ENGINEERING

Engineering in Synthetic Biology follows a ‘Design-Build-Test-Learn’ (DBTL) Cycle, which serves as our guiding principle throughout our project. As an illustrative example, let's consider our endeavor concerning the hypoxia response module. Our primary objective entails the identification of a hypoxia-inducible promoter distinguished by a robust disparity between hypoxia-triggered and basal expression levels. To accomplish this, we judiciously employed the DBTL cycle, meticulously outlined as follows.

Cycle 1 Characterization of EcN PepT Promoter

Design

In our project, our aim was to specifically target Escherichia coli Nissle 1917 (EcN) to hypoxic tumour regions. Following an extensive review of the literature, we selected the endogenous hypoxia-inducible promoter pPepT from EcN (Fig. 1). This choice was informed by its proven utility in tumour-targeting EcN$^1$.

Figure 1 | pPepT from EcN. PepT promoter is highlighted in red. Gene sequence was downloaded from the National Center for Biotechnology Information (NCBI)

Build

We use the PJUMP41-2A (sfGFP) plasmid (BBa_J428365, this plasmid referred below are abbreviated as O12, as it corresponds to the O12 well on the biobrick.) as our vector This plasmid is characterized as a low-copy plasmid with spectinomycin resistance and carries the BBa_J23100 promoter along with the sfGFP sequence.

In our experimental procedure, we replaced the original promoter and ribosome binding site (RBS) region with the pPepT sequence sourced from EcN. Specifically, we designed a pair of primers with 15 - 20 bp overlapping regions. Subsequently, we performed a reverse PCR amplification step, followed by DpnI (NEB) digestion of the plasmid template. The resulting DNA fragment was purified through gel electrophoresis.

Afterwards, we conducted a Gibson reaction to facilitate self-ligation. The resulting product was then transferred into DH5alpha competent cells for amplification. To ensure the integrity of the modified plasmid, we conducted a thorough sequencing analysis, confirming the correctness of the target plasmid sequence. Then we got plasmid O12-pept (sfGFP), which was subsequently transferred into EcN.

We applied the same procedures to another plasmid pJUMP49-2C(sfGFP) (abbreviated as K12, medium copy) and obtained the K12-pept plasmid concurrently.

Figure 2 | Construction of O12-pept plasmid. The specific building steps are given in the build section.

Test

The bacterial (EcN) culture containing the O12-pept / K12-pept plasmid was inoculated into LB (with 50 μg/ml spectinomycin) at a ratio of 1:100, then was grown separately under normoxic and anaerobic conditions for 16 hours. As a blank control, EcN was inoculated into LB using the same method and grew for the same duration. OD600 and fluorescence (with 488 nm as the excitation wavelength and 520 nm as the emission wavelength) was measured using Multi-Mode Microplate Reader (Biotek).

Figure 3 | Characterization results of two plasmids. N=3 biological replicates. Data are mean ± SEM.

Learn

The endogenous expression of pPepT under anaerobic conditions exhibited poor fluorescence intensity. We speculated that the following factors may contribute to this:

1. The endogenous pPepT (with its natural RBS) induces only weak sfGFP expression.
2. sfGFP undergoes incomplete folding under anaerobic conditions, resulting in inadequate chromophore formation.

The unsatisfactory characterization results prompted us to search the literature to gain a better understanding of the mechanism behind hypoxia-inducible promoters (HIP). HIPs are regulated by the Fumarate and Nitrate Reduction regulator (FNR)$^2$. Under hypoxic conditions, FNR homodimerizes and binds to a specific DNA site termed ‘FNR box’. Upon its binding, the homodimer interacts with the RNA polymerase binding to TATA box at -10 region, thus activating transcription (Fig. 3). Mengesha et al.$^3$ made improvements to the pPepT from Salmonella by mutating its FNR-box and TATA-box, resulting in an enhanced oxygen-responsive curve.

Figure 4 | Mechanism of HIP. FNR homodimer binds to FNR binding site as it interacts with RNA polymerase to activate transcription.

Simultaneously, we conducted an in-depth investigation into the mechanism behind sfGFP luminescence. sfGFP is a modified variant of GFP and features a highly stable and well-folded structure, characterized by a beta-barrel structure, which consists of 11 strands of beta-sheet organized into a cylinder-like shape. The chromophore, Ser65-Tyr66-Gly67 tripeptide, is located in the center of the beta-barrel core. The generation of its fluorescence involves a complex biochemical mechanism, with oxygen playing an irreplaceable role. Therefore, under anaerobic conditions, sfGFP struggles to form the correctly folded chromophore, explaining its underexpression in oxygen-deprived environments.

