Overview

Our designed biolens allow engineered bacteria to remain dormant when not in use and resuscitate when removed and worn to the eyes. In the case of elevated intraocular pressure, the water inside the contact lens is squeezed out, causing an increase in internal arabinose concentration. The arabinose promoter activates the expression of salidroside production metabolism pathway (triggering mechanism refers to Tainan University in 2020), achieving the goal of preventing myopia. Because it is a medical device used for human contact, we chose Escherichia coli Nissle1917 (EcN) and designed liposome vesicles and suicide switches to ensure safety. Overall, we have designed the following modules to achieve the project objectives. Figure 1 shows the contact lens usage process we designed.

Fig.1 Contact lens design. (a) It shows the storage conditions of biolens. (b) It shows the growth trigger conditions. (c) and (d) show the production trigger mechanism)

1. Production of salidroside module: We choose to start with L-tyrosine and synthesize 4-hydroxyphenylpyruvate through endogenous action in Escherichia coli. Then, we convert it into tyrosol through yeast derived pyruvate decarboxylase and endogenous alcohol dehydrogenase. Finally, salidroside was synthesized through glycosyltransferase from Schizonepeta tenuifolia [1] [2].

2. Dormancy switch module: Only when the temperature reaches 37 °C and there is blue light irradiation, the engineering bacteria enter the active state from a dormant state

3. Liposome vesicle module: Restricting the presence of engineering bacteria in contact lenses and preventing the discharge of metabolic waste

4. Quorum-Sensing module: Control the number of engineering bacteria

5. Kill switch module: Preventing the escape of engineering bacteria from causing biological pollution

6. Vitamin E module: Enhancing salidroside distribution in the aqueous oil phase

Module 1: Production of salidroside

The construction of a salidroside synthesis pathway is the foundation of our Module 1. In order to achieve better production release effects, we have designed and attempted various modification methods to increase the yield of salidroside

1.1 Construction of salidroside synthesis pathway

Fig.2 Artificial synthesis pathway of salidroside[1]

Salidroside, as a natural phenolic antioxidant, has various effects. Scientists have successfully synthesized salidroside in microorganisms using different methods. Our team chose one of the synthesis pathways, starting from L-tyrosine, to synthesize 4-hydroxyphenylpyruvate through endogenous action in Escherichia coli, and then convert it into tyrosol through yeast derived pyruvate decarboxylase ARO10 (BBa_K4761000) and endogenous alcohol dehydrogenase ADH6 (BBa_K4761001). Finally, the glycosyltransferase UGT85A1 (BBa_K4761002) from schizonepeta was used to catalyze the attachment of glucose to the phenol position of tyrosol to synthesize salidroside [1] [2].

So, theoretically, we only need to transfer the two genes of pyruvate decarboxylase ARO10 and glycosyltransferase UGT85A1 into Escherichia coli to achieve the synthesis of salidroside. Thank very much to Prof. Qipeng Yuan, Beijing University of Chemical Technology for his assistance. His laboratory has gifted our team with three plasmids: pSA-ARO10-adh6, pET-UGT85A1, and pCS-pgm-galu. Among them, galu (BBa_K4761004) expresses glucose-1-phosphate uridyltransferase; pgm (BBa_K4761003) expresses phosphoglucose mutase, both of which can regulate metabolic flux [3] [4].

1.2 Copy number optimization

In metabolic pathways, different levels of gene expression affect the final yield. Different expression levels of genes will lead to different metabolic fluxes, which will lead to different yields. We have three plasmid skeletons with different copy numbers: low copy pSA (BBa_K4761100), medium copy pCS (BBa_K4761102) and high copy pET (BBa_K4761101) [3]. By exchanging skeleton genes, six new plasmids can theoretically be constructed. They are pSA-UGT85A1, pSA-pgm-galu, pET-ARO10-adh6, pET-pgm-galu, pCS-ARO10-adh6, pCS-UGT85A1.

In addition, we also obtained duet series plasmids containing double T7 expression frames from the laboratory of Associate Prof. Lidan Ye of Zhejiang University, namely pACYCDuet (BBa_K4761103), pCDFDuet (BBa_K4761104) and pETDuet (BBa_K4761105). Gibson Assembly is used to integrate ARO10-ADH6 into the first T7 expression box and UGT85A1 into the second T7 expression box.

1.3 Metabolic pathway modification (gene knockout)

Fig.3 Synthesis and Metabolism Pathway of Rhodiola Salidroside[1]

Due to the use of the strain's own tyrosine as a raw material for synthesis, in order to increase the yield of salidroside, it is possible to find ways to supplement and consume tyrosine in the organism. For example, the pykA and pykF genes encoding pyruvate kinase isoenzymes are both knocked out to block the metabolic flux associated with the conversion of phosphoenolpyruvate (PEP) to pyruvate, further enhancing the availability of PEP for DAHP synthesis. Remove the pheA gene to prevent competitive L-phenylalanine biosynthesis [5].

