DeepTMHMM is a tool for predicting protein transmembrane regions based on transmembrane topology prediction and deep learning models. DeepTMHMM was used to conduct structural analysis of TlpC and Tsr proteins and found that they are both twice-spanning membrane proteins. 1-9 and 322-673 of TlpC protein are intramembrane regions, 10-30 and 301-321 are transmembrane regions, and 31-300 are extramembrane regions. 1-16 and 220-560 of the Tsr protein are the intramembrane region, 17-37 and 199-219 are the transmembrane region, and 38-198 are the extramembrane region.
Fig. 1 Analysis results of the transmembrane region of TlpC protein
Fig. 2 Analysis results of the transmembrane region of Tsr protein
According to the X-ray diffraction data of co-crystallization of TlpC protein and lactic acid included in the PDB database. The extracellular ligand-binding domain of TlpC protein has five active amino acid sites, namely ASN 213, TYR 249, LEU 252, SER253, and TYR 285, forming a ligand pocket and forming 6 hydrogen bonds with lactic acid.
Figure.3 Lactic acid co-crystal X-ray diffraction 3D structure of TlpC protein in PDB
Tsr protein lacks PDB ligand co-crystallization X-ray diffraction data and Tsr protein X-ray diffraction data. The binding status of this part was obtained by AlphaFold2's structural prediction of Tsr protein and MOE's molecular docking. Using the Tsr protein structural model with the highest prediction score from AlphaFold2, in the MOE molecular docking simulation, the far membrane end of the Tsr protein formed a ligand pocket. There are three active amino acid sites around this ligand pocket, namely MET 81, ASP 84, and LYS 93, and three hydrogen bonds are formed with the ligand molecule serine.
Figure.4 The complex structure formed by molecular docking of the Tsr protein structure predicted by AlphaFold2 and the serine molecule through MOE
The recombinant protein eTlpC is a recombinant protein in which the extramembrane ligand-binding domain and transmembrane part of TlpC are the same as the intramembrane cytoplasmic signaling domain of Tsr. Through the simulation of AlphaFold2, we obtained its predicted 3D structure. It better retains the structure of the TlpC extramembrane ligand-binding domain and the Tsr intramembrane cytoplasmic signaling domain. Its RMSD with Tsr protein is 0.796, and its RMSD with TlpC protein is 0.411. Therefore, it can be predicted that the extramembrane ligand-binding domain and intramembrane cytoplasmic signaling domain of eTlpC can complete their respective functions. The extramembrane binding to ligand molecules activates chemotactic signals, and the intramembrane binding to cascade proteins transmits chemotactic signals.
Figure.5 AlphaFold2 prediction model of eTlpC, the left end is the proximal membrane end.
Using DeepTMHMM to predict, eTlpC is still a twice-spanning membrane protein, retaining the characteristics of TlpC protein and Tsr protein. 1-16 and 328-669 are intramembrane regions, 17-35 and 307-327 are transmembrane regions, and 36-306 are extramembrane regions. The transmembrane region highly overlaps with TlpC and Tsr, so it is predicted that the transmembrane structure of eTlpC will not change.
Figure.6 The DeepTMHMM prediction results of the transmembrane region of eTlpC indicate that it is a twice-spanning protein and has a high degree of overlap with the original proteins TlpC and Tsr in each region.
Molecular docking of the ligand-binding domain of eTlpC was performed using MOE and lactic acid molecules. The docking results show that there are 4 amino acid active sites in its ligand pocket, namely ASN 220, TYR 256, LEU 259, and SER 260, and 5 hydrogen bonds are formed with the ligand lactic acid molecule. The lactate-binding activity of the eTlpC ligand pocket is therefore predicted to be unchanged.
Figure.7 the lactic acid molecule formed 5 hydrogen bonds with the 4 active amino acid sites as shown in the figure.
Use AlphaFold2 to predict the structures of scFv-EAK2-PE38 and scFv-GS3-PE38. Compared with scFv-EAK2-PE38 and Pseudomonas aeruginosa exotoxin A (PDB ID: 1IKQ), the RMSD is 0.690<3 and there is no significant difference; with When compared with trastuzumab (PDB ID: 5TDN), the RMSD is 0.519<3, which is considered to be no significant difference. When comparing scFv-GS3-PE3 with Pseudomonas aeruginosa exotoxin A, the RMSD is 0.859<3, which is considered to be no significant difference; when compared with trastuzumab, the RMSD is 0.363<3, which is considered to be no significant difference.
