In order to reduce the detection limit of the biological detection system, we modified the ligand binding domain (LBD) of olfactory receptor by AutoDock prediction to improve the specificity and affinity of receptor.

HarmOR10 is an odor receptor gene cloned from the antennae ofHelicoverpa armigera, and its encoded olfactory receptor is a member of the class A rhodopsin family of G protein coupled receptors (GPCR). Olfactory receptors (ORs) are chemoreceptors expressed in the cell membrane of olfactory receptor neurons, which are responsible for detecting odors that produce olfaction. The activated olfactory receptors trigger nerve impulses and transmit information about smell to the brain to form the sense of smell. Helicoverpa armigera olfactory receptor 10 can specifically detect benzaldehyde in the environment. We introduced it into our engineering bacteria to build a detection system to automatically identify whether tea plants are attacked by tea aphids.

The three-dimensional structure of Helicoverpa armigera olfactory receptor 10 has not been included in the PDB database. Therefore, we use Alphafold2 to model and speculate on its three-level structure. A total of five simulation structures are obtained, and the scores are shown in the figure below.

AlphaFold2 is an artificial intelligence program of DeepMind Company. First, it was proposed that the protein structure can be predicted with atomic precision based on the calculation method, even in the absence of homologous templates. Alphafold2 integrates the physical and biological knowledge of protein structure into the design of deep learning algorithm by using multiple sequence alignment, and shows very high accuracy on CASP14. The prediction of most protein structures is only one atom wide from the real structure, reaching the level of human observation and prediction by using complex instruments such as cryoelectron microscope. After using AlphaFold2 to simulate the three-dimensional structure of the protein, we selected the model 3 with the highest score. The simulation results are shown in the figure below. The small ligand molecule forms a hydrogen bond with Lys at the 89 position of the protein.

In order to improve the degree of binding between the receptor and benzaldehyde, saturation mutation was carried out on the amino acid residues near the binding site. According to the preliminary speculation of charge and interaction, virtual saturation mutation was carried out at five sites, namely 21, 20, 36, 23 and 34, to observe the change of their binding ability.

Saturation mutagenesis was used with computational design. In order to evaluate of the effect of mutations, we used Autodock4 to calculate the binding energies, which brings much convenience to our model comparisons. AutoDock4 is a computational docking program based on an empirical free energy force field and rapid Lamarckian genetic algorithm search method. Docking is performed in two steps: first, a grid-based lookup table of interaction energies is calculated for the receptor using AutoGrid, then ligands are docked with AutoDock using this information. The detailed calculation is described as following.
Evaluation of binding free energy:

where,
∆E_ff--the interaction energy between the ligand and the protein calculated by the CHARMm force filed; P_HB--hydrogen bonding penalty.

The equation for calculating ∆E_ff:

where,
∆E_vdW--the intermolecular van der Waals energy;
∆E_coul--the intermolecular Coulombic energy in vacuo;
∆E_strain--the strain energy of ligand upon binding;
∆G_solv--the change in solvation energy of ligand and protein upon binding.

The method for calculating P_HB:

where,
w--hydrogen bonding weights;

f_hb—the fraction of hydrogen bonding to that of an optimum geometry;

r--the distance between the hydrogen atom and the acceptor;
θ--the angle centered at hydrogen among donor, hydrogen and acceptor.

The results about binding free energies with mutations are shown below.Simulation results show that mutants (K34V) may have better performances.

[1] Ningcan Liu.(2014).Cloning and function analysis of general odorant receptor genes from cotton bollworm, Helicoverpa armigera (master's thesis, Northeast Forestry University).
[2] Jumper J, Evans R, Pritzel A, Green T, Figurnov M, Ronneberger O, Tunyasuvunakool K, Bates R, Žídek A, Potapenko A, Bridgland A, Meyer C, Kohl SAA, Ballard AJ, Cowie A, Romera-Paredes B, Nikolov S, Jain R, Adler J, Back T, Petersen S, Reiman D, Clancy E, Zielinski M, Steinegger M, Pacholska M, Berghammer T, Bodenstein S, Silver D, Vinyals O, Senior AW, Kavukcuoglu K, Kohli P, Hassabis D. Highly accurate protein structure prediction with AlphaFold. Nature. 2021 Aug;596(7873):583-589. doi: 10.1038/s41586-021-03819-2. Epub 2021 Jul 15. PMID: 34265844; PMCID: PMC8371605.