Our team chose to use the CRISPR-Cas9 system [4] to knock out the three genes pykA, pykF, and pheA. The specific experimental steps can be found in the protocol.

1.4 Fusion expression

Fusion expression of enzymes corresponding to two adjacent reactions in the metabolic pathway may have an impact on the reaction rate. In addition, the type of connector also affects the fusion expression effect. Therefore, we designed three flexible connectors (BBa_K4761030, BBa_K4761031, BBa_K4761032) and three rigid connectors (BBa_K4761033, BBa_K4761034, BBa_K4761035) to connect the adh6 and ARO10 genes [6], and observed their impact on yield.

1.5 Gene integration

Considering the potential loss of plasmids for subsequent production purposes, integrating exogenous genes into the genome of engineering bacteria is a better choice. We conducted gene integration experiments using the CRISPR-CAS guided transposition system [7][8]. Unlike traditional homologous recombination, this system has higher integration efficiency; Compared with ordinary transposition systems, it has an additional specificity function.

Fig.4 Schematic of INTEGRATE using a Type I-F V. cholerae CRISPR-transposon[7]

Module 2: Dormancy switch

In our designed contact lenses, there are limitations in terms of the availability of nutrients and oxygen. Therefore, we cannot initiate production by the engineered bacteria right from the beginning. To address this issue, we have devised a growth switch mechanism to ensure that the engineered bacteria start producing rhodiola glycoside only after the contact lenses are worn. After discussions, we have decided to employ a light-responsive, temperature-coupled AND gate switch for control.

In our design, the two subunits of T7 RNA polymerase are produced by two pathways, light control and RNA thermometer, respectively. In the presence of blue light, LacI:LOV debinds to the Ptrc promoter to transcribe and translate a subunit of T7 RNA polymerase. When the temperature reached 37 ℃, the SD sequence, which was originally covered by the ASD sequence, melted, and the ribosome was able to bind and initiate translation. The ribosome binds to and expresses another subunit of T7 RNA polymerase. When both subunits are present in the bacteria, the T7 promoter activates the FLP recombinase. The FLP recombinase expresses and cuts the FRT sites at both ends of the hicA box, eliminating the toxin hicA and allowing the bacteria to enter an active state.

2.1 Blue light detection

Irradiation of the retina with blue light generates free radicals, which can contribute to the degeneration of retinal pigment epithelial cells. The deterioration of epithelial cells can result in a lack of nutrition in photosensitive cells, resulting in irreversible visual impairment.

Therefore, we plan to introduce blue light into conditions that activate the production of engineered bacteria. We refer to the work of iGEM Toronto 2017 and simplify it. Ptrc promoter, which is inhibited by LacI:LOV, is used to trigger the photosensitive switch. In the dark, LacI:LOV inhibits Ptrc promoter activity and downstream genes are not expressed. When blue light is irradiated, LacI:LOV debinds to Ptrc promoter promoter, and downstream gene expression is activated.

Fig.5 Light control switch

2.2 RNA thermometer

RNA thermometers (RNA thermosensors) are RNA-only translational control elements that sense temperature changes. They rely on a simple principle: the temperature-dependent melting of an RNA secondary structure in the 5′ untranslated region (5′ UTR). At low temperatures, the 5′ UTR adopts a conformation that masks the ribosome-binding site (Shine-Dalgarno (SD) sequence) by complementary base pairing and, in this way, makes it inaccessible to the 30S subunit of the bacterial ribosome. A shift to higher temperatures switches on translation by melting the secondary structure that harbors the SD sequence, thus allowing for ribosome binding and translation. Conversely, a shift from high to low growth temperatures switches off translation by allowing formation of the secondary structure that masks the SD sequence[9].

Fig.6 Modular design of synthetic RNA thermometers[9]

We designed five RNA thermometer elements [10] with the designations U6-1(BBa_K4761040), U6-2(BBa_K4761041), U7(BBa_K4761042), U8(BBa_K4761043), and U9(BBa_K4761044). The corresponding RNA thermometer element with the suffix 'new' differs only at the left and right end of the digestion site (BBa_K4761045-BBa_K4761049).

2.3 Construct AND gate

Our inspiration comes from the work of iGEM Glasgow 2017, which utilized two subunits that express T7 RNA polymerase to construct an AND gate. A portion of the purified T7 RNAp protein is often cleaved into two subunits, a small subunit (179 aa) and a large subunit (701 aa). If these two cleavage products are expressed as separate peptides within the cell, they can drive gene expression under the control of the pT7 promoter. As shown in the diagram below, each module expresses one subunit of T7 RNA polymerase. In the presence of both subunits, they reassemble into a functional enzyme that drives transcription from the T7 promoter, enabling downstream gene expression.[11]

Fig.7 AND gate

2.4 toxin-antitoxin system

We drew inspiration from the work of iGEM NUS 2019,, which utilized a toxin-antitoxin system to regulate bacterial dormancy. In our design, the hicA toxin and hicB antitoxin were constitutively expressed at a basal level. Flanking the entire hicA cassette were FRT sites, which could be deleted by FLP recombinase.

