Team:St Andrews/Poster
Shinescreen - A Reef-safe Probiotic Sunscreen
Presented by Team St Andrews 2020
Overview
The St Andrews team is working on a two-phase project called Shinescreen. Our
goals are to develop a coral-friendly alternative to conventional sunscreens and raise
awareness of the vital roles played by corals as ecosystem engineers. Our novel probiotic
would use live bacteria actively producing the mycosporine-like amino acid (MAA), shinorine,
synthesised naturally by diverse marine microorganisms for UV protection. Due to the ongoing
restrictions in the UK, we have been unable to access a lab this summer, and instead are
executing the first virtual phase this year. Namely, our focus has been the design,
modelling and human practices of the project to inform the second phase. The concerns and
opinions of stakeholders have fed back into our project design and proposed implementation
in phase two.
Insight
There is a growing urgency to protect reef ecosystems from bleaching caused by
the chemicals such as oxybenzone from conventional sunscreens and climate change. Corals
support 25% of all marine species and constant exposure to these stressors have a cumulative
effect leading to their decline, which has a devastating effect on the marine ecosystem.
Therefore, we were motivated to create a sustainable probiotic sunscreen, and using synthetic biology, we designed Shinescreen, an approach to
Unlike conventional sunscreens, we designed the in-silico project as a probiotic that is topically applied to the skin, containing bacteria with a gene cluster that will overexpress shinorine and provide users protection from UV rays. Additionally, the gene circuit contains a novel kill-switch that will prevent integration of the bacteria into the skin microbiome or the environment. Since our product does not contain the toxic chemicals that have been shown to disrupt hormone functions and increase skin irritation in individuals with skin conditions such as acne, it will be a potential reef-friendly alternative that is both environmentally and user-safe. With Shinescreen, we aim to provide a non-toxic sunscreen alternative that would allow for the protection of the skin from the UV rays, with several advantages in protection and cosmetics, while also protecting marine organisms.
Therefore, we were motivated to create a sustainable probiotic sunscreen, and using synthetic biology, we designed Shinescreen, an approach to
- Save coral reefs and marine organisms
- Contribute to Sustainable Development Goal 14: to conserve and sustainably use the oceans and their resources
- Combat UV-induced skin cancer with a more responsive and longer-lasting sunscreen. Nearly 90% of cutaneous melanomas are due to phototoxicity
- Consider a proof-of-concept skin product using synthetic biology to deliver key compounds to the epidermis
Unlike conventional sunscreens, we designed the in-silico project as a probiotic that is topically applied to the skin, containing bacteria with a gene cluster that will overexpress shinorine and provide users protection from UV rays. Additionally, the gene circuit contains a novel kill-switch that will prevent integration of the bacteria into the skin microbiome or the environment. Since our product does not contain the toxic chemicals that have been shown to disrupt hormone functions and increase skin irritation in individuals with skin conditions such as acne, it will be a potential reef-friendly alternative that is both environmentally and user-safe. With Shinescreen, we aim to provide a non-toxic sunscreen alternative that would allow for the protection of the skin from the UV rays, with several advantages in protection and cosmetics, while also protecting marine organisms.
Design
Our gene circuit design consists of two divisions: the shinorine-producing
pathway and the kill-switch. These two components are further split across two plasmids
pSB3B1-DOPH and pSB3E1-AN to make up a novel two-plasmid dependency design, where both
constructs must be contained by our E.coli chassis to express the UV-absorbing
compound shinorine and survive.
Figure 1. The four enzymes required to produce the UV(A)-absorbing compound, shinorine. Like melanin, shinorine is an intracellular UV-protectant. When accumulated, the bacterial sunscreen layer could protect the underlying epidermis from UV irradiation. Furthermore, shinorine is a free-radical scavenger and antioxidant, combatting photo-ageing.
Figure 2. Light-mediated ccdA expression and glucose-regulated ccdB output control bacteriostasis and the expression of an mf-Lon protease.
The kill-switch depends on two inputs: green light, a proxy for UV, and glucose concentration. The light component controls production of an antitoxin, ccdA, endogenous to E. coli, to combat the toxin ccdB, which is produced by a glucose-regulated component at low concentrations of the sugar (Figure 2). The light system acts as a two-component system responding to light and regulating gene expression. The light sensor will be deactivated in the absence of light over extended durations.
