Figures Legends
Figure 1 Design and reconstruction of main enzyme module and auxiliary module for the synthesis of PAPS. (A) Schematic representation of PAPS biosynthesis from ATP by the main enzymes module (ATPS and APSK) and auxiliary module (PPA and PPK); (B) Effect of different activity ratio of Kl ATPS to Pc APSK on PAPS production. The dual-enzyme system was supplemented with ATP, with Kl ATPS activity fixed at 3.0 U·mL-1. The ratio ofKl ATPS to Pc APSK was changed from 1:1 to 1:3; (C) Effect of the amount of Ec PPA on the conversion rate of PAPS; (D) Effect of the amount of Rs PPK on the conversion rate of PAPS.
Figure 2 Construction and optimization of catalytic systemin vivo . (A) Effect of substrate loading on PAPS production by strain E. coli 01; The reactions were supplemented with varying concentrations of ATP from 20 to 80 mM added at a fixed whole-cell biocatalyst (wet) 30 g·L-1 at 30°C; (B) Intracellular enzyme activities of three recombinant strains (E. coli 01,E. coli 04, and E. coli 10); (C) Intracellular enzyme activity of Rc PPK in recombinant strains with different chaperones; (D) Intracellular enzyme activity of Pc APSK in recombinant strains with different RBS and the effect on the conversion rate of PAPS.
Figure 3 Computer-assisted identification of enzyme structure. (A) The catalytic progress of Pc APSK; (B) Molecular docking and highlighted the predicted hot spots as green sticks; (C) U-shaped release tunnel of ADP and its bottleneck; (D) Root-mean-square fluctuation (RMSF) of Pc APSK; (E) Open and Closed conformation of the lid, and the structure comprises lid constituent residues (red), hinge residues (blue), and binding residues (yellow).
Figure 4 Protein engineering of Pc APSK to accelerate ADP release and mechanism analysis. (A) Alanine scanning of selected hot spots; (B) Engineering scheme of Pc APSK; Libraries A and B were constructed using the iterative saturation mutation (ISM) strategies, respectively; Library A had three residues, K38, T40, S36, ISM was used to build the library with an NNK codon; There were three residues, D139, K151, and G167 in library B, which are not close to each other in the structure, and therefore, ISM was used to build the library with an NNK codon. The order is based on the increased enzyme activity (from high to low) of alanine scanning; (C) Hydrogen bonds calculated from MD simulations of L0 and L7; (D) RMSD from MD simulations of L0 and L7; (E) Bottleneck changes in U-shaped tunnel of L0 and L7; (F) Structure alignment of L0 (brown) and L7 (green); (G) Root‐mean‐square fluctuation (RMSF) value calculated from MD simulations of L0 and L7.
Figure 5 Intracellular enzyme activity and whole cell catalysis. (A) Determination of four intracellular enzyme activities ofE. coli 11; (B) Conversion experiments of E. coli 04 andE. coli 11 at the 3‐L scale with 150 mM ATP. The E. coli04 and E. coli 11 were cultivated in a 5‐L bioreactor (with the 3‐L working volume). Then, the cells were harvested by centrifugation at 6,000g for 8 min and then placed at 35°C for 22 hr for fully self‐processing. Moreover, 30 g cells were resuspended into a 3-L of bioconversion mixture (150 Mm ATP, pH was adjusted to 8.0 with NaOH). Finally, a total of 3-L volume conversion broth was obtained.