Introduction

After deciding to design an AND-gate for sensing hypoxia and high lactic acid, we considered how to detect the expression intensity in different oxygen environments and various lactic acid concentrations. A relatively established measurement method is ELISA, but Multi-Mode Microplate Reader (Biotek) lacks control over the concentration of lactic acid and oxygen. In our actual experiment, we cultured E. coli in a stable environment for a specific duration before transferring it to Biotek for measurement. The fluorescence intensity data obtained from Biotek also require adjustment to exclude the influence of bacterial culture concentration. In the experiment, the fluorescence intensity is divided by the OD value. We obtain differentiated results by Biotek, but Biotek can only perform fluorescence intensity measurements at a macroscopic level, with relatively lower accuracy. Additionally, the plasmids we introduced into E. coli are of low copy number, resulting in inherently low fluorescence intensity. This factor increases the potential for measurement error in our results. A more accurate measurement method can enhance the reliability of our data. Our ultimate plan is to employ microfluidic chip technology to acquire single-cell precise fluorescence intensity data at specific oxygen and lactate concentrations.

Design of the microfluidic chip

Microfluidic chips need to meet the following functions:

1, There is an oxygen concentration gradient in the chip environment.

2, The chip can bind bacteria, allowing them to grow in a confined area.

3, The chip can simultaneously measure various E.coli strains without interference between them.

Based on the above three requirements, we make changes on the basis of the previous chip and design a chip that forms six oxygen concentrations and measure four types of E.coli simultaneously (see Figure 1).

We designed the chamber with a very low height of only 1.3 microns and a narrow chamber outlet of only 3 microns, ensuring that E. coli becomes trapped once it enters. The method of loading bacterial culture and culture solution is shown in Figure 2.

Figure 2 The method of loading. Step 1 load the E.coli, Step 2 load the solution and flush the E. coli into the trap.

The first step of loading is to plug the culture inlet. The bacteria solution flows in from the bacteria inlet and out through the bacteria outlet. At this stage, a large number of E.coli will be distributed in the flow channel. The second step of loading is to plug the bacteria inlet and open the culture inlet. The culture solution will pass through the trap due to the pressure difference, thereby flushing the E. coli into the trap. This method not only allows real-time updating of the culture fluid in the trap but also enables trapping of the E.coli.

Manufacture process of the microfluidic chip

We first use laser printing to create a mask consisting of five layers of chips (see Figure 3a). Next, we apply a certain height of photoresist and expose it to develop a mold consisting of two layers of chips (see Figure 3b and Figure 3c). PDMS is then spread on the mold to obtain the upper and lower chips, which are subsequently spliced together (see Figure 3d). The chip is then baked overnight in an oven at 70℃. The following day, the chip is punched (see Figure 3e) and combined with the slide (see Figure 3f). Finally, the chip is baked for over 3 hours.

Test of trap chamber

We tested the feasibility of the chip and proved that our chip can bind E. coli in the trap chamber. We used a 60x microscope to take time series images of E. coli activity in a trap Figure 4, and found that E. coli was bound in the trap, and we could take clear and stable growth images.

Test of oxygen concentration

We have tested the feasibility of the chip and proved that the chip can form an oxygen concentration gradient. The measurements were made using the ruthenium tris(2,2 '-dipyridyl) dichloride hydrate (RTDP) indicator. We loaded RTDP into the culture layer. The loading layer only load air at first, and then load nitrogen and air at the same time after a period of time to obtain the change of fluorescence intensity of the six channels over time Figure 5. Due to the defects of the air pump, the concentration gradient is not stable, but after standardization, there is a difference in fluorescence intensity between the six channels Figure 7, which indicates that our chip does construct six different oxygen concentration environments.

Using the Stern-Volmer equation, $I_0/I=1+K_q [O_2]$ where $[O_2]$ represents the concentration of oxygen, $I$ represents the fluorescence intensity at this concentration, $I_0$ represents the fluorescence intensity in the absence of oxygen, and $K_q$ is the Stern-Volmer quenching constant. We can get the fluorescence intensity $I_0$ and $I_{21}$ at 0% and 21% by passing only nitrogen and air, and then calculate $K_q$, and get the corresponding relationship between oxygen concentration and fluorescence intensity: $[O_2 ]=(I_0/I -1)/(I_0/I_{21} -1)×21\%$ Due to lack of time, we did not do a quantitative oxygen concentration calculation.

Conclusion

We design and successfully construct a microfluidic chip, and detect the chip's ability to trap Escherichia coli and to facilitate the creation of oxygen concentration gradients. Due to time constraints, we have not used the chip for measuring bacterial fluorescence intensity, but this chip is proven to be feasible. Our chip also allows for experimental measurements of expression variation caused by different lactate concentrations in the culture medium.

Our chip has a wide range of application scenarios where fluorescence measurements targeting E.coli can be conducted on it, while enabling precise control over oxygen concentration and culture medium concentration.