Team:Exeter/Safety

One Shot® TOP10 E.coli are being used for all cloning of genes and strain E.coli BL21(DE3) to express our polysaccharides (WHO 1). The glycosyltranferases that were utilised to construct the operon are all from different E. coli strains. We had these genes synthesised and the enzymes encoded by these genes are not pathogenic. The following enzymes used in the laboratory were WbnJ, WbnK, Wbbc, WfcA and WclY and the amino acid sequenced were from E. coli strains O86:H2 (WHO 1), O7:K1 (WHO 2), O141 (WHO 3) O117 (WHO3) respectively.

We have synthesised a codon-optimised hyaluronan synthase gene from Streptococcus pyogenes strain S43. Hyaluronan is produced inside the human body and is not toxic.

We have synthesised a codon-optimised cyclodextran glycosyltransferase from Bacillis subtilis BPED101 (WHO 1). Cyclodextrin is not toxic.

The use of 3A assembly to construct our operons required the use of several different antibiotics including chloramphenicol, ampicillin, tetracycline and kanamycin:

• Chloramphenicol is a broad spectrum bacteriostatic antibiotic effective against both Gram-positive and Gram-negative bacteria as it inhibits protein synthesis. Its most severe side effects being bone marrow suppression, although usually reversible, and aplastic anemia which is generally fatal.
• Ampicillin is a broad spectrum bacteriocidal antibiotic targeting bacterial cell wall production and so is much less toxic to eukaryotes although can cause severe allergic reactions.
• Tetracycline is another broad spectrum bacteriostatic antibiotic that inhibits protein synthesis in both Gram-positive and negative bacteria. Its side effects are much more diverse including teeth discolouration, skin photosensitivity, drug-induced hepatitis, anaphylactic shock etc. The more severe and diverse side effects meant we avoided use of tetracycline where possible.
• Kanamycin inhibits protein synthesis also but at a different point. It is a broad spectrum bacteriostatic antibiotic that works against Gram-positive and negative bacteria and is used very commonly in laboratories against E.coli. Its serious side effects include kidney toxicity, loss of hearing and allergic reactions.

All antibiotics were handled with care and used in the lowest dosages required.

For the confirmation of our gene constructs, we digested our vectors and ran them on a gel in electrophoresis, requiring the use of ethidium bromide. Ethidium bromide is a mutagen that intercalates with DNA. It was handled in a designated fume cupboard with designated gloves and pipettes and all waste was emptied into special bins.

The DNA bands were visualised with UV light inside a dark room that when in use, had a light on the outside to say “in use”. UV light was only switched on when the drawer containing the gel was closed and visualised with a camera rather than looking at the UV image directly, as UV light is potentially carcinogen.

Chemicals which were toxic or irritants were identified and reviewed rigorously under the Control of Substances Hazardous to Health (COSHH) assessment prior to commencement of the project. These were subsequently handled by the more experienced staff members for diluting for us to use to minimise risk to our team members and other lab workers due to the correct safety practices being followed.

Our team followed all standard safety protocols including wearing lab coats and protective gloves. All waste was separated and disposed of by the correct means.

The unintentional release of our genetically modified E.coli into the environment is a public safety concern for all synthetic biology projects. One Shot® TOP10 E.coli and E.coli BL21(DE3) are both common laboratory strains that would not replicate in the environment, nor infect a host. The plasmids we have used containing our glycosyltransferase genes are non-mobilisable, so there is no threat to the public.

The generation of novel, extracellular polysaccharides should not provide any fitness advantage to the bacterium and an increase in pathogenicity is unlikely. The polysaccharides we are producing, ONLY when attached to lipid A, could represent a chemical (but NOT a biological) hazard. This is not part of the Wzy-dependent mechanism and involves a separate ligation step involving the enzyme WaaL. Toxicity is due entirely to lipid A and the polysaccharide component is not responsible for the toxicity. Therefore we believe that synthetic polysaccharides will not pose a toxic hazard because they mimic the polysaccharide of LPS. Although toxicity is not associated with the polysaccharide unit, we still avoided producing a polysaccharide which was identical to those on known toxic LPS, based on an extensive literature review conducted far before any laboratory work commenced. In the future, we expect to generate WaaL mutant E.coli defective in the production of lipopolysaccharides.

These issues were discussed in depth with our safety advisors and at the human practices panel.

Our system is confined to the laboratory and in the future would still be confined to laboratory settings, growing the chosen organism to hold our system in fermenters. These are all designed with safety in mind to prevent accidental release and only handled by trained operatives so the release of our organism into the environment is prevented. As discussed in public safety, there are safety measures in place in case of release.

