Chondroitin sulfate A (CSA) is a valuable glycosaminoglycan that has great market demand. However, current synthetic methods are limited by requiring the expensive sulfate group donor 3′-phosphoadenosine-5′-phosphosulfate (PAPS) and inefficient enzyme carbohydrate sulfotransferase 11 (CHST11). Herein, we report the design and integration of the PAPS synthesis and sulfotransferase pathways to realize whole-cell catalytic production of CSA. Using mechanism-based protein engineering, we improved the thermostability and catalytic efficiency of CHST11; its T m and half-life increased by 6.9°C and 3.5 h, respectively, and its specific activity increased 2.1-fold. Via cofactor engineering, we designed a dual cycle strategy of regenerating ATP and PAPS to increase the supply of PAPS. Through surface display engineering, we realized the outer membrane expression of CHST11 and constructed a whole-cell catalytic system of CSA production with a 89.5% conversion rate. This whole-cell catalytic process provides a promising method for the industrial production of CSA.

The compound 3′-phosphoadenosine-5′-phosphosulfate (PAPS) serves as a sulfate group donor in the production of valuable sulfated compounds, such as glycosaminoglycan and oxamniquine. However, elevated costs and low conversion efficiency limit the industrial applicability of PAPS. Here, we designed and constructed an efficient and controllable catalytic system for the conversion of ATP (disodium salt) into PAPS without inhibition from by-products. In vitro and in vivo testing in Escherichia coli identified adenosine-5′-phosphosulfate kinase from Penicillium chrysogenum (PcAPSK) as the rate-limiting enzyme. Based on analysis of the catalytic steps and molecular dynamics simulations, a mechanism-guided “ADP expulsion” strategy was developed to generate an improved PcAPSK variant (L7), with a specific activity of 48.94 U·mg-1 and 73.27-fold higher catalytic efficiency (kcat/Km) that of the wild-type enzyme. The improvement was attained chiefly by reducing the ADP-binding affinity of PcAPSK, as well as by changing the enzyme’s flexibility and lid structure to a more open conformation. By introducing PcAPSK L7 in an in vivo catalytic system, 73.59 mM (37.32 g·L-1) PAPS was produced from 150 mM ATP in 18.5 h using a 3-L bioreactor. The achieved titer is the highest reported to date and corresponds to a 98.13% conversion rate. The proposed strategy will facilitate industrial production of PAPS as well as the engineering of similar enzymes.