When ccdA ≥ ccdB, the repressive complex formed would bind an associated operator and prevent expression of mf-Lon protease which would otherwise result in a feedback cascade resulting in a second kill-switch response causing plasmid DNA shredding (Figure 3). Control of the expression of the endonuclease, R.CviJI, is regulated by the lac repressor which allows efficient prevention of plasmid degradation in favourable conditions (Table 1). In the absence of glucose and green light however, mf-Lon will target lacI by an associated degradation tag such that R.CviJI can be expressed to allow nuclease activity.
Figure 3. Second kill-switch mechanism to further reduce plasmid escape into the surrounding environment and ensure simultaneous bacteriostasis and DNA degradation in our chassis organism.
Table 1. Plausible scenarios of kill-switch activation/deactivation and their outcomes.
Figure 1. The four enzymes required to produce the UV(A)-absorbing compound, shinorine. Like melanin, shinorine is an intracellular UV-protectant. When accumulated, the bacterial sunscreen layer could protect the underlying epidermis from UV irradiation. Furthermore, shinorine is a free-radical scavenger and antioxidant, combatting photo-ageing.
Figure 2. Light-mediated ccdA expression and glucose-regulated ccdB output control bacteriostasis and the expression of an mf-Lon protease.
The kill-switch depends on two inputs: green light, a proxy for UV, and glucose concentration. The light component controls production of an antitoxin, ccdA, endogenous to E. coli, to combat the toxin ccdB, which is produced by a glucose-regulated component at low concentrations of the sugar (Figure 2). The light system acts as a two-component system responding to light and regulating gene expression. The light sensor will be deactivated in the absence of light over extended durations.
When ccdA ≥ ccdB, the repressive complex formed would bind an associated operator and prevent expression of mf-Lon protease which would otherwise result in a feedback cascade resulting in a second kill-switch response causing plasmid DNA shredding (Figure 3). Control of the expression of the endonuclease, R.CviJI, is regulated by the lac repressor which allows efficient prevention of plasmid degradation in favourable conditions (Table 1). In the absence of glucose and green light however, mf-Lon will target lacI by an associated degradation tag such that R.CviJI can be expressed to allow nuclease activity.
Figure 3. Second kill-switch mechanism to further reduce plasmid escape into the surrounding environment and ensure simultaneous bacteriostasis and DNA degradation in our chassis organism.
Table 1. Plausible scenarios of kill-switch activation/deactivation and their outcomes.
Molecular Modelling
Without lab access, in silico modelling was the main instrument in our synbio toolbox. Molecular dynamics were examined to inform gene circuit design and confirm expected behaviour. We tested the shinorine-producing gene circuit (which we call the Shinogen) and the genes of the kill-switch (the Thanogen), that takes inputs of glucose and green light. Systems of ODEs were built for the thanogen and shinogen independently. Then, both deterministic and stochastic models were produced using kinetic parameters derived from the literature and approximated where unavailable.
For the shinogen, we identified
Figure 1. Behaviour of thanogen to environmental inputs. (green line represents antitoxin, red line represents toxin, orange line represents the complex formed when the toxin is in excess).
For the thanogen, we confirmed
Figure 2. Illustration of the process involved in the evolutionary mutation modelling undertaken
We then outputted the function of the kill-switch in a biosafety model testing the evolutionary stability of the kill-switch over a certain number of generations of bacteria. We identified an epistatic positive correlation between mutation rate and loss of function of the kill-switch which occurred more quickly in smaller populations and that selection accelerates loss of function. We further confirmed which parts of the kill-switch were most sensitive to mutation.
For the shinogen, we identified
- The stable steady state concentrations of the four biosynthetic enzymes for shinorine production
- The rate-controlling enzyme(s) in the pathway and the dependence on metabolite influx from the pentose phosphate pathway
Figure 1. Behaviour of thanogen to environmental inputs. (green line represents antitoxin, red line represents toxin, orange line represents the complex formed when the toxin is in excess).
For the thanogen, we confirmed
- That the gene circuit could produce bistability: the bacterium was either committed to life or death
- All parts of the kill-switch were necessary for its intended behaviour
Figure 2. Illustration of the process involved in the evolutionary mutation modelling undertaken
We then outputted the function of the kill-switch in a biosafety model testing the evolutionary stability of the kill-switch over a certain number of generations of bacteria. We identified an epistatic positive correlation between mutation rate and loss of function of the kill-switch which occurred more quickly in smaller populations and that selection accelerates loss of function. We further confirmed which parts of the kill-switch were most sensitive to mutation.