Figure 5 | Outline of superfolder green fluorescent protein (sfGFP). (a) Crystal structure of a sfGFP protein$^4$ (obtain from https://www.rcsb.org). (b) Excitation and emission spectra of sfGFP. The data is from FPbase (https://www.fpbase.org). Based on this graph, we used 488 nm as the excitation wavelength and 530 nm as the emission wavelength, because this approach helps minimize the influence of excitation light on the measurements. (c) The luminescence principle of sfGFP$^5$.

Additionally, we observed that the baseline expression level in K12-pept was remarkably high, which contradicted our initial intentions. Consequently, in our subsequent experiments, we have opted to exclusively utilize O12 as our expression vector.

Cycle 2 Construction of More Hypoxia-inducible Promoters

Design

After going through the previous DBTL cycle, we decided to employ various FNR-box (wild type/mutant), TATA-box (wile type/mutant) and RBS (BBa_B0034/BBa_J61101) sequences to modulate the expression strength of pPepT, resulting in the design of 8 variants (Table 1). Additionally, we selected the FF+20 variant, which is also an FNR-binding-dependent promoter, for parallel comparison. In the work of Chien et al.$^1$, this variant appears to outperform the endogenous pPepT of EcN. To explain the immature folding and lower expression of sfGFP under anaerobic conditions, we employed the O12 plasmid as a control for fluorescence intensity.

Table 1 The promoter + RBS sequences used in the second cycle.P.S. WT: wild type.

Build

We utilized O12 plasmid as the vector and employed a similar approach as described above (Build in Cycle 1) to construct 9 variants listed in Table 1. The only distinction was that we employed 2 rounds of reverse PCR to generate 15 - 20 bp overlapping regions. To mimic the impact of spectinomycin on bacterial growth, we also constructed an empty vector of O12 (O12-empty). This plasmid deleted the promoter and sfGFP sequence, retaining only the backbone (ori and resistance gene), and was obtained using the method described in Cycle 1.

Test

The plasmids were transformed into DH5alpha and sequenced, yet many had peculiar mutations and even large fragment insertions (Fig. 6).

Figure 6 | Alignment of sequencing result. Red boxs show mutations and insertions.

Learn

With regard to the sequencing result, we hypothesized that there were non-specific bindings of primers to PCR templates and products. These bindings may come from two sources:

1. The reverse primer largely coincided with a 'prefix' feature, which had high sequence similarity with 'suffix' (another feature);
2. Since the second round PCR product had 19bp overlap on both sides for ligation, primers were able to bind for further extension, which may result in unexpected insertion.

Cycle 3 Large Scale Screening of HIPs

Design

To deal with the above problems, we decided to try one round of overlap PCR with redesigned primers. PCR reaction was simulated with SnapGene to ensure uniqueness of the binding site.

Build

One round of overlap PCR, product DNA extraction and subsequent transformation were executed, yielding 8 DH5alpha strains with corresponding O12-pepT plasmids.

Test

We successfully obtained the correctly sequenced plasmids and characterized them as follows:

The overnight-cultured bacterial (DH5alpha) suspension, which transformed with the corresponding plasmid, was serially diluted in culture medium to reach OD600 = 0.4. Subsequently, it was inoculated into fresh LB medium (with 50 μg/ml spectinomycin) at a 1% inoculation rate, then grew separately under normoxic and anaerobic conditions for a period of time. Specifically, we designated 16 hours as the endpoint for anaerobic cultivation, while 7 hours served as the endpoint for aerobic cultivation. These values were determined by monitoring the bacterial growth curves under different conditions and roughly correspond to their logarithmic growth phase. The O12-empty plasmid (as blank control) and the O12 plasmid (as fluorescence intensity reference) were both characterized using the same method. Each group had 8 biological replicates.

Then, OD600 and fluorescence (with 488nm as the excitation wavelength and 530nm as the emission wavelength) was measured using Multi-Mode Microplate Reader (Biotek). Relative Expression level was analyzed with the following formula.

Relative Expression level was analyzed with the following formula:

The results visualized as Fig. 7.

Figure 7 | Characterization results of 10 HIP variants. (a) Outcome of relative sfGFP / OD600 nm. N = 8 biological replicates. Data are mean ± SEM. (b) Fold changes of biological replicates. N = 8 biological replicates. Data are mean ± SEM.

Learn

Our results show that the FF+20 variant exhibits the strongest sfGFP fluorescence under anaerobic conditions, although it also has the highest baseline expression level. Mutating the TATA-box enhances expression levels under anaerobic conditions, but it also leads to increased expression under aerobic conditions. Similar effects are observed when changing the RBS. However, altering the FNR-box did not result in significant expression distinction. Moreover, dry lab provided basal expression levels for each promoter, and our wet lab data aligns relatively well with this model. Overall, we consider PepT-JFT to be promising anaerobic promoters.

