Engineering Our Path to Innovation
Dive into the engineering journey of the Hydro Guardian and discover how we've applied the iGEM Engineering Cycle to develop groundbreaking solutions in synthetic biology.
Dive into the engineering journey of the Hydro Guardian and discover how we've applied the iGEM Engineering Cycle to develop groundbreaking solutions in synthetic biology.
The Hydro Guardian Biosensor integrates dual sensing of antibiotic and metal residues by combining prokaryotic and eukaryotic components in two subunits, as described in the following sections. To engineer the Hydro Guardian, we employed the synthetic biology engineering cycle, encompassing the key phases: Design, Build, Test, and Learn. We iteratively applied this cycle to both subunits of our system, as outlined below.
In addition, we extended the application of the engineering cycle to experiments that supported the Design and Learn phases. For example, Spectroscopy, detailed separately, provided critical insights into the limitations, challenges, and requirements of our biosensor. Also, our sophisticated Model further enhanced our analysis of metal and antibiotic residues through coupled differential equations and stakeholder interviews provided additional insights that reinforced our approach in the Learn and Design phases.
Below, we detail how our laboratory work on the Hydro Guardian remained centered on the iterative engineering cycle, refining each stage progressively. While some preliminary results are already mentioned here, the full dataset and findings can be found on the dedicated Results page.
To successfully implement our β-lactam detection unit, extensive literature research was conducted at the outset of the project to explore potential detection and signalling mechanisms. This phase was further supported by insights gained from spectroscopy and stakeholder interviews. Our goal was to achieve detection sensitivity at least comparable to that of spectroscopy and to exceed the legal concentration thresholds for β-lactam detection.
As outlined in the description, both detection units are intended to function together in HEK cells. Therefore, a significant challenge emerged: there is no known mammalian or eukaryotic β-lactam detection pathway. This led to our primary engineering question: Can we leverage the prokaryotic PASTA domain, coupled with downstream prokaryotic or known eukaryotic targets, to function within eukaryotic cells? We will answer this in the following. For plasmid design, we utilized SnapGene software, generously provided by SnapGene, which streamlined the process and facilitated the development of our constructs.
In the first phase, we focused on selecting the genes for four proteins that could play a role in the detection and signal transduction of beta-lactam antibiotics: pknB, graR, ccpA and atf2. Our β-lactam detection unit is based on a system that combines both bacterial and human components to detect β-lactam antibiotics in HEK cells. The function of our system is based on signal transduction, where the antibiotic acts as an external signal and triggers a signalling cascade leading to a measurable fluorescent response (Fig. 3). The central component is PknB (BBa_K5317013), a serine/threonine protein kinase from Staphylococcus aureus, which specifically recognizes β-lactam antibiotics via its PASTA domains. This domain is characterized by its ability to recognize β-lactam compounds, which activates downstream signalling components through autophosphorylation.Text
These signalling components include the transcription factors GraR (BBa_K5317015), CcpA (BBa_K5317014) and ATF2 (BBa_K5317016). All can be activated by phosphorylation by PknB and play a critical role in signal transduction. GraR and CcpA are also derived from S. aureus and are normally responsible for the activation of genes involved in bacterial adaptation to stress.TextText ATF2 is a human transcriptional activator that regulates the transcription of various genes. Particularly noteworthy is its ability to be phosphorylated by the bacterial kinase protein PknB, which activates it and influences the transcription of specific target genes.Text
All genes were codon-optimized for expression in HEK cells and fused with fluorescent reporter proteins (mRuby2 (BBa_K5317001) or EGFP (BBa_K3338006)) to monitor their localization and activity in the cells.
After design, we introduced the genes pknB, graR, ccpA and atf2 into the HEK cells by cloning them into expression plasmids under the control of a CMV promoter. These plasmids were expressed in the cells and the fluorescent labelled proteins were then examined for their localization and expression.