Multiscale Modelling
To understand how our hypothetical ‘Shinescreen’ might perform under realistic application, we also devised a model which takes the molecular modelling applies this to the large-scale behaviour of the bacterial population.
The aim was to predict the threshold and duration of protection our sunscreen may provide. This depends on the density of shinorine across each area of skin surface. Since the shinorine is expressed and accumulated intracellularly, it is enough to understand how E.coli spreads once it is applied to the skin surface since such growth allows for the in vivo synthesis of shinorine.
The model considers E.coli growing in a spatially homogeneous liquid culture bound to the 2-dimensional surface of the human skin. It constitutes a pair of partial differential equations (PDEs) describing the time evolution of just the bacteria and glucose. It is an idealised description of the real biological processes. The challenge was to devise a model which was ‘coarse-grained’ enough for practical predictions, while still containing the essential biology:
Statistical analysis of expected behaviour of the thanogen motivated the development of a completely unique mathematical term to describe bacterial death, with parameter values inferred from the molecular modelling.
Figure 1. Earlier MATLAB simulation showing the spatiotemporal change in (top left to right) the gel, glucose concentration, bacterial density, shinorine concentration, the native microbiome and UV flux.
The aim was to predict the threshold and duration of protection our sunscreen may provide. This depends on the density of shinorine across each area of skin surface. Since the shinorine is expressed and accumulated intracellularly, it is enough to understand how E.coli spreads once it is applied to the skin surface since such growth allows for the in vivo synthesis of shinorine.
The model considers E.coli growing in a spatially homogeneous liquid culture bound to the 2-dimensional surface of the human skin. It constitutes a pair of partial differential equations (PDEs) describing the time evolution of just the bacteria and glucose. It is an idealised description of the real biological processes. The challenge was to devise a model which was ‘coarse-grained’ enough for practical predictions, while still containing the essential biology:
- Both the bacteria and glucose diffuse within the gel or lotion formulation
- The bacteria feed on the glucose at a rate which depends on the availability of glucose
- The skin microbiome partially consumes glucose at some set of rates, which can be informed from the literature
- The kill-switch causes degradation of the bacteria. Probabilities of cell death caused by the kill-switch when exposed to different concentrations of glucose and light are translated into proportions of population-wide cell death
Statistical analysis of expected behaviour of the thanogen motivated the development of a completely unique mathematical term to describe bacterial death, with parameter values inferred from the molecular modelling.
Simulations
MATLAB software is under development to simulate and visualise the dynamics of our model. Figure 1 below shows a simulation generated with the same software on an earlier version of the model. We hope to simulate our adapted model to predict the protective threshold our sunscreen could offer and check the viability of the kill-switch at the population scale.
Figure 1. Earlier MATLAB simulation showing the spatiotemporal change in (top left to right) the gel, glucose concentration, bacterial density, shinorine concentration, the native microbiome and UV flux.
Interviews
Through the course of our project, we made sure to meet with a variety of stakeholders, experts, and the public. We wanted to ensure our scientific developments closely aligned with the needs of experts, latest science, and trends in the sunscreen market.
Coral experts
We met with coral researchers such as Dr Allison from the University of St Andrews as well as Drs Hennige and Tagliati from the University of Edinburgh. From these conversations, we concluded our sunscreen could have significant positive effects on marine ecosystem health, but this also influenced our extensive kill-switch and mutation modelling. After meeting Dr Allison, we grew cognisant of the need of incorporating mechanisms to prevent our chassis bacterium from impacting the coral microbiome. Considering the issues raised from stakeholders, St Andrews team developed micro and macro models, predicting the behaviour and the requirements of our genetically engineered chassis organism, while considering the needs of potential consumers. We completely redesigned our kill switch in response to stakeholder concerns, deciding on a two-plasmid system that would minimise horizontal gene transfer. We also incorporated an endonuclease into the kill switch, which adds a degree of redundancy into the kill switch, as well as reducing the chance of horizontal gene transfer.Biosafety
We talked to Professor Firbank from the University of Leeds, an agroecologist and member of European Food Safety Authority GMO panel. He summarised the very strict regulations of GMO products and we were able to more fully explore biosafety, gaining insights into the practicality of creating such a product in accordance with safety regulations. This conversation greatly influenced the dry-lab experimental design of the project, mainly in the context of biosafety and efficiency.Medical
We also met with a professor in dermatology who was very interested in the photosensitivity patient treatment angle. She saw permanent integration into the skin microbiome as a possible benefit in the management of light sensitivity disorders such as polymorphic light eruption and porphyria, which can be extremely debilitating. She has asked to remain anonymous on our wiki and poster.Industry
Furthermore, after having gained more of a scientific perspective from these stakeholder meetings, we wanted to gain some insights into the industry and the creation of a sustainable sunscreen. We met with Louise Laing, founder of People4Ocean, and Dr Cornelia Schürch, Head of Development and Compliance at Mibelle biochemistry. We also met with investors such as Nagase R&D to gain sponsorship and possibly form a future partnership.Overall, our project was greatly influenced by our conversations, and it allowed us to in turn share our perspective with the stakeholders, ensuring a feedback loop between our stakeholders and our project.
