Team:Tec-Monterrey/Design

DESIGN

Because our main goal consists in the development of a platform that allows us to find new antibiotics for treatment against Pseudomonas aeruginosa we designed the following working plan concerning whole cell biosensors and a mucolytic.

Microfluidic platform designs

As we needed a way to apply our biosensors, we thought on a novel, low-cost and highly efficient system that allows mixing a culture of a microorganism of interest with our biosensors so they could accomplish their function. We decided that the best kind of system to fulfill these requirements was a microfluidic device due to the low amount of material requirements, work speed and measurement precision [7]. Our main goal was to have both the biosensors and the microorganism of interest in tiny separate droplets to then merge them so bacteria could interact in a microenvironment [3].

We designed and fabricated the device with different elements; A droplet generator to encapsulate the microorganism of interest, a passive spacer to separate these droplets, a pico injection system to mix the culture of biosensors with the droplets, a reservoir to count and measure different parameters of the droplets and, finally, an outlet to recover the droplets to store and further analyse.

When one of our biosensors emits a fluorescent signal, lysis can be performed on an HPLC-MS to determine if there is a molecule that produces a spike that does not match any previously known component of the media or antibiotic. This method is useful since it could potentially save time, effort and materials invested in determining microorganisms of interest that are suspected to produce metabolites with antimicrobial properties [2].

HARDWARE

Check our hardware section!

Whole cell biosensors designs

As we planned our dispositive to be able to differentiate between antibiotics, we came up with the idea of separating the chemical entities by their mechanisms of action. With this in mind, we started an intensive research to find the best promoters that control transcription under antibiotic stress.

We decided to use only 5 promoters, which where reported by Urban et al. and belong to B. subtillis. Each one of them consists of a coding sequence that, when induced by an antibiotic, produces a specific fluorescent signal according to the mechanism of action that is beíng affected, interrupting the biosynthesis of essential molecules for the cell. [5]

Promoter	Inhibited biosynthetic pathway
helD	RNA biosynthesis
bmrC	Protein biosynthesis
yorB	DNA biosynthesis
fabHB	Fatty acid biosynthesis
ypuA	Cell wall biosynthesis

Considering that the production of the GFP by B. subtillis may be too low due to the stress the bacteria is being subjected to, we decided to complement the biosensor design by using an amplification cascade to magnify the signal of the fluorescent protein.

Amplification cascade design

The previously mentioned cascade was originally reported by Wang et al. [6]. It consists of a series of amplifier cascades that all altogether reduce the limit of detection (LOD) of heavy metals like arsenic and mercury and increase the output of a reporter gene, which in our case is case was a GFP. A whole cell biosensor can be splitted into three main modules: the sensor module, the amplifier module, and the output module (Fig. 1).

The sensor module is comprised of an inducible promoter and its corresponding repressor protein and is responsible for the manipulation of the LOD of the biosensor by varying the strength of the constitutive promoter. A low strength causes a decrease in the LOD of the biosensor, which means that even smaller concentrations of the given metal can be detected. Additionally, we describe this module in our mathematical model.

The amplifier module is where the cascade comes in, it is responsible for the amplification of the initial signal given by the inducible promoter from the first module by linking several amplifiers, creating a (sort of) cascade. Each amplifier increases the output of a given input signal. Finally, the output module is responsible for the expression of the reporter gene.

Despite being initially designed for heavy metals, we came up with the idea of using this cascade in our project by changing the metal-inducible promoter from the first module for each one of our antibiotic-inducible promoters. Then we linked this promoters to a cascade comprised of 2 amplifiers and a GFP.

The final construct ended up being composed of one promoter, 2 proteins from the first amplifier, one protein from the second amplifier, and one GFP protein at the end, with all of these proteins having their own double terminator. This made a 4400 bp long construct, which drove us to our next problem: ligating these parts.

We thought of Golden Gate cloning as a good choice for this ligation because this cloning method allows for the ligation of multiple parts in one step.

Assembly strategy

In order to synthesize all the modules that compose the cascade, we decided to split them into gBlocks, which lead us to the creation of AMP1, AMp2, AMP3 and AMP4. AMP1 and AMP2 correspond to an amplifier module, AMP3 belongs to another amplifying module and AMP4 is only a reporter gene. To assemble all these parts with the correspondent promoter, we came to the conclusion that Golden Gate was the best assembly strategy for the ligation of the AMP(1-4) but it wasn’t enough for integrating the whole cascade into B. subtilis genome. To address this, we started by making the cascade compatible with the pSB1C3 modified (Bba_P10500) standard assembly to select the transformed colonies with the blue-whithe screening method. Then, we planned on moving the entire sequence from E. coli to B. subtilis using the genome integrative plasmid pBS1C.