Click here to look at the plasmids! (Fig. 2)
In the first test phase, we examined the function and localization of the proteins in the cells using fluorescence microscopy. We found that only three of the four proteins showed activity and were localized; ATF2 in particular was characterized by strong expression and precise localization in the cell nucleus, while GraR shows a weak and CcpA no expression. Particularly noteworthy here is the very nice localization of the prokaryotic membrane protein PknB in the eukaryotic cell membrane. These results indicated that ATF2 was the most effective transcription factor for our detection unit. We have outlined this in much detail in the Results section.
The qualitative validation via microscopy showed that GraR and CcpA are not the suitable parts for our biosensor. Both parts were either not expressed or only expressed with a weak signal. Moreover, we learned that ATF2 and PknB showed the best performances, both in terms of its expression and its localization in the cell. With this knowledge, we decided to focus the next cycle of the engineering cycle specifically on ATF2.
In the design process of the next cycle, we focused on the development of a new promoter specifically optimized for activation by ATF2. Based on this protein, we developed a system in which antibiotic activation leads to the binding of ATF2 to the specially designed ATF2-3xCRE3xAP1 promoter (BBa_K5317017) (Fig. 5).
This promoter contains specific binding sites for ATF2 and activates the expression of a fluorescent reporter protein (miRFP670) when ATF2 binds to the promoter. The binding sites include three CRE, short for cyclic AMP-Response Element, and three AP-1, short for Activator Protein-1, sequences that increase the specificity and strength of the gene regulatory response Text. Beside ATF2, CRE-sites are getting bind by CRE binding proteins (CREB), which are part in the regulation of several cellular responses Text. CREs have the following consensus sequence 5′-TGACGTCA-3′ Text. Moreover, ATF2 is able to bind to AP-1 sites with the consensus sequence 5’-TGACTCA-3’. AP-1 proteins are containing several groups of proteins and transcription factors like Fos proteins, Jun proteins or activating transcription factors Text. This specific customization is intended to develop a system that responds to antibiotic activation and induces a measurable fluorescent response. To finalize the promoter construct, a miniCMV promoter unit was used to reach a strong expression level in the HEK cells.
Once the promoter was designed, we cloned it into a plasmid carrying the miRFP670 reporter. This allowed us to measure a fluorescent response to activation by beta-lactam antibiotics.
Click here to look at the plasmids! (Fig. 4)
We then tested the new promoter by adding ampicillin (0 µg/mL, 2.5 µg/mL, 5 µg/mL, 10 µg/mL, 25 µg/mL, 100 µg/mL) to the cells. This led to the activation of ATF2, which in turn triggered the specially designed promoter and induced expression of the miRFP670 reporter. The fluorescent reporter showed a response, confirming the efficacy of our design. We have outlined this in much detail in the Results section.
After the second cycle, we found that our sensor responded to antibiotics, but only to small amounts. This showed us that the basic design of the detection system works, but the sensitivity needs to be optimized to detect larger antibiotic concentrations or different β-lactam compounds more reliably. From these results, we were able to identify several potential areas for improvement:
These findings will form the basis for the next cycle in the engineering process, in which we would like to further refine our detection system and significantly improve its sensitivity.
Next to the successful implementation of our β-lactam detection unit, we aimed to engineer a metal detection unit applying the same challenges (e.g. minimal concentration) as above. Here, we could use a system which is already characterized to some extend in mammals and based on the Metal-responsive Transcription Factor-1 (MTF-1) as we explain in detail in the description. However, it is not yet clear if this system can be applied for metal residue detection and our following engineering cycle addressed this aim.
The first important part is the transcription factor MTF-1 (BBa_K5317007). MTF-1, short for Metal-responsive Transcription Factor-1, is found in humans, but also in mice TextText. In animal cells, MTF-1 serves as part of a defense mechanism against heavy metals, which enter the cell. Divalent ions such as cadmium, zinc or copper are recognized and bound Text (Fig. 7). Recognition of the metal ions leads to translocation from the cytoplasm into the cell nucleus. There it binds to the consensus sequence 5'-TGCRCNC-3' (R, purine; N, any nucleotide) of the MREs (metal-responsive element) in the promoter of the target gene, such as that of the metalthioneins (MT) Text. We obtained the sequence for the MTF-1 about the open source “Alliance of Genome Resources”. The sequence for mRuby2 was obtained from the pLenti Lifeact-mRuby2 PuroR plasmid.