Figure. 8 alphaFold2 predicted structure of scFv-GS3-PE3
Figure. 9 alphaFold2 predicted structure of scFv-EAK2-PE38
Rosetta Dock was used to dock scFv, scFv-GS3-PE3 and scFv-EAK2-PE38 with the HER2 protein. The number of dockings was 10,000. The docking score between scFv and HER2 (I_sc, the smaller the better) was -25.802. The docking score between scFv-GS-PE38 and HER2 is -20.814, and the docking score between scFv-EAK-PE38 and HER2 is -24.501. Since scFv-EAK-PE38 uses a rigid linker, it may have a better effect of isolating the domain, making the scFv binding activity higher. Therefore, it is also judged that scFv-EAK-PE3 has stronger specificity. Use PDBe PISA v1.52 to analyze scFv and HER2 to generate 5 hydrogen bonds and 5 salt bridges, and the solvation free energy gain ΔiG is -0.5kcal/mol. The converted free energy gain of each hydrogen bond is -0.44kcal/mol, and the converted free energy gain of a salt bridge is -0.15kcal/mol. , the total free energy gain is -3.45 kcal/mol; scFv-GS3-PE3 solvation free energy gain ΔiG is -6.9kcal/mol, 4 hydrogen bonds are generated, and the free energy gain is -8.66kcal/mol; scFv-EAK-PE38 solvation free energy gain ΔiG is -3.1kcal/mol, generating 9 hydrogen bonds and 2 salt bridges. The free energy gain is -7.36kcal/mol. As shown in Figures 13, 14, and 15, since only the RMSD-I_Sc plot of scFv shows a convergence trend, scFv-GS3-PE3 and scFv-EAK-PE38 may not be credible.
Figure. 10 Comparative data of each single-chain antibody/single-chain antibody drug conjugate I_sc: Rosetta Dock docking score ΔiG: PDBe PISA free energy gain
Figure. 11 a. Rosetta docking model of scFv and HER2 b. Rosetta docking model of scFv-GS3-PE3 and HER2 c. Rosetta docking model of scFv-EAK-PE38 and HER2
Figure. 12 a. Compare scFv-GS-PE3 with PE protein b. Compare scFv-EAK-PE3 with PE protein
Figure. 13 RMSD-I_Sc plot of RosettaDock scFv and HER2
Figure. 14 RMSD-I_Sc plot of RosettaDock scFv-GS-PE3 and HER2
Figure. 15 RMSD-I_Sc plot of RosettaDock scFv-EAK-PE3 and HER2
Start by pre-docking eTlpC to understand the docking situation. After partial truncation, the protein (protein a) near the binding site is retained. 
Choose the CHARMM 36 force field, prepare the ligand topology for LAC (lactic acid), and integrate it with protein a. Build an appropriately sized box. 
Neutralize the system by adding 14 chloride ions to balance the 14 positive charges.
Perform energy minimization using the steepest descent method. After position restraints, carry out NVT and NPT equilibrations at 300K and 1 bar pressure, respectively.
Conduct the following analyses:
[1] Djender S, Schneider A, Beugnet A, Crepin R, Desrumeaux KE, Romani C, Moutel S, Perez F, de Marco A. Bacterial cytoplasm as an effective cell compartment for producing functional VHH-based affinity reagents and Camelidae IgG-like recombinant antibodies. Microb Cell Fact. 2014 Sep 16;13:140. doi: 10.1186/s12934-014-0140-1. PMID: 25223348; PMCID: PMC4172947.
[2] Ahmad ZA, Yeap SK, Ali AM, Ho WY, Alitheen NB, Hamid M. scFv antibody: principles and clinical application. Clin Dev Immunol. 2012;2012:980250. doi: 10.1155/2012/980250. Epub 2012 Mar 15. PMID: 22474489; PMCID: PMC3312285.
[3] Freudl R. Signal peptides for recombinant protein secretion in bacterial expression systems. Microb Cell Fact. 2018 Mar 29;17(1):52. doi: 10.1186/s12934-018-0901-3. PMID: 29598818; PMCID: PMC5875014.
[4] Ahmadzadeh M, Farshdari F, Nematollahi L, Behdani M, Mohit E. Anti-HER2 scFv Expression in Escherichia coli SHuffle®T7 Express Cells: Effects on Solubility and Biological Activity. Mol Biotechnol. 2020 Jan;62(1):18-30. doi: 10.1007/s12033-019-00221-2. PMID: 31691197.
[5] Machuca MA, Johnson KS, Liu YC, Steer DL, Ottemann KM, Roujeinikova A. Helicobacter pylori chemoreceptor TlpC mediates chemotaxis to lactate. Sci Rep. 2017 Oct 26;7(1):14089. doi: 10.1038/s41598-017-14372-2. PMID: 29075010; PMCID: PMC5658362.
© 2023 - Content on this site is licensed under a Creative Commons Attribution 4.0 International license.
The repository used to create this website is available at gitlab.igem.org/2023/syphu-china.
