Ligation of the dumbbell DNA template

The first step of structure-switching locked hairpin triggered rolling circle amplification is to ligate the dumbbell DNA template (DT). We used T4 DNA ligase for the reaction, next added Exo I and Exo III to degrade the excess ssDNA together with the nicked dsDNA. The agarose gel electrophoresis image of ligation product was shown as Fig. 1.

Feasibility verification of RCA Reation

After the ligation of the dumbbell DNA template, we designed a rolling circle amplification (RCA) primer to verificate it's feasibility. This primer can hybridize with the dumbbell template (DT) and open it for the following RCA. As shown in Fig. 2a (lanes 1 and 2), RCA products was formed after adding the RCA primers. Then we explored the effect of different primers' concentrations on the amount of RCA products. Fig. 2b shows that within the same reaction time, the efficiency of RCA amplification increases with concentration. For detailed protocol, please CLICK THE LINK.

Detection of OTA of varied concentrations

Firstly, through the streptavidin-biotin interaction, we attached the biotin-labelled locked hairpin (LHP) to the streptavidin-coated magnetic beads.

Then we incubated magnetic beads (MBs, 1 μm in diameter, 10 mg/mL and 300 nm in diameter, 10 mg/mL), dumbbell template, and different concentrations of OTA together. We applied the detection process shown in Fig.3 to detect OTA accurately.

As shown in Fig. 4, the fluorescence intensity rises with the increase of concentrations of OTA. Europe and China have established a maximum allowable limit of 2 μg/L for OTAs. In our experiments, 2 μg/L (5000 pM/L) OTA had a significant variation in florescence intensity compared to the blank control. MBs of different diameters gave a different performance in detection. MBs of 1 μm in diameter showed a more rapid and obvious upward trends than MBs of 300 nm in diameter.

Validation of the sIPN system

To verify the sIPN system, we engineered bacteria expressing T3-YFP (SpyTag-ELPs-SpyTag-ELPs-SpyTag-YFP) and bacteria expressing C3 (SpyCatcher-ELPs-SpyCatcher-ELPs-SpyCatcher). The constructed plasmids were transformed into E. coli BL21 (DE3) and recombinant proteins were expressed using LB medium.

Purified T3-YFP and C3 were subjected to reactions under predefined time and temperature radients. The proteins after reaction were validated by electrophoresis on polyacrylamide gels (SDS-PAGE), followed by Coomassie brilliant blue staining. A distinct target band can be observed at 130 kDa, demonstrating that T3-YFP (62.4 kDa) and C3 (54.5 kDa) are capable of forming the Spy Network (Fig.5).This reaction can occur at a variety of temperatures and has good reaction characteristics.

Fig. 5 Verification of the fabrication between T3-YFP and C3. Lane1: T3-YFP. Lane2: C3. M: Marker. Lane3: T3-YFP and C3.

At the same time, we also attempted to directly use purified protein T3-YFP to produce the above-mentioned microcapsules (Fig.6 a-2.). After placing two types of immobilized microcapsules at room temperature for 12 hours, we found that the microcapsules containing T3-YFP exhibited significant yellow fluorescence outside the microcapsules, proving that proteins can leak out of the immobilized microcapsules (Fig.6 b-1.). On the contrary, the microcapsules containing engineered bacteria exhibit brighter yellow fluorescence only in the microcapsules, without any leakage (Fig.6 b-2.).This means that the engineered bacteria and sIPN system will not leak, but small molecules of proteins will leach out of the microcapsules.

Enzyme activity assay of CPA and ADH3

To degrade Ochratoxin A (OTA) in a more efficient way, we chose two enzymes, Carboxypeptidase A (CPA) and ADH3. We used the methods described by Xiong L et al. (1992) to assay CPA and ADH3 activity. Fig.7 shows that the activity of CPA and ADH3. ADH3 was estimated at approximately 1.939 unit. CPA was estimated at approximately 0.646 unit. These results indicated that ADH3 exhibited 3.0-fold higher activity than CPA.

Moreover, we used High-Performance Liquid Chromatography (HPLC) to determine the detoxification rate of CPA and ADH3 against OTA. The HPLC chromatograms of degradation products of OTA were shown in Fig. 8. The retention times (RT) of OTA and its degradation product was 1.650 min (CPA), 1.652 min (ADH3) and 0.691 min (CPA), 0.709 min (ADH3). After the treatment of OTA with CPA and ADH3, the peak area of OTA decreased significantly compared with the control group, and the new product appeared at 0.692 min (CPA), 0.709 min (ADH3). The detoxification rates of CPA and ADH3 were 98.9% and 100%. It proved that CPA and ADH3 can degrade OTA to OTα. ADH3 gave a better performance in degrading than CPA because it took less reaction time to degrade OTA completely in higher concentrations.