Outreach
Twitter Sentiment Analysis
As part of our outreach campaign, we performed a Twitter sentiment analysis with the primary goals of determining prevailing opinions on genetically modified organisms (GMOs) and finding out the emotional valences associated with GMOs. We collected 9576 tweets with a key word #GMO and used 3 different algorithms in ‘R’ which informed us about the emotional valence, sentiment score and content of the tweets. The key findings were:- Prevalence of neutral to slightly positive sentiments towards #GMO
- ‘Food’, ‘crop’, ‘risk’ and ‘new’ were among the most frequently found words
Figure 1. Distribution of the frequency of the tweets in relation to their sentiment score, where <0 is negative, 0 is neutral, > 0 is positive sentiment.
We implemented the results in our market research.
Survey
In order to understand what people think of a potential probiotic sunscreen, we conducted a multinational, online survey. The responses to the question: ‘If there were a probiotic sunscreen, such as "Shinescreen", on the market at a similar price to your current favourite, how likely would you be to buy it?’, were mostly positive:
Moreover, we used the survey to assess the level of environmental awareness among the participants and were happy to discover that over 95% of the participants realised the adverse effect of human activity on the environment.
SynthBio Forum
At the end of August, the team hosted the Synthetic Biology Forum, hoping to inform the public on synthetic biology and related topics. The online event was live streamed in several social media websites, reaching 2300 viewers. The panel comprised experts from an impressive diversity of fields including gene therapy and agriculture with a synthbio focus. Members of the public were invited to participate and ask questions on “will GMOs shape the future of humanity?” and discuss this with panellists. The forum provided us with the opportunity to discuss GMOs and synthetic biology with attendees while mutually attaining their opinions on GMOs for our human practices and market research. The feedback from the polls we carried out at the end of the event showed that we lessened concerns surrounding synthetic biology and raised awareness of its raw potential.Antibiotics Under Our Feet
As part of our ongoing education and public engagement efforts, we furthered the Antibiotics Under our Feet project started by the St Andrews iGEM 2018 and 2019 teams. The aim of this project is to raise science capital in children aged 9-11. The project explores extraction of novel compounds from soil microbes to treat multidrug-resistant organisms.Advancing previous teams’ work, we made lab-based educational videos for primary school students, designed to educate students about carrying out a soil sample analysis. Raising science capital is key to more mainstream use of synthetic biology solutions to real world problems.
CoralWare App
The development of CoralWare app was a result of our research on the currently existing apps providing information about cosmetic ingredients. We noticed that they were lacking in data on the environmental impact of the chemicals and their sources of information. CoralWare was designed to allow its users to take a photo of the ingredients list of a sunscreen, analyse the photo by optical character recognition and provide the required health and environmental data. The idea of the app is to make people more aware of the recent research, regarding the effects of cosmetic ingredients, and allow them to make their own decisions based on clear, unbiased information.
Enterprise
Early in our project, we realised that we had a unique value proposition and started thinking of how Shinescreen could be bigger than just an iGEM project. We focused heavily on gaining insights into the market and pursuing business opportunities on top of the modelling and human practices work.
We conducted a series of interviews with sunscreen companies to explore our business responsibilities and opportunities. With promising results, we conducted direct market research through surveying of several target markets and consumer groups, running Twitter sentiment analyses and hosting a public discussion forum on GMOs. We then performed further research to explore business opportunities and markets to expand into, concluding that there are unmet consumer demands which Shinescreen could satisfy.
SWOT analysis
We also consulted with several entrepreneurial experts, patent lawyers, and the St Andrews Technology Transfer Centre to explore Intellectual Property protection.
