Team:Oxford/Safety

The main aim of the ProQuorum probiotic is to provide a safer and more convenient alternative to the existing antibiotic therapies for the C. difficile infection. As such, eliminating the possible side-effects of our system has been at the forefront of our endeavours.

In contrast, typically prescribed antibiotics for clostridial infections such as metronidazole, vancomycin and fidaxomicin, can cause common gastro-intestinal adverse effects, but may, in rare cases, lead to more severe effects such as liver and/or renal insufficiency, as well as low leukocyte counts. To this is added the risk of antimicrobial resistance evolution, which is observed frequently in the case of nosocomial infections such as C. difficile. At the same time, antibiotic treatment further destabilizes the commensal and symbiotic gut flora, which appears to be a contributing factor to the induction of virulence in otherwise asymptomatic C. difficile colonization in the first place.

Thus, in order to avoid all issues associated with antibiotic administration, we are proposing a more specific targeting strategy by delivering the CD27L endolysin to the large intestine with the aid of the Lactobacillus reuteri chassis, which has been shown to safely colonise the gut milieu.

Regarding the genetic engineering of L. reuteri, there is little reason to believe there will be any significant risk to the patient; the activity of the CD27L endolysin appears to be highly specific to a narrow class of glycopeptide molecules, while the two-component signalling system should not in itself be detrimental to human health in any way.

That being said, we are aware that unforeseen adverse effects can always occur and, as with almost every substance/micro-organism introduced in the human body, a risk of developing an immune response leading to an allergic reaction exists in certain individuals. In order to ascertain whether such risks are negligible or not, preliminary tests on in vitro models of the human gut, as well as on commonly used mammalian model organisms will need to be performed. Once a good safety profile is established in these conditions, carefully designed clinical trials can be employed to provide a clear and objective assessment of all risks to the human host.

In addition, we believe that while our inducible system was designed to limit the stress on the Lactobacillus chassis and aid in its proliferation and survival in the gut, it will also limit the amount of collateral damage on commensal gut species whose peptidoglycan cell wall CD27L might show cross-reactivity for. Thus, the ProQuorum system shows double specificity: both regarding induction and antimicrobial activity, ensuring its suitability as a preventative therapeutic probiotic against pathogenic C. difficile infections.

Further concerns about our probiotic could be raised about the potential issue of its spread into the environment and the risk of lateral genetic transfer of our parts to other microorganisms. To address the former concern first, we would like to note that, from our preliminary growth assays of the transformants, they appear to have reduced fitness relative to the wild type, and indeed, there is little reason to believe that the genes we are introducing could increase the inclusive fitness of the strain. In fact, we hypothesize that the ProQuorum probiotic would require continual administration during the therapeutic window to maintain effective population sizes in the gut.

However, we are of the opinion that exercising prudence would be most appropriate both from an environmental and ethical standpoint, such that we are considering the introduction of multiple auxotrophies in our chassis. We believe this has a two-fold advantage to the use of environmental kill switches:

  1. Most kill switches require the expression of several genes, which would further increase the burden on our strain already tasked with the expression of the ProQuorum system. In contrast, if the auxotrophies are established though gene deletions, the burden is decreased.
  2. Evolutionary pressures to lose either the auxotrophy or the kill switch will be very high outside the gut environment; however, it is conceivable that the probability of occurrence of a loss of function mutation for the kill switch is much higher than the probability of concomitant acquisition of the missing genes by lateral gene transfer. Indeed, if the deletions take place in separate operons (e.g. thymidine and vitamin B auxotrophy), the risk of such an occurrence is minimal.

At the same time, concerns could be raised about the possibility of curtailing treatment in a safe manner once it has been started. Such a problem could, in principle, be solved through the introduction of pharmacologically inducible kill switch. However, after careful consideration and consultation with bioethicist Frances Butcher, we decided we would find such a kill switch undesirable for several reasons:


  1. To our knowledge, no currently existing kill switches can be induced by readily available substances that are at once non-toxic and not abundant in the human gut in the first place. Many known probiotic kill switches rely on induction by antibiotics – the use of such a mechanism would go against one of the initial premises of our project, which is to reduce our reliance on antibiotics.
  2. The ProQuorum probiotic is unlikely to be capable of establishing long term colonization of the gut at significant levels. In our conversation with Zachary Abbott from Zbiotics we learned that there is currently little evidence to suggest L. reuteri can become a dominant species in the gut; coupled to the relatively decreased fitness of our strain and its presumed safety, a pharmacologically-inducible kill switch is most likely unnecessary.

Further detail can be found on our Human Practices page.

Finally, the last issue we feel needs to be addressed is the problem of lateral gene transfer of our genetic constructs. Seemingly the best solution for our system is to use genomic integration, as that reduces the likelihood of uptake by other microorganisms compared to a plasmid. There are several biological advantages to this approach that override the need to use a toxin-antitoxin system:

  1. Genomic integration increases the stability of our parts, which is essential if the probiotic is to be therapeutically effective. At the same time, this ensures the need for a selectable marker is avoided.
  2. It reduces the heterogeneity within the population owing to the variability of plasmid copy number per cell, thus allowing us more precise fine tuning of the system according to the parameters optimised in the dry lab.
  3. A priori, the acquisition of any of our parts by other microorganisms is unlikely to increase their inclusive fitness and instead only introduce additional cellular stress, thus making the retention of our genes improbable.

References

Reference
Jacobsen, C.N. et al., Screening of Probiotic Activities of Forty-Seven Strains of Lactobacillus spp. by In Vitro Techniques and Evaluation of the Colonization Ability of Five Selected Strains in Humans, Applied and Environmental Microbiology, Nov. 1999, p. 4949–4956
Mu Qinghui, Tavella Vincent J., Luo Xin M., Role of Lactobacillus reuteri in Human Health and Diseases, Frontiers in Microbiology, 19 Apr 2018
Mayer M. J. et al., Structure-Based Modification of a Clostridium difficile-Targeting Endolysin Affects Activity and Host Range, Journal of Microbiology, Oct 2011