2 In vivo construction of the PAPS biosynthesis pathway
To prove the effectiveness of the proposed catalytic pathway in vivo , the genes encoding Kl ATPS and Pc APSK in the main enzyme module and Ec PPA and Rs PPK in the auxiliary module were inserted into pACYCDuet-1 and pETDuet-1 plasmids, respectively. These were then used to transform E. coli BL21 (DE3), resulting in E. coli strain 01. Co-expression of the four enzymes was confirmed by protein gel electrophoresis of cell-free lysates. Then, the conversion performance of strain E. coli 01 was investigated with 30 g·L-1 wet cell (treated with 1g·L-1 Tween 80) at 30°C, it was found that the PAPS titer increased from 6.03 to 11.59 mM as the substrate ATP concentration was increased from 20 to 60 mM (Figure 2A ). When the ATP concentration was increased >60 mM, the PAPS titer did not augment any further. In contrast, the conversion rate of PAPS decreased from 60.13% to 25.62%, when the ATP concentration increased from 20 to 80 mM. This drop was due to APS, an intermediate of the enzymatic cascade, which accumulated at increasing amounts (from 3.86 to 16.07 mM) in the conversion broth (Figure 2A ). Intracellular enzymatic activity of Kl ATPS, Pc APSK, Ec PPA, and Rs PPK in E. coli 01 was 135.06, 12.75, 188.19, and 243.31 U·g-1 wet cells, respectively (the ratio was 3:0.28:4.18:5.41) (Figure 2B ). This finding revealed the need to increase the intracellular enzymatic activity of Rs PPK andPc APSK.
Previous studies had shown that Rs PPK could easily form inclusion bodies in cells. Therefore, four molecular chaperones (SUMO, MBP, TrxA, and GST) involved in expression of soluble proteins were co-expressed with Rs PPK to increase its activity in the cytoplasm. TrxA alone caused the intracellular activity of Rs PPK to increase from 243.31 (E. coli 01) to 317.38 U·mg-1 wet cells (E. coli 04) (Figure 2C ), while Kl ATPS,Pc APSK, and Ec PPA reached 132.61, 12.85, and 186.63 U·g-1 wet cells, respectively (corresponding to a 3:0.29:4.22:7.18 ratio) (Figure 2A ). To improve Pc APSK intracellular activity, its expression was increased by adding six ribosome-binding site (RBS) sequences (RBS1–RBS6) of different intensities (Table S4 ). RBS5 produced the best results and increased Pc APSK activity from 10.75 (E. coli 01) to 22.47 U·g-1 wet cells (E. coli 06) (Figure 2D ). However, increasing the expression of Pc APSK decreased the transformation efficiency of the system (Figure 2D ). Moreover, the intracellular activity of the other three enzymes to drop significantly to 95.07 (Kl ATPS), 87.35 (Ec PPA), and 194.05 (Rs PPK) U·g-1 wet cells (Figure 2A ). These results suggested that increasing enzyme expression alone had limited efficacy on improving overall performance and might lead to an imbalance in intracellular resource allocation. To overcome this limitation, it is necessary to improve the inherent properties ofPc APSK through protein engineering.