3. Platform application

3.1 Synthesis of L-Erythrose

We synthesized the genes sldAB and l-Ri and gained their DNA samples by pcr (Fig. 23, 25). Then we constructed them under the control of pJ23101, B0034 or pT7-lac. Sequencing results showed that the construction was successful (Fig.24, 26).

We used colony pcr to test the transformation to BL21(Fig. 27), and added IPTG to induce the expression of SldAB(SLDH) or L-RI. As shown in Fig. 28, L-RI and SLDH (SldAB) succeeded expressed in BL21(DE3).

Subsequently, we tested the enzymatic activity of SLDH (SldAB) using sorbitol dehydrogenase (SLDH) Assay Kit. The kit is based on the reduction of the tetrazolium salt MTT in a NADH-coupled or PQQH-coupled enzymatic reaction to a reduced form of MTT which exhibits an absorption maximum at 565 nm. We first set up a standard curve (Fig. 29). But then, we added 1μg/ml PQQ and performed enzyme activity measurement (Table 1), and we found that the value of enzyme activity measurements were negative. We suspected that either our standard curves were produced incorrectly (we differ significantly from those produced by biology company) or that SldAB does not function properly in E.coli.

In general, we successfully expressed L-RI and SldAB in Escherichia coli, but unfortunately, due to our lack of chromatographic columns that could resolve erythritol, L- erythrulose, and L- erythrose, we were unable to carry out further characterization beyond enzymatic activity. It is hoped that future teams can further characterize them.

3.2 synthesis of PQQ

We cloned the pqqABCDEF from the Klebsiella Pneumoniae genome (Fig. 30), and constructed this gene cluster under the control of pJ23100, pJ23106, pJ23114, pJ23101 and B0034. Colony pcr shows the transformation succeed (Fig. 31).

As for demonstrating the producing of PQQ, we tried UPLC and NBT-based non-enzyme method to test the concentration of PQQ (Fig. 32). We found that for PQQ the detection limit of UPLC was relatively high (PQQ concentration of 6mg/ml basically did not have a peak). So we chose the NBT-based non-enzyme method to test the concentration of PQQ(Flückiger et al., 1988). NBT is nitro-blue tetrazolium, which could be reduced by PQQ and shows an absorption in 530nm. We first used NBT to establish standard curves of PQQ concentration and A530nm (Fig.33, 34).

We tested the production of PQQ(Fig. 35) in E.coli DH5alpha. Compared with control group (blank vector), pqq cassette carried strains presented significant production of PQQ (Fig. 35). Four promoters of different strengths presented similar unit productivity of PQQ (Fig. 35, right Fig ), which may be related to the decrease of bacterial activity caused by the high intensity of constitutive expression. Then we chose the medium strength promoter pJ23106 and transferred the plasmid into E.coli BL21. We found the production of PQQ was much higher than in DH5alpha and we plotted the concentration of PQQ over time in BL21 (Fig. 36).

Then we transferred those four pqq cassette plasmid into the probiotic E.coli Nissle 1917 (EcN), using colony pcr to test transformation (Fig. 37). As shown in Fig. 38, EcN strains successfully produced PQQ, and the promoter pJ23106 presented the highest PQQ production in EcN, showing an unit product of 2mg/L.

Overall, we successfully cloned and characterized PQQ production in probiotic EcN, building a platform for enzymes that require pqq as a coenzyme to function properly in the EcN. At the same time, EcN which produces PQQ also has the potential to become a health care probiotic product.

3.3 Degradation of trimethylamine

The plasmids pUC57-pJ23106-tmd and pUC57-pJ23106-dmd were a gift from our partner HZAU-China. More detailed characterization results can be found from their wiki.

Here, we use dmd and tmd as curing genes, validating the role of our platform in the following two parts: (1) Verify the differences in expression of curing genes in the use of erythritol as a carbon source (2) Verify that erythritol induction systems can initiate the expression of curing genes.

Firstly, we transferred the pEryU with pUC57-pJ23106-tmd or pUC57-pJ23106-dmd into E.coli, using different carbon source group culture for 24 hours. Then collect the bacteria and isolate RNA, use RT-qpcr to test the expression of dmd and tmd. But due to a lack of effective qpcr primers for dmd, we only tested the expression of tmd.

