Creating enzymes and self-sufficient cells for biosynthesis of the non-natural cofactor nicotinamide cytosine dinucleotide

Abstract

Nicotinamide adenine dinucleotide (NAD) and its reduced form are indispensable cofactors in life. Diverse NAD mimics have been developed for applications in chemical and biological sciences. Nicotinamide cytosine dinucleotide (NCD) has emerged as a non-natural cofactor to mediate redox transformations, while cells are fed with chemically synthesized NCD. Here, we create NCD synthetase (NcdS) by reprograming the substrate binding pockets of nicotinic acid mononucleotide (NaMN) adenylyltransferase to favor cytidine triphosphate and nicotinamide mononucleotide over their regular substrates ATP and NaMN, respectively. Overexpression of NcdS alone in the model host Escherichia coli facilitated intracellular production of NCD, and higher NCD levels up to 5.0 mM were achieved upon further pathway regulation. Finally, the non-natural cofactor self-sufficiency was confirmed by mediating an NCD-linked metabolic circuit to convert L-malate into D-lactate. NcdS together with NCD-linked enzymes offer unique tools and opportunities for intriguing studies in chemical biology and synthetic biology.

A Phosphite-Based Sscreening Platform for Identification of Enzymes Favoring Nonnatural Cofactors

“…As a proof-of-concept, we rst applied the selection method in directed nicotinate-mononucleotide adenylyltransferase (NadD) evolution aimed at creating NCD synthetase (NcdS). [22] When Pdh* and the NCD-cycling partner are present, cell growth will depend on the intracellular NCD level which is associated with the activity of NCD synthetase (Figure 1D, E). Hence, we randomly selected WT NadD and several suspects formed during the directed evolution, including 22C8 (D22R), 23F7 (V23Q), 109H9 (D109R), 1C1 (P22K/C132L/W176L), and 22D8 (D22K), whose NCD synthesis capacity was enhanced sequentially according to previous report (Figure 3A).…”

“…Hence, we randomly selected WT NadD and several suspects formed during the directed evolution, including 22C8 (D22R), 23F7 (V23Q), 109H9 (D109R), 1C1 (P22K/C132L/W176L), and 22D8 (D22K), whose NCD synthesis capacity was enhanced sequentially according to previous report (Figure 3A). [22] The NCD cycle metabolism module was assembled on redesigned plasmid pUC18 (bla::cat) with Pdh* and ME* coexpression controlled under the lac and ara operon respectively, to give plasmid pUC-chl-(P araB )ME*+Pdh*. Plasmid expressing WT NadD or the variants and pUC-chl-(P araB )ME*+Pdh* were cotransformed into BW14329.…”

“…To further prove our hypothesis, we tested the capacity of NcdS to regulate the growth of strains holding NCD-cycling pathway. We introduced NcdS-2, NcdS-3, and the V23Q/W176E mutant of NadD (3G8) [22] on plasmid pUC-18 with Pdh* coexpression controlled under the lac operon, giving pUC-chl-NcdS2+Pdh*, pUC-chl-NcdS3+Pdh*, and pUC-chl-3G8+Pdh*. According to the previous research, NcdS-2 showed higher activity and preference of NCD biosynthesis than NcdS-3 (Table 1), whereas 3G8 had the lowest speci city.…”

“…According to the previous research, NcdS-2 showed higher activity and preference of NCD biosynthesis than NcdS-3 (Table 1), whereas 3G8 had the lowest speci city. [22] The reconstructed plasmids were separately cotransformed with pTrc99K-ME* into BW25113, to give strain BW-PB01, BW-PB03, and BW-PB05, respectively. Growth behavior of engineered strains in MOPS media with Phi as the sole P source was observed (Figure 3B).…”

“…[4] Our early study indicated that NCD was the only arti cial coenzyme which has been successfully biosynthesized and applied in orthogonal redox chemistry intracellular. [20][21][22] However, ine cient evolution of NCD-dependent enzymes limits the widespread application of NCD for biocatalysis and biotransformation. To overcome this challenge, herein, we built and characterized a phosphite-based selection platform in E. coli to identify NCD-active oxidoreductases rapidly using growth as a facile readout.…”