. coli, such as the “Cryodeath” kill switch (Stirling et al., 2017). However, no such kill switch has been designed for Neisseria species. Therefore, we had to design our own conceptual temperature sensitive kill switch with parts native to Neisseria species. Our designed circuit utilizes the FitAB toxin-antitoxin system (Mattison et al., 2006), combined with the RNA thermometer CssA (Barnwal et al., 2016).
FitAB is a toxin-antitoxin (TA) system that was first isolated in Neisseria gonorrhoeae (Mattison et al., 2006). FitAB is a type II TA system, meaning that the antitoxin (FitA) is able to inhibit the toxin (FitB) via direct protein-protein binding (Lobato-Márquez et al., 2016). When FitA is not present, FitB is able to exert its toxic activity via ribonuclease and RNA binding activity (Warrender, 2019).
CssA is a temperature sensitive RNA thermometer native to Neisseria meningitidis (Barnwal et al., 2016). At temperatures below 37 degrees Celsius, CssA adopts a hairpin structure that blocks ribosomal access to the downstream RBS, preventing protein translation (Barnwal et al., 2016). As the temperature reaches 37 degrees Celsius, CssA is able to gradually unfold, and allow for protein translation to occur, with maximal protein translation occurring past 42 degrees Celsius (Barnwal et al., 2016).
For our promoter, we decided to use the native FitAB promoter (Wilbur et al., 2005). For an RBS, we decided to use the CssA RBS sequence for FitB, and the full CssA sequence for FitA. Our coding sequences are FitA and FitB, and we decided to use the ThiC terminator region, found in Neisseria meningitidis, as our terminator (Righetti et al., 2005).
Inside the Body Kill Switch
In addition to choosing a bacterial species that does not cause disease in healthy individuals, along with being a common commensal organism, we also ensure that our probiotic is safe within the body by incorporating a chemically-inducible kill switch. This way, the probiotic can rapidly be eliminated, without requiring antibiotic treatment that would disrupt the nasopharyngeal microbiome. An excellent candidate for this inducible kill switch is the aspirin-inducible expression system described by Chen et al., 2019, as aspirin is a widely used and safe drug. The aspirin-inducible kill switch works by utilizing the aspirin-inducible promoter, Psal, that originates from Acinetobacter baylyi ADP1 (Chen et al., 2019). Psal is repressed by its cognate transcriptional effector, SalR. Upon addition of aspirin, SalR transitions from its effector-free form (SalRr) to its effexor-bound form (SalRa) (Chen et al., 2019). SalRr and SalRa compete for DNA binding, with the former acting as a transcriptional repressor, while the latter acts as a transcriptional activator (Chen et al., 2019).
To get the desired aspirin-inducible behavior, we need to constitutively express SalR. The promoter we have chosen to express SalR is PopaB, a strong constitutive promoter found in Neisseria gonorrhoeae (Ramsey et al., 2012).
Next, to accomplish the cytotoxic effect, we need to conditionally express a toxin that will result in the cell death of Neisseria cinerea, but will not harm the host or other bacterial cells in the microbiome. To this end, we investigated other toxin-antitoxin systems native to Neisseria, and decided to use the protein ngζ_1 (Rocker et al., 2018). Ngζ_1 is a zeta toxin found in the Neisseria gonorrhoea ngε_1/ngζ_1 toxin-antitoxin system (Rocker et al., 2018). Ngζ_1 functions by phosphorylating UDP-sugars that are necessary for peptidoglycan synthesis (Rocker et al., 2018). Phosphorylation of these UDP-sugars prevents their utilization in peptidoglycan synthesis, which results in weakened cell wall integrity and eventual cell lysis (Rocker et al., 2018).
Our final circuit functions and is composed as described: first, SalR is constitutively expressed by the PopaB promoter, with protein translation being mediated by the CssA RBS, and termination being mediated by the ThiC terminator, as described earlier. Next, the ngζ_1 toxin is placed under the control of the Psal promoter, and the same CssA RBS and ThiC terminator. In the absence of aspirin, SalR adopts its repressor form, SalRr, preventing the transcription of the zeta toxin. Upon addition of aspirin SalR is converted into its active form, SalRa, resulting in expression of ngζ_1 from the Psal promoter, which then inhibits peptidoglycan synthesis and leads to cell lysis.
.
References
Chen, J. X., Steel, H., Wu, Y. H., Wang, Y., Xu, J., Rampley, C. P., ... & Huang, W. E. (2019). Development of aspirin-inducible biosensors in Escherichia coli and SimCells. Applied and environmental microbiology, 85(6).
Custodio, R., Johnson, E., Liu, G., Tang, C. M., & Exley, R. M. (2020). Commensal Neisseria cinerea impairs Neisseria meningitidis microcolony development and reduces pathogen colonisation of epithelial cells. PLoS pathogens, 16(3), e1008372.
Deasy, A. M., Guccione, E., Dale, A. P., Andrews, N., Evans, C. M., Bennett, J. S., ... & Read, R. C. (2015). Nasal inoculation of the commensal Neisseria lactamica inhibits carriage of Neisseria meningitidis by young adults: a controlled human infection study. Clinical infectious diseases, 60(10), 1512-1520.
Lobato-Márquez, D., Díaz-Orejas, R., & García-del Portillo, F. (2016). Toxin-antitoxins and bacterial virulence. FEMS microbiology reviews, 40(5), 592-609.
Mattison, K., Wilbur, J. S., So, M., & Brennan, R. G. (2006). Structure of FitAB from Neisseria gonorrhoeae bound to DNA reveals a tetramer of toxin-antitoxin heterodimers containing pin domains and ribbon-helix-helix motifs. Journal of Biological Chemistry, 281(49), 37942-37951.
Ramsey, M. E., Hackett, K. T., Kotha, C., & Dillard, J. P. (2012). New complementation constructs for inducible and constitutive gene expression in Neisseria gonorrhoeae and Neisseria meningitidis. Applied and environmental microbiology, 78(9), 3068-3078.
Righetti, F., Materne, S. L., Boss, J., Eichner, H., Charpentier, E., & Loh, E. (2020). Characterization of a transcriptional TPP riboswitch in the human pathogen Neisseria meningitidis. RNA biology, 17(5), 718-730.
Rocker, A., Peschke, M., Kittilä, T., Sakson, R., Brieke, C., & Meinhart, A. (2018). The ng_ζ1 toxin of the gonococcal epsilon/zeta toxin/antitoxin system drains precursors for cell wall synthesis. Nature communications, 9(1), 1-11.
Stirling, F., Bitzan, L., O’Keefe, S., Redfield, E., Oliver, J. W., Way, J., & Silver, P. A. (2017). Rational design of evolutionarily stable microbial kill switches. Molecular cell, 68(4), 686-697.
Warrender, A. (2019). Characterisation of the FitAB Toxin-Antitoxin System in Neisseria gonorrhoeae (Thesis, Master of Science (Research) (MSc(Research))). The University of Waikato, Hamilton, New Zealand.
Wilbur, J. S., Chivers, P. T., Mattison, K., Potter, L., Brennan, R. G., & So, M. (2005). Neisseria gonorrhoeae FitA interacts with FitB to bind DNA through its ribbon− helix− helix motif. Biochemistry, 44(37), 12515-12524.