Lastly, we pitched our business proposition to several sponsors and investors:
- We won the St Andrews enterprise pitch event, receiving £980 to fund our entrepreneurial ventures.
- Very recently, we approached Nagase R&D center, met with them, and they are have decided to donate €1000 towards our project.
Moreover, we have now launched an experiment.com fundraising platform to raise funds for phase II of our project. In parallel, we have begun drafting a business development plan and setting up a private limited company.
Next Steps
Restrictions imposed by the national lockdown made lab access impossible.
Consequently, we implemented a two-stage project. Our next steps shall be the improvement of
our modelling while implementing the genetic circuits we designed in vivo to test the
viability of Shinescreen as a UV protectant. Environmental modelling and refined tests of
sunscreen toxicity on corals shall inform its viability as well as the environmental impact
compared to conventional sunscreens. Additionally, while we focused on shinorine expression,
we plan to explore full coverage of the UV spectrum by expanding our gene circuit to
incorporate a wider variety of optimised for alternative microsporine-like amino acids
(MAAs) synthesis.
As a commercially available sunscreen, our engineered bacterium could be maintained within a glucose starter culture. First the culture of our genetically engineered Nissle E. coli would be mass incubated, thereby accumulating shinorine (and other potential MAAs). The culture could then be packaged and viable for application of the live bacterial culture similar to buying Actimel. Once the starter medium is depleted or is no longer exposed to the sun, the kill-switch will eliminate residual bacteria, preventing escape and establishment in the environment. This would permit control of the viability and thus length of protection conferred by Shinescreen.
What shape will Shinescreen take ex silico?
We conceive of a biointelligent and multifunctional system incorporating our sensitive kill-switch and a gene circuit building on our shinorine-producing part expressing a family of UV protectants produced in real time. Shinescreen could offer prolonged protection against UV toxicity – without cost to the environment!As a commercially available sunscreen, our engineered bacterium could be maintained within a glucose starter culture. First the culture of our genetically engineered Nissle E. coli would be mass incubated, thereby accumulating shinorine (and other potential MAAs). The culture could then be packaged and viable for application of the live bacterial culture similar to buying Actimel. Once the starter medium is depleted or is no longer exposed to the sun, the kill-switch will eliminate residual bacteria, preventing escape and establishment in the environment. This would permit control of the viability and thus length of protection conferred by Shinescreen.
What lies ahead?
We plan to create the next generation of sustainable reef-safe sun-protection, all using naturally derived genes.
References and Acknowledgements
Acknowledgements
PIDr. V Anne Smith (Senior Lecturer at the School of Biology)
Dr. Chris Hooley (Senior Lecturer at the School of Physics and Astronomy)
Dr. Paula Miles (Senior Lecturer at the School of Psychology and Neuroscience)
Student Advisors
James Hammond
Eleonora Shantsila
Partners
Hainan China iGEM 2020
Manchester iGEM 2020
Sponsors
References
Ay, Ahmet, and David N. Arnosti. 2011. “Mathematical Modeling of Gene Expression: A Guide for the Perplexed Biologist.” Critical Reviews in Biochemistry and Molecular Biology 46 (2): 137–51. https://doi.org/10.3109/10409238.2011.556597.Balskus, E. P., Walsh, C.T. 2010. The genetic and molecular basis for sunscreen biosynthesis in cyanobacteria. Science. 329 (5999), pp1653-1656.
Chandran, D., W. B. Copeland, S. C. Sleight, and H. M. Sauro. 2008. “Mathematical Modeling and Synthetic Biology.” Drug Discovery Today: Disease Models 5 (4): 299–309. https://doi.org/10.1016/j.ddmod.2009.07.002.
Danovaro, R et al. 2008. “Sunscreens Cause Coral Bleaching by Promoting Viral Infections”. Environmental Health Perspectives 116:4. https://doi.org/10.1289/ehp.10966
Hedges, Whitehead and. 2005. “Photodegradation and photosensitization of mycosporine-like amino acids.” Journal of Photochemistry and Photobiology B: Biology 115-121.
Wijgerde, T. et al. (2020) ‘Adding insult to injury: Effects of chronic oxybenzone exposure and elevated temperature on two reef-building corals’, Science of The Total Environment. The Authors, 733, p. 139030. doi: 10.1016/j.scitotenv.2020.139030.
Wright, O., Stan, G-Y., Ellis, T. 2013. Building-in biosafety for synthetic biology. Microbiology. 159, p1221-1235.