Safety in Laboratories

Conducting research in the lab means working with substances, systems and organisms that are not yet fully understood. Additionally, in life sciences, we work with components and procedures that are meant to visualize and alter the function of DNA, proteins and cells. Therefore, it is essential to inform oneself beforehand about the properties and dangers of the substances used in the lab and the project.

Personal Labsafety

Before starting to work in our lab, we had an extensive safety instruction by our PI Dr. Nicole Gensch who is also the safety officer of our institute. For the chemical lab, there were different aspects to consider in terms of safety, so we received a separate instruction for working there. We always made sure to wear our lab coats and gloves when handling dangerous material and installed a clear separation between areas and surfaces that should be treated with and without gloves respectively.

Safe handling of bacteriophages

As we decided to establish phage display in our lab, it was important to avoid any hazards or cross-contamination coming from the phages. Therefore, we designated a part of our lab as phage area. This included pipets, materials and specialized waste containers. Phages are very sturdy, and can even withstand autoclaving. Therefore, we consulted Dr. Hanna Wagner who is an expert on working with phages on how to eliminate phage containing waste. She suggested to eliminate phages with bleach before autoclaving and to regularly clean our laboratory with bleach. However, since bleach contains chloride we were very cautious to not leak it in the environment while handling it.

In vitro work with PSMα3

D- and L-PSMα3 were synthesized using solid phase peptide synthesis to avoid the safety risks that would come along with recombinant expression in genetically modified organisms. Transforming E. coli with the gene for PSMα3 would imply to equip a labsafe strain with the means to produce a toxin from the more pathogenic Staphylococcus aureus which is classified as a biosafety 2 level organism. As this did not represent an option for us, we focused our project on the PSMα3, which still required additional safety precautions. Prof. Friedrich Götz, an expert in the PSMα3 field, was consulted on how to handle PSMα3, and more specifically, how to deactivate it. His advice to chemically inactivate PSMα3 with 0.5M NaOH was implemented in our working routine. However, since PSMα3 has never before been synthesized in D-form we were reluctant to simply apply the same risk assessment as for the L-form of PSMα3. Therefore we did an additional rational risk assessment by researching the literature about D-proteins and the mechanism of PSMα3 toxicity in general.

Risk assessment of D-PSMα3

The safety profile of D-proteins in general is well known. D-proteins are less immunogenic and show less interactions with biological systems¹. Therefore, we did not expect any hazards coming from the D-amino acids themselves.

The toxicity of L-PSMα3 is well-described². The lytic properties of this toxin depend on its amphipathic nature. L-PSMα3 can insert itself into lipid bilayers like cell membranes and form pores, acting similar to a detergent like Triton-X-100. We compared the toxicity of L-PSMα3 to lysis buffers based on Triton X-100 and observed that PSMα3 at a concentration of 2 uM had approximately 20% of the cytotoxic effect. At 8 uM and higher concentrations the toxicity of PSMα3 didn’t exceed 50% of the cytotoxicity effect of Triton -X-100.

Because the mechanism of PSMα3-mediated cell lysis is based neither on specific protein-protein interactions nor the structural recognition of a binding motiv, we assumed that D-PSMα3 would carry the same properties as L-PSMα3 in regard to its cytotoxicity. We verified this assumption in experiments, comparing L- and D-PSMα3.

The second function of PSMα3 relies on the activation of the formyl peptide receptor 2 to stimulate neutrophils to secrete proinflammatory cytokines thereby causing inflammation. Because the FPR2 recognizes the structure of L-PSMα3 in a specific way it is unlikely that D-PSMα3 which has a different structure would bind to it as well.

Fig. 1: PSMα3 causes inflammation by activation of FPR2 and lyses cells by pore formation.

Our risk assessment pointed towards the assumption that D-PSMα3 is comparable to L-PSMα3 in its toxicity or probably even less toxic. Nonetheless, we handled PSMα3 with the utmost care and adhered to special safety guidelines. For instance, while handling the toxin in dried lyophilized form after synthesis, respirators were worn and work was carried out under a fume hood to reduce the risk of exposure to L-PSMα3.

Assessing the possibility of safe in vivo production of PSMα3

Solid phase peptide synthesis (SPPS) is a tedious technique when attempting to synthesize large proteins, especially when posttranslational modifications are present such as an N-terminal formylation. In addition, the chemical reactions that are required for these modifications can significantly reduce the protein yield. Therefore it would greatly facilitate research on protein based toxins if researchers would be able to express toxins in vivo, also from a financial point of view. However, as described before, the in vivo production of toxins comes along with several problems such as for example a higher biosafety level and a generally elevated risk. Furthermore, toxins with lytic mechanisms, like PSMα3, are difficult to express in significant amounts in biological systems because they would lyse the E. coli organelles.
Therefore we thought of an approach based on synthetic biology, that could improve the safety of projects focusing on toxins and pathogens. To enable future iGEM-Teams to safely express toxic proteins in E. coli and produce them in sufficient yields, we implemented an expression system based on split inteins. Split inteins are naturally occuring proteins, which, when binding with their respective counterpart, reconstitute to a full intein that autocatalytically splices out of the precursor proteins and attaches the flanking regions together by forming a new peptide bond³.

Fig. 2: Intein-based expression system for unsafe products.

By splitting the toxic protein in two safe and non-toxic fragments and then fusing them to the N- and C-Intein of gp41.1 (BBa_K1362160 and BBa_K1362161) one can now efficiently express both constructs in separate cultures (Figure 2). After the expression the cells of both cultures are lysed and mixed so that the desired protein can now be assembled under safe in vitro conditions.

To demonstrate the feasibility of this system we optimized BBa_K1362160 and BBa_K1362161 by adding maltose-binding protein (MBP) as extein to the gp41.1 N-Intein (BBa_K3009016) and thioredoxin (TRX) as extein to the gp41.1 C-Intein (BBa_K3009015). With our improved versions we could now demonstrate the in vitro protein splicing.

We chemically synthesized the two fragments of PSMα3 using SPPS. This enabled us to perform an LDH cytotoxicity assay in order to assess, whether the fragments exhibit reduced cytotoxicity compared to our chemically synthesized full-length PSMα3 toxin (Figure 3).

Fig. 3: Reduced cytotoxicity of PSMα3 fragments compared to the complete toxin

Three biological replicas showed a significant reduction in cytotoxicity of the two halves of PSMα3, hereby termed fragments A and B, compared to the natural L-PSMα3. This indicates the improvement of safety when splitting toxins into fragments for their expression. In conclusion, we demonstrated the feasibility and the safety of this system and demonstrated this systems potential to further improve the safety in iGEM and synthetic biology in general.