Find appropriate protein combinations to functionalize the living sIPN

1.Expression and optimization of C3

We first cloned C3 into the pQE-80L, constructed pQE-80L-C3 and expressed the recombinant protein in E. coli BL21 (DE3) using Terrific Broth medium and 2xYT medium.

After incubation at 20℃ overnight or 37℃ for 4h, respectively, we found that C3 expression level in the supernatant was very low, and no obvious bands were found at 54.5 kDa as shown in Fig. 9(b-c). Considering the weak strength of the T5 promoter, we cloned C3 into a vector containing a stronger T7 promoter.

We cloned C3 into the pET-29a(+)(Fig. 2a), and expressed it in E. coli BL21 (DE3) using LB medium.

After overnight incubation at 20℃, C3 (54.5 kDa) was determined to be soluble and purified on a HiTrap Ni-NTA column. The purified protein was verified by SDS-PAGE. After that, obvious target bands can be seen at 54.5 kDa, confirming the successful expression of C3 in pET-29a(+) vector as shown in Fig. 4b (lanes 8 and 9).

2.Expression and optimization of T3-M-CPA

We cloned T3-M-CPA (SpyTag-ELPs-SpyTag-ELPs-SpyTag-Linker-M-CPA) into the PQE-80L, constructed pQE-80L-T3 and expressed the recombinant protein in E. coli BL21 (DE3) using Terrific Broth medium and 2xYT medium.

After incubation at 25℃ overnight or 37℃ for 4h and 8h, respectively, the expression of T3-M-CPA (62.4 kDa) was roughly the same as that of C3. The expression levels of both were very low. Therefore, we considered cloning T3 into pET-29a(+) vector with the same method to try to increase the expression of T3-M-CPA.

We constructed pET-29a(+)-T3-M-CPA and expressed the recombinant protein in E. coli BL21 (DE3) using LB medium.

After overnight incubation at 20℃, T3-M-CPA was purified on a HiTrap Ni-NTA column. The purified protein was verified by SDS-PAGE. As shown in Fig. 12b (lanes 1 and 2), T3-M-CPA mainly appear in the precipitation and almost non-existent in the supernatant, which proves that T3 formed inclusion body. We suspected that the eukaryotic origin of M-CPA leads to the formation of protein inclusion bodies. After reviewing literature, we found that the reducing conditions in the E. coli cytoplasm doesn't seem to truly favor the formation of disulfide bonds in M-CPA.

To reduce the formation of inclusion body, we tried SHuffle T7 E. coli expression cell to achieve soluble expression of T3-M-CPA. The SHuffle T7 E. coli strain constitutively expresses a chromosomal copy of the disufide bond isomerase DsbC, which promotes the correction of mis-oxidized proteins into their correct form, and the cytoplasmic DsbC is also a chaperone that can assist the folding of proteins that do not require disulfide bonds.

In this case, we cloned T3-M-CPA into pET-29a(+), and expressed in SHuffle T7 E. coli using 2xYT medium. After incubation at 20℃ overnight, the soluble expression of T3-M-CPA in SHuffle T7 E. coli did not increase significantly. Therefore, we considered adding a small ubiquitin-like modifier (SUMO) protein to further help the expression of T3-M-CPA.

We cloned M-CPA into pET-PC-SUMO vector, and expressed the protein in SHuffle T7 E. coli expression cell using 2xYT medium.

After incubation at 20℃ overnight, the soluble expression of SUMO-M-CPA (48.5 kDa) was slightly increased. However, most M-CPA still existed in precipitation. Thus, we tried to increase the expression level to obtain more soluble M-CPA.

The large expression of SUMO-M-CPA was purified and verified by SDS-PAGE. Only a very small amount of SUMO-M-CPA was found at the 48.5 kDa band shown in Fig. 13b (lane 3). Therefore, we considered replacing the degrading enzyme with amidohydrolase 3 (ADH3) from Stenotrophomonas acidaminiphila.

3.Expression and optimization of T3-ADH3

We obtained the plasmid pET-46 Ek/LIC-ADH3 from Associate Professor Longhai Dai of Hubei University, and then we cloned ADH3 into PET29a(+)-T3-M-CPA vector in Fig. 14a. ADH3 and T3-ADH3 were expressed by E. coli BL21 (DE3) using LB medium.

After overnight incubation at 20℃, ADH3 (43.4 kDa) was purified. The purified protein was verified by SDS-PAGE. After that, obvious target bands can be seen at 43.4 kDa and 73.6 kDa shown in Fig. 14c (lanes 4 and 5) and Fig. 14d (lanes 1 and 2), respectively, confirming the successful expression of ADH3 and T3-ADH3. Therefore, we chose T3-ADH3 and C3 as the two monomers of the sIPN system.