The individual BioBricks submitted for the separate glycosyltranferases, which make the individual sugar units, are mostly characterised enzymes and are non-toxic individually as they occur naturally in different E.coli strains. The only component not directly involved in our system that may raise safety issues is the ligating enzyme WaaL responsible for attaching polymerised polysaccharide repeat units from the Wzy-dependent system onto lipid A moieties. This summer, attachment to lipid A may occur when producing our novel polysaccharide and may represent a chemical hazard. However, this does not represent a biological hazard because polysaccharides are known to be secreted and are therefore not associated with the GMO. As previously mentioned, lipid A causes the entire toxicity associated with LPS. Therefore the production of polysaccharides using synthetic biology would not pose a toxic hazard just because they are identical to the polysaccharide component on LPS. In the future, we aim to generate WaaL mutants and as a result the toxicity issue is largely avoided because it is the lipids that evoke an immune response in humans and animals. These mutants would then not represent a biological or a chemical safety issue.

We linked a signal peptide to the BioBrick for SacB which is toxic to E.coli so that it would be exported from the cell and could produce levansurose extracellularly. Levansucrose is not toxic to the public but on accumulating inside Gram-negative bacteria such as E.coli, the chassis we used in this project, causes cell lysis. On submitting the OmpA-SacB to the registry we documented this issue we had overcome.

In linking our enzymes to signal peptides we were able to demonstrate that the polysaccharide levansucrose, and the non-toxic cyclodextran, could be produced extracellularly. Other teams can use the OmpA BioBrick we submitted and produce substances extracellularly for better isolation of their product or avoid toxicity to the cell. Producing our novel polysaccharide meant that more background research needed to be done to ensure characterised enzymes were used from known strains and related compounds researched to ensure it is as safe as possible. We would suggest to other teams making novel products to consider safety at every step of the construction process to ensure that they weren’t making a toxic compound.

If a new safety issue arises with any of the BioBrick parts produced in our project, we will document any new safety concerns and update the Registry with immediate effect.

At the University of Exeter, projects involving the use of genetically engineered organisms (GMOs) must undergo a strict risk assessment by the local GM committee based in Biosciences and reviewed in accordance with the college’s code of practice (Biosciences’ Code of Safety Practice) in conjunction with the Scientific Advisory Committee on Genetic Modified Organisms (Contained Use), abbreviated to SACGM (CU), which includes Genetic Modified Organisms (Contained Use) Regulations 2000. Specifically, most of the risk assessment that was applicable to our project was soughted from Part 2: Risk assessment of genetically modified microorganisms (other than those associated with plants) from The SACGM Compendium of Guidance. Considerations in the risk assessment for our project included information on the host species and strain that we will genetically engineer (E.coli MG-1655), vectors to be used (mainly pSB1C3/pSB1A3), origin of glycosyltransferase genes and system (as well as their function and potential harmful properties), and the survivability of the genetically modified organism in the environment if an accidental release event were to occur.

Before we could undertake our project, we had to complete the following forms:

• Genetic Modification Registration Form which is to be completed by workers intending to use GMOs within the Biosciences.
• Declaration of Compliance with the Biosciences’ Code of Practice 2011-2012 which included safety procedures covered by our lab induction.

As a result, all local regulations were adhered too and all people involved with the project were trained for the handling of GMOs. Overall, the GM committee were pleased with project’s outline and aims, and the chair of the committee signed the GM risk assessment form on 18th June and approved that our project was deemed safe.

The Wzy-dependent system is fairly ubiquitous in the bacterial kingdom as it is one of the three main pathways used to produce polysaccharides. However, the Wzy-dependent pathway is mostly found in Gram-negative bacteria because of the close connection to LPS production and the involvement of the periplasmic space. This easily lended itself to use E.coli as our chassis since a wide range of safe strains of K-12 and BL21 (we used BL21(DE3)) are available despite public perception of E.coli as “dangerous”.

Lipopolysaccharides (LPS) and their toxic properties (broadly known as endotoxins) are present on the outer membrane of Gram-negative bacteria and are well known for eliciting high immune responses. Lipopolysaccharides are defined as a polysaccharide covalently bound to a lipid A core, and occurs when the polysaccharide has been polymerised by the Wzy-dependent system followed by attachment to lipid A by the ligase enzyme WaaL, producing a nascent LPS. Since the production of LPS is closely associated to the Wzy-dependent system, we intend to make the system safer through biosafety engineering. Since the transport of LPS are extremely complicated and involves 7 genes of the lpt family, in addition to over 10 genes that code for proteins involved in modifying the LPS, we envisage that knocking-out WaaL in the gmhD operon is an easier approach to prevent LPS assembly. There has been conflicted evidence that WaaL mutants maybe lethal to E.coli. Therefore in the future, we hope that we can generate WaaL mutants that are defective in LPS assembly and use these mutants in high-throughput manufacture of bespoke polysaccharides.