The following images show in a graphic way the assembly strategy:

Figure 1: The BBa_P10500 iGEM part counts with restriction sites for type IIs enzymes, such as BsaI and BsmBI, that allow this plasmid to be used in Golden Gate cloning. As it is an iGEM part it also counts with a BioBrick Prefix and Suffix.

Figure 2: The first step of our assembly strategy consisted in pre cloning by Standard Assembly our gBlocks into the iGEM part BBa_I732902, which is comprised of a lacI regulated promoter linked to a lacZ alpha fragment allowing for blue-white screening.

Figure 3: The BBa_P10500 iGEM part counts with restriction sites for type IIs enzymes, such as BsaI and BsmBI, that allow this plasmid to be used in Golden Gate cloning. As it is an iGEM part it also counts with a BioBrick Prefix and Suffix.

Figure 4: The pBS1C plasmid is used for genomic integration of a desired DNA construct into the B. subtilis genome, specifically in the amyE region. The construct of interest must be added so that it ends up flanked by the amyE sequences in the plasmid in order for it to be integrated into the genome.

Figure 5: Once the construct is successfully ligated to the BBa_P10500 plasmid by Golden Gate assembly method, it will be digested with the BioBrick prefix and suffix enzymes and ligated to the pBS1C plasmid for its final genomic integration into B. subtilis.

Mucolytic design

Having in mind the fact that Cystic Fibrosis (CF) patients mucus provides the perfect environment for bacteria proliferation and prevents them from being easily expelled, we realized that the discovery of new antibiotics will not be enough to help patients with CF. Thus we also propose the production of a recombinant protein able to reduce the density of the mucus, allowing for a proper permeability and diffusion of the sumministrated antibiotic.

To do this, we looked for an enzyme able to break down the main components of the mucus: mucins; these proteins present a high amount of sialic acid which confers density to the CF mucus. Mucins are polar glycoproteins that normally form bonds with water molecules, so if there is an increment in the amount of mucins, it will also be an increase in the density of the mucus.

The chosen protein for the breakage is the sialidase from Micromonospora viridifaciens (a harmless bacteria for humans), which according to the literature it has the capacity of degrading sialic acid [4]. To address this, we designed three different protein expression constructs. The first one is only for the expression of sialidase. For the second construct we decided to implement a super-folding green fluorescent protein (sfGFP) because is has been proved to have auto-secretory activity, making easier the detection and recovery of chimeric proteins fused to it [7]. The last construct is a chimeric protein which consists in the union of the above mentioned parts. All of the constructs are under the control of a T7-LacO promoter and C-terminal 6His-tagged, which will facilitate the recovery of the sialidase from the cell culture.

Possible secretory mechanism ofthe sfGFP fusion protein

References:

1. Hutter, B., Fischer, C., Jacobi, A., Schaab, C., & Loferer, H. (2004). Panel of Bacillus subtilis reporter strains indicative of various modes of action. Antimicrobial agents and chemotherapy, 48(7), 2588–2594. doi:10.1128/AAC.48.7.2588-2594.2004
2. Mahler, L., Wink, K., Beulig, R. J., Scherlach, K., Tovar, M., Zang, E., … Roth, M. (2018). Detection of antibiotics synthetized in microfluidic picolitre-droplets by various actinobacteria. Scientific Reports, 8(1). doi: 10.1038/s41598-018-31263-2
3. Scanlon, T. C., Dostal, S. M., & Griswold, K. E. (2014). A high-throughput screen for antibiotic drug discovery. Biotechnology and bioengineering, 111(2), 232–243. doi:10.1002/bit.25019
4. UniProtKB. ID: Q02834 (NANH_MICVI)|
5. Urban, A., Eckermann, S., Fast, B., Metzger, S., Gehling, M., Ziegelbauer, K., Rübsamen-Waigmann, H., Freiberg, C. (2007). Novel whole-cell antibiotic biosensors for compound discovery. Applied and environmental microbiology, 73(20), 6436–6443. doi:10.1128/AEM.00586-07
6. Zhang, Z., Tang, R., Zhu, D., Wang, W., Yi, L. & Ma, L.. (2017). Non-peptide guided auto-secretion of recombinant proteins by super-folder green fluorescent protein in Escherichia coli . mayo 29, 2019, de Nature Sitio web: https://www.nature.com/articles/s41598-017-07421-3
7. Zhu, Pingan, and Liqiu Wang. (2017). “Passive and Active Droplet Generation with Microfluidics: A Review.” Lab on a Chip, 17(1): 34–75.