However, we also acknowledge that our characterization was conducted on a macroscopic scale, and some measurement errors may be inherent. Therefore, we aimed to observe individual or small groups of cells at the microscopic level to obtain more precise data. To achieve this, we employed microfluidic technology.

Cycle 4 Microfluidics for Precise Characterization at the Single-cell Scale

Design

In order to measure fluorescence intensity with single cell accuracy at precise oxygen concentrations, we specially designed a microfluidic chip. With the help of Prof. Chunxiong Luo, we designed a chip that can simultaneously measure the fluorescence intensity of 4 E. coli strains under 6 oxygen concentration gradients$^6$, as shown in Fig. 8.

Figure 8 | Schematic diagram of microfluidic chip. The above figure shows the details of the culture layer (Mold 1) and loading layer (Mold 2). The culture layer (Mold 1) is composed of 3 masks on the left, and the loading layer (Mold 2) on the right is composed of 2 masks. To clearly show the detailed pattern of each layer of the two molds, the size in the figure does not represent the true proportion of the structure. All our masks are drawn using L-edit software, and the tdb format file is also uploaded in Gitlab.

Build

We have made the microfluidic chip according to the drawing, and the real picture is shown in Fig. 9.

Figure 9 | Actual diagram of microfluidic chip. (a) The actual effect of the microfluidic chip. (b) The chip under the microscope. We use the lithography machine and coating machine in Chunxiong Luo's lab for chip fabrication.

Test

We demonstrated the viability of the microfluidic chip by verifying the presence of an oxygen concentration gradient and the capture of E. coli by the trap chamber. As shown in Fig. 10a, when the mixture was changed from air to nitrogen air at 10 minutes, the fluorescence intensity was found to be increased, and the 6 channels were different. The following data (Fig. 10b - c, Video 1) further confirms the reliability of our chips.

Figure 10 | On-machine test of our microfluidic chip. (a) The oxygen concentration changes over time, switching from total air to air nitrogen mixture at 10 minutes. (b - c) Green fluorescent of E. coli in a trap chamber.

Video 1 | Sample loading process of our microfluidic chip. This video was captured under a microscope.

Unfortunately, we haven’t had time to apply our microfluidic chip on measurement. However, we believe that microscopic result will be similar to the macroscopic result.

Learn

Achieving complex and precise regulation requires multiple debugging, and sufficient time needs to be reserved for experiments on microfluidic chips.

Conclusion & Further Thinking

After 4 iterations of DBTL cycles, we have succeeded in constructing and screening for a series of sensitive hypoxia-inducible promoters (BBa_K4713116 - BBa_K4713125), which could be utilized both in our project as well as our hybrid promoter parts (BBa_K4713001 - BBa_K4713014).

Despite predominant success, there is still room for improvement in our characterization of the parts. The fluorescence protein we used, superfolder Green Fluorescence Protein (sfGFP), rely on oxygen during maturation. Therefore, it would be more rigorous to change to miniGFP, an oxygen-independent fluorescence protein developed by Guo-Teng Liang et al$^7$, for characterization. In addition, we observed fluorescent expression at baseline levels, indicating that there is some escape rate and that antisense promoters can be used to further suppress unexpected expression. Unfortunately, we don't have enough time to do this. We hope to further improve our design in the future.

References

1. Chien, T. et al. Enhancing the tropism of bacteria via genetically programmed biosensors. Nat. Biomed. Eng. 6, 94–104 (2022).
2. Wing, H. J., Williams, S. M. & Busby, S. J. Spacing requirements for transcription activation by Escherichia coli FNR protein. J. Bacteriol. 177, 6704–6710 (1995).
3. Mengesha, A. et al. Development of a flexible and potent hypoxia-inducible promoter for tumor-targeted gene expression in attenuated salmonella. Cancer Biol. Ther. 5, 1120–1128 (2006).
4. Pédelacq, J.-D., Cabantous, S., Tran, T., Terwilliger, T. C. & Waldo, G. S. Engineering and characterization of a superfolder green fluorescent protein. Nat. Biotechnol. 24, 79–88 (2006).
5. Barondeau, D. P., Putnam, C. D., Kassmann, C. J., Tainer, J. A. & Getzoff, E. D. Mechanism and energetics of green fluorescent protein chromophore synthesis revealed by trapped intermediate structures. Proc. Natl. Acad. Sci. 100, 12111–12116 (2003).
6. Sun, Y. et al. Two-Layered Microfluidic Devices for High-Throughput Dynamic Analysis of Synthetic Gene Circuits in E. coli. ACS Synth. Biol. 11, 3954–3965 (2022).
7. Liang, G.-T. et al. Enhanced small green fluorescent proteins as a multisensing platform for biosensor development. Front. Bioeng. Biotechnol. 10, (2022).