The second important part for our biosensor is the before mentioned MRE, in front of the MTs. These are small metal-binding stress proteins that play an important role in metal homeostasis and much more interestingly in detoxifying the cell from toxic metal ions Text. The wild-type (WT) promoter contains five different consensus sequences (a to e). Of these, sequences “a” (5'-CTTTGCGCCCGGACT-3') and “d” (5'-CTCTGCACTCCGCCCGA-3') are bound by MTF-1 Text. For the design of the MRE promoter, we followed the work of Searle et al. (1985)[11] and adapted the sequence order and composition accordingly. We used the following MREs for the promoter - “a” (5'-CTTTGCGCCCGGACT-3') and “d” (5'-CTCTGCACTCCGCCCGA-3') as well as the WT promoter (BBa_K5317003) with the MREs “a”, “b” (5'-CTGCTGGGTGCAAAC-3'), “c” (5'-AAAGTGCGCTCGGCTC-3') and “d” Text. In addition to the WT sequence, the following arrangements were chosen for the other three promoters - 4x “a” (BBa_K5317004), 4x “d” (BBa_K5317005) and “dada” (BBa_K5317006) (see Figure 8). To show the activity and cell localization of our components, we decided to label the MTF-1 with mRuby2 (BBa_K5317001) and the respective promoters with eGFP (BBa_K3338006, iGEM Hannover 2020).
Based on this, we developed our metal biosensor and designed the corresponding sequences. For cloning, the promoters were amplified by PCR with the designed primers HGm_MRE-prom_fw and HGm_MRE-prom_rev with overhangs. The MTF-1 and the corresponding mRuby2 were ordered as ready-to-use GeneBlocks with matching overhangs from IDT. With the help of HiFi DNA Assembly, the respective parts were integrated into the linearized backbone vector EGFP-C2 (BBa_K3338020, iGEM Hannover 2020), making it possible to recreate the desired plasmids in a simple and reliable way. This resulted in the following composite parts, shown in figure 6: (BBa_K5317012) & (BBa_K5317008, BBa_K5317009, BBa_K5317010, BBa_K5317011)
Click here to look at the plasmids! (Fig. 6)
The finished plasmids were validated and confirmed by colony PCRs and sequencing. Subsequently, the HEK293T cells were transfected with the constructs using the GenJet™ reagent. The successful transfection was confirmed the next day by fluorescence microscopy and localization in the cell was checked.
After successful confirmation of localization and basal activity of the constructs, stimulation with copper sulfate (CuSO4) could be performed. Different concentrations (0 µM, 50 µM, 100 µM, 200 µM, 300 µM & 500 µM) were tested. The different concentration levels were used to test the sensitivity of the promoter (Fig. 8). If Cu2+ ions are present in the medium, MTF-1 is activated and the MRE sequences of the synthetic promoter are bound. This leads to the expression of the reporter gene for the fluorescent protein EGFP (Fig. 9). With increasing concentration of Cu2+ ions, fluorescence should increase in comparison to the basal level.
The tests indicated the best activity for the MRE-containing promoter “dada”. There was the highest increase in fluorescent signal between basal level to stimulation (500 µM) detectable. We have outlined this in much detail in the Results section.
What we have learned through the testing phase was that the promoter “dada” has the strongest activity. Sadly, there was no time left to improve this promoter. For the future, this promoter could be extended/shortened by “a” and “d” sequences. Moreover, the impact of the consensus sequences “b” and “c” on the synthesized promoter could be considered. Moreover, the functionality of the biosensor could and should be tested for their sensitivity in the presence of other heavy metals like zinc, cadmium or the more toxic lead.