As shown in Fig. 39, in LB broth, single plasmid (pUC57-pJ23106-tmd) strain presented the lowest relative expression. Then single plasmid strain in M9 culture with 4g/L glucose and 2g/L erythritol showed the highest expression of tmd, which seemed to indicate that the erythritol could enhance the expression of exogenous genes. Moreover, the consuming of the erythritol might decrease this effect (column 2 and 6). Using erythritol as an independent carbon source, although the growth is slower, the expression of curing gene does not drop much (column 4). Even the relative expression was higher than in the group using two carbon sources, erythritol and glucose (column 6). We suspect that this phenomenon may be due to the expression of metabolic genes that are less used by expression machines (such as ribosomes) due to lower growth efficiency, so more exogenous genes are expressed. Accordingly, when using erythritol as a carbon source, our engineered bacteria can normally express exogenous curing genes

In order to test our induction platform, we replaced the sfGFP region with tmd gene as the plasmid in Fig.19 (T7-lac controls the eryD and pJ23101-eryO controls the tmd gene). We clone the tmd and the plasmid backbone (Fig. 40) and transferred the plasmid into the E.coli BL21(DE3), using colony pcr demonstrate the successful transformation (Fig. 41). Then then we incubated this strain in different conditions (adding of IPTG, adding of erythritol) and measured the relative expression intensity with RT-qpcr. As presented in Fig. 42, we found that with the induction of erythritol, the expression of curing gene tmd tripled than those groups without adding erythritol, indicating that our induction system functions successfully.

4. Future work

Even though we found that erythritol did increase the growth of the strain to some extent, the promotion effect was very limited. We should carry out community competition experiments to observe whether adding erythritol can increase the competitiveness of pEryU-carrying strains, even our dry lab has done the simulation of competitiveness. At the same time, since the gut is a very complex system, it is necessary to conceive animal experiments to examine whether the diet of erythritol increases the colonization and community density of pEryU-carrying probiotics.

The production of PQQ did not represent in our pEryU-carrying strains. In the future, we hope to observe the effect of erythritol utilizing on PQQ production.

The endogenous proteins of E.coli that may transport erythritol have been predicted by modeling, but the proteins that actually transport erythritol into E.coli are unknown. In the future, the corresponding protein can be knocked down to observe the utilization ability of erythritol. At the same time, we can also introduce eryEFG into Escherichia coli to observe whether it can increase the utilization effect of erythritol.

Whether the sensitivity of the erythritol induced expression system is as high as the concentration of erythritol in the gut is not known, and more data on the dynamics of erythritol concentration in the gut is needed if it is to be used for induction in vivo. There might be a need for directed evolution of eryD or eryO to increase or decrease their sensitivity

In addition, we did not detect the function of SldAB and L-RI because there was no corresponding column. We hope to carry out this part of experiment in the future.

Reference

Flückiger, R., Woodtli, T., Gallop, P.M., 1988. The interaction of aminogroups with pyrroloquinoline quinone as detected by the reduction of nitroblue tetrazolium. Biochemical and Biophysical Research Communications 153, 353–358. https://doi.org/10.1016/S0006-291X(88)81230-0

He, L., 2020. Metformin and Systemic Metabolism. Trends in Pharmacological Sciences 41, 868–881. https://doi.org/10.1016/j.tips.2020.09.001

Huo YX, Zhang YT, Xiao Y, et al. IHF-binding sites inhibit DNA loop formation and transcription initiation. Nucleic Acids Res. 2009;37(12):3878-3886. doi:10.1093/nar/gkp258

Šoltysová M, Sieglová I, Fábry M, Brynda J, Škerlová J, Řezáčová P. Structural insight into DNA recognition by bacterial transcriptional regulators of the SorC/DeoR family. Acta Crystallogr D Struct Biol. 2021;77(Pt 11):1411-1424. doi:10.1107/S2059798321009633

Yost, C.K., Rath, A.M., Noel, T.C., Hynes, M.F., 2006. Characterization of genes involved in erythritol catabolism in Rhizobium leguminosarum bv. viciae. Microbiology 152, 2061–2074. https://doi.org/10.1099/mic.0.28938-0