Prior to the use of a contact lens, the conditions did not meet the requirements of the AND gate, and the T7 promoter downstream of the AND gate was not activated. The recombinase system remained inactive, and the bacteria were in a dormant state. When we used the contact lens, the temperature and light conditions were met, and the T7 promoter was activated. The recombinase system was then activated, resulting in the excision of the hicA cassette, and the bacteria entered an active state.

Fig.8 toxin-antitoxin system

Module 3: Liposome vesicle

Liposomes are self-closed spherical structures consisting of one or several concentrically curved lipid bilayers and cholesterol[12]. They are biocompatible and biodegradable. Due to their biofilm-like structure, they can be used for many drug delivery applications, including the delivery of ophthalmic drugs through contact lenses[13].

In this part, we plan to encapsulate E. coli and the corresponding culture medium by liposomes by using Thin-film Rehydration[14][15] and further immobilize or encapsulate them inside the chamber of contact lenses, thus ensuring that engineering bacteria stay inside the lens for biosafety[16]. In addition, since vitamin E can combine with hydrophobic drugs[17], we plan to add vitamin E and glucose facilitated diffusion protein to liposomes to promote the diffusion of Salidroside, thus ensuring drug delivery.

When we add inducer(normally IPTG)in the culture medium, the engineering bacteria will start producing Salidroside and release it into the culture medium. Then Salidroside can spread from the contact lenses to eyes through the glucose facilitated diffusion protein or by combination with vitamin E.

Fig.9 Ideal schematic of lipsome vesicle

Module 4: Quorum-Sensing

As a key factor in determining microbial interactions, quorum sensing (QS) allows cells in a colony to receive signaling molecules that then trigger a series of cascading reactions and gene expression[18].

We use LuxI/R system to perform QS, thus controlling the population density of our engineering bacteria. We design to use diffusible acyl-homoserine lactone (AHL) as a signalling molecule and CcdB as our killer gene[19]. Engineering bacteria can grow normally when cell density is not high and will produce AHL. As the cell density increases, the AHL accumulates in the experimental medium and inside the cells. When the concentration of cells is high enough, AHL binds and activates the LuxR transcriptional regulator, which in turn induces the expression of killer gene (CcdB) under the control of a luxI promoter (PluxI)[19]. Sufficient CcdB protein causes death of the cell, thus inhibiting cell growth when the cell density is sufficiently high.

Fig.10 Schematic diagram of quorum sensing circuit. E stands for CcdB, I,R and R* stand for LuxI, LuxR and active LuxR, respectively.

Module 5: Bio-safety

In our design, to prevent potential contamination from escaping engineered bacteria, we have planned to construct a nutrient-deficient strain of Escherichia coli. Through a comprehensive literature review [20][21], we were delighted to discover that the pheA gene we targeted for knockout during the metabolic pathway engineering process encodes phenylalanine dehydratase.

Module 6: Vitamin E

From the exterior to the inside, the cornea is made up of five layers of tissue: the epithelial layer, anterior elastic layer, stromal layer, posterior elastic layer, and endothelial layer. The epithelial layer is the primary barrier to the permeation of hydrophilic and macromolecular medicines, but the stromal layer's high moisture restricts the penetration of macromolecular lipophilic medications. The endothelium layer is also hydrophobic. The hydrophobic - hydrophilic - hydrophobic "sandwich" shape of the cornea necessitates a high drug distribution coefficient in both the aqueous and oil phases, severely limiting therapeutic delivery to the eye. Salidroside, on the other hand, has excellent hydrophilicity but poor lipophilicity, making it difficult to pass through the cornea and necessitating modification to boost its lipophilicity.

We found a solution to add vitamin E after searching for relevant literature on eye delivery of hydrophilic drugs. Many hydrophilic medicines, such as pyrifenidone, have been released using vitamin E as an adjuvant[22]. Its aromatic ring can form hydrophobic interactions with the aromatic ring of salidroside, enhancing salidroside distribution in the aqueous oil phase.

Fig.11 Aromatic structure of salidroside and vitamin E

In addition, vitamin E can extend drug release period and give ultraviolet protection for the eyes. Contact lenses with 20% vitamin E have also been shown to retain all important lens properties, including adequate ion, oxygen permeability and translucency. Vitamin E also has biological benefits such as antioxidant, antimicrobial, and anti-inflammatory qualities that can benefit the ocular surface.[23]

Fig.12 Schematic diagram of adding vitamin E to prolong the release time of drugs[23]

---

References